WO2020023713A1 - Tri-cytokine txm compositions and methods - Google Patents

Abstract

Contemplated compositions and methods comprise chimeric molecule complexes that advantageously provide various activating signals to immune competent cells and especially to T cells to differentiate/proliferate to T_SCM cells. Where desired, further proliferation and/or activation can be achieved using agonists for TNF superfamily receptors.

Description

TRI-CYTOKINE TxM COMPOSITIONS AND METHODS

[0001] This application claims priority to our copending US Provisional Application with the serial number 62/703,804, which was filed 7/26/2018.

Field of the Invention

[0002] The field of the invention is immune stimulation for cancer therapy, particularly as it relates to cancer therapy with multifunctional chimeric molecule complexes and TNF family receptor ligands to stimulate TSCM (stem cell memory T) cells.

Background of the Invention

[0003] The background description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.

[0004] All publications and patent applications herein are incorporated by reference to the same extent as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference, except that where a definition or use of a term in an incorporated reference is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies and the definition of that term in the reference does not apply.

[0005] Human patients that had received an allogeneic hematopoietic stem cell transplant (HSCT) together with relatively low doses of streptamer-enriched human CMV-specific CD8⁺ T cells produced a robust pathogen- specific T-cell expansion. Notably, antiviral T cells, including CD62L⁺ TSCM are capable of massive proliferation and at least potential indefinite persistence (see e.g., Blood 2014 124:476-477).

[0006] Unfortunately, TSCM are relatively rare. While it is known that T cell proliferation for certain T cells can be stimulated with various cytokines, the quantities required are often undesirably high (see e.g., J Exp Med 2005;202(7):907-9l2).

[0007] More recently, it was reported that TSCM were generated from precursors using IL7 and IL15 (see e.g., Blood 2013 121:573-584). Interestingly, CD19-CAR modified TSCM could also be produced by incubation of specific T cell precursor cells with IL7 and IL21 (see e.g., Blood 2016 128:519-528). While at least conceptually promising, the stimulation of T cells required high quantities of cytokines that, if administered in vivo, would likely trigger adverse events and would also lack persistence in blood.

[0008] In an effort to increase persistence in blood, certain chimeric molecules have been prepared in which two scFv portions were fused to generate a bi-specific therapeutic agent. For example, one bispecific therapeutic agent had single-chain variable fragments (scFv) against CD 16 on NK cells and against EpCAM on tumor cells to so facilitate antigen-specific antibody dependent cell-mediated cytotoxicity (ADCC). To improve on the ADCC of such therapeutics, the two scFv portions were coupled together via an IL15 portion to form a multi-component chimeric protein (TriKE, tri-specific killer engager; see e.g., Molecular Therapy Vol. 24 No. 7 Jul. 2016, 1312-1322; Human Vaccines & Immunotherapeutics 2016, vol. 12, No. 11, 2790-2796; Blood 2016 128:4291). Still other compositions using scFvs in multi-specific protein complexes with IL15 are described in WO 2018/075989. While desirable for antigen binding, scFv domains in these chimeric molecules do not have cytokine activity.

[0009] Even though there are numerous therapeutic agents known in the art, all or almost all of them suffer from various disadvantages. Therefore, there remains a need for improved compositions and methods to stimulate immune responses, especially as it relates to cancer therapy using cancer specific TSCM cells.

Summary of the Invention

[0010] The inventors have discovered compositions and methods in which multifunctional chimeric molecule complexes are used for stimulation of T cells, and in especially preferred aspects, contemplated multifunctional chimeric molecule complexes will stimulate T cell proliferation and differentiation of naive T cells into TSCM cells.

[0011] In one aspect of the inventive subject matter, the inventors contemplate a method of generating memory T cells that includes a step of contacting a plurality of lymphocytes with (i) a chimeric molecule complex comprising an IL15 portion, and at least one of an IL7 portion and an IL21 portion, or (ii) a chimeric molecule complex comprising an IL15 portion, and at least one of an IL18 portion, and an IL12 portion; and contacting the plurality of lymphocytes with an agonist for a TNF superfamily receptor. Most typically, the step of contacting is performed under conditions and for a time sufficient to allow differentiation of non-memory T cells to memory T cells and to allow proliferation of the memory T cells.

[0012] While not limiting to the inventive subject matter, it is generally preferred that the memory T cells are TSCM cells. In some aspects, the lymphocytes are contacted with the chimeric molecule complex and/or the agonist for a TNF superfamily receptor in vitro. For example, the lymphocytes can be contacted first with the chimeric molecule complex, and after several hours (e.g., 12-24, or 18-36 hours), with the agonist for a TNF superfamily receptor.

[0013] In still further contemplated aspects, the chimeric molecule complex comprises the IL7 portion and the IL21 portion, while in other aspects the chimeric molecule complex comprises the IL18 portion and the IL12 portion. Most typically, but not necessarily, the agonist for the TNF superfamily receptor is a multimeric 4-1BBL ligand (e.g., multimer of trimers) or an agonistic antibody against 4-1BB. Additionally, it is contemplated that the lymphocytes may further comprise NK cells and/or recombinant CAR-T cells.

[0014] Where desired, the proliferated memory T cells can be administered to a patient where the cells are contacted in vitro. Otherwise, the lymphocytes may also be contacted with the chimeric molecule complex and/or the agonist for a TNF superfamily receptor in vivo. In such case, the chimeric molecule complex preferably comprises the IL7 portion and the IL21 portion and/or the chimeric molecule complex preferably comprises the IL18 portion and the IL12 portion, and the agonist for the TNF superfamily receptor is a multimeric 4-1BBL ligand or an agonistic antibody against 4-1BB.

[0015] The inventors also contemplate a kit that includes a chimeric molecule complex comprising: an IL15 portion, and at least one of an IL7 portion and an IL21 portion; and an agonist for a TNF superfamily receptor. Most typically, the chimeric molecule complex is a TxM molecule, and the TxM molecule comprises the IL7 portion and the IL21 portion. It is further contemplated that the agonist for the TNF superfamily receptor is a multimeric 4- 1BBL ligand or an agonistic antibody against 4-1BB.

[0016] Alternatively, or additionally the inventors also contemplate a kit that includes a chimeric molecule complex comprising: an IL15 portion, and at least one of an IL18 portion, and an IL12 portion; and an agonist for a TNF superfamily receptor. As noted before, it is contemplated that the chimeric molecule complex is a TxM molecule, and/or that the TxM molecule comprises the IL18 portion and the IL12 portion. It is further contemplated that the agonist for the TNF superfamily receptor is a multimeric 4-1BBL ligand or an agonistic antibody against 4-1BB.

[0017] Various objects, features, aspects and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing.

Brief Description of the Drawing

[0018] Figure 1 is a schematic illustration of various memory T cells and selected associated characteristics.

[0019] Figure 2A shows an exemplary chimeric molecule complex.

[0020] Figure 2B shows an exemplary dimer formed of two chimeric molecule complexes of Figure 2A.

[0021] Figure 3A is an exemplary dimer for chimeric molecule complexes that include hIL7 and IL21.

[0022] Figure 3B is an exemplary dimer for chimeric molecule complexes that include mIL7 and IL21.

[0023] Figure 3C is an exemplary dimer for chimeric molecule complexes that includes hIL2l.

[0024] Figure 3D is an exemplary dimer for chimeric molecule complexes that includes hIL7.

[0025] Figure 3E is an exemplary dimer for chimeric molecule complexes that include L18 and IL12.

[0026] Figure 3F is an exemplary dimer for chimeric molecule complexes that includes L18.

[0027] Figure 4 depicts various graphs comparing ELISA data for selected TxM constructs (IL7, IL15, IL21) and N-803 (ALT-803). [0028] Figure 5 depicts selected results for induction of specific activation of IL7, IL15, and IL21 receptors.

[0029] Figure 6 shows exemplary results for expansion of sorted T cell populations following exposure to selected TxM constructs (IL7, IL15, IL21).

[0030] Figure 7 depicts effective expansion of sorted CD8+ T cells after brief stimulation with antiCD3/CD28 beads.

[0031] Figure 8 shows exemplary results for enhancement of granzyme B expression in human NK cells using selected TxM constructs (IL7, IL15, IL21).

[0032] Figure 9 depicts exemplary results for cytotoxicity and ADCC activity of human NK cells activated with selected TxM constructs (IL7, IL15, IL21) against SW1990 pancreatic cancer cells.

[0033] Figure 10 depicts exemplary results for cytotoxicity and ADCC activity of human NK cells activated with selected TxM constructs (IL7, IL15, IL21) against Ramos lymphoma cells.

[0034] Figures 11 and 12 depict exemplary results indicating that selected TxM constructs (IL7, IL15, IL21) better expands purified NK cells that individual cytokines.

