EA043914B1 - WAYS TO INCREASE RED CYTE AND/OR HEMOGLOBIN LEVELS AND TREAT MYELOFIBROSIS - Google Patents

Description

Cross reference to related applications

This application claims priority to US Provisional Application No. 62/367289, filed July 27, 2016. The contents of the foregoing application are incorporated by reference in their entirety.

Prior Art

Myelofibrosis is a rare disease that primarily affects older people. Myelofibrosis is a BCR-ABL1-negative myeloproliferative neoplasm that occurs de novo (primary disease) or may be preceded by polycythemia vera or essential thrombocythemia. Clinical signs include progressive anemia, severe splenomegaly, fibrosis (eg, bone marrow fibrosis), constitutional symptoms (eg, fatigue, night sweats, bone pain, itching and cough) and weight loss [Tefferi A (2000) N Engl J Med 342:1255-1265]. Median survival rates range from less than 2 years to more than 15 years, depending on currently known prognostic factors. Mutations affecting JAK2, MPL, TET2, ASXL1, IdH1/IDH2, CBL, IKZF1, LNK and EZH2 have been described in patients with myelofibrosis [James C et al. (2005) Nature 434:1144-1148, 2005; Scott L M et al. (2007) N Engl J Med 356:459-468, 2007; Pikman Y et al. (2006) PLoS Med 3:e270; Delhommeau F et al. (2009) N Engl J Med 360:2289–2301; Carbuccia N et al. (2009) Leukemia 23:2183–2186; Green A et al. (2010) N Engl J Med 362:369–370; Tefferi A et al. (2010) Leukemia 24:1302–1309; Grand F H et al. (2009) Blood 113:6182–6192; Jager R et al. (2010) Leukemia 24:1290’1298; Oh S T et al. (2010) Blood 116:988992; and Ernst T et al., Nat Genet. 42:722-726]. Certain mutations occur with high frequency in myelofibrosis (eg, JAK2 mutations in approximately 50% of patients) and either directly (eg, JAK2 or MPL mutations) or indirectly (eg, LNK or CBL mutations) induce JAK-STAT hyperactivation.

The only existing treatment for myelofibrosis is bone marrow transplantation. However, such treatment is associated with high mortality and only a small proportion of patients are suitable candidates for transplantation. Many other currently available treatments are ineffective in reversing the course of myelofibrosis, whether primary or secondary. Treatment of myelofibrosis includes, for example, cytoreductive therapy (eg, treatment with hydroxyurea), treatment of anemia with androgens and/or erythropoietin, and splenectomy. These treatments have not demonstrated improved survival and are mostly considered palliative [Cervantes F., Myelofibrosis: Biology and treatment options, European Journal of Haematology, 2007, 79 (suppl. 68) 13-17]. More recently, JAK inhibitors have been used to treat myelofibrosis. JAK inhibitors have been effective in reducing splenomegaly in patients with myelofibrosis, but otherwise their effect on the disease is largely palliative [Gupta et al. (2012) Blood 120:1367–1379]. In particular, JAK inhibitors have little or no effect on many manifestations (complications) of the disease, including, for example, cytopenias, transfusion dependence, exacerbation or blast phase of the disease, and fibrosis. In addition, JAK inhibitors have been shown to induce or worsen thrombocytopenia, anemia, and neutropenia in some patients.

Thus, there is a great need to provide effective therapies for the treatment of myelofibrosis. It is therefore an object of the present invention to provide new methods for treating or preventing myelofibrosis, in particular, treating or preventing one or more complications of myelofibrosis.

The essence of the invention

In part, this description relates to the observation that an ActRIIB (activin receptor type IIB) antagonist (inhibitor) can be used to treat myelofibrosis, in particular, improve various complications of the disease, including, for example, splenomegaly, extramedullary hematopoiesis and fibrosis. In particular, the data presented indicate that the GDF decoy polypeptide reduces splenomegaly, extramedullary hematopoiesis, and fibrosis in the JAK2V617F model of myelofibrosis. Accordingly, in some aspects, the disclosure relates to compositions and methods for treating myelofibrosis, particularly treating or preventing one or more complications of myelofibrosis (eg, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis), by administering to a patient in need thereof an effective amount one or more than one ActRIIB antagonist, optionally in combination with one or more maintenance treatments or other active agent for the treatment of myelofibrosis. Although the effects of GDF trap polypeptides on myelofibrosis may be mediated by mechanisms other than ActRIIB antagonism [e.g., inhibition of one or more of: GDF11, GDF8, Activin B, BMP6, GDF3, and BMP10 may be an indicator of the propensity of the agent to inhibit the activity of a number of additional agents, including possibly other members of the TGF-beta superfamily, and such collective inhibition may result in desired effects in, for example, myelofibrosis], however, the disclosure demonstrates that desired therapeutic agents can be selected based on ActRIIB antagonism. Thus, without wishing to be bound by any particular mechanism of action, other ActRIIB antagonists [e.g., ActRIIB receptor antagonists, antagonists of one or more ligands] are expected to

- 1 043914

ActRIIB (eg, GDF11, GDF8, activin B, BMP6, GDF3 and BMP10), antagonists of one or more type I receptors (eg, ALK4, ALK5 and/or ALK7), antagonists of one or more co-receptors and/ or antagonists of one or more downstream components of the ActRIIB signaling pathway (eg, Smad)], or combinations of such antagonists] may find use in the treatment of myelofibrosis, in particular in the treatment or prevention of one or more than one complications of myelofibrosis (eg, splenomegaly , extramedullary hematopoiesis, anemia and fibrosis). Such agents are collectively referred to herein as ActRIIB antagonists or ActRIIB inhibitors.

Accordingly, in some aspects, the disclosure relates to methods of treating myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some embodiments, the disclosure relates to methods of treating one or more than one complication of myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some aspects, the disclosure relates to methods of preventing myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some embodiments, the disclosure relates to methods of preventing one or more than one complication of myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some aspects, the disclosure relates to reducing the rate of progression of myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some embodiments, the description relates to reducing the rate of progression of one or more than one complication of myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some aspects, the disclosure relates to methods of reducing the severity of myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some embodiments, the disclosure relates to methods of reducing the severity of one or more than one complication of myelofibrosis, comprising administering to a patient in need thereof an effective amount of an ActRIIB antagonist. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient suffers from primary myelofibrosis. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has myelofibrosis that was preceded by polycythemia vera. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has myelofibrosis that is preceded by essential thrombocythemia. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has low-risk myelofibrosis according to the International Prognosis Scoring Score (IPSS). In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has intermediate-1 risk myelofibrosis according to IPSS. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has intermediate-2 risk myelofibrosis according to IPSS. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has high-risk myelofibrosis according to IPSS. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has low-risk myelofibrosis according to dynamic IPSS (DIPSS). In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has intermediate-1 risk myelofibrosis according to DIPSS. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has intermediate-2 risk myelofibrosis according to DIPSS. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has high-risk myelofibrosis according to DIPSS. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has low-risk myelofibrosis according to DIPSS-plus. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, wherein the patient

- 2 043914 has myelofibrosis with intermediate-1 risk according to DIPSS-plus. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has intermediate-2 risk myelofibrosis according to DIPSS-plus. In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has high-risk myelofibrosis according to DIPSS-plus. In some aspects, an ActRIIB antagonist can be used to prevent or delay progression of risk in myelofibrosis according to any of the established risk stratification models for myelofibrosis (eg, IPSS, DIPPS and DIPPSplus). For example, in some embodiments, an ActRIIB antagonist can be used to prevent or delay progression of IPSS, DIPPS, or DIPPS-plus risk in low to intermediate-1 myelofibrosis. In other embodiments, an ActRIIB antagonist can be used to prevent or delay the progression of IPSS, DIPPS, or DIPPS-plus risk in myelofibrosis from intermediate-1 to intermediate-2. In further embodiments, an ActRIIB antagonist can be used to prevent or delay the progression of intermediate-2 to high risk for myelofibrosis according to LPSS, DIPPS or DIPPS-plus. In some aspects, an ActRIIB antagonist can be used to promote or enhance risk reduction in myelofibrosis according to any accepted risk stratification model for myelofibrosis (eg, IPSS, DLPPS and DIPPS-plus). For example, in some embodiments, an ActRIIB antagonist can be used to help reduce or enhance risk reduction in high to intermediate-2 myelofibrosis according to LPSS, DLPPS, or DIPPS-plus. In other embodiments, an ActRIIB antagonist can be used to promote or enhance risk reduction in intermediate-2 to intermediate-1 myelofibrosis according to LPSS, DLPPS, or DLPPS-plus. In further embodiments, an ActRLLB antagonist can be used to help reduce or enhance risk reduction in myelofibrosis from intermediate-1 to low according to LPSS, DLPPS or DLPPS-plus. In some aspects, the disclosure relates to methods of using ActRLLB antagonists to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has one or more than one genetic mutation associated with myelofibrosis. For example, in some embodiments, an ActRLLB antagonist may be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, wherein the myelofibrosis is associated with one or more than one genetic mutation selected from the group consisting of: null zygosity for haplotype JAK2 46/1, JAK2V617F, IDH1, IDH2, EZH2, SRSF2, ASXL1, JAK1, JAK2, JAK3, TYK2, MPL, CALR, CALR+ASXL1-, CALR-ASKL1+, CALR+ASKL1+, CALR-ASKL1 -, TET2, TNRO and LNK. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, wherein the myelofibrosis is associated with one or more than one Janus kinase (JAK) genetic mutation (e.g., JAK1 , JAK2 and/or JAK3). In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the myelofibrosis is associated with one or more than one JAK2 genetic mutation. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the myelofibrosis is associated with the JAK2V617F mutation. In some aspects, the disclosure relates to methods of using an ActRIIB antagonist to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more than one complication of myelofibrosis, wherein the myelofibrosis is associated with an increase in one or more than one serum marker selected from the group consisting of: of: increased serum IL-8 levels, increased serum IL-2R levels and increased serum free light chain levels. In some aspects, the disclosure relates to methods of using ActRIIB antagonists to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has been treated with a Janus kinase inhibitor (e.g., ruxolitinib, fedratinib (SAR302503), monoelotinib (CYT387), pacritinib, lestaurtinib, AZD-1480, BMS-911543, NS-018, LY2784544, SEP701, XL019 and AT-9283). In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient is intolerant to a Janus kinase inhibitor. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient has an unsatisfactory response to a Janus kinase inhibitor. In some aspects, the disclosure relates to methods of using an ActRIIB antagonist to treat, prevent, or reduce the rate of progression and/or severity of myeloma.

- 3 043914 fibrosis or one or more than one complication of myelofibrosis, where the patient was treated with hydroxyurea. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis, where the patient is intolerant to hydroxyurea. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or one or more complications of myelofibrosis where the patient has an unsatisfactory response to hydroxyurea.

As described herein, myelofibrosis is a clonal neoplasm of hematopoietic tissue that is associated with a variety of clinical complications that the patient may experience as the disease progresses. The examples provided herein demonstrate that an ActRIIB antagonist can be used to alleviate a number of these clinical complications, indicating that an ActRIIB antagonist can be used more broadly to treat various complications of myelofibrosis, in contrast to the variety of currently existing treatments for myelofibrosis, which treat only a limited number of complications of the disease. Thus, in some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of ineffective hematopoiesis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of extramedullary hematopoiesis in a patient with myelofibrosis. For example, an ActRIIB antagonist may be used to treat, prevent, or reduce the rate of progression and/or severity of extramedullary hematopoiesis in the spleen (splenic extramedullary hematopoiesis) in a patient with myelofibrosis. In other embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of extramedullary hematopoiesis in the liver (hepatic extramedullary hematopoiesis) in a patient with myelofibrosis. In further embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of extramedullary hematopoiesis in the lungs (pulmonary extramedullary hematopoiesis) in a patient with myelofibrosis. In other embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of extramedullary hematopoiesis in the lymph nodes (lymphatic extramedullary hematopoiesis) in a patient with myelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of inflammation and/or increase (size) of an organ or tissue in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of inflammation and/or enlargement (size) of the spleen in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of inflammation and/or liver enlargement in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of inflammation and/or increase in lung(s) in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of inflammation and/or enlargement of lymph node(s) in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of splenomegaly in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of hepatomegaly in a patient with myelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of fibrosis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of bone marrow fibrosis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of splenic fibrosis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of liver fibrosis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of pulmonary fibrosis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of lymph node fibrosis in a patient with myelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of osteosclerosis in a patient with myelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of osteomyelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of one or more hematologic complications

- 4 043914 myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of anemia in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of thrombocytopenia in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of pancytopenia in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of poikilocytosis in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of bleeding in a patient with myelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of one or more constitutional symptoms of myelofibrosis (e.g., fatigue, itching, weight loss, night sweats, fever, abdominal pain or discomfort, paresthesia and feelings of rapid satiety). In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of tissue and/or organ pain in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of bone pain in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of arthralgia in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of myalgia in a patient with myelofibrosis. In some aspects, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of cachexia in a patient with myelofibrosis. In some aspects, the disclosure relates to increasing red blood cell levels in a patient with myelofibrosis by administering an effective amount of an ActRIIB antagonist. In some aspects, the disclosure relates to increasing hemoglobin levels in a patient with myelofibrosis by administering an effective amount of an ActRIIB antagonist. In some aspects, a patient with myelofibrosis who is to be treated with the methods of this invention is anemic. In some embodiments, an ActRIIB antagonist can be used to treat, prevent, or reduce the rate of progression and/or severity of anemia in a patient with myelofibrosis. In some aspects, the disclosure relates to methods of using an ActRIIB antagonist to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or a complication of myelofibrosis in a patient who has received one or more blood cell transfusions (whole blood or red blood cell transfusions). In some aspects, the disclosure relates to methods of using an ActRIIB antagonist to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis or a complication of myelofibrosis in a patient who is blood cell transfusion dependent. In some aspects, an ActRIIB antagonist can be used to reduce the transfusion burden of blood cells in a patient with myelofibrosis. For example, an ActRIIB antagonist can be used to reduce blood cell transfusion by more than about 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% over 4 to 8 weeks compared with an equal period of time before starting treatment with an ActRIIB antagonist. In some embodiments, an ActRIIB antagonist can be used to reduce the transfusion of blood cells in a patient with myelofibrosis by more than about 50% within 4-8 weeks compared to an equal period of time before treatment with an ActRIIB antagonist. In some aspects, an ActRIIB antagonist can be used to reduce iron overload in a patient with myelofibrosis. For example, an ActRIIB antagonist can be used to reduce iron overload of an organ or tissue in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to reduce splenic iron overload in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to reduce hepatic iron overload in a patient with myelofibrosis. In some embodiments, an ActRIIB antagonist can be used to reduce cardiac iron overload in a patient with myelofibrosis.

In any of the methods described herein, a patient with myelofibrosis may be further administered one or more additional active ingredients and/or one or more supportive therapies (in addition to the administration of one or more ActRIIB antagonists) to treat, prevent, or reducing the rate of progression and/or severity of myelofibrosis or one or more than one complication of myelofibrosis. For example, in some embodiments, the patient may additionally receive one or more than one maintenance therapy modality or may additionally administer one or more additional active agents selected from the group consisting of: blood transfusion (whole blood or red blood cell transfusion), iron chelators (e.g. , deferoxamine, deferiprone and deferasirox), corticosteroids, prednisolone, erythropoiesis stimulants (eg, erythropoietin, epoetin alfa, epoetin beta, darbepoetin alfa and methoxypolyethylene glycol epoetin beta), androgens, danazol, thalidomide, lenalidomide, cytoreductive agent, hydroxyurea, busulfan, melphalma, cladribine , splenecto

- 5 043914 myotherapy, radiation therapy, aspirin, pomalidomide, Janus kinase inhibitors, mTOR inhibitors (for example, rapamycin, sirolimus, deforolimus, everolimus, temsirolimus, NVP-BEZ235, BGT226, SF1126, PK1-587, INK128, AZD8055 and AZD20 14) and histone deacetylase inhibitors (eg, givinostat, panobinostat and pracinostat). In some aspects, the disclosure relates to methods of treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis or one or more than one complication of myelofibrosis, comprising administering to a patient in need thereof: a) a Janus kinase inhibitor, and b) an ActRIIB antagonist, wherein the inhibitor Janus kinases and the ActRIIB antagonist are administered in an effective amount. In some embodiments, the ActRIIB antagonist is administered prior to treatment with the Janus kinase inhibitor. In other embodiments, the ActRIIB antagonist is administered following treatment with a Janus kinase inhibitor. In further embodiments, the ActRIIB antagonist is administered concomitantly with the Janus kinase inhibitor. Janus kinase inhibitors for use in the methods described herein can be agents that inhibit one or more Janus kinases selected from the group consisting of: JAK1, JAK2 and JAK3. For example, a Janus kinase inhibitor may be an agent that inhibits the signaling pathway of one or more of JAK1, JAK2 and JAK3 in a cell assay. In some embodiments, the Janus kinase inhibitor for use according to the methods described herein is selected from the group consisting of: ruxolitinib, fedratinib (SAR302503), monoelotinib (CYT387), pacritinib, lestaurtinib, AZD-1480, BMS-911543, NS- 018, LY2784544, SEP-701, XL019 and AT-9283. In some preferred embodiments, the Janus kinase inhibitor for use according to the methods described herein is ruxolitinib.

Janus kinase inhibitors (eg, ruxolitinib) are approved for the treatment of various disorders, including, for example, myelofibrosis. In addition, a number of other clinical studies are being conducted to determine the effectiveness of Janus kinase inhibitors for the treatment of a number of other diseases. A common side effect of Janus kinase inhibitor therapy is anemia. While blood cell transfusion and EPO (erythropoietin) receptor activator therapy can be used to treat anemia in patients treated with a Janus kinase inhibitor, such anemia therapy is also associated with side effects in patients (eg, promotes iron overload or worsens iron overload, poor response to EPO and EPO intolerance). Thus, there is a need in the art for alternative methods of increasing red blood cell/hemoglobin levels and treating anemia in patients treated with a Janus kinase inhibitor. In part, this description relates to the observation that an ActRIIB antagonist (inhibitor) can be used to increase red blood cell and hemoglobin levels in patients treated with a Janus kinase inhibitor. Accordingly, in some aspects, the disclosure relates to compositions and methods of increasing red blood cell/hemoglobin levels and treating or preventing anemia in patients treated with a Janus kinase inhibitor by administering to the patient in need an effective amount of one or more ActRIIB antagonists, possibly in combination with one or more than one maintenance therapy or other active agent for the treatment of anemia. Although the effects of GDF decoy polypeptides on red blood cell and/or hemoglobin levels may be mediated by a mechanism other than ActRIIB antagonism [e.g., inhibition of one or more of: GDF11, GDF8, Activin B, BMP6, GDF3, and BMP10 may be an indicator of the agent's propensity to inhibit activity of a number of additional agents, including possibly other members of the TGF-beta superfamily, and such collective inhibition may result in desired effects on, for example, red blood cell and/or hemoglobin levels], however, the disclosure demonstrates that the desired therapeutic agents can be selected by ActRIIB antagonism. Thus, without wishing to be bound by any particular mechanism of action, other ActRIIB antagonists [e.g., ActRIIB receptor antagonists, antagonists of one or more ActRIIB ligands (e.g., GDF11, GDF8, Activin B, BMP6, GDF3, and BMP10) are expected to ), antagonists of one or more type I receptors (e.g. ALK4, ALK5 and/or ALK7), antagonists of one or more co-receptors and/or antagonists of one or more downstream components of the ActRIIB signaling pathway (e.g. Smad)], or combinations of such antagonists] may be useful in the treatment of patients treated with a Janus kinase inhibitor, particularly in the treatment or prevention of one or more complications associated with Janus kinase inhibitor therapy (eg, anemia, thrombocytopenia and/or neutropenia). Such agents are collectively referred to herein as ActRIIB antagonists or ActRIIB inhibitors.

In some aspects, the disclosure relates to methods of increasing red blood cell and/or hemoglobin levels in a patient being treated with a Janus kinase inhibitor by administering to the patient in need thereof an effective amount of an ActRIIB antagonist. In some embodiments, an ActRIIB antagonist can be used to treat or prevent anemia in a patient treated with a Janus kinase inhibitor. In some embodiments, a patient receiving treatment with a Janus kinase inhibitor may receive one or more blood cell transfusions prior to treatment with an ActRIIB antagonist. In some embodiments, a patient treated with a Janus kinase inhibitor is blood cell transfusion dependent. In some aspects, the disclosure relates to methods of using an ActRIIB antagonist to reduce blood cell transfusion burden in a patient,

- 6 043914 treated with a Janus kinase inhibitor. For example, an ActRIIB antagonist can be used to reduce blood cell transfusion by more than about 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% over 4 to 8 weeks compared with an equal period of time before starting treatment with ActRIIB antagonists in a patient treated with a Janus kinase inhibitor. In some embodiments, an ActRIIB antagonist can be used to reduce blood cell transfusion in a patient treated with a Janus kinase inhibitor by more than about 50% over a period of 4 to 8 weeks compared to an equal period of time before initiation of treatment with an ActRIIB antagonist. In some aspects, the disclosure relates to methods of using an ActRIIB antagonist to reduce iron overload in a patient being treated with a Janus kinase inhibitor. In some embodiments, an ActRIIB antagonist can be used to reduce liver iron in a patient being treated with a Janus kinase inhibitor. In some embodiments, an ActRIIB antagonist can be used to reduce the iron content of the spleen of a patient treated with a Janus kinase inhibitor. In some embodiments, an ActRIIB antagonist can be used to reduce iron content in the heart of a patient being treated with a Janus kinase inhibitor. In some embodiments, the ActRIIB antagonist is administered prior to treatment with the Janus kinase inhibitor. In other embodiments, the ActRIIB antagonist is administered following treatment with a Janus kinase inhibitor. In further embodiments, the ActRIIB antagonist is administered concomitantly with the Janus kinase inhibitor. In some aspects, a patient being treated with a Janus kinase inhibitor receives an agent that inhibits one or more Janus kinases selected from the group consisting of JAK1, JAK2, and JAK3. In some embodiments, the Janus kinase inhibitor inhibits signaling of one or more of JAK1, JAK2, and JAK3 in a cell assay. For example, a patient could be treated with one or more Janus kinase inhibitors selected from the group consisting of: ruxolitinib, fedratinib (SAR302503), monoelotinib (CYT387), pacritinib, lestaurtinib, AZD-1480, BMS-911543, NS-018 , LY2784544, SEP-701, XL019 and AT-9283. In some embodiments, the patient may be treated with ruxolitinib.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least GDF11 (eg, a GDF11 antagonist). The effect of GDF11 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least GDF11. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds at least 1x10'7 M to GDF11 KD (e.g., at least 1x10'8 M, at least 1x10'9 M, at least ^1x10'10 M , at least 1x10'11 M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit GDF11 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of antagonists that inhibit GDF11 may further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least GDF8 (eg, a GDF8 antagonist). The effect of GDF8 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least GDF8. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds to at least GDF8 with a KD of at least 1x10'7 M (e.g., at least 1x10'8 M, at least 1x10'9 M, at least ^1x10'10 M, at least 1x10'11 M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit GDF8 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of antagonists that inhibit GDF8 can further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF11, GDF3, BMP6, BMP10, ActRIIB, ALK4,

- 7 043914

ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least GDF3 (eg, a GDF3 antagonist). The effect of GDF3 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least GDF3. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds to at least GDF3 with a KD of at least ^1x10'7 M (e.g., at least ^1x10'8 M, at least ^1x10'9 M, at least 1x10'1 ° M, at least 1x10'11 M or at least 1x1°' ¹² M). As described herein, various ActRIIB antagonists that inhibit GDF3 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of antagonists that inhibit GDF3 can further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF11, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least BMP6 (eg, a BMP6 antagonist). The effect of BMP6 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least BMP6. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRUB antagonist or combination of antagonists as described binds to at least BMP6 with a KD of at least 1 x 10' ⁷ M (e.g., at least 1 x 1°' ⁸ M, at least 1 x 10' ⁹ M, at least 1 x 10' 1° M, at least 1x1°' ^p M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit BMP6 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of antagonists that inhibit BMP6 can further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, GDF11, BMP10, ActRIIB, ALK4, ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least BMP10 (eg, a BMP10 antagonist). The effect of BMP10 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least BMP10. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds at least BMP10 with a KD of at least 1 x 10' ⁷ M (e.g., at least 1 x 10' ⁸ M, at least 1 x 10' ⁹ M, at least 1 x 10' ¹⁰ M, at least 1x10' ¹¹ M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit BMP10 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of antagonists that inhibit BMP10 can further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, GDF11, BMP6, ActRIIB, ALK4, ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin

- 8 043914

BC, activin E, activin AE and/or activin BE) (for example, activin antagonist). The effect of activin inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least activin. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds at least activin with a K _D of at least 1 x 10' ⁷ M (e.g., at least 1 x 1°' ⁸ M, at least 1 x 10' ⁹ M, at least 1 x 10 '1° M, at least 1x10'11 M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit activin can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of activin-inhibiting antagonists can further inhibit one or more of the following: GDF8, GDF3, GDF11, BMP6, BMP10, ActRIIB, ALK4, ALK5, and ALK7. In some preferred embodiments, the ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least activin B. In some embodiments, the ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein document, substantially does not bind to activin A (e.g., binds to activin A with a KD greater than 1 x 10' ⁷ M or has relatively moderate binding, e.g., approximately 1 x 1°' ⁸ M or approximately 1 x 10' ⁹ M) and/or inhibits activin activity A. In some preferred embodiments, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least activin B, but does not substantially bind to activin A (e.g., binds to activin A with KD exceeding 1x10' ⁷ M or has relatively moderate binding, for example, approximately 1x1°' ⁸ M or approximately 1x10' ⁹ M) and/or inhibits the activity of activin A.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least ActRIIB (eg, an ActRIIB antagonist). The effect of ActRIIB inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least ActRIIB. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds at least ActRIIB with a K _D of at least 1 x 10' ⁷ M (e.g., at least 1 x 1°' ⁸ M, at least 1 x 10' ⁹ M, at least 1 x 1 °' ¹⁰ M, at least 1x1°' ^p M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit ActRIIB can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of antagonists that inhibit ActRIIB can further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, GDF11, BMP6, BMP10, ALK4, ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least ALK4 (eg, an ALK4 antagonist). The effect of ALK4 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least ALK4. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists as described binds to at least AL4K with a KD of at least 1 x 10' ⁷ M (e.g., at least 1 x 1°' ⁸ M, at least 1 x 10' ⁹ M, at least 1 x 1° ' ¹⁰ M, at least 1x1°' ^p M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit ALK4 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules,

- 9 043914 nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of ALK4 inhibitory antagonists may further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, GDF11, BMP6, BMP10, ActRIIB, ALK5 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least ALK5 (eg, an ALK5 antagonist). The effect of ALK5 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least ALK5. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists described binds to at least ALK5 with a KD of at least ^1x10'7 M (e.g., at least ^1x10'8 M, at least ^1x10'9 M, at least 1x10'1 ° M, at least 1x10'11 M or at least 1x1°' ¹² M). As described herein, various ActRIIB antagonists that inhibit ALK5 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of ALK5 inhibitory antagonists may further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, GDF11, BMP6, BMP10, ActRIIB, ALK4 and ALK7.

In some aspects, an ActRIIB antagonist or combination of antagonists for use according to the methods and uses described herein is an agent that inhibits at least ALK7 (eg, an ALK7 antagonist). The effect of ALK7 inhibition can be determined, for example, using cell-based assays, including those described herein (eg, Smad signaling pathway reporter assay). Thus, in some embodiments, an ActRIIB antagonist or combination of antagonists as described may bind to at least ALK7. Ligand binding activity can be determined, for example, using a binding affinity test, including those described herein. In some embodiments, an ActRIIB antagonist or combination of antagonists described binds to at least ALK7 with a KD of at least ^1x10'7 M (e.g., at least ^1x10'8 M, at least ^1x10'9 M, at least 1x10'1 ° M, at least 1x10' ^p M or at least 1x10' ¹² M). As described herein, various ActRIIB antagonists that inhibit ALK7 can be used according to the methods and uses described herein, including, for example, ligand decoys (e.g., ActRIIB polypeptides, GDF decoys, follistatin polypeptides, and FLRG polypeptides), antibodies, small molecules, nucleotide sequences and their combinations. In some embodiments, an ActRIIB antagonist or combination of ALK7 inhibitory antagonists may further inhibit one or more of the following: activin (e.g., activin A, activin B, activin AB, activin C, activin AC, activin BC, activin E, activin AE and/or activin BE), GDF8, GDF3, GDF11, BMP6, BMP10, ActRIIB, ALK5 and ALK4.

Part of the description relates to ActRIIB antagonists, which are ActRIIB polypeptides. The term ActRIIB polypeptide includes naturally occurring ActRIIB polypeptides, as well as truncated forms and variants thereof, such as those described herein (eg, GDF trap polypeptides). Preferably, the ActRIIB polypeptides contain, consist essentially of, or consist of a ligand binding domain of an ActRIIB polypeptide or a modified form (variant) thereof. For example, in some embodiments, ActRIIB polypeptides contain, consist essentially of, or consist of a ligand binding domain of an ActRIIB polypeptide, for example, part of an ActRIIB extracellular domain. Preferably, ActRIIB polypeptides for use according to the methods described herein are soluble polypeptides.

In some aspects, the description relates to compositions containing an ActRIIB polypeptide and their use. For example, in some embodiments, the ActRIIB polypeptide is described to contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96% , 97%, 98%, 99%, or 100% identical to amino acid sequence 29-109 of SEQ ID NO: 1. In some embodiments, an ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85 %, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acid sequence 29-109 of SEQ ID NO: 1, where the polypeptide is ActRIIB contains a negatively charged amino acid [naturally occurring (E or D) or a negatively charged amino acid of artificial origin] at position 79 relative to

- 10 043914 specifically SEQ ID NO: 1. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94 %, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acid sequence 25-131 of SEQ ID NO: 1. In some embodiments, an ActRIIB polypeptide may comprise an amino acid sequence that is at least 70% identical 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acid sequence 25-131 of SEQ ID NO: 1, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90% , 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 1. In some embodiments, the ActRIIB polypeptide may contain an amino acid sequence that at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid identical sequence of SEQ ID NO: 1, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85% , 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 2. In other embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100 % identical to the amino acid sequence of SEQ ID NO: 2, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In further embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80% 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 3. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99 % or 100% identical to the amino acid sequence of SEQ ID NO: 3, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75% identical , 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 4. In some In embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98 %, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 4, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 4. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70% identical to SEQ ID NO: 4. , 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 5 In some embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97 %, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 5, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 4. In other embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 6. In some embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96 %, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 6, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 4. In further embodiments, the ActRIIB polypeptide may contain an amino acid sequence that at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid identical sequence SEQ ID NO: 24. In further embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95 %, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 25. In further embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85 %, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 28. In further embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 29. In some embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 29, wherein the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In the following embodiments, the polypeptide ActRIIB

- 11 043914 may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98 %, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 30. In some embodiments, an ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92 %, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 30, where the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1 In further embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97 %, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 31. In some embodiments, an ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 31, where the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In the following embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96 %, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 45. In some embodiments, an ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90 %, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 45, where the ActRIIB polypeptide contains a negatively charged amino acid at position 79 with respect to SEQ ID NO: 1. In the following embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95 %, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 50. In some embodiments, an ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85 %, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 50, where the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In further embodiments, the ActRIIB polypeptide may comprise an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94 %, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 53. In some embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80 %, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 53, where the ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In the following embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93 %, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 54. In some embodiments, an ActRIIB polypeptide may contain an amino acid sequence that is at least 70% identical to SEQ ID NO: 54. %, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 54, where The ActRIIB polypeptide contains a negatively charged amino acid at position 79 relative to SEQ ID NO: 1. In the following embodiments, the ActRIIB polypeptide may contain an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92 %, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 58. In some embodiments, ActRIIB polypeptides for use according to the methods and uses described herein , do not contain a negatively charged amino acid at the position corresponding to L79 of SEQ ID NO: 1.

As described herein, ActRIIB polypeptides and variants thereof (GDF traps) can be homomultimers, eg, homodimers, homotrimers, homotetramers, homopentamers, and higher order homomultimeric complexes. In some preferred embodiments, ActRIIB polypeptides and variants thereof are homodimers. In some embodiments, the ActRIIB polypeptide dimers described herein comprise a first ActRIIB polypeptide covalently or non-covalently linked to a second ActRIIB polypeptide, wherein the first polypeptide comprises an ActRIIB domain and the amino acid sequence of the first member (or second member) of the interacting pair (e.g., an immunoglobulin constant domain ), and the second polypeptide comprises an ActRIIB polypeptide and the amino acid sequence of the second member (or first member) of the interacting pair.

In some aspects, ActRIIB polypeptides, including variants thereof (eg, GDF traps), may be fusion proteins. For example, in some embodiments, an ActRIIB polypeptide may be a fusion protein comprising an ActRIIB polypeptide domain and one or more heterologous (non-ActRIIB) polypeptide domains. In some embodiments, the ActRIIB polypeptide may be a fusion protein that has as one domain an amino acid sequence derived from the ActRIIB polypeptide (e.g., a receptor ligand binding domain

- 12 043914 ActRIIB torus or variant thereof) and one or more heterologous domains that provide a desired property such as improved pharmacokinetics, easier purification, targeting specific tissues, etc. For example, the domain of a fusion protein may enhance one or more of the following: in vivo stability, in vivo half-life, absorption/administration, tissue localization or distribution, protein complex formation, multimerization of the fusion protein, and/or purification. Optionally, the ActRIIB polypeptide domain of the fusion protein is directly fused to one or more domains of a heterologous polypeptide, or a mediating sequence, such as a linker, may be located between the amino acid sequence of the ActRIIB polypeptide and the amino acid sequence of one or more domains. In some embodiments, the ActRIIB fusion protein contains a relatively unstructured linker located between the heterologous domain and the ActRIIB domain. This unstructured linker may correspond to an unstructured region of approximately 15 amino acids located at the C-terminus of the ActRIIB extracellular domain (tail), or it may be an artificial sequence of 3 to 15, 20, 30, 50 or more amino acids that is relatively free of secondary structures. The linker may be enriched in glycine and proline residues and may, for example, contain repeated sequences of threonine/serine and glycine. Examples of linkers include, but are not limited to, the sequences TGGG (SEQ ID NO: 18), SGGG (SEQ ID NO: 19), TGGGG (SEQ ID NO: 16), SGGGG (SEQ ID NO: 17), GGGGS (SEQ ID NO: 20), GGGG (SEQ ID NO: 15) and GGG (SEQ ID NO: 14). In some embodiments, the ActRIIB fusion protein may comprise an immunoglobulin constant domain, including, for example, an immunoglobulin Fc portion. For example, an amino acid sequence that is derived from the Fc domain of the immunoglobulin IgG (IgG1, IgG2, IgG3 or IgG4), IgA (IgA1 or IgA2), IgE or IgM. For example, the Fc portion of an immunoglobulin domain may comprise, consist essentially of, or consist of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94% , 95%, 96%, 97%, 98%, 99%, or 100% identical to any of SEQ ID NO: 9-13. Such immunoglobulin domains may contain one or more amino acid modifications (eg, deletion, insertion and/or substitution) that impart altered activity to the Fc, eg, impair one or more than one Fc effector function. In some embodiments, the ActRIIB fusion protein contains the amino acid sequence shown in the formula A-B-C. For example, part B is an N- and C-terminally truncated ActRIIB polypeptide described herein. Parts A and C may independently represent zero, one, or more than one amino acid, and both parts A and C are heterologous to B. Parts A and/or C may be attached to part B via a linker sequence. In some embodiments, the ActRIIB fusion protein contains a leader sequence. The leader sequence may be a native ActRIIB leader sequence or a heterologous leader sequence. In some embodiments, the leader sequence is a tissue plasminogen activator (TPA) leader sequence.

The ActRIIB polypeptide, including variants thereof (eg, GDF traps), may contain a purification sequence such as an epitope tag, a FLAG tag, a polyhistidine sequence, and a GST fusion. Optionally, the ActRIIB polypeptide contains one or more modified amino acid residues selected from: a glycosylated amino acid, a PEGylated amino acid, a farnesylated amino acid, an acetylated amino acid, a biotinylated amino acid, and/or an amino acid conjugated to a lipid moiety. ActRIIB polypeptides may contain at least one N-linked sugar, and may include two, three or more N-linked sugars. Such polypeptides may also contain O-linked sugars. In general, it is preferred that ActRIIB polypeptides be expressed in a mammalian cell line that mediates proper glycosylation of the polypeptide to reduce the likelihood of an unwanted immune response in the patient. ActRIIB polypeptides can be produced in a variety of cell lines that glycosylate the protein in a manner suitable for use in a patient, including engineered insect or yeast cells and mammalian cells such as COS cells, CHO cells, HEK cells and NSO cells. In some embodiments, the ActRIIB polypeptide is glycosylated and has a glycosylation pattern derived from a Chinese hamster ovary cell line. In some embodiments, ActRIIB polypeptides are described to have a half-life in the serum of a mammal (e.g., mouse or human) of at least 4, 6, 12, 24, 36, 48, or 72 hours. Optionally, ActRIIB may have a half-life in the serum of a mammal (e.g. , mouse or human) for at least 6, 8, 10, 12, 14, 20, 25 or 30 days.

Some aspects of the disclosure provide pharmaceutical preparations comprising one or more ActRIIB antagonists as defined herein and a pharmaceutically acceptable carrier. The pharmaceutical preparation may also contain one or more additional active agents, such as a compound, which is used for the treatment of myelofibrosis, in particular, the treatment or prevention of one or more than one complications of myelofibrosis (for example, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis), and/or treating a patient treated with a Janus kinase inhibitor. In general, the pharmaceutical preparation should preferably be apiro

- 13 043914 genetic (means free from pyrogens to a degree that meets regulatory requirements for the quality of products used in therapy).

In some cases, when administering an ActRIIB antagonist or a combination of antagonists as described for the disorders or conditions described herein, it may be desirable to monitor the effect on red blood cells when administering the ActRIIB antagonist or to determine or adjust the dosage of the ActRIIB antagonist to reduce the undesirable effect on red blood cells. For example, increases in red blood cell levels, hemoglobin levels, or hematocrit levels may cause an undesirable increase in blood pressure.

