US20110144198A1

US20110144198A1 - Breast cancer prognostics

Info

Publication number: US20110144198A1
Application number: US12/992,615
Authority: US
Inventors: Mathias Uhlén; Karin Jirström; Fredrik Pontén
Original assignee: Atlas Antibodies AB
Current assignee: Atlas Antibodies AB
Priority date: 2008-05-16
Filing date: 2008-05-16
Publication date: 2011-06-16
Also published as: CN102089442B; AU2009246994B2; CA2722411A1; WO2009139681A8; WO2009138130A1; JP5798916B2; US20110142828A1; US8945832B2; JP2011525105A; AU2009246994A1; WO2009139681A1; CN102089442A

Abstract

The present invention provides new methods, uses and means for breast cancer prognostics. The provided method for establishing a prognosis for a mammalian subject having a breast cancer, comprises the steps of: obtaining a hormone receptor status of the subject; obtaining an HMGCR protein value of the subject; and correlating the hormone receptor status and the HMGCR protein of the subject to a prognosis for the subject.

Description

FIELD OF THE INVENTION

The present invention generally relates to breast cancer prognostics, and in particular to molecular markers having prognostic value and uses thereof.

BACKGROUND

Cancer

Cancer is one of the most common causes of disease and death in the western world. In general, incidence rates increase with age for most forms of cancer. As human populations continue to live longer, due to an increase of the general health status, cancer may affect an increasing number of individuals. The cause of most common cancer types is still largely unknown, although there is an increasing body of knowledge providing a link between environmental factors (dietary, tobacco smoke, UV radiation etc) as well as genetic factors (germ line mutations in “cancer genes” such as p53, APC, BRCA1, XP etc) and the risk for development of cancer.
No definition of cancer is entirely satisfactory from a cell biological point of view, despite the fact that cancer is essentially a cellular disease and defined as a transformed cell population with net cell growth and anti-social behavior. Malignant transformation represents the transition to a malignant phenotype based on irreversible genetic alterations. Although this has not been formally proven, malignant transformation is believed to take place in one cell, from which a subsequently developed tumor originates (the “clonality of cancer” dogma). Carcinogenesis is the process by which cancer is generated and is generally accepted to include multiple events that ultimately lead to growth of a malignant tumor. This multi-step process includes several rate-limiting steps, such as addition of mutations and possibly also epigenetic events, leading to formation of cancer following stages of precancerous proliferation. The stepwise changes involve accumulation of errors (mutations) in vital regulatory pathways that determine cell division, asocial behavior and cell death. Each of these changes may provide a selective Darwinian growth advantage compared to surrounding cells, resulting in a net growth of the tumor cell population. A malignant tumor does not only necessarily consist of the transformed tumor cells themselves but also surrounding normal cells which act as a supportive stroma. This recruited cancer stroma consists of connective tissue, blood vessels and various other normal cells, e.g., inflammatory cells, which act in concert to supply the transformed tumor cells with signals necessary for continued tumor growth.
The most common forms of cancer arise in somatic cells and are predominantly of epithelial origin, e.g., prostate, breast, colon, urothelial and skin, followed by cancers originating from the hematopoetic lineage, e.g., leukemia and lymphoma, neuroectoderm, e.g., malignant gliomas, and soft tissue tumors, e.g., sarcomas.

Cancer Diagnostics and Prognostics

Microscopic evaluation of a tissue section taken from a tumor remains the golden standard for determining a diagnosis of cancer. For microscopic diagnosis, biopsy material from suspected tumors is collected and examined under the microscope. To obtain a firm diagnosis, the tumor tissue is fixated in formalin, histo-processed and paraffin embedded. From the resulting paraffin block, tissue sections can be produced and stained using both histochemical, i.e., hematoxylin-eosin staining, and immunohistochemical methods. The surgical specimen is then evaluated with pathology techniques, including gross and microscopic analysis. This analysis forms the basis for assigning a specific diagnosis, i.e., classifying the tumor type and grading the degree of malignancy, of a tumor.
Malignant tumors can be categorized into several stages according to classification schemes specific for each cancer type. The most common classification system for solid tumors is the tumor-node-metastasis (TNM) staging system. The T stage describes the local extent of the primary tumor, i.e., how far the tumor has invaded and imposed growth into surrounding normal tissues, whereas the N stage and M stage describe how the tumor has developed into metastasis, with the N stage describing spread of tumor to lymph nodes and the M stage describing growth of tumor in other distant organs. Early stages include: T0-1, N0, M0, representing localized tumors with negative lymph nodes. More advanced stages include: T2-4, N0, M0, localized tumors with more widespread growth and T1-4, N1-3, M0, tumors that have metastasized to lymph nodes and T1-4, N1-3, M1, tumors with a metastasis detected in a distant organ. Staging of tumors is often based on several forms of examination, including surgical, radiological and histopathological analyses. In addition to staging, there is also a classification system to grade the level of malignancy for most tumor types. The grading systems rely on morphological assessment of a tumor tissue sample and are based on the microscopic features found in a given tumor. These grading systems may be based on the degree of differentiation, proliferation and atypical appearance of the tumor cells. Examples of generally employed grading systems include Gleason grading for prostatic carcinomas and the Nottingham Histological Grade (NHG) grading for breast carcinomas.
Accurate staging and grading is crucial for a correct diagnosis and provides an instrument to predict a prognosis. The diagnostic and prognostic information for a specific tumor subsequently determines an adequate therapeutic strategy for a given cancer patient. The most commonly used method, in addition to histochemical staining of tissue sections, to obtain more information regarding a tumor is immunohistochemical staining. IHC allows for the detection of protein expression patterns in tissues and cells using specific antibodies. The use of IHC in clinical diagnostics allows for the detection of immunoreactivity in different cell populations, in addition to the information regarding tissue architecture and cellular morphology that is assessed from the histochemically stained tumor tissue section. IHC can be involved in supporting the accurate diagnosis, including staging and grading, of a primary tumor as well as in the diagnostics of metastases of unknown origin. The most commonly used antibodies in clinical practice today include antibodies against cell type “specific” proteins, e.g., PSA (prostate), MelanA (melanocytes) and Thyroglobulin (thyroid gland), and antibodies recognizing intermediate filaments (epithelial, mesenchymal, glial), cluster of differentiation (CD) antigens (hematopoetic, sub-classification of lympoid cells) and markers of malignant potential, e.g., Ki67 (proliferation), p53 (commonly mutated tumor suppressor gene) and HER-2 (growth factor receptor).
Aside from IHC, the use of in situ hybridization for detecting gene amplification and gene sequencing for mutation analysis are evolving technologies within cancer diagnostics. In addition, global analysis of transcripts, proteins or metabolites all add relevant information. However, most of these analyses still represent basic research and have yet to be evaluated and standardized for the use in clinical medicine.
In order for physicians to give a cancer patient the right type of treatment as early as possible, the provision of new molecular markers that allow for more accurate stratification of cancer patients into different risk categories is of interest. In summary, there exists a need for new means to advance prognostics and staging of cancer.

Breast Cancer

Breast cancer is the second most common form of cancer worldwide and by far the most frequent cancer of women. Data from the GLOBOCAM 2002 database presented by Parkin et al. reveal 1.15 million new cases in 2002 and 0.41 million deaths during the same period (Parkin D M et al. (2005) CA Cancer J Clin 55, 74-108). If detected at an early stage, the prognosis is relatively good for a patient living in a developed country, with a general five-year survival rate of 73%, compared to 57% in a developing country. The incidence is slowly increasing and about one in every nine women in the developed world is believed to get breast cancer in her lifetime. Although lifestyle changes related to female steroid hormones, including exposure to exogenous hormones, affect the risk of developing breast cancer, these factors only make up for a small fraction of the etiology, and the benefit of preventive manipulation is believed to be low. The decreased mortality is due to earlier detection by mammography screening and the use of modern adjuvant systemic treatment.

Treatment of Breast Cancer

Since its introduction in the late seventies, breast-conserving therapy, combining breast conserving surgery and postoperative radiotherapy, has become the primary treatment of choice in women where radical removal of the tumor can be combined with a good cosmetic result. Mastectomy is still preferable in some patients, i.e., women with small breasts, large tumors (>4 cm) or multifocal/multicentric disease.
Axillary dissection is primarily performed for diagnostic purposes and removal of at least 10 lymph nodes gives a good staging guidance with 97-98% sensitivity (Axelsson C K et al. (1992) Eur J Cancer 28A:1415-8; Recht A and Houlihan M J (1995) Cancer 6(9):1491-1512). However, the next step towards minimal surgery in the treatment of primary cancer has been the introduction of the sentinel node biopsy technique with mapping of axillary lymph nodes instead of axillary lymph node clearance, which is associated with a high complication rate. This technique was introduced as a consequence of the knowledge that most of the lymphatic drainage to the axilla from the breast initially passes through one (or a few) lymph node(s)—the sentinel node(s)—supporting that analysis of this lymph node may be a sufficient indicator of axillary node status (Veronesi U et al. (2003) New Engl J Med 349(6): 546-53.)
The concept of breast cancer as a systemic disease, i.e., the presence of disseminating micro-metastases at the time of diagnosis that may explain treatment failure after locoregional therapy, paved the way for adjuvant randomized trials in the 1970s, including endocrine therapy and chemotherapy. Adjuvant polychemotherapy is standard treatment for hormone-receptor negative patients with high risk of recurrence, irrespective of nodal status. A beneficial effect on both overall- and relapse-free survival has been demonstrated, especially in premenopausal patients (EBCTCG (1998) Lancet 352(9132): 930-42). For patients with hormone-responsive disease, i.e., estrogen receptor (ER) and/or progesterone receptor (PR) positive disease, adjuvant polychemotherapy has been delivered in combination with endocrine therapy as sequential chemo-endocrine therapy. Also, adjuvant chemotherapy generally induces amenorrhea, causing a secondary endocrine effect in addition to the cytotoxic (Pagani O et al. (1998) Eur J Cancer 34(5):632-40).
Endocrine therapy is recommended for patients with hormone receptor positive tumors irrespective of age, stage and menopausal status.
In hormone-responsive premenopausal patients, ovarian ablation by surgery or irradiation, or ovarian suppression by LHRH agonists are efficient adjuvant treatment modalities (Emens L A and Davidson N A (2003) Clin Ca Res (1 Pt 2): 468S-94S). In postmenopausal patients, ovarian ablation has no place, since the primary source of estrogen is not from ovarian synthesis but from the conversion of androstenedione to estrone and estradiol in peripheral tissues including the breast.
Tamoxifen is a selective estrogen receptor modulator (SERM) with an agonistic effect on the ER, making it a suitable treatment for advanced breast cancer in both pre- and postmenopausal women. Five years of tamoxifen as adjuvant treatment after primary surgery clearly reduces the breast cancer mortality in patients with ER positive (ER+) tumors, irrespective of lymph node status (EBCTCG (1998) Lancet 351(9114):1451-67). While tamoxifen has a protective effect against cardiovascular disease, the risk of developing endometrial cancer is increased, due to an agonistic effect on the ER in the endometrium (EBCTCG (2005) Lancet 365(9472):1687-717)
Aromatase inhibitors (AIs) function by inhibiting aromatase, the enzyme converting androgens into estrogens. AIs are not suitable for treatment of premenopausal women, as it stimulates the ovaries to an increased androgen production through the hypothalamus and pituitary gland. AIs can be given as adjuvant treatment to postmenopausal women, either alone or following tamoxifen treatment and they have been shown to significantly reduce the mortality, possibly even more if given alone (Howell A et al. (1995) Lancet 345(8941):29-30; Ellis M J and Rigden C E (2006) Curr Med Res Opin 22(12):2479-87; Coates A S et al. (2007) J Clin Oncol 25(5):486-92). However, this therapy is relatively new and the long-term side effects are not yet fully known (Buzdar A et al. (2006) Lancet Oncol 7(8):633-43), but the most important are cardiovascular complications and osteoporosis.
Newly developed pure anti-estrogens such as fulvestrant, which completely blocks the ER, are currently only used in advanced breast cancer and not in the adjuvant setting (Rutqvist L E (2004) Best Pract Res Clin Endocrinol Metab 18(1): 81-95).
Adjuvant endocrine therapy has no place in hormone receptor negative breast cancer, although some studies indicate that some ER negative (ER−), i.e., ERα negative (ERα−), tumors respond to tamoxifen treatment (EBCTCG (1998) Lancet 351:1451-1467)
The HER2/neu gene is overexpressed in about 20% of all, and in up to 70% of lowly differentiated, breast cancers (Berger M S et al. (1988) Cancer Res 48(5):1238-43; Borg Å et al. (1990) Cancer Res 50(14): 4332-7). Patients with HER2 overexpressing tumors may benefit from treatment with the monoclonal antibody trastuzumab. Experimental data support a relationship between HER2 overexpression and resistance to endocrine treatment (Shou J et al. (2004) J Natl Cancer Inst 96(12):926-35) while clinical data are not consistent (Borg Å et al. (1994) Cancer Lett 81(2):137-44, De Placido S et al. (2003) Clin Ca Res 9(3):1039-46, Rydén L et al. (2005) J Clin Oncol 23(21):4695-704).

Breast Cancer Diagnostics

Morphologic criteria are still involved in the establishment of a breast cancer diagnosis, both in situ and invasive cancer. Among invasive breast carcinomas, invasive ductal carcinoma is the most common tumor type (˜80%) and lobular carcinoma is the second largest entity (˜10-15%). Tubular and medullary carcinomas are other distinct types with lower prevalence (WHO, Histological typing of breast tumors, in International histological classification of tumors no: 2, 1981, WHO, Geneva).

Breast Cancer Prognostics and Treatment Predictive Factors

A correct histological classification of the tumor type may be of prognostic relevance, since certain subtypes, such as medullary carcinomas, in general have a more favorable prognosis. Nevertheless, assessment of the histological grade using the Nottingham Histological Grade (NHG) system is still a prognostic tool (Elston C W and Ellis I O (1991) 19(5):403-10; Sundquist M et al. (1999) Breast Cancer Res Treat 53(1):1-8).
The majority of breast cancers are hormone receptor responsive, i.e., express the ER and/or PR. The action of estrogen is mediated by the two receptors ERα and ERβ. ERα and PR are routinely assessed in order to select patients for endocrine therapy, and tumors with >10% nuclear positivity are considered positive. ERα is today considered a predictor of tamoxifen response. However, there are studies that indicate that PR positivity may even be a more powerful predictor of tamoxifen response than ERα. A study of premenopausal patients randomly treated with tamoxifen revealed that a high, >75% nuclear fraction, PR level was significantly correlated with an increased recurrence-free and overall survival, irrespectively of ERα status. At a lower PR level, <75%, no positive effect was observed (Stendahl M et al. (2006) Clin Cancer Res 12:4614-18). Yu et al. recently studied the predictive value of PR for adjuvant endocrine therapy, and found that older patients (≧60 years) with ERα+/PR+ tumors had a significantly longer disease free survival when treated with tamoxifen than patients that received no adjuvant treatment. In younger patients (<60 years), no significant effect was observed (Yu K D et al. (2007) The Breast 16:307-315). Interestingly, a report from the ATAC trial (postmenopausal women treated with arimidex, tamoxifen or in combination), showed that the recurrence rate was halved for anastrozole-treated patients with ERα+/PR− tumors over the follow-up period of 6 years, compared to patients treated with tamoxifen (Dowsett M et al. (2005) J Clin Oncol 23(30):7512-7).
The role of ERβ in breast cancer is not yet fully clarified, although recent studies implicate that ERβ-expression may be associated with a better tamoxifen response (Borgquist S et al, (2008) J Clin Pathol 61(2):197-203), particularly in ERα− tumors (Gruvberger-Saal S K et al. (2007) Clin Cancer Res 13:1987-1994). However, determination of ERβ is today generally not considered clinically relevant.
A major problem to day is that 30-40% of the ERα positive (ERα+) patients do not respond to tamoxifen treatment (Riggins R B et al. (2007) Cancer Letters 1:1-24, Gruvberger-Saal S K et al. (2007) Clin Cancer Res 13:1987-1994), which results in unnecessary treatment. In addition, a fraction of the ERα− patients do respond to tamoxifen treatment, and the reason for that is currently not known. Gruvberger-Saal suggests that ERβ expression may be a positive predictor of tamoxifen response in ERα− patients (Gruvberger-Saal S K et al. (2007) Clin Cancer Res 13:1987-1994).
HER2 status is also assessed routinely, primarily by IHC and in cases with moderate expression, gene amplification status is determined by fluorescence in situ hybridization (FISH) analysis. Patients with a HER2 positive tumor may be treated with trastuzumab.
Breast cancer is a truly heterogeneous disease and despite the increasing understanding of its nature, the arsenal of available prognostic and treatment predictive markers is still not sufficient and some patients may therefore receive unnecessary treatment while others may get insufficient or even ineffective treatment. Additional molecular markers are needed in order to better define different subgroups of breast cancer and increase the options for tailored therapies.

Endpoint Analysis

Endpoint analysis is used to evaluate trials with adjuvant treatments for cancer as this gives information on how the patients respond to a certain therapy. Overall survival (OS) has been considered the standard primary endpoint. OS takes in to account time to death, irrespective of cause, e.g., if the death is due to cancer or not. Loss to follow-up is censored and regional recurrence, distant metastases, second primary breast cancers, and second other primary cancers are ignored.
To date, an increasing number of effective treatments available in many types of cancer have resulted in the need for surrogate endpoints to allow for a better evaluation of the effect of adjuvant treatments. Thus, the much longer follow-up required to demonstrate that adjuvant treatments improve OS is often complemented with other clinical endpoints that give an earlier indication on how successful the treatment is.
Endpoint analysis may also be useful for studies of a potential biomarker. In the present disclosure, the inventors show that the level of expression of a particular protein (the proposed biomarker) significantly correlates with prognoses. For these observations, primarily two surrogate endpoints where used, namely recurrence-free survival (RFS) and breast cancer-specific survival (BCSS). Analysis of RFS includes time to any event related to the same cancer, i.e., all cancer recurrences and deaths from the same cancer are events. Also distant, local and regional metastases as well as breast cancer specific death are considered. On the other hand, second primary same cancers and other primary cancers are ignored, as well as contralateral breast cancer. Deaths from other cancers, non-cancer-related deaths, treatment-related deaths, and loss to follow-up are censored observations. Breast cancer-specific survival (BCSS) includes time to death caused by breast cancer due to the original tumor. Both endpoints are relevant, since similarities or differences may reflect different tumor biological behaviors. Biomarkers associated with a locally aggressive behavior may for instance have greater impact on RFS than BCSS, while biomarkers associated with the development of distant metastases may be reflected in both RFS and BCSS.