[0035] Figure 13 depicts further exemplary results comparing ex vivo expansion of NK cells with expansion using selected TxM constructs (IL7, IL15, IL21).

[0036] Figure 14 illustrates exemplary results for phenotype determination of NK cells following stimulation with selected TxM constructs (IL7, IL15, IL21).

[0037] Figure 15 depicts exemplary results showing induction of certain cytokine expression in NK cells after exposure to selected TxM constructs (IL7, IL15, IL21).

[0038] Figure 16 depicts results for direct and antibody-mediated cytotoxicity

of hIL7/IL2l/TxM-expanded NK cells on Ramos lymphoma cells.

[0039] Figure 17 depicts exemplary results for activation of specific receptors of selected cells after activation with selected TxM constructs (IL15, IL21). [0040] Figure 18 depicts exemplary results for activation of specific receptors of selected cells after activation with selected TxM constructs (IL15, IL7).

[0041] Figure 19 are images for exemplary SDS-PAGE gels under reducing and non reducing conditions for various TxM constructs

[0042] Figure 20 depicts exemplary results comparing NK cell expansion using selected TxM constructs presented herein.

[0043] Figure 21 depicts exemplary results for cell activation, IFN-gamma production, and cell proliferation using selected TxM constructs (IL7, IL15, IL21).

[0044] Figure 22 depicts exemplary results comparing MCP-l induction in vivo using N-803 and selected TxM constructs (IL7, IL15, IL21).

[0045] Figures 23-28 depict exemplary results comparing proliferation of selected cells in vivo using N-803 and selected TxM constructs (IL7, IL15, IL21).

[0046] Figure 29 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induce reduced levels of Treg cells in the spleen of BALB/c mice.

[0047] Figure 30 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induce reduced proliferation of NK and CD8⁺ T cells in C57BL/6 mice.

[0048] Figure 31 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induce reduced proliferation of CD4⁺ T cells in C57BL/6 mice.

[0049] Figure 32 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induce reduced proliferation of NK and CD8⁺ T cells in C57BL/6 mice.

[0050] Figure 33 shows exemplary data for comparison of N-803 and selected TxM constructs (IL7, IL15, IL21) with respect to CD4⁺ T and B cell proliferation in C57BL/6 mice. [0051] Figure 34 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induces reduced numbers of NK and CD8⁺ T cells in C57BL/6 mice.

[0052] Figure 35 shows exemplary data for comparison of N-803 and selected TxM constructs (IL7, IL15, IL21) with respect to CD4⁺ T and B cell proliferation in C57BL/6 mice.

[0053] Figure 36 shows data establishing that NK cells expanded with selected TxM constructs (IL7, IL15, IL21) can still be primed other selected TxM constructs (IL18, IL15, IL12).

[0054] Figures 37-38 depict exemplary results for hILl8/ILl2/TxM priming after hIL7/IL2l/TxM expansion.

[0055] Figure 39 depicts exemplary results demonstrating that Ll8/ILl2/TxM induces specific activation of IL18, IL12, and IL15 receptors.

[0056] Figure 40 depicts exemplary results demonstrating the Effect of hILl8/ILl2/TxM on human NK cells.

[0057] Figure 41 depicts exemplary results demonstrating that Ll8/ILl2/TxM induces primary human NK cells to differentiate into cytokine induced memory like NK cells in vitro.

[0058] Figure 42 depicts exemplary results for Enhancement of granzyme B expression of human NK cells induced by MLl8/ILl2/TxM.

[0059] Figure 43 depicts exemplary results demonstrating that Ll8/ILl2/TxM induces similar NK cell killing as separate cytokines.

[0060] Figure 44 depicts exemplary results demonstrating cytotoxicity and ADCC activity of hILl8/ILl2/TxM-activated human NK cells against SW1990 pancreatic cancer cells.

[0061] Figure 45 depicts exemplary results for h2*ILl8/TxM IL15 activity.

[0062] Figure 46 depicts exemplary results for a h2*ILl8/TxM ELISA (ILl8:ILl5).

[0063] Figure 47 depicts exemplary results for h2*ILl8/TxM stimulation of aNK cells. [0064] Figure 48 depicts exemplary results for stimulation of T cells with a 4-1BB ligand. Figure 48 also sets out an amino acid sequence for the 4-1BB ligand.

[0065] Figure 49 is an exemplary schematic for the generation of antigen reactive T cells.

[0066] Figure 50 shows T cell phenotypes in single cell suspension of a kidney tumor.

[0067] Figure 51 depicts T cell subpopulations after expansion with selected compounds

[0068] Figure 52 is a photomicrograph of tumor cells after contact with TIL treated with SK or IL-2.

[0069] Figure 53 is an exemplary schematic for the generation/identification of neoepitope reactive T cells from the blood of a cancer patient.

[0070] Figure 54 shows exemplary results for the experiments of Figure 53.

[0071] Figure 55 shows further results for the experiments of Figure 53.

[0072] Figure 56 illustrates use of a tag for quantification of neoantigens and exemplary results.

[0073] Figure 57 shows exemplary results for short term T cells lines as presented herein on in vitro generated antigens.

Detailed Description

[0074] The inventive subject matter is directed to compositions and methods that increase the T cell memory cell repertoire and that are particularly suitable to generate and proliferate TSCM cells for immune therapy, especially for immune therapy of cancer.

[0075] TSCM cells are a rare subset of memory lymphocytes that have stem cell-like ability to self-renew, that have multipotent capacity to reconstitute the entire spectrum of memory and effector T cells subsets, and that are minimally differentiated cells. Moreover, TSCM cells exhibit extreme longevity and have robust potential for immune reconstitution and long term persistence after encounter with antigen. Figure 1 exemplarily depicts various T cell memory cells and their respective characteristics. Here APC denotes an antigen presenting cell, TN denoted a naive T cell, TSCM denotes a stem cell memory T cell, TCM denotes a central memory T cell, TEM denotes an effector memory T cell, and TEFF denotes an effector T cell.

[0076] Advantageously, the inventors have now discovered that the above memory T cells and especially TSCM cells can be generated and proliferated in an effective manner using one or more therapeutic proteins that stimulate differentiation and proliferation of the desired cells. Most preferably, the therapeutic proteins are chimeric TxM proteins that can be further used in combination with one or more stimulatory factors that interact with TNF superfamily receptors. For example, especially suitable TxM molecules will be IL15 based and further include one or more of IL7 and IL21. Additionally or alternatively, suitable TxM molecules will be IL15 based and further include one or more of IL18 and IL12. Stimulatory factors that interact with the TNF superfamily receptors especially include single and multimeric forms of 4-1BBL (which may form higher order multimers) and/or an agonistic mAh against 4- 1BB.

[0077] Figure 2A schematically depicts one exemplary embodiment of a chimeric molecule complex 100. Chimeric molecule complex 100 comprises first fusion protein 110 and second fusion protein 160. The first fusion protein 110 has Fc portion 120, IL15 receptor portion 130, and a first effector portion 140. Most typically, the first effector portion is a cytokine or biologically active mutant or portion thereof. The IL15 receptor portion is preferably an IL15 receptor alpha chain. First effector portion 140 is coupled to IL15 receptor portion 130 via first linker 135. Second fusion protein 160 has IL15 portion 170 coupled to second effector portion 180 via second linker 175. Most typically, the second effector portion is a cytokine or biologically active mutant or portion thereof. Moreover, it should be noted that the first or second effector portion may be optional. It should also be noted that as used herein, the term “IL15 portion” also expressly includes its mutant forms, and particularly IL15N72D. Thus, the IL15 portion may be a native (typically human) IL15 or an IL15 superagonist. In preferred embodiments, first and second fusion proteins 110 and 160 form a complex via non-covalent but specific binding of the IL15 portion 170 to the IL15 receptor portion 130.

[0078] It should be appreciated that chimeric molecule complexes can be combined in dimers (or higher multimers, depending on the type of Fc portion selected), which can increase not only the effectiveness but also the number of functions of treatment. It should also be noted that such dimers need not necessarily comprise identical chimeric molecule complexes, but may indeed comprise distinct chimeric molecule complexes that so allow additional effector or affinity portions (e.g., the first chimeric molecule complex includes IL7 and IL21, while the second chimeric molecule complex targets one or more specific epitopes using scFv portions).

[0079] Figure 2B schematically and exemplarily illustrates one embodiment of a dimer 200 formed of first chimeric molecule complex 210 and second chimeric molecule complex 260. First chimeric molecule complex 210 comprises first fusion protein 212 and second fusion protein 232. The first fusion protein has an Fc portion 215, IL15 receptor portion 220, and a first effector portion 230. The IL15 receptor portion is preferably a (human) IL15 receptor alpha chain. First effector portion 230 is coupled to IL15 receptor portion 220 via first linker 225. Second fusion protein 232 has a (human) IL15 portion 235 coupled to second effector portion 245 via second linker 240.

