Allogeneic CAR Research Articles

Preclinical activity of allogeneic SLAMF7-specific CAR T-cells (UCARTCS1) in multiple myeloma

BackgroundAutologous BCMA-specific CAR T-cell therapies have substantial activity in multiple myeloma (MM). However, due to logistical limitations and BCMAlow relapses, there is a need for alternatives. UCARTCS1 cells are ‘off-the-shelf’...

"Off-The-Shelf" allogeneic chimeric antigen receptor T-cell therapy for B-cell malignancies: current clinical evidence and challenges.

Chimeric antigen receptor T-cell therapy (CAR T) has revolutionized the treatment landscape for hematologic malignancies, notably B-cell non-Hodgkin lymphoma (B-NHL) and B-cell acute lymphoblastic leukemia (B-ALL). While autologous CAR T products have shown remarkable efficacy, their complex logistics, lengthy manufacturing process, and high costs impede widespread accessibility and pose therapeutic challenge especially for patients in rapid need for therapy. "Off-the-shelf" allogeneic CAR T-cell therapy (alloCAR T) has emerged as a promising alternative therapy, albeit experimental to date. AlloCARTs are derived from healthy donors, manufactured by batches and stored, making them available off-the-shelf which lowers financial burden. Various gene editing techniques have been employed to mitigate graft-versus-host disease (GVHD) and host-versus-graft (HvG) to enhance alloCAR T persistence. In this review, we summarize available manufacturing techniques, current evidence, and discuss challenges faced with the use of alloCAR Ts.

The interim analysis of a first-in-human phase 1 trial of ET-901, a CRISPR edited allogeneic immune-cloaked anti-CD19 CAR-T cell therapy in patients with r/r B-NHL.

e19010 Background: Allogeneic CAR T-cell therapy faces challenges like limited persistence and host rejection. Current research has improved its viability by modifying genes like B2M, despite new challenges like NK cell targeting have emerged. Through CRISPR screening, we discovered a novel gene, named GeneX, and its role in enhancing T cell persistence against allogeneic attacks and AICD. GeneX influences glycan composition, affecting immune responses and AICD initiation. Engineered GeneXKO/TCRKO/anti-CD19 CAR T cells (ET-901) resisted allogeneic killing and maintained anti-tumor effects without causing GvHD, marking a significant advancement in CAR T-cell therapy. Methods: ET-901 is being tested in a phase 1 trial for r/r B-NHL at a single center, utilizing a 3+3 dose escalation scheme and planning for expansion at the recommended Phase 2 dose. Participants must have an ECOG status of 0–2, without a history of allo-HSCT or CAR-T therapy. The study's main goals are to assess ET-901's safety and establish the optimal Phase 2 dosage. Patients undergo a 3-day lymphodepletion with fludarabine (30-50 mg/m2/d) and cyclophosphamide (500-1000 mg/m2/d), followed by a single-dose infusion of ET-901 at escalating doses (DL1:1*10e6/kg, DL2:3*10e6/kg, and DL3:10*10e6/kg). Pharmacokinetic monitoring occurs during follow-up. Results: From September 2023, a total of six patients were enrolled in the escalation cohort, including 4 patients with LBCL and 2 with FL. In these patients (3 males), the median age was 53 years (range: 46–55 years), and the median number of prior treatment lines was 5. The median SPD was 7,650 mm2 (range: 2,700-43,400 mm2). Notably, two cases in DL2 exhibited unusual lymphoma infiltration sites: patient No.4 in the forearm muscle and patient No.5 in the spleen. Update to February 2024, we have completed the safety evaluation of Dose Level (DL) 1 and DL2. No DLTs were observed. The most common TRAEs of any grade included leukopenia (6/6), granulocytopenia (6/6), anemia (6/6), thrombocytopenia (6/6), CRS (6/6), infection (2/6), ICANS (2/6), and GvH like reaction (1/6). All patients achieved objective responses, with 2 cases of PR and 4 cases of CR (1 in DL1, 3 in DL2). The peak values of CAR+ cells count and VCN in peripheral blood were 36.0/μl (range: 14.0-48.0/μl, median: day 7), and 9.6e4 copies/μg (range: 7.7e4-1.2e5 copies/μg, median: day 7), respectively. In all patients, a second expansion of CAR-T cells was observed alongside B cell recovery. This confirms that GeneX knockout enables allo CAR-T cells to maintain long-term immune privilege while preserving their biological function. Conclusions: The study demonstrated the safety and feasibility of ET-901 in patients with r/r B-NHL. These findings highlight the potential of ET-901 as an effective therapeutic approach for improving outcomes for these patients. Clinical trial information: NCT06014073 .

