Kenderian, Saad S.

Abstract B cell maturation antigen (BCMA)-targeted chimeric antigen receptor T (CART) cell therapies have emerged as a potentially curative treatment for multiple myeloma (MM). However, 50-70% of patients relapse within 3 years. The MM tumor microenvironment (TME) is highly immunosuppressive with an abundance of bone marrow (BM) cancer-associated fibroblasts (CAFs), which contribute to BCMA-CART cell failure. Therapeutic targeting of BM-CAFs has been beset by 1) an incomplete understanding of CAF interactions with CART cells and other BM cells and 2) toxicity risks due to prolonged depletion of healthy fibroblasts and inefficient drug delivery to the BM. To elucidate how BM-CAFs drive BCMA-CART resistance, we performed single cell RNA sequencing (scRNAseq) of baseline BM aspirates from responders (R, n = 6) and non-responders (NR, n = 6) to BCMA-CART cell therapy. NR BMs were enriched for inhibitory CAFs (iCAFs) that highly express fibroblast activation protein (FAP). Spatial analyses of BM cells revealed greater iCAF-iCAF and iCAF-inhibitory macrophage interactions in NR. These findings suggest that FAP+ iCAFs in NR are abundant in the BM and exhibit significant interactions with immunosuppressive macrophages. Next, we sought to validate our scRNAseq data in preclinical models. Xenograft mice were intravenously (IV) engrafted with luciferase+ (luc+) BCMA+ OPM2 cells and/or patient-derived (pd) BM-CAFs (1x106 each). Following engraftment, mice received 1x106 BCMA-CART cells (IV). BCMA-CART cells eradicated tumors in mice engrafted with OPM2 cells, but not in mice engrafted with pdBM-CAFs and OPM2 cells (p<0.01). In NSG and humanized mice (NSG mice engrafted with human CD34+ cells in the first 48 hours of life), tumor growth was significantly accelerated in the presence of pdBM-CAFs (NSG: p<0.01; humanized: p<0.001). BCMA-CART, OPM2, and pdBM-CAF co-cultures also showed significant increases in immunosuppressive cytokines including IL-1Rα, IL-4 and IL-10 and inhibited CART cell proliferation compared to co-cultures without pdBM-CAFs (IL-1Rα, IL-10: p<0.01; IL-4: p<0.001; proliferation: p<0.05). To overcome CAF-mediated CART dysfunction, we engineered BCMA-CART cells to secrete molecules to target BM FAP+ iCAFs and remodel the MM-TME (STriKEs). We designed two BCMA-CART-STriKE strategies: 1) a tri-specific composed of IL-15, anti-FAP, and anti-CD16 molecules to activate innate immune cells to kill CAFs and 2) bi-specific anti-FAP, anti-SIRPα molecules linked to an antibody Fc region (Fc fusion) targeting the CD47/SIRPα axis to induce phagocytosis of FAP+ iCAFs. We hypothesized that BCMA-CART-STriKEs would surmount CAF-mediated immunosuppression without toxicity. We generated BCMA-CART-STriKEs from healthy human donor T cells. CAR expression, STriKE secretion, and binding were confirmed via flow cytometry. Co-cultures of BCMA-CART-STriKEs, donor-matched macrophages, OPM2 cells or BCMA+ MM1S cells, and pdBM-CAFs showed significantly higher OPM2 or MM1S killing (p<0.001), immune cell activation (p<0.01), and phagocytosis (p<0.05), compared to control BCMA-CART cells secreting IL-15 or Fc fusions targeting only FAP or only SIRPα. Next, we assessed BCMA-CART-STriKE efficacy in vivo. NOD-SCID-γ-/- (NSG) mice received 1x106 pdBM-CAFs and 1x106 luc+ OPM2 or luc+ MM1S cells via IV or subcutaneous (SQ) injection. Mice were randomized after confirming tumor engraftment (via bioluminescence imaging for IV or tumor caliper for SQ) and received IV treatment with 1x106 BCMA-CART-STriKEs (or controls) and 5x105 donor-matched macrophages or CD16+ immune cells. BCMA-CART-STriKE-treated mice achieved total tumor clearance. Treatment with BCMA-CART cells secreting IL-15 led to lethal weight loss despite tumor clearance while BCMA-CART cells secreting Fc fusions targeting only FAP or SIRPα failed to eradicate tumors. Lastly, we assessed the potential toxicities of BCMA-CART-STriKEs in humanized mice engrafted with OPM2 cells and pdBM-CAFs (1x106 each). CART cells expressing BCMA and FAP CARs (BCMA-FAP CART) served as a control. BCMA-FAP CART treatment resulted in significant toxicity and weight loss while BCMA-CART-STriKE treatment did not (p<0.01). In summary, we have identified iCAF interactions with immunosuppressive immune cells as key contributors to BCMA-CART cell failure and designed novel BCMA-CART-STriKEs that overcome this failure in MM preclinical models. These findings support the translation of BCMA-CART-STriKEs to a phase I clinical trial.