[0080] Similarly, second chimeric molecule complex 260 comprises third fusion protein 262 and fourth fusion protein 282. The third fusion protein has Fc portion 265, IL15 receptor portion 270, and a third effector portion 280, coupled to IL15 receptor portion 270 via third linker 275. The IL15 receptor portion is preferably a (human) IL15 receptor alpha chain. Fourth fusion protein 282 has a (human) IL15 portion 285 coupled to fourth effector portion 295 via fourth linker 290. Where desired, a fifth effector or affinity portion 267 may be coupled to the Fc portion 265 of the third fusion protein 262 via fifth linker 266. Of course, it should be appreciated that the fifth effector or affinity portion may also be coupled to the Fc portion 215 or to any one of first through fourth effector portions, not illustrated. In the example of Figure 2B, the first chimeric molecule complex 210 and the second chimeric molecule complex 260 are bound together via disulfide bond 250 and hydrophobic interactions between Fc portions. As noted above, it should be appreciated that in at least some embodiments dimer 200 is a heterodimer, where first chimeric molecule complex 210 and second chimeric molecule complex 260 are different, most typically by virtue of different effector portions. Therefore, multiple advantageous functions can be combined in one dimer.

[0081] Figures 3A-3E illustrate specific examples of selected chimeric molecule complexes used in the examples provided herein. In particular, Figure 3A depicts one TxM chimeric molecule that includes a human IL7 portion and an IL21 portion fused to different elements in the chimeric molecule as indicated. Similarly, Figure 3B depicts another TxM chimeric molecule that includes a murine IL7 portion and an IL21 portion fused to different elements in the chimeric molecule as indicated. Figure 3C shows a further chimeric molecule that includes human IL21 portions fused to selected elements in the chimeric molecule, and

Figure 3D shows yet another chimeric molecule that includes human IL7 portions fused to selected elements in the chimeric molecule. Figure 3E shows another chimeric molecule that includes human IL18 and IL12 portions fused to different elements in the chimeric molecule, and Figure 3F shows yet another chimeric molecule that includes human IL18 portions fused to selected elements in the chimeric molecule.

[0082] Regardless of the particular arrangement, it should be appreciated that the so formed chimeric molecule complexes will allow for a combination of specific biological effects that provide activity in an enhanced manner over a prolonged period of time. Moreover, where the chimeric complexes include an affinity portion (e.g., scFv), such effects can be further targeted to specific locations. For example, the affinity portions may target tissue (and especially tumor) specific markers to so direct and maintain the effects at a desired location. Among other biological effects, particularly preferred biological effects are those that trigger development and/or proliferation of various memory T cells and especially TSCM cells as is shown in more detail below. In addition, it should be appreciated that the differentiation and proliferation of naive T cells to various T memory cells can be further enhanced by ligands to TNF superfamily receptors, and especially single and multimeric forms of 4-1BBL (which may form higher order multimers) and/or an agonistic mAh against 4-1BB. Advantageously, IL15 activates T cells and promotes expression of 4-1BB, whereas IL7 and IL21 promote differentiation of T cells to selected memory T cells, and particularly TSCM· In the presence of 4-1BBL or agonistic mAh against 4-1BB, proliferation of such memory T cells is further stimulated.

[0083] Furthermore, it should be recognized that contemplated chimeric molecule complexes have significantly increased persistence (/._<?., extended serum half life time), which allows the molecule to perform its functions while delaying degradation in the blood stream. Indeed, it was observed that the Fc portion significantly increased the lifespan of IL15 in circulation as compared to IL15 per se. This provides the chimeric molecule complex with persistence, thus prolonging the activity of the molecule in circulation. Another contemplated benefit, as is discussed in more detail below, is that the Fc portion greatly simplifies purification processes during the preparation of purified chimeric molecule complexes. As already noted above, the Fc portion may include at least an Fc portion of an IgG, IgM, IgA, IgD, or IgE antibody. [0084] It is contemplated that a preferred arrangement of the first fusion protein, from N-to C-terminus, is as follows: first affinity portion, IL15 receptor, Fc portion. However, the first fusion protein may also be arraigned as follows: IL15 receptor, Fc portion, first affinity portion. Ideally, the second affinity portion is arranged, from N- to C-terminus as follows: second affinity portion, IL15 portion.

[0085] With respect to the production of contemplated chimeric molecule complexes and dimers/multimers, various methods of synthesis are contemplated. For example, and in general, it is contemplated that chimeric molecule complexes can be expressed from one or more nucleic acid in patient cells in vivo, in patient or mammalian production cells (e.g.,

CHO cells) in vitro, or where desired in other cells such as bacteria, yeast, or non-mammalian cells. Most preferably, the individual components of the fusion proteins (e.g., IL15 or ILl5Ra, Fc portion, affinity portions, etc.) are expressed from a recombinant nucleic acid as a single polypeptide, typically with short and flexible linker sequences in between the individual components. However, in less preferred aspects, the individual components can be expressed individually and are then coupled together after expression. Such coupling will typically include use of high-affinity binding pairs such as biotin/avidin, protein A/G, short nucleic acids with sequence complementarity, etc. Most preferably, the fusion protein (components) of the chimeric molecule complex will be co-expressed within the same cell.

[0086] For example, contemplated chimeric molecule complexes can be prepared as an IL7 (or ILl8)/huILl5N72D fusion protein in complex with (/._<?., non-covalently bound to) an IL2l(or ILl2)/huILl5RaSu/huIgGl Fc fusion protein using sequences known in the art. For example, with respect to IL15 it is contemplated that all known IL15 sequences are deemed suitable for use herein, and particularly human or humanized IL15. Thus, the term“IL15” as used herein refers to all IL15 forms (including isoforms, prepro, and pro forms), preferably mammalian, and most preferably human or humanized forms. Moreover, contemplated IL15 proteins also include mutant forms, and particularly mutants with higher biological activity such as N72D mutant form (see e.g., Zhu et al., J Immunol, 183: 3598-3607, 2009). For example, suitable IL15 sequences can be found under Genbank accession numbers X91233 (Genomic DNA), CH471056 (Genomic DNA), or X94223 (mRNA), AK290619 (mRNA), BC100961 (mRNA), and the protein sequence is known under UniProtKB identifier P40933.

[0087] Similarly, with respect to the IL15 receptor it is contemplated that all known IL15 receptor sequences are deemed suitable for use herein. However, especially preferred receptor sequences are high- affinity IL15R alpha chain (ILl5Ra) receptors, and particularly human or humanized ILl5Ra. Thus, the term“IL15 receptor” as used herein refers to all IL15R forms (including all isoforms), preferably mammalian, and most preferably human or humanized forms. For example, suitable ILl5Ra sequences can be found under Genbank accession number AY316538 (Genomic DNA), CH471072 (Genomic DNA), or CR457064 (mRNA), AK304211 (mRNA), BC121140 mRNA, and the protein sequence is known under UniProtKB identifier Q13261.

[0088] As discussed above, the fusion proteins may also comprise an Fc domain of an antibody. Most typically, Fc domains will ultimately be present as a dimer as exemplarily shown in Figure 2B. Moreover, it is generally preferred that immunoglobulin from which the Fc domain is obtained is a mammalian, and most preferably human immunoglobulin, and especially IgGl and IgG2. Other suitable Fc domains may be derived from different Ig classes (such as IgG, IgA, IgE) or subclasses (e.g., IgGl, IgG2, IgG3, IgAl, IgGA2) or variants thereof (see e.g., WO 97/34631 and WO 96/32478). Fc domains are well known in the art and suitable sequences can be obtained from publically available sources such as GenBank, EMBL, SwissProt, etc.

[0089] In a particularly preferred example, a first fusion protein containing ILl5Ra (or portion thereof) comprises a first cytokine portion coupled via a linker to a hu-ILl5RaSu portion coupled via a linker to a hu-IgGl Fc portion. Such first fusion protein forms a complex (via ILl5/ILl5Ra binding) with a second fusion protein containing IL15 (or portion thereof) that comprises a second cytokine portion coupled via a linker (as discussed below) to a hu-ILl5N72D portion. Such complex may then form an antibody-like dimer via interaction and disulfide bond formation of the two Fc portions.