CAR T cells outperform CAR NK cells in CAR-mediated effector functions in head-to-head comparison

BackgroundCAR NK cells as vehicles for engineered “off-the-shelf” cellular cancer immunotherapy have attracted significant interest. Nonetheless, a comprehensive comparative assessment of the anticancer activity of CAR T cells and CAR NK cells carrying approved benchmark anti-CD19 CAR constructs is missing. Here, we report a direct head-to-head comparison of CD19-directed human T and NK cells.MethodsWe generated CAR T and CAR NK cells derived from healthy donor PBMC by retroviral transduction with the same benchmark second-generation anti-CD19 CAR construct, FMC63.28z. We investigated IFN-γ secretion and direct cytotoxicity in vitro against various CD19+ cancer cell lines as well as in autologous versus allogeneic settings. Furthermore, we have assessed anticancer activity of CAR T and CAR NK cells in vivo using a xenograft lymphoma model in an autologous versus allogeneic setting and a leukemia model.ResultsOur main findings are a drastically reduced capacity for CAR-mediated IFN-γ production and lower CAR-mediated cytotoxicity of CAR NK cells relative to CAR T cells in vitro. Consistent with these in vitro findings, we report superior anticancer activity of autologous CAR T cells compared with allogeneic CAR NK cells in vivo.ConclusionsCAR T cells had significantly higher CAR-mediated effector functions than CAR NK cells in vitro against several cancer cell lines and autologous CAR T cells outperformed allogeneic CAR NK cells both in vitro and in vivo. CAR NK cells will likely benefit from further engineering to enhance anticancer activity to ultimately fulfill the promise of an effective off-the-shelf product.

CD70-Targeted Allogeneic CAR T-Cell Therapy for Advanced Clear Cell Renal Cell Carcinoma.

Therapeutic approaches for clear cell renal cell carcinoma (ccRCC) remain limited; however, chimeric antigen receptor (CAR) T-cell therapies may offer novel treatment options. CTX130, an allogeneic CD70-targeting CAR T-cell product, was developed for the treatment of advanced or refractory ccRCC. We report that CTX130 showed favorable preclinical proliferation and cytotoxicity profiles and completely regressed RCC xenograft tumors. We also report results from 16 patients with relapsed/refractory ccRCC who received CTX130 in a phase I, multicenter, first-in-human clinical trial. No patients encountered dose-limiting toxicity, and disease control was achieved in 81.3% of patients. One patient remains in a durable complete response at 3 years. Finally, we report on a next-generation CAR T construct, CTX131, in which synergistic potency edits to CTX130 confer improved expansion and efficacy in preclinical studies. These data represent a proof of concept for the treatment of ccRCC and other CD70+ malignancies with CD70- targeted allogeneic CAR T cells. Significance: Although the role of CAR T cells is well established in hematologic malignancies, the clinical experience in solid tumors has been disappointing. This clinical trial demonstrates the first complete response in a patient with RCC, reinforcing the potential benefit of CAR T cells in the treatment of solid tumors.

Abstract 5240: Novel CD3-Fusion Receptor enables combination of T-cell engagers and allogeneic CAR T cells to promote enhanced antitumor activity and overcome antigen escape