In some aspects, the ActRIIB antagonist is an antibody or combination of antibodies. In some embodiments, the antibody binds to at least ActRIIB. In some embodiments, an antibody that binds to ActRIIB inhibits ActRIIB signaling, possibly in a cell assay such as that described herein. In some embodiments, an antibody that binds to ActRIIB inhibits the binding of one or more TGF-beta superfamily ligands, TGF-beta superfamily type I receptor, or TGFbeta superfamily co-receptor to ActRIIB. In some embodiments, an antibody that binds to ActRIIB inhibits the binding to ActRIIB of one or more TGF-beta superfamily ligands selected from the group consisting of: activin (e.g., activin A, activin B, activin C, activin AB, activin AC , activin BC, activin E, activin AE and activin BE), GDF8, GDF11, GDF3, BMP6, BMP10, BMP9 and BMP5. In some embodiments, the antibody binds to at least GDF11. In some embodiments, an antibody that binds to GDF11 inhibits ActRIIB signaling, possibly in a cellular assay such as that described herein. In some embodiments, an antibody that binds to GDF11 inhibits the binding of GDF11-ActRIIB. In some embodiments, the antibody binds to at least GDF8. In some embodiments, an antibody that binds to GDF8 inhibits ActRIIB signaling, possibly in a cellular assay such as that described herein. In some embodiments, an antibody that binds to GDF8 inhibits the binding of GDF8ActRIIB. In some embodiments, the antibody binds to at least BMP6. In some embodiments, an antibody that binds to BMP6 inhibits ActRIIB signaling, possibly in a cell assay such as that described herein. In some embodiments, an antibody that binds to BMP6 inhibits BMP6-ActRIIB binding. In some embodiments, the antibody binds to at least BMP10. In some embodiments, an antibody that binds to BMP10 inhibits ActRIIB signaling, possibly in a cellular assay such as that described herein. In some embodiments, an antibody that binds to BMP10 inhibits the binding of BMP10'ActRIIB. In some embodiments, the antibody binds to at least GDF3. In some embodiments, an antibody that binds to GDF3 inhibits ActRIIB signaling, possibly in a cellular assay such as that described herein. In some embodiments, an antibody that binds to GDF3 inhibits the binding of GDF3-ActRIIB. In some embodiments, the antibody binds to at least activin (eg, activin A, activin B, activin C, activin AB, activin AC, activin BC, activin E, activin AE, and activin BE). In some embodiments, an antibody that binds to activin (e.g., activin A, activin B, activin C, activin AB, activin AC, activin BC, activin E, activin AE, and activin BE) inhibits ActRIIB signaling, possibly in a study on cells such as those described herein. In some embodiments, an antibody that binds to activin (eg, activin A, activin B, activin C, activin AB, activin AC, activin BC, activin E, activin AE, and activin BE) inhibits activin-ActRIIB binding. In some embodiments, the antibody binds to activin B. In some embodiments, an antibody that binds to activin B inhibits ActRIIB signaling, possibly in a cell assay such as that described herein. In some embodiments, an antibody that binds to activin B inhibits the binding of activin B-ActRIIB. In some embodiments, the antibody is a multispecific antibody or a combination of multispecific antibodies that binds to one or more of: ActRIIB, GDF11, GDF8, activin A, activin B, BMP6 and BMP10. In some embodiments, the antibody binds to at least ALK4. In some embodiments, an antibody that binds to ALK4 inhibits ALK4 signaling, possibly in a cellular assay such as that described herein. In some embodiments, an antibody that binds to ALK4 inhibits the binding of one or more ActRIIB ligand, type II receptor, or co-receptor to ALK4. In some embodiments, an antibody that binds to ALK4 inhibits the binding to ALK4 of one or more ActRIIB ligands selected from the group consisting of: activin (e.g., activin A, activin B, activin C, activin AB, activin AC, activin BC , activin E, activin AE and activin BE), GDF8, GDF11, BMP6, BMP10 and GDF3. In some embodiments, the antibody binds to at least ALK5. In some embodiments, an antibody that binds to ALK5 inhibits ALK5 signaling, possibly in a cellular assay such as that described herein. In some embodiments, an antibody that binds to ALK5 inhibits the binding to ALK5 of one or more ActRIIB ligands, type II receptor, or co

- 14 043914 receptor. In some embodiments, an antibody that binds to ALK5 inhibits the binding to ALK5 of one or more ActRIIB ligands selected from the group consisting of: activin (e.g., activin A, activin B, activin C, activin AB, activin AC, activin BC , activin E, activin AE and activin BE), GDF8, GDF11, BMP6, BMP10 and GDF3. In some embodiments, the antibody binds to at least ALK7. In some embodiments, an antibody that binds to ALK7 inhibits ALK7 signaling, possibly in a cell assay such as those described herein. In some embodiments, an antibody that binds to ALK7 inhibits the binding of one or more ActRIIB ligand, type II receptor, or co-receptor to ALK7. In some embodiments, an antibody that binds to ALK7 inhibits the binding to ALK7 of one or more ActRIIB ligands selected from the group consisting of: activin (e.g., activin A, activin B, activin C, activin AB, activin AC, activin BC , activin E, activin AE and activin BE), GDF8, GDF11, BMP6, BMP10 and GDF3. In some embodiments, the antibody binds to at least GDF11. In some aspects, the multispecific antibody or combination of multispecific antibodies in a cell assay inhibits signaling of one or more of ActRIIB, GDF11, GDF8, activin A, activin B, GDF3, BMP6, and BMP10. In some embodiments, the antibody is a chimeric antibody, a humanized antibody, or a human antibody. In some embodiments, the antibody is a single chain antibody, an F(ab')2 fragment, a single chain diabody, tandem single chain Fv fragments, a tandem single chain diabody, or a fusion protein comprising a single chain diabody and at least a portion of an immunoglobulin heavy chain constant region.

In some aspects, the ActRIIB antagonist is a small molecule inhibitor or a combination of small molecule inhibitors. In some embodiments, the small molecule inhibitor is an inhibitor of at least ActRIIB. In some embodiments, the small molecule inhibitor is an inhibitor of at least ALK4. In some embodiments, the small molecule inhibitor is an inhibitor of at least ALK5. In some embodiments, the small molecule inhibitor is an inhibitor of at least ALK7. In some embodiments, the small molecule inhibitor is an inhibitor of at least GDF11. In some embodiments, the small molecule inhibitor is an inhibitor of at least GDF8. In some embodiments, the small molecule inhibitor is an inhibitor of at least BMP6. In some embodiments, the small molecule inhibitor is an inhibitor of at least BMP10. In some embodiments, the small molecule inhibitor is an inhibitor of at least GDF3. In some embodiments, the small molecule inhibitor is an inhibitor of at least activin (eg, activin A, activin B, activin C, activin AB, activin AC, activin BC, activin E, activin AE, and activin BE). In some embodiments, the small molecule inhibitor is an inhibitor of at least activin B.

In some aspects, the ActRIIB antagonist is a nucleic acid inhibitor or a combination of nucleic acid inhibitors. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least ActRIIB. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least ALK4. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least ALK5. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least ALK7. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least GDF11. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least GDF8. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least BMP6. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least BMP10. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least GDF3. In some embodiments, the nucleic acid inhibitor is an inhibitor of at least activin (eg, activin A, activin B, activin C, activin AB, activin AC, activin BC, activin E, activin AE, and activin BE). In some embodiments, the nucleic acid inhibitor is an inhibitor of at least activin B.

In some aspects, the ActRIIB antagonist is a follistatin polypeptide. In some embodiments, the follistatin polypeptide contains an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98 %, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 63. In some embodiments, the follistatin polypeptide contains an amino acid sequence that is at least 70%, 75% 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 64. In some embodiments, the follistatin polypeptide contains an amino acid sequence that is at least 70% identical to 75 % 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 65. In some embodiments of the follistatin polypeptide contains an amino acid sequence

- 15 043914 rate, which is at least 70%, 75% 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 66. In some embodiments, the follistatin polypeptide contains an amino acid sequence that is at least 70%, 75% 80%, 85%, 90%, 91%, 92%, 93%, 94 %, 95%, 96%, 97%, 98%, 99%, or 100% identical to the amino acid sequence of SEQ ID NO: 67.

In some aspects, the ActRIIB antagonist is an FLRG polypeptide. In some embodiments, the FLRG polypeptide contains an amino acid sequence that is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98 %, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 68.

Brief description of drawings

In fig. 1 shows an alignment of the extracellular domains of human ActRIIA (SEQ ID NO: 36) and human ActRIIB (SEQ ID NO: 2), with residues predicted from complete analysis of multiple crystal structures of ActRIIB and ActRIIA that directly contact the ligand being boxed.

In fig. 2 shows an alignment of multiple ActRIIB protein sequences from various vertebrates and human ActRIIA (SEQ ID NOs: 37-43), as well as a consensus ActRII sequence deduced from the alignment (SEQ ID NO: 44).

In fig. 3 shows the complete amino acid sequence of the GDF trap ActRIIB(L79D 20134)-hFc (SEQ ID NO: 45), including the TPA leader sequence (double underline), the extracellular domain of ActRIIB (residues 20-134 in SEQ ID NO: 1; single underline), and hFc domain. The aspartate substituted at position 79 in the native sequence is double underlined and highlighted, as is glycine, which is sequenced to be the N-terminal residue in the mature fusion protein.

In fig. 4A and 4B show the nucleotide sequence encoding ActRIIB(L79D 20-134)hFc. SEQ ID NO: 48 corresponds to the sense strand, and SEQ ID NO: 49 corresponds to the antisense strand. The TPA leader sequence (nucleotides 1–66) is underlined by a double line, and the extracellular domain of ActRIIB (nucleotides 76–420) is underlined by a single line.

In fig. 5 shows the complete amino acid sequence of the truncated GDF trap ActRIIB(L79D 25-131)-hFc (SEQ ID NO: 50), including the TPA leader sequence (double underline), the truncated extracellular domain of ActRIIB (residues 25-131 in SEQ ID NO: 1; single underscore) and hFc domain. The aspartate substituted at position 79 in the native sequence is double underlined and highlighted, as is glutamate, which is sequenced to be the N-terminal residue in the mature fusion protein.

In fig. 6A and 6B show the nucleotide sequence encoding ActRIIB(L79D 25-131)hFc. SEQ ID NO: 51 corresponds to the sense strand, and SEQ ID NO: 52 corresponds to the antisense strand. The TPA leader sequence (nucleotides 1–66) is underlined by a double line, and the truncated extracellular domain of ActRIIB (nucleotides 76–396) is underlined by a single line. Also shown is the amino acid sequence of the extracellular domain of ActRIIB (residues 25-131 in SEQ ID NO: 1).

In fig. 7 shows the complete amino acid sequence of the truncated GDF trap ActRIIB(L79D 25-131)-hFc without the leader sequence (SEQ ID NO: 53). The truncated extracellular domain of ActRIIB (residues 25-131 in SEQ ID NO: 1) is underlined. The aspartate substituted at position 79 in the native sequence is double underlined and highlighted, as is glutamate, which is sequenced to be the N-terminal residue in the mature fusion protein.

In fig. 8 shows the complete amino acid sequence of the truncated GDF trap ActRIIB(L79D 25-131)-hFc without the leader sequence, hFc domain and linker (SEQ ID NO: 54). The aspartate substituted at position 79 in the native sequence is underlined and highlighted, as is glutamate, which is sequenced to be the N-terminal residue in the mature fusion protein.

In fig. 9A and 9B show an alternative nucleotide sequence encoding ActRIIB(L79D 25-131)-hFc. SEQ ID NO: 55 corresponds to the sense strand, and SEQ ID NO: 56 corresponds to the antisense strand. The TPA leader sequence (nucleotides 1-66) is double-underlined and the truncated ActRIIB extracellular domain (nucleotides 76-396) is underlined, and substitutions in the wild-type extracellular domain nucleotide sequence are double-underlined and highlighted (compare with SEQ ID NO: 51, FIG. 6A and 6B). Also shown is the amino acid sequence of the extracellular domain of ActRIIB (residues 25-131 in SEQ ID NO: 1).

In fig. 10 shows nucleotides 76-396 (SEQ ID NO: 57) of the alternative nucleotide sequence shown in FIG. 9A and 9B (SEQ ID NO: 55). The same nucleotide substitutions indicated in Fig. 9A and 9B are also underlined and highlighted. SEQ ID NO: 57 encodes only the truncated extracellular domain of ActRIIB (corresponding to residues 25-131 in SEQ ID NO: 1) with the L79D substitution, for example, ActRIIB(L79D25-131).

In fig. 11 shows a multiple sequence alignment of human IgG Fc domains of different isotypes using Clustal 2.1. The hinge areas are marked with dots.

In fig. 12 shows the complete amino acid sequence of ActRIIB (25-131)-hFc (SEQ ID NO:

- 16 043914

58), not subjected to processing. The TPA leader sequence (residues 1-22) and the doubly truncated extracellular domain of ActRIIB (residues 24-131, using numbering based on the native sequence in SEQ ID NO: 1) are underlined. Glutamate was isolated and was sequenced to be the N-terminal amino acid in the mature fusion protein, located at position 25 relative to SEQ ID NO: 1.

In fig. 13A and 13B show the nucleotide sequence encoding ActRIIB(25-131)-hFc (coding strand shown at top, SEQ ID NO: 59, and complementary strand shown at bottom 3'-5', SEQ ID NO: 60). Sequences encoding the TPA leader sequence (nucleotides 1–66) and the extracellular domain of ActRIIB (nucleotides 73–396) are underlined. The corresponding amino acid sequence of ActRIIB(25-131) is also shown.

In fig. 14A and 14B show an alternative nucleotide sequence encoding ActRIIB(25-131)-hFc (coding strand shown at top, SEQ ID NO: 61, and complementary strand shown at bottom 3'-5', SEQ ID NO: 62). This sequence ensures a higher level of protein expression in the first transformants, which accelerates the development of the cell line. Sequences encoding the TPA leader sequence (nucleotides 1-66) and the extracellular domain of ActRIIB (nucleotides 73-396) are underlined, and substitutions of the wild-type extracellular domain nucleotide sequence (see FIGS. 13A and 13B) are highlighted. The corresponding amino acid sequence of ActRIIB(25-131) is also shown.

Detailed Description of the Invention

1. General information.

The transforming growth factor-beta (TGF-beta) superfamily includes various growth factors that share common sequence elements and structural motifs. These proteins are known to have biological effects on various cell types in vertebrates and invertebrates. Members of the family perform important functions during embryogenesis in tissue patterning and specialization and can influence various differentiation processes, including adipogenesis, myogenesis, chondrogenesis, cardiogenesis, hematopoiesis, neurogenesis, and epithelial cell differentiation. By regulating the activity of a member of the TGF-beta family, it is often possible to cause significant biological changes in the body. For example, the Piedmontese and Belgian Blue cattle breeds carry a loss-of-function mutation in the GDF8 gene (also referred to as myostatin), resulting in a significant increase in muscle mass [see, for example, Grobet et al. (1997) Nat Genet. 17(1):71-4]. Additionally, in humans, inactive GDF8 alleles are associated with increased muscle mass and, according to available data, exceptional strength [see, for example, Schuelke et al. (2004) N Engl J Med 350:2682-8].

TGF-β signaling is mediated by heteromeric type I and type II receptor serine/threonine kinase complexes that, when stimulated by ligands, phosphorylate and activate downstream signaling pathway proteins, the SMAD proteins (e.g., SMAD proteins 1, 2, 3, 5, and 8) [see ., for example, Massague (2000) Nat. Rev. Mol. Cell Biol. 1:169-178]. These type I and II receptors are transmembrane proteins consisting of a ligand-binding extracellular domain with a cysteine-rich region, a transmembrane domain, and a cytoplasmic domain with predicted serine/threonine specificity. Type I receptors play an essential role in signal transmission. Type II receptors are required for ligand binding and for activation of type I receptors. Type I and II activin receptors form a stable complex upon binding of ligands, resulting in phosphorylation of type I and II receptors.

Two related type II receptors (ActRII), ActRIIA and ActRIIB, have been identified as type II activin receptors [see, for example, Mathews and Vale (1991) Cell 65:973-982; and Attisano et al. (1992) Cell 68: 97-108]. In addition to the activins, ActRIIA and ActRIIB can interact biochemically with several other TGF-β family proteins, including, for example, BMP6, BMP7, Nodal, GDF8 and GDF11 [see, for example, Yamashita et al. (1995) J. Cell Biol. 130:217-226; Lee and McPherron (2001) Proc. Natl. Acad. Sci. USA 98:9306-9311; Yeo and Whitman (2001) Mol. Cell 7:949-957; and Oh et al. (2002) Genes Dev. 16:2749-54]. ALK4 is the primary type I receptor for activins, particularly activin A, and ALK-7 may also serve as a receptor for other activins, particularly activin B.

Activins are dimeric polypeptide growth factors that belong to the TGF-beta superfamily. There are three main forms of activin (A, B and AB), which are homo/heterodimers of two closely related β subunits (βАв _А , eBeB and β _А β _B , respectively). The human genome also encodes activin C and activin E, which are predominantly expressed in the liver, and heterodimeric forms containing BC or β _& are also known

Within the TGF-beta superfamily, activins are unique and multifunctional factors that can stimulate hormone production in ovarian and placental cells, support neuronal survival, have positive or negative effects on cell cycle progression depending on the cell type, and induce mesoderm differentiation at least in amphibian embryos [DePaolo et al. (1991) Proc Soc Ep Biol Med. 198:500-512; Dyson et al.

- 17 043914 (1997) Curr Biol. 7:81-84; and Woodruff (1998) Biochem Pharmacol. 55:953-963]. In addition, erythroid differentiation factor (EDF) isolated from stimulated monocyte-like cells from human acute monocytic leukemia was found to be identical to activin A [Murata et al. (1988) PNAS, 85:2434]. It has been suggested that activin A promotes erythropoiesis in the bone marrow. In several tissue types, the antagonist of the activin signaling pathway is its heterodimer, inhibin. For example, when follicle-stimulating hormone (FSH) is released by the pituitary gland, activin promotes FSH secretion and synthesis, whereas inhibin prevents FSH secretion and synthesis. Other proteins capable of regulating the biological activity of activin and/or binding to activin include follistatin (FS), follistatin-like protein (FSRP, also known as FLRG or FSTL3) and _α2 -macroglobulin.

As used herein, agents that bind to activin A are agents that specifically bind to the e _A subunit, either an isolated e _A subunit or as part of a dimeric complex (eg, a homodimer in _A in _A or a heterodimer in Abb). In the case of a heterodimeric complex (e.g., a β _A β _B heterodimer), agents that bind to activin A are specific for epitopes present in the e _A subunit but do not bind to epitopes present in a non-subunit of the complex e _A (for example, a subunit of the e _B complex). Likewise, agents that antagonize (inhibit) activin A are agents that inhibit one or more activities mediated by the _eA subunit, either as an isolated _eA subunit or as part of a dimeric complex (eg, a bAbA homodimer or heterodimer β _A β _B ). In the case of β _A β _B heterodimers, agents that inhibit activin A are agents that specifically inhibit one or more activities of the e _A subunit but do not inhibit the activity of a subunit of the complex that is not the e _A subunit (e.g., the e _B subunit of the complex ). This principle also applies to agents that bind to and/or inhibit activin B, activin C and activin E. Agents disclosed herein that are activin AB antagonists are agents that inhibit one or more than one subunit-mediated activity e _B .

Growth and differentiation factor -8 (GDF8) is also known as myostatin. GDF8 is a negative regulator of skeletal muscle mass. GDF8 is highly expressed in developing and adult skeletal muscle. The GDF8 null mutation in transgenic mice is characterized by severe hypertrophy and hyperplasia of skeletal muscle [McPherron et al., Nature (1997) 387:83-90]. A similar increase in skeletal muscle mass is evident with naturally occurring mutations in cattle [see, for example, Ashmore et al. (1974) Growth, 38:501-507; Swatland and Kieffer (1994) J. Anim. Sci. 38:752-757; McPherron and Lee (1997) Proc. Natl. Acad. Sci. USA 94:12457-12461; and Kambadur et al. (1997) Genome Res. 7:910-915] and, strikingly, in humans [see, for example, Schuelke et al. (2004) N Engl J Med 350:2682-8]. Studies have also shown that loss of muscle mass associated with HIV infection in humans is accompanied by an increase in GDF8 protein expression [see, for example, Gonzalez-Cadavid et al. (1998) PNAS 95:14938-43]. In addition, GDF8 can regulate the production of muscle-specific enzymes (eg, creatine kinase) and regulate myoblast proliferation (see, for example, international patent application WO 00/43781]. The GDF8 propeptide can non-covalently bind to the dimerized domains of mature GDF8, inactivating its biological activity [See, for example, Miyazono et al. (1988) J. Biol. Chem., 263: 6407-6415; Wakefield et al. (1988) J. Biol. Chem., 263: 7646-7654; and Brown et al. (1990) Growth Factors, 3: 35-43] Other proteins that bind to GDF8 or structurally related proteins and inhibit their biological activity include follistatin and possibly follistatin-like proteins [see, for example, Gamer et al. (1999) Dev. Biol., 208: 222-232].

Growth and differentiation factor-11 (GDF11), also known as BMP11, is a secreted protein [McPherron et al. (1999) Nat. Genet. 22. 260-264]. GDF11 is expressed in the tail bud, limb bud, maxillary and mandibular arches and dorsal root ganglia during mouse ontogeny [see, for example, Nakashima et al. (1999) Mech. Dev. 80: 185-189]. GDF11 plays a unique role in the formation of mesoderm and neural tissue [see, for example, Gamer et al. (1999) Dev Biol., 208:222-32]. GDF11 has been shown to be a negative regulator of chondrogenesis and myogenesis during chick wing development [see, for example, Gamer et al. (2001) Dev Biol. 229:407-20]. Expression of GDF11 in muscle also suggests a role in regulating muscle growth similar to GDF8. In addition, the expression of GDF11 in the brain suggests that GDF11 may also have activities that are relevant to nervous system function. Interestingly, GDF11 inhibits neurogenesis in the olfactory epithelium [see, for example, Wu et al. (2003) Neuron. 37:197-207].

In part, this description relates to the observation that an ActRIIB antagonist (inhibitor) can be used to treat patients with myelofibrosis, in particular, improve various complications of the disease, including, for example, splenomegaly, extramedullary hematopoiesis and fibrosis. In particular, the data presented indicate that the GDF decoy polypeptide reduces splenomegaly, extramedullary hematopoiesis, and fibrosis in the JAK2V617F model of myelofibrosis. Accordingly, at some point

- 18 043914 In certain aspects, the description relates to compositions and methods of treating myelofibrosis, in particular, treating or preventing one or more complications of myelofibrosis (for example, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis), by administering to a patient in need thereof an effective amount one or more than one ActRIIB antagonist, optionally in combination with one or more maintenance treatments or other active agent for the treatment of myelofibrosis.

The terms used in this description generally have their standard meaning in the art, in the context of this invention and in the particular context in which each term is used. Certain terms are discussed below or further in the description and provide additional guidance to the practitioner in describing the compositions and methods of the invention, their preparation and use. The scope or meaning of any term used will be apparent from the particular context in which the term is applied.

Homologous and its grammatical forms and spellings refer to the interaction between two proteins that share a common evolutionary origin, including superfamily proteins in organisms of the same biological species, as well as homologous proteins from organisms of different species. Such proteins (and the nucleic acids encoding them) have homologous sequences, as evidenced by the similarity of their sequences, both in terms of percentage identity and the presence of specific residues or motifs and conserved amino acid positions. In common usage and as used herein, the term homologous with the adverb such as highly may refer to sequence similarity and may or may not refer to common evolutionary origin.

The term sequence similarity and all its grammatical forms refers to the degree of identity or correspondence between nucleic acid or amino acid sequences that may or may not share a common evolutionary origin.

Percentage (%) identity to a reference polypeptide (or nucleotide) sequence is defined as the percentage of amino acid residues (or nucleic acids) in a candidate sequence that are identical to amino acid residues (or nucleic acids) in the reference polypeptide (nucleotide) sequence after sequence alignment and gapping ( gaps) if necessary to achieve the maximum percentage of sequence identity, and without accepting any conservative substitutions as part of an identical sequence. Alignment to determine percent amino acid sequence identity can be performed by various methods known in the art, for example using publicly available computer programs such as BLAST, BLAST-2, ALIGN or Megalign (DNASTAR). Those skilled in the art can determine the appropriate parameters for sequence alignment, including any algorithms necessary to achieve maximum alignment over the entire length of the sequences being compared. However, in this description, the % amino acid (nucleic acid) sequence identity values are obtained using the ALIGN-2 sequence comparison computer program. The ALIGN-2 sequence comparison computer program was developed by Genentech, Inc., and the source code and user documentation have been filed with the United States Copyright Office, Washington D.C. 20559, under copyright number TXU510087. The ALIGN-2 program is available from Genentech, Inc., South San Francisco, California, or the program code can be compiled from the source code. The ALIGN-2 program must be compiled to run on the UNIX platform, including the digital UNIX platform. V4.0D. All parameters for sequence comparison are specified by the ALIGN2 program and remain unchanged.

To be an agonist and all grammatical forms of this expression refer to the process of activating a protein and/or gene (for example, by activating or enhancing the expression of a gene for that protein or by inducing the transition of an inactive protein to an active state) or increasing the activity of a protein and/or gene.

To be an antagonist and all grammatical forms of this expression refer to the process of inhibiting a protein and/or gene (for example, by inhibiting or reducing gene expression of that protein or by inducing the transition of an active protein to an inactive state) or reducing the activity of a protein and/or gene.

The terms approximately and approximate, as used in the specification and claims in relation to numerical values, denote precision within the meaning of one skilled in the art. Typically this accuracy is ±10%. Alternatively, and particularly in biological systems, the terms approximately and about may mean values that are within the same order of magnitude, preferably not more than a factor of 5, and more preferably not more than a factor of 2.

The ranges of numerical values in this description include values indicating these ranges.

Singular forms include reference to plural forms if from the context,

- 19 043914 in which the term is used does not explicitly state otherwise. The singular terms, one or more than one, and at least one may be used interchangeably herein. In addition, and/or is to be understood herein as a specific description of each of two or more specific features or components, together or separately. Thus, the term and/or as used in an expression such as A and/or B herein includes A and B, A or B, A (alone) and B (alone). Likewise, the term and/or used in an expression such as A, B and/or C herein covers each of: A, B and C; A, B or C; A or C; A or B; B or C; A and C; A and B; B and C; A (separately); B (separately) and C (separately).

As used herein, the term contain, or variations thereof such as contains or containing, should be understood to include the specified integer value or group of integer values, but not to exclude any other integer value or group of integer values.

2. ActRIIB antagonists.

In part, this description relates to the observation that an ActRIIB antagonist (inhibitor) can be used to treat patients with myelofibrosis, in particular, improve various complications of the disease, including, for example, splenomegaly, extramedullary hematopoiesis and fibrosis. Specifically, the data presented here suggest that the GDF decoy polypeptide reduces splenomegaly, extramedullary hematopoiesis, and fibrosis in the JAK2V617F model of myelofibrosis. Although the effect of soluble GDF decoy polypeptides on myelofibrosis may be mediated by a mechanism other than ActRIIB antagonism [e.g., inhibition of one or more of: GDF11, GDF8, activin B, BMP6, GDF3, and BMP10 may be an indicator of the propensity of the agent to inhibit the activities of a number of additional agents , including possibly other members of the TGF-beta superfamily, and such collective inhibition may result in desired effects in, for example, myelofibrosis], however, the disclosure demonstrates that desired therapeutic agents can be selected for ActRIIB antagonism. Thus, without wishing to be bound by any particular mechanism of action, other ActRIIB antagonists [e.g., ActRIIB receptor antagonists, antagonists of one or more ActRIIB-binding ligands (e.g., GDF11, GDF8, activin, BMP6, GDF3) are expected to and BMP10), antagonists of one or more type I receptors (e.g. ALK4, ALK5 and/or ALK7), antagonists of one or more co-receptors, antagonists of one or more downstream components of the ActRIIB signaling pathway (e.g. Smad)], or combinations of such antagonists] may be useful in the treatment of myelofibrosis, in particular in the treatment or prevention of various complications associated with myelofibrosis (eg, splenomegaly, extramedullary hematopoiesis and fibrosis). Such agents are collectively referred to herein as ActRIIB antagonists or ActRIIB inhibitors.

A. ActRIIB polypeptides and variants thereof.

In some aspects, this disclosure relates to ActRIIB polypeptides and variants thereof (eg, GDF traps). In particular, the description provides methods of using ActRIIB polypeptides alone or in combination with one or more additional supportive therapies for the treatment of myelofibrosis, in particular, the treatment or prevention of one or more than one complications of myelofibrosis (for example, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis) and/or treatment of a patient treated with a Janus kinase inhibitor. As used herein, the term ActRIIB refers to proteins of the activin receptor type IIB (ActRIIB) family of any biological species and variants that are derived from such ActRIIB proteins by mutagenesis or other modification. Reference to ActRIIB in this specification should be understood to include any of the currently identified forms. Members of the ActRIIB family of receptors are typically transmembrane proteins consisting of a ligand-binding extracellular domain with a cysteine-rich region, a transmembrane domain, and a cytoplasmic domain with predicted serine/threonine kinase activity.

The term ActRIIB polypeptide includes polypeptides containing any polypeptide of a naturally occurring member of the ActRIIB family, as well as any variants thereof (including mutants, fragments, fusion proteins and peptidomimetics) that retain beneficial activity. Examples of such variant ActRIIB polypeptides are provided herein as well as in international patent applications WO 2006/012627 and WO 2008/097541, the contents of which are incorporated by reference in their entirety. The amino acid numbering for all ActRIIB-related polypeptides described herein is based on the sequence numbering of the human ActRIIB precursor protein given below (SEQ ID NO: 1), unless otherwise explicitly stated.

Below is the sequence of the human ActRIIB precursor protein.

- 20 043914

MTAPWVALAL LWGSLCAGSG RGEAETRECI YYNANWELER THQSGLERCE

GEQDKRLHCY ASWRgSSGTI ELVKKGCWLD DFNCYDRQEC VATEENPQVY

101 FCCCEGNFCN ERFTHLPEAG GPEVTYEPPP TAPTLLTVLA YSLLPIGGLS

151 LIVLLAFWMY RHRKPPYGHV DIHEDPGPPP PSPLVGLKPL QLLEIKARGR

201 FGCVWKAQLM NDFVAVKIFP LQDKQSWQSE REIFSTPGMK HENLLQFIAA

251 EKRGSNLEVE LWLITAFHDK GSLTDYLKGN IITWNELCHV AETMSRGLSY

301 LHEDVPWCRG EGHKPSIAHR DFKSKNVLLK SDLTAVLADF GLAVRFEPGK

351 PPGDTHGQVG TRRYMAPEVL EGAINFQRDA FLRIDMYAMG LVLWELVSRC

401 KAADGPVDEY MLPFEEEIGQ HPSLEELQEV WHKKMRPTI KDHWLKHPGL

451 AQLCVTIEEC WDHDAEARLS AGCVEERVSL IRRSVNGTTS DCLVSLVTSV

501 TNVDLPPKES SI (SEQIDNO:1).

The signal peptide is underlined with a single line, the extracellular domain is indicated in bold, and potential endogenous N-glycosylation sites are underlined with a double line.

Below is the sequence of the extracellular ActRIIB polypeptide that has undergone processing (mature) polypeptide:

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVKKGCWLDDFN

CYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEVTYEPPPTAPT (SEQ Ш NO: 2).

In some embodiments, the protein may be prepared with the sequence SGR... at the N-terminus. The C-terminal tail of the extracellular domain is underlined with a single line. Below is the sequence with the tail deletion (sequence A15):

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVKKGCWLDDFN

CYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEA (SEQ Ш NO: 3).

The form of ActRIIB with an alanine at position 64 of SEQ ID NO: 1 (A64) is also described in the literature [Hilden et al. (1994) Blood, 83(8): 2163-2170]. Applicants have demonstrated that an ActRIIB-Fc fusion protein containing the extracellular domain of ActRIIB with an A64 substitution has relatively low affinity for activin and GDF11. In contrast, the same ActRIIB-Fc fusion protein with arginine at position 64 (R64) has low nanomolar to high picomolar affinity for activin and GDF11. Thus, in this document, sequences from R64 were used as wild-type reference sequences for human ActRIIB.

The form of ActRIIB with alanine at position 64 is shown below.

MTAPWVALAL LWGSLCAGSG RGEAETRECI YYNANWELER TNQSGLERCE

GEQDKRLHCY ASWANSSGTI ELVKKGCWLD DFNCYDRQEC VATEENPQVY

101 FCCCEGNFCN ERFTHLPEAG GPEVTYEPPP TAPTLLTVLA YSLLPIGGLS

151 LIVLLAFWMY RHRKPPYGHV DIHEDPGPPP PSPLVGLKPL QLLEIKARGR

201 FGCVWKAQLM NDFVAVKIFP LQDKQSWQSE REIFSTPGMK HENLLQFIAA

251 EKRGSNLEVE LWLITAFHDK GSLTDYLKGN IITWNELCHV AETMSRGLSY

301 LHEDVPWCRG EGHKPSIAHR DFKSKNVLLK SDLTAVLADF GLAVRFEPGK

351 PPGDTHGQVG TRRYMAPEVL EGAINFQRDA FLRIDMYAMG LVLWELVSRC

401 KAADGPVDEY MLPFEEEIGQ HPSLEELQEV WHKKMRPTI KDHWLKHPGL

451 AQLCVTIEEC WDHDAEARLS AGCVEERVSL IRRSVNGTTS DCLVSLVTSV

501 TNVDLPPKES SI (SEQ ID NO: 4).

The signal peptide is underlined with a single line, and the extracellular domain is highlighted in bold.

The sequence of the extracellular processed (mature) ActRIIB polypeptide of the alternative form with A64 is given below:

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWANSSGTIELVKKGCWLDDFN

CYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEVTYEPPPTAPT (SEQ ID NO: 5).

In some embodiments, the protein may be prepared with an SGR sequence at the N-terminus. The terminal tail of the extracellular domain is underlined with a single line. Below is the sequence with the tail deletion (sequence A15):

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWANSSGTIELVKKGCWLDDFN

CYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEA (SEQIDNO: 6)

The nucleic acid sequence encoding the human ActRIIB precursor protein is shown below (SEQ ID NO: 7) and consists of nucleotides 25-1560 of the Genbank reference sequence NM_001106.3, encoding amino acids 1-513 of the ActRIIB precursor. In the above sequence, arginine is located at position 64 and can be replaced by alanine. The signal sequence is underlined.

- 21 043914

1 ATGACGGCGC CCTGGGTGGC CCTCGCCCTC CTCTGGGGAT CGCTGTGCGC 51 CGGCTCTGGG CGTGGGGAGG STGAGACACG GGAGTGCATC TACTACAACG 101 CCAACTGGGA GCTGGAGCGC ACCAACCAGA GCGGCCTGGA GCGCTGCGAA 151 GGCGAGCAGG ACAAGCGGCT GCACTGCTAC GCCTCCTGGC GCAACAGCTC 201 TGGCACCATTC GAGCTCGTGA AGAAGGGCTG CTGGCTAGAT GACTTCAACT 251 GCTACGATAG GCAGGAGTGT GTGGCCACTG AGGAGAACCC CCAGGTGTAC 301 TTCTGCTGCT GTGAAGGCAA CTTCTGCAAC GAACGCTTCA CTCATTTGCC 351 AGAGGCTGGG GGCCCGGAAG TCACGTACGA GCCACCCCCG ACAGCCCCCA 401 CCCTGCTCAC GGTGCTGGCC TACTCASTGC TGCCCATCGG GGGCCTTTCC 451 CTCATCGTCC TGCTGGCCTT TTGGATTGTAC CGGCATCGCA AGCCCCCCTA 501 CGGTCATGTG GACATCCATG AGGACCCTGG GCCTCCACCA SSATSSSSTS 551 TGGTGGGCCT GAAGCCACTG CAGCTGCTGG AGATCAAGGC TCGGGGGCGC 601 TTTGGCTGTG TCTGGAAGGC CCAGCTCATG AATGACTTTG TAGCTGTCAA 651 GATCTTCCCA CTCCAGGACA AGCAGTCGTG GCAGAGTGAA CGGGAGATCT 701 TCAGCACACC TGGCATGAAG CACGAGAACC TGCTACAGTT CATTGCTGCC 751 GAGAAGCGAG GCTCCAACCT CGAAGTAGAG CTGTGGCTCA TCACGGCCTT 801 CCATGACAAG GGCTCCCTCA CGGATTACCT CAAGGGGAAC ATCATCASAT 851 GGAACGAACT GTGTCATGTA GCAGAGACGA TGTCACGAGG SSTSTSATAS 901 CTGCATGAGG ATGTGCCCTG GTGCCGTGGC GAGGGCCACA AGCCGTCTAT 951 TGCCCACAGG GACTTTAAAA GTAAGAATGT ATTGCTGAAG AGCGACCTCA 1001 CAGCCGTGCT GGCTGACTTT GGCTTGGCTG TTCGATTTGA GCCAGGGAAA 1051 CCTCCAGGGG ACACCCACGG ACAGGTAGGC ACGAGACGGT ACATGGCTCC 1101 TGAGGTGCTC GAGGGAGCCA TSAASTTSSA GAGAGATGCC TTCCTGCGCA 1151 TTGACATGTA TGCCATGGGG TTGGTGCTGT GGGAGCTTGT GTCTCGCTGC 1201 AAGGCTGCAG ACGGACCGT GGATGAGTAC ATGCTGCCCT TTGAGGAAGA 1251 GATTGGCCAG CACCCTTCGT TGGAGGAGCT GCAGGAGGTG GTGGTGCACA 1301 AGAAGATGAG GCCCACCAT AAAGATCACT GGTTGAAACA CCCGGGCCTG 1351 GCCCAGCTTT GTGTGACCAT CGAGGAGTGC TGGGACCATG ATGCAGAGGC 1401 TCGCTTGTCC GCGGGCTGTG TGGAGGAGCG GGTGTCCCTG ATTCGGAGGT 1451 CGGTCAACG CACTACCTCG GACTGTCTCG TTTCCCTGGT GACCTCTGTC 1501 ACCAATGTGG ACCTGCCCC TAAAGAGTCA AGCATC (SEQ Ш NO: 7).

The following is the nucleotide sequence encoding the processed extracellular human ActRIIB polypeptide (SEQ ID NO: 8):

GGGCGTGGGG AGGCTGAGAC ACGGGAGTGC ATCTACTACA ACGCCAACTG

GGAGCTGGAG CGCACCAACC AGAGCGGCCT GGAGCGCTGC GAAGGCGAGC

101 AGGACAAGCG GCTGCACTGC TACGCCTCCT GGCGCAACAG CTCTGGCACC

151 ATCGAGCTCG TGAAGAAGGG CTGCTGGCTA GATGACTTCA ACTGCTACGA

201 TAGGCAGGAG TGTGTGGCCA CTGAGGAGAA CCCCCAGGTG TACTTCTGCT

251 GCTGTGAAGG CAACTTCTGC AACGAACGCT TCACTCATTT GCCAGAGGCT

301 GGGGGCCCGG AAGTCACGTA CGAGCCACCC CCGACAGCCC CCACC (SEQ ID NO:8)

In the above sequence, arginine is located at position 64 and can be replaced by alanine.

An amino acid sequence alignment of the extracellular domain of human ActRIIB and the extracellular domain of human ActRIIA is shown in FIG. 1. This alignment shows the amino acid residues in both receptors that are believed to directly contact ActRII ligands. For example, ActRII group structures indicated that the ActRIIB ligand-binding pocket is defined, in part, by residues Y31, N33, N35, L38 - T41, E47, E50, Q53 - K55, L57, H58, Y60, S62, K74, W78 - N83, Y85 , R87, A92 and E94 - F101. It is assumed that conservative mutations will be tolerated at these positions.

In addition, vertebrate ActRIIB is generally conserved, with large regions of the extracellular domain being completely conserved. For example, in FIG. Figure 2 shows a multiple sequence alignment of the extracellular domain of human ActRIIB compared to various ActRIIB orthologues. Many of the ligands that bind to ActRIIB are also highly conserved. Accordingly, based on these alignments, it is possible to predict key amino acid positions within the ligand-binding domain that are important for the normal ligand-binding activity of ActRIIB, as well as predict amino acid positions at which substitutions may be tolerated without significantly altering the normal ligand-binding activity of ActRIIB. Thus, an active variant of a human ActRIIB polypeptide that may find use in the present methods may include one or more amino acids at corresponding positions in the ActRIIB sequence of another vertebrate, or may include a residue similar to that of human or other vertebrate sequences.

While not exhaustive, the examples provided illustrate this approach for identifying the active ActRIIB variant. L46 in the human extracellular domain (SEQ ID NO: 2) represents

- 22 043914 is a valine in ActRIIB Xenopus (SEQ ID NO: 42), and therefore at this position a substitution can be made and possibly a substitution with another hydrophobic residue such as V, I or F, or with a non-polar residue such as A E52 in the human extracellular domain represents K in Xenopus, indicating that various changes are allowed at this site, including polar residues such as E, D, K, R, H, S, T, P, G, Y, and possibly , A. T93 in the human extracellular domain represents K in Xenopus, indicating that much structural variability is tolerated at this position, with polar residues favored such as S, K, R, E, D, H, G, P, G and Y. F108 in the human extracellular domain is Y in Xenopus, and therefore Y or another hydrophobic group such as I, V or L should be allowed. E111 in the human extracellular domain represents K in Xenopus, indicating that charged residues including D, R, K and H, as well as Q and N, will be tolerated at this position. R112 in the human extracellular domain represents K in Xenopus, indicating that basic residues including R and H will be tolerated at this position. A at position 119 in the human extracellular domain is relatively weakly conserved and represents P in rodents (SEQ ID NOs: 37 and 39) and V in Xenopus, so at this position it should Substantially any amino acid may be acceptable.

In addition, the structural/functional characteristics of ActRII proteins are described in the art, particularly with regard to binding to ligands [Attisano et al. (1992) Cell 68(1):97–108; Greenwald et al. (1999) Nature Structural Biology 6(1): 18-22; Allendorph et al. (2006) PNAS 103(20: 76437648; Thompson et al. (2003) The EMBO Journal 22(7): 1555-1566; and US patents 7709605, 7612041 and 7842663] In addition to this description, the documents cited provide detailed information to create ActRII variants that retain one or more of the required activities (eg, ligand binding activity).