[0090] Notably, contemplated Fc fusion proteins will have an in vivo pharmacokinetic profile comparable to that of human IgG with a similar isotype, and as such substantially increase the halflife time (up to 21 days) as compared to IL15 alone. For example, the pharmacokinetics and biological activity of a IL15 super-agonist (IL15N72D) was dramatically increased by binding the IL15N72D to a ILl5Ra/Fc fusion protein (ALT-803), such that the super agonist complex has at least 25 -times the activity of the native cytokine in vivo (see e.g., Cytokine,

56: 804-810, 2011). Such advantage is expected to persist where such or similar scaffolds are used to carry one or more affinity portions. [0091] As noted above, contemplated fusion proteins will preferably include a (preferably flexible) peptide linker between each of the individual components complexes. Most typically, suitable linker sequences will include between about 7 to 20 amino acids, and preferably between about 10 to 20 amino acids. In further preferred aspects, suitable linker sequences are preferably flexible to reduce steric hindrance and/or facilitate proper folding/tertiary structure. Thus, linkers may include amino acids with small side chains, such as glycine, alanine, and serine, to provide for flexibility (G4S linker). Where desired, such mini- sequences may be repeated to achieve a desired length. However, numerous alternative linker sequences are also deemed suitable and appropriate linkers are known in the art (e.g., Whitlow, M. et ak, (1991) Methods : A Companion to Methods in Enzymology, 2:97-105). Therefore, most typically a linker sequence is between the affinity portion and the IL15 or ILl5Ra, and between the ILl5Ra and the Fc portion.

[0092] Suitable cytokine sequences include all known cytokine sequences, and especially those for IL7, IL12, IL18, and IL21. Preferably, the cytokine sequences are mammalian cytokines, and especially human cytokine sequences. For example, suitable mammalian and human IL7 protein sequences and isoforms are known from UniProtKB record P13232, and suitable mammalian and human IL12 protein sequences and isoforms for alpha and beta subunits are known from UniProtKB records E7ENE1 and E9PGR3, while suitable mammalian and human IL18 protein sequences and isoforms are known from UniProtKB record Q14116, and suitable mammalian and human IL21 protein sequences and isoforms are known from UniProtKB record Q9HBE4. As will be readily appreciated, all amino acid sequences can be back-translated into corresponding DNA sequences using appropriate codon usage or codon optimization in the respective production cell where the chimeric molecules are recombinantly produced. Moreover, it should be appreciated that all mutant forms are also expressly deemed suitable for use herein, particularly where the mutant protein has increased biological activity as compared with the wildtype form.

[0093] Most preferably, contemplated fusion proteins are encoded on a recombinant nucleic acid, typically with codon usage adapted to the host cell that is used to express the fusion proteins. Moreover, it should be appreciated that the chimeric fusion protein complexes may be produced in vivo or in vitro. Therefore, the recombinant nucleic acid may be part of a viral nucleic acid (e.g., recombinant adenovirus that also encodes one or more neoepitopes or polytopes, optionally with co- stimulatory molecules) or part of a DNA vaccine. Alternatively, the recombinant nucleic acid may be a RNA or DNA that is transfected into a production cell to generate recombinant protein. Preferably, the recombinant nucleic acid is part of a vector for extrachromosomal replication such as a phage, vims, plasmid, phagemid, cosmid, or YAC. In particularly preferred aspects, a DNA plasmid is constructed to encode the fusion proteins contemplated herein and is used to prepare the complexes in clinically meaningful quantities. To that end, the recombinant nucleic acid can be inserted into an appropriate expression vector (/._<?., a vector that contains the necessary elements for the transcription and translation of the inserted protein-coding sequence). A variety of host- vector systems may be utilized to express the protein-coding sequence, including mammalian cell systems infected with a virus (e.g., vaccinia virus, adenovirus, etc.), insect cell systems infected with virus (e.g., baculovirus), microorganisms such as yeast containing yeast vectors, or bacteria transformed with bacteriophage DNA, plasmid DNA or cosmid DNA. Depending on the host- vector system utilized, any one of a number of suitable transcription and translation elements may be used. The fusion proteins described herein are preferably produced by standard recombinant DNA techniques.

[0094] Another advantage of contemplated compositions and methods is that the step of purification is greatly simplified because the entire chimeric molecule complex can be pulled out of solution by the binding of the Fc portion to a corresponding affinity protein. Therefore, in preferred embodiments, the Fc portion is ideally long enough to (a) facilitate the isolation of the chimeric molecule complex with any affinity protein that binds to the Fc portion (e.g. Protein A or G), and/or (b) facilitate the binding to the Sudlow II domain in albumin. Thus it is contemplated that in at least some embodiments the chimeric molecule complex can be administered as protein complexes with albumin. It is also contemplated that IL15 can be further coupled to alpha-CD3, cellulose binding protein, and/or oligohistidine for the purpose of easier affinity purification. Alternatively, IL15 could be bound to IL2 or another cytokine portion to further enhance activation capability.

[0095] When forming dimers, it is contemplated that first and second chimeric molecule complexes can be made in separate batches as homodimers. The batches can then be combined and a reducing agent can then be added to split the bonds between homodimers so the chimeric molecule complexes dissociate. The reducing agent can then be pulled out (e.g. via dialysis), which will cause chimeric molecule complexes to re-associate, but this time in lst-lst, 2nd-2nd, and lst-2nd combinations. Further details and methods of producing fusion proteins based on ILl5/ILl5Ra are described in WO 2017/205726, which is incorporated by reference herein.

[0096] It should further be appreciated that the fusion proteins of the inventive subject matter can be combined with any appropriate pharmaceutically acceptable amount in any quantities, and will generally present in an amount of 1-95 % by weight of the total weight of the composition. Contemplated compositions may be provided in a dosage form that is suitable for parenteral (e.g., subcutaneous, intravenous, intramuscular, intravesicular, intratumoral or intraperitoneal) administration route. As will be appreciated, pharmaceutical compositions contemplated herein may be formulated according to conventional pharmaceutical practice (see, e.g., Remington: The Science and Practice of Pharmacy (20th ed.), ed. A. R. Gennaro, Lippincott Williams & Wilkins, 2000 and Encyclopedia of Pharmaceutical Technology, eds.

J. Swarbrick and J.C. Boylan, 1988-1999, Marcel Dekker, New York). In addition, the formulations may further include appropriate excipients that, upon administration, release the therapeutic agent in a controlled manner. Examples include single or multiple unit tablet or capsule compositions, oil solutions, suspensions, emulsions, microcapsules, microspheres, molecular complexes, nanoparticles, patches, and liposomes.

[0097] Most preferably, however, the pharmaceutical compositions comprising a fusion protein complex of the invention may be in a form suitable for sterile injection. To prepare such a composition, the suitable active therapeutic(s) are dissolved or suspended in a parenterally acceptable liquid vehicle. Among acceptable vehicles and solvents that may be employed are water, water adjusted to a suitable pH by addition of an appropriate amount of hydrochloric acid, sodium hydroxide or a suitable buffer, l,3-butanediol, Ringer's solution, and isotonic sodium chloride solution and dextrose solution. The aqueous formulation may also contain one or more preservatives (e.g., methyl, ethyl or n-propyl p-hydroxybenzoate).

In cases where one of the compounds is only sparingly or slightly soluble in water, a dissolution enhancing or solubilizing agent can be added, or the solvent may include one or more cosolvents, e.g. 10-60% w/w of propylene glycol.

[0098] Human dosage amounts are initially determined by extrapolating from the amount of compound used in mice or non-human primates, as a skilled artisan recognizes it is routine in the art to modify the dosage for humans compared to animal models. For example, the dosage may vary from between about 1 pg compound/kg body weight to about 50 mg compound/kg body weight; or from about 5 mg/kg body weight to about 40 mg/kg body weight or from about 10 mg/kg body weight to about 30 mg/kg body weight; or from about 50 mg/kg body weight to about 20 mg/kg body weight; or from about 100 mg/kg body weight to about 10 mg/kg body weight; or from about 150 mg/kg body weight to about 500 mg/kg body weight. For example, the dose is about 0.01, 0.05, 0.1, 1, 5, 10, 25, 50, 75, 100, 150, 200,250, 300, 350,400,450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, or 1,000 mg/kg body weight. Alternatively, doses are in the range of about 5 mg compound/Kg body weight to about 20 mg compound/kg body weight. In another example, the doses are about 0.01, 0.05, 0.1, 1, 5,

8, 10, 12, 14, 16 or 18 mg/kg body weight.

[0099] It is contemplated that in preferred embodiments, chimeric molecule complexes will be administered in formulations suitable for injection in therapeutically effective amounts. Therapeutically effective amounts can range from 1 ng to 1,000 mg, from 1 pg to 500 mg, from 100 pg to 100 mg. Solubility co-solvents or detergents can be used to increase the storage stability. Additionally, or alternatively, mixed-phase or two-phase liquid systems can be used. Most preferably, chimeric molecule complexes will be formulated in a liquid carrier ready to use. However, it is contemplated that the chimeric molecule complexes can also be formulated in a dried form for reconstitution, by lyophilization or freeze drying.