HMGCR

The enzyme 3-hydroxy-3-methylglutharyl-coenzyme A reductase (HMGCR) is rate-limiting in the mevalonate pathway, i.e., the cholesterol synthesis (Goldstein J L and Brown M S (1990) Nature 343:425-30). However, the non-sterol products of the mevalonate pathway are needed for cell survival. HMGCR is required for the synthesis of isoprenoids, which, in turn, are needed for post-translational modification of proteins that regulate cellular proliferation and apoptosis (Shachaf C M et al. (2007) Blood 110: 2674-84). Animal studies have shown that estrogens influence cholesterol metabolism (Di Croce L et al. (1996) Biochem Biophys Res Commun 224:345-50; Ma P T, (1998) PNAS 83:792-6), which may be due to an estrogen-responsive region located in the promoter of the HMGCR gene (Di Croce L (1999) Mol Endocrinol 13:1225-36). Despite indications of a multifaceted role in cancer as well as estrogen-related activities, little is known about the tumor-specific expression of HMGCR in breast cancer.

DISCLOSURE OF THE INVENTION

Summarizing the state of the art, there is a great need for new methods and tools for breast cancer prognostics, i.e., the branch of medicine dealing with prognoses for subjects having breast cancer.
To meet this need and for other objects apparent to the skilled person from the present disclosure, the present invention provides, in its different aspects, new means, including methods and uses, for determining a prognosis for a mammalian subject having a breast cancer, for determining whether such a subject should undergo a breast cancer treatment and for treatment of such a subject. The present invention is defined by the appended claims.
Thus, according to a first aspect, there is provided a method for determining a prognosis for a mammalian subject having a breast cancer, comprising the steps of:

- i) obtaining a hormone receptor status of the subject;
- ii) obtaining an HMGCR protein value of the subject; and
- iii) correlating the hormone receptor status and the HMGCR protein value of the subject to a prognosis for the subject.

“Determining a prognosis” refers to establishing a specific prognosis or a prognosis interval. “Hormone receptor status” refers to whether the subject is positive or negative for a hormone receptor, i.e., the estrogen receptor and/or the progesterone receptor. The establishment of a hormone receptor status is further discussed below.
The correlating of step iii) refers to any way of associating survival data to the obtained hormone receptor status and HMGCR protein value so as to determine a prognosis for the subject.
In the correlating of step iii), for hormone receptor positive subjects, a high HMGCR protein value, e.g., an HMGCR protein value which is higher than a predetermined reference value, may preferably be indicative of a better prognosis than a low HMGCR protein value, e.g., an HMGCR protein value which is lower than, or equal to, the predetermined reference value. For hormone receptor negative subjects, the relationship may preferably be reversed, i.e., a high HMGCR protein value, e.g., an HMGCR protein value which is higher than a predetermined reference value, is indicative of a worse prognosis than a low HMGCR protein value, e.g., an HMGCR protein value which is lower than, or equal to, the predetermined reference value.
As an embodiment of the first aspect, there is provided a method for determining a prognosis for a mammalian subject having a hormone receptor positive or a hormone receptor negative breast cancer, comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value;
- d) and, if said sample value is higher than said reference value,
- d1) concluding that the prognosis for said subject is better than a first reference prognosis associated with the reference value if the subject has a hormone receptor positive breast cancer; or
- d2) concluding that the prognosis for said subject is worse than a second reference prognosis associated with the reference value if the subject has a hormone receptor negative breast cancer.

For example, the above method may further comprise the step:

- e) and, if said sample value is lower than, or equal to, said reference value,
- e1) concluding that the prognosis for said subject is worse than, or equal to, the first reference prognosis associated with the reference value if the subject has a hormone receptor positive breast cancer; or
- e2) concluding that the prognosis for said subject is better than, or equal to, the second reference prognosis associated with the reference value if the subject has a hormone receptor negative breast cancer.

In a variant of this embodiment, the reference value may be adapted to the hormone receptor status of the cancer. That is, if the cancer is hormone receptor positive, a first reference value is used for comparison with the sample value in step c), and if the subject is hormone receptor negative, a second reference value is used for comparison with the sample value in step c). Further, in such case, the first reference prognosis is associated with the first reference value, and the second reference prognosis is associated with the second reference value.
As a configuration of the first aspect, there is provided a method for determining whether a prognosis for a mammalian subject having a hormone receptor positive breast cancer is better than a reference prognosis, comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value associated with said reference prognosis; and
- d) if said sample value is higher than said reference value, concluding that the prognosis for said subject is better than said reference prognosis.

In an embodiment of the configuration above, the method may further comprise the step:

- e) and if the sample value is equal to, or lower than, the reference value, concluding that the prognosis for the subject is worse than, or equal to, the reference prognosis.

As a similar configuration, there is provided a method for determining whether a prognosis for a mammalian subject having a hormone receptor positive breast cancer is worse than, or equal to, a reference prognosis, comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value associated with a reference prognosis; and
- d) if said sample value is lower than, or equal to, said reference value, concluding that the prognosis for said subject is worse than, or equal to, said reference prognosis.

In an embodiment of the configuration above, the method may further comprise the step:
e) and if the sample value is higher than the reference value, concluding that the prognosis for the subject is better than the reference prognosis.
According to another similar configuration of the first aspect, there is provided a method for determining whether a prognosis for a mammalian subject having a hormone receptor positive breast cancer is better than a reference prognosis or worse than or equal to the reference prognosis, the method comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value associated with a reference prognosis; and
- d1) if said sample value is higher than said reference value, concluding that the prognosis for said subject is better than said reference prognosis, or
- d2) if said sample value is lower than, or equal to, said reference value, concluding that the prognosis for said subject is equal to or worse than said reference prognosis.

According to a complementary configuration of the first aspect, there is provided a method for determining whether a prognosis for a mammalian subject having a hormone receptor negative breast cancer is worse than a reference prognosis, comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value associated with said reference prognosis; and
- d) if said sample value is higher than said reference value, concluding that the prognosis for said subject is worse than said reference prognosis.

- e) and if the sample value is lower than, or equal to, the reference value, concluding that the prognosis for the subject is better than, or equal to, the reference prognosis.

As a similar configuration, there is provided a method for determining whether a prognosis for a mammalian subject having a hormone receptor negative breast cancer is better than, or equal to, a reference prognosis, comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value associated with a reference prognosis; and
- d) if said sample value is lower than, or equal to, said reference value, concluding that the prognosis for said subject is better than, or equal to, said reference prognosis.

In an embodiment of the configuration above, the method may further comprise the step:
e) and if the sample value is higher than the reference value, concluding that the prognosis for the subject is worse than the reference prognosis.
According to another complementary configuration of the first aspect, there is provided a method for determining whether a prognosis for a mammalian subject having a hormone receptor negative breast cancer i: worse than a reference prognosis or better than, or equal to, the reference prognosis, the method comprising the steps of:

- a) providing a sample earlier obtained from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value associated with a reference prognosis; and
- d1) if said sample value is higher than said reference value, concluding that the prognosis for said subject is worse than said reference prognosis, or
- d2) if said sample value is lower than, or equal to, said reference value, concluding that the prognosis for said subject is better than, or equal to, said reference prognosis.

Regarding step b) of the above methods, an increase in the amount of HMGCR protein typically results in an increase in the sample value, and not the other way around.
In the present disclosure, a subject having a hormone receptor positive breast cancer is sometimes referred to as “HR+ subject”, and a subject having a hormone receptor negative breast cancer is sometimes referred to as “HR− subject”.
This first aspect of the present invention is based on the previously unrecognized fact that the amount of HMGCR protein present in samples earlier obtained from a subject having a breast cancer may serve as a disease status indicator in the subject. More particularly, the present invention identifies for the first time, in subjects suffering from a breast cancer, a correlation between an amount of HMGCR protein on the one hand and a prognosis for survival on the other. Typically, high HMGCR protein values have been shown to correlate with a good prognosis in HR+ subjects, probably due to a less aggressive or low-risk form of the cancer and with a poor prognosis in HR− subjects, probably due to a more aggressive or high-risk form of the cancer. The present invention based on HMGCR protein expression as a breast cancer status indicator has a number of benefits. In general, early identification of the aggressiveness of a breast cancer is of vital importance as it helps a physician select an appropriate treatment strategy. For example, early identification of aggressive forms of breast cancer is of vital importance as it helps a physician select an appropriate treatment strategy. Also, by identifying less aggressive forms at an early stage, over-treatment may be avoided. As a further example, the HMGCR protein, as a marker for which a certain level of expression is correlated with disease progression information, has a great potential for example in a panel for prognostics.
In the present disclosure, different HMGCR protein values (sample values) corresponding to various prognoses are presented. Typically for HR+ subjects, a low sample value is associated with a poorer prognosis than a high sample value. In the above methods, the sample value is compared to a reference value, and if the sample value is higher than the reference value, it is concluded that the prognosis for the subject is better than a reference prognosis associated with the reference value.
Consequently, the methods of the above aspect may be adapted to a given reference value. In such case, assuming a HR+ subject and starting from a given sample value which under certain circumstances is considered to be relevant, a reference value which is equal to, or lower than, that sample value, may be selected. Subsequently, a reference prognosis being associated with that reference value may be established. Guided by the present disclosure, the person skilled in the art should understand how to establish a reference prognosis which corresponds to a given reference value. For example, the relationship between sample values and survival data in a group of breast cancer patients may be examined as in Examples, section 4, below, and the procedure described therein may be adapted to a given reference value, and a prognosis corresponding to the given reference value may then be selected as the reference prognosis.
Also, the methods of the above aspect may be adapted to a given reference prognosis. In such case, again assuming a HR+ subject and starting from a given reference prognosis which under certain circumstances is considered to be relevant, for example for selecting an appropriate treatment strategy, a corresponding reference value may be established. Guided by the present disclosure, the person skilled in the art should understand how to establish a reference value which corresponds to a given reference prognosis. For example, the relationship between sample values and survival data in a group of cancer patients may be examined as in Examples, section 4, below, but the procedure described therein is adapted to establish reference values corresponding to a given reference prognosis. For example, different reference values may be tested until one which correlates with the given reference prognosis is found. A skilled person may do such testing without undue burden, e.g., because the present disclosure presents suitable starting values which substantially reduces the required amount of testing.
With the knowledge of the present disclosure, the skilled person should understand how to adapt, e.g., in certain aspects reverse, the above reasoning to a HR− subject.
Accordingly, in embodiments of the methods of the above aspect, the reference prognosis may be based on a previously established prognosis, e.g., obtained by an examination of the same subject or population of subjects. Further, the reference prognosis may be adapted to a background risk in the general population, a statistical prognosis/risk or an assumption based on an examination of the subject. Such examination may comprise the subject's age, general condition, sex, race and/or medical status and history, such as cancer history or breast cancer status. For example, a physician may adapt the reference prognosis to the subject's breast cancer history, the stage of the tumor, the morphology of the tumor, the location of the tumor, the presence and spread of metastases and/or further cancer characteristics.
The identified correlation between HMGCR protein levels and breast cancer characteristics may also form the basis for applying various treatment regimes.
For example, as shown in the attached FIGS. 1-8, the prognoses for HR+ subjects having high HMGCR protein levels are generally better than those for HR+ subjects having low HMGCR protein levels. Provided the teachings of the present disclosure, a physician may consider the prognosis of an HR+ and HMGCR protein high subject as being so favorable that a less comprehensive adjuvant breast cancer therapy, or even no adjuvant breast cancer therapy, is necessary. The present disclosure may thus relieve subjects from over-treatment. Further, for HR+ subjects, a low HMGCR protein level may lead to the application of a certain adjuvant therapy.
As another example, as shown in FIGS. 9-10, among HR− subjects, the prognosis of an HMGCR protein high subject may be considerable worse than that of an HMGCR protein low subject. For example, such subjects may be in need of a broad adjuvant therapy having serious side effects which would not have been given to them if not diagnosed as HMGCR protein high.
Thus, according to a second aspect of the present disclosure, there is provided a method of treatment of a subject having a hormone receptor positive breast cancer, comprising:

- a) providing a sample from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value; and,
- d) if said sample value is equal to or lower than said reference value, treating said subject with an adjuvant breast cancer treatment.

Regarding step b) of the above method, an increase in the amount of HMGCR protein typically results in an increase in the sample value, and not the other way around.
As discussed above, by obtaining HMGCR protein information, such as a prognosis, regarding a HR+ subject according to the second aspect, a physician may refrain from treating the subject with an adjuvant breast cancer treatment. As an example, a reference prognosis or reference value may be adapted to suit an individual case for making such judgment. Various conditions to which the reference value and/or prognosis may be adapted are described above.
Consequently, in an embodiment of the second aspect, the method may comprise the additional step:
e) and if said sample value is higher than said reference value, refraining from treating said subject with the adjuvant breast cancer treatment regimen.
Also, in one embodiment, the reference value of step c) of the above method may be associated with a reference prognosis and said breast cancer treatment regimen of step d) may be adapted to a prognosis which is worse than or equal to the reference prognosis. In such an embodiment, the method may comprise the additional step: e) and if said sample value is higher than said reference value, treating said subject with a treatment regimen adapted to a prognosis which is better than the reference prognosis, which treatment regimen may be no treatment.
In conclusion, step e) of the method above (relating to ER+) is based on the inventors insight that a high HMGCR protein value may indicate that an adjuvant breast cancer treatment is unnecessary.
According to a complementary embodiment of the second aspect, there is provided a method of treatment of a subject having a hormone receptor negative breast cancer, comprising:

- a) providing a sample from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value; and,
- d) if said sample value is higher than said reference value, treating said subject with an adjuvant breast cancer treatment.

Regarding step b) of the above method, an increase in the amount of HMGCR protein typically results in an increase in the sample value, and not the other way around.
Similar to what is discussed above, by obtaining HMGCR protein information, such as a prognosis, regarding a HR− subject according to the second aspect, a physician may refrain from treating the subject with an adjuvant breast cancer treatment. As an example, a reference prognosis or reference value may be adapted to suit an individual case for making such judgment. Various conditions to which the reference value and/or prognosis may be adapted are described above.
Consequently, the complementary embodiment of the second aspect may comprise the additional step:
e) and if said sample value is lower than or equal to said reference value, refraining from treating said subject with the adjuvant breast cancer treatment regimen.
Also, in one embodiment, the reference value of step c) of the above method may be associated with a reference prognosis and said breast cancer treatment regimen of step d) may be adapted to a prognosis which is worse than the reference prognosis. In such an embodiment, the method may comprise the additional step: e) and if said sample value is higher than said reference value, treating said subject with a treatment regimen adapted to a prognosis which is better than or equal to the reference prognosis, which treatment regimen may be no treatment.
For example, the refraining of step e) of the embodiments of the second aspect may be a refraining from treatment lasting at least one week from the completion of steps a)-c), such as at least one month from the completion of steps a)-c), such as at least three months from the completion of steps a)-c), such as at least six months from the completion of steps a)-c), such as at least one year from the completion of steps a)-c), such as at least two years from the completion of steps a)-c).
Alternatively, the refraining of step e) may be a refraining from treatment until the next time the method is performed or until recurrence of a breast cancer tumor.
In conclusion, step e) of the methods above is based on the inventors' insight that an HMGCR protein value may indicate that an adjuvant breast cancer treatment is unnecessary.
In embodiments of the second aspect, the sample may be an earlier obtained sample.
Further, in embodiments of the second aspect, the adjuvant breast cancer treatment may be a chemotherapy, an endocrine treatment or a combination thereof, such as a sequential chemo-endocrine therapy. Also, the adjuvant breast cancer treatment may be radiation therapy which may be combined with chemotherapy and/or endocrine therapy.
In the context of the present disclosure, an “endocrine treatment” refers to a systemic treatment having an anti-estrogenic effect. Further, “anti-estrogenic effect” refers to suppression of estrogen production or inhibition of estrogen effects in the body. In the art, an endocrine treatment is sometimes referred to as an anti-hormonal treatment. For example, the endocrine treatment may be tamoxifen treatment or aromatase inhibitor treatment.
The chemotherapy may be mono- or polychemotherapy, such as treatment with CMF (cyclophosphamide, methotrexate, 5-fluorouracil), FEC (fluorouracil, epirubicin, cyclofosfamid) and/or taxanes.
In embodiments of the second aspect, e.g., if the prognosis of the patient is found to be particularly poor, the adjuvant breast cancer treatment may be an anti-estrogen treatment which blocks the ER. An example of such treatment is fulvestrant treatment.
Also, if the breast cancer is HER2 positive (HER2+), the adjuvant breast cancer treatment may be an anti-HER2 treatment, such as a treatment with an anti-HER2 antibody, e.g., trastuzumab.
Further, in embodiments of the second aspect, the combination of breast cancer treatment and endocrine treatment may be a combination of one or more of the above-mentioned chemotherapies and one or more of the above-mentioned endocrine therapies. An example of such a combination is sequential chemo-endocrine therapy, wherein the subjects is first treated with a chemotherapy which is followed by an endocrine treatment. As an example, such sequential chemo-endocrine therapy has traditionally been given to HR+ subjects. For example, by utilizing one of the methods comprising the step e), this HR+ group may be relieved from the chemotherapy part and/or the endocrine therapy part of such a sequential chemo-endocrine therapy.
A recent in vitro study on colon cancer cells confirmed that estrogen has a regulatory effect on HMGCR protein activity (Messa C et al. (2005) Scand J Gastroenterol 40:1454-61). The importance of HMGCR protein in cancer development is further indicated from studies on statins. Statins act as reversible HMGCR inhibitors (Lennernas H and Fager G (1997) Clin Pharmacokinet 32:403-25.) and are widely used in the treatment of hypercholesterolemia. Recent findings propose that statins exhibit anti-tumoral properties as well: in vitro studies on prostate and breast cancer cell lines suggest that lipophilic statins inhibit proliferation via arrest in the G1-phase of the cell cycle, probably through induction of cdk-inhibitors, e.g., p27 (Sivaprasad U et al. (2006) Mol Cancer Ther 5:2310-6, Campbell M J et al. (2006) Cancer Res 66:8707-14.), and a decreased expression of G1-S phase stimulators (Duncan R E, et al. (2004) Biochem Pharmacol 68:1739-47). Some epidemiological studies report an up to 50% reduced risk of cancer among statin users (Graaf M R et al. (2004) J Clin Oncol 22:2388-94, Strandberg T E et al. (2004) Lancet 364:771-7, Poynter J N, et al. (2005) N Engl J Med 352:2184-92.) whereas others could not support the association between statin use and breast cancer (Boudreau D M, et al. (2007) Cancer Epidemiol Biomarkers Prey 16:416-21, Bonovas S, (2005) J Clin Oncol 23:8606-12.).
Based on e.g., what is discussed above, the inventors predict that statins may be a suitable adjuvant treatment for HMGCR protein high subjects having a poor prognosis, i.e., HMGCR protein high subjects having HR− breast cancer, such as ER− breast cancer.
Further, in this group, statins may be given in combination with or, as an alternative to, chemotherapy. Consequently, the HMGCR level may also be a predictor of response to adjuvant treatment/chemoprevention with statins after e.g. breast conserving therapy of ductal carcinoma in situ (DCIS) of the breast. This indication is particularly interesting for subjects having ER− cancers since such subjects generally do not respond to endocrine treatment.
Consequently, in an alternative or complementary embodiment of the second aspect, the adjuvant breast cancer treatment may be a statintreatment. The statin treatment may for example be treatment with a lipophilic/hydrophobic statin, such as fluvastatin, lovastatin or simvastatin. Other examples of lipophilic/hydrophoboc statins are atorvastatin and cerivastatin. In some embodiments may hydrophilic statins like pravastatin and rosuvastatin be used.
Also, statins may be given as chemoprevention to subjects at a relatively high risk for developing breast cancer, subjects with precancerous conditions or subjects who are at risk for a second primary cancer. For example, a high-risk subject may be a subject who has previously suffered from a ER− and HMGCR high breast cancer.
Further, as an alternative aspect of the present disclosure, there is provided a method for determining whether statin treatment is suitable for a mammalian subject having, or having a high risk of developing, a hormone receptor negative breast cancer, comprising the steps of:

- a) providing a sample from the subject;
- b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;
- c) comparing the sample value obtained in step b) with a reference value; and, if said sample value is higher than said reference value,
- d) concluding that statin treatment is suitable for the subject.

