ALBERT

Description: We have previously shown that marrow infiltrating lymphocytes (MILs) from myeloma patients proliferate upon stimulation with anti-CD3/CD28 beads more efficiently and exhibit greater specificity to autologous tumor than their peripheral blood lymphocyte (PBL) counterparts. In attempting to dissect the mechanisms responsible for greater tumor specificity of MILs, we examined the frequency and function of the putative CD4+/CD25+ regulatory T cells (Tregs) in both compartments. Phenotypic and functional differences in the CD4+/CD25+ populations were examined in MILs and PBLs from 12 multiple myeloma patients and 3 normal donors. CD4+/CD25+ cells were stained for the suppressive marker: FoxP3; or the activation markers: CD40L, CD71 (transferrin receptor). CD4+/CD25+ MILs of myeloma patients are CD25lo, FoxP3lo and predominantly express activation markers. They expand more readily upon anti-CD3/CD28 stimulation, produce high amounts of the Th1 cytokines: IFNγ and IL-2, and low amounts of IL-10. Importantly, they fail to suppress the tumor specific T cell proliferation. These findings are all consistent with an activated T cell phenotype. In contrast, CD4+/CD25+ cells from PBLs of myeloma patients are CD25hi, FoxP3+, primarily produce IL-10 and IL-4 and markedly suppress T cell expansion in a tumor specific assay, suggestive of a regulatory T cell phenotype. Although MILs possess a lower number of FoxP3+ cells as compared to PBLs, the higher MFI may indicate the presence of a small yet highly regulatory T cell population. Interestingly, MILs from normal donors possess a more regulatory phenotype then their myeloma counterparts as determined by high FoxP3 and low CD40L and CD71 expression as well as reduced expansion upon activation. Taken together, these data underscore major differences in the immuno-inhibitory pathways present in blood as compared to marrow. Notably, MILs of myeloma patients demonstrate a paucity of Tregs. These differences may explain the disparities seen in the tumor-specificity of T cells from these two compartments. The ability to expand a T cell population with fewer endogenous Tregs and heightened tumor-specificity may have significant implications for the implementation of adoptive immunotherapy. %CD25 (MFI) %FoxP3 (MFI) %CD40L %CD71 Fold Expansion Ability to Suppress Myeloma CD4/CD25 MILs 4.7 (29.6) 2.2 (2932.1) 88.8 90.1 12.2 − CD4/CD25 PBLs 19.1 (115.2) 52.3 (163.7) 15.7 19.6 2.5 +++++ Normal CD4/CD25 MILs 14.0 (56.8) 70.1 (82.7) 37.3 6.4 3.3 +++ CD4/CD25 PBLs 24.0 (19.4) 30.9 (261.8) 12.1 4.2 4.0 ++