[00100] With respect to agonists for TNF superfamily receptors it is generally

contemplated that suitable agonists will be 4-1BBL or modified forms thereof, 0X40 ligands or modified forms thereof, or GITR ligands or modified forms thereof. However, it is especially preferred that the agonist is a trimer or multimers of trimers of 4-1BBL. In this context, it should be noted that the chimeric molecules as described above will exert a stimulatory effect on T cells that triggers expression of 4-1BB (and potentially GITR and/or 0X40), which will then bind the agonists contemplated herein. In addition, as the chimeric molecules will also trigger the differentiation of T cells (exposed to tumor antigens or naive) towards memory T cells and especially TSCM cells, exposure of the stimulated/differentiated T cells to the agonists will enhance proliferation of this desirable sub-population of T memory cells.

[00101] For example, suitable multimeric forms of 4-1BBL that stimulate T cells are known, among other references, from EP 1736482, WO 2004/069876. Likewise, further preferred multimeric forms include those described elsewhere (see e.g., J Immunother Cancer. 2014; 2(Suppl 3): P246.). On the other hand, where agonistic antibodies against 4- 1BB are employed, such antibodies are described, for example, in WO 2003/084999, WO 2015/179236, and US 8163550. A fully humanized monoclonal antibody with agonist activity is known as utomilumab (Pfizer, PF-05082566).

[00102] As with the TxM chimeric molecules, administration of the agonists (e.g., 4-1BBL compositions or agonistic mAh against 4-1BB) will preferably be performed following protocols known in the art. Therefore, human dosage amounts are initially determined by extrapolating from the amount of compound used in mice or non-human primates, as a skilled artisan recognizes it is routine in the art to modify the dosage for humans compared to animal models.

[00103] For example, the dosage may vary from between about 1 pg compound/kg body weight to about 50 mg compound/kg body weight; or from about 5 mg/kg body weight to about 40 mg/kg body weight or from about 10 mg/kg body weight to about 30 mg/kg body weight; or from about 50 mg/kg body weight to about 20 mg/kg body weight; or from about 100 mg/kg body weight to about 10 mg/kg body weight; or from about 150 mg/kg body weight to about 500 mg/kg body weight. For example, the dose is about 0.01, 0.05, 0.1, 1, 5, 10, 25, 50, 75, 100, 150, 200,250, 300, 350,400,450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, or 1,000 mg/kg body weight. Alternatively, doses are in the range of about 5 mg compound/Kg body weight to about 20 mg compound/kg body weight. In another example, the doses are about 0.01, 0.05, 0.1, 1, 5, 8, 10, 12, 14, 16 or 18 mg/kg body weight.

[00104] Most typically, the chimeric TxM molecules and TNF superfamily receptor agonists will be co- administered, either in the same formulation, or from distinct

formulations. Such administration can be performed using the same route (e.g., intravenously or subcutaneously) or using two different routes (e.g., subcutaneous and intratumoral).

Furthermore, it should be noted that the co-administration may be staged in which the agonist is first administered and the TxM is then administered 12-24 hours, or 18-36 hours, or 24-48 hours thereafter. Most typically, the formulation(s) can be administered in a diluent or without prior dilution. Consequently, the inventors contemplate kits and multi-drug compositions that include as a first component the TxM chimeric complexes and as a second component the TNF superfamily receptor agonists as described herein.

[00105] While in some embodiments administration of the chimeric TxM molecules and TNF superfamily receptor agonists is performed in vivo, administration of the chimeric TxM molecules and TNF superfamily receptor agonists is performed in vitro, especially using a preparation that is enriched in lymphocytes (e.g., huffy coat from whole blood, or isolated cells from tumor infiltrating lymphocytes). Where administration is performed in vitro, it is contemplated that the cells (e.g. , T cells, lymphocyte enriched preparation, etc.) are further cultivated under conditions that allow for continued proliferation, typically using cell culture methods and reagents well known in the art. As will be readily appreciated, the cells can be incubated in the presence of any desirable antigen and most typically in the presence of a tumor associated or tumor and/or patient specific antigen (e.g., neoantigen, tumor biopsy portion, circulating tumor cells). Upon sufficient proliferation and differentiation, the so obtained cell population is then transfused to a patient.

[00106] Therefore, it should be appreciated that exposure of (activated or antigen stimulated) T cells to the compositions contemplated herein will not only result in (further) activation and differentiation of the cells towards the memory phenotype, but also result in significant (hyper)proliferation of the memory T cells, and particularly TSCM cells. Such cells can then be transfused to a patient where activation and proliferation was performed in vitro. On the other hand, activation and/or proliferation can be performed in vivo by injection of the compositions contemplated herein.

Examples

[00107] The following examples provide exemplary guidance and are not intended to limit the scope of the inventive concept provided herein. Moreover, and unless otherwise indicated, all experiments were performed following standard protocols for cell culture and analysis well known in the art.

[00108] To confirm the presence of the cytokine components in the TxM constructs (for TxM, see URL: altorbioscience.com; here, scFv domains were replaced by selected cytokines as described above), ELISA experiments were performed. Figure 4 depicts various graphs comparing ELISA data for selected TxM constructs (IL7, IL15, IL21) and N-803 (ALT-803). It should be noted that N-803 does not include cytokines other than IL15 portion.

[00109] With respect to biological activity of the TxM constructs compared to individual cytokines, IL7 dependent 2E8 cells (10⁵) were stimulated for 2 days with hIL7/IL2l/TxM or IL7 and cell proliferation was assessed using PrestoBlue. The EC50 of IL7 in hIL7/IL2l/TxM is 14 pM. n=4 from 2 experiments. aNK cells (2xl0⁵) were stimulated for 40 hours with hIL7/IL2l/TxM or N-803 and production of IFNy was measured by ELISA. n=2 from 1 experiment. IL2/15 dependent 32D-IL2/15RP cells (10⁴) were stimulated for 3 days with L7/IL2l/TxM or N-803 and cell proliferation was assessed using PrestoBlue. The EC₅₀ of IL15 in hIL7/IL2l/TxM is 530 pM. n=4 from 2 experiments. Figure 5 depicts selected results for specific activation of IL7, IL15, and IL21 receptors.

[00110] Figure 6 shows exemplary results for expansion of sorted T cell populations following exposure to selected TxM constructs (IL7, IL15, IL21). Here, Sorted CD8⁺ Naive, Central memory, Effector memory and Stem cell memory T cells were labeled with CFSE and stimulated with media alone (US) or IL7/IL21 (25ng), IL7/N-803 (25ng/l44ng), IL21/N- 803 (25ng/l44ng), IL7/IL21/N-803 (25ng/25ng/l44ng), TxM (l.4mg) in 200ml total volume in 96 well flat bottom plate in 37 °C, 5% C( As can be readily seen, the TxM constructs presented herein have significant activity on naive and selected memory T cells.

[00111] Similarly, Figure 7 depicts effective expansion of sorted CD8+ cells after brief stimulation with antiCD3/CD28 beads. More specifically, CD8⁺ T Cells were purified using RosetteSep, labeled with CellTrace Violet, and stained with CD45RO, CD95, CD8 and CCR7 for sorting. 10⁴ sorted Naive, Central memory, Effector memory, or Stem cell memory cells were stimulated with aCD3/28 beads (0.5: 1) with either N-803 or h2*IL7(ILl5)/TxM or hIL7/IL2l/TxM (10hM) for 2 days. Beads were removed, cells were recounted and resuspended with fresh stimuli for 6 additional days.

[00112] The effects of contemplated constructs were also tested for NK cells and typical results are shown in Figure 8. Here, the graph shows exemplary results for enhancement of granzyme B expression in human NK cells using selected TxM constructs (IL7, IL15, IL21), and Granzyme B expression was determined in pre-activated human NK cells (16 h, 50 nM, n=2). Clearly, the TxM construct (IL7, IL15, IL21) had superior activity on granzyme B expression as compared to N-803.

[00113] The effects of contemplated constructs were also tested with respect to cytotoxicity and ADCC activity. Figure 9 depicts exemplary results for cytotoxicity and ADCC activity of human NK cells activated with selected TxM constructs (IL7, IL15, IL21) against SW1990 pancreatic cancer cells. Here, fresh NK cells were mixed with SW1990 cells for 40 hr at E:T of 2:1. aTF = 0.1 nM. N-803 or ML7/IL2l/TxM = 50 nM. n=2.

Comparison is with ALT-803 (N-803). Similarly, Figure 10 depicts exemplary results for cytotoxicity and ADCC activity of human NK cells activated with selected TxM constructs (IL7, IL15, IL21) against Ramos lymphoma cells. Here, fresh NK cells were mixed with Ramos cells for 40 hr at E:T of 1:1. aCD20 = 1 nM. N-803 or L7/IL2l/TxM = 0.5 nM. n=2.