Regarding step b) of the above method, an increase in the amount of HMGCR protein typically results in an increase in the sample value, and not the other way around.
In one embodiment, the sample is an earlier obtained sample.
This alternative aspect is based on e.g., the facts that subjects having HR−, e.g., ER−, cancers generally do not respond to endocrine treatment and that other chemotherapies frequently entails considerable side-effects.
A mammalian subject having a high risk of developing a hormone receptor negative breast cancer may be a subject who has previously suffered from a HR− breast cancer which is now considered cured. Such subject may also present other indicators of high risk, such as hereditary characteristics, e.g. a history of breast cancer in the family. The risk may also be assessed according to a model, such as the Gail model.
For example, a subject, such as the subject of the above method, may have had a DCIS removed and be considered disease free, i.e. free from breast cancer. If a sample obtained from such removed DCIS shows to be HR− and HMGCR+, e.g. by means of the above method, it may be concluded that statin treatment is suitable for the subject, even though the subject is considered to be disease free. Consequently, in such case statins may be used as chemoprevention. Today, many subjects only undergo radiotherapy after removal of DCIS, e.g., by breast conserving surgery, but according to the present disclosure, they may be considered likely to benefit from statin treatment if their breast cancer was ER− and HMGCR+.
A treatment being “suitable for” a subject refers to the case wherein the benefits of the treatment compensate for the disadvantages of the same. The benefits of the adjuvant breast cancer treatment refers to a higher probability of survival or recovery if undergoing the treatment than if not undergoing the treatment. The “disadvantages of the adjuvant breast cancer treatment” refers to the drawbacks, such as side-effects, pain or other inconveniences and costs.
In an embodiment of the alternative aspect above, the breast cancer of the subject is further HER2−. The negativity for HER2 leaves the subject with even fewer treatment alternatives, i.e., anti-HER2 treatment is excluded, and therefore, statin treatment becomes an even more suitable alternative. Also, HMGCR high subjects having HER2− and ER− have a particularly poor prognosis (FIG. 11).
In general, when deciding on a suitable treatment strategy for a patient having breast cancer, the physician responsible for the treatment may take several parameters into account, such as the result of an immunohistochemical evaluation, patient age, menopausal status, hormone receptor status, general condition and medical history, such as breast cancer history. To be guided in such decision, the physician may perform an HMGCR protein test, or order an HMGCR protein test performed, according to an embodiment of one of the method aspects presented above.
In the context of the present disclosure, “prognosis” refers to the prediction of the course or outcome of a disease and its treatment. For example, prognosis may also refer to a determination of chance of survival or recovery from a disease, as well as to a prediction of the expected survival time of a subject. It may also refer to the likelihood of disease recurrence, e.g., local, regional or distant events. Further, a prognosis may involve establishing the likelihood for survival of a subject during a period of time into the future, such as three years, five years, ten years or any other period of time.
Further, in the context of the present disclosure, “a mammalian subject having a breast cancer” refers to a mammalian subject having a primary or secondary breast tumor or a mammalian subject which has had a tumor removed from the breast, wherein the removal of the tumor refers to killing or removing the tumor by any type of surgery or therapy. “Breast tumor” includes ductal carcinoma in situ (DCIS). In the method and use aspects of the present disclosure, “a mammalian subject having a breast cancer” also includes the case wherein the mammalian subject is suspected of having a breast cancer at the time of the performance of the use or method and the breast cancer diagnosis is established later.
Further, in the context of the inventive methods of present disclosure, “earlier obtained” refers to obtained before the inventive method is performed. Consequently, if a sample earlier obtained from a subject is provided in a method, the method does not involve obtaining the sample from the subject, i.e., the sample was previously obtained from the subject in a step separate from the method.
Still further, in the context of the present disclosure, the “reference value” refers to a predetermined value found to be relevant for making decisions, or drawing conclusions, regarding the prognosis, or a suitable treatment strategy, for the subject.
Also, in the context of the present disclosure, a reference prognosis being “associated” with a reference value refers to the reference value being assigned to the reference prognosis, based on empirical data and/or clinically relevant assumptions. The reference value does not have to be assigned to a reference prognosis directly derived from prognosis data of a group of subjects exhibiting the reference value. The reference prognosis may for example correspond to the prognosis for subjects exhibiting the reference value or lower. That is, if the reference value is 1 on a scale from 0 to 3, the reference prognosis may be the prognosis of the subjects exhibiting the values 0 or 1. Consequently, the reference prognosis may also be adapted to the nature of the available data.
Step b) of the methods of the above aspects involve evaluating the amount of HMGCR protein present in at least part of the sample, and determining a sample value corresponding to the amount. The “at least part of the sample” refers to a relevant part, or relevant parts, of the sample for establishing the prognosis or drawing conclusions regarding suitable treatments. The person skilled in the art understands which part or parts that are relevant under the circumstances present when performing the method. For example, if evaluating at a sample comprising cells, the skilled person may only consider the tumor cells, or only the cytoplasms of tumor cells, of the sample.
Further, in step b) an amount is evaluated and a sample value corresponding to the amount is determined. Consequently, an exact measurement of the amount of HMGCR protein is not required for obtaining the sample value. For example, the amount of HMGCR protein may be evaluated by visual inspection of a stained tissue sample and the sample value may then be categorized, e.g., as high or low based on the evaluated amount.
In the context of the present disclosure, unless otherwise is stated, “hormone receptor positive” or “HR+” refer to estrogen receptor positive (ER+) and/or progesterone receptor positive (PR+), and accordingly, “hormone receptor negative” or “HR−” refer to estrogen receptor negative (ER−) and progesterone receptor negative (PR−).
However, in alternative embodiments, “hormone receptor positive” refers to “estrogen receptor positive” and “hormone receptor negative” refers to “estrogen receptor negative” (see e.g., FIG. 9A). Further, in other alternative embodiments, “hormone receptor positive” refers to “progesterone receptor positive” and “hormone receptor negative” refers to “progesterone receptor negative” (see e.g., FIG. 9B). Consequently, in these embodiments, it is not necessary to know both the ER status and the PR status to establish the HR status.
The person skilled in the art knows how to obtain the ER and/or PR status of a patient. For example, such information may be obtained from the result of a test using the commercially available ER/PR pharmDX kit (DakoCytomation). As another example, the method disclosed by Allred et al. (Allred et al. (1998) Mod Pathol 11(2), 155) may be used to obtain a total score (Allred score), and, for both ER and PR, an Allred score of higher than two is considered positive and an Allred score of two or lower is considered negative. Alternatively, when classifying a sample as being positive or negative for ER or PR, a cutoff of 10% positive cells may be used, which is a recognized limit within the art.
For example, a specimen from an earlier surgical removal of the breast cancer tumor or an earlier obtained biopsy of the tumor may be used for the establishment of HR status. The sample of step a) of the methods of the above aspects may also be obtained from such specimen or biopsy.
As shown in FIG. 11, within the group of subjects having ER− cancers, the correlation between high HMGCR values and a poor prognosis is particularly accentuated for HER2 negative (HER2−) cancers.
Consequently, in embodiments of the methods of the above aspects, the HR− cancer is further HER2−. In such embodiments, HR− may preferably refer to ER−.
The person skilled in the art knows how to obtain the HER2 status of a patient. For example, such information may be obtained from the result of a test using the commercially available HER2 FISH pharmDx™ kit (Dako).
Statin treatment may be particularly suitable for subjects having a HR− and HER2− cancer and a poor prognosis (HMGCR high), since these subjects are not considered to respond to endocrine or anti-HER2 (such as anti-HER2 antibody, e.g., trastuzumab) treatment.
Consequently, if the cancer is HR− and HER2−, the above-mentioned adjuvant breast cancer treatment of the second aspect may be statin treatment.
However, the actual step of determining the HR, ER, PR and/or HER2 status of the subject does not form part of the above-mentioned methods.
In embodiments of the methods of the above aspects, the sample may be a body fluid sample, such as a sample of blood, plasma, serum, cerebral fluid, urine, or exudate. Alternatively, the sample may be a cytology sample or a stool sample. The body fluid, cytology or stool sample may for example comprise tumor cells.
In further embodiments of the methods of the above aspects, the sample may be a tissue sample. As an example, the tissue sample may be derived from a primary breast tumor, such as in situ or invasive carcinoma, or secondary tumor (metastasis) of the subject.
In embodiments of the methods of the above aspects, the evaluation of step b) may be limited to evaluating the amount of HMGCR protein expression in tumor cells, such as cells originating from epithelial cells, e.g., glandular cells, of the sample, e.g., an earlier obtained tumor tissue biopsy material or specimen from a surgical removal of a breast cancer. The inventors have found that HMGCR protein is expressed in the cytoplasm. Consequently, in embodiments of the methods of the above aspects, the evaluation of step b) may be limited to evaluating the amount of HMGCR protein expression in the cytoplasm of tumor cells of the tissue sample.
T1N0M0 subjects are a frequently over-treated group of breast cancer subjects. The findings of the present disclosure may relieve this group from unnecessary treatment. Also, at this early stage it is particularly beneficial to identify aggressive breast cancer forms to avoid under-treatment which may lead to death. Accordingly, in embodiments of the methods of the above aspects, the breast cancer may be a breast cancer previously classified as T1N0M0 according to the TNM staging system.
Also, in embodiments of the methods of the above aspects, the breast cancer may be a node negative cancer. A “node negative cancer” refers to a cancer that has not spread to the lymph nodes. The above methods, especially those relating to the non-treatment of the subject, may be particularly useful for such less advanced stages of breast cancer.
In embodiments of the methods of the above aspects, said prognosis may be a probability of survival, that is, said prognosis for said subject is a probability of survival and said reference prognosis is a probability of survival. For example, the probability of survival may be a probability of five-year survival, ten-year survival or 15-year survival. Further, the probability of survival may for example be selected from the group consisting of probability of overall survival, probability of recurrence-free survival and probability of breast cancer specific survival. In some embodiments, especially if the group of HR+ subjects is particularly relevant, the probability of survival may be a probability of recurrence-free survival or a probability of breast cancer specific survival
As shown in the attached figures, the correlation between high HMGCR protein levels and a good prognosis among HR+ subjects is particularly accentuated if recurrence-free survival (RFS) (FIG. 1-5) is analyzed. Consequently, in embodiments of the methods of the above aspects, the prognosis may be a probability of recurrence-free survival. If the prognosis is a probability of recurrence-free survival it follows that the reference prognosis preferably should be a probability of recurrence-free survival.
As also shown in the attached figures, the correlation between high HMGCR protein levels and a poor prognosis among HR− subjects is particularly accentuated if overall survival (OS) is analyzed (FIG. 9A-C). Consequently, in embodiments of the methods of the above aspects, the prognosis may be a probability of overall survival. If the prognosis is a probability of overall survival it follows that the reference prognosis preferably should be a probability of overall survival.
The “first reference prognosis” and the “second reference prognosis” do not have to be the same type of prognosis. For example, the first reference prognosis may be a probability of recurrence-free survival whereas the second reference prognosis is a probability of overall survival. In such case, the prognosis for the subject is a probability of recurrence-free survival if the first reference prognosis applies, i.e., if the breast cancer is HR+ and a probability of overall survival if the second reference prognosis applies, i.e., if the breast cancer is HR−.
Consequently, in embodiments of the methods of the above aspects, the first and second reference prognosis may each independently be a probability of recurrence-free, breast cancer specific or overall survival.
A sample value of HMGCR protein being higher than the reference value, or a subject from which such sample value is obtained, is sometimes referred to herein as “HMGCR protein high”. Further, a sample value of HMGCR protein being lower than, or equal to, the reference value, or a subject from which such sample value is obtained, is sometimes referred to herein as “HMGCR protein low”.
In the context of the present disclosure, the terms “sample value” and “reference value” are to be interpreted broadly. The quantification of HMGCR protein to obtain these values may be done via automatic means, via a scoring system based on visual or microscopic inspection of samples, or via combinations thereof. However, it is also possible for a skilled person, such as a person skilled in the art of histopathology, to determine the sample and reference values merely by inspection, e.g., of tissue slides that have been stained for HMGCR protein expression. The determination of the sample value being higher than the reference value may thus correspond to the determination, upon visual or microscopic inspection, that a sample tissue slide is more densely stained and/or exhibit a larger fraction of stained cells than is the case for a reference tissue slide. The sample value may also be compared to a reference value given by a literal reference, such as a reference value described in wording. Consequently, the sample and/or reference values may be thought of as mental values that the skilled person determines upon inspection and comparison.
For example, the skilled person may categorize a sample as being HMGCR protein high or low, wherein the sample is categorized as high if it contains more HMGCR protein than a previously inspected reference sample and low if it contains less or equally much. Such evaluation may be assisted by staining the sample, and, if necessary, a reference sample, with a staining solution comprising e.g., antibodies selective for HMGCR protein.
A reference value found to be relevant for the provision of a prognosis for a subject having a breast cancer, or for making treatment decisions regarding such subjects, for use as comparison with the sample value from the subject, may be provided in various ways. With the knowledge of the teachings of the present disclosure, the skilled artisan can, without undue burden, provide relevant reference values for performing the methods of the above aspects.
The person performing the methods of the above aspects may, for example, adapt the reference value to desired prognostic information. For example, the reference value may be adapted to yield the most significant prognostic information, e.g., the largest separation between the HMGCR protein high survival curve and the HMGCR protein low survival curve (see e.g., the FIGS. 2-11). Examples of reference values that may yield a large separation are: if the commercial anti-HMGCR antibody is used, an absent cytoplasmic intensity (CI); and if the HPA anti-HMGCR antibody is used, a moderate cytoplasmic intensity. “Absent” and “moderate” CI, respectively and the “commercial” and “HPA” antibody, respectively, are defined below (Examples, Section 4).
Alternatively, the reference value may be adapted to identify a group of subjects having a predetermined prognosis, e.g., the group of subjects having a probability of five-year recurrence-free survival of higher than 80%.
In embodiments of the methods of the above aspects, the reference value may correspond to the amount of HMGCR protein expression in a healthy tissue, such as healthy breast or stroma tissue, of the subject of the method. As another example, the reference value may be provided by the amount of HMGCR protein expression measured in a standard sample of normal tissue from another, comparable subject. As another example, the reference value may be provided by the amount of HMGCR protein expression measured in a reference sample comprising tumor cells, such as a reference sample of tumor tissue, e.g., breast cancer tissue. The amount of protein expression of the reference sample may preferably be previously established. Consequently, the reference value may be provided by the amount of HMGCR protein measured in a reference sample comprising cells expressing a predetermined amount of HMGCR protein.
As another example, the reference value may be provided by the amount of HMGCR protein expression measured in a reference sample comprising cell lines, such as cancer cell lines, expressing a predetermined, or controlled, amount of HMGCR protein. The person skilled in the art understands how to provide such cell lines, for example guided by the disclosure of Rhodes et al. (2006) The biomedical scientist, p 515-520.