Abstract Chimeric antigen receptor (CAR) T cells have shown remarkable clinical success in the treatment of hematologic malignancies. However, the durability of response of CAR T therapy can be impaired by heterogenous expression of the CAR-specific target antigen (1° Ag) on tumors. In combination with a CAR, T-cell engagers (TCEs) that target additional tumor antigen (2° Ag) can be a unique approach to address tumor heterogeneity and overcome antigen escape. Engineered induced pluripotent stem cell (iPSC) master cell lines are a renewable source material from which iPSC-derived CAR T (CAR-iT) cells can be manufactured and administered in an off-the-shelf manner. In developing allogeneic CAR-iT cells, the T-cell receptor (TCR) must be eliminated to prevent graft-versus-host disease. However, the absence of TCR expression leads to loss of surface CD3 and a corresponding loss of TCE compatibility. Here we discuss a novel CD3ε fusion receptor (CD3-FR) with a modified transmembrane and endodomain, which allows for CD3 surface expression on TCR-less T cells and effective combination of off-the-shelf allogeneic CAR-T cells and TCEs. CAR+ CD3-FR+ iPSCs were differentiated into CAR-iT cells, uniformly expressing both modalities (>95% CAR+, >90% CD3ε+) and absent of TCR. To demonstrate that CD3-FR+ CAR-iT cells can elicit TCE-dependent activity, 1° Ag- (CD19-)/2° Ag+ (EpCAM+) target cells were used in cytotoxicity assays. As shown, CD3-FR- CAR-iT cells failed to kill 1° Ag- target cells. In contrast, CD3-FR+ CAR-iT cells demonstrated potent activity toward the target cells (>80% cytolysis) but only in the presence of the TCE targeting 2° Ag. To test the ability of CD3-FR+ CAR-iT cells to mitigate antigen escape, we used a tumor cell population heterogenous for the CAR-specific target (50% 1° Ag-) but homogenous for TCE-specific target (100% 2° Ag+). In co-culture assays, CD3-FR- CAR-iT cells failed to prevent tumor cell growth due to 1° Ag- tumor escape. However, CD3-FR+ CAR-iT cells in the presence of TCE exhibited robust control (80% cytolysis) of the heterogenous tumor, mitigating 1° Ag- tumor escape via targeting of 2° Ag. Next, we tested the ability of CD3-FR+ CAR-iT cells to control heterogeneous tumor in vivo in a disseminated leukemia model. As expected, CD3-FR+ CAR-iT cells exhibited significantly better tumor growth inhibition in the presence of TCE (p-value <0.005) compared to controls lacking the CD3-FR modality. Ex vivo analysis of tumor in the bone marrow confirmed 1° Ag- tumor escape in the control treated mice, but not in mice treated with CD3-FR+ CAR-iT and TCE targeting 2° Ag. Taken together, these studies present a novel opportunity to leverage a novel CD3 fusion receptor and enable combination of TCEs with engineered off-the-shelf CAR T-cell products, uniquely harnessing two potent therapeutic modalities for improving cancer therapy. Citation Format: Dan Lu, Eigen Peralta, Hui-Yi Chu, Soo Park, Masanao Tsuda, Earl Avramis, Matthew Denholtz, Alec Witty, Tom Lee, Bahram Valamehr. Novel CD3-Fusion Receptor enables combination of T-cell engagers and allogeneic CAR T cells to promote enhanced antitumor activity and overcome antigen escape [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5240.

Current state and future prospects of off-the-shelf allogeneic CAR NK therapy

Chimeric antigen receptor-transduced autologous T (CAR-T) cell therapy targeting CD19 has revolutionized the treatment of CD19-positive hematological tumors, including acute lymphoblastic leukemia and large B-cell lymphoma. However, despite the high response rate, many problems such as exceedingly high cost, complex logistics, insufficient speed, and manufacturing failures have become apparent. One solution for these problems is to use an allogeneic cell as an effector cell for genetic modification with CAR. Allogeneic, or "off-the-shelf", CAR-expressing immune-effector cells include 1) genome-edited, T-cell receptor (TCR) gene-deleted CAR-T cells generated using healthy adult donor T cells, 2) induced pluripotent stem cell-derived CAR-T cells, and 3) CAR NK cells. NK cells are notorious for their poor ex-vivo expansion and low susceptibility to genetic modification. In this article, I will review the current state and future prospects of allogeneic CAR cell therapies, with special reference to CAR NK cells.

Chimeric Antigen Receptor-T Cell Therapy for Lymphoma: New Settings and Future Directions.