For example, for ligand binding by type I and type II receptors, it is important to identify a structural motif known as the three-finger toxin fold, formed by conserved cysteine residues located at different positions within the extracellular domain of each monomeric receptor [Greenwald et al. (1999) Nat Struct Biol 6:18–22; and Hinck (2012) FEBS Lett 586:1860-1870]. Accordingly, the major ligand-binding domains of human ActRIIB, defined by the conserved cysteines closest to the outer surface, correspond to positions 29-109 of SEQ ID NO: 1 (ActRIIB precursor). Thus, structurally less ordered amino acids flanking these cysteine-limited key sequences can be truncated by approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15. 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 residues N-terminal and/or approximately 1, 2, 3, 4, 5, 6, 7, 8 , 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or 25 residues C-terminal, which will not necessarily change ligand binding. Examples of ActRIIB extracellular domains for N-terminal and/or C-terminal truncation include SEQ ID NOs: 2, 3, 5 and 6.

Attisano et al. showed that deletion of the proline node at the C-terminal end of the extracellular domain of ActRIIB reduced the receptor's affinity for activin. The ActRIIB-Fc fusion protein containing amino acids 20 ^-11 9 of the given sequence SEQ ID NO: 1, ActRIIB(20-119)-Fc, binds to GDF11 and activin more weakly than ActRIIB(20-134)-Fc, which includes the region proline knot and a complete juxtamembrane domain (see, for example, US Pat. No. 7,842,663). However, the ActRIIB(20-129)-Fc protein retains similar, but slightly reduced activity compared to the wild-type protein, even when the proline knot region is destroyed. Thus, all extracellular domains of ActRIIB that end at amino acids 134, 133, 132, 131, 130 and 129 (relative to SEQ ID NO: 1) are predicted to have activity, but constructs ending at amino acids 134 or 133 will have the greatest activity . Likewise, mutations to any of residues 129-134 (relative to SEQ ID NO: 1) are not expected to significantly alter ligand binding affinity. In support of this assumption, there is evidence in the art that mutations P129 and P130 (relative to SEQ ID NO: 1) do not significantly reduce ligand binding. Thus, the ActRIIB polypeptide herein may end as early as amino acid 109 (the final cysteine), but forms ending at or between 109 and 119 (e.g., 109, 110, 111, 112, 113, 114, 115, 116, 117, 118 or 119) are expected to bind the ligand more weakly. Amino acid 119 (relative to SEQ ID NO: 1) is weakly conserved and therefore can be easily replaced or trimmed. ActRIIB polypeptides ending at amino acid 128 (relative to SEQ ID NO: 1) or beyond should retain ligand binding activity. ActRIIB polypeptides that end on or between amino acids 119 and 127 (eg, 119, 120, 121, 122, 123, 124, 125, 126, or 127) relative to SEQ ID NO: 1 will have intermediate binding ability. Any of these forms may be suitable for use, depending on the clinical or experimental conditions.

An ActRIIB protein starting at the N-terminus at amino acid 29 or earlier (relative to SEQ ID NO: 1) is predicted to retain ligand-binding activity. Amino acid 29 is the first cysteine. Mutation at position 24 (relative to SEQ ID NO: 1) from alanine to asparagine creates an N-glycosylation sequence without significantly affecting ligand binding [US Patent 7,842,663]. This confirms that mutations in the region between the cleavage signal

- 23 043914 peptide and cysteine bridged region corresponding to amino acids 20-29 are acceptable. In particular, ActRIIB polypeptides starting at positions 20, 21, 22, 23 and 24 (relative to SEQ ID NO: 1) should retain overall ligand binding activity, and ActRIIB polypeptides starting at positions 25, 26, 27, 28 and 29 (relative to SEQ ID NO: 1) are also believed to retain ligand binding activity. For example, US Pat. No. 7,842,663 showed that ActRIIB constructs starting with amino acids 22, 23, 24, or 25 were surprisingly most active.

Collectively, the general formula of the activity portion of ActRIIB (e.g., the ligand binding portion) contains amino acids 29-109 of the sequence SEQ ID NO: 1. Thus, ActRIIB polypeptides may, for example, contain, essentially consist of, or consist of the amino acid sequence, which is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the portion of ActRIIB beginning with a residue corresponding to any of amino acids 20-29 (e.g., beginning with any of amino acids 20, 21, 22, 23, 24, 25, 26, 27, 28, or 29) the sequence SEQ ID NO: 1 and ending at a position corresponding to any of amino acids 109-134 (for example, ending with any of amino acids 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120 , 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, or 134) SEQ ID NO: 1. Other examples include polypeptides that begin at positions 20-29 (e.g. in any of positions 20, 21, 22, 23, 24, 25, 26, 27, 28 or 29) or 21-29 (for example, in any of positions 21, 22, 23, 24, 25, 26, 27, 28 or 29) and end at positions 119-134 (for example, at any of positions 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133 or 134), 119-133 (for example, in any of provisions 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132 or 133), 129-134 (for example, in any of positions 129, 130, 131, 132, 133, or 134) or 129-133 (e.g., at any of positions 129, 130, 131, 132, or 133) SEQ ID NO: 1. Other examples include polypeptides that begin at positions 20 -24 (for example, in any of positions 20, 21, 22, 23 or 24), 21-24 (for example, in any of positions 21, 22, 23 or 24) or 22-25 (for example, in any of positions 22 , 22, 23 or 25) and end at positions 109-134 (for example, at any of positions 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123 , 124, 125, 126, 127, 128, 129, 130, 131, 132, 133 or 134), 119-134 (for example, in any of the provisions 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133 or 134) or 129-134 (for example, in any of provisions 129, 130, 131, 132, 133 or 134) SEQ ID NO: 1. Variants with boundaries as specified ranges are also within the scope of the invention, in particular those that are at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92% identical , 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% of the corresponding portion of SEQ ID NO: 1.

The options described herein can be combined in various ways. In some embodiments, ActRIIB variants contain no more than 1, 2, 5, 6, 7, 8, 9, 10, or 15 conservative amino acid substitutions in the ligand binding pocket and zero, one or more non-conservative modifications at positions 40, 53, 55, 74, 79 and/or 82 in the ligand binding pocket. Sites outside the binding pocket where variability may be most tolerated include the amino and carboxy termini of the extracellular domain (as noted above) and positions 42-46 and 65-73 (relative to SEQ ID NO: 1). The substitution of asparagine for alanine at position 65 (N65A) actually improves ligand binding by the A64-carrying form and is therefore not believed to have a negative effect on ligand binding by the R64-carrying form [US Patent 7,842,663]. This change probably eliminates glycosylation at N65 of the A64-carrying form, thus demonstrating that significant change in this region is likely to be tolerated. While the R64A substitution does not hold up well, R64K does well and therefore another basic residue such as H can be tolerated at position 64 [US Patent 7,842,663]. In addition, mutagenesis results described in the art indicate that there are amino acid positions in ActRIIB that are often advantageous to retain. With respect to SEQ ID NO: 1, these positions include position 80 (negatively charged or hydrophobic amino acid), position 78 (hydrophobic amino acid and, in particular, tryptophan), position 37 (negatively charged amino acid and, in particular, aspartate or glutamic acid), position 56 (positively charged amino acid), position 60 (hydrophobic amino acid, particularly phenylalanine or tyrosine). Thus, this description provides a framework of amino acids that can be stored in ActRIIB polypeptides. Other positions that may be desirable to retain are: position 52 (negatively charged amino acid), position 55 (positively charged amino acid), position 81 (negatively charged amino acid), 98 (polar or charged amino acid, particularly E, D, R or K), all relative to SEQ ID NO: 1.

It has previously been demonstrated that the addition of an additional N-glycosylation site (NX-S/T) to the extracellular domain of ActRIIB is permissible (see, for example, US Pat. No. 7,842,663). Thus, N-X-S/T sequences can generally be inserted into positions outside the ligand binding pocket of the ActRIIB polypeptide herein, indicated in FIG. 1. The most suitable sites for insertion of non-endogenous N-X-S/T sequences are amino acids 20-29, 20-24, 22-25, 109-134, 120-134 or 129-134 (relative to SEQ ID NO: 1). Follower

- 24 043914 N-X-S/T properties can also be introduced into the linker between the ActRIIB sequence and the Fc domain or other component of the fusion protein. Such a site can be inserted with minimal effort by inserting an N at the appropriate position relative to a preexisting S or T, or by inserting an S or T at a position corresponding to a preexisting N. Thus, desirable substitutions capable of producing an N-glycosylation site are: A24N, R64N, S67N (possibly in combination with modification N65A), E105N, R112N, G120N, E123N, P129N, A132N, R112S and R112T (relative to SEQ ID NO: 1).

Any S that is predicted to be glycosylated can be changed to a T without creating an immunogenic site due to the protection provided by glycosylation. Likewise, any T that is predicted to be glycosylated can be changed to S. Thus, the modifications S67T and S44T (relative to SEQ ID NO: 1) are contemplated. Similarly, for the A24N variant you can use the S26T modification. Accordingly, the ActRIIB polypeptide herein may be a variant having one or more than one additional non-endogenous N-glycosylation consensus sequence as described above.

In some embodiments, the description relates to ActRIIB antagonists that contain at least one ActRIIB polypeptide, which includes fragments, functional variants, and modified forms thereof. Preferably, ActRIIB polypeptides for use in this invention are soluble (eg, ActRIIB extracellular domain). In some embodiments, ActRIIB polypeptides for use herein inhibit (antagonize) the activity (e.g., Smad signaling) of one or more TGF-beta superfamily ligands [e.g., GDF11, GDF8, activin (activin A, activin B, activin AB , activin C, activin E) BMP6, GDF3 and/or BMP10. In some embodiments, ActRIIB polypeptides for use in this invention bind to one or more TGF-beta superfamily ligands [e.g., GDF11, GDF8, activin (activin A, activin B, activin AB, activin C, activin E) BMP6, GDF3, BMP10 and/or BMP9. In some embodiments, the ActRIIB polypeptide of this invention contains, consists essentially of, or consists of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90% , 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the portion of ActRIIB beginning with the residue corresponding to amino acids 20-29 (e.g., beginning with any of amino acids 20 , 21, 22, 23, 24, 25, 26, 27, 28, or 29) of the sequence SEQ ID NO: 1 and ending at the position corresponding to amino acids 109-134 (for example, ending with any of amino acids 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, or 134) sequences SEQ ID NO: 1. In some embodiments, the ActRIIB polypeptides of the invention contain, consist essentially of, or consist of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acids 29-109 of SEQ ID NO: 1. In some embodiments, ActRIIB polypeptides the invention contain, consist essentially of, or consist of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92% , 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acids 29-109 of SEQ ID NO: 1, where the position corresponding to L79 in SEQ ID NO: 1 is a negatively charged amino acid (negatively charged naturally occurring amino acids D and E or a negatively charged artificially occurring amino acid). In some preferred embodiments, the ActRIIB polypeptides of the invention comprise, are substantially composed of, or consist of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90 %, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acids 25-131 of SEQ ID NO: 1. In some preferred embodiments, the ActRIIB polypeptides the invention contain, consist essentially of, or consist of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92% , 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to amino acids 25-131 of SEQ ID NO: 1, where the position corresponding to L79 in SEQ ID NO: 1 is negatively charged amino acid. In some embodiments, the ActRIIB polypeptides of the invention contain, consist essentially of, or consist of an amino acid sequence that is at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90% , 91%, 92%, 93%, 94%, 95%, 97%, 98%, 99%, 99%, or 100% identical to the amino acid sequences of any of SEQ ID NO: 1, 2, 3, 4, 5, 6 , 24, 25, 28, 29, 30, 31, 33, 34, 35, 45, 50, 53, 54, and 58. In some embodiments, the ActRIIB polypeptides of the invention comprise, are essentially composed of, or consist of an amino acid sequence that at least 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 97%, 98% , 99%, 99% or 100% identical to the amino acid sequences of any of SEQ ID NO: 1, 2, 3, 4, 5, 6, 24, 25, 28, 29, 30, 31, 33, 34, 35, 45, 50, 53, 54 and 58, wherein the position corresponding to L79 in SEQ ID NO: 1 is a negatively charged amino acid. In some embodiments, the ActRIIB polypeptides of the invention comprise at least one ActRIIB polypeptide wherein the position corresponding to L79 of SEQ ID NO: 1 is not negatively charged

- 25 043914 amino acid (i.e. is not the negatively charged naturally occurring amino acids D and E or the negatively charged amino acid residue of artificial origin).

In some aspects, this disclosure relates to GDF decoy polypeptides (also referred to as GDF decoys). In some embodiments, the GDF traps of this invention are variants of ActRIIB polypeptides that contain one or more mutations (e.g., additions, deletions, amino acid substitutions, and combinations thereof) in the extracellular domain (also referred to as ligand binding domain) of the ActRIIB polypeptide (e.g. , wild-type or unmodified ActRII polypeptide), such that the variant ActRIIB polypeptide has one or more ligand-binding activities altered relative to the wild-type ActRIIB polypeptide. In preferred embodiments, the GDF trap polypeptides herein retain at least one similar activity as the corresponding wild-type ActRIIB polypeptide. For example, preferred GDF traps bind and inhibit (eg, antagonize) the functions of GDF11 and/or GDF8. In some embodiments, the GDF traps herein further bind to and inhibit one or more TGF-beta superfamily ligands. Accordingly, provided herein are GDF trap polypeptides that have altered binding specificity for one or more ActRIIB ligands.

By way of illustration, one or more mutations may be selected to increase the selectivity of the altered ligand-binding domain for GDF11 and/or GDF8 over one or more ActRIIB-binding ligands such as activins (activin A, activin B, activin AB, activin C and/or activin E), in particular activin A. It is possible that the altered ligand binding domain has a ratio of KD for activin binding and KD for GDF11 and/or GDF8 binding of at least 2-, 5-, 10 -, 20-, 50-, 100-, or even 1000-fold higher than that of the wild-type ligand binding domain. It is possible that the altered ligand binding domain has a ratio of IC ₅₀ for activin inhibition to IC ₅₀ for GDF11 and/or GDF8 inhibition of at least 2-fold, 5-fold, 10-fold, 20-fold, 50-fold, 100-fold, or even 1000-fold higher than that for the wild-type ligand-binding domain. It is possible that the altered ligand binding domain inhibits GDF11 and/or GDF8 with an _IC50 of at least 2-, 5-, 10-, 20-, 50-, 100-, or even 1000-fold less than the _IC50 for activin inhibition (for example, activin A).

In some preferred embodiments, the GDF traps herein are designed to preferentially bind GDF11 and/or GDF8 (also known as myostatin). It is possible that decoys that bind GDF11 and/or GDF8 may additionally bind to activin B. It is possible that decoys that bind GDF11 and/or GDF8 may additionally bind to BMP6. It is possible that decoys that bind GDF11 and/or GDF8 may additionally bind to BMP10. It is possible that decoys that bind GDF11 and/or GDF8 may additionally bind to activin B and BMP6. In some embodiments, the GDF traps herein have reduced binding affinity to activins (eg, activin A, activin A/B, activin B, activin C, activin E), for example, compared to wild-type ActRIIB polypeptide. In some preferred embodiments, the GDF decoy polypeptide herein has reduced binding affinity for activin A.

The amino acid residues of the ActRIIB proteins (eg, E39, K55, Y60, K74, W78, L79, D80 and F101) are located in the ligand-binding pocket of ActRIIB and promote indirect binding to its ligands, including, for example, activin A, GDF11 and GDF8. Thus, provided herein are GDF trap polypeptides comprising an altered ligand binding domain (eg, GDF8/GDF11 binding domain) of the ActRIIB receptor that contains one or more mutations of said amino acid residues.

As a specific example, the positively charged amino acid residue Asp (D80) of the ActRIIB ligand binding domain can be mutated to another amino acid residue to produce a GDF decoy polypeptide that preferentially binds to GDF8 but not to activin. Preferably, residue D80 with respect to SEQ ID NO: 1 is replaced by an amino acid residue selected from the group consisting of: an uncharged amino acid residue, a negatively charged amino acid residue and a hydrophobic amino acid residue. As another specific example, hydrophobic residue L79 of SEQ ID NO: 1 can be modified to impart altered activin-GDF11/GDF8 binding properties. For example, the L79P substitution reduces GDF11 binding to a greater extent than activin binding. In contrast, substitution of L79 with a negatively charged amino acid [aspartic acid or glutamic acid; substitution L79D or L79E] greatly reduces the binding affinity of activin A while maintaining the binding affinity of GDF11. In exemplary embodiments, the described methods employ a GDF decoy polypeptide, which is a variant of the ActRIIB polypeptide containing a negatively charged amino acid, e.g., D or E) at a position corresponding to position 79 in SEQ ID NO: 1, possibly in combination with one or more than one additional amino acid substitution, addition or deletion.

In some embodiments, this description provides for the creation of functional variants

- 26 043914 by modifying the structure of the ActRIIB polypeptide for purposes such as increasing therapeutic efficacy or stability (eg, storage time and resistance to proteolytic degradation in vivo). Variants can be obtained by substitution, deletion, addition of amino acids, or combinations thereof. For example, it is reasonable to expect that isolated substitution of isoleucine or valine for leucine, glutamate for aspartate, serine for threonine, or similar amino acid substitution for a structurally related amino acid (eg, conservative mutations) will not have a significant effect on the biological activity of the resulting molecule. Conservative substitutions are substitutions that occur within a family of amino acids that have similar side chains. Whether a substitution in the amino acid sequence of a polypeptide as described will result in a functional homologue can be readily determined by assessing the ability of the variant polypeptide to elicit a cellular response in a manner similar to the wild-type polypeptide or to bind one or more TGF-beta ligands, including, for example, BMP2, BMP2 /7, BMP3, BMP4, BMP4/7, BMP5, BMP6, BMP7, BMP8a, BMP8b, BMP9, BMP10, GDF3, GDF5, GDF6/BMP13, GDF7, GDF8, GDF9b/BMP15, GDF11/BMP11, GDF15/MIC1, TGF -e1, TGF-e2, TGF-e3, activin A, activin B, activin C, activin E, activin AB, activin AC, nodal protein, glial cell-derived neurotrophic factor (GDNF), neuroturin, artemin, persefin, Müllerian inhibitory substance ( MIS) and Lefty protein.

In some embodiments of the present invention, specific mutations of the ActRIIB polypeptide are provided to alter the glycosylation of the polypeptide. Such mutations may be selected to introduce or remove one or more glycosylation sites, such as an Oglycosylation or N-glycosylation site. Asparagine glycosylation sites typically contain a tripeptide sequence, asparagine-X-threonine or asparagine-X-serine (where X is any amino acid), which is specifically recognized by the corresponding cellular glycosylation enzymes. The change can also be made by adding or replacing one or more serine or threonine residues to the polypeptide sequence (for O-glycosylation sites). A series of amino acid substitutions or deletions at one or both positions of the first and third amino acids of the glycosylation site (and/or deletion of an amino acid at the second position) results in a lack of glycosylation of the modified tripeptide sequence. Another way to increase the number of carbohydrate moieties on a polypeptide is to chemically or enzymatically link glycosides to the polypeptide. Depending on the method of attachment used, the carbohydrate(s) may be attached to: (a) arginine and histidine; (b) free carboxyl groups; (c) free sulfhydryl groups, such as cysteine; (d) free hydroxyl groups, such as those of serine, threonine or hydroxyproline; (e) aromatic residues such as those of phenylalanine, tyrosine or tryptophan, or (f) the amide group of glutamine. Removal of one or more carbohydrate moieties present on a polypeptide can be accomplished by chemical and/or enzymatic means. Chemical deglycosylation may include, for example, exposing the polypeptide to a trifluoromethanesulfonic acid compound or an equivalent compound. This action results in the elimination of most or all carbohydrates, with the exception of the binding carbohydrate (N-acetylglucosamine or N-acetylgalactosamine), while the amino acid sequence remains intact. Enzymatic cleavage of carbohydrate moieties on a polypeptide can be achieved using various endo- and exo-glycosidases, such as those described by Thotakura et al. [Meth. Enzymol. (1987) 138:350]. The sequence of the polypeptide can be adjusted as appropriate depending on the type of expression system used, since mammalian, yeast, insect and plant cells can produce different glycosylation patterns, which may depend on the amino acid sequence of the peptide. In general, the polypeptides herein for use in humans can be expressed in a mammalian cell line that ensures proper glycosylation, such as the HEK293 or CHO cell lines, although other mammalian expression cell lines may conceivably be used.

Also provided herein is a method for generating mutants, in particular a number of combinatorial mutants of the ActRIIB polypeptide, as well as truncating mutants. Combinatorial mutant pools are particularly effective at detecting functionally active (eg, TGF-beta superfamily ligand-binding) ActRIIB sequences. The goal of screening such combinatorial libraries may be to generate, for example, polypeptide variants that have altered properties, such as altered pharmacokinetics or altered ligand binding. Below are a variety of screening methods and such tests can be used to evaluate options. For example, ActRIIB variants can be screened for the ability to bind one or more TGF-beta superfamily ligands (e.g., BMP2, BMP2/7, BMP3, BMP4, BMP4/7, BMP5, BMP6, BMP7, BMP8a, BMP8b, BMP9, BMP10, GDF3, GDF5, GDF6/BMP13, GDF7, GDF8, GDF9b/BMP15, GDF11/BMP11, GDF15/MIC1, TGF-e1, TGF-e2, TGF-e3, activin A, activin B, activin AB, activin AC, nodal protein, glial cell-derived neurotrophic factor (GDNF), neuroturin, artemin, persefin, Müllerian inhibitory substance (MIS) and Lefty protein) to prevent super ligand binding

- 27 043914 TGF-beta family with a TGF-beta superfamily receptor and/or to interfere with signal transduction initiated by a TGF-beta superfamily ligand.

The activity of ActRIIB polypeptides can also be studied in cell studies or in vivo studies. For example, the effect of the ActRIIB polypeptide on the expression of genes mediating the severity of myelofibrosis can be assessed. If desired, this can be done in the presence of one or more recombinant ActRII ligand proteins (e.g. BMP2, BMP2/7, BMP3, BMP4, BMP4/7, BMP5, BMP6, BMP7, BMP8a, BMP8b, BMP9, BMP10, GDF3, GDF5, GDF6/BMP13, GDF7, GDF8, GDF9b/BMP15, GDF11/BMP11, GDF15/MIC1, TGF-p1, TGF-e2, TGF-e3, activin A, activin B, activin C, activin E, activin AB, activin AC, nodal protein, glial cell-derived neurotrophic factor (GDNF), neuroturin, artemin, persefin, MIS and Lefty protein), and cells can be transfected to produce the ActRIIB polypeptide and possibly the ActRIIB ligand. Likewise, the ActRIIB polypeptide can be administered to mice or other animals and the effect on myelofibrosis assessed by methods known in the art. Likewise, the activity of an ActRIIB polypeptide or a variant thereof can be tested in hematopoietic progenitor cells to detect any effect on the growth of these cells, for example, in the studies described herein or known in the art. To monitor the effect on subsequent steps in the signaling pathway in such cell lines, a SMAD-dependent reporter gene can be used.

Combinatorial variants can be generated that have increased selectivity or increased overall activity compared to the reference ActRIIB polypeptide. When expressed in recombinant DNA constructs, such variants can be used in gene therapy protocols. Likewise, mutagenesis can give rise to variants whose intracellular half-life is strikingly different from the corresponding unmodified ActRIIB polypeptide. For example, the modified protein can be made more stable or less stable to proteolytic cleavage or other cellular processes that result in the destruction or other inactivation of the unmodified polypeptide. Such variants and the genes encoding them can be used to alter the levels of the polypeptide complex by adjusting the half-life of the polypeptide. For example, a short half-life may result in more transient biological effects and, when part of an inducible expression system, provide finer regulation of the levels of the recombinant polypeptide complex within the cell. To change the half-life of the ActRIIB polypeptide in Fc fusion proteins, mutations may affect the linker (if present) and/or the Fc moiety.

A combinatorial library can be constructed in the form of a degenerate library of genes encoding a library of polypeptides, each of which includes at least a portion of potential ActRIIB sequences. For example, a mixture of synthetic oligonucleotides can be enzymatically ligated to a gene sequence such that a series of degenerate nucleotide sequences potentially encoding ActRIIB are expressed as individual polypeptides, or alternatively, as a series of larger fusion proteins (e.g. for phage display) .

There are many ways to create a library of potential homologs from a degenerate oligonucleotide sequence. Chemical synthesis of a degenerate gene sequence can be performed using an automated DNA synthesizer, and the synthesized genes can then be ligated into an appropriate expression vector. Synthesis of degenerate oligonucleotides is well known in the art [Narang, SA (1983) Tetrahedron 39:3; Itakura et al. (1981) Recombinant DNA, Proc. 3rd Cleveland Sympos. Macromolecules, ed. A.G. Walton, Amsterdam: Elsevier pp273-289; Itakura et al. (1984) Annu. Rev. Biochem. 53:323; Itakura et al. (1984) Science 198:1056; and Ike et al. (1983) Nucleic Acid Res. 11:477]. Such methods have been used to target other proteins [Scott et al., (1990) Science 249:386-390; Roberts et al. (1992) PNAS USA 89:2429-2433; Devlin et al. (1990) Science 249: 404406; Cwirla et al, (1990) PNAS USA 87: 6378-6382; as well as US patents 5223409, 5198346 and 5096815].

Alternatively, other forms of mutagenesis can be used to generate a combinatorial library. For example, ActRIIB polypeptides as described can be generated and isolated from a library by screening, for example using, for example, alanine scanning mutagenesis [Ruf et al. (1994) Biochemistry 33:1565-1572; Wang et al. (1994) J Biol. Chem. 269:3095-3099; Balint et al. (1993) Gene 137:109-118; Grodberg et al. (1993) Eur. J. Biochem. 218:597-601; Nagashima et al. (1993) J. Biol. Chem. 268:2888-2892; Lowman et al. (1991) Biochemistry 30:10832–10838; and Cunningham et al. (1989) Science 244:1081-1085], via linker scanning [Gustin et al. (1993) Virology 193:653-660; and Brown et al. (1992) Mol. Cell Biol. 12:2644-2652; McKnight et al. (1982) Science 232:316], by saturation mutagenesis [Meyers et al., (1986) Science 232:613]; mutagenesis by PCR (polymerase chain reaction) [Leung et al. (1989) Method Cell Mol Biol 1:11-19]; or random mutagenesis, including chemical mutagenesis [Miller et al. (1992) A Short Course in Bacterial Genetics, CSHL Press, Cold Spring Harbor, NY; and Greener et al. (1994) Strategies in Mol Biol 7:32-34]. Linker scanning, particularly in combinatorial approaches, is an attractive way to detect truncated (biologically active) forms of ActRIIB polypeptides.

A wide range of methods are known in the art for screening gene products of combinatorial

- 28 043914 libraries created as a result of point mutations and truncations and even screening of cDNA libraries for gene products that have certain properties. Such methods are typically adapted for rapid screening of gene libraries generated by combinatorial mutagenesis of ActRIIB polypeptides. The most widely used methods for screening large gene libraries typically involve cloning the gene library into replicable expression vectors, transforming appropriate cells with the resulting vector libraries, and expressing combinatorial genes under conditions where detection of the desired activity allows for relatively easy isolation of the vector encoding the gene whose product is being detected. Preferred studies include TGF-beta ligand binding studies (e.g., BMP2, BMP2/7, BMP3, BMP4, BMP4/7, BMP5, BMP6, BMP7, BMP8a, BMP8b, BMP9, BMP10, GDF3, GDF5, GDF6/BMP13, GDF7, GDF8, GDF9b/BMP15, GDF11/BMP11, GDF15/MIC1, TGF-e1, TGF-e2, TGF-e3, activin A, activin B, activin C, activin E, activin AB, activin AC, nodal protein, glial neurotrophic factor cells (GDNF), neuroturin, artemin, persefin, MIS and Lefty protein), and/or studies of TGF-beta ligand-mediated cell signaling.

Those skilled in the art will appreciate that most of the described mutations, variants or modifications described herein can be accomplished at the nucleic acid level or, in some cases, through post-translational modification or chemical synthesis. Such methods are well known in the art and some of them are described herein. In part, this description allows us to identify functionally active parts (fragments) and variants of ActRIIB polypeptides, which can be used as a guide for the creation and use of other variants of ActRIIB polypeptides that are within the scope of the proposed inventions.

In some embodiments, functionally active fragments of ActRIIB polypeptides herein can be obtained by screening polypeptides obtained by recombinant methods from the corresponding nucleic acid fragment encoding the polypeptide (e.g., SEQ ID NOs: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59, 60, 61 and 62). In addition, the fragments can be chemically synthesized by methods well known in the art, such as Merrifield's standard solid phase method using f-Moc or t-Boc protected amino acids. Fragments can be prepared (recombinantly or by chemical synthesis) and screened for those peptidyl fragments that may function as ActRII receptor antagonists and/or one or more ligands (e.g., BMP2, BMP2/7, BMP3, BMP4, BMP4/7, BMP5, BMP6, BMP7, BMP8a, BMP8b, BMP9, BMP10, GDF3, GDF5, GDF6/BMP13, GDF7, GDF8, GDF9b/BMP15, GDF11/BMP11, GDF15/MIC1, TGF-p1, TGFβ2, TGF-e3, activin A, activin B, activin C, activin E, activin AB, activin AC, nodal protein, glial cell-derived neurotrophic factor (GDNF), neuroturin, artemin, persefin, MIS and Lefty protein).

In some embodiments, the ActRIIB polypeptides of this invention may further contain post-translational modifications other than those naturally present in the ActRIIB polypeptide. Such modifications include, but are not limited to, acetylation, carboxylation, glycosylation, phosphorylation, lipidation and acylation. As a result, the ActRIIB polypeptide may contain non-amino acid elements such as polyethylene glycols, lipids, polysaccharides or monosaccharides and phosphates. The effect of such non-amino acid elements on the functionality of the ligand decoy polypeptide can be studied as described herein for other ActRIIB variants. When a polypeptide is said to be produced in cells by cleavage of the synthesized form of the polypeptide, post-translational processing may also be important for proper folding and/or function of the protein. Different cells (eg, CHO, HeLa, MDCK, 293, WI38, NIH-3T3 or HEK293) have specific cellular machinery and characteristic mechanisms for such post-translational activity and can be selected to ensure proper modification and processing of ActRII polypeptides).

In some aspects, ActRIIB polypeptides herein include fusion proteins having at least a portion of an ActRIIB polypeptide and one or more heterologous portions of a domain. Well-known examples of such fusion domains include, but are not limited to, polyhistidine, Glu-Glu, glutathione-S-transferase (GST), thioredoxin, protein A, protein G, immunoglobulin heavy chain constant region (Fc), maltose binding protein (MBP) or human serum albumin. The fusion domain can be selected to impart the desired property. For example, some fusion domains are particularly effective for isolating fusion proteins by affinity chromatography. For affinity purification, appropriate matrices are used, such as resins conjugated with glutathione, amylase, and nickel or cobalt. Many of these matrices are commercially available in kit form, such as Pharmacia's GST-tagged protein purification system and the QIAexpress™ system (Qiagen) for hexahistidine (HIS6) fusion binding partners. As another example, the fusion domain may be selected to facilitate detection of an ActRIIB polypeptide. Examples of such domains to detect include various fluores

- 29 043914 cent proteins (for example, GFP), as well as epitope tags, which are usually short peptide sequences to which there is a specific antibody. Well-known epitope tags for which specific monoclonal antibodies are readily available include the FLAG, influenza virus hemagglutinin (HA) and c-myc epitopes. In some cases, the fusion domains have a protease cleavage site, such as for factor Xa or thrombin, which allows the respective proteases to partially cleave the fusion proteins and thereby release the recombinant proteins. The released proteins can then be isolated from the fusion domains through subsequent chromatographic separation. Other types of fusion domains that may be selected include multimerizing (eg, dimerizing, tetramerizing) domains and functional domains (which confer additional biological function), including, for example, immunoglobulin constant domains (eg, Fc domains).

In some aspects, the ActRIIB polypeptides herein have one or more modifications capable of stabilizing the polypeptides. Stabilization refers to anything that increases the in vitro half-life, serum half-life, whether due to decreased degradation, decreased renal clearance, or other pharmacokinetic effects of the agent. For example, such modifications improve the storage time of the polypeptides, improve the half-life of the polypeptides in circulation, and/or reduce the proteolytic degradation of the polypeptides. Such stabilizing modifications include, but are not limited to, fusion proteins (including, for example, fusion proteins comprising an ActRIIB polypeptide domain and a stabilizing domain), glycosylation site modifications (including, for example, adding a glycosylation site to a polypeptide of the invention), and carbohydrate moiety modifications (including, for example, removal of carbohydrate groups from a polypeptide according to the invention). As used herein, the term stabilizing domain refers not only to the fusion domain (eg, immunoglobulin Fc domain) as in the case of fusion proteins, but also includes non-protein modifications such as a carbohydrate moiety or non-protein moieties such as polyethylene glycol. In some preferred embodiments, the ActRIIB polypeptide is fused to a heterologous domain that stabilizes the polypeptide (stabilizing domain), preferably a heterologous domain that increases the stability of the polypeptide in vivo. Fusion to an immunoglobulin constant domain (eg, Fc domain) is known to impart desirable pharmacokinetic properties to a wide range of proteins. Likewise, fusion with human serum albumin can impart the desired properties.

An example of a native amino acid sequence that can be used for the Fc portion of human IgG1 (G1Fc) is given below (SEQ ID NO: 9). The dotted line indicates the hinge region and the solid line indicates the location with naturally occurring variants. In part of the invention, the proposed polypeptides that contain, essentially consist of, or consist of amino acid sequences that are 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91% , 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identical to SEQ ID NO: 9. Naturally occurring G1Fc variants would include E134D and M136L according to the numbering system used for SEQ ID NO: 9 (See Uniprot P01857).

TNTSRRSRAR ELLGGPSVFL FPPKPKDTLM ISRTPEVTCV WDVSHEDPE

VKFNWYVDGV EVHNAKTKPR EEQYNSTYRV VSVLTVLHQD WLNGKEYKCK

101 VSNKALPAPI EKTISKAKGQ PREPQVYTLP PSREEMTKNQ VSLTCLVKGF

151 YPSDIAVEWE SNGQPENNYK TTPPVLDSDG SFFLYSKLTV DKSRWQQGNV

01 FSCSVMHEAL HNHYTQKSLS LSPGK (SEQ ID NO: 9).

It is possible that the Fc domain of IgG1 has one or more mutated residues, such as Asp265, lysine 322, and Asn-434. In some cases, an IgG1 Fc domain mutant having one or more such mutations (eg, an Asp-265 mutation) has a reduced ability to bind to the Fcy receptor compared to the wild-type Fc domain. In other cases, an IgG1 Fc domain mutant having one or more such mutations (eg, the Asn-434 mutation) has an increased ability to bind to the MHC class I Fc receptor-like proteins (FcRN) compared to the IgG1 Fc domain wild type.

An example of a native amino acid sequence that can be used for the Fc portion of human IgG2 (G2Fc) is given below (SEQ ID NO: 10). The dotted line indicates the hinge region, and the double line indicates positions where inconsistencies exist in the sequence between databases (according to UniProt P01859). Part of the invention provides polypeptides that contain, consist essentially of, or consist of amino acid sequences that are 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91% , 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10.

VECPPCPAPP VAGPSVFLFP PKPKDTLMIS RTPEVTCVW DVSHEDPEVQ

FNWYVDGVEV HNAKTKPREE QFNSTFRWS VLTWHQDWL NGKEYKCKVS

101 NKGLPAPIEK TISKTKGQPR EPQVYTLPPS REEMTKNQVS LTCLVKGFYP

151 SDIAVEWESN GQPENNYKTT PPMLDSDGSF FLYSKLTVDK SRWQQGNVFS

01 CSVMHEALHN HYTQKSLSLS PGK (SEQ ID NO 10)

- 30 043914

Below are two examples of amino acid sequences that can be used for the Fc portion of human IgG3 (G3Fc). The G3Fc hinge region can be up to 4 times longer than other Fc chains and contains three identical 15-residue segments preceded by a similar 17-residue segment. The first G3Fc sequence shown below (SEQ ID NO: 11) contains a short hinge region consisting of a single segment of 15 residues, and the second G3Fc sequence (SEQ ID NO: 12) contains a full-length hinge region. In each case, the dotted line indicates the hinge region and the solid line indicates positions with naturally occurring variants according to UniProt P01859. Part of the invention provides polypeptides that contain, consist essentially of, or consist of amino acid sequences that are 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91% , 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 11 and 12.

EPKSCDTPPP CPRCPAPELL GGPSVFLFPP KPKDTLMISR TPEVTCVWD

VSHEDPEVQF KWYVDGVEVH NAKTKPREEQ YNSTFRWSV LTVLHQDWLN

101 GKEYKCKVSN KALPAPIEKT ISKTKGQPRE PQVYTLPPSR EEMTKNQVSL 151 TCLVKGFYPS DIAVEWESSG QPENNYNTTP PMLDSDGSFF LYSKLTVDKS 201 RWQQGNIFSC SVMHEALHNR FTQKSLSLSP GK (SEQ ID NO AND)

ELKTPLGDTT HTCPRCPEPK SCDTPPPCPR CPEPKSCDTP PPCPRCPEPK

SCDTPPPCPR CPAPELLGGP SVFLFPPKPK DTLMISRTPE VTCVWDVSH

101 EDPEVQFKWY VDGVEVHNAK TKPREEQYNS TFRWSVLTV LHQDWLNGKE 151 YKCKVSNKAL PAPIEKTISK TKGQPREPQV YTLPPSREEM TKNQVSLTCL 201 VKGFYPSDIA VEWESSGQPE NNYNTTPPML DSDGSFFLYS KLTVDKSRWQ 251 Q GNIFSCSVM HEALHNRFTQ KSLSLSPGK (SEQ ID NO 12)

Naturally occurring G3Fc variants (for example, see Uniprot P01860) include E68Q, P76L, E79Q, Y81F, D97N, N100D, T124A, S169N, S169del, F221Y when moving to the numbering system used for SEQ ID NO: 11, and in this specification Proposed are fusion proteins containing G3Fc domains having one or more of these variations. In addition, the human immunoglobulin IgG3 (IGHG3) gene exhibits structural polymorphism characterized by varying hinge lengths [see Uniprot P01859]. In particular, the WIS variant lacks most of the V region and the entire CH1 region. It has an additional disulfide bond at position 7 of the hinge region in addition to that normally present at position 11. The ZUC variant lacks most of the V region, all of the CH1 region, and part of the hinge region. The OMM variant may be an allelic form or have a gamma chain of a different subclass. Proposed herein are additional fusion proteins comprising G3Fc domains containing one or more of these variants.

An example of a native amino acid sequence that can be used for the Fc portion of human IgG4 (G4Fc) is given below (SEQ ID NO: 13). The dotted line indicates the hinge region. Part of the invention provides polypeptides that contain, consist essentially of, or consist of amino acid sequences that are 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91% , 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 13.

ESKYGPPCPS CPAPEFLGGP SVFLFPPKPK DTLMISRTPE VTCVWDVSQ

EDPEVQFNWY VDGVEVHNAK TKPREEQFNS TYRWSVLTV LHQDWLNGKE

101 YKCKVSNKGL PSSIEKTISK AKGQPREPQV YTLPPSQEEM TKNQVSLTCL

151 VKGFYPSDIA VEWESNGQPE NNYKTTPPVL DSDGSFFLYS RLTVDKSRWQ 201 EGNVFSCSVM HEALHNHYTQ KSLSLSLGK (SEQ ID NO 13)

Various genetically engineered mutations of the G1Fc sequence (SEQ ID NO: 9) are presented herein, and similar mutations can be generated in G2Fc, G3Fc and G4Fc using the G1Fc alignment shown in FIG. 11. Due to the unequal length of the hinge regions, different amino acids in SEQ ID NOs are found at similar Fc positions in the isotype alignment (Fig. 11): 9, 10, 11, 12 and 13. In addition, it should be noted that the position of a given amino acid in immunoglobulin sequences consisting of the hinge region, C _H 2 and C _H 3 regions (for example, SEQ ID NOs: 9, 10, 11, 12 and 13) will be designated by a different number if the numbering covers the entire IgG1 heavy chain constant domain (consisting of regions CH1, hinge region, _CH2 and _CH3 ), as in the Uniprot database. For example, the following are some relevant positions in the C _H 3 region of the human G1Fc sequence (SEQ ID NO: 9), the constant domain of the IgG1 heavy chain (Uniprot P01857) and the human IgG1 heavy chain.