Consequently, in embodiments of the methods of the above aspects, the reference value may be a predetermined value corresponding to the amount of HMGCR protein expression in a reference sample.
However, as discussed further below, the amount of HMGCR protein in the reference sample does not have to directly correspond to the reference value. The reference sample may also provide amounts of HMGCR protein that help a person performing the method to assess various reference values. For example, the reference sample(s) may help in creating a mental image of the reference value by providing a “positive” and/or a “negative” value.
The inventors have shown that subjects who suffer from HR+ breast cancer and have lost essentially all their HMGCR protein expression generally have a poor prognosis. Further, a relatively good prognosis for subjects that suffer from HR− breast cancer and have lost essentially all their HMGCR protein expression has been observed. Thus, in embodiments of the methods of the above aspects, the sample value of step b) may be either 1, corresponding to detectable HMGCR protein in the sample, or 0, corresponding to no detectable HMGCR protein in the sample. Consequently, in such embodiments, the evaluation of the sample is digital: HMGCR protein is considered to be either present or not. In the context of the present disclosure, “no detectable HMGCR protein” refers to an amount of HMGCR protein that is so small that it is not, during normal operational circumstances, detectable by a person or an apparatus performing the method according to any one of the above aspects. The “normal operational circumstances” refer to the laboratory methods and techniques a person skilled in the art would find appropriate for performing the invention.
Accordingly, in embodiments of the methods of the above aspects, the reference value of step c) may be 0. And it follows that, in further embodiments of the methods of the above aspects, the reference value of step c) may correspond to a reference sample having no detectable HMGCR protein.
One alternative for the quantification of HMGCR protein expression in a sample, such as the sample earlier obtained from the subject or the reference sample, is the determination of the fraction of cells in the sample that exhibit HMGCR protein expression over a certain level. The fraction may for example be: a “cellular fraction”, wherein the HMGCR protein expression of the whole cells is taken into account; a “cytoplasmic fraction”, wherein the HMGCR protein expression of only the cytoplasms of the cells is taken into account; or a “nuclear fraction”, wherein the HMGCR protein expression of only the nuclei of the cells is taken into account. The cytoplasmic fraction may for example be classified as <2%, 2-25%, >25-75% or >75% immunoreactive cells of the relevant cell population. This determination may for example be performed as described below in the Examples, Section 3, with reference to “cytoplasmic fraction”. The “cytoplasmic fraction” corresponds to the percentage of relevant cells in a sample that exhibits a positive staining in the cytoplasm, wherein a medium or distinct and strong immunoreactivity in the cytoplasm is considered positive and no or faint immunoreactivity in the cytoplasm is considered negative. The person skilled in the art of pathology understands which cells that are relevant under the conditions present when performing the method and may determine a cytoplasmic fraction based on his general knowledge and the teachings of the present disclosure. The relevant cells may for example be tumor cells. Further, the skilled artisan understands how to perform corresponding measurements employing the “cellular fraction” or the “nuclear fraction”.
Another alternative for the quantification of HMGCR protein expression in a sample, such as the sample earlier obtained from the subject or the reference sample, is the determination of the overall staining intensity of the sample. The intensity may for example be: a “cellular intensity”, wherein the HMGCR protein expression of the whole cells is taken into account; a “cytoplasmic intensity”, wherein the HMGCR protein expression of only the cytoplasms of the cells is taken into account, or a “nuclear intensity”, wherein the HMGCR protein expression of only the nuclei of the cells is taken into account. Cytoplasmic intensity is subjectively evaluated in accordance with standards used in clinical histopathological diagnostics. Outcome of a cytoplasmic intensity determination may be classified as: absent=no overall immunoreactivity in the cytoplasm of relevant cells of the sample, weak=faint overall immunoreactivity in the cytoplasm of relevant cells of the sample, moderate=medium overall immunoreactivity in the cytoplasm of relevant cells of the sample, or strong=distinct and strong overall immunoreactivity in the cytoplasm of relevant cells of the sample. The person skilled in the art understands which cells that are relevant under the conditions present when performing the method and may determine a cytoplasmic intensity based on his general knowledge and the teachings of the present disclosure. The relevant cells may for example be tumor cells. This determination may for example be performed as described below in the Examples, Section 4, definition of “cytoplasmic intensity”. Further, the skilled artisan understands how to perform corresponding measurements employing the “cellular intensity” or the “nuclear intensity”.
The inventors have found that the cytoplasmic expression of HMGCR protein is particularly relevant for establishing a prognosis. Thus, in embodiments of the methods of the above aspects, the reference value may be a cytoplasmic fraction, a cytoplasmic intensity or a combination thereof. Accordingly, the sample value may be a cytoplasmic fraction, a cytoplasmic intensity or a combination thereof.
Preferably, the sample value and the reference value are both the same type of value. Accordingly, in embodiments of the methods of the above aspects, the sample value and the reference value may each be a cytoplasmic fraction, a cytoplasmic intensity or a combinations thereof.
In embodiments of the methods of the above aspects, the criterion for the conclusion in step d) is a sample value for the cytoplasmic fraction of HMGCR protein positive cells, i.e., a “cytoplasmic fraction”, which is higher than a reference value which is 1%, such as higher than 5%, such as higher than 10%, such as higher than 15%, such as higher than 20%, such as higher than 25%, such as higher than 30%, such as higher than 35%, such as higher than 40%, such as higher than 50%, such as higher than 55%, such as higher than 60%, such as higher than 65%, such as higher than 70%, such as higher than 75%, such as higher than 80%, such as higher than 85%, such as higher than 90%, such as higher than 95%.
In alternative or complementing embodiments of the methods of the above aspects, the reference value of step c) is a cytoplasmic fraction of 1% or higher, such as 5% or higher, such as 10% or higher, such as 15% or higher, such as 20% or higher, such as 25% or higher, such as 30% or higher, such as 35% or higher, such as 40% or higher, such as 45% or higher, such as 50% or higher, such as 55% or higher, such as 60% or higher, such as 65% or higher, such as 70% or higher, such as 75% or higher, such as 80% or higher, such as 90% or higher, such as 95% or higher.
Further, in embodiments of the methods of the above aspects, the criterion for the conclusion in step d) may be a sample value for staining intensity of a sample, i.e., a cytoplasmic intensity, which is higher than absent cytoplasmic intensity, such as higher than weak cytoplasmic intensity, such as higher than moderate cytoplasmic intensity. In alternative or complementing embodiments of the methods of the above aspects, the reference value of step c) may be an absent cytoplasmic intensity of HMGCR protein expression or higher, such as a weak cytoplasmic intensity of HMGCR protein expression or higher, such as moderate cytoplasmic intensity of HMGCR protein expression.
Further, in embodiments of the methods of the above aspects, the reference value may be constituted of two values, wherein the criterion for the conclusion in step d) is a sample value being higher than any one of these two values.
Alternatively, in embodiments of the methods of the above aspects, the reference value may be a combination of a fraction value and an intensity value, such as a cytoplasmic fraction value and a cytoplasmic intensity value.
Also, in embodiments of the methods of the above aspects, the reference value may be a function of a cytoplasmic fraction value and a cytoplasmic intensity value. For example, such a function may be a staining score. The “staining score” is calculated as described in Examples, Section 3 and table 1 below. For example, the reference value may be a staining score of 0 or higher, such as 1 or higher, such as 2.
The person skilled in the art realizes that other reference values being an intensity value or a fraction value also fall within the scope of the present invention. Likewise, the person skilled in the art realizes that other combinations of fractions and intensities also fall within the scope of the present invention. Consequently, the reference value may involve two, and possibly even more, criteria.
In general, the selection of a cytoplasmic intensity value and/or a cytoplasmic fraction value as the reference value may depend on the staining procedure, e.g., on the employed anti-HMGCR antibody and on the staining reagents. For example, as shown in Examples and the attached figures of the present disclosure, the commercial anti-HMGCR antibody yields a stronger staining than the HPA anti-HMGCR antibody.
Guided by the present disclosure, and especially Examples, Sections 3-5 below, a person skilled in the art, e.g., a pathologist, understands how to perform the evaluation yielding a fraction, such as a cellular, cytoplasmic or nuclear fraction, or an intensity, such as a cellular, cytoplasmic or nuclear intensity. For example, the skilled artisan may use a reference sample comprising a predetermined amount of HMGCR protein for establishing the appearance of a certain fraction or intensity.
However, a reference sample may not only be used for the provision of the actual reference value, but also for the provision of an example of a sample with an amount of HMGCR protein that is higher than the amount corresponding to the reference value. As an example, in histochemical staining, such as in immunohistochemical staining, the skilled artisan may use a reference sample for establishing the appearance of a stained sample having a high amount of HMGCR protein, e.g., a positive reference. Subsequently, the skilled artisan may assess the appearances of samples having lower amounts of HMGCR protein, such as the appearance of a sample with an amount of HMGCR protein corresponding to the reference value. In other words, the skilled artisan may use a reference sample to create a mental image of a reference value corresponding to an amount of HMGCR protein which is lower than that of the reference sample. Alternatively, or as a complement, in such assessments, the skilled artisan may use another reference sample having a low amount of HMGCR protein, or essentially lacking HMGCR protein, for establishing the appearance of such sample, e.g., as a “negative reference”.
Consequently, in the evaluation, the skilled artisan may use a reference sample for establishing the appearance of a sample with a high amount of HMGCR protein. Such reference sample may be a sample comprising tissue expressing a high amount of HMGCR protein, such as a sample comprising breast tumor tissue having a pre-established high expression of HMGCR protein.
Accordingly, the reference sample may provide an example of a strong cytoplasmic intensity (CI). With the knowledge of the appearance of a sample with strong CI, the skilled artisan may then divide samples into the CI categories presented in Examples, Sections 3 and 4, below, i.e., absent, weak, moderate and strong. This division may be further assisted by a reference sample essentially lacking HMGCR protein (negative reference), i.e., a reference sample providing an absent cytoplasmic intensity. Also, the reference sample may provide an example of a sample with a cytoplasmic fraction (CF) of 75% or higher. With the knowledge of the appearance of a sample with more than 75% positive cells, the skilled artisan may then evaluate the cytoplasmic fraction of other samples having e.g., a lower percentage of positive cells. This division may be further assisted by a reference sample essentially lacking HMGCR protein (negative reference), i.e., a reference sample providing a low CF (e.g., <5%, such as <2%), or a CF of 0.
As mentioned above, cell lines expressing a controlled amount of HMGCR protein may be used as the reference, in particular as a positive reference.
As discussed above, the methods according to the above aspects may be adapted to a selected reference value, such as one of the reference values presented above, and the reference prognosis will in such case be a consequence of the selected reference value. As a non-limiting example, if an absent cytoplasmic intensity (CI=0) is used as the reference value (see Examples, Section 4), an associated reference prognosis may be derived from FIG. 4A (subjects having HR+ breast cancer, staining with the commercial anti-HMGCR antibody and RFS analyzed) by looking at the CI=0 curve (dotted line). At a given time from diagnosis, the cumulative survival corresponding to the CI=0 curve may be read from the figure, e.g., 67% after 5 years (60 months), which results in a reference prognosis being a probability of five-year recurrence-free survival of 67%. Consequently, subjects of the same group having sample values which are higher than the reference value CI=0 have a probability of five year recurrence-free survival which is higher than 67%. Using the same logic, the associated reference prognosis may be a probability of five-year recurrence-free survival of higher than 74% if using a moderate cytoplasmic intensity (CI=2) as the reference value (FIG. 4B, subjects having HR+ breast cancer and staining with the HPA anti-HMGCR antibody).
In the latter example, wherein a moderate cytoplasmic intensity (CI=2) is used as the reference value, the reference prognosis is based on subjects exhibiting a CI of 0, 1 or 2, i.e., the reference prognosis is not directly derived from a group of subjects exhibiting the reference value. However, the conclusion that the subjects having a sample value which is higher than such reference value have a prognosis which is better than such reference prognosis is still true. It may be of interest to use a more detailed approach wherein survival curves for each level of CI are established. An example of such an approach is shown in FIG. 1, wherein the RFS for subjects having ER+ breast cancer have been analyzed (staining with the HPA anti-HMGCR antibody). Starting from data like those presented in FIG. 1, a more precise reference prognosis may be obtained, which for example may be particularly useful if conclusions regarding sample values being lower than, or equal to, the reference value are to be drawn. When establishing such reference prognoses, it is beneficial to have data from a large patient group to make reliable predictions because the number of patients representing each CI value is relatively small.
If looking at HR− subjects, the logic is similar. Turning for example to FIG. 9C, wherein the OS for subjects having HR− breast cancer have been analyzed (staining with the commercial anti-HMGCR antibody). In FIG. 9C, an absent cytoplasmic intensity (CI=0) is used as the reference value (see Examples, Section 4) and an associated reference prognosis may be derived by looking at the CI=0 curve (dotted line). At a given time from diagnosis, the cumulative survival corresponding to the CI=0 curve may be read from the figure, e.g., 78% after 5 years (60 months), which results in a reference prognosis being a probability of five-year overall survival of 78%. Consequently, subjects of the same group having sample values which are higher than the reference value CI=0 have a probability of five year overall survival which is lower than 78%. However, as may be seen in FIG. 9C, much lower probabilities of survival are probably detectable if the group CI>0 is dived into the categories CI=1, CI=2 and CI=3, respectively, as done in FIG. 1.
Consequently, in embodiments of the methods of the above aspects, the first reference prognosis may be a likelihood of five-year recurrence-free survival of 60-80%, such as 67-74%. Further, in embodiments of the methods of the above aspects, the second reference prognosis may be a likelihood of five-year overall survival of 68-88%, such as 73-83%.
The data of the present disclosure is based on groups of human females. Consequently, in embodiments of the methods of the above aspects, the mammalian subject may be a human. In further embodiments of the methods of the above aspect, the mammalian subject may be a female, such as a premenopausal or postmenopausal female.
The skilled person should recognize that the usefulness of the present invention is not limited to the quantification of any particular variant of the HMGCR protein present in the subject in question, as long as the protein is encoded by the relevant gene and presents the relevant pattern of expression. As a non-limiting example, the HMGCR protein has an amino acid sequence which comprises a sequence selected from:
i) SEQ ID NO:1; and
ii) a sequence which is at least 85% identical to SEQ ID NO:1.
In some embodiments, sequence ii) above is at least 90% identical, at least 91% identical, at least 92% identical, at least 93% identical, at least 94% identical, at least 95% identical, at least 96% identical, at least 97% identical, at least 98% identical or at least 99% identical to SEQ ID NO:1.
As another non-limiting example, the HMGCR protein has an amino acid sequence which comprises a sequence selected from:
i) SEQ ID NO:2; and
ii) a sequence which is at least 85% identical to SEQ ID NO:2.
In some embodiments, sequence ii) above is at least 90% identical, at least 91% identical, at least 92% identical, at least 93% identical, at least 94% identical, at least 95% identical, at least 96% identical, at least 97% identical, at least 98% identical or at least 99% identical to SEQ ID NO:2.
In embodiments of the methods of the aspects above, the HMGCR protein may be detected and/or quantified through the application to a sample of a detectable and/or quantifiable affinity ligand, which is capable of selective interaction with the HMGCR protein. The application of the affinity ligand is performed under conditions that enable binding of the affinity ligand to any HMGCR protein in the sample.
To concretize, in embodiments of the methods of the aspects above, step b) may comprise:
b1) applying to the sample a quantifiable affinity ligand capable of selective interaction with the HMGCR protein to be evaluated, said application being performed under conditions that enable binding of the affinity ligand to any HMGCR protein present in the sample;
b2) removing non-bound affinity ligand; and
b3) quantifying any affinity ligand remaining in association with the sample to evaluate said amount.
“Affinity ligand remaining in association with the sample” refers to affinity ligand which was not removed in step b2), e.g., the affinity ligand bound to the sample.
It is regarded as within the capabilities of those of ordinary skill in the art to select or manufacture the proper affinity ligand and to select the proper format and conditions for detection and/or quantification, once the connection between HMGCR protein and a prognosis for breast cancer is known through the teaching of the present disclosure. Nevertheless, examples of affinity ligands that may prove useful, as well as examples of formats and conditions for detection and/or quantification, are given below for the sake of illustration.
Thus, in some embodiments of the methods of the above aspects, an affinity ligand is used, which is selected from the group consisting of antibodies, fragments thereof and derivatives thereof, i.e., affinity ligands based on an immunoglobulin scaffold. For example, the antibodies may be isolated and/or mono-specific. For example, antibodies comprise monoclonal and polyclonal antibodies of any origin, including murine, rabbit, human and other antibodies, as well as chimeric antibodies comprising sequences from different species, such as partly humanized antibodies, e.g., partly humanized mouse antibodies. Polyclonal antibodies are produced by immunization of animals with the antigen of choice, whereas monoclonal antibodies of defined specificity can be produced using the hybridoma technology developed by Köhler and Milstein (Köhler G and Milstein C (1976) Eur. J. Immunol. 6:511-519). Antibody fragments and derivatives comprise Fab fragments, consisting of the first constant domain of the heavy chain (CH1), the constant domain of the light chain (CL), the variable domain of the heavy chain (VH) and the variable domain of the light chain (VL) of an intact immunoglobulin protein; Fv fragments, consisting of the two variable antibody domains VH and VL (Skerra A and Plückthun A (1988) Science 240:1038-1041); single chain Fv fragments (scFv), consisting of the two VH and VL domains linked together by a flexible peptide linker (Bird R E and Walker B W (1991) Trends Biotechnol. 9:132-137); Bence Jones dimers (Stevens F J et al. (1991) Biochemistry 30:6803-6805); camelid heavy-chain dimers (Hamers-Casterman C et al. (1993) Nature 363:446-448) and single variable domains (Cai X and Garen A (1996) Proc. Natl. Acad. Sci. U.S.A. 93:6280-6285; Masat L et al. (1994) Proc. Natl. Acad. Sci. U.S.A. 91:893-896), and single domain scaffolds like e.g., the New Antigen Receptor (NAR) from the nurse shark (Dooley H et al. (2003) Mol. Immunol. 40:25-33) and minibodies based on a variable heavy domain (Skerra A and Plückthun A (1988) Science 240:1038-1041). In embodiments of the methods of the above aspects, the quantifiable affinity ligand, e.g., an antibody or fragment thereof, is obtainable by a process comprising a step of immunizing an animal with a protein whose amino acid sequence comprises the sequence SEQ ID NO:1, preferably with a protein having the sequence SEQ ID NO:1.
Polyclonal and monoclonal antibodies, as well as their fragments and derivatives, represent the traditional choice of affinity ligands in applications requiring selective biomolecular recognition, such as in the detection and/or quantification of HMGCR protein according to the method aspects above. However, those of skill in the art know that, due to the increasing demand of high throughput generation of selective binding ligands and low cost production systems, new biomolecular diversity technologies have been developed during the last decade. This has enabled a generation of novel types of affinity ligands of both immunoglobulin as well as non-immunoglobulin origin that have proven equally useful as binding ligands in biomolecular recognition applications and can be used instead of, or together with, immunoglobulins.
The biomolecular diversity needed for selection of affinity ligands may be generated by combinatorial engineering of one of a plurality of possible scaffold molecules, and specific and/or selective affinity ligands are then selected using a suitable selection platform. The scaffold molecule may be of immunoglobulin protein origin (Bradbury A R and Marks J D (2004) J. Immunol. Meths. 290:29-49), of non-immunoglobulin protein origin (Nygren P Å and Skerra A (2004) J. Immunol. Meths. 290:3-28), or of an oligonucleotide origin (Gold L et al. (1995) Annu. Rev. Biochem. 64:763-797).
A large number of non-immunoglobulin protein scaffolds have been used as supporting structures in development of novel binding proteins. Non-limiting examples of such structures, useful for generating affinity ligands against HMGCR protein for use according to the present disclosure, are staphylococcal protein A and domains thereof and derivatives of these domains, such as protein Z (Nord K et al. (1997) Nat. Biotechnol. 15:772-777); lipocalins (Beste G et al. (1999) Proc. Natl. Acad. Sci. U.S.A. 96:1898-1903); ankyrin repeat domains (Binz H K et al. (2003) J. Mol. Biol. 332:489-503); cellulose binding domains (CBD) (Smith G P et al. (1998) J. Mol. Biol. 277:317-332; Lehtiö J et al. (2000) Proteins 41:316-322); γ crystallines (Fiedler U and Rudolph R, WO01/04144); green fluorescent protein (GFP) (Peelle B et al. (2001) Chem. Biol. 8:521-534); human cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) (Hufton S E et al. (2000) FEBS Lett. 475:225-231; Irving R A et al. (2001) J. Immunol. Meth. 248:31-45); protease inhibitors, such as Knottin proteins (Wentzel A et al. (2001) J. Bacteriol. 183:7273-7284; Baggio R et al. (2002) J. Mol. Recognit. 15:126-134) and Kunitz domains (Roberts B L et al. (1992) Gene 121:9-15; Dennis M S and Lazarus R A (1994) J. Biol. Chem. 269:22137-22144); PDZ domains (Schneider S et al. (1999) Nat. Biotechnol. 17:170-175); peptide aptamers, such as thioredoxin (Lu Z et al. (1995) Biotechnology 13:366-372; Klevenz B et al. (2002) Cell. Mol. Life Sci. 59:1993-1998); staphylococcal nuclease (Norman T C et al. (1999) Science 285:591-595); tendamistats (McConell S J and Hoess R H (1995) J. Mol. Biol. 250:460-479; Li R et al. (2003) Protein Eng. 16:65-72); trinectins based on the fibronectin type III domain (Koide A et al. (1998) J. Mol. Biol. 284:1141-1151; Xu L et al. (2002) Chem. Biol. 9:933-942); and zinc fingers (Bianchi E et al. (1995) J. Mol. Biol. 247:154-160; Klug A (1999) J. Mol. Biol. 293:215-218; Segal D J et al. (2003) Biochemistry 42:2137-2148).
The above-mentioned examples of non-immunoglobulin protein scaffolds include scaffold proteins presenting a single randomized loop used for the generation of novel binding specificities, protein scaffolds with a rigid secondary structure where side chains protruding from the protein surface are randomized for the generation of novel binding specificities, and scaffolds exhibiting a non-contiguous hyper-variable loop region used for the generation of novel binding specificities.
In addition to non-immunoglobulin proteins, oligonucleotides may also be used as affinity ligands. Single stranded nucleic acids, called aptamers or decoys, fold into well-defined three-dimensional structures and bind to their target with high affinity and specificity. (Ellington A D and Szostak J W (1990) Nature 346:818-822; Brody E N and Gold L (2000) J. Biotechnol. 74:5-13; Mayer G and Jenne A (2004) BioDrugs 18:351-359). The oligonucleotide ligands can be either RNA or DNA and can bind to a wide range of target molecule classes.
For selection of the desired affinity ligand from a pool of variants of any of the scaffold structures mentioned above, a number of selection platforms are available for the isolation of a specific novel ligand against a target protein of choice. Selection platforms include, but are not limited to, phage display (Smith G P (1985) Science 228:1315-1317), ribosome display (Hanes J and Plückthun A (1997) Proc. Natl. Acad. Sci. U.S.A. 94:4937-4942), yeast two-hybrid system (Fields S and Song O (1989) Nature 340:245-246), yeast display (Gai S A and Wittrup K D (2007) Curr Opin Struct Biol 17:467-473), mRNA display (Roberts R W and Szostak J W (1997) Proc. Natl. Acad. Sci. U.S.A. 94:12297-12302), bacterial display (Daugherty P S (2007) Curr Opin Struct Biol 17:474-480, Kronqvist N et al. (2008) Protein Eng Des Sel 1-9, Harvey B R et al. (2004) PNAS 101(25):913-9198), microbead display (Nord O et al. (2003) J Biotechnol 106:1-13, WO01/05808), SELEX (System Evolution of Ligands by Exponential Enrichment) (Tuerk C and Gold L (1990) Science 249:505-510) and protein fragment complementation assays (PCA) (Remy I and Michnick S W (1999) Proc. Natl. Acad. Sci. U.S.A. 96:5394-5399).
Thus, in embodiments of the methods of the above aspects, an affinity ligand may be used, which is a non-immunoglobulin affinity ligand derived from any of the protein scaffolds listed above, or an oligonucleotide molecule.
The HMGCR polypeptide SEQ ID NO:1 was designed to consist of a unique sequence with low homology with other human proteins and to minimize cross reactivity of generated affinity reagents. Consequently, in embodiments of the methods of the above aspects, the quantifiable affinity ligand may be capable of selective interaction with an HMGCR protein having the sequence SEQ ID NO:1. A protein or polypeptide “having the sequence” refer to a protein or polypeptide “consisting of the sequence”.
In Examples below, an antibody binding to an HMGCR epitope having the sequence SEQ ID NO:4 is employed. This antibody is sometimes referred to herein as the commercial antibody. Consequently, in further embodiments of the methods of the above aspects, the quantifiable affinity ligand may be capable of selective interaction with an HMGCR protein having the sequence SEQ ID NO:4.
In some embodiments of the methods of the above aspects, an affinity ligand capable of selective interaction with the HMGCR protein is detectable and/or quantifiable. The detection and/or quantification of such an affinity ligand may be accomplished in any way known to the skilled person for detection and/or quantification of binding reagents in assays based on biological interactions. Thus, any affinity ligand, as described above, may be used quantitatively or qualitatively to detect the presence of the HMGCR protein. These “primary” affinity ligands may be labeled themselves with various markers or may in turn be detected by secondary, labeled affinity ligands to allow detection, visualization and/or quantification. This can be accomplished using any one or more of a multitude of labels, which can be conjugated to the affinity ligand capable of interaction with HMGCR protein or to any secondary affinity ligand, using any one or more of a multitude of techniques known to the skilled person, and not as such involving any undue experimentation.
Non-limiting examples of labels that can be conjugated to primary and/or secondary affinity ligands include fluorescent dyes or metals (e.g., fluorescein, rhodamine, phycoerythrin, fluorescamine), chromophoric dyes (e.g., rhodopsin), chemiluminescent compounds (e.g., luminal, imidazole) and bioluminescent proteins (e.g., luciferin, luciferase), haptens (e.g., biotin). A variety of other useful fluorescers and chromophores are described in Stryer L (1968) Science 162:526-533 and Brand L and Gohlke J R (1972) Annu. Rev. Biochem. 41:843-868. Affinity ligands can also be labeled with enzymes (e.g., horseradish peroxidase, alkaline phosphatase, beta-lactamase), radioisotopes (e.g., ³H, ¹⁴C, ³²P, ³⁵S or ¹²⁵I) and particles (e.g., gold). In the context of the present disclosure, “particles” refer to particles, such as metal particles, suitable for labeling of molecules. Further, the affinity ligands may also be labeled with fluorescent semiconductor nanocrystals (quantum dots). Quantum dots have superior quantum yield and are more photostable compared to organic fluorophores and are therefore more easily detected (Chan et al. (2002) Curr Opi Biotech. 13: 40-46). The different types of labels can be conjugated to an affinity ligand using various chemistries, e.g., the amine reaction or the thiol reaction. However, other reactive groups than amines and thiols can be used, e.g., aldehydes, carboxylic acids and glutamine.
The method aspects above may be put to use in any of several known formats and set-ups, of which a non-limiting selection is discussed below.
In a set-up based on histology, the detection, localization and/or quantification of a labeled affinity ligand bound to its HMGCR protein target may involve visualizing techniques, such as light microscopy or immunofluoresence microscopy. Other methods may involve the detection via flow cytometry or luminometry.
A biological sample, such as a tumor tissue sample (biopsy), for example from breast tissue, may be removed from the subject for detection and/or quantification of HMGCR protein. Alternatively, the biological sample, such as the biopsy, may be an earlier obtained sample. If using an earlier obtained sample, no steps of any one of the embodiments of the methods of the above aspects are practiced on the human or animal body. The affinity ligand may be applied to the biological sample for detection and/or quantification of the HMGCR marker protein. This procedure enables not only detection of HMGCR protein, but may in addition show the distribution and relative level of expression thereof.
The method of visualization of labels on the affinity ligand may include, but is not restricted to, fluorometric, luminometric and/or enzymatic techniques. Fluorescence is detected and/or quantified by exposing fluorescent labels to light of a specific wavelength and thereafter detecting and/or quantifying the emitted light in a specific wavelength region. The presence of a luminescently tagged affinity ligand may be detected and/or quantified by luminescence developed during a chemical reaction. Detection of an enzymatic reaction is due to a color shift in the sample arising from chemical reaction. Those of skill in the art are aware that a variety of different protocols can be modified in order for proper detection and/or quantification.
In embodiments of the methods of the above aspects, a biological sample may be immobilized onto a solid phase support or carrier, such as nitrocellulose or any other solid support matrix capable of immobilizing any HMGCR protein present in the biological sample applied to it. Some well-known solid state support materials useful in the present invention include glass, carbohydrate (e.g., Sepharose), nylon, plastic, wool, polystyrene, polyethene, polypropylene, dextran, amylase, films, resins, cellulose, polyacrylamide, agarose, alumina, gabbros and magnetite. After immobilization of the biological sample, primary affinity ligand specific to HMGCR protein may be applied, e.g., as described in Examples, Sections 2 and/or 3, of the present disclosure. If the primary affinity ligand is not labeled in itself, the supporting matrix may be washed with one or more appropriate buffers known in the art, followed by exposure to a secondary labeled affinity ligand and washed once again with buffers to remove unbound affinity ligands. Thereafter, selective affinity ligands may be detected and/or quantified with conventional methods. The binding properties for an affinity ligand may vary from one solid state support to the other, but those skilled in the art should be able to determine operative and optimal assay conditions for each determination by routine experimentation.
Consequently, in embodiments of the methods of the above aspects, the quantifiable affinity ligand of b1) may be detected using a secondary affinity ligand capable of recognizing the quantifiable affinity ligand. The quantification of b3) may thus be carried out by means of a secondary affinity ligand with affinity for the quantifiable affinity ligand. As an example, the secondary affinity ligand may be an antibody or a fragment or a derivative thereof.
As an example, one available method for detection and/or quantification of the HMGCR protein is by linking the affinity ligand to an enzyme that can then later be detected and/or quantified in an enzyme immunoassay (such as an EIA or ELISA). Such techniques are well established, and their realization does not present any undue difficulties to the skilled person. In such methods, the biological sample is brought into contact with a solid material or with a solid material conjugated to an affinity ligand against the HMGCR protein, which is then detected and/or quantified with an enzymatically labeled secondary affinity ligand. Following this, an appropriate substrate is brought to react in appropriate buffers with the enzymatic label to produce a chemical moiety, which for example is detected and/or quantified using a spectrophotometer, fluorometer, luminometer or by visual means.
As stated above, primary and any secondary affinity ligands can be labeled with radioisotopes to enable detection and/or quantification. Non-limiting examples of appropriate radiolabels in the present disclosure are ³H, ¹⁴C, ³²P, ³⁵S or ¹²⁵I. The specific activity of the labeled affinity ligand is dependent upon the half-life of the radiolabel, isotopic purity, and how the label has been incorporated into the affinity ligand. Affinity ligands are preferably labeled using well-known techniques (Wensel T G and Meares C F (1983) in: Radioimmunoimaging and Radioimmunotherapy (Burchiel S W and Rhodes B A eds.) Elsevier, New York, pp 185-196). A thus radiolabeled affinity ligand can be used to visualize HMGCR protein by detection of radioactivity in vivo or in vitro. Radionuclear scanning with e.g., gamma camera, magnetic resonance spectroscopy or emission tomography function for detection in vivo and in vitro, while gamma/beta counters, scintillation counters and radiographies are also used in vitro.
As a further aspect of the present disclosure, there is provided a kit for carrying out a method according to an embodiment of the above aspects, which comprises:
a) a quantifiable affinity ligand capable of selective interaction with an HMGCR protein; and
b) reagents necessary for quantifying the amount of the affinity ligand.
Various components of the kit according to the kit aspect may be selected and specified as described above in connection with the method aspects of the present disclosure.
As a non-limiting example, the affinity ligand is capable of selective interaction with an HMGCR protein having an amino acid sequence which comprises a sequence selected from:

- i) SEQ ID NO:1; and
- ii) a sequence which is at least 85% identical to SEQ ID NO:1.

In some embodiments, sequence ii) above is at least 90% identical, at least 91% identical, at least 92% identical, at least 93% identical, at least 94% identical, at least 95% identical, at least 96% identical, at least 97% identical, at least 98% identical or at least 99% identical to SEQ ID NO:1.
As another non-limiting example, the affinity ligand is capable of selective interaction with an HMGCR protein having an amino acid sequence which comprises a sequence selected from:

- i) SEQ ID NO:2; and
- ii) a sequence which is at least 85% identical to SEQ ID NO:2.

In some embodiments, sequence ii) above is at least 90% identical, at least 91% identical, at least 92% identical, at least 93% identical, at least 94% identical, at least 95% identical, at least 96% identical, at least 97% identical, at least 98% identical or at least 99% identical to SEQ ID NO:2.
Thus, the kit according to the present disclosure comprises an affinity ligand against an HMGCR protein, as well as other means that help to quantify the specific and/or selective affinity ligand after it has bound specifically and/or selectively to the HMGCR protein. For example, the kit may contain a secondary affinity ligand for detecting and/or quantifying a complex formed by the HMGCR protein and the affinity ligand capable of selective interaction with the HMGCR protein. The kit may also contain various auxiliary substances other than affinity ligands, to enable the kit to be used easily and efficiently. Examples of auxiliary substances include solvents for dissolving or reconstituting lyophilized protein components of the kit, wash buffers, substrates for measuring enzyme activity in cases where an enzyme is used as a label, target retrieval solution to enhance the accessibility to antigens in cases where paraffin or formalin-fixed tissue samples are used, and substances such as reaction arresters, e.g., endogenous enzyme block solution to decrease the background staining and/or counterstaining solution to increase staining contrast, that are commonly used in immunoassay reagent kits.
Thus, in embodiments of the kit aspect, the quantifiable affinity ligand is selected from the group consisting of antibodies, fragments thereof and derivatives thereof. As an example, such quantifiable affinity ligand may be obtainable by a process comprising a step of immunizing an animal, such as a rabbit, with a protein whose amino acid sequence comprises the sequence SEQ ID NO:1 or a sequence which is at least 85% identical to SEQ ID NO:1, preferably the sequence SEQ ID NO:1, and more preferably, the protein of the immunization has the sequence SEQ ID NO:1. Such immunization may for example be performed as described in Examples, Section 2, below. Also, the antibodies, fragments thereof and derivatives may for example be isolated and/or mono-specific.
Alternatively, the quantifiable affinity ligand is a protein ligand derived from a scaffold selected from the group consisting of staphylococcal protein A and domains thereof, lipocalins, ankyrin repeat domains, cellulose binding domains, y crystallines, green fluorescent protein, human cytotoxic T lymphocyte-associated antigen 4, protease inhibitors, PDZ domains, peptide aptamers, staphylococcal nuclease, tendamistats, fibronectin type III domain and zinc fingers. As a further alternative, the quantifiable affinity ligand is an oligonucleotide molecule.
The inventors have designed the sequence SEQ ID NO:1 to be a particularly suitable antigen for being recognized by the affinity ligands of the present disclosure. Consequently, in embodiments of the kit aspect, the quantifiable affinity ligand may be capable of selective interaction with an HMGCR protein having the sequence SEQ ID NO:1.
Further, in Examples below, an antibody binding to an HMGCR epitope having the sequence SEQ ID NO:4 is employed (the commercial antibody). This antibody resulted in strong staining. Consequently, in further embodiments of the kit aspect, the quantifiable affinity ligand may be capable of selective interaction with an HMGCR protein having the sequence SEQ ID NO:4.
Further, in embodiments of the kit aspect, the detectable affinity ligand may comprise a label selected from the group consisting of fluorescent dyes and metals, chromophoric dyes, chemiluminescent compounds and bioluminescent proteins, enzymes, radioisotopes, particles and quantum dots. Alternatively, the reagents necessary for quantifying the amount of the affinity ligand comprise a secondary affinity ligand capable of recognizing the quantifiable affinity ligand. As an example, the secondary affinity ligand capable of recognizing the quantifiable affinity ligand comprises a label selected from the group consisting of fluorescent dyes or metals, chromophoric dyes, chemiluminescent compounds and bioluminescent proteins, enzymes, radioisotopes, particles and quantum dots.
The kit according to the kit aspect may also advantageously comprise a reference sample for provision of, or yielding, the reference value to be used for comparison with the sample value. Preferably, the reference sample comprises a predetermined amount of HMGCR protein. Such a reference sample may for example be constituted by a tissue sample having the predetermined amount of HMGCR protein. The tissue reference sample may then be used by the person of skill in the art in the determination of the HMGCR expression status in the sample being studied, by manual, such as ocular, or automated comparison of expression levels in the reference tissue sample and the subject sample. As another example, the reference sample may comprise cell lines, such as cancer cell lines, expressing a predetermined, or controlled, amount of HMGCR protein. The person skilled in the art understands how to provide such cell lines, for example guided by the disclosure of Rhodes et al. (2006) The biomedical scientist, p 515-520. As an example, the cell lines may be formalin fixed. Also, such formalin fixed cell lines may be paraffin embedded.
The wording “reference sample for provision of the reference value” is to be interpreted broadly in the context of the present disclosure. The reference sample may comprise an amount of HMGCR protein actually corresponding to the reference value, but it may also comprise an amount of HMGCR protein corresponding to a value being higher than the reference value. In the latter case, the “high” value may be used by a person performing the method as an upper reference (positive reference) for assessing, e.g., the appearance of, a reference value which is lower than the “high” value. The person skilled in the art of immunohistochemistry understands how to do such an assessment. Further, as an alternative or a complementing example, the skilled person may use another reference sample comprising a low amount of HMGCR protein for provision of a “low” value in such an assessment, e.g., as a negative reference.
Consequently, in embodiments of the kit aspect, the reference sample may comprise an amount of HMGCR protein corresponding to the reference value. As an example, the reference sample may comprise an amount of HMGCR protein corresponding to a cytoplasmic fraction of 1% or higher, such as 5% or higher, such as 10% or higher, such as 15% or higher, such as 20% or higher, such as 25% or higher, such as 30% or higher, such as 35% or higher, such as 40% or higher, such as 45% or higher, such as 50% or higher, such as 55% or higher, such as 60% or higher, such as 65% or higher, such as 70% or higher, such as 75% or higher, such as 80% or higher, such as 90% or higher, such as 95% or higher. Alternatively, or as a complement, the reference sample may comprise an amount of HMGCR protein corresponding to an absent cytoplasmic intensity of HMGCR protein expression or higher, such as a weak cytoplasmic intensity of HMGCR protein expression or higher, such as moderate cytoplasmic intensity of HMGCR protein expression.
Further, the reference sample may comprise an amount of HMGCR protein corresponding a staining score of 0 or higher, such as 1 or higher, such as 2 or higher, such as 3.
The provision of cytoplasmic fraction values or cytoplasmic intensity values is discussed above in connection with the method aspects.
Further, in alternative or complementing embodiments of the kit aspect, the kit may comprise a reference sample comprising an amount of HMGCR protein corresponding to a value being higher than the reference value. In these embodiments, the reference sample may for example comprise an amount of HMGCR protein corresponding to a cytoplasmic fraction of 75% or higher and/or a strong cytoplasmic intensity of HMGCR expression.
In other alternative or complementing embodiments of the kit aspect, the kit may comprise a reference sample comprising an amount of HMGCR protein corresponding to a value being lower than the reference value, e.g., an absent cytoplasmic intensity and/or a cytoplasmic fraction of <2% HMGCR positive cells, such as 0% HMGCR positive cells.
Also, in alternative or complementing embodiments of the kit aspect, the kit may comprise: a reference sample comprising an amount of HMGCR protein corresponding to a predetermined reference value; a reference sample comprising an amount of HMGCR protein corresponding to a value being higher than a predetermined reference value; and/or a reference sample comprising an amount of HMGCR protein corresponding to a value being lower than a predetermined reference value.
Consequently, embodiments of the kit may comprise: a first reference sample comprising an amount of HMGCR protein being higher than a predetermined reference value; and a second reference sample comprising an amount of HMGCR protein being lower than the predetermined reference value.
In embodiments of the kit aspect, the reference sample may be a tissue sample, such as a tissue sample adapted to ocular or microscopic evaluation. As an example, the tissue reference sample may be fixated in paraffin or buffered formalin and/or histo-processed to μm-thin sections that are mounted on microscopic glass-slides. The tissue reference sample may be further adapted to staining with affinity ligands, such as antibodies, for an HMGCR protein.
Consequently, in embodiments of the kit aspect, the reference sample may be adapted to directly, or indirectly, provide any relevant reference value, such as any one of the reference values discussed above.
Accordingly, further embodiments of the reference sample of the kit aspect are discussed above in connection with the reference values and reference samples of the method aspects.
The kit may also include means for establishing the hormone receptor status and/or the HER2 status of a subject.
Consequently, in embodiments of the kit aspect, the kit may further comprise:
a′) a quantifiable affinity ligand capable of selective interaction with the estrogen receptor and b′) reagents necessary for quantifying the amount of such affinity ligand;
a″) a quantifiable affinity ligand capable of selective interaction with the progesterone receptor and b″) reagents necessary for quantifying the amount of such affinity ligand; and/or
a′″) a quantifiable affinity ligand capable of selective interaction with the HER2 receptor and b′″) reagents necessary for quantifying the amount of such affinity ligand.
The quantifiable affinity ligands of a′), a″) and a′″) are provided for the determination of the ER status, PR status and HER2 status, respectively.
The reagents of b), b′), b″) and b′″), respectively, may be the same of different.
Quantifiable affinity ligands appropriate for steps a′), a″) and a′″), respectively, are well known to the skilled person and commercially available.
Consequently, the kit may include means for provision of the status information necessary for drawing conclusions according to embodiments the above method aspect.
As a further aspect of the present disclosure, there is provided the use of an HMGCR protein as a prognostic marker. Also provided is the use of an HMGCR protein as a prognostic marker for cancer, such as breast cancer, e.g., a HR+ or HR− breast cancer.
In the context of the present disclosure, a “prognostic marker” refers to a species which presence is of value in an establishment of a prognosis.
The prognostic relevance of HMGCR protein found by the inventors entails the application of HMGCR protein in various assays or other laboratory set-ups. An HMGCR protein having the amino acid sequence of SEQ ID NO:1, which has been designed to have desired antigenic properties, may be particularly suitable for such purposes.
Accordingly, as a further use aspect of the present disclosure, there is provided the use of an HMGCR protein, or an antigenically active fragment thereof, for the production, selection or purification of a prognostic agent for establishing a prognosis for a patient having a breast cancer, e.g., a HR+ or a HR− breast cancer.
In the context of the present disclosure, an “antigenically active fragment” of an HMGCR protein is a fragment of sufficient size to be useful for the generation of an affinity ligand, e.g., an antibody, which should interact with an HMGCR protein comprising the fragment. Further, in the context of the present disclosure, a “prognostic agent” refers to an agent having at least one property being valuable in an establishment of a prognosis.
For example, the prognostic agent may be an affinity ligand capable of selective interaction with the HMGCR protein, or antigenically active fragment thereof. In turn, the affinity ligand may preferably be capable of selective interaction with a protein having the sequence SEQ ID NO:1. Also, the affinity ligand may preferably be an antibody or a fragment or a derivative thereof. For example, the antibody may be isolated and/or mono-specific. Guided by the teachings of the present disclosure, the person skilled in the art understands how to use HMGCR protein in the production, selection or purification of the prognostic agent. For example, such use may comprise affinity purification on a solid support onto which the HMGCR protein has been immobilized. The solid support may for example be arranged in a column. Further, the use may comprise selection of prognostic agents having specificity for the HMGCR protein using a soluble matrix in which the HMGCR protein has been immobilized. Such soluble matrix may for example be a dextran matrix, e.g., for use in a surface plasmon resonance instrument, such as Biacore™ instrument, and the selection may for example comprise monitoring the affinity for the immobilized HMGCR protein of a number of potential prognostic agents. Also, the use may comprise immunizing a mammal with the HMGCR protein in order to produce prognostic agents that are collected from the serum of the immunized mammal.
In embodiments of the above use aspects, the HMGCR protein may for example comprise a sequence selected from: i) SEQ ID NO:1; and ii) a sequence which is at least 85% identical to SEQ ID NO:1. In some embodiments, sequence ii) is at least 90% identical, at least 91% identical, at least 92% identical, at least 93% identical, at least 94% identical, at least 95% identical, at least 96% identical, at least 97% identical, at least 98% identical or at least 99% identical to SEQ ID NO:1.
Also, the HMGCR protein may for example comprise a sequence selected from: i) SEQ ID NO:2; and ii) a sequence which is at least 85% identical to SEQ ID NO:2. In some embodiments, sequence ii) is at least 90% identical, at least 91% identical, at least 92% identical, at least 93% identical, at least 94% identical, at least 95% identical, at least 96% identical, at least 97% identical, at least 98% identical or at least 99% identical to SEQ ID NO:2.
In embodiments of the above use aspects, the use may be use in vitro.
The inventors have found a previously unknown correlation between HMGCR protein and the prognosis of a subject suffering from a breast cancer. Further, the inventors have developed an affinity ligand with affinity for the HMGCR protein. However, the inventive scope is not limited to a single type of affinity ligand, and the inventive idea may be put into practice using any type of affinity ligand capable of selective interaction with an HMGCR protein.
Thus, as a further aspect of the present disclosure, there is provided an affinity ligand, capable of selective interaction with an HMGCR protein, for establishing a prognosis for a mammalian subject having a breast cancer, e.g., a HR+ or HR− breast cancer. As an alternative or complementary affinity ligand aspect, there is provided an affinity ligand, capable of selective interaction with an HMGCR protein, for evaluating the amount of HMGCR protein present in at least part of a sample obtained from a subject having a breast cancer.
The HMGCR polypeptide SEQ ID NO:1 was designed to consist of a unique sequence with low homology with other human proteins and to minimize cross reactivity of generated affinity reagents. Consequently, in one embodiment, the affinity ligand may be capable of selective interaction with an HMGCR protein having the amino acid sequence SEQ ID NO:1. In one embodiment, the affinity ligand is an antibody or a fragment or a derivative thereof. For example, the antibody or a fragment or a derivative thereof may be isolated and/or mono-specific. Further, the antibody or a fragment or a derivative thereof may be capable of selective interaction with an HMGCR protein having the amino acid sequence SEQ ID NO:4. SEQ ID NO:4 is part of SEQ ID NO:1.
SEQ ID NO:1 was also especially designed for immunizations, e.g., designed to lack transmembrane regions to ensure efficient expression in E. coli, and to lack any signal peptide, since those are cleaved off in the mature protein. Consequently, the antibody, or fragment or derivative thereof, may for example be one that is obtainable by a process comprising a step of immunizing an animal, such as a rabbit, with a protein whose amino acid sequence comprises the sequence SEQ ID NO:1 or a sequence which is at least 85% identical to SEQ ID NO:1, preferably the sequence SEQ ID NO:1. For example, the immunization process may comprise primary immunization with the protein in Freund's complete adjuvant. Also, the immunization process may further comprise boosting at least two times, in intervals of 2-6 weeks, with the protein in Freund's incomplete adjuvant. Processes for the production of antibodies or fragments or derivatives thereof against a given target are known in the art, and may be applied in connection with this aspect of the present disclosure. Any of those variants of the HMGCR protein (e.g., SEQ ID NO:2) or the antigenically active fragment thereof (e.g., SEQ ID NO:1) that are discussed above may, of course, be used in such a process for generating an antibody or a fragment or derivative thereof.
As a further aspect of the present disclosure, there is provided the use of the affinity ligand according to the above aspects as a prognostic agent. A preferred embodiment of this use is as a prognostic agent for the prognosis of a cancer, such as a breast cancer, e.g., a HR+ or HR− breast cancer. As a related aspect thereof, there is provided a use of the affinity ligand in the manufacture of a prognostic agent for the prognosis of a breast cancer. For example, the uses may be uses in vitro.
In the context of the present disclosure, “specific” or “selective” interaction of e.g., an affinity ligand with its target or antigen means that the interaction is such that a distinction between specific and non-specific, or between selective and non-selective, interaction becomes meaningful. The interaction between two proteins is sometimes measured by the dissociation constant. The dissociation constant describes the strength of binding (or affinity) between two molecules. Typically the dissociation constant between an antibody and its antigen is from 10⁻⁷to 10⁻¹¹M. However, high specificity does not necessarily require high affinity. Molecules with low affinity (in the molar range) for its counterpart have been shown to be as specific as molecules with much higher affinity. In the case of the present disclosure, a specific or selective interaction refers to the extent to which a particular method can be used to determine the presence and/or amount of a specific protein, the target protein or a fragment thereof, under given conditions in the presence of other proteins in a tissue sample or fluid sample of a naturally occurring or processed biological fluid. In other words, specificity or selectivity is the capacity to distinguish between related proteins. Specific and selective are sometimes used interchangeably in the present description. For example, the specificity or selectivity of an antibody may be determined as in Examples, section 2, below, wherein analysis is performed using a protein array set- up and a western blot, respectively. Specificity and selectivity determinations are also described in Nilsson P et al. (2005) Proteomics 5:4327-4337.
In the context of the present disclosure, a “mono-specific antibody” is one of a population of polyclonal antibodies which has been affinity purified on its own antigen, thereby separating such mono-specific antibodies from other antiserum proteins and non-specific antibodies. This affinity purification results in antibodies that bind selectively to its antigen. In the case of the present invention, the polyclonal antisera are purified by a two-step immunoaffinity based protocol to obtain mono-specific antibodies selective for the target protein. Antibodies directed against generic affinity tags of antigen fragments are removed in a primary depletion step, using the immobilized tag protein as the capturing agent. Following the first depletion step, the serum is loaded on a second affinity column with the antigen as capturing agent, in order to enrich for antibodies specific for the antigen (see also Nilsson P et al. (2005) Proteomics 5:4327-4337).

BRIEF DESCRIPTION OF THE FIGURES

In the figures, tissue cores were scored for level of cytoplasmic intensity (CI) of HMGCR protein expression, wherein: CI=3 represents strong CI; CI=2 represents moderate CI; CI=1 represents weak CI; and CI=0 represents absent CI. Consequently: CI>0 represents strong, moderate or weak CI; and CI<3 represents moderate, weak or absent CI.

FIG. 1 shows RFS based on IHC staining of 428 subjects diagnosed with ER+ breast carcinoma. Using the HPA antibody, CI was determined. The subjects were divided into four groups based on CI level (CI=0, CI=1, CI=2 and CI=3).

FIG. 2 shows RFS based on IHC staining of 428 subjects diagnosed with ER+ breast carcinoma. FIG. 2A shows the result from staining with the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 2B shows the result from staining with the HPA antibody, wherein the subjects were divided into two groups based on CI levels: CI=3 and CI<3, respectively.

FIG. 3 shows RFS based on IHC staining of 410 subjects diagnosed with PR+ breast carcinoma. FIG. 3A shows the result from staining with the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 3B shows the result from staining with the HPA antibody, wherein the subjects were divided into two groups based on CI levels: CI=3 and CI<3, respectively.

FIG. 4 shows RFS based on IHC staining of 454 subjects diagnosed with HR+ breast carcinoma. FIG. 4A shows the result from staining with the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 4B shows the result from staining with the HPA antibody, wherein the subjects were divided into two groups based on CI levels: CI=3 and CI<3, respectively.

FIG. 5 shows RFS based on IHC staining of 428 subjects diagnosed with ER+ breast carcinoma. FIG. 5A and 5B, respectively, show the result from staining with the commercial antibody. The subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 5A shows patients treated with endocrine therapy. FIG. 5B shows patients not treated with endocrine therapy.

FIG. 6 shows BOSS based on IHC staining of 428 subjects diagnosed with ER+ breast carcinoma. FIG. 6A shows the result from staining with the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 6B shows the result from staining with the HPA antibody, wherein the subjects were divided into two groups based on CI levels: CI=3 and CI<3, respectively.

FIG. 7 shows BOSS based IHC staining of 410 subjects diagnosed with PR+ breast carcinoma. FIG. 7A shows the result from staining with the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 7B shows the result from staining with the HPA antibody, wherein the subjects were divided into two groups based on CI levels: CI=3 and CI<3, respectively.

FIG. 8 shows BOSS based on IHC staining of 454 subjects diagnosed with HR+ breast carcinoma. FIG. 8A shows the result from staining with the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 8B shows the result from staining with the HPA antibody, wherein the subjects were divided into two groups based on CI levels: CI=3 and CI<3, respectively.

FIG. 9 shows OS based on IHC staining using the commercial antibody, wherein the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 9A, shows the result of 65 subjects diagnosed with ER− breast carcinoma. FIG. 9B, shows the result of 74 subjects diagnosed with PR− breast carcinoma. FIG. 9C, shows the result of 74 subjects diagnosed with HR− breast carcinoma.

FIG. 10 shows the results from IHC staining of 38 subjects diagnosed with HR− breast carcinoma. The commercial antibody was used and the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively. FIG. 10A shows RFS and FIG. 10B shows BOSS.

FIG. 11 shows RFS based on IHC staining of 28 subjects diagnosed with ER− and HER2− breast carcinoma. The commercial antibody was used and the subjects were divided into two groups based on CI levels: CI=0 and CI>0, respectively.