In the last decade, anti-CD19 CAR-T cell therapy has led to a treatment paradigm shift for B-cell non-Hodgkin lymphomas, first with the approval for relapsed/refractory (R/R) large B-cell lymphomas and subsequently for R/R mantle cell and follicular lymphoma. Many efforts are continuously being made to extend the therapeutic setting in the lymphoma field. Several reports are supporting the safety and efficacy of CAR-T cells in patients with central nervous system disease involvement. Anti-CD30 CAR-T cells for the treatment of Hodgkin lymphoma are in development and early studies looking for the optimal target for T-cell malignancies are ongoing. Anti-CD19/CD20 and CD19/CD22 dual targeting CAR-T cells are under investigation in order to increase anti-lymphoma activity and overcome tumor immune escape. Allogeneic CAR product engineering is on the way, representing a rapidly accessible 'off-the-shelf' and potentially more fit product. In the present manuscript, we will focus on recent advances in CAR-T cell therapy for lymphomas, including new settings and future perspectives in the field, reviewing data reported in literature in the last decade up to October 2023.

Abstract C117: Preclinical characterization of allogeneic CAR γδ T Cell therapy for prostate cancer targeting a novel dimeric epitope on PSMA

Abstract Introduction: PSMA is a transmembrane glycosylated homodimer overexpressed in >80% of prostate cancers, with increased expression in advanced stages of the disease. Formation of homodimeric PSMA is necessary for enzymatic function and displays epitopes that are potentially more distinct from alternative PSMA-like proteins. Clinically, autologous anti-PSMA αβ CAR T cells have shown initial efficacy with limited therapeutic index. Compared to αβ T cells and other innate cells, γδ T cells are capable of multifunctional innate and adaptive targeting and infiltrate into tumor-associated tissues, including associated tropism for prostate tumor tissues. Additionally, γδ CAR T cells have demonstrated enhanced tumoricidal activity and activation-induced cytokine profiles that may decrease CRS-associated toxicities. We characterized γδ T cells modified from a set of novel scFv-based CARs targeting PSMA for prostate cancer and mapped unique epitopes for binders associated with the most potent CARs. Methods: Phage panning was used to identify anti-PSMA scFv sequences, which were reformatted into IgGs and characterized for binding to both cells expressing endogenous PSMA and recombinant PSMA. Binders with favorable profiles were reformatted into CARs in VH-VL and VL-VH orientations and transduced into Vδ1 T cells, a primarily tissue-resident subset, activated from healthy donor PBMCs. We identified lead CARs based on anti-tumor efficacy both in vitro in coculture assays and in vivo in tumor xenograft models and compared their activity to Vδ1 CAR T cells transduced with a clinically validated benchmark, J591. Epitope mapping of binders in the lead CARs was performed using a funnel of molecular assays including dot blots, competition assays, ELISAs on recombinant PSMA from different species and cross-linking mass spectrometry (XL-MS). Results: High-affinity PSMA binders yielded potent CARs. Lead PSMA CARs demonstrated robust in vitro cytotoxicity and antigen-specific binding to PSMA-expressing cell lines in a manner comparable to, or greater than the J591-based CAR T cells. Additionally, lead PSMA CARs showed anti-tumor control in 22Rv1 human prostate cancer xenografts, a highly resistant model with heterogenous PSMA expression. Epitope mapping studies on binders in lead PSMA CARs identified conformational, membrane-distal epitopes that are distinct from the predicted linear epitope of J591 and relevant comparators. Conclusion: The epitope for a lead PSMA antibody includes residues in a helix involved in PSMA dimer formation. The γδ CAR T construct derived from this binder demonstrates enhanced preclinical performance when compared to the J591 reference CAR T. Combining the features of this novel target and the potential of the γδ-based CAR T approach, the lead PSMA γδ CAR T cells described here are candidates for further clinical development in the context of additional armoring technologies and CAR architecture modifications designed to complement therapeutic potential. Citation Format: Nitya S. Ramadoss, Elizabeth Speltz, Tonya Capillo, Erika Meaddough, Katherine Wang, Natalia Friedland, Jonathan Wong, Alexander G Teague, Aruna Azameera, Smitha Rao Yelaguli Gundurao, Ramandeep Kaur, Morgan Smith-Boeck, Michael Salum, Yvan Chanthery, Shon Green, Marissa Herrman, Kevin P Nishimoto, Blake T Aftab, Arun Bhat. Preclinical characterization of allogeneic CAR γδ T Cell therapy for prostate cancer targeting a novel dimeric epitope on PSMA [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr C117.