- 31 043914

Correspondence of SNZ provisions in different numbering systems GIFc (Numbering begins with the first threonine of the hinge region) IgGl heavy chain constant domain (Numbering starts from _H 1) IgGl heavy chain (EU numbering scheme according to Kabat et al., 1991*) Y127 Y232 Y349 S132 S237 S354 E134 E239 E356 T144 T249 T366 L146 L251 L368 K170 K275 KZ 92 D177 D282 D399 Y185 Y290 Y407 K187 K292 K409 * Kabat et al. (eds) 1991; pp. 688-696 in Sequences of Proteins of Immunological Interest, ^5th ed., Vol. 1, NIH, Bethesda, MD.

It should be understood that the various elements of fusion proteins (eg, immunoglobulin Fc fusion proteins) can be arranged in any manner consistent with the desired functionality. For example, the domain of an ActRIIB polypeptide may be C-terminal to a heterologous domain, or, alternatively, the heterologous domain may be C-terminal to a domain of an ActRIIB polypeptide. The ActRIIB polypeptide domain and the heterologous domain need not be contiguous in the fusion protein, and additional domains or amino acid sequences may be located between the domains, either C-terminal or N-terminal to any domain or between domains.

For example, the ActRIIB receptor fusion protein may contain the amino acid sequence shown in the formula A-B-C. Part B corresponds to the ActRIIB polypeptide domain. Parts A and C may independently be 0, 1, or more than one amino acid, and both parts A and C are heterologous to B. Parts A and/or C may be attached to part B via a linker sequence. The linker may be enriched in glycine (e.g., 2-10, 2-5, 2-4, 2-3 glycine residues) or glycine and proline residues, for example, contain a single sequence consisting of threonine/serine or glycines or repeat sequences consisting from threonine/serine or glycines, e.g. GGG (SEQ ID NO: 14), GGGG (SEQ ID NO: 15), TGGGG(SEQ ID NO: 16), SGGGG(SEQ ID NO: 17), TGGG (SEQ ID NO : 18), SGGG (SEQ ID NO: 19) or GGGGS (SEQ ID NO: 20) in either single or repeat form. In some embodiments, ActRIIB fusion proteins comprise an amino acid sequence shown in the formula A-B-C, wherein A is a leader sequence, B consists of a domain of an ActRIIB polypeptide, and C is a portion of a polypeptide that enhances one or more one of the following: in vivo stability, in vivo half-life, absorption/administration, tissue localization or distribution, protein complexation, and/or purification. In some embodiments, the ActRIIB fusion proteins contain an amino acid sequence shown in the formula A-B-C, wherein A is the TPA leader sequence, B consists of an ActRIIB polypeptide domain, and C is an immunoglobulin Fc domain. Preferred fusion proteins contain the amino acid sequence set forth in any of SEQ ID NOs: 24, 25, 28, 29, 31, 33, 34, 45, 50, 53 and 58.

In preferred embodiments, ActRIIB polypeptides for use according to the methods described herein are isolated polypeptides. As used herein, an isolated protein or polypeptide is a protein or polypeptide isolated from components of its natural environment. In some embodiments, the polypeptide of the invention is purified to a purity of greater than 95%, 96%, 97%, 98%, or 99%, for example, by electrophoresis (e.g., SDS-PAGE). , isoelectric focusing (IEF), capillary electrophoresis), or chromatographic determination (e.g., ion exchange or reverse phase high performance liquid chromatography (HPLC)). Methods for assessing the purity of antibodies are well known in the art [see, for example, Flatman et al., (2007) J. Chromatogr. At 848:79-87]. In some embodiments, ActRIIB polypeptides for use according to the methods described herein are recombinant polypeptides.

The ActRIIB polypeptides of the invention can be produced by various methods known in the art. For example, the polypeptides of the invention can be synthesized by standard protein synthesis methods such as those described by Bodansky, M. Principles of Peptide Synthesis, Springer Verlag, Berlin (1993) and Grant G.A. (ed.), Synthetic Peptides: A User's Guide, W.H. Freeman and Company, New York (1992). In addition, automated peptide synthesizers are commercially available (eg, Advanced ChemTech Model 396; Milligen/Biosearch 9600). Alternatively, the polypeptides of the invention, including fragments or variants thereof, can be produced recombinantly using

- 32 043914 various expression systems (eg, E. coli, Chinese hamster ovary (CHO) cells, COS cells, baculovirus], as is well known in the art. In a further embodiment, the modified or unmodified polypeptides of the invention can be obtained by cleavage of the resulting recombinantly of ActRIIB polypeptides using, for example, a protease such as trypsin, thermolysin, chymotrypsin, pepsin or paired amino acid cleavage enzyme (PACE).Computer analysis (using commercially available programs) can be used to detect protease cleavage sites , e.g., MacVector, Omega, PCGene, Molecular Simulation, Inc.) Alternatively, such polypeptides can be prepared from full-length ActRIIB polypeptides produced recombinantly using chemical digestion (e.g., cyanogen bromide, hydroxylamine, etc.)

B. Nucleic acids encoding ActRIIB polypeptides.

Some embodiments of this disclosure provide isolated and/or recombinant nucleic acids encoding ActRIIB polypeptides (including fragments, functional variants (eg, GDF traps), and fusion proteins thereof). For example, SEQ ID NO: 7 encodes the naturally occurring human ActRIIB precursor polypeptide (variant R64 described herein), whereas SEQ ID NO: 8 encodes the processed extracellular domain of ActRIIB (variant R64 described herein). The nucleic acids in question may be single-stranded or double-stranded. Such nucleic acids may be DNA or RNA molecules. Such nucleic acids can be used, for example, in methods for creating ActRII-based ligand decoy polypeptides, as described herein.

Isolated nucleic acid as used herein refers to a nucleic acid molecule isolated from components of its natural environment. An isolated nucleic acid encompasses a nucleic acid molecule contained in cells typically containing a nucleic acid molecule, wherein the nucleic acid molecule is located extrachromosomally or in a chromosomal region different from its natural position on the chromosome.

In some embodiments, nucleic acids encoding ActRIIB polypeptides of the invention are understood to include nucleic acids that are variants of any of SEQ ID NO: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59 , 60, 61 and 62. Nucleotide sequence variants include sequences that differ by one or more nucleotide substitutions, additions or deletions, including allelic variants, and therefore include coding sequences different from the nucleotide sequence designated by any of SEQ ID NOs : 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59, 60, 61 and 62.

In some embodiments, the ActRIIB polypeptides of the invention are encoded by isolated or recombinant nucleic acid sequences that are at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical to any of SEQ ID NO: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59, 60, 61 and 62 One skilled in the art will appreciate that the nucleic acid sequence is at least 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96% , 97%, 98%, 99% or 100% identical to sequences complementary to SEQ ID NO: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59, 60, 61 and 62 and their variants are also included in the scope of this invention. In further embodiments, the nucleic acid sequences of the invention may be isolated, recombinant and/or fused to a heterologous nucleotide sequence or in a DNA library.

In other embodiments, the nucleic acids of this invention also include nucleotide sequences that hybridize under very stringent conditions to the nucleotide sequence set forth in SEQ ID NO: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57 , 59, 60, 61 and 62, sequences complementary to SEQ ID NO: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59, 60, 61 and 62 or fragments thereof. As discussed above, one skilled in the art will readily understand that appropriate stringent conditions to promote DNA hybridization may vary. As discussed above, one skilled in the art will readily understand that appropriate stringent conditions to promote DNA hybridization may vary. For example, a 6.0 x sodium chloride/sodium citrate (SSC) hybridization can be performed at approximately 45°C, followed by a 2.0 x SSC wash at 50°C. For example, the salt concentration in the washing step can be selected from a low hardness of approximately 2.0 x SSC at 50°C to a high hardness of approximately 0.2 x SSC at 50°C. In addition, the temperature in the washing step can be increased from room temperature under low severity conditions, at approximately 22°C, to approximately 65°C under high severity conditions. Both temperature and salt concentration can vary, or temperature and salt concentration can remain constant when another variable changes. One embodiment of the disclosure provides nucleic acids that are hybridized under low stringency conditions at 6 x SSC at room temperature followed by a wash with 2 x SSC at room temperature.

Isolated nucleic acids that differ from the nucleic acids shown in SEQ ID NO: 7, 8, 26, 32, 48, 49, 51, 52, 55, 56, 57, 59, 60, 61 and 62 due to genetic degeneracy

- 33 043914 code are also included in the scope of this invention. For example, a number of amino acids are encoded by more than one triplet. Codons that encode the same amino acid, or synonymous codons (eg, CAU and CAC are synonymous codons encoding histidine), can result in silent mutations that do not affect the amino acid sequence of the protein. However, it is assumed that in mammalian cells there is a DNA sequence polymorphism that leads to changes in the amino acid sequence of the proteins in question. One skilled in the art will appreciate that within individuals of a given species, such variations may exist in one or more than one nucleotide (up to approximately 3-5% of the nucleotides) of the nucleic acids encoding a particular protein due to naturally occurring allelic diversity. Any of these nucleotide variants and the resulting amino acid polymorphisms are within the scope of this invention.

In some embodiments, the recombinant nucleic acids of this invention may be operably linked to one or more regulatory nucleotide sequences in an expression construct. Regulatory nucleotide sequences typically correspond to the host cell used for expression. Various expression vectors and suitable regulatory sequences are known in the art and can be used in a variety of host cells. Typically, the one or more regulatory sequences may include, but are not limited to, promoter sequences, leader or signal sequences, ribosome binding sites, transcription initiation and termination sequences, and enhancer or activator sequences. Constitutive or inducible promoters known in the art are included within the scope of the invention. Promoters can be either naturally occurring promoters or hybrid promoters combining elements of one or more than one promoter. The expression construct may be located in a cell or episome, such as a plasmid, or the expression construct may be inserted into a chromosome. In some embodiments, the expression vector contains a selectable marker gene that allows selection of transformed host cells. Selectable marker genes are well known in the art and may differ depending on the host cell used.

In some aspects, a subject nucleic acid of the present invention is provided as an expression vector comprising a nucleotide sequence encoding an ActRIIB polypeptide and operably linked to at least one regulatory sequence. Regulatory sequences are known in the art and are selected to direct expression of the ActRIIB polypeptide. Accordingly, the term regulatory sequence includes promoters, enhancers and other expression control elements. Examples of regulatory sequences are described by Goeddel; Gene Expression Technology: Methods in Enzymology, Academic Press, San Diego, CA (1990). For example, to express DNA sequences encoding an ActRIIB polypeptide in such vectors, any of the expression control sequences that can control the expression of a DNA sequence operably linked thereto can be used. Such expression control sequences include, for example, the SV40 early and late promoters, the tet promoter, the adenovirus or cytomegalovirus immediate early promoter, the RSV promoters, the lac system, the trp system, the TAC or TRC system, the T7 promoter, whose expression is controlled by T7 RNA polymerase , the aha lambda core operator and promoter region, control regions of the bacteriophage fd coat protein, the promoter of 3-phosphoglycerate kinase or other glycolytic enzymes, acid phosphatase promoters such as Pho5, the yeast mating factor α promoters, the polyhedron promoter of the baculovirus system, and other sequences that, are known to control the expression of genes of prokaryotic or eukaryotic cells or their viruses and various combinations thereof. It should be understood that the design of the expression vector may depend on factors such as the choice of host cell to be transformed and/or the type of protein desired to be expressed. In addition, the copy number of the vector, the ability to control that copy number, and the expression of any other protein encoded by the vector, such as antibiotic resistance markers, should be considered.

The recombinant nucleic acid of this invention can be produced by ligating a cloned gene or a portion thereof into a vector suitable for expression in either prokaryotic or eukaryotic cells (yeast, birds, insects or mammals), or both. Expression vehicles for producing recombinant ActRIIB polypeptide include plasmids and other vectors. For example, suitable vectors include the following types of plasmids: pBR322-based plasmids, pEMBL-based plasmids, pEX-based plasmids, pBTac-based plasmids, and pUC-based plasmids for expression in prokaryotic cells such as E. coli.

Some mammalian cell expression vectors include both prokaryotic sequences that facilitate vector production in bacteria and one or more eukaryotic transcription units that are expressed in eukaryotic cells. Vectors based on pcDNAI/amp, pcDNAI/neo, pRc/CMV, pSV2gpt, pSV2neo, pSV2-dhfr, pTk2, pRSVneo, pMSG, pSVT7, pkoneo and pHyg are examples of mammalian expression vectors suitable for

- 34 043914 transfection of eukaryotic cells. Some of these vectors are modified with bacterial plasmid sequences, such as pBR322, to improve replication and selection for antibiotic resistance in both prokaryotic and eukaryotic cells. Alternatively, derivatives of viruses such as bovine papilloma virus (BPV-1) or Epstein-Barr virus (pHEBo, pREP and p205 derivatives) can be used to transiently express proteins in eukaryotic cells. Examples of other viral (including retrovirus)-based expression systems are provided below in the description of gene therapy delivery systems. Various methods used to obtain plasmids and to transform host organisms are well known in the art. Other suitable expression systems for prokaryotic and eukaryotic cells, as well as general recombinant methods, are described, for example, in Molecular Cloning A Laboratory Manual, 3rd Ed., ed. by Sambrook, Fritsch and Maniatis (Cold Spring Harbor Laboratory Press, 2001). In some cases, it may be desirable to express recombinant polypeptides using a baculovirus-based expression system. Examples of such baculovirus-based expression systems include pVL derivative vectors (such as pVL1392, pVL1393 and pVL941), pAcUW derivative vectors (such as pAcUW1) and pBlueBac derivative vectors (such as pBlueBac III containing β-gal).

In a preferred embodiment, the vector will be designed to produce the subject ActRIIB polypeptides in CHO cells, such as the Pcmv-Script vector (Stratagene, La Jolla, Calif.), pcDNA4 vectors (Invitrogen, Carlsbad, Calif.), and pCI-neo vectors (Promega, Madison, Wise). It will be appreciated that the discussed gene constructs can be used to express the subject ActRII polypeptides in cultured cells, for example, to produce proteins, including fusion proteins or protein variants, for purification.

This description also covers host cells transfected with a recombinant gene comprising the coding sequence of one or more of the subject ActRIIB polypeptides. Host cells can be any prokaryotic or eukaryotic cells. For example, the ActRIIB polypeptide of the invention can be expressed in bacterial cells such as E. coli, insect cells (eg, using a baculovirus-based expression system), yeast or mammalian cells [eg, Chinese hamster ovary (CHO) cell lines]. Other suitable host cells are known to those skilled in the art.

Accordingly, this description also covers methods for preparing the subject ActRIIB polypeptides. For example, a host cell transfected with an expression vector encoding an ActRIIB polypeptide can be cultured under appropriate conditions to allow expression of the ActRIIB polypeptide. The polypeptide can be secreted and isolated from a mixture of cells and media containing the polypeptide. Alternatively, the ActRIIB polypeptide may remain in the cytoplasm or membrane fraction and the cells may be harvested, lysed, and protein isolated. Cell culture includes host cells, media, and other byproducts. Suitable media for cell culture are well known in the art. The subject polypeptides can be isolated from culture fluid, host cells, or both by methods known in the art for protein purification, including ion exchange chromatography, gel filtration, ultrafiltration, electrophoresis, immunoaffinity purification using antibodies specific to specific epitopes of ActRIIB polypeptides and affinity purification using an agent that binds to the domain fused to the ActRIIB polypeptide (eg, a protein A column can be used to purify the ActRIIB-Fc fusion protein). In some embodiments, the ActRIIB polypeptide is a fusion protein containing a domain that facilitates its purification.

In some embodiments, purification is achieved through a series of column chromatography steps including, for example, three or more of the following, in any order: protein A chromatography, Q-Sepharose chromatography, phenyl Sepharose chromatography, size exclusion chromatography, and cation exchange chromatography. Cleaning can be completed by virus filtration and buffer replacement. The ActRIIB protein can be purified to a purity of greater than 90%, greater than 95%, greater than 96%, greater than 98%, or greater than 99% by size exclusion chromatography and greater than 90%, greater than 95%, greater than 96%, greater than 98%, or greater than 99 % based on SDS PAGE results. The target purity level must be sufficient to achieve the desired results in mammalian systems, particularly non-human primates, rodents (mice), and humans.

In another embodiment, a fusion gene encoding a purification leader sequence, such as a poly(His)/enterokinase cleavage site sequence at the N-terminus of a desired portion of a recombinant ActRIIB polypeptide, may allow the expressed fusion protein to be purified by affinity chromatography using a Ni ² resin. ⁺ . The purification leader sequence can then be removed by treatment with enterokinase to obtain the purified ActRIIB polypeptide; see, for example, Hochuli et al. (1987) J Chromatography 411:177; and Janknecht et al. (1991) PNAS USA 88:8972.

Methods for creating fusion genes are well known. Essentially, the joining of different DNA fragments encoding different polypeptide sequences is carried out according to standard methods using blunt or sticky ends for ligation, restriction digestion

- 35 043914 enzymes to obtain proper ends, proper joining of sticky ends, alkaline phosphatase treatment to avoid unwanted connection and enzymatic ligation. In another embodiment, the fusion gene can be synthesized by standard methods, including automated DNA synthesizers. Alternatively, gene fragments can be amplified by PCR using different anchor primers that give rise to complementary overlapping regions of two successive gene fragments, which can then be hybridized to form a chimeric gene sequence; see, for example, Current Protocols in Molecular Biology, eds. Ausubel et al, John Wiley & Sons: 1992.

B. Antibody antagonists.

In some aspects, the ActRIIB antagonist for use according to the methods and uses described herein is an antibody (ActRIIB antagonist antibody) or a combination of antibodies. An ActRIIB antagonist antibody or combination of antibodies may bind, for example, to one or more ActRIIB-binding ligands (eg, activin, GDF11, GDF8, GDF3, BMP10, and BMP6), ActRIIB receptor, type I receptor (eg, ALK4, ALK5, and /or ALK7) and/or co-receptor ActRIIB. As described herein, ActRIIB antagonist antibodies can be used alone or in combination with one or more additional maintenance therapy or other active agent for the treatment of myelofibrosis, in particular, the treatment or prevention of one or more than one complication of myelofibrosis (for example, splenomegaly , extramedullary hematopoiesis, anemia and fibrosis) and/or treatment of a patient treated with a Janus kinase inhibitor.

In some aspects, an ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, and/or activin BE) . Thus, in some embodiments, an ActRIIB antagonist antibody or combination of antibodies is described to bind at least activin. As used herein, an anti-activin antibody generally refers to an antibody that binds to activin with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting activin. In some embodiments, the degree of binding of the anti-activin antibody to the non-activin foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1 % of antibody binding to activin as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-activin antibody binds to an epitope of activin that is conserved among activins of different biological species. In some preferred embodiments, the anti-activin antibody binds to human activin. In some embodiments, an anti-activin antibody may inhibit the binding of activin to type I and type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit activin-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-activin antibody may inhibit the binding of activin to the ActRIIB co-receptor and thereby inhibit activin-mediated signaling (eg, Smad signaling pathway). It should be noted that activin A has a homologous sequence to activin B and therefore antibodies that bind to activin A may also bind to activin B and/or inhibit activin B, which also applies to antibodies to activin B. In some embodiments, the invention relates to multispecific an antibody (eg, a bispecific antibody) that binds to activin and further binds, for example, to one or more additional ActRIIB ligands [eg, GDF11, GDF8, GDF3, BMP10 and BMP6], one or more than one type I receptor, and /or a type II receptor (eg ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as uses thereof. In some embodiments, a multispecific antibody that binds to activin does not bind or substantially does not bind to BMP9 (e.g., binds to BMP9 with a KD greater than ^1x10'7 M or has relatively moderate binding, e.g., about 1x10'8 M or about 1x10' 9 M). In some embodiments, a multispecific antibody that binds to activin does not bind or substantially does not bind to activin A (e.g., binds to activin A with a KD greater than ^1x10'7 M or has relatively moderate binding, e.g., about 1x10'8 M or about 1x10'9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination includes an anti-activin antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB superfamily ligands [for example, GDF8, GDF11, GDF3 and BMP6 ], one or more type I receptors and/or type II receptors (eg ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as uses thereof. In some embodiments, the antibody combination that includes an anti-activin antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-activin antibody does not include an anti-activin A antibody.

- 36 043914

In some aspects, an antagonist antibody or combination of antibodies is an antibody that inhibits at least activin B. Thus, in some embodiments, an ActRIIB antagonist antibody or combination of antibodies binds to at least activin B. As used herein, an anti-activin B antibody generally refers to an antibody , which binds to activin B with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent that targets activin B. In some embodiments, the degree of binding of the anti-activin B antibody to a non-activin B protein is less than less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to activin as determined, for example, by radioimmunoassay (RIA), Biacore or other protein interaction or binding affinity assay. In some embodiments, the anti-activin B antibody binds to an epitope of activin B that is conserved among the activin Bs of various species. In some preferred embodiments, an anti-activin B antibody can inhibit the binding of activin B to type I and type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit activin B-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-activin B antibody may inhibit the binding of activin B to the ActRIIB co-receptor and thereby inhibit activin B-mediated signaling (eg, the Smad signaling pathway). It should be noted that activin B has a homologous sequence to activin A and therefore antibodies that bind to activin B may also bind to activin A and/or inhibit activin A. In some embodiments, the invention provides a multispecific antibody (e.g., a bispecific antibody) that binds to activin B and further binds, for example, to one or more additional ActRIIB ligands [e.g., GDF11, GDF8, GDF3, BMP10, and BMP6], one or more type I receptors, and/or type II receptors (e.g., ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as applications thereof. In some embodiments, a multispecific antibody that binds activin B does not bind or substantially does not bind BMP9 (e.g., binds BMP9 with a KD greater than 1x10'7 M or has relatively moderate binding, e.g., about 1x10'8 M or about 1x10 '9 M). In some embodiments, a multispecific antibody that binds to activin B does not bind or substantially does not bind to activin A (e.g., binds to activin A with a KD greater than 1x10'7 M or has relatively moderate binding, e.g., approximately 1x10'8 M or approximately 1x10'9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-Activin B antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB ligands [e.g., GDF8, GDF11, GDF3, BMP6 and BMP10], one or more type I receptors and/or type II receptors (eg ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as uses thereof. In some embodiments, the antibody combination that includes an anti-Activin B antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-Activin B antibody does not include an anti-Activin A antibody.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least GDF8. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least GDF8. As used herein, an anti-GDF8 antibody generally refers to an antibody that binds to GDF8 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting GDF8. In some embodiments, the degree of binding of the anti-GDF8 antibody to the non-GDF8 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to GDF8 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-GDF8 antibody binds to a GDF8 epitope that is conserved across GDF8 across different species. In some preferred embodiments, the anti-GDF8 antibody binds to human GDF8. In some embodiments, an anti-GDF8 antibody may inhibit the binding of GDF8 to type I and type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit GDF8-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-GDF8 antibody may inhibit the binding of GDF8 to co-receptors and thereby inhibit GDF8-mediated signaling (eg, the Smad signaling pathway). It should be noted that GDF8 has sequence homology to GDF11 and therefore antibodies that bind to GDF8 may, in some cases, also bind to GDF11 and/or inhibit GDF11. In some embodiments, the invention provides a multispecific antibody (e.g., a bispecific antibody) that binds to GDF8 and further binds, for example, to one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF11, GDF3, BMP10 and

- 37 043914

BMP6], one or more type I receptors and/or type II receptors (eg ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as applications thereof. In some embodiments, a multispecific antibody that binds to GDF8 does not bind or substantially does not bind to BMP9 (e.g., binds to BMP9 with a KD greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or about 1x10 ^{- 9} M). In some embodiments, a multispecific antibody that binds to GDF8 does not bind or substantially does not bind to activin A (e.g., binds to activin A with a KD greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or about 1x10 ^-9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-GDF8 antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF11, GDF3, BMP6, BMP10 and BMP15], one or more type I receptors and/or type II receptors ( for example, ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as their uses. In some embodiments, the antibody combination that includes an anti-GDF8 antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-GDF8 antibody does not include an anti-activin A antibody.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least GDF11. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least GDF11. As used herein, an anti-GDF11 antibody generally refers to an antibody that binds to GDF11 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting GDF11. In some embodiments, the degree of binding of the anti-GDF11 antibody to the non-GDF11 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to GDF11 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-GDF11 antibody binds to an epitope of GDF11 that is conserved across GDF11 from different species. In some preferred embodiments, the anti-GDF11 antibody binds to human GDF11. In some embodiments, an anti-GDF11 antibody may inhibit the binding of GDF11 to type I and/or type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit GDF11-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-GDF11 antibody may inhibit the binding of GDF11 to co-receptors and thereby inhibit GDF11-mediated signaling (eg, the Smad signaling pathway). It should be noted that GDF11 has sequence homology to GDF8 and therefore antibodies that bind to GDF11 may, in some cases, also bind to GDF8 and/or inhibit GDF8. In some embodiments, the invention provides a multispecific antibody (e.g., a bispecific antibody) that binds to GDF11 and further binds, for example, to one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF3, BMP10 and BMP6], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as applications thereof. In some embodiments, a multispecific antibody that binds to GDF11 does not bind or substantially does not bind to BMP9 (e.g., binds to BMP9 with a KD greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or about 1x10 ^{- 9} M). In some embodiments, a multispecific antibody that binds to GDF11 does not bind or substantially does not bind to activin A (e.g., binds to activin A with a KD greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or about 1x10 ^-9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-GDF11 antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF3, BMP6 and BMP10], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more coreceptors, as well as their uses. In some embodiments, the antibody combination that includes an anti-GDF11 antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-GDF11 antibody does not include an anti-activin A antibody.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least BMP6. Thus, in some embodiments, the antibody

- 38 043914 ActRIIB antagonist or combination of antibodies binds to at least BMP6. As used herein, an anti-BMP6 antibody generally refers to an antibody that binds to BMP6 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting BMP6. In some embodiments, the degree of binding of the anti-BMP6 antibody to the non-BMP6 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to GDF11 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, an anti-BMP6 antibody binds to an epitope of BMP6 that is conserved among BMP6s across different species. In some preferred embodiments, the anti-BMP6 antibody binds to human BMP6. In some embodiments, an anti-BMP6 antibody may inhibit the binding of BMP6 to type I and/or type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit BMP6-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-BMP6 antibody may inhibit the binding of BMP6 to co-receptors and thereby inhibit BMP6-mediated signaling (eg, the Smad signaling pathway). In some embodiments, the invention provides a multispecific antibody (e.g., a bispecific antibody) that binds to BMP6 and further binds, for example, to one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF3, BMP10 and GDF11], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as applications thereof. In some embodiments, a multispecific antibody that binds to BMP6 does not bind or substantially does not bind to BMP9 (e.g., binds to BMP9 with a KD greater than 1x10'7 M or has relatively moderate binding, e.g., about 1x10'8 M or about 1x10' 9 M). In some embodiments, a multispecific antibody that binds to BMP6 does not bind or substantially does not bind to activin A (e.g., binds to activin A with a KD greater than 1x10'7 M or has relatively moderate binding, e.g., about 1x10'8 M or about 1x10'9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination includes an anti-BMP6 antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF11, GDF3 and BMP10], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as their uses. In some embodiments, the antibody combination that includes an anti-BMP6 antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-BMP6 antibody does not include an anti-activin A antibody.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least GDF3. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least GDF3. As used herein, an anti-GDF3 antibody generally refers to an antibody that binds to GDF3 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting GDF3. In some embodiments, the degree of binding of the anti-GDF3 antibody to the non-GDF3 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to GDF11 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-GDF3 antibody binds to a GDF3 epitope that is conserved across GDF3 across species. In some preferred embodiments, the anti-GDF3 antibody binds to human GDF3. In some embodiments, an anti-GDF3 antibody may inhibit the binding of GDF3 to type I and type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit GDF3-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-GDF3 antibody may inhibit the binding of GDF3 to co-receptors and thereby inhibit GDF3-mediated signaling (eg, the Smad signaling pathway). In some embodiments, the invention provides a multispecific antibody (e.g., a bispecific antibody) that binds to GDF3 and further binds, for example, to one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, BMP6, BMP10 and GDF11], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as applications thereof. In some embodiments, a multispecific antibody that binds to GDF3 does not bind

- 39 043914 or substantially does not bind to BMP9 (eg, binds to BMP9 with a K _D greater than 1x10 ^-7 M or has relatively moderate binding, eg, about 1x10 ^-8 M or about 1x10 ^-9 M). In some embodiments, a multispecific antibody that binds to GDF3 does not bind or substantially does not bind to activin A (e.g., binds to activin A with a K _D greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or approximately 1x10 ^-9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-GDF3 antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF11, BMP6 and BMP10], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as their uses. In some embodiments, the antibody combination that includes an anti-GDF3 antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-GDF3 antibody does not include an anti-activin A antibody.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least BMP10. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least BMP10. As used herein, an anti-BMP10 antibody generally refers to an antibody that binds to BMP10 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting BMP10. In some embodiments, the degree of binding of the anti-BMP10 antibody to the non-BMP10 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to BMP10 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-BMP10 antibody binds to a BMP10 epitope that is conserved among BMP10s from different species. In some preferred embodiments, the anti-BMP10 antibody binds to human BMP10. In some embodiments, an anti-BMP10 antibody may inhibit the binding of BMP10 to type I and type II receptors (eg, ActRIIB, ALK4, ALK5 and/or ALK7) and thereby inhibit BMP10-mediated signaling (eg, Smad signaling pathway). In some embodiments, an anti-BMP10 antibody may inhibit the binding of BMP10 to co-receptors and thereby inhibit BMP10-mediated signaling (eg, the Smad signaling pathway). In some embodiments, the invention provides a multispecific antibody (e.g., a bispecific antibody) that binds to BMP10 and further binds, for example, to one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF11, GDF3 and BMP6], one or more type I receptors and/or type II receptors (e.g. ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as applications thereof. In some embodiments, a multispecific antibody that binds to BMP10 does not bind or substantially does not bind to BMP9 (e.g., binds to BMP9 with a K _D greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or about 1x10 ^-9 M). In some embodiments, a multispecific antibody that binds to BMP10 does not bind or substantially does not bind to activin A (e.g., binds to activin A with a K _D greater than 1x10 ^-7 M or has relatively moderate binding, e.g., about 1x10 ^-8 M or approximately 1x10 ^-9 M). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-BMP10 antibody and one or more additional antibodies that bind, for example, one or more additional ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), GDF8, GDF3, BMP6, BMP10 and GDF11], one or more type I receptors and/or type II receptors ( for example, ActRIIB, ALK4, ALK5 and/or ALK7) and/or one or more co-receptors, as well as their uses. In some embodiments, the antibody combination that includes an anti-BMP10 antibody does not include an anti-BMP9 antibody. In some embodiments, the antibody combination that includes an anti-BMP10 antibody does not include an anti-activin A antibody.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least ActRIIB. Thus, in some embodiments, an ActRIIB antagonist antibody or combination of antibodies binds to at least ActRIIB. As used herein, an anti-ActRIIB antibody generally refers to an antibody that binds to ActRIIB with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting ActRIIB. In some embodiments, the degree of binding of the anti-ActRIIB antibody to the non-ActRIIB foreign protein is less than about 10%, 9%, 8%,

- 40 043914

7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to ActRIIB as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, an anti-ActRIIB antibody binds to an ActRIIB epitope that is conserved across ActRIIBs from different species. In some preferred embodiments, the anti-ActRIIB antibody binds to human ActRIIB. In some embodiments, an anti-ActRIIB antibody may inhibit the binding of ActRIIB to one or more ActRIIB ligands [e.g., GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), GDF11, BMP6, GDF3 and BMP10]. In some embodiments, an anti-ActRIIB antibody is a multispecific antibody (e.g., a bispecific antibody) that binds to ActRIIB and one or more ActRIIB ligands [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC), GDF3, BMP6 and BMP10], type I receptor (for example, ALK4, ALK5 and/or ALK7], co-receptor and/or additional type II receptor (for example, ActRIIA). In some In embodiments, the invention relates to antibody combinations, wherein the antibody combination comprises an anti-ActRIIB antibody and one or more additional antibodies that bind, for example, one or more than one ActRIIB ligand [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE, activin BE), BMP6, GDF3 and BMP10], co-receptor, type I receptor (for example, ALK4, ALK5 and/or ALK7) and/or an additional type II receptor (eg, ActRIIA), as well as applications thereof. It should be noted that ActRIIB has sequence similarity to ActRIIA and therefore antibodies that bind to ActRIIB may, in some cases, also bind to GDF8 and/or inhibit ActRIIA.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least ALK4. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least ALK4. As used herein, an anti-ALK4 antibody generally refers to an antibody that binds to ALK4 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting ALK4. In some embodiments, the degree of binding of the anti-ALK4 antibody to the non-ALK4 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to ALK4 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-ALK4 antibody binds to an epitope of ALK4 that is conserved across ALK4 across species. In some preferred embodiments, the anti-ALK4 antibody binds to human ALK4. In some embodiments, an anti-ALK4 antibody may inhibit the binding of ALK4 to one or more ActRIIB ligands [e.g., GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), GDF11, BMP6, GDF3 and BMP10]. In some embodiments, the anti-ALK4 antibody is a multispecific antibody (e.g., a bispecific antibody) that binds to ALK4 and one or more GDF/BMP ligands [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC), GDF3, BMP6 and BMP10], type II receptor (eg, ActRIIB), co-receptor and/or additional type I receptor (eg, ALK5 and/or ALK7). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-ALK4 antibody and one or more additional antibodies that bind, for example, one or more than one ActRIIB ligand [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), BMP6 and BMP10], co-receptor, type II receptor (for example, ActRIIB) and/or accessory receptor I type (eg ALK5 and/or ALK7), as well as their applications.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least ALK5. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least ALK5. As used herein, an anti-ALK5 antibody generally refers to an antibody that binds to ALK5 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting ALK5. In some embodiments, the degree of binding of the anti-ALK5 antibody to the non-ALK5 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to ALK5 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-ALK5 antibody binds to an epitope of ALK5 that is conserved among ALK5s across species. In some preferred embodiments, the anti-ALK5 antibody binds to human ALK5. In some embodiments, an anti-ALK5 antibody may inhibit the binding of ALK5 to one or more

- 41 043914 ActRIIB gang [for example, GDF8, activin (for example, activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), GDF11, BMP6, GDF3 and BMP10] . In some embodiments, the anti-ALK5 antibody is a multispecific antibody (e.g., a bispecific antibody) that binds to ALK5 and one or more ActRIIB ligands [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC), GDF3, BMP6 and BMP10], type II receptor (eg ActRIIB), co-receptor and/or additional type I receptor (eg ALK4 and/or ALK7). In some embodiments, the invention provides antibody combinations, wherein the antibody combination includes an anti-ALK5 antibody and one or more additional antibodies that bind, for example, one or more than one ActRIIB ligand [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), BMP6 and BMP10], co-receptor, type II receptor (for example, ActRIIB) and/or accessory receptor I type (eg ALK4 and/or ALK7), as well as their applications.

In some aspects, the ActRIIB antagonist antibody or combination of antibodies is an antibody that inhibits at least ALK7. Thus, in some embodiments, the ActRIIB antagonist antibody or combination of antibodies binds to at least ALK7. As used herein, an anti-ALK7 antibody generally refers to an antibody that binds to ALK7 with sufficient affinity such that the antibody can be used as a diagnostic and/or therapeutic agent targeting ALK7. In some embodiments, the degree of binding of the anti-ALK7 antibody to the non-ALK7 foreign protein is less than about 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or less than about 1% of antibody binding to ALK7 as determined by, for example, radioimmunoassay (RIA), Biacore, or other protein interaction or binding affinity assay. In some embodiments, the anti-ALK7 antibody binds to an epitope of ALK7 that is conserved among ALK7s across different species. In some preferred embodiments, the anti-ALK7 antibody binds to human ALK7. In some embodiments, an anti-ALK7 antibody may inhibit the binding of ALK7 to one or more ActRIIB ligands [e.g., GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), GDF11, BMP6, GDF3 and BMP10]. In some embodiments, the anti-ALK7 antibody is a multispecific antibody (e.g., a bispecific antibody) that binds to ALK7 and one or more ActRIIB ligands [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC), GDF3, BMP6 and BMP10], type II receptor (eg ActRIIB), co-receptor and/or additional type I receptor (eg ALK4 and/or ALK5). In some embodiments, the invention provides antibody combinations, wherein the antibody combination comprises an anti-ALK7 antibody and one or more additional antibodies that bind, for example, one or more ActRIIB ligands [e.g., GDF11, GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and activin BE), BMP6 and BMP10], co-receptor, type II receptor (for example, ActRIIB) and/or accessory receptor I type (eg ALK4 and/or ALK5), as well as their applications.

The term antibody is used herein in its broadest sense to cover a variety of antibody structures, including, but not limited to, monoclonal antibodies, polyclonal antibodies, multispecific antibodies (eg, bispecific antibodies), and antibody fragments, as long as they exhibit the desired antigen-binding activity. An antibody fragment refers to a molecule, other than the intact antibody, containing a portion of the intact antibody that binds to the antigen to which the intact antibody binds. Examples of antibody fragments include, but are not limited to, Fv, Fab, Fab', Fab'-SH, F(ab') ₂ , diabodies, linear antibodies, single chain antibody molecules (eg, scFv) and multispecific antibodies formed by antibody fragments [see eg Hudson et al. (2003) Nat. Med. 9:129-134; Pluckthun, in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds., (Springer-Verlag, New York), pp. 269-315 (1994); WO 93/16185; and patents US 5571894; 5587458 and 5869046]. Diabodies are fragments with two antigen-binding sites, which can be divalent or bispecific [see, for example, EP 404,097; WO 1993/01161; Hudson et al. (2003) Nat. Med. 9:129-134 (2003); and Hollinger et al. (1993) Proc. Natl. Acad. Sci. USA 90: 6444-6448]. Tribodies and tetrabodies are also described by Hudson et al. (2003) Nat. Med. 9:129-134. Single-domain antibodies and antibody fragments containing all or part of the variable domain of the heavy chain of an antibody or all or part of the variable domain of the light chain of an antibody. In some embodiments, the single domain antibody is a human single domain antibody [see, for example, US Pat. No. 6,248,516]. The antibodies set forth herein may be polyclonal antibodies or monoclonal antibodies. In some embodiments, the antibodies of this invention contain a detectable label attached thereto (e.g., the label may be a radioactive isotope, a fluorescent compound, an enzyme

- 42 043914 or enzyme cofactor). In some preferred embodiments, the antibodies of this invention are isolated antibodies. In some preferred embodiments, the antibodies of this invention are recombinant antibodies.

The antibodies of this invention can be of any class. The class of an antibody refers to the type of constant domain or constant region of its heavy chain. There are five main classes of antibodies: IgA, IgD, IgE, IgG and IgM, and some of them can be further divided into subclasses (isotypes), for example, IgG1, IgG2, _IgG3 , _IgG4 , IgA1 and IgA2. The heavy chain constant domains corresponding to the various classes of immunoglobulins are designated alpha, delta, epsilon, gamma and mu.

Typically, an antibody for use in the methods disclosed herein binds specifically to its target antigen, preferably with high binding affinity. Affinity may be expressed as a KD value and reflects intrinsic binding affinity (eg, minimizing avidity). Typically, binding affinity is determined in vitro, both in cell-free and cell-based assays. Any number of assays known in the art can be used to determine binding affinity, including those described herein, for example, Biacore, radiolabeled antigen (RIA) binding assay, and ELISA. In some embodiments, antibodies herein bind to their target antigens (e.g., ALK4, ALK5, ALK7, ActRIIB, GDF3, activin, GDF11, GDF8, BMP10, and/or BMP6) with a KD of at least 1x10 or stronger than 1x10 '8 or stronger, 1x10' ⁹ or stronger, 1x10'1° or stronger, 1x10'11 or stronger, 1x10'12 or stronger, 1x10' ¹³ or stronger, or 1x10' ¹⁴ or stronger.