[00114] Figures 11 and 12 depict further exemplary results indicating that selected TxM constructs (IL7, IL15, IL21) better expand purified NK cells than individual cytokines in combination with N-803 (which notably provides the same cytokine components) following an experimental setup as indicated in the Figure. Figure 13 depicts further exemplary results comparing ex vivo expansion of NK cells with selected TxM constructs (IL7, IL15, IL21) vis- a-vis known methods. Comparative protocols are found elsewhere (Granzin M, et al. (2015) Cytotherapy. l7:5;62l-632; Denman CJ, et al. (2012) PLoS ONE 7(1): e30264).

[00115] Figure 14 illustrates exemplary FACS results for phenotype determination of NK cells following stimulation with selected TxM constructs (IL7, IL15, IL21). As can be readily seen, the stimulated cells had an activated phenotype. Figure 15 depicts exemplary results showing induction of certain cytokine expression in NK cells after exposure to selected TxM constructs (IL7, IL15, IL21). Here, hIL7/IL2l/TxM-expanded NK cells were stimulated by separate cytokines or TxM overnight.

[00116] Thusly activated NK cells were also tested for cell killing activity and typical results are shown in Figure 16. Results are shown for direct and antibody-mediated cytotoxicity of hIL7/IL2l/TxM-expanded NK cells on Ramos lymphoma cells. Here, purified human NK cells (0.5xl0⁶/mL) were expanded with 20 nM hIL7/IL2l/TxM for 9 days, washed once, and mixed with CellTrace Violet labeled CD20⁺ Ramos Burkitt’s lymphoma cells (10⁵) for 4 hr, followed by flow cytometry analysis in the presence of 7-AAD viability reagent. As can be readily seen, cell killing was significant at both TxM concentrations.

[00117] Figure 17 depicts exemplary results for activation of specific receptors of selected cells after activation with selected TxM constructs (IL15, IL21). Here, aNK cells (2xl0⁵) were stimulated for 40 hours with h2*IL2l/TxM or N-803 and production of IFNy was measured by ELISA. n=2 from 1 experiment. IL2/15 dependent 32D-IL2/15RP cells (10⁴) were stimulated for 3 days with h2*IL2l/TxM or N-803 and cell proliferation was assessed using PrestoBlue. The EC₅₀ of IL15 in h2*IL2l/TxM is 56 pM. n=4 from 2 experiments.

[00118] Figure 18 depicts exemplary results for activation of specific receptors of selected cells after activation with selected TxM constructs (IL15, IL7). More specifically, IL7 dependent 2E8 cells (10⁵) were stimulated for 2 days with h2*IL7(ILl5)/TxM or IL7 and cell proliferation was assessed using PrestoBlue. The EC50 of IL7 in h2*IL7(ILl5)/TxM is 13.3 pM. n=4 from 2 experiments. aNK cells (2xl0⁵) were stimulated for 40 hours with h2*IL7(ILl5)/TxM or N-803 and production of IFNy was measured by ELISA. n=2 from 1 experiment. IL2/15 dependent 32D-IL2/15RP cells (10⁴) were stimulated for 3 days with h2*IL7(ILl5)/TxM or N-803 and cell proliferation was assessed using PrestoBlue. The EC50 of IL15 in h2*IL7(ILl5)/TxM is 81.3 pM. n=4 from 2 experiments.

[00119] The TxM constructs were prepared using recombinant nucleic acid constructs and purified following standard protocols. Figure 19 are images for exemplary SDS-PAGE gels under reducing and non-reducing conditions for various TxM constructs.

[00120] Figure 20 depicts exemplary results comparing NK cell expansion using selected TxM constructs presented herein. Here, purified human NK cells were stimulated with 19.4 nM hIL7/IL2l/TxM, h2*IL2l/TxM, or hIL7(ILl5)/TxM, and cell number was maintained in the range of 0.5-2 xl0⁶/ml. Cell number assessed with Vi-CELL XR. n=2 from 1 experiment. As can be seen, the combined activity on cell proliferation of all three cytokines in a single TxM is substantially higher than TxM constructs providing only two (/._<?., IL15/IL7, or IL15/IL21). Thus, in view of the above, it should be noted that hIL7/IL2l/TxM effectively expands human T cells and T cell subsets including TSCM in the presence and absence of aCD3/28 beads. Moreover hIL7/IL2l/TxM also expands human NK cells in the absence of feeder cells.

[00121] Figure 21 depicts exemplary results cell activation IFN-gamma production and cell proliferation using selected TxM constructs (IL7, IL15, IL21). More specifically, IL7 dependent 2E8 cells were stimulated for 15 minutes with mIL7/IL2l/TxM or IL7/IL21/N-803 combinations and phosphorylation of STAT5 was assessed using flow cytometry. n=l from 1 experiment. aNK cells (2xl0⁵) were stimulated for 40 hours with mIL7/IL2l/TxM or N-803 and production of IFNy was measured by ELISA. n=2 from 1 experiment. IL2/15 dependent 32D-IL2/l5R cells (10⁴) were stimulated for 3 days with mIL7/IL2l/TxM or N-803 and cell proliferation was assessed using PrestoBlue. The EC50 of IL15 in mIL7/IL2l/TxM is 1.7 nM. n=4 from 2 experiments.

[00122] Figure 22 depicts exemplary results comparing MCP-l induction in vivo using N- 803 and selected TxM constructs (IL7, IL15, IL21). Here, BALB/c mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and serum cytokines were analyzed by flow cytometry using BD Cytometric Bead Array. Similarly, Figures 23-28 depict exemplary results comparing proliferation of selected cells in vivo using N-803 and selected TxM constructs (IL7, IL15, IL21). Here, BALB/c mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell proliferation in the peripheral blood was assessed by flow cytometry using Ki67 (Figures 23- 26). BALB/c mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell number in the spleen was assessed by flow cytometry (Figures 27-28). BALB/c mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and T_reg frequency in the spleen was assessed by flow cytometry using CD25 and FoxP3 (Figure 29). Figure 30 shows exemplary results indicating that compared to N- 803, selected TxM constructs (IL7, IL15, IL21) induce reduced proliferation of NK and CD8⁺ T cells in C57BL/6 mice. Figure 31 shows exemplary results indicating that compared to N- 803, selected TxM constructs (IL7, IL15, IL21) induce reduced proliferation of CD4⁺ T cells in C57BL/6 mice. Here, C57BL/6 mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell proliferation in the peripheral blood was assessed by flow cytometry using Ki67.

[00123] Figure 32 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induce reduced proliferation of NK and CD8⁺ T cells in C57BL/6 mice. Here, C57BL/6 mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell proliferation in the spleen was assessed by flow cytometry using Ki67. Figure 33 shows exemplary data for comparison of N-803 and selected TxM constructs (IL7, IL15, IL21) with respect to CD4⁺ T and B cell proliferation in C57BL/6 mice. Here, C57BL/6 mice

(n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell proliferation in the spleen was assessed by flow cytometry using Ki67.

[00124] Figure 34 shows exemplary results indicating that compared to N-803, selected TxM constructs (IL7, IL15, IL21) induces reduced numbers of NK and CD8⁺ T cells in C57BL/6 mice. Here, C57BL/6 mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell number in the spleen was assessed by flow cytometry. Figure 35 shows exemplary data for comparison of N-803 and selected TxM constructs (IL7, IL15, IL21) with respect to CD4⁺ T and B cell proliferation in C57BL/6 mice. Here, C57BL/6 mice (n=2/group) were injected with N-803 (0.2 mg/kg) or an equivalent level of mIL7/IL2l/TxM (20 mg/kg) with regard to IL15 activity and cell number in the spleen was assessed by flow cytometry.

[00125] It is noted that mIL7/IL2l/TxM functions similarly to N-803 in BALB/c mice, which more closely represents the effect of N-803 in humans. Compared to N-803, mIL7/IL2l/TxM is less effective in C57BL/6 mice, however, still has desirable effects as described above.

[00126] Figure 36 shows data establishing that NK cells expanded with selected TxM constructs (IL7, IL15, IL21) can still be primed other selected TxM constructs (IL18, IL15, IL12). Here, purified human NK cells (0.5xl0⁶/ml) were expanded with 20 nM

hIL7/IL2l/TxM for 9 days, washed once, primed with 19.4 nM TxM for 20 hr (2xl0⁶/mL), in the presence of brefeldin A and monensin for the final 4 hr, and intracellular IFNy expression was analyzed by flow cytometry. Figures 37-38 depict results for hILl8/ILl2/TxM priming after hIL7/IL2l/TxM expansion. Here, purified human NK cells were expanded with 20 nM hIL7/IL2l/TxM for 12 days, washed once, primed with 19.4 nM hILl8/ILl2/TxM or hIL7/IL2l/TxM for 16 hours, washed once, and rested in 77.6 pM N-803 for 7 more days, maintaining the cell density at 0.5-2xl0⁶/mL, and intracellular IFNy expression was assessed by flow cytometry in the presence of brefeldin A and monensin for 4 hr following restimulation (Figure 37). Purified human NK cells were expanded with 20 nM

hIL7/IL2l/TxM for 12 days, washed once, primed with 19.4 nM hIL18/IL12/TxM or hIL7/IL21/TxM overnight, washed once, and rested in 77.6 pM N-803 for 6 more days, maintaining the cell density at 0.5-2xl0⁶/mL (Figure 38).