Development of a Novel, Allogeneic GPRC5D-Directed CAR for Treatment of Multiple Myeloma Patients

Despite the success of BCMA-targeted therapy in advanced multiple myeloma, the disease is still considered incurable. Relapse following BCMA-directed cellular and antibody-based therapies is commonly associated with decreased surface expression of BCMA on malignant plasma cells. The G protein-coupled receptor, class C group 5 member D (GPRC5D) represents an additional target whose expression is limited to bone marrow and hair follicles, with selectively increased expression on malignant plasma cells. Thus far, the use of GPRC5D-targeted CAR T cells has provided a strong clinical benefit in multiple myeloma with limited on-target, off-tumor effects. To this end, we have validated novel, fully human GPRC5D-specific CARs for use with our hypoimmune (HIP), allogeneic CAR T cell platform. This will allow rapid delivery of allogeneic GPRC5D CAR T cells from healthy donors that simultaneously circumvent “non-self” recognition through TCR, B2M and CIITA gene disruption and promote persistence through innate immune evasion by leveraging CD47 overexpression. For early selection of candidate GPRC5D binders, T cells from healthy donors were transduced with VSV-G pseudotyped lentivirus containing second-generation, fully-human GPRC5D CARs. A clinically validated GPRC5D CAR was produced in similar fashion and used as benchmark control for initial characterization experiments. Transduction efficiency of GPRC5D CARs was analyzed via flow cytometry. GPRC5D CAR efficacy was analyzed in vitro via 24-hour luciferase or long-term Incucyte cytotoxicity assays against RPMI-8226 and MM.1S multiple myeloma target cells. Systemic MM.1S in vivo tumor models were performed in NSG mice to further assess efficacy of lead CAR candidates. Here we show our candidate GPRC5D CARs demonstrate in vitro cytotoxicity and effector cytokine production that is comparable to clinically validated benchmark controls. In 5-day Incucyte cytotoxicity assays, GPRC5D CAR T cells expand up to 10-fold and reduce tumor cell counts at starting E:T ratios of 1:8. Furthermore, in MM.1S in vivo tumor challenge models, these GPRC5D CARs significantly reduce flux and prolonged survival relative to tumor only controls with tumor clearance that was in line with clinical benchmark controls. Using ex vivo VSV-G production, our GPRC5D CAR T cells control multiple myeloma tumor cells both in vitro and in vivo, demonstrating efficacy that is on par with GPRC5D clinical benchmark CAR T cells. Studies to further define the efficacy of HIP-modified GPRC5D CAR T cells are ongoing. Moving forward, further interrogating GPRC5D CAR activity against multiple myeloma tumor cells with varying levels of target antigen expression will be critical to understanding the benefit of deploying GPRC5D CAR T cells within the multiple myeloma clinical landscape.

Durable Multiplex Epigenetic Editing for Generation of Allogeneic CAR T without Chromosomal Rearrangements