In some embodiments, KD is determined using an RIA that is performed with the Fab version of the antibody of interest and its target antigen, as described in the following study. The binding affinity of a Fab to an antigen in solution is determined by equilibrating the Fab with a minimal concentration of radiolabeled antigen (eg, labeled with ¹²⁵ I) in the presence of serial dilutions of unlabeled antigen, followed by capture of the bound antigen by anti-Fab antibody adsorbed on the plate [see, for example, Chen et al. (1999) J. Mol. Biol. 293:865-881]. The experiment was performed under the following conditions: anti-Fab capture antibody (e.g., Cappel Labs) was immobilized (e.g., overnight) onto the surface of multiwell plates (e.g., MICROTITER® from Thermo Scientific) and then blocked with bovine serum albumin, preferably at room temperature ( approximately 23°C). In a non-adsorbent plate, the radiolabeled antigen and a series of dilutions of the Fab of interest were mixed [eg, similar to the determination of the anti-VEGF antibody (Fab-12) described by Presta et al, (1997) Cancer Res. 57:4593-4599]. The Fab of interest was then incubated, preferably overnight, however incubation could be continued for a longer time (eg, approximately 65 hours) to achieve equilibrium. The mixtures were then transferred to a capture plate and incubated, preferably at room temperature, for approximately one hour. The solution was then removed and the plate was washed several times, preferably with a mixture of polysorbate 20 and PBS (phosphate buffered saline). After the plate had dried, scintillation fluid (eg, MICROSCINT® from Packard) was added and counts were counted using a gamma counter (eg, TOPCOUNT® from Packard).

In another embodiment, KD was determined by surface plasmon resonance, for example, on a BIACORE® 2000 or BIACORE® 3000 biosensor (BIAcore, Inc., Piscataway, N.J.) with antigen immobilized on the surface of CM5 chips to a level of approximately 10 response units (RU). Briefly, biosensor chips with a carboxymethylated dextran surface (CM5, BIACORE, Inc.) were activated with N-ethyl-N-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) according to the supplier's instructions.

For example, the antigen can be diluted with 10 mM sodium acetate, pH 4.8, to 5 μg/ml (approximately 0.2 μM) before injection at a rate of 5 μl/minute to achieve protein binding causing a shift in the sensorgram of approximately 10 response units (RU ). After injection of the antigen, 1 M ethanolamine is injected to block unreacted groups. For kinetic measurements, 2-fold dilutions of Fab (0.78 to 500 nM) were injected into PBS with 0.05% polysorbate 20 (TWEEN-20®) surfactant (PBST) at a flow rate of approximately 25 μL/min. Association rates (k _on ) and dissociation rates (k _off ) are calculated, for example, using the simple one-to-one binding model proposed by Langmuir (BIACORE® Evaluation Software version 3.2). The equilibrium dissociation constant (KD) is calculated as the ratio k _off / k _on [see, for example, Chen et al., (1999) J. Mol. Biol. 293:865-881]. If the association rate exceeds, for example, 106 M' ¹ s' ¹ in a surface plasmon resonance study as described above, the association rate can be determined by a fluorescence quenching method, which allows one to determine the increase or decrease in the intensity of the emitted fluorescence (for example, excitation - 295 nm; emission - 340 nm, bandwidth 16 nm) 20 nM antibody to antigen (in Fab form) in PBS in the presence of increasing concentrations of antigen, measured using a spectrophotometer, such as a spectrophotometer equipped with a stop-flow measurement attachment (Aviv Instru

- 43 043914 ments) or spectrophotometer SLM-AMINCO® series 8000 (ThermoSpectronic) with stirred cuvette.

Antibody fragments can be produced by a variety of methods, including, but not limited to, proteolytic cleavage of the intact antibody, as well as production by recombinant host cells (eg, E. coli or phages), as described herein. Nucleic acids and amino acid sequences of human ALK4, ALK5, ALK7, ActRIIB, activin (Activin A, Activin B, Activin C and Activin E), GDF11, GDF8, BMP10, GDF3 and BMP6 are known in the art. In addition, numerous methods for generating antibodies are known in the art, some of which are described herein. Thus, an antagonist antibody for use herein can be prepared routinely by one skilled in the art based on knowledge known in the art and the information provided herein.

In some embodiments, the provided antibody is a chimeric antibody. The term chimeric antibody refers to an antibody in which a portion of the heavy and/or light chain is derived from a particular source or species, while the remainder of the heavy and/or light chain is derived from a different source or species. Some chimeric antibodies are described, for example, in US patent 4816567 and Morrison et al, (1984) Proc. Natl. Acad. Sci. USA, 81:6851-6855. In some embodiments, the chimeric antibody comprises a non-human variable region (eg, a variable region originating from a mouse, rat, hamster, rabbit, or non-human primate such as a monkey) and a constant region of human origin. In the following example, a chimeric antibody is a class-switched antibody in which the class or subclass has been changed compared to the parent antibody. Typically, chimeric antibodies also include their antigen-binding fragments.

In some embodiments, the proposed chimeric antibody is a humanized antibody. A humanized antibody refers to a chimeric antibody containing non-human hypervariable region (HVR) amino acid residues and human framework (FR) amino acid residues. In some embodiments, a humanized antibody will comprise substantially all of at least one, and typically two, variable domains in which all or substantially all of the HVRs (e.g., CDRs) correspond to those of non-human immunoglobulins, and all or substantially all of all FRs correspond to those of the human antibody. The humanized antibody may optionally contain at least a portion of the antibody constant region derived from a human antibody. A humanized form of an antibody, eg a non-human antibody, refers to an antibody that has been humanized. Humanized antibodies and methods for their production are described, for example, Almagro and Fransson (2008) Front. Biosci. 13:1619-1633 and further described, for example, Riechmann et al, (1988) Nature 332:323-329; Queen et al. (1989) Proc. Nat'l Acad. Sci. USA 86:10029-1оОзЗ; in US patents 5821337; 7527791; 6982321 and 7087409; Kashmiri et al, (2005) Methods 36:25-34 [SDR transplant description (a-CDR)]; Padlan, Mol. Immunol. (1991) 28:489-498 (description of surface changes); Dall'Acqua et al. (2005) Methods 36:43-60 (describing FR shuffling); Osbourne et al. (2005) Methods 36:61–68 and Klimka et al. Br. J. Cancer 2000 (2000) 252-260 (describing the FR shuffling directional selection approach). Framework regions of human origin that can be used for humanization include, but are not limited to: framework regions selected using the best-fit method (see, for example, Sims et al. (1993) J. Immunol. 151:2296]; framework regions derived from a consensus sequence of human antibodies with variable regions of light or heavy chains of a particular subgroup (see, for example, Carter et al. (1992) Proc. Natl. Acad. Sci. USA, 89:4285 and Presta et al. (1993) J . Immunol., 151:2623]; mature framework regions of human origin (with somatic mutations) or framework regions of primary antibodies (see, for example, Almagro and Fransson (2008) Front. Biosci. 13:1619-1633]; and framework regions , obtained from screening FR libraries (see, for example, Vasa et al., (1997) J. Biol. Chem. 272:10678-10684 and Rosok et al., (1996) J. Biol. Chem. 271:2261122618] .

In some embodiments, the provided antibody is a human antibody. Human antibodies can be produced by various methods known in the art. Human antibodies are generally described by van Dijk and van de Winkel (2008) Curr. Opin. Pharmacol. 5: 368-74 (2001) and Lonberg, Curr. Opin. Immunol. 20:450-459. For example, human antibodies can be produced by administering an immunogen (eg, a GDF11 polypeptide, an Activin B polypeptide, an ActRIIA polypeptide, or an ActRIIB polypeptide) to a transgenic animal modified to produce intact human antibodies or intact antibodies with variable regions of human origin in response to administration of the antigen. Such animals typically have all or part of the human immunoglobulin loci, replacing endogenous immunoglobulin loci that are extrachromosomal or randomly integrated into the animal's chromosomes. In such transgenic animals, the endogenous immunoglobulin loci are usually inactivated. A review of methods for producing human antibodies in transgenic animals is provided, for example, by Lonberg (2005) Nat. Biotech.

- 44 043914

23:1117-1125; in US patent 6075181 and 6150584 (describing XENOMOUSE™ technology); US Pat. No. 5,770,429 (describing HuMab® technology); patent US 7041870 (describing K-M MOUSE® technology) and published patent application US 2007/0061900 (describing VelociMouse® technology). The human-derived variable regions of intact antibodies produced in such animals can be further modified, for example, by combination with various human-derived constant regions.

The proposed human antibodies can also be obtained using hybridoma technology. Human myeloma and mouse-human heteromyeloma cell lines have been described for the production of human monoclonal antibodies [see, for example, Kozbor J. Immunol., (1984) 133: 3001; Brodeur et al. (1987) Monoclonal Antibody Production Techniques and Applications, pp. 51-63, Marcel Dekker, Inc., New York; and Boerner et al. (1991) J. Immunol., 147: 86]. Human antibodies produced using hybridoma technology using human B cells are described by Li et al., (2006) Proc. Natl. Acad. Sci. USA, 103:3557-3562. Additional methods include those described, for example, in US Pat. No. 7,189,826 (describing the production of human IgM monoclonal antibodies from hybridoma cell lines) and Ni, Xiandai Mianyixue (2006) 26(4):265-268 (2006) (describing human-human hybridomas) . Human hybridoma technology (triome technology) is also described by Vollmers and Brandlein (2005) Histol. Histopathol., 20(3):927–937 (2005) and Vollmers and Brandlein (2005) Methods Find Exp. Clin. Pharmacol., 27(3): 185-91. The human antibodies provided herein can also be produced by isolating the variable domain sequences of Fv clones selected from phage display libraries constructed from human genes. Such variable domain sequences can then be combined with the required constant domain of human origin. Technologies for selecting human antibodies from antibody libraries are known in the art and are described herein.

For example, antibodies of this invention can be isolated by screening combinatorial libraries for antibodies with the desired activity or activities. Various methods are known in the art for constructing phage display libraries and methods for screening such libraries for antibodies having the desired binding characteristics. An overview of such methods is presented, for example, by Hoogenboom et al. (2001) in Methods in Molecular Biology 178:1-37, O'Brien et al., ed., Human Press, Totowa, N.J. and is further described, for example, by McCafferty et al. (1991) Nature 348:552-554; Clackson et al., (1991) Nature 352: 624-628; Marks et al. (1992) J. Mol. Biol. 222:581-597; Marks and Bradbury (2003) in Methods in Molecular Biology 248:161–175, Lo, ed., Human Press, Totowa, N.J.; Sidhu et al. (2004) J. Mol. Biol. 338(2):299-310; Lee et al. (2004) J. Mol. Biol. 340(5): 1073-1093; Fellowes (2004) Proc. Natl. Acad. Sci. USA 101(34):12467-12472; and Lee et al. (2004) J. Immunol. Methods 284(1-2): 119-132.

In some phage display methods, the VH and VL gene repertoires are cloned separately using polymerase chain reaction (PCR) and recombined randomly in phage libraries, with the phages further screened for antigen binding, as described by Winter et al. (1994) Ann. Rev. Immunol., 12: 433–455. Typically, phages present antibody fragments, either as single-chain Fv fragments (scFv) or Fab fragments. Libraries from immunized sources allow the generation of high-affinity antibodies to an immunogen (eg, ALK4, ALK5, ALK7, ActRIIB, activin, GDF11, GDF8, GDF3, BMP10, or BMP6) without the need for hybridoma construction. Alternatively, a repertoire of primary antibody genes (eg, human) can be cloned to provide a single source of antibodies to a wide range of foreign as well as self antigens without any immunization, as described by Griffiths et al. (1993) EMBO J, 12: 725-734. Finally, naïve or primary libraries can also be synthesized by cloning unrearranged segments comprising stem cell V genes and using PCR primers containing random sequences that encode highly variable CDR3 regions and allow rearrangement in vitro, as described by Hoogenboom and Winter (1992). ) J. Mol. Biol., 227: 381-388. Publications describing phage libraries of human antibodies include, for example: US patent 5750373 and US publications 2005/0079574, 2005/0119455, 2005/0266000, 2007/0117126, 2007/0160598, 2007/0237764, 2007/0292 936 and 2009/0002360.

In some embodiments, the antibody provided is a multispecific antibody, such as a bispecific antibody.

Multispecific antibodies (typically monoclonal antibodies) that specifically bind to at least two different epitopes (for example, two, three, four, five, six or more) or one or more (for example, two, three, four, five, six or more) antigens.

Techniques for producing multispecific antibodies involve co-recombinant expression of heavy chain-light chain pairs of two immunoglobulins with different specificities [see, for example, Milstein and Cuello (1983) Nature 305: 537; international publication. WO 93/08829 and Traunecker et al. (1991) EMBO J, 10: 3655 and US patent 5731168 (knob-in-hole technology)], but are not limited to them. Multispecific antibodies can also be obtained using the electrostatic effect

- 45 043914 technical interaction for the construction of heterodimeric antibody molecules with an Fc fragment (see, for example, WO 2009/089004 A1); cross-linking two or more antibodies or fragments [see, for example, US Pat. No. 4,676,980 and Brennan et al. (1985) Science 229: 81]; the use of leucine zippers to produce bispecific antibodies [see, for example, Kostelny et al. (1992) J. Immunol., 148(5): 1547-1553]; the use of diabody technology to create bispecific antibody fragments [see, for example, Hollinger et al. (1993) Proc. Natl. Acad. Sci. USA, 90:6444-6448]; use of single-chain Fv (sFv) dimers [see, for example, Gruber et al. (1994) J. Immunol., 152:5368] and the production of trispecific antibodies (see, for example, Tutt et al. (1991) J. Immunol. 147: 60. Multispecific antibodies can be obtained in the form of full-length antibodies or antibody fragments. Invention also covers recombinant antibodies with three or more functional antigen-binding sites, for example, octopus antibodies [see, for example, US 2006/0025576A1].

In some embodiments, an antibody disclosed herein is a monoclonal antibody. The term monoclonal antibody refers to an antibody obtained from a population of essentially homogeneous antibodies, i.e. The individual antibodies that make up the population are identical and/or bind to the same epitope, with the exception of possible antibody variants, for example those containing natural mutations or those formed during the preparation of a monoclonal antibody preparation, these variants are usually present in small quantities. Unlike polyclonal antibody preparations, which typically include different antibodies directed against different epitopes, each monoclonal antibody in a monoclonal antibody preparation is directed against a single antigen determinant. Thus, the adjective monoclonal indicates the nature of the antibody obtained from an essentially homogeneous population of antibodies, and should not be construed as requiring that the antibody be produced by any particular method. For example, monoclonal antibodies for use in the present invention can be produced by a variety of methods, including, but not limited to, hybridoma technology, DNA recombination methods, phage display methods, and methods that employ transgenic animals containing all or a portion of human immunoglobulin loci. these and other exemplary methods for producing monoclonal antibodies are described herein.

For example, using activin-derived immunogens, antiserum or monoclonal antibodies to a protein or peptide can be generated according to standard protocols [see, for example, Antibodies: A Laboratory Manual ed. by Harlow and Lane (1988) Cold Spring Harbor Press: 1988]. It is possible to immunize a mammal, such as a mouse, hamster or rabbit, with an immunogenic form of an activin polypeptide, an antigenic fragment that is capable of eliciting a humoral response, or a fusion protein. Methods for rendering a protein or peptide immunogenic include conjugation with carriers or other methods also well known in the art. The immunogenic portion of the activin polypeptide can be administered in the presence of an adjuvant. The progress of immunization can be monitored by determining antibody titers in plasma or serum. Standard enzyme-linked immunosorbent assay (ELISA) or other immunological methods using an immunogen as the antigen can be used to assess the levels of antibody produced and/or binding affinity.

After immunization of an animal with an antigen preparation of activin, antiserum can be obtained and, if necessary, polyclonal antibodies can be isolated from the serum. To produce monoclonal antibodies, antibody-producing cells (lymphocytes) can be obtained from an immunized animal and fused with immortalizing cells, such as myeloma cells, using standard somatic cell fusion procedures to form hybrid cells. Such methods are well known in the art and include, for example, hybridoma technology [see, for example, Kohler and Milstein (1975) Nature, 256: 495-497], human B cell hybridoma technology [see, for example, Kozbar et al. (1983) Immunology Today, 4:72], and EBV-hybridoma technology for the production of human monoclonal antibodies [Cole et al. (1985) Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc. pp. 77-96]. It is possible to immunochemically screen hybridoma cells for the production of antibodies that specifically interact with the activin polypeptide, and isolate monoclonal antibodies from a culture containing such hybridoma cells.

In some embodiments, one or more amino acid modifications may be introduced into the Fc region of an antibody described herein to form a variant Fc region. The Fc region variant may comprise an Fc region sequence of human origin (eg, a human IgG1, IgG2, IgG3, or IgG4 Fc region) containing an amino acid modification (eg, substitution, deletion, and/or insertion) at one or more than one amino acid position.

For example, this disclosure provides a variant antibody that has some, but not all, effector functions that make it a suitable candidate for applications in which the in vivo half-life of the antibody is important, but some effector functions [eg, complement-dependent cellular cytotoxicity (CDC) ) and antibody-dependent cellular cytotoxicity (ADCC)] are optional or undesirable. In vitro and/or in vivo cytotoxicity studies can be performed to confirm the reduction/elimination of CDC and/or ADCC activities. For example, Fc receptor (FcR) binding assays can be performed for

- 46 043914 asserting that the antibody does not bind to FcyR (hence, in all likelihood, does not have ADCC activity), but retains the ability to bind to FcRn. The primary cells mediating ADCC, NK cells, express only FcyRIII, whereas monocytes express FcyRI, FcyRII, and FcyRIII. Data regarding FcR expression on hematopoietic cells are summarized, for example, by Ravetch and Kinet (1991) Annu. Rev. Immunol. 9:457-492. Non-exhaustive examples of in vitro studies to evaluate the ADCC activity of a molecule of interest are described in US Pat. No. 5,500,362; Hellstrom, I. et al. (1986) Proc. Natl. Acad. Sci. USA 83:7059-7063]; Hellstrom, I et al. (1985) Proc. Natl. Acad. Sci. USA 82:1499-1502; US patent 5821337; Bruggemann, M. et al. (1987) J. Exp. Med. 166:1351–1361. Alternatively, non-radioactive assays may be used (see, for example, the ACTI™ Non-Radioactive Flow Cytometry Assay (CellTechnology, Inc. Mountain View, Calif., and the CytoTox 96® Non-Radioactive Cytotoxicity Assay (Promega, Madison, Wis.). Suitable effector cells for such studies include peripheral blood mononuclear cells (PBMCs) and natural killer (NK) cells.Alternatively, or as a complement to ADCC, the activity of the molecule of interest can be assessed in vivo, for example, in an animal model such as described by Clynes et al (1998) Proc Natl Acad Sci USA 95:652-656 C1q binding can also be assayed to confirm that the antibody is unable to bind C1q and therefore does not have CDC activity [see eg, enzyme immunoassay for C1q and C3c binding in WO 2006/029879 and WO 2005/100402] To assess complement activation, a CDC assay can be performed [see, for example, Gazzano-Santoro et al., (1996) J. Immunol. Methods 202:163; Cragg, M. S. et al. (2003) Blood 101:1045–1052; and Cragg, M. S, and M. J. Glennie (2004) Blood 103:2738-2743]. You can also study the binding of FcRn and clearance/half-life in vivo using methods known in the art [see, for example, Petkova, S.B. et al. (2006) Intl. Immunol. 18(12): 1759-1769]. Antibodies of the invention with reduced effector function include those with substitution of one or more residues at positions 238, 265, 269, 270, 297, 327 and 329 of the Fc region (US Patent No. 6,737,056). Such Fc region mutants include Fc region mutants with substitutions of two or more amino acids at positions 265, 269, 270, 297 and 327, including the so-called DANA Fc mutants with substitutions of residues at positions 265 and 297 with alanine (US patent 7332581).

In some embodiments, it may be desirable to create cysteine-modified antibodies, for example, thioMAb, in which one or more residues of the antibody are replaced with cysteine residues. In specific embodiments, the replaced residues are located at accessible sites on the antibody. By replacing these residues with cysteine, reactive thiol groups are placed at accessible sites on the antibody and can be used to conjugate the antibody to other moieties, such as drug moieties or linker-drug moieties, to create an antibody conjugate, as described below. In some embodiments, cysteine may be replaced by any or all of the following residues: light chain V205 (Kabat numbering); A118 heavy chain (EU numbering) and S400 heavy chain (EU numbering) Fc region of the heavy chain. Cysteine-modified antibodies can be prepared, for example, as described in US Pat. No. 7,521,541.

In addition, the methods used to screen antibodies to detect the desired antibody may depend on the properties of the resulting antibody. For example, if the antibody will be used to bind an antigen in solution, it may be desirable to examine solution binding. A variety of methods exist to study the interaction between antibodies and antigens in order to detect specific antibodies of interest. Such methods include ELISA, surface plasmon resonance assays (eg, Biacore biosensor binding assay, Biacore AB, Uppsala, Sweden), sandwich assay (eg, paramagnetic particle system from IGEN International, Inc., Gaithersburg, Maryland), Western -blotting, immunoprecipitation studies and immunohistochemical studies.

In some embodiments, variations in the amino acid sequences of the antibodies and/or binding polypeptides provided herein are contemplated. For example, it may be desirable to improve the binding affinity and/or other biological properties of the antibody or binding polypeptide. Variants of the amino acid sequences of the antibody and/or binding polypeptides can be obtained by introducing appropriate modifications into the nucleotide sequence encoding the antibody and/or binding polypeptide, or by peptide synthesis. Such modifications include, for example, deletions and/or insertions and/or substitutions of residues within the amino acid sequences of the antibody and/or binding polypeptide. Any combination of deletions, insertions and substitutions can be made to produce the final construct, as long as the final construct has the desired characteristics, eg, target binding (eg, activin binding, such as activin E and/or activin C).

Changes (eg, substitutions) can be made within the HVR, for example, to improve the affinity of the antibody. Such changes can be made at HVR hotspots, i.e. in residues encoded by codons, which undergo mutation with high frequency during somatic maturation

- 47 043914 tations [see, for example, Chowdhury (2008) Methods Mol. Biol. 207:179-196 (2008)] and/or SDRs (a-CDRs), with the resulting VH or VL variant tested for binding affinity. Affinity maturation by design and re-selection from secondary libraries is described in the art [see, for example, Hoogenboom et al., in Methods in Molecular Biology 178:1-37, O'Brien et al., ed., Human Press, Totowa, N. J., (2001). In some embodiments, affinity maturation is achieved by increasing the diversity of variable genes selected for affinity maturation by any of existing methods (eg, reduced fidelity PCR, combining light and heavy chain variable domains, or oligonucleotide-directed mutagenesis). After this, a secondary library is created. The library is then screened to detect any antibody variants with the required affinity. Another way to increase diversity involves methods targeting HVR variation that randomize multiple HVR residues (e.g., 4–6 residues at a time). HVR residues involved in antigen binding can be specifically identified, for example, using alanine scanning mutagenesis or modeling. In particular, CDR-H3 and CDR-L3 are often subject to changes.

In some embodiments, substitutions, insertions or deletions may occur within one or more than one HVR, provided that such changes do not significantly reduce the ability of the antibody to bind antigen. For example, conservative substitutions in HVR (eg, the conservative substitutions presented herein) can be made that do not significantly reduce binding affinity. Such changes may be outside of the HVR or SDR hotspots. In some embodiments, each HVR is either left unchanged or contains no more than one, two, or three amino acid substitutions within the VH and VL sequence variants above.

An effective method for detecting residues or regions of an antibody and/or binding polypeptide that may be targets for site-directed mutagenesis is so-called alanine scanning mutagenesis, described by Cunningham and Wells (1989) Science, 244: 1081-1085. In this method, an amino acid residue or group of target residues (for example, charged residues such as Arg, Asp, His, Lys and Glu) are detected and replaced with a neutral or negatively charged amino acid (for example, alanine or polyalanine) to determine whether the interaction will change antibodies with antigen. If the initial amino acid substitution changes the functional properties of the molecule, further substitutions can be made at these positions. Alternatively or in addition, the crystal structure of the antigen-antibody complex can be analyzed to detect points of contact between the antibody and the antigen. Such residues involved in contacts, as well as residues adjacent to them, may or may not be selected as candidates for substitutions. Variants can be screened to determine whether they have the required properties.

Amino acid sequence insertions include amino- and/or carboxy-terminal fusions ranging in length from a single residue to polypeptides containing hundreds or more residues, as well as insertions of one or more than one amino acid residue within a sequence. Examples of terminal inserts include an antibody with an N-terminal methionyl residue. Other variants of the insertion antibody molecule include fusing the N- or C-terminus of the antibody with an enzyme (eg, ADEPT) or polypeptide that increases the serum half-life of the antibody.

In some embodiments, the antibody and/or binding polypeptide provided herein may also be modified to introduce additional non-protein moieties known in the art and readily available. Moieties suitable for derivatization of the antibody and/or binding polypeptide include, but are not limited to, water-soluble polymers. Non-exhaustive examples of water-soluble polymers include polyethylene glycol (PEG), ethylene glycol-propylene glycol copolymers, carboxymethylcellulose, dextran, polyvinyl alcohol, polyvinylpyrrolidone, poly-1,3-dioxolane, poly-1,3,6trioxane, ethylene-maleic anhydride copolymer, polyamino acids ( both homopolymers and random copolymers) and dextran or poly(p-vinyl pyrrolidone) polyethylene glycol, propylene glycol homopolymers, propylene oxide-ethylene oxide copolymers, polyoxyethylated polyols (eg, glycerol), polyvinyl alcohol, and mixtures thereof, but are not limited to. Polyethylene glycol propionaldehyde may have manufacturing advantages due to its stability in water. The polymer can have any molecular weight and can be branched or unbranched. The number of polymers attached to the antibody and/or binding polypeptide may vary, and if more than one polymer is attached, their molecules may be identical or different. In general, the number and/or type of polymers used for derivatization may be determined based on what specific properties or functions of the antibody and/or binding polypeptide need to be improved and whether the derivative of the antibody and/or binding polypeptide will be used in therapy under certain conditions, etc., but not limited to these considerations.

- 48 043914

D. Low molecular weight antagonists.

In other aspects, the ActRIIB antagonist for use according to the methods and uses described herein is a small molecule (small molecule ActRIIB antagonist) or a combination of small molecule antagonists. A small molecule ActRIIB antagonist or combination of small molecule antagonists can bind, for example, to one or more than one ActRIIB ligand (e.g., activin, GDF11, GDF8, GDF3, BMP6, and/or BMP10), receptor, type I receptor (e.g., ALK4, ALK5, and /or ALK7), type II receptor (eg ActRIIB) and/or co-receptor. In some embodiments, a small molecule ActRIIB antagonist or combination of small molecule antagonists inhibits signal transduction mediated by one or more ActRIIB ligands, for example, as determined by a cell assay such as those described herein. As described herein, a small molecule ActRIIB antagonist can be used alone or in combination with one or more maintenance therapies or one or more active agents to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis, in particular the treatment , preventing or reducing the rate of progression and/or severity of one or more than one complication of myelofibrosis (eg, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis) and/or treating a patient treated with a Janus kinase inhibitor.

In some embodiments, an ActRIIB small molecule antagonist or combination of small molecule antagonists inhibits at least GDF11, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, a small molecule antagonist of ActRIIB or a combination of small molecule antagonists inhibits at least GDF8, possibly additionally inhibiting one or more one of the following: GDF11, activin (for example, activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and/or activin BE), GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, the small molecule ActRIIB antagonist or combination of small molecule antagonists inhibits at least activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and/or activin BE), optionally additionally inhibiting one or more of the following: GDF8, GDF11, GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, the small molecule ActRIIB antagonist or combination of small molecule antagonists inhibits at least activin B, possibly additionally inhibiting one or more of the following: GDF8, GDF11, GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, the small molecule ActRIIB antagonist or combination of small molecule antagonists inhibits at least BMP6, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), GDF3, GDF11, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB small molecule antagonist or combination of small molecule antagonists inhibits at least GDF3, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP15, BMP6, GDF11, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, a small molecule ActRIIB antagonist or combination of small molecule antagonists inhibits at least BMP10, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP15, BMP6, GDF11, GDF3, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, a small molecule antagonist of ActRIIB or a combination of small molecule antagonists inhibits at least ActRIIB, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP15, BMP6, GDF11, GDF3, BMP10, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB small molecule antagonist or combination of small molecule antagonists inhibits at least ALK4, optionally further inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP15, BMP6, GDF11, GDF3, ActRIIB, BMP10, ALK5 and ALK7. In some embodiments, an ActRIIB small molecule antagonist or combination of small molecule antagonists inhibits at least ALK5, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP15, BMP6, GDF11, GDF3, ActRIIB, ALK4, BMP10 and ALK7. In some embodiments, an ActRIIB small molecule antagonist or combination of small molecule antagonists inhibits at least ALK7, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin

- 49 043914

BE), BMP15, BMP6, GDF11, GDF3, ActRIIB, ALK4, ALK5 and BMP10. In some embodiments, the small molecule ActRIIB antagonist or combination of small molecule antagonists described herein does not or substantially does not inhibit BMP9. In some embodiments, the small molecule ActRIIB antagonist or combination of small molecule antagonists described herein does not or substantially does not inhibit activin A.

The small molecule ActRIIB antagonist may be a direct or indirect inhibitor. For example, a small molecule antagonist or combination of small molecule antagonists can inhibit the expression (e.g., transcription, translation, cellular secretion, or combinations thereof) of at least one or more than one ActRIIB ligand [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and/or activin BE), GDF11, BMP10, BMP9, BMP6, BMP5, GDF3 and/or GDF8], type I receptor (for example, ALK4, ALK5 and/or ALK7), a type II receptor (eg, ActRIIB), a co-receptor, and/or one or more downstream components of the ActRIIB signaling pathway (eg, Smad). Alternatively, a small molecule antagonist or combination of small molecule antagonists can directly bind to and inhibit, for example, one or more ActRIIB ligands [e.g., activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, active AE and/or activin BE), GDF11, BMP10, BMP9, BMP6, BMP5, GDF3 and/or GDF8], type I receptor (for example, ALK4, ALK5 and/or ALK7), type II receptors (for example, ActRIIB ), a co-receptor and/or one or more downstream components of the ActRIIB signaling pathway (eg, Smad). Combinations of one or more indirectly acting small molecule ActRIIB antagonists can be used according to the methods described herein.

The binding small molecule antagonists of this invention can be determined and synthesized chemically using known methods (see, for example, PCT publication numbers WO 00/00823 and WO 00/39585). In general, the small molecule antagonists of the invention typically have a size of less than about 2000 daltons, alternatively, a size of less than about 1500, 750, 500, 250 or 200 daltons, such organic small molecule compounds being capable of binding, preferably specifically, to a polypeptide described in this document. Such small molecule antagonists can be identified without unnecessary experimentation using routine methods. In this regard, it should be noted that techniques for screening organic small molecule libraries for molecules capable of binding to a target polypeptide are well known in the art (see, for example, international publications WO00/00823 and WO00/39585).

Binding low molecular weight compounds herein may be, for example, aldehydes, ketones, oximes, hydrazones, semicarbazones, carbazides, primary amines, secondary amines, tertiary amines, N-substituted hydrazines, hydrazides, alcohols, ethers, thiols, thioethers, disulfides, carboxylic acids, esters, amides, ureas, carbamates, carbonates, ketals, thioketals, acetals, thioacetals, aryl halides, arylsulfonates, alkyl halides, alkyl sulfonates, aromatic compounds, heterocyclic compounds, anilines, alkenes, alkynes, diols, amino alcohols, oxazolidines, oxazolines, thiazolidines , thiazolines, enamines, sulfonamides, epoxides, aziridines, isocyanates, sulfonyl chlorides, diazo compounds and acid chlorides.

D. Polynucleotide antagonists.

In other aspects, the ActRIIB antagonist for use in the methods and applications described herein is a polynucleotide (ActRIIB antagonist polynucleotide) or a combination of polynucleotides. An ActRIIB polynucleotide antagonist or a combination of polynucleotide antagonists can inhibit, for example, one or more ActRIIB ligands (eg, activin, GDF11, GDF8, GDF3, BMP6 and/or BMP10), type I receptors (eg, ALK4, ALK5 and/or ALK7 ), type II receptors (eg ActRIIB), a co-receptor and/or a downstream component of the signaling pathway (eg Smad). In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits signal transduction mediated by one or more ActRIIB ligands, for example, as determined by a cell assay such as those described herein. As described herein, an ActRIIB polynucleotide antagonist can be used alone or in combination with one or more maintenance therapies or one or more active agents to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis, in particular the treatment , preventing or reducing the rate of progression and/or severity of one or more than one complication of myelofibrosis (eg, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis) and/or treating a patient treated with a Janus kinase inhibitor.

In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least GDF11, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or a combination of polynucleotide antagonists

- 50 043914 inhibits at least GDF8, possibly additionally inhibiting one or more of the following: GDF11, activin (for example, activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and/or activin BE), GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and/or activin BE), optionally additionally inhibiting one or more of the following: GDF8, GDF11, GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least activin B, possibly additionally inhibiting one or more of the following: GDF8, GDF11, GDF3, BMP6, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least BMP6, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), GDF3, GDF11, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least GDF3, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP6, GDF11, BMP10, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least BMP10, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP6, GDF11, GDF3, ActRIIB, ALK4, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least ActRIIB, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP6, GDF11, GDF3, BMP10, ALK4, ALK5 and ALK7. In some embodiments, a GDF/BMP polynucleotide antagonist, an ActRIIB polynucleotide antagonist, or a combination of polynucleotide antagonists inhibits at least ALK4, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC, activin BC, activin AE and/or activin BE), BMP6, GDF11, GDF3, ActRIIB, BMP10, ALK5 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least ALK5, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP6, GDF11, GDF3, ActRIIB, ALK4, BMP10 and ALK7. In some embodiments, an ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists inhibits at least ALK7, possibly additionally inhibiting one or more of the following: GDF8, activin (e.g., activin A, activin B, activin C, activin E, activin AB, activin AC , activin BC, activin AE and/or activin BE), BMP6, GDF11, GDF3, ActRIIB, ALK4, ALK5 and BMP10. In some embodiments, the ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists described herein does not or substantially does not inhibit BMP9. In some embodiments, the ActRIIB polynucleotide antagonist or combination of polynucleotide antagonists described herein does not or substantially does not inhibit activin A.

In some embodiments, the polynucleotide antagonist of the invention may be an antisense nucleic acid, an RNAi molecule [eg, small interfering RNA (siRNA), small hairpin RNA (shRNA), microRNA (miRNA)], an aptamer, and/or a ribozyme. The nucleic acids and amino acid sequences of human GDF11, GDF8, activin (Activin A, Activin B, Activin C and Activin E), BMP6, GDF3, ActRIIB, ALK4, ALK5, ALK7 and BMP10 are known in the art. In addition, various methods for creating polynucleotide antagonists are well known in the art. Thus, polynucleotide antagonists for use in accordance with this description can be prepared routinely by one skilled in the art based on knowledge known in the art and the information provided herein.

Antisense technology can be used to control gene expression through antisense DNA or RNA or through triple helix formation. Antisense technologies are discussed, for example, Okano (1991) J. Neurochem. 56:560; Oligodeoxynucleotides as Antisense Inhibitors of Gene Expression, CRC Press, Boca Raton, Fla. (1988). Triple helix formation is discussed, for example, by Cooney et al. (1988) Science 241:456; and Dervan et al., (1991) Science 251:1300. These methods are based on the binding of a polynucleotide to complementary DNA or RNA. In some embodiments, the antisense nucleic acids comprise a single-stranded RNA or DNA sequence that is complementary to at least the RNA portion of a gene transcript described herein. In this case, absolute complementarity is preferable, but not mandatory.

A sequence complementary to at least a portion of an RNA, as used herein, means a sequence having sufficient complementarity to hybridize to the RNA,

- 51 043914 forming a stable duplex; in the case of double-stranded antisense nucleic acids of the gene described herein, one strand of a DNA duplex can be examined in this manner, or triplex formation can be assessed. The ability to hybridize will depend on both the degree of complementarity and the length of the antisense nucleic acids. In general, the longer the nucleic acid that hybridizes, the more RNA base mismatches it can contain and still form a stable duplex (or, under appropriate circumstances, a triplex). One skilled in the art can determine the acceptable degree of mismatch by using standard procedures for determining the melting point of the hybridizing complex.

Polynucleotides that are complementary to the 5' end of the transcript, such as the 5' untranslated sequence up to and including the AUG start codon, will be most effective in inhibiting translation. However, it has been shown that sequences complementary to the 3' untranslated sequences of mRNA are also effective in inhibiting the translation of mRNA [see, for example, Wagner, R., (1994) Nature 372:333-335]. Thus, oligonucleotides complementary to either the 5' or 3' untranslated non-coding regions of the gene of the invention can be used in antisense technology to inhibit the translation of endogenous mRNA. Polynucleotides complementary to the 5' untranslated region of the mRNA must include the complement of the AUG start codon. Antisense polynucleotides complementary to the coding regions of the mRNA are less effective translation inhibitors but can be used according to the methods of the present invention. Whether designed to hybridize to the 5', 3', or coding region of the mRNA of the invention, antisense nucleic acids should be at least 6 nucleotides in length and preferably be oligonucleotides ranging in length from 6 to about 50 nucleotides. In certain aspects, the oligonucleotide consists of at least 10 nucleotides, at least 17 nucleotides, at least 25 nucleotides, or at least 50 nucleotides.

In one embodiment, the antisense nucleic acid of this invention is produced intracellularly by transcription of an exogenous sequence. For example, a vector or portion thereof is transcribed to produce an antisense nucleic acid (RNA) gene of the invention. Such a vector will contain a sequence encoding the desired antisense nucleic acid. Such a vector may remain episomal or integrate into the chromosome, provided that it can be transcribed to produce the desired antisense RNA. Such vectors can be constructed using recombinant DNA technology known in the art. Vectors may be plasmid, viral, or others known in the art, used for replication and expression in vertebrate cells. Expression of the sequence encoding the necessary genes of this invention or fragments thereof can be mediated by any promoter known in the art operating in vertebrate cells, preferably human cells. Such promoters may be inducible or constitutive. Such promoters include, but are not limited to, the SV40 early promoter region [see, for example, Benoist and Chambon (1981) Nature 290:304-310], the promoter contained in the long 3' terminal repeat of Rous sarcoma virus [see, for example, Yamamoto et al. (1980) Cell 22:787-797], thymidine kinase promoter of herpes virus [see, for example, Wagner et al. (1981) Proc. Natl. Acad. Sci. U.S.A. 78:1441-1445] and metallothionein gene regulatory sequences [see, for example, Brinster, et al. (1982) Nature 296:39-42].

In some embodiments, the polynucleotide antagonists are RNA interfering (RNAi) molecules that target the expression of one or more of the following: GDF11, GDF8, activin (Activin A, Activin B, Activin C, and Activin E), BMP6, ActRIIB, GDF3, ALK4 , ALK5, ALK7 and BMP10. RNA interference refers to the expression of RNA interfering with the expression of a target RNA. In particular, an interfering RNA molecule silences target genes by interacting with a specific mRNA indirectly through siRNA (small interfering RNA). The double-stranded RNA complex is then sent to the cell for degradation. The siRNA molecule is a double-stranded RNA duplex of 10 to 50 nucleotides in length that interferes with the expression of a target gene that is sufficiently complementary (eg, at least 80% identical to the gene). In some embodiments, the siRNA molecule contains a nucleotide sequence that is at least 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identical to the nucleotide sequence of the target gene.

Additional interfering RNA molecules include short hairpin RNA (shRNA); as well as short interfering hairpin RNA and microRNA (miRNA). The shRNA molecule contains the sense and antisense sequences of the target gene, connected by a loop. ShRNA is transported from the nucleus to the cytoplasm and undergoes degradation along with mRNA. Pol III or U6 promoters can be used to express RNA into interfering RNA molecules. Paddison et al. [Genes & Dev. (2002) 16:948–958, 2002] used hairpin-folded small RNA molecules as a means of influencing RNAi. Accordingly, such short hairpin RNA (shRNA) molecules are also useful in the methods described herein. The stem and loop lengths of functional shRNAs vary; trunk length varies from approximately 25 to approximately 30

- 52,043,914 nucleotides, and the loop size can vary from 4 to approximately 25 nucleotides without affecting the silencing activity. Without wishing to be bound by any particular theory, it is believed that such shRNAs are similar to the double-stranded RNA (dsRNA) products of the DICER RNase and, in any case, have the same ability to inhibit the expression of a particular gene. Expression of shRNA can be achieved by a lentiviral vector. MicroRNAs are single-stranded RNAs, approximately 10 to 70 nucleotides in length, that are initially transcribed as pre-miRNAs characterized by a stem-loop structure and then processed into mature miRNAs followed by RISC assembly.