[00127] Therefore, the inventors contemplate that TSCM can be generated and/or expanded in vitro, and that even TSCM CAR-T cells can be generated and/or expanded. As such persistence and efficacy of TSCM CAR-T in vivo is especially contemplated. Such activated cells may therefore be used as an injectable to support CAR-T therapy. Similarly, with respect to NK cells it is contemplated that contemplated compositions and methods are suitable as a cell-free method for expansion of NK cells and that such treated NK cells will have anti-tumor efficacy in vitro and in vivo. [00128] Figure 39 depicts further exemplary results demonstrating that hILl8/ILl2/TxM induces specific activation of IL18, IL12, and IL15 receptors. More specifically, IL18 dependent HEK-Blue IL18 cells (5xl0⁴) were stimulated for 20-22 hours with

MLl8/ILl2/TxM or IL18 and cell activation was assessed using QuantiBlue. The EC50 of IL18 in hILl8/ILl2/TxM is 7.1 pM. n=l4 from 6 experiments. IL12 dependent HEK-Blue IL12 cells (5xl0⁴) were stimulated for 20-22 hours with hILl8/ILl2/TxM or IL12 and cell activation was assessed using QuantiBlue. The EC50 of IL12 in hILl8/ILl2/TxM is 99.1 pM. n=9 from 4 experiments. IL2/15 dependent 32D-IL2/ 15RP cells (10⁴) were stimulated for 3 days with hILl8/ILl2/TxM or N-803 and cell proliferation was assessed using PrestoBlue. The EC50 of IL15 in hILl8/ILl2/TxM is 1.7 nM. n=4 from 2 experiments.

[00129] Figure 40 depicts exemplary results demonstrating the effect of hILl8/ILl2/TxM on human NK cells. Here, human NK cells (5xl0⁵ @ 5xl0⁶/mL, >95% CD56⁺) were stimulated for 12-14 hours with hILl8/ILl2/TxM and surface expression of CD25 was assessed by flow cytometry n = 6 from 3 experiments (left panel). Human NK cells (5xl0⁵ @ 5xl0⁶/mL, >95% CD56⁺) were stimulated for 16-18 hours with hILl8/ILl2/TxM, in the presence of brefeldin A and monensin for the final 4 hours, and intracellular expression of IFNy was assessed by flow cytometry n = 6 from 3 experiments (right panel).

[00130] Figure 41 depicts exemplary results demonstrating that hILl8/ILl2/TxM induces primary human NK cells to differentiate into cytokine induced memory like NK cells in vitro. Here, primary human NK cells (2xl0⁶/mL) were stimulated for 16 hours with

hILl8/ILl2/TxM (38.8 nM) or N-803 as a control (77.6 pM equivalent to 1 ng/ml IL15) and cells were rested in 77.6 pM N-803 for 6 additional days to allow for differentiation. Figure 42 depicts exemplary results for Enhancement of granzyme B expression of human NK cells induced by hILl8/ILl2/TxM. Here, Granzyme B expression of pre-activated human NK cells was measured (16 hr, 50 nM). Killing capability is exemplarily shown in Figure 43. Here it can be seen that hILl8/ILl2/TxM induces similar NK cell killing as separate cytokines. Here, cells were primed with cytokines or the TxM for a time and at concentrations as indicated. Similarly, Figure 44 depicts exemplary results demonstrating cytotoxicity and ADCC activity of hILl8/ILl2/TxM-activated human NK cells against SW1990 pancreatic cancer cells. Fresh NK cells were mixed with SW1990 cells for 40h at E:T of 2:1. aTF = 0.1 nM. N- 803 or hILl8/ILl2/T xM = 50 nM. [00131] Figure 45 depicts further exemplary results for h2*ILl8/TxM IL15 activity. Here, IL2/15 dependent 32D-IL2/15RP cells (10⁴) were stimulated for 3 days with h2*ILl8/TxM or N-803 and cell proliferation was assessed using PrestoBlue. The EC₅₀ of IL15 in

h2*ILl8/TxM is 36.8 pM. n=4 from 2 experiments. Figure 46 depicts exemplary results for a h2*ILl8/TxM ELISA (ILl8:ILl5). ELISA was performed with h2*ILl8/TxM using aILl8 capture and aILl5 detection. Figure 47 depicts exemplary results for h2*ILl8/TxM stimulation of aNK cells. Here, aNK cells were treated with TxM for 40 hours. Finally, exemplary stimulation of T cells was performed using a 4-1BB Megaligand using pp65. The results are depicted in Figure 48, and the amino acid sequence for the megaligand is also shown in Figure 48.

[00132] In yet a further set of experiments, the inventors established that a chimeric molecule complex comprising an IL15 portion, an IL7 portion and an IL21 portion (IL- 7/15/21 superkine (SK)) can expand antigen specific CD8 T cells to a significantly treater extent than IL-2 or a combination of separate IL-2, IL-15, and IL-7 alone. To that end, monocyte derived dendritic cells (MoDC) of a healthy, HLA-A2+ subject were pulsed with pp65 peptide pool (composed of 138 peptides l5aa long with l laa overlap) overnight (37 °C). Unbound peptide was washed away and MoDC were added to autologous T cells (5:1 ratio T cells to DC) and cultured at 37 °C for five days. Activated T cells were enriched using density gradient medium (Ficoll), then split into three cultures, T cells propagated in either IL-2 (6000IU/ml), a cocktail of IL- 2/7/15, or IL-7/15/21 superkine (SK) for approximately two weeks. The cells were harvested, labeled with a CD8 Ab as well as HLA-A2-NLV- dextran (Immudex) to identify antigen reactive CD8 T cells. The frequency of dextramer labeled cells, an indicator of antigen-reactive CD8 T cells, was higher in the cells cultured with superkine as is shown in the schematic illustration of Figure 49.

[00133] The left side of Figure 49 shows the above workflow, while the middle of Figure 49 depicts FACS results for the three different experimental set-ups (/._<?. , IL-2 alone, top; individual IL-2 and IL-7 and IL-15 combined, middle; SK, bottom). As can be readily seen from the results in the bar graph on the right, stimulation of the cells with SK has superior activity with respect to the expansion of antigen specific CD8 T cells.

[00134] When starting from T cells of a kidney tumor similar results were obtained as is shown in the following. First, T cell phenotypes were determined from a single cell suspension of a kidney tumor and cells waere labeled with an Antibody to CD3 to detect the frequency of T cells in the tumor. As can be seen from the FACS scan in Figure 50, normal sized T cells were present at a larger fraction that small T cells. T cells from the tumor were then split in two batches and either batch was cultured with IL-7/15/21 superkine (SK) or with high dose IL-2 (6000 U/ml), which is an amount of growth factor typically used.

Notably, more T cells overall (71.0%) and a higher frequency of CD8 T cells (71.6%) was identified in the cells cultured with SK as compared to those cultured with IL-2 (47.1% overall T cells, 51.9% CD8 T cells) as can be seen from the results presented in Figure 51.

[00135] The inventors then investigated that effect of tumor infiltrating lymphocytes on tumor cells in vitro. To that end, tumor cells were cultured and remaining adherent cells after exposure to the activated tumor infiltrating lymphocytes were e amined in the microscope.

As can be seen from Figure 52, the T cells grown in SK (left) yielded no attached tumor cells whereas the cells grown in IL-2 clearly had residual cells that grew well.

[00136] The inventors further developed a rapid and reliable method to generate/identify neoepitope reactive T cells from the blood of a cancer patient, and Figure 53 shows an exemplary schematic for such method. Here, mononuclear dendritic cells are isolated and pulsed overnight with an exemplary antigen (here pp65 peptide pool as described above or pp65 expressed in a bacterial clear coli system; of course, it should be recognized that any suitable antigen or antigen mixture can be used). Dendritic cells were then harvested and co cultured with T cells (selected by CD3 selection). Antigen reactive T cells were enriched via Ficoll separation. Different cytokines (IL-2 alone; mixture of IL-2, IL-15, and IL-7; and SK IL-7/15/21) were then used to identify optimal stimulation. After two weeks antigen reactive T cells were identified and harvested, which were subsequently used in an ELISPOT assay to determine the frequency of antigen reactive T cells. More specifically, a model antigen (CMV pp65) was used to test the expansion of T cells from the blood of a healthy donor using pp65 cloned into LPS-deficient E. coli (aka ClearColi or CC). Monocyte-derived dendritic cells (MoDC), derived from the blood of a healthy HLA-A2+ subject were pulsed with peptide or CC expressing pp65 (overnight at 37 °C). Unbound peptide/CC was washed from the MoDC, then autologous T cells were combined with the antigen-pulsed MoDC for five days at 37 °C. Antigen- specific T cells were enriched using a density gradient solution and centrifugation, then cells were split into three groups, to be cultured with IL-2

(6000U/ml), a cocktail of IL-2, IL-7, and IL-15, or a superkine (IL-7/15/21). Cells were cultured for an additional two weeks, then evaluated for antigen reactive T cells using a peptide loaded dextramer (HLA-A2 loaded with pp65 peptide 495-503, aka NLV peptide).