Description: Hallmarks of effective adoptive T cell therapy include: 1) the ability to access and expand T cells of interest 2) measurable anti-tumor efficacy of these cells upon reinfusion and 3) T cell persistence over time. In examining the long-term follow-up of patients treated with MILs, we show evidence of persistence of adoptively transferred MILs for over five years post transfer. In a clinical trial conducted in 2007, Prevnar naïve myeloma patients were immunized against the modified diphtheria toxin, CRM197 with the Prevnar vaccine administered 2 weeks prior to collecting marrow infiltrating lymphocytes (MILs) from bone marrow. Patients received a CD34-selected (T cell depleted) autologous stem cell transplant (SCT) and had activated and expanded MILs infused 3 days later. As such, 99% of infused T cells were Prevnar-primed, ex vivo expanded MILs. This approach enabled us to track the fate of the non gene-modified MILs by examining CRM197 specific T cells. As originally reported, persistence at one year of both CRM197 and myeloma antigen-specific T cells correlated with clinical responses.1 Long-term follow up was performed in a subset of the patients that had undergone bone marrow biopsies for clinical purposes. The results showed that patients who had been in a complete response at the end of the clinical trial maintained a greater number of both CRM197 and myeloma antigen-specific T cells over time for up to 7 years after the clinical trial. Patients who did not achieve a complete response and subsequently relapsed showed a loss of both CRM197 as well as myeloma antigen- specific T cells in most cases. Considering that the patients received a T cell depleted stem cell graft, the CRM197-specific T cells that were measured were likely a result of the Prevnar-primed MILs. From these results, we conclude that: 1) adoptively transferred CRM-197 specific MILs showed persistence up 7 years following their infusion. 2) the magnitude of CRM197 and myeloma antigen-specific T cell responses correlated with clinical responsiveness. Taken together, these data demonstrate several unique attributes of MILs which are critical for successful adoptive T cell therapyAdoptive transfer of activated marrow-infiltrating lymphocytes induces measurable antitumor immunity in the bone marrow in multiple myeloma. Noonan KA, Huff CA, Davis J, Lemas MV, Fiorino S, Bitzan J, Ferguson A, Emerling A, Luznik L, Matsui W, Powell J, Fuchs E, Rosner GL, Epstein C, Rudraraju L, Ambinder RF, Jones RJ, Pardoll D, Borrello I. Sci Transl Med. 2015 May 20;7(288). Disclosures Noonan: WindMIL Therapeutics: Employment, Equity Ownership, Patents & Royalties. Rudraraju:WindMIL Therapeutics: Employment, Equity Ownership. Lutz:WindMIL Therapeutics: Employment, Equity Ownership. Borrello:BMS: Honoraria, Research Funding; Celgene: Honoraria, Research Funding, Speakers Bureau; WindMIL Therapeutics: Equity Ownership, Patents & Royalties, Research Funding.

Description: The bone marrow of myeloma patients is a reservoir of marrow infiltrating lymphocytes (MILs) that upon activation with anti-CD3 and anti-CD28 antibodies show significant tumor specificity against both mature myeloma plasma cells and their clonotypic precursors, whereas activated peripheral blood lymphocytes (PBLs) fail to demonstrate any tumor specificity. Here we examine the major differences between MILs and PBLs of both myeloma patients and normal donors. In myeloma, CD4+/CD25+ MILs lack a regulatory T-cell (Treg) phenotype as shown by the absence of FoxP3 expression and the inability to suppress tumor-specific proliferation of effector T cells. In contrast, the CD4+/CD25+ PBLs express FoxP3 and suppress tumor specific proliferation. Analysis of MILs and PBLs from normal individuals reveals the exact opposite (more Tregs in the marrow than blood). Considering the abundance of plasma cell-derived IL-6 in myeloma marrow, we hypothesize that IL-6 skewing of MILs towards the effector Th17 phenotype could explain the reciprocal paucity of Tregs in myeloma and relative abundance in normal bone marrows. Intracellular staining shows increased expression of IL-17 in myeloma CD4 MILs (2.7%) as compared to either myeloma PBLs (0.02%) or normal MILs (0.7%). The Th17 cytokine secretion pattern also appears more pronounced in myeloma-derived MILs as compared to either myeloma PBLs or normal MILs respectively (see table). Th17 T cells have recently been implicated in osteoclastogenesis of rheumatoid arthritis. In fact, bone marrow serum IL-17 levels correlate with the degree of lytic bone disease. In experiments utilizing M-CSF and RANK-L as the osteoclast differentiation medium, we show that the addition of IL-17 to the culture medium increases the number of mature osteoclasts by 2.3 fold whereas γIFN reduces mature osteoclast numbers by 80%. These findings explain the profound differences seen in MILs of normal and myeloma patients and implicate Th17 MILs in the generation of lytic bone disease. Activation of myeloma MILs with anti-CD3/CD28 skews the population from Th17 to a γIFN-producing Th1 phenotype. As such, targeting Th17 T cells or utilizing adoptive T cell therapy with activated, γIFN-producing MILs may represent a novel therapeutic mechanism to target osteoclastogenesis and reduce bone disease in myeloma. Th-17 Cytokine Profile N=56 MM MILs MM PBLs NL MILs IL-6 (pg/ml) 30 3.7 6.5 IL-23 (pg/ml) 246 19.6 35.9 IL-17 (pg/ml) 12.2 0.4 5.1