Abstract Chimeric antigen receptor (CAR) T-cell therapy has revolutionized the treatment of cancer. However, the durable response to this therapy remains low, and there is a risk of moderate-to-severe toxicities following treatment that requires close monitoring. Over the past decade, we have learned that cytokines play an important role in mediating CAR T cell–associated toxicities and efficacy. As such, cytokine modulation has become a popular area of investigation to improve therapeutic responses. Although the relationship of many cytokines with CAR T-cell therapy has been investigated, several recent studies suggest paradoxical roles for cytokines such as IFNγ, IL2, IL4, and IL10 in CAR T-cell response and toxicity. In this review, we summarize the history of these cytokines in immunotherapies, detail the contexts in which these cytokines have been beneficial or harmful in the context of CAR T-cell therapy, and discuss factors that may be dictating their pleiotropy.

Abstract Description B7-H3 is a member of the B7 superfamily, which includes co-inhibitory receptors such as B7-H1/PD-L1 and co-stimulatory receptors such as B7-1/CD80 and B7-2/CD86. B7-H3 is upregulated by almost all types of tumors and is associated with worse prognosis. However, the exact mechanism by which B7-H3 suppresses anti-tumor immune response remains to be discovered as its ligand has yet to be identified. Sialic acid modifications are also increased in many cancers and are similarly associated with poor outcome. Here, we demonstrate for the first time that sialylated B7-H3 binds to the immune inhibitory receptor Siglec-9. Siglec-9 is expressed on many immune cells, including NK cells and T cells. We have shown that B7-H3 expression on target cells reduces killing by Siglec-9 expressing NK cells in vitro. We then hypothesized that antibodies that block B7-H3 binding to Siglec-9 will be an effective intervention to boost anti-tumor immunity. We have generated a high affinity anti-B7-H3 monoclonal antibody, F7, which blocks B7-H3 binding to Siglec-9. F7 functions to enhance NK cell killing of B7-H3 expressing target cells in vitro. F7 also dramatically enhances CART19 mediated killing of B7-H3 expressing Nalm6 cells in a relapse model in NSG mice in vivo. In summary, our work has demonstrated that B7-H3 binds Siglec-9, and blockade of such interaction can enhance anti-tumor immunity. Funding Sources • Mayo Clinic Center for Biomedical Discovery • NIH Glycobiology Common Fund supplement to R01CA243545-01A1 to VSS Topic Categories Tumor Immunology: Checkpoints, Prevention, and Treatment (TIPT)

Abstract BACKGROUND Glioblastoma (GBM) is a universally lethal brain cancer with limited durable responses to immunotherapy, largely due to its immunosuppressive tumor microenvironment (TME). Ephrin receptor A3 (EphA3), a receptor tyrosine kinase minimally expressed in normal tissue, is enriched on GBM cells, glioma stem cells (GSCs), and M2-like tumor-associated macrophages (TAMs). We hypothesize that dual targeting of TAMs and tumor cells by EphA3-CAR-T is a viable treatment strategy. METHODS We developed EphA3-directed second-generation CAR-T cells (EphA3-CAR-T) incorporating the single-chain variable fragment of ifabotuzumab, an anti-EphA3 monoclonal antibody, achieving >90% transduction efficiency. We evaluated antitumor efficacy in vitro; the functional activities of EphA3-CAR-T on GSCs were assessed via selective cytotoxicity, T cell proliferation, degranulation assays, and depletion of M2-like macrophages. In vivo, 1 × 10⁶ EphA3⁺ U251 MG cells and/or 1 × 10⁶ ex vivo–polarized M2-like macrophages were implanted in the flanks of NOD-SCID-γ⁻/⁻ mice. Comparative studies with EGFRvIII-directed CAR-T cells were also performed. RESULTS Immunohistochemistry confirmed EphA3 expression across both tumor and immune compartments in patient-derived specimens. EphA3-CAR-T cells demonstrated potent, selective cytotoxicity (p<0.001), T cell proliferation (p<0.001), and CD107α degranulation (p<0.001) against EphA3⁺ GBM and GSCs, while also reversing TAM-mediated immunosuppression via depletion of ex vivo–polarized M2-like macrophages (p<0.05), compared to other validated CAR-Ts. In vivo, EphA3-CAR-T significantly reduced tumor volume (p<0.001) and improved survival (p<0.01), in the presence of TAMs. In contrast, EGFRvIII CAR-T cells were inactivated under the same TME conditions. EphA3-CAR-T cells remained active in models co-engrafted with macrophages, suggesting preserved function within the immunosuppressive TME. CONCLUSIONS EphA3-CAR-T therapy addresses a major barrier to CAR-T success in GBM: TME-mediated immunosuppression. This tumor- and TME-targeting approach demonstrated robust preclinical activity. These findings support the clinical translation of EphA3-CAR-T cells as a next-generation immunotherapy for GBM.