Molecules that mediate RNA interference, including but not limited to siRNA, can be produced in vitro by chemical synthesis (Hohjo, FhEBS Lett 521:195-199, 2002), dsRNA hydrolysis (Yang et al, Proc Natl Acad Sci USA 99: 9942-9947, 2002), via in vitro transcription by T7 RNA polymerase (Donzeet et al, Nucleic Acids Res 30:e46, 2002; Yu et al, Proc Natl Acad Sci USA 99:6047-6052, 2002) and hydrolysis double-stranded RNA using a nuclease such as E. coli RNase III (Yang et al., Proc Natl Acad Sci USA 99:9942-9947, 2002).

According to another aspect of the invention, there is provided a polynucleotide antagonist, including, without limitation, DNA decoys, dsDNA, single-stranded DNA, complexed DNA, encapsulated DNA, viral DNA, plasmid DNA, naked RNA, encapsulated RNA, viral RNA, double-stranded RNA, molecules, capable of causing RNA interference or combinations thereof.

In some embodiments, the polynucleotide antagonists of the invention are aptamers. Aptamers are nucleic acid molecules, including double-stranded DNA and single-stranded RNA molecules, that bind and form tertiary structures that specifically bind to a target molecule. The creation and use of aptamers for therapeutic purposes is well known in the art (see, for example, US Pat. No. 5,475,096). Additional information on aptamers can be found in published patent application US 20060148748. Nucleic acid aptamers are selected by methods known in the art, for example, using the process of systematic evolution of ligands by exponential enrichment (SELEX). SELEX is a method for the evolution of nucleic acid molecules in vitro through highly specific binding to target molecules, as described, for example, in US patents 5475096; 5580737; 5567588; 5707796; 5763177; 6011577 and 6699843. Another screening method for the detection of aptamers is described in US patent 5270163. The SELEX method is based on the ability of nucleic acids to form a variety of two- and three-dimensional structures, as well as the chemical diversity of nucleotide monomers acting as ligands (forming specific binding pairs) with virtually any chemical compound, whether monomeric or polymeric, including other nucleic acid molecules and polypeptides. The targets can be molecules of any size or composition. The SELEX method involves selection from a mixture of candidate oligonucleotides and stepwise iterations of binding, separation and amplification using the same general selection scheme to achieve the required binding affinity and selectivity. Taking as a starting point a mixture of nucleic acids, which may contain a segment of randomized sequence, the SELEX method includes the steps of bringing the mixture into contact with a target under conditions conducive to binding; separation of unbound nucleic acids and nucleic acids specifically bound to target molecules; dissociation of nucleic acid-target complexes; amplification of nucleic acids dissociated from target nucleic acid complexes to produce a mixture of nucleic acids enriched with ligands. The binding, separation, dissociation and amplification steps are repeated as many rounds as necessary to produce nucleic acid ligands that bind with high affinity and specificity to the target molecule.

Typically, such binding molecules are administered separately to the animal [see, for example, O'Connor (1991) J. Neurochem. 56:560], however, such binding molecules can also be expressed in vivo from polynucleotides taken up by the host cell and expressed in vivo [see, for example, Oligodeoxynucleotides as Antisense Inhibitors of Gene Expression, CRC Press, Boca Raton, Fla. (1988)].

E. Follistatin and FLRG antagonists.

In other aspects, the ActRIIB antagonist is follistatin or a FLRG polypeptide. As described herein, follistatin and/or FLRG polypeptides can be used alone or in combination with one or more maintenance therapies or one or more active agents to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis, in in particular, treating, preventing or reducing the rate of progression and/or severity of one or more than one complication of myelofibrosis (eg, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis) and/or treating a patient treated with a Janus kinase inhibitor.

The term follistatin polypeptide includes polypeptides containing any naturally occurring follistatin polypeptide, as well as any variants thereof (including mutants, fragments, fusion proteins and peptidomimetics) that retain beneficial activity, and also covers any

- 53 043914 functional monomers or multimers of follistatin. In some preferred embodiments, the follistatin polypeptides of the invention bind and/or inhibit the activity of activin, in particular activin A. Variants of follistatin polypeptides that retain activin binding properties can be identified based on previous studies of follistatin and activin interactions. For example, WO2008/030367 describes follistatin specific domains (FSDs) that are important for activin binding. As shown below in SEQ ID NO: 65-67, the N-terminal domain of follistatin (FSND SEQ ID NO: 65), FSD2 (SEQ ID NO: 67) and to a lesser extent FSD1 (SEQ ID NO: 66) are shown in as an example, domains of follistatin, which are important for activin binding. In addition, methods for generating and testing polypeptide libraries with respect to ActRII polypeptides are described above, and such methods also relate to creating and testing follistatin variants. Follistatin polypeptides include polypeptides derived from the sequence of any known follistatin, having a sequence that is at least 80% identical to the sequence of follistatin, and possibly at least 85%, 90%, 95%, 96%, 97%, 98% identical. , 99% or more. Examples of follistatin polypeptides include mature follistatin polypeptides or shorter isoforms or other variants of the human follistatin polypeptide precursor (SEQ ID NO: 63) as described, for example, in WO2005/025601.

The human follistatin polypeptide precursor isoform FST344 is given below:

MVRARHQPGG LCLLLLLLCQ FMEDRSAQAG NCWLRQAKNG RCQVLYKTEL

SKEECCSTGR LSTSWTEEDV NDNTLFKWMI FNGGAPNCIP CKETCENVDC

101 GPGKKCRMNK KNKPRCVCAP DCSNITWKGP VCGLDGKTYR NECALLKARC

151 KEQPELEVQY QGRCKKTCRD VFCPGSSTCV VDQTNNAYCV TCNRICPEPA

201 SSEQYLCGND GVTYSSACHL RKATCLLGRS IGLAYEGKCI KAKSCEDIQC

251 TGGKKCLWDF KVGRGRCSLC DELCPDSKSD EPVCASDNAT YASECAMKEA

301 ACSSGVLLEV KHSGSCNSIS EDTEEEEEDE DQDYSFPISS ILEW (SEQ GO NO: 63; NCBI Reference No. NP_O3754L1)

The signal peptide is underlined, and the last 27 residues are also underlined, representing the C-terminal extension that distinguishes this follistatin isoform from the shorter follistatin isoform FST317 below.

The human follistatin polypeptide precursor isoform FST317 is given below:

MVRARHQPGG LCLLLLLLCQ FMEDRSAQAG NCWLRQAKNG RCQVLYKTEL

SKEECCSTGR LSTSWTEEDV NDNTLFKWMI FNGGAPNCIP CKETCENVDC

101 GPGKKCRMNK KNKPRCVCAP DCSNITWKGP VCGLDGKTYR NECALLKARC

151 KEQPELEVQY QGRCKKTCRD VFCPGSSTCV VDQTNNAYCV TCNRICPEPA

201 SSEQYLCGND GVTYSSACHL RKATCLLGRS IGLAYEGKCI KAKSCEDIQC

251 TGGKKCLWDF KVGRGRCSLC DELCPDSKSD EPVCASDNAT YASECAMKEA

301 ACSSGVLLEV KHSGSCN (SEQ ID NO: 64; NCBI Reference No. NP_006341.1)

The signal peptide is underlined.

The sequence of the follistatin N-terminal domain (FSND) is given below:

GNCWLRQAKNGRCQVLYKTELSKEECCSTGRLSTSWTEEDVNDNTLFKWMI

FNGGAPNCIPCK (SEQ GO NO: 65; FSND)

The sequences of FSD1 and FSD2 are given below:

ETCENVDCGPGKKCRMNKKNKPRCV (SEQ ID NO: 66; FSD1)

KTCRDVFCPGSSTCVVDQTNNAYCVT (SEQ ID NO: 67; FSD2)

In another aspect, the agent for use according to the methods set forth herein is follistatin-like gene (FLRG), also known as follistatin-like protein 3 (FSTL3). The term FLRG polypeptide includes polypeptides containing any naturally occurring FLRG polypeptide, as well as any variants thereof (including mutants, fragments, fusion proteins and peptidomimetics) that retain beneficial activity. In some preferred embodiments, the FLRG polypeptides of the invention bind and/or inhibit the activity of activin, in particular activin A. Variants of FLRG polypeptides that retain activin binding properties can be identified by routine methods of studying FLRG and activin interactions (see, for example, US 6,537,966) . In addition, methods for generating and testing polypeptide libraries with respect to ActRII polypeptides are described above, and such methods also relate to generating and testing FLRG variants. FLRG polypeptides include polypeptides derived from the sequence of any known FLRG, having a sequence that is at least 80% identical to the FLRG sequence, and possibly at least 85%, 90%, 85%, 90%, 95%, 97% identical. , 99% or more.

- 54 043914

The human FLRG polypeptide precursor (follistatin-like protein 3 precursor) is given below:

MRPGAPGPLW PLPWGALAWA VGFVSSMGSG NPAPGGVCWL QQGQEATCSL

VLQTDVTRAE CCASGNIDTA WSNLTHPGNK INLLGFLGLV HCLPCKDSCD

101 GVECGPGKAC RMLGGRPRCE CAPDCSGLPA RLQVCGSDGA TYRDECELRA

151 ARCRGHPDLS VMYRGRCRKS CEHVVCPRPQ SCVVDQTGSA HCVVCRAAPC

201 PVPSSPGQEL CGNNNVTYIS SCHMRQATCF LGRSIGVRHA GSCAGTPEEP

251 PGGESAEEEE NFV (SEQ Ш NO: 68; NCBI Reference No. NP 005851.1)

The signal peptide is underlined.

In some embodiments, functional variants or modified forms of follistatin polypeptides and FLRG polypeptides include fusion proteins having at least a portion of a follistatin polypeptide or FLRG polypeptide and one or more fusion domains, such as, for example, domains that facilitate isolation, detection, stabilization, or multimerization of the polypeptide . Suitable fusion domains are discussed in detail above in relation to ActRIIB polypeptides. In some embodiments, the antagonist agent of the invention is a fusion protein comprising the activin binding moiety of a follistatin polypeptide fused to an Fc domain. In another embodiment, the antagonist agent of the invention is a fusion protein comprising the activin binding portion of a FLRG polypeptide fused to an Fc domain.

3. Screening studies.

In some aspects, this disclosure relates to the use of the subject ActRIIB polypeptides and variants thereof (eg, GDF8 decoys) for the discovery of compounds (agents) that are agonists or antagonists of ActRIIB polypeptides. Compounds discovered through such screening can be tested to assess their ability to treat myelofibrosis, for example in animal models.

There are many approaches to screen therapeutic agents for the treatment of myelofibrosis by targeting the ActRIIB signaling pathway (eg, Smad signaling). In some embodiments, high-throughput screening of compounds can be performed to detect agents that disrupt ActRIIB-mediated effects on a selected cell line. In some embodiments, the assay is conducted to screen for and discover compounds that specifically inhibit or reduce the binding of an ActRIIB polypeptide to its binding partner, such as an ActRIIB ligand (e.g., activin A, activin B, activin AB, activin C, GDF8, GDF3, GDF11, or VMR10). Alternatively, the assay can be used to detect compounds that enhance the binding of an ActRIIB polypeptide to its binding partner, such as an ActRIIB ligand. In a further embodiment, compounds can be detected by their ability to interact with an ActRIIB polypeptide.

There is a sufficient number of studies of various formats and in light of this invention, even those that are not specifically described herein will nevertheless be understandable to one skilled in the art. As described herein, the test compounds (agents) of the invention can be created by any combinatorial chemical method. Alternatively, the compounds discussed may be naturally occurring biomolecules synthesized in vivo or in vitro. The compounds (agents) that will be tested for their ability to act as tissue growth modulators may be produced by, for example, bacteria, yeast, plants or other organisms (e.g., naturally occurring products), may be produced by chemical means (e.g., low molecular weight compounds, including peptidomimetics) or obtained by recombinant methods. The compounds of interest contemplated by this invention include non-peptidyl organic molecules, peptides, polypeptides, peptidomimetics, sugars, hormones and nucleic acid molecules. In some embodiments, the test agent is a low molecular weight organic compound having a molecular weight of less than about 2000 Da.

The test compounds of the invention may be presented as individual single objects or presented as libraries of greater complexity, such as those obtained by combinatorial chemistry methods. Such libraries may contain, for example, alcohols, alkyl halides, amines, amides, esters, aldehydes, ethers and other classes of organic compounds. The test compounds can enter the test system either in isolated form or in the form of mixtures of compounds, in particular at the initial stages of screening. It is possible that compounds can be derivatized with other compounds and have derivatizing groups to facilitate isolation of the compounds. Non-exhaustive examples of derivatizing groups include biotin, fluorescein, digoxigenin, green fluorescent protein, isotopes, polyhistidine, magnetic particles, glutathione Stransferase (GST), photoactivatable cross-linkers, or any combination thereof.

In many drug screening programs that examine compound libraries and naturally occurring extracts, high-throughput assays are desired

- 55 043914 to maximize the number of connections monitored over a given period of time. Assays that are performed in cell-free systems, such as those that can be performed with purified or semi-purified proteins, are often preferred as primary screens because they can be designed to provide rapid and relatively simple detection of a change in a target molecule mediated by the compound of interest. Additionally, in an in vitro system, effects of cellular toxicity or bioavailability of the test compound can generally be neglected, instead focusing the study primarily on the effects of the drug on the target molecule, which may be manifested by changes in binding affinity between the ActRIIB polypeptide and its binding partner (eg ActRIIB ligand).

For purposes of illustration only, in an exemplary screening study herein, a compound of interest is contacted with an isolated and purified ActRIIB polypeptide that is typically capable of binding to an ActRIIB ligand relevant to the purposes of the study. The mixture of compound and ActRIIB polypeptide is then added to a composition containing an ActRIIB ligand (eg, GDF11). Detection and quantitation of ActRIIB/ActRIIB ligand complexes provides a means for determining the effectiveness of a compound in inhibiting (or potentiating) the formation of a complex between an ActRIIB polypeptide and its binding protein. The potency of a compound can be assessed by constructing concentration response curves based on data obtained at various concentrations of the test compound. In addition, a control study can also be performed to provide a baseline against which to compare. For example, in a control study, isolated and purified ActRIIB ligand is added to a composition containing an ActRIIB polypeptide and the formation of an ActRIIB/ActRIIB ligand complex is quantified in the absence of the test compound. It should be understood that the order in which the components of the reaction mixture are added may vary and they may be added simultaneously. In addition, instead of purified proteins, cell extracts and lysates can be used to obtain a suitable cell-free test system.

The formation of a complex between the ActRIIB polypeptide and its binding protein can be detected in various ways. For example, quantification of changes in complex formation can be carried out, for example, using proteins bearing a detectable label, such as radiolabeled (eg, ³² P, ³⁵ S, ¹⁴ C or ³ H), fluorescently labeled (eg, FITC), enzymatically labeled ActRIIB polypeptide and/or its binding protein, immunologically or chromatographically detectable.

Some embodiments of this disclosure provide for the use of a fluorescence polarization technique and a fluorescence resonance energy transfer (FRET) technique in measuring, directly or indirectly, the degree of interaction between an ActRIIB polypeptide and its binding protein. In addition, detection methods such as those based on optical waveguides (see, for example, PCT Publication WO 96/26432 and US Pat. No. 5,677,196), surface plasmon resonance (SPR), surface charge detectors, and surface force detectors are compatible with many embodiments of the invention. .

In addition, the present invention provides the use of a protein-protein interaction detection method, also known as a two-hybrid assay, to detect agents that disrupt or potentiate the interaction between an ActRIIB polypeptide and its binding partner; see, for example, US Pat. No. 5,283,317; Zervos et al. (1993) Cell 72:223-232; Madura et al. (1993) J Biol Chem 268:12046–12054; Bartel et al. (1993) Biotechniques 14:920-924; and Iwabuchi et al. (1993) Oncogene 8:1693-1696). A specific embodiment of the present invention provides for the use of a reverse two-hybrid system for the detection of compounds (eg, small molecules or peptides) that uncouples the interaction between an ActRII polypeptide or GDF decoy and its binding protein [see, for example, Vidal and Legrain, (1999) Nucleic Acids Res 27:919-29; Vidal and Legrain, (1999) Trends Biotechnol 17:374-81 and US patents 5525490; 5955280 and 5965368].

In some embodiments, the subject compounds are detected by their ability to interact with an ActRIIB polypeptide. The interaction between the compound and the ActRIIB polypeptide may be covalent or non-covalent. For example, such interactions can be detected at the protein level using in vitro biochemical techniques, including light-induced cross-linking, radiolabeled ligand binding, and affinity chromatography [see, for example, Jakoby WB et al. (1974) Methods in Enzymology 46:1]. In some cases, screening of compounds can be carried out in studies based on mechanism, such as studies to detect compounds that bind to the ActRIIB polypeptide. These may involve binding occurring in the solid phase or in the liquid phase. Alternatively, one can transfect a cell with a gene encoding an ActRIIB polypeptide along with a reporter system (eg, β-galactosidase, luciferase, or green fluorescent protein) and screen against the library, preferably a high-throughput screen, or individual members of the library. Other studies that have a mechanism as their basis can be used, e.g.

- 56 043914 in which changes in free energy are detected. Binding studies can be performed with a target fixed to a well, particle or chip, or captured by an immobilized antibody, or separation by capillary electrophoresis. Bound compounds can be identified using endpoint colorimetric or fluorescence measurements or surface plasmon resonance.

4. Examples of application in therapy.

As described in the examples provided herein, it has been discovered that an ActRIIB antagonist (inhibitor) can be used to treat patients with myelofibrosis, in particular, improve various complications of the disease, including, for example, splenomegaly, extramedullary hematopoiesis and fibrosis. In particular, the data presented indicate that the GDF decoy polypeptide reduces splenomegaly, extramedullary hematopoiesis, and fibrosis in the JAK2V617F model of myelofibrosis. Accordingly, in some aspects, the disclosure relates to compositions and methods for treating myelofibrosis, particularly for treating or preventing one or more complications of myelofibrosis (eg, splenomegaly, extramedullary hematopoiesis, anemia and fibrosis), by administering to a patient in need thereof an effective amount one or more than one ActRIIB antagonist, optionally in combination with one or more maintenance treatments or other active agent for the treatment of myelofibrosis.

As used herein, the term therapeutic that prevents a disorder or condition refers to a compound that reduces the incidence of a disorder or condition in a statistical sample compared to an untreated control sample or delays the onset or reduces the severity of one or more than one symptom of a disorder or condition compared to untreated control sample. The term treatment as used herein covers the improvement or elimination of a condition after it has developed. In any case, prevention or treatment can be differentiated by the diagnosis made by the physician or other health care professional and the intended outcome of the administration of the therapeutic agent.

In general, treatment or prevention of a disease or condition described herein is achieved by administering an ActRIIB antagonist in an effective amount. An effective amount of an agent means an amount effective to achieve the desired therapeutic or prophylactic result at the required dosages and for the required period of time. The therapeutically effective amount of an agent herein may vary depending on factors such as the course of the disease, the age, sex and body weight of the individual, as well as the ability of the agent to produce the desired response in the individual. A prophylactically effective amount means an amount effective to achieve the desired prophylactic result at the required dosages and for the required period of time.

The terms subject, individual, or patient are used interchangeably herein and generally refer to mammals. Mammals include, but are not limited to, domesticated animals (eg, cows, sheep, cats, dogs and horses), primates (eg, humans and non-human primates such as monkeys), rabbits and rodents (eg, mice and rats).

Myelofibrosis is a clonal neoplastic disorder of hematopoiesis, usually characterized by progressive fibrosis of the bone marrow, leading to increasingly ineffective hematopoiesis, extramedullary hematopoiesis, various inflammatory complications and decreased survival [Mascarenhas et al. (2012) Curr Med Chem 19:4399–4413; and Vannucchi et al. (2011) Hematol Am Soc Hematol Educ Prog 2011:222-230]. It is one of the myeloproliferative disorders of the bone marrow with excessive cell production. The production of cytokines such as fibroblast growth factor by abnormal cell lineages leads to the replacement of bone marrow hematopoietic tissue with connective tissue through collagen fibrosis. A decrease in hematopoietic tissue impairs the patient's ability to form new blood cells, resulting in the progression of pancytopenia, a deficiency of all types of blood cells. However, fibroblast proliferation and collagen deposition is a secondary phenomenon, and the fibroblasts themselves are not part of the abnormal cell lineage. As a result of progressive scarring, or fibrosis, of the bone marrow, the patient develops extramedullary hematopoiesis as hematopoietic cells are forced to migrate to other areas, particularly the liver and spleen. This causes these organs to enlarge. If the liver is affected, the condition is called hepatomegaly. An enlarged spleen is called splenomegaly, which also contributes to pancytopenia, especially thrombocytopenia and anemia. In addition, extramedullary hematopoiesis has been reported to occur in the lungs and lymph nodes. Another complication of extramedullary hematopoiesis is poikilocytosis, the presence of abnormally shaped red blood cells. Typical clinical manifestations of myelofibrosis include progressive hepatosplenomegaly, abnormal blood cell counts, and wasting symptoms such as fatigue, weight loss, night sweats, fever, itching, bone pain, early satiety, abdominal pain or discomfort, arthralgia, myalgia , paresthesia, cachexia, splenic infarction and bleeding. Until recently, the only treatment with clearly demonstrated

- 57 043914 allogeneic hematopoietic stem cell transplantation alloHSCT has had an impact on disease progression, but treatment is associated with high mortality and only a small proportion of patients are suitable candidates for such intensive therapy [Gupta et al. (2012) Blood 120: 1367-1379].

In some aspects, an ActRIIB antagonist may be used alone or in combination with one or more maintenance therapies or active agents to treat, prevent, or reduce the rate of progression and/or severity of myelofibrosis (eg, primary myelofibrosis, myelofibrosis preceded by polycythemia vera, or myelofibrosis preceded by essential thrombocythemia). In particular, ActRIIB antagonists may be used alone or in combination with one or more supportive therapies or active agents to treat, prevent, or reduce the rate of progression and/or severity of one or more complications of myelofibrosis, including, for example, ineffective hematopoiesis, anemia, inflammation, fibrosis (eg, bone marrow fibrosis, splenic fibrosis, and liver fibrosis), pancytopenia, thrombocytopenia, extramedullary hematopoiesis (eg, splenic extramedullary hematopoiesis, hepatic extramedullary hematopoiesis, pulmonary extramedullary hematopoiesis, and lymphatic extramedullary hematopoiesis), hepatomegaly, splenomegaly, osteosis sclerosis , osteomyelofibrosis, poikilocytosis, fatigue, weight loss, night sweats, fever, itching, bone pain, early satiety, abdominal pain or discomfort, arthralgia, myalgia, paresthesia, cachexia, splenic infarction and bleeding.

Currently, the diagnosis of primary myelofibrosis (PMF) is based on the criteria of the World Health Organization (WHO) and includes a comprehensive assessment of clinical and laboratory parameters [Tefferi A et al. (2007) Blood. 110:1092-1097]. There are three main diagnostic criteria according to WHO: 1) proliferation of megakaryocytes with signs of atypia (small to large megakaryocytes with an abnormal nuclear-cytoplasmic ratio and hyperchromatic nuclei, dense chromatin and dyskaryosis) in combination with reticulin and/or collagen fibrosis or in the absence of reticulin fibrosis megakaryocyte changes are accompanied by increased cellularity, granulocyte proliferation, often decreased erythropoiesis (ie, pre-fibrotic primary myelofibrosis), 2) does not meet WHO criteria for the diagnosis of chronic myeloid leukemia, polycythemia vera, myelodysplastic syndrome or other myeloproliferative neoplasms, and 3) detection of JAK2V617F or other clonal marker or absence of evidence of reactive bone marrow fibrosis. In addition, there are four additional WHO diagnostic criteria: 1) leukoerythroblastosis, 2) elevated serum lactate dehydrogenase, 3) anemia, and 4) palpable splenomegaly. Peripheral blood leukoerythroblastosis (i.e., the presence of nucleated red blood cells, immature granulocytes, and teardrop red blood cells) is a typical but not consistent feature of PMF; prefibrotic PMF may not be accompanied by overt leukoerythroblastosis [Kvasnicka et al. (2010) Am J Hematol. 85:62-69]. Bone marrow fibrosis in PMF is usually associated with JAK2V617F or mutant CALR or MPL, trisomy 9, or del(13q) [Hussein et al. (2009) Eur J Haematol. 82:329-338]. The presence of these genetic markers supports the diagnosis of PMF in the presence of myeloid neoplasms associated with bone marrow fibrosis. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of primary myelofibrosis, in particular, treating, preventing, or reducing the rate of progression and/or severity of one or more than one complication of primary myelofibrosis.

Currently, the diagnosis of myelofibrosis that develops on the background of previous polycythemia vera (post-polycythaemia MF) and myelofibrosis that develops on the background of previous essential thrombocythemia (post-thrombocythaemia MF) is based on criteria published by the International Working Group on the Study and Treatment of Myeloproliferative Disorders (IWG-MRT). [Barosi G. et al. (2008) Leukemia. 22:437-438]. There are two main criteria for post-polycythaemia MF according to IWG-MRT: 1) documentation of a previous diagnosis of polycythaemia vera according to WHO criteria and 2) bone marrow fibrosis grade 2-3 (on a scale of 0-3) or grade 3-4 (on a scale 0-4 ). Grade 2-3 according to the European classification: diffuse, often coarse fibrous network without signs of collagenization (negative trichrome staining) or diffuse coarse fibrous network with areas of collagenization (positive trichrome staining) [Thiele et al. (2005) Haematologica. 90:1128-1132]. 3-4 degrees according to the European classification: diffuse or dense increase in the amount of reticulin with multiple intersections; sometimes only focal collagen bundles and/or areas of osteosclerosis, or diffuse and dense increase in reticulin, characterized by multiple intersections, with coarse collagen bundles, often accompanied by significant osteosclerosis [Manoharan et al. (1979) Br J Haematol 43:185–190]. In addition, there are four additional diagnostic criteria according to IWG-MRT; to make a diagnosis of post-polycythaemic MF in a patient, the basic IWGMRT criteria and two additional criteria are required: 1) anemia or long-term lack of need for phlebotomy in the absence of cytoreductive therapy, 2) leukoerythroblastic pattern of peripheral

- 58 043914 blood, 3) an increase in the size of the spleen, defined as either palpable splenomegaly > 5 cm or the appearance of palpable splenomegaly; 4) the appearance of 1 or more of 3 constitutional symptoms: loss of more than 10% of body weight over 6 months, night sweats, unexplained fever. There are two main criteria for post-thrombocythaemic MF according to IWG-MRT: 1) documentation of a previous diagnosis of essential thrombocythemia according to WHO criteria, 2) bone marrow fibrosis grade 2-3 (on a scale of 0-3) or grade 3-4 (on a scale 0-4 ). In addition, there are five additional diagnostic criteria according to IWG-MRT; to make a diagnosis of post-thrombocythaemic MF in a patient, the main criteria of IWG-MRT and two additional criteria are required: 1) anemia and a decrease in hemoglobin by 20 g/l below the initial value, 2) leukoerythroblastic peripheral blood picture, 3) enlargement of the spleen, defined as either palpable splenomegaly > 5 cm or the appearance of palpable splenomegaly; 4) an increase in serum lactate dehydrogenase levels; and 5) the appearance of 1 or more of 3 constitutional symptoms: loss of more than 10% of body weight in 6 months, night sweats, unexplained fever. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of postpolycythaemic myelofibrosis, in particular, treating, preventing, or reducing the rate of progression and/or severity of one or more than one complication of postpolycythaemic myelofibrosis. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of post-thrombocythemic myelofibrosis, in particular, treating, preventing, or reducing the rate of progression and/or severity of one or more than one complication of post-thrombocythemic myelofibrosis.

Active modeling of myelofibrosis prognosis began with the development of the International Prognosis Score (IPSS) in 2009 [Cervantes F et al. (2009) Blood 113:2895–2901]. The IPSS for myelofibrosis is used to determine the prognosis of patients at diagnosis and uses five independent predictors of decreased survival: age >65 years, hemoglobin <100 g/L, white blood cell count >25 x 10 /L, peripheral blood blasts >1% and the presence of constitutional symptoms. The presence of 0, 1, 2, and >3 adverse factors defines low-, intermediate-1, intermediate-2, or high-risk disease, respectively. The corresponding median survival rates were 11.3, 7.9, 4, and 2.3 years, respectively. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis in a patient having low, intermediate-1, intermediate-2, or high risk myelofibrosis according to IPSS. In some embodiments, the disclosure relates to methods and uses of ActRIIB antagonists for preventing or delaying progression of IPSS risk in myelofibrosis (e.g., preventing or delaying progression from low to intermediate-1, intermediate-1 to intermediate-2, and intermediate-2 to high risk according to IPSS). In some embodiments, the disclosure relates to methods and uses of ActRIIB antagonists to help reduce or enhance IPSS risk reduction in myelofibrosis (e.g., promote or enhance risk reduction from high to intermediate-2, intermediate-2 to intermediate-1, and intermediate -1 to low risk according to IPSS).

Subsequently, IWG-MRT developed a dynamic prognostic model (Dynamic International Prognosis Score [DIPSS]) that uses the same prognostic variables as the IPSS but can be applied at any time during the course of the disease [Passamonti F et al. (2010) Blood. 115:1703-1708]. According to DIPSS, hemoglobin levels <100 g/L are assigned two points instead of one, and the risk group classification changes accordingly: low (0 points), intermediate-1 (1 or 2 points), intermediate-2 (3 or 4 points) and high (5 or 6 points). The corresponding median survival values of 14.2, 4 and 1.5 years were not achieved. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis in a patient having low, intermediate-1, intermediate-2, or high risk myelofibrosis according to DIPSS. In some embodiments, the disclosure relates to methods and uses of ActRIIB antagonists for preventing or delaying risk progression in myelofibrosis according to DIPSS (e.g., preventing or delaying risk progression from low risk to intermediate-1, intermediate-1 to intermediate-2, and intermediate-2 to high risk according to DIPSS). In some embodiments, the disclosure relates to methods and uses of ActRIIB antagonists to help reduce or enhance risk reduction in myelofibrosis according to DIPSS (e.g., promote or enhance risk reduction from high risk to intermediate-2, intermediate-2 to intermediate-1, and intermediate-1 to low risk according to DIPSS).

Subsequently, IPSS- and DIPSS-independent risk factors for survival in myelofibrosis were found to include an unfavorable karyotype (i.e., combined karyotype or isolated

- 59 043914 anomaly or two anomalies including +8, -7/7q-, i(17q), inv(3), -5/5q-, 12p- or 11q23 rearrangement) [Hussein et al. (2010) Blood. 115:496-499], the need for red blood cell transfusion [Tefferi et al. (2009) Am J Hematol. 85:14-17] and platelet count <100 x 10 ⁹ /L [Patnaik et al. (2010) Eur J Haematol. 84:105108]. Accordingly, the DIPSS was modified to DIPSS-plus to include the following three additional DIPSS-independent risk factors: platelet count <100 x ¹⁰⁹ /L, red blood cell transfusion requirement, and unfavorable karyotype. Based on the above eight risk factors, there are four DIPSS-plus risk categories: low (no risk factors), intermediate-1 (one risk factor), intermediate-2 (two or three risk factors) and high (four or more risk factors) with the corresponding median survival values were 15.4, 6.5, 2.9 and 1.3 years. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis in a patient having low, intermediate-1, intermediate-2, or high risk myelofibrosis according to DIPSSplus. In some embodiments, the disclosure relates to methods and uses of ActRIIB antagonists for preventing or delaying risk progression in myelofibrosis according to DIPSS-plus (e.g., preventing or delaying risk progression from low risk to intermediate-1, intermediate-1 to intermediate-2, and intermediate -2 to high risk according to DIPSS-plus). In some embodiments, the disclosure relates to methods and uses of ActRIIB antagonists to promote or enhance risk reduction in myelofibrosis according to DIPSS-plus (e.g., promote or enhance risk reduction from high to intermediate-2, intermediate-2 to intermediate-1 and intermediate-1 to low risk according to DIPSS-plus).

Since the publication of DIPSS-plus, several studies have been published that provide additional prognostic information. For example, monosomal karyotype, inv(3)/i(17q) abnormalities, and any two of the following: circulating blasts >9%, leukocytes > ^40x109 /L, or other unfavorable karyotype were predictors of 2-year myelofibrosis mortality >80% [ Tefferi et al. (2011) Blood. 118:4595-4598]. Similarly, poorer survival in myelofibrosis has been associated with null zygosity for the JAK2 46/1 haplotype, low JAK2V617F allelic load, or the presence of IDH, EZH2, SRSF2, or ASXL1 mutations [Tefferi, Ayalew (2014) Am. J. Hematol. 89:916925]. In contrast, the presence or absence of JAK2V617F, MPL, or TET2 mutations did not affect survival. Survival in myelofibrosis was also affected by elevated serum levels of IL-8 and IL-2R, as well as serum free light chain levels, both independent of DIPSS-plus. Recently, Tefferi et al. studied 254 patients with myelofibrosis and described mutation rates of 58% for JAK2, 25% CALR, 8% MPL, and 9% wild type for all three mutations (ie, triple negative) [Tefferi et al. (2014) Leukemia, advance publication DOI 10.1038/leu.2014.3]. The CALR mutation rate in cases without JAK2/MPL mutations was 74%. CALR mutations were associated with younger age, lower platelet counts, and lower DLPSS-plus scores. Patients with a CALR mutation were also less likely to have anemia, require transfusion, or have leukositosis. Spliceosome mutations were rare in patients with CALR mutations. In a subsequent international study of 570 patients, the authors found the longest survival in CALR+ASXL1- patients (median 10.4 years) and the shortest in CALR-ASXL1+ patients (median 2.3 years) [Tefferi et al. (2014) Leukemia, advance publication DOI 10.1038/leu.2014.57]. CALR+ASXL1+ and CALR-ASXL1 patients had similar survival and were grouped together in the intermediate-risk category (median survival 5.8 years). As is evident for overall survival, leukemia-free survival is also significantly reduced in patients carrying several mutations, including IDH and SRSF2 [Tefferi et al. (2012) Leukemia. 26:475-480; Lasho et al. (2012) Blood. 120:4168-4171]. In addition, LNK and TPO mutations have also been associated with myelofibrosis.

The discovery of the Janus kinase 2 (JAK2) gain-of-function mutation, JAK2V617F, has greatly improved understanding of the biological basis of myelofibrosis and has also enabled the development of ruxolitinib, a JAK2 inhibitor, which is the first drug approved by the FDA for the treatment of myelofibrosis [Baxter et al. (2005) Lancet 365:1054-1061; James S. et al. (2005) Nature 434:1144-1148; Kralovics et al. (2005) N Engl J Med. 352:1779-1790; and Levine et al. (2005) Cancer Cell 7:387-397]. The Janus kinase family of receptor tyrosine kinases includes four distinct proteins (JAK1, JAK2, JAK3, and TYK2), and this family of proteins is known to play a key role in the growth and development of myeloid and lymphoid cells. In particular, they mediate cell-to-cell interactions of cytokine receptors, resulting in the activation of members of the signal transducer and activator of transcription (STAT) family and subsequent activation of genes that regulate cell proliferation and differentiation [Quintas-Cardama et al. (2011) Nat Rev Drug Discov 10:127–140]. The JAK2V617F mutation results in constitutive activation of JAK2 and thus promotes the proliferation and differentiation of myeloid cells. Other Janus kinase inhibitors in clinical trials include, for example: fedratinib (SAR302503), monoelotinib (CYT387), pacritinib, lestaurtinib, AZD-1480, BMS-911543, NS-018, LY2784544, SEP-701, XL019 and AT- 9283.

- 60 043914

In some aspects, the invention provides methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis in a patient having one or more of the following: monosomal karyotype, inv(3)/i(17q) abnormalities, circulating blasts >9% and/or leukocytes >40 x 10 ⁹ /l, null zygosity for JAK2 haplotype 46/1, JAK2V617F mutation, IDH1 mutation, IDH2 mutation, EZH2 mutation, SRSF2 mutation, ASXL1 mutation, increased serum IL-8 levels , increased serum IL-2R levels, increased free light chain levels, JAK1 mutation, JAK2 mutation, JAK3 mutation, TYK2 mutation, MPL mutation, CALR mutation, CALR+ASXL1-, CALR-ASKL1+, CALR+ASKL1+, CALR-ASKL1- , TET2 mutation, THPO mutation and LNK mutation.

Controlling the development of anemia can be one of the most difficult aspects of treating patients with myelofibrosis [Tefferi A. (2011) Blood 117(13):3949–3504; Barosi et al. (2011) Expert Opin Pharmacother 12(10): 1597-1611]. Blood transfusion (whole blood or red blood cell transfusion) is standard therapy for myelofibrosis patients with symptoms of anemia. In addition to transfusion, various standard agents are used to treat anemia in these patients. For example, erythropoiesis stimulants [eg, erythropoiesis stimulating agents such as erythropoietin (EPO) and its derivatives], androgens (eg, testosterone enanthate and fluoxymesterone), prednisone, danazol, thalidomide, prednisone, and lenalidomide are routinely used to treat anemia in patients with myelofibrosis. In general, erythropoiesis stimulants are used in patients with mild, nontransfusion-dependent anemia and low serum erythropoietin levels. The percentage of patients with an objective response varies from 20 to 60%, with no clear data supporting darbepoetin alfa over standard recombinant erythropoietin. Responses to erythropoiesis stimulants are usually short-term (approximately 1 year). If erythropoiesis stimulants do not work or have low efficacy, danazol or androgen drugs are usually used to treat patients with anemia, the percentage of patients with an objective response is approximately 20%. Low doses of thalidomide combined with tapering doses of prednisone provided an objective response in approximately 20-40% of anemic patients [Thapaliya et al. (2011) Am J Hematol 86(1):86–98]. However, treatment with thalidomide is often poorly tolerated by patients, causing peripheral neuropathies, constipation and drowsiness, forcing discontinuation of the drug. In myelofibrosis patients with del(5q31)-associated anemia, lenalidomide is recommended as first-line therapy because it produces significant improvement with resolution of anemia and, in some cases, molecular remission [Tefferi et al. (2007) Leukemia 21(8): 1827-1828]. In some aspects, the disclosure relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of myelofibrosis in a patient having anemia. In some embodiments, the description relates to methods and uses of ActRIIB antagonists for treating, preventing, or reducing the rate of progression and/or severity of anemia in a patient having myelofibrosis. In some embodiments, the description relates to a method of treating, preventing or reducing the rate of progression and/or severity of myelofibrosis in a patient in need thereof, comprising administering one or more than one ActRIIB antagonist together with one or more active ingredients selected from the group consisting of from: erythropoiesis stimulant [eg, such as erythropoietin (EPO) and its derivatives], androgen (eg, testosterone enanthate and fluoxymesterone), prednisone, danazol, thalidomide, prednisone and lenalidomide. In some embodiments, the description relates to a method of treating, preventing or reducing the rate of progression and/or severity of anemia in a patient with myelofibrosis in need thereof, comprising administering one or more than one ActRIIB antagonist together with one or more active ingredients selected from the group , consisting of: an erythropoiesis stimulant [eg, such as erythropoietin (EPO) and its derivatives], androgen (eg, testosterone enanthate and fluoxymesterone), prednisone, danazol, thalidomide, prednisone and lenalidomide. In some embodiments, the disclosure relates to a method of treating, preventing, or reducing the rate of progression and/or severity of anemia in a patient with myelofibrosis in need thereof, comprising administering one or more ActRIIB antagonists in conjunction with a blood transfusion (whole blood or red blood cell transfusion).

When monitoring a person's hemoglobin and/or hematocrit levels, levels below normal for the appropriate age and sex category may indicate the presence of anemia, although individual variations should be taken into account. For example, hemoglobin levels of 10 to 12.5 g/dL and usually approximately 11.0 g/dL are considered the normal range for healthy adults, although from a therapeutic perspective, a lower target level may cause fewer cardiovascular side effects. systems; see, for example, Jacobs et al. (2000) Nephrol Dial Transplant 15, 15-19. Alternatively, hematocrit (the percentage of the volume of a blood sample occupied by cells) can be used as an indicator of anemia. Hematocrit levels in healthy individuals range from approximately 41 to 51% in healthy men and from 35% to 45% in healthy women. In some embodiments, the patient may receive treatment with a regimen designed to restore the patient's target red blood cell, hemoglobin, and/or hematocrit levels, or

- 61 043914 can reduce or eliminate red blood cell transfusions (reduce the transfusion load), while maintaining an acceptable level of red blood cells, hemoglobin and/or hematocrit. Because hemoglobin and hematocrit levels can vary from individual to individual, it is optimal that the target hemoglobin and/or hematocrit level can be individualized for each patient.