[00137] T cells (isolated from peripheral blood) with reactivity to neoepitopes might be present at a frequency that is detectable directly from the blood using this approach.

Alternatively, the neoepitope reactive T cells may be in such a low frequency in the blood that they may require expansion before testing in the in vitro transcription/translation system. Exemplary results are shown in Figure 54 (left two columns peptide pool, right two columns ClearColi expressed antigen). As can be seen from the results, the tested antigen was readily detectable using this system, with a stronger signal (or higher sensitivity) where bacterial expression was used to present the antigen. Figure 55 shows an exemplary graph that shows higher frequency of CD8 positive antigen reactive T cells using E.coli expressed antigen and/or IL-7/15/21 superkine. Alternatively, it should be appreciated that the neoepitopes need not necessarily be presented as peptide mixtures or expressed in a cell. Indeed, it should be noted that an in vitro transcription/translation system may also be employed. For example, a DNA oligo may be designed that includes a (universal) primer annealing region to start transcription, which is followed by a ribosome binding site upstream of a start codon that is in frame of a sequence encoding the neoantigen (typically 7-30 amino acids). Most commonly, such synthetic oligo will terminate with a stop codon. Advantageously, such DNA template requires no enzymatic manipulation, purification, cloning. Indeed, one only needs to add annealed oligos to an expression mix, and 2 hour reaction time (which can be done with various known transcription/translation systems).

[00138] Where desired, various functional elements such as flags and other tags can be included. For example, where it is desired to quantitate the peptide it is contemplated that a tag can be added to help quantify the expressed peptide. Among other suitable tags, a tetracysteine tag reactive fluorophore can be used as is exemplarily shown in Figure 56. Such tags may be particularly useful where the peptide is produced in an in vitro transcription and translation system. Figure 57 shows exemplary results for short term T cell lines (established using ClearColi-expressing pp65) that respond to in vitro transcribed and translated pp65 peptide.

[00139] The approach takes advantage of an in vitro transcription and translation system that can generate an antigenic peptide from an oligonucleotide (slide 4). In this case, we propose to use a strain of E. coli that lacks LPS as a vector for the neoepitopes. See slides 2 and 3. Remaining T cells were maintained in culture for ten more days, then evaluated for antigen reactive T cells using ELISPOT with the antigen supplied by the the in vitro transcription and translation system described in slides 4 and 5. The results of the ELISPOT are shown in slide 6.

[00140] It should be apparent to those skilled in the art that many more modifications besides those already described are possible without departing from the inventive concepts herein. The inventive subject matter, therefore, is not to be restricted except in the scope of the appended claims. Moreover, in interpreting both the specification and the claims, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms“comprises” and“comprising” should be interpreted as referring to elements, components, or steps in a non-exclusive manner, indicating that the referenced elements, components, or steps may be present, or utilized, or combined with other elements, components, or steps that are not expressly referenced. As used in the description herein and throughout the claims that follow, the meaning of“a,”“an,” and“the” includes plural reference unless the context clearly dictates otherwise. Also, as used in the description herein, the meaning of“in” includes“in” and“on” unless the context clearly dictates otherwise. Where the specification claims refers to at least one of something selected from the group consisting of A, B, C .... and N, the text should be interpreted as requiring only one element from the group, not A plus N, or B plus N, etc.

Claims

CLAIMS What is claimed is:

1. A kit, comprising:

a chimeric molecule complex comprising an IL15 portion, and at least one of an IL7 portion and an IL21 portion; and

an agonist for a TNF superfamily receptor.

2. The kit of claim 1 wherein the chimeric molecule complex is a TxM molecule.

3. The kit of claim 2 wherein the TxM molecule comprises the IL7 portion and the IL21 portion.

4. The kit of any one of claims 1-3 wherein the agonist for the TNF superfamily receptor is a multimeric 4-1BBL ligand or an agonistic antibody against 4-1BB.

5. A kit, comprising:

a chimeric molecule complex comprising an IL15 portion, and at least one of an IL18 portion, and an IL12 portion; and

an agonist for a TNF superfamily receptor.

6. The kit of claim 5 wherein the chimeric molecule complex is a TxM molecule.

7. The kit of claim 6 wherein the TxM molecule comprises the IL18 portion and the IL12 portion.

8. The kit of any one of claims 5-7 wherein the agonist for the TNF superfamily receptor is a multimeric 4-1BBL ligand or an agonistic antibody against 4-1BB.

9. A method of generating memory T cells, comprising

contacting a plurality of lymphocytes with

(a) a chimeric molecule complex comprising an IL15 portion, and a second portion comprising

i. a portion selected from an IL7 portion, an IL21 portion, and a combination thereof, or

ii. a portion selected from an IL18 portion, an IL12 portion, and a combination thereof; and (b) an agonist for a TNF superfamily receptor;

wherein the step of contacting is performed under conditions and for a time sufficient to allow differentiation of non-memory T cells to memory T cells and to allow proliferation of the memory T cells.

10. The method of claim 9 wherein the memory T cell is a TSCM cell.

11. The method of any one of claims 9-10 wherein the plurality of lymphocytes are contacted with at least one of the chimeric molecule complex and the agonist for a TNF superfamily receptor in vitro.

12. The method of any one of claims 9-11 wherein the plurality of lymphocytes are contacted first with the chimeric molecule complex, and after a plurality of hours, with the agonist for a TNF superfamily receptor.

13. The method of any one of claims 9-12 wherein the chimeric molecule complex comprises the IL7 portion and the IL21 portion.

14. The method of any one of claims 9-12 wherein the chimeric molecule complex comprises the IL18 portion and the IL12 portion.

15. The method of any one of claims 9-14 wherein the agonist for the TNF superfamily

receptor is a multimeric 4-1BBL ligand or an agonistic antibody against 4-1BB.

16. The method of any one of claims 9-15 wherein the plurality of lymphocytes further

comprises NK cells.

17. The method of any one of claims 9-15 wherein the plurality of lymphocytes further

comprises recombinant CAR-T cells.

18. The method of any one of claims 9-17 wherein the plurality of lymphocytes are contacted with at least one of the chimeric molecule complex and the agonist for a TNF superfamily receptor in vivo.

19. The method of claim 18 wherein the chimeric molecule complex comprises the IL7

portion and the IL21 portion and/or wherein the chimeric molecule complex comprises the IL18 portion and the IL12 portion, and wherein agonist for the TNF superfamily receptor is a multimeric 4-1BBL ligand or an agonistic antibody against 4-1BB.

20. The method of any one of claims 9-19, further comprising a step of administering the proliferated memory T cells to a patient.

21. A memory T cell generated by the method of any one of claims 9-19.

22. Use of the memory T cell of claim 21 to treat cancer in a patient in need thereof.

23. The use of claim 22, wherein the cancer is pancreatic cancer.

24. A method of generating antigen reactive short term T cells, comprising:

isolating dendritic cells from a donor blood sample, and exposing the dendritic cells to an antigen of interest to generate exposed dendritic cells;

co-culturing the exposed dendritic cells with T cells, and enriching antigen reactive T cells from the co-culture;

exposing the antigen reactive T cells with a IL-7/15/21 superkine in an activation culture; and

isolating antigen reactive short term T cells from the activation culture.

25. The method of claim 24 wherein the antigen of interest comprises a tumor antigen of the donor.

26. The method of claim 24 further comprising detecting activity of the antigen reactive short term T cells in an ELISPOT assay.

27. A method of expanding antigen reactive CD8 T cells, comprising:

isolating dendritic cells from a donor blood sample, and exposing the dendritic cells to an antigen of interest to generate exposed dendritic cells;

co-culturing the exposed dendritic cells with T cells, and enriching antigen reactive T cells from the co-culture;

exposing the antigen reactive T cells with a IL-7/15/21 superkine in an activation culture; and

isolating antigen reactive T cells from the activation culture.

28. The method of claim 27 wherein the dendritic cells are from the donor.

29. The method of claim 27 wherein the antigen of interest comprises a tumor antigen of the donor.

30. The method of claim 27 further comprising administering the antigen reactive T cells to the donor.