Abstract Front-line pharmaceutical interventions for treating acute graft-versus-host disease (GVHD) are not uniformly effective and have toxic side effects. Myeloid-derived suppressor cells (MDSCs) are a heterogeneous population of immature myeloid cells with potent in vitro and in vivo immunosuppressive functions. Clinical translation of in vitro–generated MDSCs has been limited because of requirements for multiple, high infusion doses, the relatively low yield from peripheral blood–sourced MDSCs (PB-MDSCs), and inconsistent product quality. To circumvent these obstacles, we developed a methodology to generate MDSCs using human induced pluripotent stem cell (iPSC)–derived CD34+ cells. Compared with PB-MDSCs, iPSC-derived MDSCs (iMDSCs) shared similar morphology, phenotype, and suppressive function. We found that the CD14+ iMDSC subset possessed the highest suppressor function. In previous studies, we reported that MDSCs transferred into mice with GVHD lost suppressor function because of inflammasome activation and immature myeloid cell maturation. In striking contrast to human PB-MDSCs, we show herein that iMDSCs retained 95% of suppressor function in vitro despite exposure to lipopolysaccharide (LPS) plus adenosine triphosphate (ATP), which are stimuli that activate the inflammasome via danger-associated molecular patterns released during early posttransplant conditioning and GVHD-induced injury. In an in vivo xenogenic GVHD model with PB mononuclear cells, iMDSCs significantly increased recipient survival without loss of antileukemia effects. iMDSC RNA sequencing and gene knockdown studies revealed that the maintenance of the purine metabolizing enzyme, phosphoglycerate dehydrogenase, during LPS plus ATP treatment, was linked to iMDSC inflammasome resistance. Taken together, these findings provide a platform for translating in vitro–generated, off-the-shelf iMDSCs into the clinic for suppressing a spectrum of adverse immune responses, including GVHD.

SummaryIn individuals with high‐count monoclonal B‐cell lymphocytosis (MBL), we investigated if lymphadenopathy or splenomegaly found by imaging was associated with shorter time to first chronic lymphocytic leukaemia (CLL) therapy (TTFT) and overall survival (OS). Individuals with MBL seen at Mayo Clinic (2002–2019) were retrospectively divided into three cohorts based on imaging studies within 1 year of diagnosis: no imaging studies (Cohort A); imaging with no evidence of lymphadenopathy/splenomegaly (Cohort B); imaging with evidence of lymphadenopathy/splenomegaly (Cohort C). We compared baseline characteristics, TTFT and OS across the MBL cohorts to a cohort of individuals with small lymphocytic lymphoma (SLL). A total of 1078 patients were included: 640 with MBL and 438 with SLL. Compared to Cohort B, individuals in Cohort C were more likely to have unmutated immunoglobulin heavy chain variable region (IGHV) (43% vs. 25% p = 0.016), high‐risk fluorescence in situ hybridization (FISH) (del17p and del11q in 15% vs. 4%, p = 0.038) and higher expression of CD38 (31% vs. 16%; p = 0.011). After adjusting for sex and CLL‐International Prognostic Index (IPI), lymphadenopathy/splenomegaly was associated with a shorter TTFT (hazard ratio [HR] = 2.04, 95% confidence interval [CI]: 1.02–4.04, p = 0.042) but not OS (HR = 1.09, 95% CI: 0.62–1.92, p = 0.775). Lymphadenopathy/splenomegaly on imaging in individuals with high‐count MBL is associated with a more unfavourable risk profile and shorter time to first CLL‐directed therapy.