In patients who frequently receive transfusions of whole blood or red blood cells, the normal mechanism of iron homeostasis may be suppressed, ultimately leading to toxic and possibly fatal accumulation of iron in vital tissues such as the heart, liver, and endocrine glands. Regular RBC transfusions require blood units from multiple donors and therefore carry an increased risk of alloimmunization. Difficult vascular access, availability and compliance with iron chelators, and high cost are some reasons that make limiting the number of red blood cell transfusions desirable.

In some aspects, one or more ActRIIB antagonists, possibly in combination with an EPO receptor activator, can be used with one or more iron chelator molecules to promote iron excretion into urine and/or feces and thereby prevent or reverse iron overload in tissues. patients with myelofibrosis. Effective iron chelators must be able to selectively bind and neutralize iron ions, the oxidized form of non-transferrin-bound iron, which is probably the cause of most iron toxicity due to the catalytic formation of hydroxyl radicals and oxidation products [see, for example, Esposito et al. (2003) Blood 102:2670-2677]. Such agents are structurally diverse, but all have oxygen or nitrogen donor atoms capable of forming neutralizing octahedral coordination complexes with individual iron atoms in a stoichiometric ratio of 1:1 (hexadentate agents), 2:1 (tridentate agents), or 3:1 (bidentate agents) [ Kalinowski et al. (2005) Pharmacol Rev 57:547-583]. In general, effective iron chelators also have a relatively low molecular weight (eg, less than 700 daltons), with both water and lipid solubility to provide access to affected tissues. Specific examples of iron chelator molecules include deferoxamine, a bacterially derived hexadentate agent requiring daily parenteral administration, and the orally active synthetic agents deferiprone (bidentate) and deferasifox (tridentate). Combination therapy, including same-day administration of two iron chelators, is promising in patients without an objective response to chelator monotherapy and to overcome problems with low compliance to deferoxamine monotherapy [Cao et al. (2011) Pediatr Rep 3(2):e17; and Galanello et al. (2010) Ann NY Acad Sci 1202:79-86].

One or more ActRIIB antagonists of the invention can be used in combination with an EPO receptor activator to achieve an increase in red blood cell count, particularly at low doses. This may be effective in reducing known side effects and risks associated with high doses of EPO receptor activators. The main side effects of erythropoiesis stimulants include, for example, excessive increases in hematocrit or hemoglobin levels and polycythemia. Elevated hematocrit levels can lead to hypertension (more specifically, exacerbation of hypertension). Other side effects of erythropoiesis stimulants have been described, some of which are related to hypertension, and include headaches, influenza-like syndrome, shunt occlusion, myocardial infarction and cerebral paroxysms due to thrombosis, hypertensive encephalopathy and red cell aplasia; see, for example, Singibarti (1994) J. Clin Investig 72(suppl 6), S36-S43; Horl et al. (2000) Nephrol Dial Transplant 15(suppl 4), 51-56; Delanty et al. (1997) Neurology 49, 686-689; and Bunn (2002) N Engl J Med 346(7), 522-523). Some embodiments of this disclosure provide methods for treating or preventing anemia in a patient with myelofibrosis by administering to the patient a therapeutically effective amount of one or more ActRIIB antagonist and EPO receptor activator. In some embodiments, ActRIIB antagonists as disclosed may be used in combination with EPO receptor activators to reduce the required dose of these activators in patients predisposed to the side effects of these erythropoiesis stimulants. These methods can be used for therapeutic and prophylactic effects on the patient.

Provided that the ActRIIB antagonists described herein act by a mechanism other than ESAs (erythropoiesis-stimulating agents), these antagonists may have utility in increasing red blood cell counts and hemoglobin levels in a patient who has not achieved an objective response to ESAs or other receptor activators EPO. For example, the ActRIIB antagonist described herein may be useful in a patient in whom normal or increased dosage of ESA (>300 IU/kg/week) does not increase hemoglobin levels to target. An unsatisfactory response to ESA can be either constitutive (observed with the first ESA treatment) or acquired (observed with repeated ESA treatment).

Cytoreductive agents have been the drugs of choice for most patients with symptomatic splenomegaly. Hydroxyurea (hydroxyurea, HC) is the most commonly used cytoreductive agent and generally produces a moderate response in high doses. However, NS can often exacerbate cytopenias and is therefore often poorly tolerated. According to published

- 62 043914 data, up to 35% of patients treated with NS showed a decrease in spleen size from 25% to 50% [Martinez-Trillos et al. (2010) Ann Hematol. 89(12): 1233-1237]. In patients who do not achieve an objective response to NS, busulfan or melphalan can be used, especially in elderly patients, as there is evidence that these agents may increase the incidence of leukemic transformation. The percentage of patients who achieve objective improvement in their spleen when treated with low-dose thalidomide is low (<20%). However, in a study that included some patients who had failed prior lenalidomide therapy, the percentage of patients who achieved an objective response was 33%. In cases of massive refractory splenomegaly, the percentage of patients who achieved an objective response to intravenous cladribine over a month reached 50%, with severe but reversible cytopenias being the main manifestation of toxicity [Faoro et al. (2005) Eur J Haematol 74(2): 117-120]. In recent studies, ruxolitinib was superior to NS and is therefore becoming a first-line treatment for the control of symptomatic or progressive splenomegaly. Unfortunately, a common side effect of ruxolitinib is the development or worsening of anemia in patients with myelofibrosis. Thus, although JAK inhibitors can be used to treat splenomegaly, they may worsen other complications of myelofibrosis, particularly anemia and anemia-related disorders.

In addition to JAK2 inhibition, other treatment strategies are being explored for the treatment of myeloproliferative disorders, including immunomodulatory drugs (eg, pomalidomide), inhibitors of mammalian mTOR pathway targets (eg, rapamycin, sirolimus, deforolimus, everolimus, temsirolimus, NVP-BEZ235, BGT226, SF1126, PK1- 587, INK128, AZD8055 and AZD2014) and modulators of epigenetic factors (eg, histone deacetylase inhibitors such as givinostat (ITF2357), panobinostat (LBH589) and pracinostat) [Mascarenhas et al. (2013) Haematologica 98(10): 1499-1509].

This specification further provides the use of an ActRIIB antagonist in combination with one or more therapeutic modalities for the treatment of patients according to this specification. For example, ActRIIB antagonists can be administered in combination with cytotoxins, immunosuppressive agents, radiotoxic agents and/or therapeutic antibodies. Specific co-therapeutic agents provided herein include steroids (eg, corticosteroids such as prednisone), immunosuppressive and/or anti-inflammatory agents (eg, interferon gamma, cyclophosamide, azathioprine, methotrexate, penicillamine, cyclosporine, colchicine, antithymocyte globulin , mycophenolate mofetil and hydroxychloroquine), cytotoxic drugs, calcium channel blockers (eg, nifedipine), angiotensin-converting enzyme (ACE) inhibitors, para-aminobenzoic acid (PABA), dimethyl sulfoxide, transforming growth factor beta (TGFP) inhibitors, interleukin inhibitors 5 (IL-5) and pan-caspase inhibitors, but are not limited to them. Additional agents that may be used in combination with an ActRIIB antagonist include lectins (as described, for example, in US Pat. No. 7,026,283, the contents of which are incorporated herein by reference in their entirety), as well as the antifibrotic agents described by Wynn et al. (2007, J Clin Invest 117:524-529), but not limited to them. For example, additional antithrombotic agents and therapies include anti-inflammatory/immunosuppressive/cytotoxic drugs (including colchicine, azathioprine cyclophosphamide, prednisone, thalidomide, pentoxifylline, and theophylline), TGFP signaling pathway modulators (including relaxin, SMAD7, HGF, and BMP7, and TGFe1 inhibitors, TeRI, TeRII, EGR-I and CTGF), antagonists of cytokines and cytokine receptors (inhibitors of LL-1e, LL-5, IL6, IL-13, IL-21, IL-4R, IL-13Ra1, GM-CSF, TNF- α, oncostatin M, WlSP-I and PDGF), cytokines and chemokines (LFN-γ, LFN-α/β, LL-12, IL-10, HGF, CXCL10 and CXCL11), chemokine antagonists (inhibitors of CXCL1, CXCL2, CXCL12 , CCL2, CCL3, CCL6, CCL17 and CCL18), chemokine receptor antagonists (CCR2, CCR3, CCR5, CCR7, CXCR2 and CXCR4 inhibitors), TLR antagonists (TLR3, TLR4 and TLR9 inhibitors), angiogenesis antagonists (VEGF-specific antibodies and therapy for adenosine deaminase deficiency), antihypertensive drugs (beta blockers and ANG 11, ACE and aldosterone inhibitors), vasoactive substances (ET-1 receptor antagonists and bosentan), inhibitors of enzymes involved in the synthesis and processing of collagen (prolyl hydroxylase inhibitors), B antagonists cells (rituximab), integrin/adhesion molecule antagonists (molecules that block α1β1 and ave6 integrins, as well as integrin-linked kinase inhibitors and antibodies specific for ICAM-I and VCAM-I), proapoptotic drugs targeting myofibroblasts, MMP inhibitors (MMP2, MMP9 and MMP12 inhibitors) and TLMP inhibitors (TIMP-1 specific antibodies), but are not limited to them.

In some embodiments, ActRIIB antagonists as disclosed may be used as monotherapy to treat a patient in need thereof. Alternatively, ActRIIB antagonists may be used in combination with standard therapeutic approaches aimed at treating or preventing the proliferative disorders described herein, including, for example, surgery (eg, splenectomy), cytotoxic agents, radiotherapy, including radiation or administration of radioactive substances, chemotherapeutic agents, antihormonal agents, growth inhibitory agents, antineoplastic compositions, as well as treatment with antineoplastic agents listed herein and known in the art, or combinations thereof.

In general, a cytotoxic agent refers to a substance that inhibits or prevents cell function and/or causes cell destruction. The term includes radioactive isotopes (e.g. At ²¹¹ , I ¹³¹ , I ¹²⁵ , Y ⁹⁰ , Re ¹⁸⁶ , Re ¹⁸⁸ , Sm ¹⁵³ , Bi ²¹² , P ³² and Lu radioisotopes), chemotherapeutic agents (e.g. methotrexate, adriamycin, vinca alkaloids (vincristine, vinblastine, etoposide), doxorubicin, melphalan, mitomycin C, chlorambucil, daunorubicin or other intercalating agents, enzymes and fragments thereof, such as nucleolytic enzymes, antibiotics and toxins, such as low molecular weight toxins or enzymatically active toxins of bacteria, fungi, plants or animals, including fragments and/or variants thereof, and various antitumor or anticancer agents described below.Other cytotoxic agents are described below.The tumoricidal agent causes the destruction of tumor cells.

In general, a chemotherapeutic agent is a chemical compound that may find use in the treatment of cancer. Examples of chemotherapeutic agents include alkylating agents such as thiotepa and Cytoxan® cyclophosamide; alkylsulfonates such as busulfan, improsulfan and piposulfan; aziridines such as benzodopa, carboquone, meturedopa and uredopa; ethyleneimines and methylmelamines, including altretamine, triethylenemelamine, triethylenephosphoramide, triethylenethiophosphoramide and trimethylmelamine; acetogenins (in particular bullatacin and bullatacinone); delta-9tetrahydrocannabinol (dronabinol, MARINOL®); beta-lapachone, lapachol; colchicines; betulinic acid; camphothecin (including synthetic analogue of topotecan (HYCAMTIN®), CPT-11 (irinotecan, CAMPTOSAR®), acetylcamphothecin, scopolectin and 9-aminocampotecin); bryostatin; kallistatin; CC1065 (including its synthetic analogues adozelesin, carzelesin and bizelesin); podophyllotoxin; podophyllic acid; teniposide; cryptophycins (especially cryptophycin 1 and cryptophycin 8); dolastatin; duocarmycin (including synthetic analogues KW-2189 and CBl-TMl); eleutherobin; pancratistatin; sarcodictin; spongistatin; nitrogen mustards such as chlorambucil, chlornaphazine, cyclophosphamide, estramustine, ifosfamide, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, novembicine, phenesterine, prednimustine, trofosfamide, uracil mustard; nitrosoureas such as carmustine, chlorosotocin, fotemustine, lomustine, nimustine and ranimustine; antibiotics such as enediyne antibiotics (eg, calicheamicin, especially calicheamicin gamma 11 and calicheamicin omega 11 (see, for example, Agnew, Chem Intl. Ed. Engl., 33: 183-186 (1994)); dinemycin, including dinemycin A; esperamycin; as well as neocarzinostatin chromophore and related chromoprotein chromophores of enediyne antibiotics), aclacinomysins, actinomycin, anthramycin, azaserin, bleomycins, cactinomycin, carabicin, carminomycin, carzinophyllin, chromomycins, dactinomycin, daunorubicin, detorubicin, 6-diazo-5oxo- L-norleucine, adriamycin® doxorubicin (including morpholino-doxorubicin, cyanomorpholino-doxorubicin, 2-pyrrolino-doxorubicin and deoxydoxorubicin), epirubicin, ezorubicin, idarubicin, marcellomycin, mitomycins such as mitomycin C, mycophenolic acid, nogalamycin, olivomycins, peplomycin, porphyromycin cin, puromycin, quelamycin, rhodorubicin, streptonigrin, streptozocin, tubercidin, ubenimex, zinostatin, zorubicin; antimetabolites such as methotrexate and 5-fluorouracil (5-FU); folic acid analogs such as denopterin, methotrexate, pteropterin, trimetrexate; purine analogues such as fludarabine, 6-mercaptopurine, thiamiprin, thioguanine; pyrimidine analogues such as ancytabine, azacitidine, 6-azauridine, carmofur, cytarabine, dideoxyuridine, doxifluridine, enocytabine, floxuridine; androgens such as calusterone, dromostanolone propionate, epithiostanol, mepithiostane, testolactone; adrenal inhibitors such as aminoglutethimide, mitotane, trilostane; folic acid derivatives such as folinic acid; aceglatone, aldophosphamide glycoside; aminolevulinic acid; eniluracil; amsacrine; bestrabucil; bisantrene; edatraxate; demecolcine; diaziquon; elfornitine; elliptinium acetate; epothilone; etoglucide; gallium nitrate; hydroxyurea; lentinan; lonidamine; maytansinoids such as maytansine and ansamitocins; mitoguazone; mitoxantrone; mopidanmol; nitraerin; pentostatin; phenomet; pirarubicin; Losoxantrone; 2ethylhydrazide; procarbazine; PSK® polysaccharide complex (JHS Natural Products, Eugene, OR); razoxane; rhizoxin; sisofiran; spirogermanium; tenuazonic acid; triaziquon; 2,2',2trichlorotriethylamine; trichothecenes (especially T-2 toxin, verracurine A, roridin A and anguidine); urethane; vindesine (eldizine®, fieldesine®); dacarbazine; mannomustin; mitobronitol; mitolactol; pipobromance; gacytosine; arabinoside (Ara-C); thiotera; taxoids, such as Taxol® (paclitaxel; Bristol-Myers Squibb Oncology, Princeton, NJ.), Abraxane™ (Cremophor-free), paclitaxel albumin nanoformulations (American Pharmaceutical Partners, Schaumberg, Illinois), and Taxotere® doxetaxel (RhonePoulenc Rorer, Anthony, France); chlorambucil; gemcitabine (gemzar®); 6-thioguanine; mercaptopurine; methotrexate; platinum analogues such as cisplatin and carboplatin, vinblastine (Velban®), platinum; etoposide (VP-16); ifosfamide; mitoxantrone; vincristine (Oncovin®); oxaliplatin; leucovorin; vinorelbine (navelbine®); Novantrone; edatrexate; daunomycin; aminopterin; ibandronate; topoisomerase inhibitor

- 64 043914

RFS 2000; difluoromethylornithine (DMFAO); retinoids such as retinoic acid; capecitabine (Xeloda®); pharmaceutically acceptable salts, acids or derivatives of any of the above, as well as combinations of two or more of the above, such as CHOP, an acronym for cyclophosphamide, doxorubicin, vincristine and prednisone combination therapy, and FOLFOX, an acronym for oxaliplatin (eloxatin) treatment regimen ™) in combination with 5-fluorouracil and leucovorin.

Also included are antihormonal agents that act by regulating, reducing, blocking or inhibiting hormonal effects that may promote cancer growth and are often administered systemically or throughout the body. They themselves may be hormones. Examples include antiestrogens and selective estrogen receptor modulators (SERMs), including, for example, tamoxifen (including tamoxifen Nolvadex®), raloxifene Evista®, droloxifene, 4-hydroxytamoxifene, trioxifene, keoxifene, LY117018, onapristone and toremifene Fareston®; antiprogesterone drugs; estrogen receptor negative regulators (ERD); agents that function to suppress or shut down the ovaries, for example, luteinizing hormone releasing factor (LHRH) agonists such as leuprolide acetate Lupron® and Eligard®, goserelin acetate, buserelin acetate and tripterelin; other antiandrogens, such as flutamide, nilutamide and bicalutamide and aromatase inhibitors, which inhibit the enzyme aromatase, which regulates the formation of estrogens in the adrenal glands, such as, for example, 4(5)-imidazoles, aminoglutethimide, megestrol acetate (Megaza®, exemestane aromazin®, formestane, fadrozole, vorozole rivizor®, letrozole femara® and anastrozole arimidex® In addition, this definition of chemotherapeutic agents includes bisphosphonates such as clodronate (e.g. bonefos® or ostak®, etidronate didrocal®, NE58095, zoledronic acid/zoledronate zometa® , alendronate fosamax®, pamidronate aredia®, tiludronate skelid® or risedronate actonel®; and troxacitabine (1,3-dioxolane cytosine nucleoside analogue); antisense oligonucleotides, especially those that inhibit gene expression in signaling pathways involved in impaired cell proliferation , such as, for example, PKCalpha, Raf, H-Ras and epidermal growth factor receptor (EGF-R), vaccines such as theratop® vaccine and gene therapy vaccines, for example, allovectin® vaccine, leuvectin vaccine and Vaxid® vaccine; topoisomerase 1 inhibitor Lurtotecan®; rmRH abarelix®; lapatinib ditosylate (small molecule inhibitor of ErbB-2 and EGFR dual tyrosine kinase, also known as GW572016) and pharmaceutically acceptable salts, acids or derivatives of any of the above.

A growth inhibitory agent generally refers to a compound or composition that inhibits cell growth in vitro or in vivo. Thus, a growth inhibitory agent may be an agent that significantly reduces the percentage of cells in S phase. Examples of growth inhibitory agents include agents that block cell cycle progression (in a state other than S phase), agents such as agents that induce G1 phase arrest and M phase arrest. Classic phase M blockers include the vinca alkaloids (vincristine and vinblastine), taxanes, and topoisomerase II inhibitors such as doxorubicin, epirubicin, daunorubicin, etoposide, and bleomycin. Agents that cause arrest in G1 phase also cause arrest in S phase, for example, DNA alkylating agents such as tamoxifen, prednisone, dacarbazine, mechlorethamine, cisplatin, methotrexate, 5-fluorouracil, and ara-C. Additional information can be found in The Molecular Basis of Cancer, Mendelsohn and Israel, eds., Chapter 1, entitled Cell cycle regulation, oncogenes, and antineoplastic drugs by Murakami et al. (WB Saunders: Philadelphia, 1995), especially p. 13. Taxanes (paclitaxel and docetaxel) are anticancer agents, both of which are derived from the yew tree. Docetaxel (Taxotere®, Rhone -Poulenc Rorer), derived from European yew, is a semisynthetic analogue of paclitaxel (Taxol®, Bristol-Myers Squibb). Paclitaxel and docetaxel promote the assembly of microtubules from tubulin dimers and stabilize microtubules by preventing depolymerization, resulting in inhibition of cell mitosis.

The ActRIIB antagonist and the additional active ingredient (eg, a co-administered therapeutic agent) can be administered in the same composition or separately. When administered separately, the ActRIIB antagonist can be administered before, after, or simultaneously with the additional active substance. Administration of one substance may precede or follow administration of another substance at intervals ranging from a few minutes to several weeks. In embodiments where two or more therapeutic agents of different types are separately administered to a subject, care is generally taken to ensure that no significant time elapses between each administration so that the different types of agents can have an effective combined effect on the target tissues or cells.

Some embodiments of the present invention provide methods for managing a patient who has been treated or who is a candidate for treatment with one or more ActRIIB antagonist agents of the invention by measuring one or more hematologic parameters in the patient. Hematological indicators can be used to assess the appropriate dosage for a patient who is a candidate for treatment with an antagonist of this invention, to monitor hematological indicators during treatment, to assess the need for selection

- 65 043914 dose during treatment with one or more than one antagonist of the invention and/or to evaluate the appropriate maintenance dose of one or more than one antagonist of the invention. If one or more than one hematologic parameters are outside the normal range, the dose of one or more ActRIIB antagonists may be reduced, delayed, or discontinued.

Hematological parameters that can be determined in accordance with the proposed methods include, for example, red blood cell count, blood pressure, iron stores, and other agents found in body fluids that correlate with elevated red blood cell counts using methods recognized in the art. Such indicators can be determined using a patient's blood sample. Increases in red blood cell counts, hemoglobin levels, and/or hematocrit levels may cause an undesirable increase in blood pressure.

In one embodiment, if one or more hematologic parameters are outside the normal range or at the upper limit of normal in a patient who is a candidate for treatment with one or more ActRIIB antagonists, the initiation of one or more antagonists as described may be delayed until hematological parameters return to normal or acceptable levels, either naturally or through therapeutic intervention. For example, if a candidate patient has hypertension or prehypertension, the patient may be treated with a blood pressure lowering agent to lower the patient's blood pressure. Any blood pressure-lowering agent appropriate for the patient's individual condition may be used, including, for example, diuretics, adrenergic receptor inhibitors (including alpha blockers and beta blockers), vasodilators, calcium channel blockers, angiotensin-converting enzyme (ACE) inhibitors, or blockers. angiotensin II receptors. Alternatively, blood pressure can be controlled through diet and exercise. Likewise, if a candidate patient's iron stores are below normal or at the lower limit of normal, the patient can be treated with an appropriate diet and/or iron supplementation until the patient's iron stores return to normal or acceptable levels. For patients whose red blood cell count and hemoglobin level are above normal, administration of one or more than one antagonists of the invention may be delayed until these values return to normal or acceptable levels.

In some embodiments, if one or more hematological parameters are outside the normal range or at the upper limit of normal in a patient who is a candidate for treatment with one or more ActRIIB antagonists, the initiation of administration of one or more antagonists as described may not be delayed. . However, the dose level or frequency of administration of one or more than one antagonists of the invention may be adjusted to reduce the risk of unacceptable increases in hematologic parameters encountered when administering one or more antagonists of the invention. Alternatively, a treatment regimen may be developed for a patient in which one or more ActRIIB antagonists are combined with a therapeutic agent aimed at normalizing the undesirable hematologic parameters. For example, if a patient has high blood pressure, a treatment regimen may be developed that includes administration of one or more than one ActRIIB antagonist agent and a blood pressure lowering agent. For a patient who has less than desirable iron stores in the body, a treatment regimen may be developed that includes one or more ActRIIB antagonists of the invention and an iron supplement.

In one embodiment, a patient who is a candidate for treatment with one or more ActRIIB antagonists of the invention is assessed with baseline values of one or more hematological parameters and, based on the baseline values, an administration regimen suitable for the patient is developed. Alternatively, baseline values determined from the patient's history can be used to inform the appropriate antagonist regimen for the patient. For example, if a healthy patient is found to have a baseline blood pressure above the established normal range, it may not be necessary to bring the patient's blood pressure into the range considered normal for the general population before treatment with one or more than one antagonist of the invention. The patient's baseline values of one or more hematologic parameters prior to treatment with one or more ActRIIB antagonists of the invention may be used as appropriate reference values when monitoring any changes in hematologic parameters during treatment with one or more than one antagonists of the invention.

In some embodiments, one or more hematological parameters are determined in patients who are receiving treatment with one or more ActRIIB antagonists. Hematological parameters can be used to monitor a patient during treatment and allow the administration of one or more than one antagonist of the invention to be adjusted or discontinued or another therapeutic agent to be additionally administered. For example, if administration of one or more ActRIIB antagonists results in an increase in blood pressure, red blood cell count, or hemoglobin level or a decrease in body iron stores, administration of one or more

- 66 043914 A particular antagonist of the invention may be reduced in amount or frequency to reduce the effect of one or more than one antagonist of the invention on one or more hematological parameters. If the administration of one or more than one ActRIIB antagonists results in a change in one or more hematological parameters that is undesirable for the patient, the administration of one or more than one antagonists of the invention may be discontinued either temporarily until the hematological parameters return to an acceptable level, or forever. Likewise, if one or more hematological parameters do not come within an acceptable range when the dose or frequency of administration of one or more than one antagonists of the invention is reduced, administration may be discontinued. Alternatively, or in addition to reducing or stopping the administration of one or more than one antagonists of the invention, an additional therapeutic agent may be administered to the patient that is intended to normalize undesirable hematologic values, such as, for example, a blood pressure lowering agent or an iron supplement. For example, if a patient receiving treatment with one or more ActRIIB antagonists has elevated blood pressure, administration of one or more antagonists of the invention may be continued at the same level and a blood pressure lowering agent may be added to the treatment regimen, administration of one or more than one antagonist of the invention may be reduced (eg, in amount and/or frequency) and a blood pressure lowering agent added to the treatment regimen, or administration of one or more antagonists of the invention may be discontinued and the patient treated with a blood pressure lowering agent.

As used herein, the terms in combination with, combinations or co-administration refer to any form of administration such that additional therapeutic agents (eg, second, third, fourth, etc.) are still effective in the body (eg, multiple compounds simultaneously effective in the patient, including the synergistic effects of these compounds). Efficacy may not correlate with measured blood, serum, or plasma concentrations of the agent. For example, different therapeutic compounds can be administered either in the same composition or in different compositions, simultaneously or sequentially and according to different schedules. Thus, an individual receiving such treatment will benefit from the combined effects of various therapeutic agents. One or more ActRIIB antagonists or combinations of such polypeptides of the invention can be administered simultaneously, before or after the administration of one or more additional agents or maintenance therapies. In general, each therapeutic agent will be administered at a dose and/or time specified for that particular agent. The specific combination used in the regimen will take into account the compatibility of the antagonist of this invention with the therapy and/or desired.

5. Pharmaceutical compositions.

The therapeutic agents described herein (eg, ActRIIB antagonist) can be formulated as pharmaceutical compositions. Pharmaceutical compositions for use herein can be formulated in a standard manner using one or more physiologically acceptable carriers or excipients. Such compositions will be substantially pyrogen-free, in accordance with most regulatory requirements.

In some embodiments, the therapeutic method described includes administering the composition systemically or locally in the form of an implant or device. When administered, the therapeutic composition for use herein is pyrogen-free and is in a physiologically acceptable form. Therapeutically effective agents other than ActRIIB signaling pathway antagonists, which may be formulated as described above, can be administered simultaneously or sequentially with the subject compounds (eg, ActRIIB polypeptides) in the methods described herein.

Typically, protein therapeutic agents herein will be administered parenterally and, in particular, intravenously or subcutaneously. Pharmaceutical compositions suitable for parenteral administration may contain one or more than one ActRIIB antagonist in combination with one or more pharmaceutically acceptable sterile isotonic aqueous or non-aqueous solution, dispersion, suspension or emulsion or sterile powder from which sterile solutions can be prepared for injection or dispersion immediately before use, which may contain antioxidants, buffers, bacteriostatic agents, solutes to render the composition isotonic with the blood of the intended recipient, or suspending agents or thickening agents. Examples of suitable aqueous or non-aqueous carriers that may be used in the pharmaceutical compositions herein include water, ethanol, polyols (such as glycerin, propylene glycol, polyethylene glycol, etc.) and suitable mixtures thereof, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. To maintain proper fluidity, you can, for example, use coatings such as lecithin, maintain the required particle size in the case of dispersions, and also use surfactants.

If necessary, the compositions and formulations can be placed in a package or dispenser, which can contain one or more unit dosages containing the active ingredient. For example,

- 67 043914 the packaging may contain metal foil or plastic film, for example as a blister pack. The packaging or dispenser may include accompanying instructions for use.

In addition, the composition may be in encapsulated or injectable form for delivery to the target tissue site. In some embodiments, the compositions of this invention may include a matrix capable of delivering one or more therapeutically active compounds (eg, ActRIIB polypeptides) to a target tissue site, imparting structure to developing tissue, and possibly being resorbable by the body. For example, the matrix may provide a slow release of the ActRIIB antagonist. Such matrices can be formed by materials currently used for other implantable medical devices.

The choice of matrix material is based on biocompatibility, biodegradability, mechanical properties, appearance and properties of the contacting surface. The specific application of the compositions discussed will determine the required composition. Possible matrices for the compositions include biodegradable and chemically characterized calcium sulfate, tricalcium phosphate, hydroxyapatite, polylactic acid and polyanhydrides. Other possible materials are biodegradable and biologically well characterized, such as bone or skin collagen. Other matrices consist of pure proteins or extracellular matrix components. Other possible matrices are non-biodegradable and chemically characterized, such as sintered hydroxyapatite, bioceramics, aluminates or other types of ceramics. The matrices may consist of combinations of any of the above-mentioned types of material, such as polylactic acid and hydroxyapatite or collagen and tricalcium phosphate. Bioceramics may have a modified composition, such as calcium aluminate phosphate, and be processed to change pore size, particle size, particle shape, and biodegradability.

In some embodiments, the methods of the invention involve oral administration, for example, in the form of capsules, sachets, pills, tablets, lozenges (using a flavored base, typically sucrose, gum or tragacanth), powders, granules, or in the form of solutions or suspensions in aqueous or non-aqueous liquid, or in the form of oil-in-water or water-in-oil emulsions, or in the form of an elixir or syrup, or in the form of lozenges (using an inert base such as gelatin or glycerin, or sucrose and gum) and/or in the form of mouth rinses and the like, each of which contains a predetermined amount of agent as an active ingredient. The agent can also be administered as a bolus, electuary, or paste.

In solid oral dosage forms (capsules, tablets, pills, dragees, powders, granules, etc.), one or more than one therapeutically active compound of this invention may be admixed with one or more pharmaceutically acceptable carriers, such as sodium citrate or dicalcium phosphate and/or any of the following: (1) fillers or bulking agents such as starches, lactose, sucrose, glucose, mannitol and/or silicic acid; (2) binders, such as, for example, carboxymethylcellulose, alginates, gelatin, polyvinylpyrrolidone, sucrose and/or gum; (3) humectants such as glycerin; (4) leavening agents such as agar-agar, calcium carbonate, potato or tapioca starch, alginic acid, some silicates and sodium carbonate; (5) dissolution retardants such as paraffin; (6) absorption accelerating substances such as quaternary ammonium compounds; (7) wetting agents, such as cetyl alcohol and glycerol monostearate; (8) absorbents such as kaolin and bentonite clay; (9) lubricants such as talc, calcium stearate, magnesium stearate, solid polyethylene glycols, sodium lauryl sulfate and mixtures thereof; and (10) dyes. In the case of capsules, tablets and pills, the pharmaceutical compositions may also contain buffering agents. Solid compositions of a similar type can also be used as fillers in soft and hard gelatin capsules using excipients such as lactose or milk sugars, as well as high molecular weight polyethylene glycols and the like.

Liquid dosage forms for oral administration include pharmaceutically acceptable emulsions, microemulsions, solutions, suspensions, syrups and elixirs. In addition to the active ingredient, liquid dosage forms may contain inert diluents commonly used in the art, such as water or other solvents, solubilizing agents and emulsifiers, such as ethyl alcohol, isopropyl alcohol, ethyl carbonate, ethyl acetate, benzyl alcohol, benzyl benzoate, propylene glycol, 1. 3-butylene glycol, oils (in particular cottonseed, groundnut, corn, sprout, olive, castor and sesame oils), glycerin, tetrahydrofuryl alcohol, polyethylene glycols and sorbitan fatty acid esters and mixtures thereof. In addition to inert diluents, oral compositions may also contain adjuvants such as wetting agents, emulsifying agents and suspending agents, sweeteners, flavoring agents, colors, flavoring agents and preservatives.

Suspensions, in addition to the active compounds, may contain suspending agents such as ethoxylated isostearyl alcohols, polyoxyethylene sorbitol and sorbitan ethers, microcrystalline cellulose, aluminum metahydroxide, bentonite, agar-agar and tragacanth and mixtures thereof.

The compositions of the invention may also contain adjuvants such as preservatives, wetting agents, emulsifying agents and dispersing agents. Action of microorganisms

- 68 043914 is prevented by including various antibacterial and antifungal agents, for example paraben, chlorobutanol, phenol, sorbic acid and the like. It may also be desirable to include isotonic agents such as sugars, sodium chloride and the like in the compositions. In addition, prolonged absorption of injectable pharmaceutical forms can be achieved through the inclusion of agents that delay absorption, such as aluminum monostearate and gelatin.

It is understood that the schedule of administration will be determined by the attending physician, taking into account various factors that modify the effect of the subject compounds of the invention (eg, TpRII polypeptides). Various factors include, but are not limited to, patient age, sex and diet, severity of disease, time of administration, and other clinical factors. It is possible that the dosage will vary depending on the matrix used for dissolution and the types of compounds in the composition. The addition of other known growth factors to the final composition may also influence the dosage. The condition can be monitored by periodic assessment of bone growth and/or repair, for example, by radiological examination (including dual-energy x-ray absorptiometry), histomorphometric studies and tetracycline labeling.

Some embodiments of the present invention also provide gene therapy for the production of an ActRIIB antagonist in vivo. The therapeutic effect of such therapy will be achieved through the introduction of ActRIIB polynucleotide sequences into cells or tissues that have disorders, as described above. Delivery of ActRIIB polynucleotide sequences can be accomplished using a recombinant expression vector, such as a chimeric virus, or a colloidal dispersion system. Preferred for the therapeutic delivery of ActRIIB polynucleotide sequences is the use of specific liposomes.

Various virus-based vectors that can be used for gene therapy herein include adenovirus, herpes virus, vaccinia virus, or preferably an RNA virus such as a retrovirus. Preferably, the retroviral vector is derived from a rodent or avian retrovirus. Examples of retroviral vectors into which a single foreign gene can be inserted include, but are not limited to: Moloney murine leukemia virus (MoMuLV), Harvey mouse sarcoma virus (HaMuSV), mouse mammary tumor virus (MuMTV), and Rous sarcoma virus (RSV). A number of additional retroviral vectors may include multiple genes. All of these vectors can carry or include a gene for a selectable marker so that transduced cells can be detected and produced. Retroviral vectors can be target specific by attaching, for example, a sugar, glycolipid or protein. Preferential targeting of the target is achieved using an antibody. Those skilled in the art will appreciate that specific polynucleotide sequences can be inserted into the retroviral genome or attached to the viral envelope to allow specific delivery of a retroviral vector containing an ActRIIB polynucleotide. In a preferred embodiment, the vector targets bone or cartilage.

Alternatively, tissue culture cells can be directly transfected with plasmids encoding the retroviral structural genes gag, pol and env using standard calcium phosphate transfection. These cells are then transfected with a plasmid vector containing the genes of interest. The resulting cells release the retroviral vector into the culture medium.

Another system for targeted delivery of ActRIIB antagonist polynucleotides is a colloidal disperse system. Colloidal dispersions include complex macromolecules, nanocapsules, microspheres, particles, and lipid-based systems, including oil-in-water emulsions, micelles, mixed micelles, and liposomes. The preferred colloidal system of this invention is a liposome. Liposomes are artificial membrane vesicles that can be used as delivery vehicles in vitro and in vivo. RNA, DNA and intact virions can be encapsulated in an aqueous internal environment and delivered to cells in a biologically active form (see, for example, Fraley, et al., Trends Biochem. Sci., 6:77, 1981). Methods are known in the art for efficient gene transfer using liposomes as carriers, see, for example, Mannino, et al., Biotechniques, 6:682, 1988. The composition of liposomes is usually a combination of phospholipids, usually in combination with steroids, in particularly with cholesterol. Other phospholipids or other lipids can also be used. The physical characteristics of liposomes depend on pH, ionic strength, and the presence of divalent cations.

Examples of lipids that can be used in the production of liposomes include phosphatidyl compounds such as phosphatidylglycerol, phosphatidylcholine, phosphatidylserine, phosphatidylethanolamine, sphingolipids, cerebrosides and gangliosides. Illustrative examples of phospholipids include egg phosphatidylcholine, dipalmitoylphosphatidylcholine, and distearoylphosphatidylcholine. Targeted delivery of liposomes is also possible, for example, according to the principle of organ specificity, cell specificity and organelle specificity, and is known in the art.

The description describes compositions that may vary due to the inclusion of acids and bases to adjust the pH, as well as buffering agents to maintain the pH within a narrow range.

- 69 043914

Examples of implementation of the invention

The invention described above in general terms will be better understood by the following examples, which are included solely for the purpose of illustrating certain embodiments of the invention and are not intended to limit the invention.

Example 1: Generation of ActRIIB-Fc fusion proteins.

Applicants have constructed an ActRIIB fusion protein having the extracellular domain of human ActRIIB fused to a human or mouse Fc domain with a minimal linker (three glycine amino acids) in between. The constructs were designated ActRIIB-hFc and ActRIIB-Fc, respectively.

ActRIIB-hFc below was isolated from CHO cell line (SEQ ID NO: 24) GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVKKGC

WLDDFNCYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEVTYEPPPTAPTGG GTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVD GVEVHNAI<TI<PREEOYNSTYRVVSVLTVLHODWLNGI<EYI<CI<VSNI<ALPVPIEI<TISI< AKGOPREPOVY TLPPSREEMTKNOVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPV LDSDGSFFLYSKLTVDKSRWOOGNVFSCSVMHEALHNHYTQKSLSLSPGK

ActRIIB-hFc and ActRIIB-Fc proteins were expressed in the CHO cell line. Three different leader sequences were considered: honeybee mellitin (HBML): MKFLVNVALVFMVVYISYIYA (SEQ ID NO: 21), ii) tissue plasminogen activator (TPA): MDAMKRGLCCVLLLCGAVFVSP (SEQ ID NO: 22) and (iii) native: MGAAAKLAFAVFLISCSSGA (SEQ ID NO : 23).

The selected form has a TPA leader sequence and has the following full-length amino acid sequence (SEQ ID NO: 25)

MDAMKRGLCCVLLLCGAVFVSPGASGRGEAETRECIYYNANWELERTNOSGLERCEGE QDKRLHCYASWRNSSGTIELVKKGCWLDDFNCYDRQECVATEENPQVYFCCCEGNFC NERFTHLPEAGGPEVTYEPPPTAPTGGGTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISR TPEVTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPREEOYNSTYRVVSVLTVLHQD WLNGKEYKCKVSNKALPVPIEKTISKAKGOPREPOVYTLPPSREEMTKNOVSLTCLVKG FYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQOGNVFSCSVMHE ALHNHYTQKSLSLSPGK

The specified polypeptide is encoded by the following nucleic acid sequence (SEQ ID NO: 26)

A TGGATGCAAT GAAGAGAGGG CTCTGCTGTG TGCTGCTGCT GTGTGGAGCA GTCTTCGTTT CGCCCGGCGC CTCTGGGCGT GGGGAGGCTG AGACACGGGA GTGCATCTAC TACAACGCCA ACTGGGAGCT GGAGCGCACC AACCAGAGCG GCCTGGAGCG CTGCGAAGGC GAGCAGGACA AGCGGCTGCA CTGCTACGCC TCCTGGCG CA ACAGCTCTGG CACCATCGAG CTCGTGAAGA AGGGCTGCTG GCTAGATGAC TTCAACTGCT ACGATAGGCA GGAGTGTGTG GCCACTGAGG AGAACCCCCA GGTGTACTTC TGCTGCTGTG AAGGCAACTT CTGCAACGAG CGCTTCACTC ATTTGCCAGA GGCTGGGGGC CCGGAAGTCA CGTACGAGCC ACCCCCGACA GCCCCACCG GTGGTGGA AC TCACACATGC CCACCGTGCC CAGCACCTGA ACTCCTGGGG GGACCGTCAG TTTTSSTT SSSSSSSAAA CCCAAGGACA CCCTCATGAT CTCCCGGACC CCTGAGGTCA CATGCGTGGT GGTGGACGTG AGCCACGAAG ACCCTGAGGT CAAGTTCAAC TGGTACGTGG ACGGCGTGGA GGTGCATAAT GCCAAGACAA AGCCGCGGGA GGAGCAGTAC AACAGCACGT ACCGTGTGGT CAGCGTCCTC ACCGTCCTGC ACCAGGACTG GCTGAAT GGC AAGGAGTACA AGTGCAAGGT STSSAASAAA GCCCTCCCAG TCCCCATCGA GAAAACCATC TCCAAAGCCA AAGGGCAGCC CCGAGAACCA CAGGTGTACA CCCTGCCCCC ATCCCGGGAG GAGATGACCA AGAACCAGGT CAGCCTGACC TGCCTGGTCA AAGGCTTCTA TCCCAGCGAC ATCGCCGTGG AGTGGGAGAG CAATGGGCAG CCGGAG AACA ACTACAAGAC CACGCCTCCC GTGCTGGACT CCGACGGCTC STTSTTTSTS TATAGCAAGC TCACCGTGGA CAAGAGCAGG TGGCAGCAGG GGAACGTCTT CTCATGCTCC GTGATGCATG AGGCTCTGCA SAASSASTAS ACGCAGAAGA GCCTCTCCCT GTCTCCGGGT AAATGA

N-terminal sequencing of the CHO cell-derived material revealed the major sequence -GRGEAE (SEQ ID NO: 27). It should be noted that the structures described in the literature begin with the sequence -SGR...