EXAMPLES

Generation of Mono-Specific Antibodies Against HMGCR and use thereof to Detect HMGCR in Normal and Cancerous Samples

1. Generation of Antigen

a) Materials and Methods

A suitable fragment of the target protein encoded by the EnsEMBL Gene ID ENSG00000113161 was selected using bioinformatic tools with the human genome sequence as template (Lindskog M et al. (2005) Biotechniques 38:723-727, EnsEMBL, www.ensembl.org). The fragment was used as template for the production of a 140 amino acid long fragment corresponding to amino acids 742-881 (SEQ ID NO:1) of the HMGCR protein (SEQ ID NO:2; EnsEMBL entry no. ENSP00000287936).
A fragment of the HMGCR gene transcript containing nucleotides 2274-2693 of EnsEMBL entry number ENST00000287936 (SEQ ID NO:3), was isolated by a Superscript™ One-Step RT-PCR amplification kit with Platinum® Taq (Invitrogen) and a human total RNA pool panel as template (Human Total RNA Panel IV, BD Biosciences Clontech). Flanking restriction sites NotI and AscI were introduced into the fragment through the PCR amplification primers, to allow in-frame cloning into the expression vector (forward primer: ATGGCTGGGAGCATAGGAG, reverse primer: TCCTTGGAGGTCTTGTAAATTG). Then, the downstream primer was biotinylated to allow solid-phase cloning as previously described, and the resulting biotinylated PCR product was immobilized onto Dynabeads M280 Streptavidin (Dynal Biotech) (Larsson M et al. (2000) J. Biotechnol. 80:143-157). The fragment was released from the solid support by NotI-AscI digestion (New England Biolabs), ligated into the pAff8c vector (Larsson M et al, supra) in frame with a dual affinity tag consisting of a hexahistidyl tag for immobilized metal ion chromatography (IMAC) purification and an immunopotentiating albumin binding protein (ABP) from streptococcal protein G (Sjölander A et al. (1997) J. Immunol. Methods 201:115-123; Ståhl S et al. (1999) Encyclopedia of Bioprocess Technology: Fermentation, Biocatalysis and Bioseparation (Fleckinger M C and Drew S W, eds) John Wiley and Sons Inc., New York, pp 49-63), and transformed into E. coli BL21(DE3) cells (Novagen). The sequences of the clones were verified by dye-terminator cycle sequencing of plasmid DNA amplified using TempliPhi DNA sequencing amplification kit (GE Healthcare, Uppsala, Sweden) according to the manufacturer's recommendations.
BL21(DE3) cells harboring the expression vector were inoculated in 100 ml 30 g/l tryptic soy broth (Merck KGaA) supplemented with 5 g/l yeast extract (Merck KGaA) and 50 mg/l kanamycin (Sigma-Aldrich) by addition of 1 ml of an overnight culture in the same culture medium. The cell culture was incubated in a 1 liter shake flask at 37° C. and 150 rpm until the optical density at 600 nm reached 0.5-1.5. Protein expression was then induced by addition of isopropyl-β-D-thiogalactopyranoside (Apollo Scientific) to a final concentration of 1 mM, and the incubation was continued overnight at 25° C. and 150 rpm. The cells were harvested by centrifugation at 2400 g, and the pellet was re-suspended in 5 ml lysis buffer (7 M guanidine hydrochloride, 47 mM Na₂HPO₄, 2.65 mM NaH₂PO₄, 10 mM Tris-HCl, 100 mM NaCl, 20 mM β-mercaptoethanol; pH=8.0) and incubated for 2 hours at 37° C. and 150 rpm. After centrifugation at 35300 g, the supernatant containing the denatured and solubilized protein was collected.
The His₆-tagged fusion protein was purified by immobilized metal ion affinity chromatography (IMAC) on columns with 1 ml Talon® metal (Co²⁺) affinity resin (BD Biosciences Clontech) using an automated protein purification procedure (Steen J et al. (2006) Protein Expr. Purif. 46:173-178) on an ASPEC XL4™ (Gilson). The resin was equilibrated with 20 ml denaturing washing buffer (6 M guanidine hydrochloride, 46.6 mM Na₂HPO₄, 3.4 mM NaH₂PO₄, 300 mM NaCl, pH 8.0-8.2). Clarified cell lysates were then added to the column. Thereafter, the resin was washed with a minimum of 31.5 ml washing buffer prior to elution in 2.5 ml elution buffer (6 M urea, 50 mM NaH₂PO₄, 100 mM NaCl, 30 mM acetic acid, 70 mM Na-acetate, pH 5.0). The eluted material was fractioned in three pools of 500, 700 and 1300 μl. The 700 μl fraction, containing the antigen, and the pooled 500 and 1300 μl fractions were stored for further use.
The antigen fraction was diluted to a final concentration of 1 M urea with phosphate buffered saline (PBS; 1.9 mM NaH₂PO₄, 8.1 mM Na₂HPO₄, 154 mM NaCl) followed by a concentration step to increase the protein concentration using Vivapore 10/20 ml concentrator with molecular weight cut off at 7500 Da (Vivascience AG). The protein concentration was determined using a bicinchoninic acid (BCA) micro assay protocol (Pierce) with a bovine serum albumin standard according to the manufacturer's recommendations. The protein quality was analyzed on a Bioanalyzer instrument using the Protein 50 or 200 assay (Agilent Technologies).

b) Results

A gene fragment corresponding to nucleotides 2274-2693 of the long transcript (SEQ ID NO:3) of the HMGCR gene and encoding a peptide (SEQ ID NO:1) consisting of amino acids 742 to 881 of the target protein HMGCR (SEQ ID NO:2) was successfully isolated by RT-PCR from a human RNA pool using primers specific for the protein fragment. The 140 amino acid fragment (SEQ ID NO:1) of the target protein (SEQ ID NO:2) was designed to lack transmembrane regions to ensure efficient expression in E. coli, and to lack any signal peptide, since those are cleaved off in the mature protein. In addition, the protein fragment was designed to consist of a unique sequence with low homology with other human proteins, to minimize cross reactivity of generated affinity reagents, and to be of a suitable size to allow the formation of conformational epitopes and still allow efficient cloning and expression in bacterial systems.
A clone encoding the correct amino acid sequence was identified, and, upon expression in E. coli, a single protein of the correct size was produced and subsequently purified using immobilized metal ion chromatography. After dilution of the eluted sample to a final concentration of 1 M urea and concentration of the sample to 1 ml, the concentration of the protein fragment was determined to be 11.7 mg/ml and was 84.2 pure according to purity analysis.

2. Generation of Antibodies

a) Materials and Methods

The purified HMGCR fragment as obtained above was used as antigen to immunize a rabbit in accordance with the national guidelines (Swedish permit no. A 84-02). The rabbit was immunized intramuscularly with 200 pg of antigen in Freund's complete adjuvant as the primary immunization, and boosted three times in four week intervals with 100 μg antigen in Freund's incomplete adjuvant.
Antiserum from the immunized animal was purified by a three-step immunoaffinity based protocol (Agaton C et al. (2004) J. Chromatogr. A 1043:33-40; Nilsson P et al. (2005) Proteomics 5:4327-4337). In the first step, 7 ml of total antiserum was buffered with 10x PBS to a final concentration of 1× PBS (1.9 mM NaH₂PO₄, 8.1 mM Na₂HPO₄, 154 mM NaCl), filtered using a 0.45 μm pore-size filter (Acrodisc®, Life Science) and applied to an affinity column containing 5 ml N-hydroxysuccinimide-activated Sepharose™ 4 Fast Flow (GE Healthcare) coupled to the dual affinity tag protein His₆-ABP (a hexahistidyl tag and an albumin binding protein tag) expressed from the pAff8c vector and purified in the same way as described above for the antigen protein fragment. In the second step, the flow-through, depleted of antibodies against the dual affinity tag His₆-ABP, was loaded at a flow rate of 0.5 ml/min on a 1 ml Hi-Trap NHS-activated HP column (GE Healthcare) coupled with the HMGCR protein fragment used as antigen for immunization (SEQ ID NO:1). The His₆-ABP protein and the protein fragment antigen were coupled to the NHS activated matrix as recommended by the manufacturer. Unbound material was washed away with 1× PBST (1× PBS, 0.1% Tween20, pH 7.25), and captured antibodies were eluted using a low pH glycine buffer (0.2 M glycine, 1 mM EGTA, pH 2.5). The eluted antibody fraction was collected automatically, and loaded onto two 5 ml HiTrap™ desalting columns (GE Healthcare) connected in series for efficient buffer exchange in the third step. The second and third purification steps were run on the ÄKTAxpress™ platform (GE Healthcare). The antigen selective (mono-specific) antibodies (msAbs) were eluted with PBS buffer, supplemented with glycerol and NaN₃to final concentrations of 40% and 0.02%, respectively, for long term storage at −20° C. (Nilsson P et al. (2005) Proteomics 5:4327-4337).
The specificity and selectivity of the affinity purified antibody fraction were analyzed by binding analysis against the antigen itself and against 383 other human protein fragments in a protein array set-up (Nilsson P et al. (2005) Proteomics 5:4327-4337). The protein fragments were diluted to 40 μg/ml in 0.1 M urea and 1× PBS (pH 7.4) and 50 μl of each were transferred to the wells of a 96-well spotting plate. The protein fragments were spotted in duplicate and immobilized onto epoxy slides (SuperEpoxy, TeleChem) using a pin-and-ring arrayer (Affymetrix 427). The slide was washed in 1× PBS (5 min) and the surface was then blocked (SuperBlock®, Pierce) for 30 minutes. An adhesive 16-well silicone mask (Schleicher & Schuell) was applied to the glass before the mono-specific antibodies were added (diluted 1:2000 in 1× PBST to appr. 50 ng/ml) and incubated on a shaker for 60 min. Affinity tag-specific IgY antibodies were co-incubated with the mono-specific antibodies in order to quantify the amount of protein in each spot. The slide was washed with 1× PBST and 1× PBS twice for 10 min each. Secondary antibodies (goat anti-rabbit antibody conjugated with Alexa 647 and goat anti-chicken antibody conjugated with Alexa 555, Molecular Probes) were diluted 1:60000 to 30 ng/ml in 1× PBST and incubated for 60 min. After the same washing procedure, as for the first incubation, the slide was spun dry and scanned (G2565BA array scanner, Agilent), thereafter images were quantified using image analysis software (GenePix 5.1, Axon Instruments).

b) Results

The quality of polyclonal antibody preparations has proven to be dependent on the degree of stringency in the antibody purifications, and it has previously been shown that depletion of antibodies directed against epitopes not originated from the target protein is necessary to avoid cross-reactivity to other proteins and background binding (Agaton C et al. (2004) J. Chromatogr. A 1043:33-40). Thus, a protein microarray analysis was performed to ensure that mono-specific polyclonal antibodies of high specificity had been generated by depletion of antibodies directed against the His₆-tag as well as of antibodies against the ABP-tag.
To quantify the amount of protein in each spot of the protein array, a two-color dye labeling system was used, with a combination of primary and secondary antibodies. Tag-specific IgY antibodies generated in hen were detected with a secondary goat anti-hen antibody labeled with Alexa 555 fluorescent dye. The specific binding of the rabbit msAb to its antigen on the array was detected with a fluorescently Alexa 647 labeled goat anti-rabbit antibody. The protein array analysis showed that the affinity purified mono-specific antibody against HMGCR is highly selective to the correct protein fragment and has a very low background to all other protein fragments analyzed on the array.

3. Tissue Profiling by Immunohistochemistry

a) Material and Methods

In total, 576 paraffin cores containing human tissues were analyzed using the mono-specific antibody sample obtained in Examples, section 2. All tissues used as donor blocks for tissue microarray (TMA) production were selected from the archives at the Department of Pathology, University Hospital, Uppsala, in agreement with approval from the local ethical committee. All tissue sections used for TMA analysis were examined to determine diagnosis and to select representative areas in donor blocks. Normal tissue was defined as microscopically normal (non-neoplastic) and was most often selected from specimens collected from the vicinity of surgically removed tumors. Cancer tissue was reviewed for diagnosis and classification. All tissues were formalin fixated, paraffin embedded, and sectioned for diagnostic purposes.
The TMA production was performed essentially as previously described (Kononen J et al. (1998) Nature Med. 4:844-847; Kallioniemi O P et al. (2001) Hum. Mol. Genet. 10:657-662). Briefly, a hole was made in the recipient TMA block and a cylindrical core tissue sample from the donor block was acquired and deposited in the recipient TMA block. This was repeated in an automated tissue arrayer from Beecher Instrument (ATA-27, Beecher Instruments, Sun Prairie, Calif., USA) until a complete TMA design was produced. TMA recipient blocks were baked at 42° C. for 2 h prior to sectioning.
The design of TMA:s was focused on obtaining samples from a large range of representative normal tissues, and on including representative cancer tissues. This has previously been described in detail in Kampf C et al. (2004) Clin. Proteomics 1:285-300. In brief, samples from 48 normal tissues and from 20 of the most common cancer types affecting humans were selected. In total, eight different designs of TMA blocks, each containing 72 cores of tissue with 1 mm diameter, were produced. Two of the TMA:s represented normal tissues, corresponding to 48 different normal tissues in triplicates from different individuals. The remaining 6 TMA:s represented cancer tissue from 20 different types of cancer. For 17 of the 20 cancer types, 12 individually different tumors were sampled, and for the remaining 3 cancer types, 4 individually different tumors were sampled, all in duplicates from the same tumor. The TMA blocks were sectioned with 4 μm thickness using a waterfall microtome (Leica), and placed onto SuperFrost® (Roche Applied Science) glass slides for IHC analysis.
Automated IHC was performed as previously described (Kampf C et al. (2004) Clin. Proteomics 1:285-300). In brief, the glass slides were incubated for 45 min in 60° C., de-paraffinized in xylene (2×15 min) and hydrated in graded alcohols. For antigen retrieval, slides were immersed in TRS (Target Retrieval Solution, pH 6.0, DakoCytomation) and boiled for 4 min at 125° C. in a Decloaking chamber® (Biocare Medical). Slides were placed in the Autostainer® (DakoCytomation) and endogenous peroxidase was initially blocked with H₂O₂(DakoCytomation). The slides were incubated for 30 min at room temperature with the primary antibody obtained as in Examples, Section 2, followed by incubation for 30 min at room temperature with goat anti-rabbit peroxidase conjugated Envision®. Between all steps, slides were rinsed in wash buffer (DakoCytomation). Finally, diaminobenzidine (DakoCytomation) was used as chromogen and Harris hematoxylin (Sigma-Aldrich) was used for counterstaining. The slides were mounted with Pertex® (Histolab).
All immunohistochemically stained sections from the eight different TMA:s were scanned using a ScanScope T2 automated slide-scanning systems (Aperio Technologies). In order to represent the total content of the eight TMA:s, 576 digital images were generated. Scanning was performed at 20 times magnification. Digital images were separated and extracted as individual tagged image file format (TIFF) files for storage of original data. In order to be able to handle the images in a web-based annotation system, the individual images were compressed from TIFF format into JPEG format. All images of immunohistochemically stained tissue were manually evaluated under the microscope and annotated by a certified pathologist or by specially educated personnel, subsequently verified by a pathologist.
Annotation of each different normal and cancer tissue was performed using a simplified scheme for classification of IHC outcome. Each tissue was examined for representativity and immunoreactivity. The different tissue specific cell types included in each normal tissue type were annotated. For each cancer, tumor cells and stroma were annotated. Basic annotation parameters included an evaluation of i) subcellular localization (nuclear and/or cytoplasmic/membranous), ii) staining intensity (SI) and iii) fraction of stained cells (FSC). Staining intensity was subjectively evaluated in accordance to standards used in clinical histo-pathological diagnostics and outcome was classified as: absent=no immunoreactivity, weak=faint immunoreactivity, moderate=medium immunoreactivity or strong=distinct and strong immunoreactivity. The fraction of stained cells was estimated and classified as <2%, 2-25%, >25-75% or >75 immunoreactive cells of the relevant cell population. Based on both the intensity and fraction of immunoreactive cells, a “staining score” was given for each tissue sample: 0=negative, 1=weak, 2=moderate and 3=strong. N.R. means that no representative tissues were present. In detail, the staining score was given according to the following criteria: 0 was given if SI=absent or weak and FSC≦25%; 1 was given if SI=weak and FSC>25% or if SI=moderate and FSC≦25%; 2 was given if SI=moderate and FSC>25% or if SI=strong and FSC≦25% and SI=moderate; and finally 3 was given if SI=strong and FSC>25%. See also table 1. The skilled artisan should recognize that this procedure is similar to a calculation of an Allred score, see e.g., Allred et al. (1998) Mod Pathol 11(2), 155.

TABLE 1

Staining score

Staining	Staining	Fraction of
score	intensity	stained cells

0	absent	<2%
0	absent	2-25%
0	absent	>25-75%
0	absent	>75%
0	weak	<2%
0	weak	2-25%
1	weak	>25-75%
1	weak	>75%
1	moderate	<2%
1	moderate	2-25%
2	moderate	>25-75%
2	moderate	>75%
2	strong	<2%
2	strong	2-25%
3	strong	>25-75%
3	strong	>75%

b) Results

The results from tissue profiling with the mono-specific antibody generated towards a recombinant protein fragment of the human target protein HMGCR obtained as in Examples, Section 2 showed a particular immunoreactivity in several normal tissues. Table 2 shows the HMGCR protein expression pattern in normal human tissues. Using IHC and TMA technology, 144 spots (1 mm in diameter) representing 48 different types of normal tissue were screened for expression of HMGCR. Immunoreactivity was observed mainly in cytoplasm of most tissues. Some cases showed additional nuclear or membranous positivity. Strongest staining was found in glandular epithelia. In a few cases no representative tissue (N.R.) were observed.

TABLE 2

Expression pattern of HMGCR in normal tissues

		Staining
Tissue type	Cell type	score

Adrenal gland	cortical cells	3
Appendix	glandular cells	3
	lymphoid tissue	1
Bone marrow	bone marrow poietic cells	3
Breast	glandular cells	2
Bronchus	respiratory epithelial cells	2
Cerebellum	cells in granular layer	2
	cells in molecular layer	3
	purkinje cells	3
Cerebral cortex	glial cells	0
	neuronal cells	3
Cervix, uterine	glandular cells	3
	squamous epithelial cells	1
Colon	glandular cells	3
Corpus, uterine 1	cells in endometrial stroma	1
	glandular cells	3
Corpus, uterine 2	cells in endometrial stroma	1
	glandular cells	3
Duodenum	glandular cells	3
Epididymis	glandular cells	2
Esophagus	squamous epithelial cells	2
Fallopian tube	glandular cells	2
Gall bladder	glandular cells	3
Heart muscle	myocytes		2
Hippocampus	glial cells	0
	neuronal cells	2
Kidney	cells in glomeruli	0
	cells in tubules	2
Lateral ventricle	glial cells	0
	neuronal cells	1
Liver	bile duct cells	1
	hepatocytes	1
Lung	alveolar cells	2
	macrophages	1
Lymph node	lymphoid cells outside reaction centra	2
	reaction center cells	2
Nasopharynx	respiratory epithelial cells	2
Oral mucosa	squamous epithelial cells	2
Ovary	follicle cells	N.R.
	ovarian stromal cells	1
Pancreas	exocrine glandular cells	3
	islet cells	1
Parathyroid gland	glandular cells	3
Placenta	decidual cells		1
	trophoblastic cells	2
Prostate	glandular cells	2
Rectum	glandular cells	3
Salivary gland	glandular cells	2
Seminal vesicle	glandular cells	2
Skeletal muscle	myocytes		2
Skin	adnexal cells	N.R.
	epidermal cells	2
Small intestine	glandular cells	3
Smooth muscle	smooth muscle cells	0
Soft tissue 1	mesenchymal cells	2
Soft tissue 2	mesenchymal cells	2
Spleen	cells in red pulp	2
	cells in white pulp	2
Stomach 1	glandular cells	3
Stomach 2	glandular cells	3
Testis	cells in seminiferus ducts	1
	leydig cells	3
Thyroid gland	glandular cells	2
Tonsil	lymphoid cells outside reaction centra	2
	reaction center cells	1
	squamous epithelial cells	2
Urinary bladder	urothelial cells		2
Vagina	squamous epithelial cells	1
Vulva/anal skin	squamous epithelial cells	2

HMGCR protein expression was further evaluated in tissue samples from various cancer types. Table 3 shows the level of HMGCR expression in 12 different breast carcinoma tissues samples. All of these samples showed representative tissue and ten showed positivity, i.e., a staining score of higher than zero. HMGCR expression was observed in cytoplasm.