Abstract Allogeneic stem cell transplant (ASCT) remains the only curative option in chronic myelomonocytic leukemia (CMML). We retrospectively analyzed 138 CMML patients who underwent ASCT at the Mayo Clinic. Patients who transitioned to ASCT while in chronic phase (Group A) displayed superior post-transplant survival (PTS), compared to those in whom ASCT was performed after blast transformation (BT; Group B) (median 95 vs. 16 months; p=0.01). In Group A, PTS was superior in patients with <5% bone marrow (BM) blasts at time of ASCT (median 164 vs. 13.5 months; p=0.01). Other predictors of superior PTS included day-100 BM blast <5% or normal cytogenetics (median 164 vs 18 months; p=0.01) or presence of chronic graft-versus-host-disease (GVHD; median 164 vs. 26 months; p=0.01). Pre-ASCT hypomethylating agent exposure (HR=2.03; p=0.03), and receiving more than one line of pre-ASCT chemotherapy (p=0.01) predicted inferior PTS. In multivariable analysis, predictors of superior GVHD-free and relapse-free survival (GRFS) included the use myeloablative conditioning and the absence of morphologically or cytogenetically apparent disease at day-100. The use of post-transplant cyclophosphamide (PTCy) was associated with higher cumulative incidence of relapse (p=0.02) and numerically inferior PTS (p=0.1). Group B patients also appeared to benefit from achieving BM blast <5% at time of ASCT (p=0.4) as well as at day-100 (p=0.01), in terms of PTS, while full chimerism and normal cytogenetics at day-100 were associated with superior GRFS. These observations support the value of ASCT in CMML, especially if performed prior to BT and in the presence of <5% BM blasts at time of ASCT. Additionally, the observed detrimental impact of PTCy requires additional studies to confirm and investigate the underlying mechanisms.

7037 Background: TRANSCEND CLL 004 (NCT03331198) is a phase 1/2, open-label, nonrandomized, multicohort study in adult patients (pt) with R/R chronic lymphocytic leukemia/small lymphocytic lymphoma with liso-cel mono and liso-cel + ibrutinib combo cohorts. Given the study design, we performed PS matching/weighting to adjust for differences in key covariates between cohorts to compare efficacy and safety. Methods: Here, PS is the probability of receiving combo given the covariates and estimated by logistic regression. Covariates were hemoglobin, LDH, number of prior therapies, platelets, prior BTKi + venetoclax exposure, R/R on prior BTKi, and bulky disease for efficacy; and ALC, bulky disease, CRP, ferritin, and number of prior therapies for safety. PS balancing methods were optimal 1:1 matching, average treatment effect for treated pts (ATT) full matching, and ATT inverse probability of treatment weighting (IPTW). Efficacy endpoints were response (CR rate, ORR) and time to event endpoints (DOR, PFS, OS). Safety endpoints were proportions of pts with any-grade (gr)/gr ≥ 3 cytokine release syndrome (CRS) and investigator-identified neurological events (NE). Results: Fifty-one and 88 efficacy-evaluable pts were treated at dose level 2 (100 × 10 6 CAR + T cells) in the combo and mono cohorts, respectively. Before matching/weighting on PS, odds of achieving CR and overall response were statistically significantly higher for combo (Table). DOR did not differ between cohorts. For PFS and OS, HRs were lower with combo vs mono (not statistically significant [NS]). Odds of gr ≥ 3 CRS/NE were lower with combo vs mono (NS). After matching/weighting, odds of CR (except IPTW ATT) and overall response remained statistically significantly higher in combo. HRs for PFS and OS and ORs for gr ≥ 3 CRS/NE were numerically lower with combo vs mono (NS, except IPTW ATT for PFS). HRs for DOR were less but close to HR of 1. Conclusions: The liso-cel + ibrutinib combo demonstrated a trend for better efficacy and safety vs liso-cel mono, with statistically significant differences for CR rate and ORR. Clinical trial information: NCT03331198 . Efficacy and safety outcomes: combo vs mono. No adjustment ATT optimal 1:1 IPTW - ATT CR rate, OR (95% CI) 2.07* (1.01–4.28) 2.40* (1.05–5.66) 2.00(0.89–4.61) ORR, OR (95% CI) 3.96* (1.68–10.51) 3.43* (1.33–9.71) 4.18* (1.63–11.78) DOR, HR (95% CI) 0.98(0.51–1.87) 0.88(0.43–1.78) 0.96(0.50–1.85) PFS, HR (95% CI) 0.62(0.37–1.02) 0.61(0.35–1.05) 0.59* (0.35–0.98) OS, HR (95% CI) 0.70(0.38–1.28) 0.73(0.37–1.43) 0.66(0.36–1.22) CRS, OR (95% CI) 0.63(0.26–1.56) 0.67(0.24–1.84) 0.69(0.24–1.88) Gr ≥ 3 CRS, OR (95% CI) 0.23(0.01–1.35) 0.49(0.02–5.28) 0.26(0.01–1.94) NE, OR (95% CI) 0.84(0.41–1.68) 0.92(0.40–2.08) 1.10(0.48–2.52) Gr ≥ 3 NE, OR (95% CI) 0.56(0.19–1.45) 0.58(0.16–1.89) 0.68(0.19–2.30) *Statistically significant difference at 5% level.