Purification can be accomplished in a series of column chromatography steps including, for example, three or more of the following, in any order: protein A chromatography, Q-Sepharose chromatography, phenyl Sepharose chromatography, size exclusion chromatography, and cation exchange chromatography. Cleaning can be completed by virus filtration and buffer replacement.

ActRIIB-Fc fusion proteins were also expressed in HEK293 cells and COS cells. Although material from all cell lines and suitable culture conditions produced a protein with in vivo muscle-building activity, its potency varied, possibly due to the choice of cell line and/or culture conditions.

- 70 043914

Applicants created a series of mutations in the extracellular domain of ActRIIB and produced these mutant proteins as soluble proteins fused between the extracellular domain of ActRIIB and the Fc domain. The ActRIIB-Fc fusion protein has the sequence SEQ ID NO: 24.

The ActRIIB-Fc protein has been subjected to various mutations, including truncation at the N- and C-terminus. Based on the data presented, it is assumed that these constructs, when expressed with the TPA leader sequence, will lack the N-terminal series. Mutations in the extracellular domain of ActRIIB were obtained by PCR mutagenesis. After PCR, fragments were purified on Qiagen columns, digested with SfoI and Agel, and purified by gel electrophoresis. These fragments were ligated into the expression vector pAID4 (see WO2006/012627) so that upon ligation they formed a fusion protein with human IgG1. After transformation with E. coli DH5 alpha, individual colonies were selected and DNA isolated. For mouse constructs (mFc), human IgG1 was replaced with mouse IgG2a. The sequences of all mutants were verified.

All mutants were obtained by transient transfection of HEK293F cells. Briefly, HEK293F cells at a concentration of 6x10 ⁵ cells/ml in Freestyle medium (Invitrogen) in a volume of 250 ml were grown overnight in 500 ml flasks with constant stirring. The next day, the DNA:PEI complex (1:1) was added to the cells to a final DNA concentration of 0.5 μg/ml. After 4 hours, 250 ml of medium was added and the cells were cultured for 7 days. The conditioned medium was collected by centrifugation of the cells and concentrated.

Mutants were isolated by various methods, including, for example, Protein A column chromatography and eluted with low pH glycine buffer (3.0). After neutralization, they were dialyzed against PBS.

Mutants were also produced in cells in a similar manner. Mutants were tested in binding assays and/or bioassays described in WO 2008/097541 and WO 2006/012627, incorporated by reference. In some cases, the assay was performed with conditioned media rather than purified proteins. Additional variants of ActRIIB are described in US Pat. No. 7,842,663.

The Applicant has produced an ActRIIB(25-131)-hFc fusion protein which contains the extracellular domain of human ActRIIB truncated at the N-terminus and C-terminus (residues 25-131 of the native protein SEQ ID NO: 1) fused at the N-terminus to a leader sequence TPA replacing the native ActRIIB leader sequence and the C-terminus with the human Fc domain via a minimal linker (three glycine residues) (Fig. 12). The nucleotide sequence encoding the fusion protein is shown in FIG. 13. Applicant modified the codons and discovered a nucleic acid variant encoding the ActRIIB (25-131)-hFc protein that provided a significant improvement in the expression levels of the original transformants (FIG. 14).

The mature protein has the following amino acid sequence (N-terminus verified by N-terminal sequencing) (SEQ ID NO: 28)

ETRECIYYNANWELERTNQS GLERCEGEQD KRLHCYASWRNSSGTIELVK

KGCWLDDFNC YDRQECVATE ENPQVYECCC EGNFCNERFT HLPEAGGPEV

TYEPPPTGGG THTCPPCPAP ELLGGPSVFL FPPKPKDTLM ISRTPEVTCV

VVDVSHEDPE VKFNWYVDGV EVHNAKTKPR EEQYNSTYRV VSVLTVLHQD

WLNGKEYKCK VSNKALPAPIEKTISKAKGQ PREPQVYTLP PSREEMTKNQ

VSLTCLVKGF YPSDIAVEWE SNGQPENNYK TTPPVLDSDG SFFLYSKLTV

DKSRWQQGNV FSCSVMHEAL HNHYTQKSLS LSPGK

Amino acids 1-107 come from ActRIIB.

The expressed molecule is purified through a series of column chromatography steps, including, for example, three or more of the following, in any order: protein A chromatography, Q-Sepharose chromatography, phenyl Sepharose chromatography, size exclusion chromatography, and cation exchange chromatography. The purification could be completed by virus filtration and buffer replacement.

The affinities of several ligands of ActRIIB(25-131)-hFc and its full-length analogue ActRIIB(20134)-hFc were assessed using the Biacore™ biosensor, the results are summarized in the table below. The values of the equilibrium constant Kd were calculated, since the very rapid association and dissociation of the complex did not make it possible to accurately determine k _on and k _off . ActRIIB(25-131)-hFc bound activin A, activin B and GDF11 with high affinity.

Affinity of ActRIIB-hFc forms to ligands

Fused design Activin A (e-11) Activin B (e-11) GDF11 (e-P) ActRIIB (20-134)-hFс 1.6 1.2 3.6 ActRIIB(25-131)-hFc 1.8 1.2 3.1

Example 2: Creating a GDF trap.

The applicant has designed a GDF trap as follows. A polypeptide having a modified extracellular domain of ActRIIB (amino acids 20-134 SEQ ID NO: 1 with substitution L79D), having significantly less binding to activin A compared to GDF11 and/or myostatin (due to the substitution of leucine for aspartate at position 79 SEQ ID NO: 1) was fused to a human or mouse Fc domain with a minimal linker (three glycine amino acids) between them. Constructions

- 71 043914 designated ActRIIB(L79D 20-134)-hFc and ActRIIB(L79D 20-134)-Fc, respectively. Alternative forms with glutamate instead of aspartate at position 79 (L79E) showed similar characteristics. Alternative forms were also created with alanine instead of valine at position 226 relative to SEQ ID NO: 44 below, which were equivalent in all respects. The aspartate at position 79 (relative to SEQ ID NO: 1, or at position 60 relative to SEQ ID NO: 29) is indicated by a double underline. Valine at position 226 relative to SEQ ID NO: 29 is also indicated by a double underline below.

The GDF trap ActRIIB(L79D 20-134)-hFc, which was isolated from CHO cells, is shown below (SEQ ID NO: 29).

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVKKGC

WDDDFNCYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEVTYEPPPTAPTG

GGTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYV

DGVEVHNAI<TI<PREEOYNSTYRVVSVLTVLHODWLNGI<EYI<CI<VSNI<ALPVPIEI<TIS

KAKGOPREPOVYTLPPSREEMTKNOVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPP

VLDSDGSFFLYSKLTVDKSRWOOGNVFSCSVMHEALHNHYTQKSLSLSPGK

The ActRIIB-derived portion of the GDF trap has the amino acid sequence shown below (SEQ ID NO: 30), and this portion could be used as a monomer or as a non-Fc fusion protein, as a monomer, dimer, or higher complex order.

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELV

KKGCWDDDFNCYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEVTYEPPPT APT (SEQ Ш NO: 30)

The GDF decoy protein was expressed in CHO cells. Three different leader sequences were considered:

(i) honey bee mellitin (HBML), (ii) tissue plasminogen activator (TPA) and (iii) native.

The selected form has a TPA leader sequence and has the following full-length amino acid sequence:

MDAMKRGLCCVLLLCGAVFVSPGASGRGEAETRECIYYNANWELERTNOSGLERCEGE

QDKRLHCYASWRNSSGTIELVKKGCWDDDFNCYDRQECVATEENPQVYFCCCEGNFC

NERFTHLPEAGGPEVTYEPPPTAPTGGGTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISR

TPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEOYNSTYRVVSVLTVLHQD

WLNGKEYKCKVSNKALPAPIEKTISKAKGOPREPOVYTLPPSREEMTKNOVSLTCLVKG

FYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHE

ALHNHYTQKSLSLSPGK (SEQ Ш NO: 31)

The specified polypeptide is encoded by the following nucleic acid sequence (SEQ ID NO: 32)

A TGGATGCAAT GAAGAGAGGG CTCTGCTGTG TGCTGCTGCT GTGTGGAGCA

GTCTTCGTTT CGCCCGGCGC CTCTGGGCGT GGGGAGGCTG AGACACGGGA

GTGCATCTAC TACAACGCCA ACTGGGAGCT GGAGCGCACC AACCAGAGCG

GCCTGGAGCG CTGCGAAGGC GAGCAGGACA AGCGGCTGCA CTGCTACGCC

TCCTGGCGCA ACAGCTCTGG CACCATCGAG CTCGTGAAGA AGGGCTGCTG

GGACGATGAC TTCAACTGCT ACGATAGGCA GGAGTGTGTG GCCACTGAGG

AGAACCCCCA GGTGTACTTC TGCTGCTGTG AAGGCAACTT CTGCAACGAG

CGCTTCACTC ATTTGCCAGA GGCTGGGGGC CCGGAAGTCA CGTACGAGCC

ACCCCCGACA GCCCCCACCG GTGGTGGAAC TCACACATGC CCACCGTGCC

CAGCACCTGA ACTCCTGGGG GGACCGTCAG TSTTTSSTTT SSSSSSAAAA

CCCAAGGACA CCCTCATGAT CTCCCCGGACC CCTGAGGTCA CATGCGTGGT

GGTGGACGTG AGCCACGAAG ACCCTGAGGT CAAGTTCAAC TGGTACGTGG

ACGGCGTGGA GGTGCATAAT GCCAAGACAA AGCCGCGGGA GGAGCAGTAC

AACAGCACGT ACCGTGTGGT CAGCGTCCTC ACCGTCCTGC ACCAGGACTG

GCTGAATGGC AAGGAGTACA AGTGCAAGGT STSSAASAAA GCCCTCCCAG

TCCCCATCGA GAAAACCATC TCCAAAGCCA AAGGGCAGCC CCGAGAACCA

CAGGTGTACA CCCTGCCCCC ATCCCGGGAG GAGATGACCA AGAACCAGGT

CAGCCTGACC TGCCTGGTCA AAGGCTTCTA TCCCAGCGAC ATCGCCGTGG

AGTGGGAGAG CAATGGGCAG CCGGAGAACA ACTACAAGAC CACGCCTCCC

GTGCTGGACT CCGACGGCTC STTSTTTSTS TATAGCAAGC TCACCGTGGA

CAAGAGCAGG TGGCAGCAGG GGAACGTCTT CTCATGCTCC GTGATGCATG

AGGCTCTGCA SAASSASTAS ACGCAGAAGA GCCTCTCCCT GTCTCCGGGT

AAATGA

Purification can be accomplished in a series of column chromatography steps including, for example, three or more of the following, in any order: protein A chromatography, Q-Sepharose chromatography, phenyl Sepharose chromatography, size exclusion chromatography, and cation exchange chromatography. The purification could be completed by virus filtration and buffer replacement. In the example purification scheme, the culture medium was passed through a Protein A column, washed with 150 mM Tris/NaCl (pH 8.0), then washed with 50 mM Tris/NaCl (pH 8.0) and eluted with 0.1 M glycine,

- 72 043914 pH 3.0. The low pH eluate was left at room temperature for 30 minutes as a virus clearance step. The eluate was then neutralized and passed through a QSepharose ion exchange column and washed with 50 mM Tris pH 8.0, 50 mM NaCl and eluted with 50 mM Tris pH 8.0 with a NaCl concentration of 150 mM to 300 mM. The eluate buffer was then replaced with 50 mM Tris pH 8.0, 1.1 M ammonium sulfate and passed through a phenyl Sepharose column, and eluted with 50 mM Tris pH 8.0 with 150 to 300 mM ammonium sulfate. The eluate was dialyzed and filtered for use.

Additional GDF decoys (ActRIIB-Fc fusion proteins modified to reduce activin A binding relative to myostatin or GDF11 binding) are described in WO 2008/097541 and WO 2006/012627, incorporated by reference.

Example 3: Biological Study of GDF-11 and Activin-Mediated Signal Transduction.

To evaluate the effect of ActRIIB-Fc proteins and GDF traps on GDF-11 and Activin A signal transduction, a reporter gene study was performed in A-204 cells. Cell line: human rhabdomyosarcoma cells (from muscle). Reporter vector: pGL3(CAGA)12 (described by Dennler et al, 1998, EMBO 17: 3091-3100). The CAGA12 motif is present in TGF-beta-dependent genes (for example, in the PAI-1 gene), so this vector is usually used for factors in which SMAD2 and 3 are involved in the signaling pathway.

day: A-204 cells are planted in a 48-well plate.

day: A-204 cells are transfected with 10 μg of pGL3(CAGA)12 or pGL3(CAGA)12 (10 μg) + pRLCMV (1 μg) using Fugene.

day: add factors (diluted in medium + 0.1% BSA). Inhibitors must be preincubated with factors for 1 hour before adding to cells. After six hours, cells are washed with PBS and lysed.

A luciferase assay is then performed. In the absence of any inhibitors, activin A demonstrated a 10 ^- fold activation of reporter gene expression and an ED50 of ~2 ng/ml. GDF-11: 16-fold activation, ED50 ~ 1.5 ng/ml.

In this assay, ActRIIB(20-134) is a potent inhibitor of activin A, GDF-8 and GDF-11. As described below, ActRIIB variants were also examined in this analysis.

Example 4. ActRIIB-Fc variants, activity in a cell study.

The activity of ActRIIB-Fc and GDF trap proteins was assayed in cells as described below. The results are presented in the table below. Some variants were investigated in various C-terminal truncated constructs. As discussed above, cutting off five or fifteen amino acids resulted in decreased activity. GDF traps (variants L79D and L79E) showed a significant reduction in activin A inhibition while maintaining GDF-11 inhibition at essentially wild-type levels.

Binding of soluble ActRIIB-Fc to GDF11 and activin A.

+ Low activity (approx. 1 x 10 ^-6 K), ++ moderate activity (approx. 1 x 10 ^-7 K), +++ good activity (wild type) (approx. 1 x 10 ^-8 Ki), ++++ activity, superior to that of the wild type.

The half-life in rat serum was studied for several variants. The serum half-life of ActRIIB(20-134)-Fc is approximately 70 hours. The serum half-life of ActRIIB(A24N 20-134)Fc is approximately 100-150 hours. Any of the options explored above can be combined with GDF decoy molecules.

- 73 043914

Example 5: Binding of GDF-11 and Activin A.

The binding of several ActRIIB-Fc proteins and GDF decoys to ligands was examined on the Biacore™ biosensor.

Immobilization of ActRIIB-Fc variants or wild-type protein in the system was carried out using an antibody to hFc. Ligands were injected and passed over the surface of immobilized receptor proteins. The results are presented in the tables below.

Ligand binding specificity of IIB variants.

GDF11 Protein Kon (1/Mc) Koff (1/c) KD(M) ActRIIB(20-134)-hFc l,34e-6 l,13e-4 8.42e-ll ActRIIB(A24N 20-134)-hFc l,21e-6 6.35e-5 5.19e-ll ActRIIB(L79D 20-134)-hFc 6.7e-5 4.39e-4 6.55e-10 ActRIIB(L79E 20-134)-hFc 3.8e-5 2.74e-4 7.16e-10 ActRIIB(R64K 20-134)-hFc 6.77e-5 2.4 le-5 3.56e-ll GDF8 Protein Kon (1/Mc) Koff (1/c) KD(M) ActRIIB(20-134)-hFc 3.69e-5 3.45e-5 9.35e-ll ActRIIB(A24N 20-134)-hFc ActRIIB(L79D 20-134)-hFc 3.85e-5 8.3 e-4 2.15e-9 ActRIIB(L79E 20-134)-hFc 3.74e-5 9e-4 2.41e-9 ActRIIB(R64K 20-134)-hFc 2.25e-5 4.71e-5 2,le-10 ActRIIB(R64K 20-129)-hFc 9.74e-4 2.09e-4 2.15e-9 ActRIIB(PI29S, P130R 20-134)-hFc l.08e-5 l,8e-4 l,67e-9 ActRIIB(K74A 20-134)-hFc 2.8e-5 2.03e-5 7.18e-ll Activin A Protein Kon (1/Mc) Koff (1/c) KD(M) ActRIIB(20-134)-hFc 5.94e6 l,59e-4 2.68e-ll ActRIIB(A24N 20-134)-hFc 3.34e6 3.46e-4 l,04e-10 ActRIIB(L79D 20-134)-hFc Loose binding ActRIIB(L79E 20-134)-hFc Loose binding ActRIIB(R64K 20-134)-hFc 6.82e6 3.25e-4 4.76e-11 ActRIIB(R64K 20-129)-hFc 7.46e6 6.28e-4 8.41e-11 ActRIIB(P129S, P130R20-134)-hFc 5.02e6 4.17e-4 8.31e-11

These data, obtained in a cell-free study, confirmed the results of a cell study demonstrating that the A24N variant retains ligand-binding activity similar to that of the ActRIIB(20-134)-hFc molecule and that the L79D or L79E molecule retains binding to myostatin and GDF11, however shows a significant decrease in binding (undetectable binding) to activin A.

Other variants were created and studied as described in WO2006/012627 (incorporated by reference in its entirety); see, for example, p. 59-60 using ligands immobilized on the device and passing the receptor along the surface of the immobilized ligands. It is noteworthy that K74Y, K74F, K74I (and presumably other hydrophobic substitutions at position K74, such as K74L) and D80I result in a decreased ratio of activin A (ActA) binding to GDF11 binding compared to the wild-type K74 molecule. A table with data related to these options is given below.

Binding of soluble ActRIIB-Fc variants to GDF11 and activin A (Biacore™ biosensor study). ________________________________________

ActRIIB ActA GDF11 WT (64A) KD=l,8e-7M (+) KD= 2.6e-7M (+) WT (64R) nd KD= 8.6e-8M (+++) +15tail KD ~2.6 e-8M (+++) KD 1.9е-8М (++++) E37A * * R40A - -

- 74 043914

D54A - * K55A ++ * R56A * * K74A KD=4,.35e-9M KD=5.3e-9M +++++ +++++ K74Y * - K74F * - K74I * - W78A * * L79A + * D80K * * D80R * * D80A * * D80F * * D80G * * D80M * * D80N * * D80I * -- F82A ++ -

*No binding observed

-- binding <1/5 wild type, - binding ~1/2 wild type, + wild type, ++ increased binding < 2x, +++ increased binding ~5x, ++++ increased binding ~10x, +++ ++ binding gain ~40x.

Example 6: Construction of a GDF Trap with a Truncated ActRIIB Extracellular Domain.

The GDF trap, designated ActRIIB(L79D 20-134)-hFc, was created by fusing the extracellular domain of ActRIIB (residues 20-134 in SEQ ID NO:1) containing a leucine to asparagine substitution (at residue 79 in SEQ ID NO:1) , the N-terminus with the TPA leader sequence and the C-terminus with the human Fc domain mediated through a minimal linker (three glycine residues) (Fig. 3). The nucleotide sequence corresponding to the specified fusion protein is shown in FIG. 4.

A GDF trap with a truncated extracellular domain of ActRIIB, designated ActRIIB(L79D 25-131)hFc, was created by fusing a truncated extracellular domain (residues 25-131 in SEQ ID NO:1) containing a leucine to aspartate substitution (at residue 79 in SEQ ID NO:1), the N-terminus with the TPA leader sequence and the C-terminus with the human Fc domain mediated through a minimal linker (three glycine residues) (Fig. 5, SEQ ID NO: 50). One nucleotide sequence encoding the specified fusion protein is shown in FIG. 6 SEQ ID NO: 51), and an alternative nucleotide sequence encoding the same fusion protein is shown in FIG. 9 (SEQ ID NO: 55).

Example 7: Selective ligand binding by a GDF trap to the doubly truncated extracellular domain of ActRIIB.

The affinities of GDF decoys and other ActRIIB-hFc proteins for several ligands were assessed in vitro using the Biacore™ biosensor. The results are presented in the table below. The values of the equilibrium constant Kd were calculated, since the very rapid association and dissociation of the complex did not make it possible to accurately determine k _on and koff.

Ligand selectivity of ActRIIB-hFc variants

Fused design Activin A (Kd e-11) Activin B (Kde-11) GDF11 (Kd e-11) ActRIIB(L79 20-134)-hFc 1.6 1.2 3.6 ActRIIB(L79D 20-134)-hFc 1350.0 78.8 12.3 ActRIIB(L79 25-13 l)-hFc 1.8 1.2 3.1 ActRIIB(L79D 25-13 l)-hFc 2290.0 62.1 7.4

The GDF trap with the truncated extracellular domain ActRIIB(L79D 25-131)-hFc had equal or superior ligand selectivity compared to that exhibited by the longer ActRIIB(L79D 20-134)-hFc variant, with a marked reduction in activin A binding, partial loss of activin B binding and almost complete preservation of GDF11 binding compared to ActRIIB-hFc, which does not have the L79D substitution. It should be noted that truncation itself (without the L79D substitution) did not significantly affect the selectivity of the listed ligands [when comparing ActRIIB(L79 25131)-hFc and ActRIIB(L79 20-134)-hFc]. ActRIIB(L79D 25-131)-hFc also retained strong to intermediate binding to GDF8, a ligand of the Smad 2/3 signaling pathway, and ligands BMP6 and BMP10 of the Smad 1/5/8 signaling pathway.

Example 8. GDF trap derived from ActRIIB 5.

Others have described an alternative soluble form of ActRIIB (designated ActRIIB5) in which exon 4, including the transmembrane domain of ActRIIB, has been replaced with a different C-terminal sequence (see, for example, WO 2007/053775).

- 75 043914

Below is the sequence of native ActRIIB5 without its leader sequence GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVK

KGCWLDDFNCYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEGPWAST TIPSGGPEATAAAGDQGSGALWLCLEGPAHE (SEQ Ш NO 33)

A leucine to aspartate substitution or other negative amino acid substitutions can be made at the native position 79 (underlined) as described to construct an ActRIIB5(L79D) variant having the following sequence

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVK

KGCWDDDFNCYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEGPWAST TIPSGGPEATAAAGDQGSGALWLCLEGPAHE (SEQ Ш NO 34)

This variant can be linked to a human Fc (double underscore) via a TGGG linker (single underscore) to produce a human ActRIIB 5(L79D)hFc fusion protein having the following sequence

GRGEAETRECIYYNANWELERTNQSGLERCEGEQDKRLHCYASWRNSSGTIELVK KGCWDDDFNCYDRQECVATEENPQVYFCCCEGNFCNERFTHLPEAGGPEGPWAST TTPSGGPEATAAAGDOGSGAEWECEEGPAHETGGGTHTCPPCPAPEELGGPSVFL FPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPREEOYN STYRVVSVLTVLHODWLNGKEYKCKVSNKALPAPIEKTISKAKGOPREPOVYTLP PSREEMTKNOVSLTCLVKGFYPSDIAVEWESNGOPENNYKTTPPVLDSDGSFFLY SKLTVDKSRWOOGNVFSCSVMHEALHNHYTOKSLSLSPGK (SEQ ID NO 35)

This construct can be expressed in CHO cells.

Example 9 Effect of GDF Trap on Animal Model JAK2V617F.

To understand the effect of ActRIIB(L79D 25-131)-Fc on myelofibrosis, a transgenic mouse with the JAK2V617F mutation [strain A described by Xing et al. (2008) Blood 111: 5109-5117].

To understand the onset and progression of myelofibrosis, complete blood counts and the degree of fibrosis in JAK2V617F mice of different ages were compared with data obtained from control animals (wild-type mice of comparable age). Red blood cell (RBC) and platelet counts in JAK2V617F mice of all ages were increased compared with wild type, with a trend toward increase in mutant animals between 2 and 5 months followed by a decrease between 8 and 12 months. Bone marrow fibrosis was detected in JAK2V617F mice starting at approximately 5 months of age and worsened with age. JAK2V617F mice also exhibited splenomegaly from approximately 3–4 months of age, which also worsened with age.

For the GDF trap study, treatment began at 12 months of age, which corresponds to late stage myelofibrosis. Mice were divided into two groups: i) treatment of JAK2V617F mice with ActRIIB(L79D 25-131)-Fc at 10 mg/kg twice weekly and ii) treatment of JAK2V617F mice with vehicle (TBS) twice weekly (i.e. control animals). After 10 weeks, animals treated with ActRIIB(L79D 25-131)-Fc showed a decrease in spleen size (-12.5%) compared to control animals. This observation was consistent with data from histopathological studies that revealed a decrease in extramedullary hematopoiesis in the spleens of mice treated with ActRIIB(L79D 25-131)-Fc compared with control animals. Histopathological examination also demonstrated decreased bone marrow fibrosis in ActRIIB(L79D 25-131)-Fc-treated mice compared to control animals.

Accordingly, GDF trap treatment is effective in improving various complications of myelofibrosis in this JAK2V617F model, particularly reducing splenomegaly, extramedullary hematopoiesis, and fibrosis. Thus, these data indicate that ActRIIB antagonists can be used to treat myelofibrosis. For example, ActRIIB antagonists are particularly effective in treating various complications of myelofibrosis, including, for example, reducing splenomegaly, reducing extramedullary hematopoiesis, increasing red blood cell counts, and/or reducing fibrosis (eg, bone marrow fibrosis).

Example 10: Effect of GDF Trap on Animals Treated with Ruxolitinib.

Ruxolitinib is a Janus kinase inhibitor approved for the treatment of myelofibrosis in patients at intermediate or high risk. In particular, ruxolitinib has a significant effect in reducing spleen size and improving symptoms associated with splenomegaly in patients suffering from myelofibrosis. However, various blood side effects, such as anemia, have been observed in patients treated with ruxolitinib. Nine-month-old C57BL/6 mice were used to understand the effects of ActRIIB(L79D 25-131)-Fc treatment on various hematological parameters in ruxolitinib-treated mice.

In this study, treatment began at 6–7 months of age. Mice were divided into four groups: i) mice treated with ActRIIB(L79D 25-131)-Fc at 10 mg/kg twice a week; ii) treatment with ruxolitinib at 60 mg/kg twice daily; iii) treatment with ActRIIB(L79D 25-131)-Fc at 10 mg/kg twice weekly and ruxolitinib at 60 mg/kg twice daily; and iv) vehicle (TBS) twice a week (ie control animals). After four weeks of treatment, mice treated with ActRIIB(L79D 25-131)-Fc had an increase in the number of red blood cells (~15%) and hemoglobin (~13%) compared to control mice (half

-

Claims (56)

treated with TBS), which indicated that ActRIIB(L79D 25-131)-Fc increases erythropoietic activity in C57BL/6 mice. In contrast, treatment with ruxolitinib resulted in a decrease in red blood cell count (~4%) and hemoglobin (~4%) compared to control animals. Mice co-treated with ActRIIB(L79D 25-131)-Fc and ruxolitinib had an increase in red blood cell count (~8%) and hemoglobin (~5%) compared to control animals. These data suggest that ActRIIB(L79D 25-131)-Fc can reverse ruxolitinib-induced anemia in normal healthy animals. Thus, these data suggest that ActRIIB antagonists may be used to alleviate Janus kinase inhibitor-induced anemia in various patient populations, including, for example, patients with myelofibrosis who have been or are receiving treatment with one or more Janus kinase inhibitors . Accordingly, ActRIIB antagonists may be used in conjunction with Janus kinase inhibitor therapy to treat various patient populations, including, for example, patients with myelofibrosis who have received or are receiving treatment with one or more Janus kinase inhibitors, particularly those with anemia is present. Incorporation by reference All publications and patents cited herein are incorporated by reference in their entirety as if each individual publication or patent were expressly stated to be incorporated by reference. The above description of specific embodiments of the objects of the invention is illustrative, but does not limit the scope of the invention. Numerous variations will be apparent to those skilled in the art having read this specification and the claims set forth below. The scope of the invention is determined by the claims, including equivalents in full, in the light of the description, taking into account such variations. CLAIM 1. A method of increasing red blood cell and/or hemoglobin levels in a patient being treated with a Janus kinase inhibitor, comprising administering to the patient an effective amount of an ActRIIB polypeptide, wherein the ActRIIB polypeptide contains an amino acid sequence that is at least 85% identical to a sequence beginning with a residue corresponding to any of amino acids 20, 21, 22, 23, 24, 25, 26, 27, 28 or 29 of the sequence SEQ ID NO: 1 and ending with a residue corresponding to any of amino acids 109, 110, 111, 112, 113, 114, 115, 116 , 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133 or 134 sequences SEQ ID NO: 1; and wherein the ActRIIB polypeptide contains a negatively charged amino acid at the amino acid position corresponding to position 79 in SEQ ID NO: 1. 2. A method of treating, preventing or reducing the rate of progression and/or severity of myelofibrosis or one or more than one complication of myelofibrosis in a patient, comprising administering to said patient in need of: a) a Janus kinase inhibitor and b) an ActRIIB polypeptide, wherein the Janus kinase inhibitor and the ActRIIB polypeptide is administered in an effective amount and the ActRIIB polypeptide contains an amino acid sequence that is at least 85% identical to the sequence beginning with a residue corresponding to any of amino acids 20, 21, 22, 23, 24, 25, 26, 27, 28 or 29 sequence SEQ ID NO: 1 and ending with a residue corresponding to any of amino acids 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133 or 134 sequences SEQ ID NO: 1; and wherein the ActRIIB polypeptide contains a negatively charged amino acid at the amino acid position corresponding to position 79 in SEQ ID NO: 1. 3. The method according to claim 2, where one or more than one complication of myelofibrosis is selected from the group consisting of ineffective hematopoiesis, anemia, inflammation, fibrosis, pancytopenia, thrombocytopenia, extramedullary hematopoiesis, hepatomegaly, splenomegaly, poikilocytosis, fatigue, weight loss, night sweats, fever, itching, bone pain, feeling of early satiety, pain or discomfort in the abdomen, arthralgia, myalgia, paresthesia, cachexia, splenic infarction, osteosclerosis, osteomyelofibrosis and bleeding. 4. The method of claim 3, wherein the one or more complication of myelofibrosis is selected from the group consisting of bone marrow fibrosis, splenic fibrosis, liver fibrosis, splenic extramedullary hematopoiesis, hepatic extramedullary hematopoiesis, pulmonary extramedullary hematopoiesis, and lymphatic extramedullary hematopoiesis. 5. A method according to any one of claims 1 to 4, wherein the method reduces one or more of the following: bone marrow fibrosis, splenic fibrosis, liver fibrosis, pulmonary fibrosis and lymph node fibrosis. 6. Method according to any one of claims 1-5, where the method reduces splenomegaly. 7. Method according to any one of claims 1-6, where the method reduces hepatomegaly. 8. A method according to any one of claims 1 to 7, wherein the method reduces inflammation in an organ/tissue selected from the group consisting of spleen, liver, lung, lymph node and bone marrow. 9. The method according to any one of claims 1 to 8, where the method reduces the size and/or weight of an organ/tissue selected from the group consisting of spleen, liver, lung and lymph node. - 77 043914 10. Method according to any one of claims 1-7, where the method reduces extramedullary hematopoiesis. 11. The method of claim 10, wherein the method reduces extramedullary hematopoiesis in an organ or tissue selected from the group consisting of the spleen (splenic extramedullary hematopoiesis), liver (hepatic extramedullary hematopoiesis), lung (pulmonary extramedullary hematopoiesis) and lymph node (lymphatic extramedullary hematopoiesis). hematopoiesis). 12. The method according to any one of claims 1 to 11, wherein the method increases the levels of red blood cells in the patient. 13. The method according to any one of claims 1 to 12, wherein the method increases hemoglobin levels in the patient. 14. The method according to any one of claims 1 to 13, where the patient has anemia. 15. The method according to claim 14, where the method treats anemia. 16. The method according to any one of claims 1 to 15, wherein prior to treatment with the ActRIIB polypeptide, the patient has undergone one or more blood cell transfusions. 17. The method according to any one of claims 1 to 16, wherein the patient is dependent on blood cell transfusion. 18. The method according to claim 17, where the method reduces the frequency of blood cell transfusions. 19. The method of claim 18, wherein the method reduces blood cell transfusion by more than about 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% over 4 to 8 weeks compared to an equal period of time before treatment. ActRIIB polypeptide. 20. The method of claim 19, wherein the method reduces blood cell transfusion by more than about 50% over a period of 4 to 8 weeks compared to an equal period of time before treatment with the ActRIIB polypeptide. 21. Method according to any one of claims 1 to 20, where the method reduces iron oversaturation. 22. The method of claim 21, wherein the method reduces the iron content of one or more than one organ/tissue selected from the group consisting of liver, spleen and heart. 23. The method according to any one of claims 1 to 22, where the patient has primary myelofibrosis. 24. The method according to any one of claims 1 to 22, where the patient has myelofibrosis, which was preceded by polycythemia vera. 25. The method according to any one of claims 1 to 22, where the patient has myelofibrosis, which was preceded by essential thrombocythemia. 26. The method according to any one of claims 1 to 25, where the patient has myelofibrosis at low risk, intermediate risk-1, intermediate risk-2 or high risk according to the International Prognosis Scoring Score (IPSS). 27. The method according to any one of claims 1 to 25, where the patient has myelofibrosis at low risk, intermediate risk-1, intermediate risk-2 or high risk according to dynamic IPSS (DIPSS). 28. The method according to any one of claims 1 to 25, where the patient has myelofibrosis at low risk, intermediate risk-1, intermediate risk-2 or high risk according to DIPSS-plus. 29. The method according to any one of claims 26-28, where the method prevents or delays the progression of risk in myelofibrosis from low risk to intermediate risk-1, from intermediate risk-1 to intermediate risk-2, or from intermediate risk-2 to high risk. 30. The method according to any one of claims 26-28, where the method promotes or accelerates a reduction in risk for myelofibrosis from high risk to intermediate risk-2, from intermediate risk-2 to intermediate risk-1, or from intermediate risk-1 to low risk. 31. The method according to any one of claims 2-4, where myelofibrosis is associated with one or more than one mutation in one or more than one gene selected from the group consisting of null zygosity for the haplotype JAK2 46/1, JAK2V617F, IDH1, IDH2, EZH2, SRSF2, ASXL1, JAK1, JAK2, JAK3, TYK2, MPL, CALR, CALR+ASXL1-, CALR-ASKL1+, CALR+ASKL1+, CALR-ASKL1-, TET2, TNPO and LNK. 32. The method of claim 31, wherein myelofibrosis is associated with one or more mutations in JAK2. 33. The method of claim 32, wherein the JAK2 mutation is JAK2V617F. 34. The method of any one of claims 2 to 4, wherein myelofibrosis is associated with an increase in one or more serum markers selected from the group consisting of increased serum levels of IL-8, increased serum levels of IL-2R and increased levels of free light chains. 35. The method according to any one of claims 2 to 34, wherein the patient has received treatment with a Janus kinase inhibitor. 36. The method according to any one of claims 1 to 35, where the patient has intolerance to a Janus kinase inhibitor. 37. The method according to any one of claims 1 to 36, wherein the patient has an unsatisfactory response to a Janus kinase inhibitor. 38. The method according to any one of claims 2 or 35-37, wherein the ActRIIB polypeptide is administered before treatment with a Janus kinase inhibitor. 39. The method according to any one of claims 2 or 35-37, wherein the ActRIIB polypeptide is administered after treatment with a Janus kinase inhibitor. 40. The method according to any one of claims 2 or 35-37, wherein the ActRIIB polypeptide is administered simultaneously with a Janus kinase inhibitor. 41. The method according to any one of claims 1-3 or 35-40, wherein the Janus kinase inhibitor inhibits one or more Janus kinases selected from the group consisting of JAK1, JAK2 and JAK3. 42. The method of claim 41, wherein the Janus kinase inhibitor inhibits signaling of one or more of JAK1, JAK2 and JAK3 in a cell assay. - 78 043914 43. The method according to any one of claims 1 to 42, where the Janus kinase inhibitor is selected from the group consisting of ruxolitinib, fedratinib (SAR302503), monoelotinib (CYT387), pacritinib, lestaurtinib, AZD1480, BMS-911543, NS-018, LY2784544 , SEP-701, XL019 and AT-9283. 44. The method of claim 43, wherein the Janus kinase inhibitor is ruxolitinib. 45. The method according to any one of claims 1 to 44, wherein the patient is additionally administered one or more than one maintenance therapy or one or more active agent selected from the group consisting of blood transfusion (whole blood or red blood cell transfusion), iron chelators (eg, deferoxamine, deferiprone and deferasirox), corticosteroids, prednisone, ESA (erythropoiesis-stimulating agents) (eg, erythropoietin, epoetin alfa, epoetin beta, darbepoetin alfa and methoxypolyethylene glycol-epoetin beta), androgens, danazol, thalidomide, lenalidomide, cytoreductive agent, hydroxyurea, busulfan, melphalm, cladribine, splenectomy, radiation therapy, aspirin, pomalidonmide, mTOR inhibitors (eg, rapamycin, sirolimus, deforolimus, everolimus, temsirolimus, NVP-BEZ235, BGT226, SF1126, PK1-587, INK128, AZD8055 and AZD2014) and histone deacetylase inhibitors (eg, givinostat, panobinostat and pracinostat). 46. The method according to claim 45, where the patient is additionally administered hydroxyurea or has previously been treated with hydroxyurea. 47. The method according to any one of claims 1 to 46, wherein the patient is intolerant to hydroxyurea or has an unsatisfactory response to hydroxyurea. 48. The method according to any one of claims 1 to 47, wherein the ActRIIB polypeptide is selected from the group consisting of: a) a polypeptide containing an amino acid sequence that is at least 85% identical to amino acids 29-109 of SEQ ID NO: 1; b) a polypeptide containing an amino acid sequence that is at least 85% identical to amino acids 25-131 of SEQ ID NO: 1; c) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 2; d) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 3; e) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 4; e) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 5; g) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 6; h) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 30; i) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 54. 49. The method of claim 48, wherein the ActRIIB polypeptide contains D at the amino acid position corresponding to position 79 in SEQ ID NO: 1. 50. The method of claim 48, wherein the ActRIIB polypeptide contains E at the amino acid position corresponding to position 79 in SEQ ID NO: 1. 51. The method according to any one of claims 48-50, wherein the ActRIIB polypeptide is a fusion protein containing an immunoglobulin Fc domain. 52. The method according to claim 51, wherein the Fc domain of the immunoglobulin is derived from the Fc domain of IgG1. 53. The method of claim 52, wherein the Fc domain of the immunoglobulin contains an amino acid sequence that is at least 85% identical to any of SEQ ID NOs: 9-13. 54. The method according to any one of claims 48 to 53, wherein the fusion protein further comprises a linker domain located between the ActRIIB domain and the immunoglobulin Fc domain. 55. The method of claim 54, wherein the linker domain is an amino acid sequence corresponding to any of SEQ ID NO: 14-20. 56. The method of claim 51, wherein the ActRIIB polypeptide is an ActRIIB-Fc fusion protein comprising a polypeptide selected from: a) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 24; b) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 25; c) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 28; d) a polypeptide containing an amino acid sequence that is at least 85 identical to the amino acid sequence of SEQ ID NO: 29; e) a polypeptide containing an amino acid sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 31; -