TABLE 3

Expression pattern of HMGCR

Tissue sample

	1	2	3	4	5	6	7	8	9	10	11	12

Staining score	3	2	2	1	1	1	1	1	1	1	0	0

4. Breast Cancer TMA (Consecutive Cohort)

a) Material and Methods

Archival formalin-fixed paraffin-embedded tissue from 498 patients diagnosed with primary breast cancer between 1988 and 1992 was collected from the Department of Pathology, Malmö University Hospital, Sweden. The median age of patients was 65 (range 27-96) years and median follow-up time was 128 months (0-207). Information regarding the date of death was obtained from the regional cause-of-death registries for all patients. Ethical permission was obtained from the Local Ethics Committee of Lund University.
All 498 cases of breast cancer were histopathologically reevaluated on slides stained with hematoxylin and eosin. TMA:s were constructed by sampling 2×1.0 mm cores per case from areas representative of invasive cancer, using a manual arraying device (ATA-27, Beecher Inc, Sun Prairie, Wis., USA). The TMA:s were prepared and automated IHC was performed as described in section 3 above, using the HMGCR antibody prepared as described in section 2 above. This antibody is sometimes referred to herein as the “HPA antibody” or the “HPA anti-HMGCR antibody”.
In addition, a commercial poly-clonal anti-HMGCR antibody (Catalog # 07-457, Upstate/Millipore, Mass., USA, was tested on the TMA:s to further verify the inventive concept. This antibody is sometimes referred to herein as the “commercial antibody” or the “commercial anti-HMGCR antibody”.
For statistical analyses, the cytoplasmic intensity (CI) level was evaluated, in line with what is described in Examples, Section 3 above. The level of staining intensity of the cytoplasm was subjectively evaluated in accordance to standards used in clinical histo-pathological diagnostics and outcome was classified as: absent=no immunoreactivity, weak=faint immunoreactivity, moderate=medium immunoreactivity or strong=distinct and strong immunoreactivity. Based on the survival trends for all different strata, a dichotomized variable were constructed for further statistical analyses. For analysis using the HPA antibody, a strong cytoplasmic intensity (CI=3) was defined as a high HMGCR expression and absent, weak or moderate cytoplasmic intensity (CI<3) as a low HMGCR expression. The commercial HMGCR antibody stained stronger and a weak, moderate or strong cytoplasmic intensity (CI>0) was defined as a high HMGCR expression and an absent cytoplasmic intensity (CI=0) as a low HMGCR expression as.
This classification of samples was used for RFS and BCSS analysis according to the Kaplan-Meier estimator, and the log-rank test was used to compare survival in different strata and p-values of <0.05% were considered significant. All calculations were made with the statistical package SPSS 16.0 (SPSS Inc. Illinois, USA).

b) Results

For the present study, IHC analysis of HMGCR protein expression could be performed on totally 444 tumors using the commercial antibody and 473 tumors using the HPA antibody. The remaining 54/25 (commercial antibody/HPA antibody) cores either did not contain invasive cancer or were lost during histoprocessing. Of the analyzed tumors, 386/409 were ER+, 375/393 were PR+ and 409/433 were HR+, i.e., ER+ and/or PR+. IHC staining of ER and PR had been performed previously and in line with current clinical praxis, a cut-off at 10% positive nuclei was used to define ER+ and PR+, respectively.
FIG. 1 shows recurrence-free survival curves for subjects having ER+ cancers. The subjects are divided into four different categories based on the level of CI, and generally, more favorable prognoses are observed for the subjects having higher levels. The figure may e.g., be interpreted as the probability of 70-month recurrence-free increases with CI level, i.e., if two CI values are compared, the higher value always correspond to a higher probability at 70 months.
Survival analysis revealed a significantly better RFS for ER+, PR+ and HR+ subjects with HMGCR protein high tumors, these results were observed using both the commercial and the HPA antibody (FIG. 2-4). Further, the positive effect of a high HMGCR protein expression, i.e., a better RFS, was evident in both endocrine treated and untreated subjects (FIG. 5). Consequently, the prognostic relevance of the level of HMGCR protein expression appears to be independent of endocrine treatment status.
The better outcome of HMGCR protein high subjects was further observed when performing BCSS analyses, in most of the cases the survival was significantly better (FIGS. 6-8). These results were not as significant as RFS but still a clear trend were observed for all groups (ER+, PR+ and HR+) of patients. Taken together, a significantly better RFS and BCSS were observed for the HMGCR protein high category, than for the HMGCR protein low category among HR+subjects. These results indicate a prognostic value for a high HMGCR protein level in hormone positive patients (ER+, PR+ or HR+).
While the beneficial effect of a high HMGCR protein expression was highly significant in the ER+, PR+ and HR+ tumors, an inverse association to outcome was observed in the ER−, PR− and HR− subgroups. Due to the relatively small number of tumors in these subgroups, this association only reached statistical significance or close to significance for ER−, PR− and HR− patients when analyzing OS (FIG. 9A-C). In other words, among subjects having ER−, PR− or HR− cancers, a worse prognosis, e.g., a shorter OS, was observed for HMGCR high subjects than for HMGCR low subjects. Further, a similar trend was observed for hormone negative tumors when looking at RFS and BOSS. For example, RFS and BOSS for HR− are viewed in FIG. 10. Taken together, a worse prognosis is observed for hormone receptor negative subjects in the HMGCR protein high category than in the HMGCR protein low category. These results indicate a prognostic value for a high HMGCR protein level in subjects diagnosed as negative for a hormone receptor.
Among subjects having ER− and HER2− breast cancers, the prognosis for the subgroup having high HMGCR levels was shown to particularly poor (FIG. 11).

Establishment of a Prognosis for a Breast Cancer Patient

5. A Non-Limiting Example

A breast cancer patient can present symptoms or signs such as a palpable lump/tumor, secretion from the mamilla or skin deformities. A proportion of breast cancers, generally without symptoms, are also detected by screening mammography.
Following the establishment of a breast cancer diagnosis in a patient, a tumor tissue sample is obtained. The tumor tissue sample may be obtained from a biopsy (a removal of a selected physical piece of tissue from the suspected tumor) performed earlier during the diagnosis of the cancer or from a specimen from an earlier surgical removal of the tumor. Further, for the provision of a “negative reference”, a sample is taken from archival material comprising tissue having low, or essentially lacking, HMGCR protein expression. Such archival tissue may for example be breast cancer tissue having a pre-established low HMGCR protein expression level or an appropriate tissue having a staining score of 0 in Table 2, such as smooth muscle cells. Further, for the provision of a “positive reference”, a sample is taken from archival material comprising tissue having high HMGCR protein expression, such as breast cancer tissue having a pre-established high HMGCR protein expression level or an appropriate tissue having a staining score of 3 in Table 2, such as glandular cells from the colon.
The sample material is fixated in buffered formalin and histo-processed in order to obtain thin sections (4 μm) of the of the sample material.
Immunohistochemistry is performed as described in Examples, Section 3. One or more sample sections from each sample are mounted on glass slides that are incubated for 45 min in 60° C., de-paraffinized in xylene (2×15 min) and hydrated in graded alcohols. For antigen retrieval, slides are immersed in TRS (Target Retrieval Solution, pH 6.0, DakoCytomation) and boiled for 4 min at 125° C. in a Decloaking chamber® (Biocare Medical). Slides are placed in the Autostainer® (DakoCytomation) and endogenous peroxidase is initially blocked with H₂O₂(DakoCytomation). The reason for mounting multiple sample sections may be to increase the accuracy of the results.
A primary HMGCR protein specific antibody is added to the slides and incubated for 30 min in room temperature. The primary HMGCR protein specific antibody may be a commercial poly-clonal anti-HMGCR antibody (Catalog # 07-457, Upstate/ Millipore, Mass., USA) or be obtained as in Examples, Section 2. Incubation of the primary HMGCR specific antibody is followed by a 30 min incubation in room temperature with a labeled secondary antibody; e.g., goat-anti-rabbit peroxidase conjugated Envision®. To detect the secondary antibody, diaminobenzidine (DakoCytomation) is used as chromogen, contrasted with a Harris hematoxylin (Sigma-Aldrich) counterstaining. Between all steps, slides are rinsed in wash buffer (DakoCytomation). The slides are then mounted with Pertex® (Histolab) mounting media.
As a tool to validate the staining procedure, two control cell-lines may be used; e.g., one slide with cells expressing HMGCR protein (positive cell line) and one slide having cells with indistinct weak or no HMGCR protein expression (negative cell line). The skilled artisan understands how to provide such cell lines, for example guided by the disclosure of Rhodes et al. (2006) The biomedical scientist, p 515-520. The control cell-line slides may be simultaneously stained in the same procedure as the breast cancer slides, i.e., incubated with the same primary and secondary antibodies.
To obtain digital images, the breast cancer tumor slides, the staining reference slides, and optionally, the slides with control cell-lines, may for example be scanned in a light microscope using a ScanScope T2 automated slide scanning system (Aperio Technologies) at ×20 magnification. However, this scanning step is not necessary, but may make the procedure easier if, for example, the preparation and staining of the slides and the evaluation of the cytoplasmic intensity (see below) are performed at different locations or by different persons.
If control cell-lines are used, these are inspected to validate the staining procedure. If the cell-lines display staining results outside acceptable criteria, e.g., staining artifacts recognized by the skilled artisan, the staining of the biopsy samples is considered invalid and the whole staining procedure is repeated with new slides. If the positive and negative cell-lines display strong staining intensity and indistinct weak or no staining intensity, respectively, the staining is considered as valid.
The stained sample slide(s) from the tumor tissue biopsy is/are evaluated manually by visual, and the immunoreactivity of the breast cancer slide(s) is/are graded as described in Examples, Section 4.
That is, the cytoplasmic intensity (CI) is subjectively evaluated in accordance to standards used in clinical histo-pathological diagnostics. The CI is classified as: absent=no immunoreactivity, weak=faint immunoreactivity, moderate=medium immunoreactivity or strong=distinct and strong immunoreactivity. The person performing the evaluation and grading is aided by visual inspection of the stained reference slides: the reference slide having high HMGCR protein provides a “positive reference”, and the reference slide having a low amount of, or no, HMGCR protein expression provides a “negative reference”. The positive and negative references help the person performing the evaluation assessing the appearance of the four different levels of CI.
In addition to the determination of HMGCR protein expression level of the stained sample slides from the tumor tissue, the hormone receptor (HR) status of the patient is obtained. That is, the estrogen receptor (ER) and/or progesterone receptor (PR) status of the subject is obtained, and if the patient is ER positive (ER+) and/or PR positive (PR+) it is concluded that the patient is HR positive (HR+), and if the subject is ER negative (ER−) and PR negative (PR−) it is concluded that the patient is HR negative (HR−). The ER and/or PR status may have been determined before or after the determination of the HMGCR protein expression level. For example, the HR status may have been previously determined during the diagnosis of the cancer and/or determined from a specimen from an earlier surgical removal of the breast cancer tumor. The person skilled in the art knows how to determine whether a breast cancer is positive or negative for ER and/or PR, respectively. For example, the commercially available kit ER/PR pharmDX (DakoCytomation) may be used for the determination. As another example, the method disclosed by Allred et al. (Allred et al. (1998) Mod Pathol 11(2), 155) may be used to obtain a a total score (Allred score), and, for both ER and PR, an Allred score of higher than two is considered positive and an Allred score of two or lower is considered negative. Alternatively, when classifying a sample as being positive or negative for ER or PR, a 10% positive cells as the cutoff may be used, which is a recognized limit within the art.
The HMGCR protein expression level of the sample slide(s) from the tumor tissue are then compared to a reference value. For example, if the commercial poly-clonal anti-HMGCR antibody (Catalog #07-457, Upstate/Millipore, Mass., USA) was used for the determination of the HMGCR protein expression level, an absent CI may constitute the reference value (C-Ref). Alternatively, if the primary HMGCR protein specific antibody obtained as in Examples, Section 2, was used for the determination of HMGCR protein expression level, a moderate CI may constitute the reference value (H-Ref).
Subsequently, if the HMGCR protein expression level was higher than the reference value (C-Ref or H-Ref), a conclusion is drawn regarding a prognosis for the patient.
For example, if the patient has a positive HR status and C-ref, which may be associated with a probability of five-year recurrence-free survival of 67% (FIG. 4A, lower curve), is used, the conclusion may be that the prognosis for the patient is a probability of five-year recurrence-free survival of higher than 67%. Alternatively, if the patient has a positive HR status and H-ref, which may be associated with a probability of five-year recurrence-free survival of 74% (FIG. 4B, lower curve), is used, the conclusion may be that the prognosis for the patient is a probability of five-year recurrence-free survival of higher than 74%.
As another example, if the patient has a negative HR status and C-ref, which may be associated with a probability of five-year recurrence-free survival of 67% (FIG. 10A, upper curve), is used, the conclusion may be that the prognosis for the patient is a probability of five-year recurrence-free survival of lower than 67%. Alternatively, if the patient has a negative HR status and H-ref, which may be associated with a probability of five-year recurrence-free survival of 60% (data not shown), is used, the conclusion may be that the prognosis for the patient is a probability of five-year recurrence-free survival of lower than 60%.
The prognosis may then form a basis for further decisions relating to the treatment of the patient.
All cited material, including but not limited to publications, DNA or protein data entries, and patents, referred to in this application are herein incorporated by reference.
The invention being thus described, it will be obvious that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the present invention, and all such modifications as would be obvious to one skilled in the art are intended to be included within the scope of the following claims.

Claims

1.-149. (canceled)

150. Method for determining a prognosis for a mammalian subject having a hormone receptor positive or a hormone receptor negative breast cancer, comprising the steps of:

a) providing a sample earlier obtained from the subject;

b) evaluating the amount of HMGCR protein present in at least part of said sample, and determining a sample value corresponding to said evaluated amount;

c) comparing the sample value obtained in step b) with a reference value;

d) and, if said sample value is higher than said reference value:

d1) concluding that the prognosis for said subject is better than a first reference prognosis associated with the reference value if the subject has a hormone receptor positive breast cancer; or

d2) concluding that the prognosis for said subject is worse than a second reference prognosis associated with the reference value if the subject has a hormone receptor negative breast cancer.

151. Method according to claim 150, further comprising the steps of:

e) and, if said sample value is lower than, or equal to, said reference value,

e1) concluding that the prognosis for said subject is worse than, or equal to, the first reference prognosis associated with the reference value if the subject has a hormone receptor positive breast cancer; or

e2) concluding that the prognosis for said subject is better than, or equal to, the second reference prognosis associated with the reference value if the subject has a hormone receptor negative breast cancer.

152. Method for determining whether a statin treatment is suitable for a mammalian subject having a hormone receptor negative breast cancer, comprising the steps of:

a) providing a sample earlier obtained from the subject;

c) comparing the sample value obtained in step b) with a reference value; and, if said s ample value is equal to or lower than said reference value,

d) concluding that statin treatment is suitable for the subject.

153. Method according to claim 150, wherein hormone receptor positive is estrogen receptor positive and/or progesterone receptor positive, and hormone receptor negative is estrogen receptor negative and progesterone receptor negative.

154. Method according to claim 150, wherein hormone receptor positive is estrogen receptor positive, and hormone receptor negative is estrogen receptor negative.

155. Method according to claim 150, wherein hormone receptor positive is progesterone receptor positive, and hormone receptor negative is progesterone receptor negative.

156. Method according to claim 150, wherein said hormone receptor positive breast cancer is estrogen receptor positive.

157. Method according to claim 150, wherein hormone receptor positive breast cancer is progesterone receptor positive.

158. Method according to claim 150, wherein said hormone receptor negative cancer is further HER2 negative.

159. Method according to claim 150, wherein said sample is a tissue sample.

160. Method according to claim 159, wherein said tissue sample is a breast tumor sample.

161. Method according to claim 150, wherein step b) comprises:

b1) applying to the sample a quantifiable affinity ligand capable of selective interaction with the HMGCR protein to be evaluated, said application being performed under conditions that enable binding of the affinity ligand to any HMGCR protein present in the sample;

b2) removing non-bound affinity ligand; and

b3) quantifying any affinity ligand remaining in association with the sample to evaluate said amount.

162. Method according to claim 161, wherein the quantifiable affinity ligand is selected from the group consisting of antibodies, fragments thereof and derivatives thereof.

163. Kit for carrying out the method according to claim 150, which comprises

a) a quantifiable affinity ligand capable of selective interaction with an HMGCR protein; and

b) reagents necessary for quantifying the amount of the affinity ligand.

164. Kit according to claim 163, further comprising at least one reference sample for provision of a reference value.

165. Kit according to claim 163, further comprising

a′) a quantifiable affinity ligand capable of selective interaction with an estrogen receptor; and

b′) reagents necessary for quantifying the amount of the affinity ligand.

166. Kit according to claim 163, further comprising

a″) a quantifiable affinity ligand capable of selective interaction with a progesterone receptor; and

b″) reagents necessary for quantifying the amount of the affinity ligand.

167. Kit according to claim 163, further comprising

a′″) a quantifiable affinity ligand capable of selective interaction with a HER2 receptor; and

b′″) reagents necessary for quantifying the amount of the affinity ligand.

168. Method for establishing a prognosis for a mammalian subject having a breast cancer, comprising the steps of:

i) obtaining a hormone receptor status of the subject;

ii) obtaining an HMGCR protein value of the subject; and

iii) correlating the hormone receptor status and the HMGCR protein value of the subject to a prognosis for the subject.

169. Method of treatment of a mammalian subject having a hormone receptor positive breast cancer, comprising the steps of:

a) providing a sample from the subject;

c) comparing the sample value obtained in step b) with a reference value;

d) and if said sample value is equal to, or lower than, said reference value, treating said subject with an adjuvant breast cancer treatment.

170. Method of treatment of a mammalian subject having a hormone receptor negative breast cancer, comprising the steps of:

a) providing a sample from the subject;

c) comparing the sample value obtained in step b) with a reference value;

d) and if said sample value is higher than said reference value, treating said subject with an adjuvant breast cancer treatment.

171. Method according to claim 170, wherein said adjuvant breast cancer treatment comprises statin therapy.

172. Method according to claim 171, wherein said breast cancer is HER2 negative.

173. Method according to claim 169, wherein the adjuvant breast cancer treatment is a radiotherapy, a chemotherapy, an endocrine treatment or a combination thereof, such as a sequential chemo-endocrine therapy.