CAR T therapies have provided significant breakthroughs for patients with CD19+ malignancies and multiple myeloma, but a major limitation is the need for individualized manufacturing of autologous T cells, increasing the cost and reducing the availability of these life-saving drugs. An allogeneic CAR T product, available off-the-shelf, could greatly reduce manufacturing burden and improve access to therapy. To date, allogeneic CAR T clinical trials have shown limited durable responses and have reduced only TCR expression to avoid a graft versus host response. Further editing to reduce MHC class I and II expression may produce a cell resistant to allogeneic CD4+ and CD8+ T cell rejection to potentially improve allogeneic CAR T persistence. Conventional genome editing technologies that rely on double or single stranded DNA breaks have the potential to introduce genomic changes including translocations, especially when multiple edits are introduced. Epigenetic editing, in contrast, enables durable modulation of gene expression without cutting or nicking the DNA sequence and is extremely well-suited for simultaneous targeting of multiple genes to enable generation of persistent allogeneic CAR T cells. In this study, we evaluated the use of a dCas9-based epigenetic editor to silence TCR and MHC class I and II expression in primary T cells to produce allogeneic CAR T transduced for expression of CAR and HLA-E. Epigenetic editor mRNA and sgRNAs targeting the genes of interest were delivered via nucleofection to activated T cells. Edited T cells were then maintained in proliferative culture, with CD3/CD28 or PMA/ionomycin restimulation, and expression of the target genes was assessed by flow cytometry. Expression of TCR and MHC class I and II was durably suppressed in both single and multiple gene target silencing, demonstrating the robustness of epigenetic editing for ex vivo cell therapies. In addition, genomic integrity assessments demonstrated that epigenetic editing did not induce genomic changes as detected by either a quantitative sequencing-based assay or a fluorescence-based technique with single-cell resolution. In primary T cells, multiplex silencing did not cause genomic rearrangements, including translocations and truncations, above the level observed in controls, whereas Cas9 multiplex editing did. Mixed lymphocyte reaction (MLR), specific killing, and degranulation assays were used to demonstrate allogeneic function of silencing edits and the overexpression of an NK shield in vitro. In these assays, epigenetically-edited cells performed similarly to control gene edited cells in which the same gene targets were knocked out. TCR-silenced T cells did not proliferate or produce IFN-γ in response to allogeneic PBMCs in an MLR assay, and MHC class I- and II-silenced T cells did not evoke a proliferative response or IFN-γ production from allogeneic CD4+ or CD8+ T cells. Furthermore, overexpression of the HLA-E NK shield protected HLA-E expressing cells relative to B2M KO cells in a specific killing assay and reduced the degranulation response of NKG2A+ NK cells without a corresponding increase in NKG2C+ NK cell degranulation observed with other single chain HLA-E constructs tested. Finally, multiplex epigenetically-edited allogeneic CAR T were produced by transduction for expression of an anti-BCMA CAR, followed by simultaneous editing to silence target genes. These cells displayed potent anti-tumor activity against the BCMA+ MM.1S cell line, comparable to activity of unedited CAR T cells. These data support multiplex epigenetic editing to produce functional allogeneic CAR T cells that (1) do not exhibit a GVHD response; (2) resist CD8+ and CD4+ T cell alloresponses; and (3) resist the NK missing-self response normally consequent of reduced MHC class I expression without introduction of genotoxic events.

Identification of Healthy Donor Phenotypic and Functional Signatures That Predict Allogeneic, Hypoimmune CD19-Directed CAR T Cell Potency

Allogeneic, hypoimmune (HIP) CAR T cells may provide off-the-shelf CAR T cells that allow for expanded access for patients, overcome manufacturing failures, and provide an efficacious alternative to CAR T cells manufactured from heavily pre-treated T cells. However, donor T cells are the primary source of variability during production of allogeneic HIP CAR T cells, and may influence clinical efficacy. Previously, we reported a series of T cell functional assays that differentiate T cell potency as a means to select high-quality CAR T cell donors for clinical trials. Here, we explore donor phenotypic and functional signatures that are associated with desirable pre-clinical attributes for the generation of allogeneic HIP CD19-directed CAR T cells. Human T cells from healthy donors were obtained by leukapheresis and used to generate allogeneic HIP CD19-directed CAR T cells by disrupting B2M, CIITA, and TRAC genes using Cas12b nuclease, and lentiviral transduction to overexpress CD47 and to express the CD19 CAR.Phenotype profiling (PBMC subsets, memory subsets, viability, exhaustion, and activation markers) was assessed on apheresis, CD4 or CD8 enriched subsets and post-CAR manufacturing. Generated CAR T cells were evaluated in vitro by repetitive tumor challenge (3x) performed over a 14-day period using the IncuCyte platform. Bulk cytokine analysis was measured by MesoScale Discovery (MSD). In vivo efficacy was evaluated using a systemic NALM-6 challenge in immune-deficient NSG mice at HIP CD19 CAR T cell doses of 5e5 and 5e6 cells per animal. Gene expression profiling was assessed on pre-production, post-CAR T cell manufacturing, and post-serial tumor challenge CD4+ and CD8+ T cells using the Nanostring nCounter CAR T panel. Functional and phenotypic attributes were used for weighted donor performance scoring. To do this, cell functional assays were defined to produce single, quantitative indices to enable the classification of CAR T cells based on functional performance. We assessed differences in CAR T cell performance and rank scored donor performance. Associations between donor attributes, phenotype/ functional T cell attributes and ranked donor performance were evaluated using Spearman's correlations. Correlative comparison of phenotype and donor attributes in relation to weighted donor performance scores indicated that cellular subtypes at leukapheresis correlate with CAR T performance. Gene expression profiling shows that outperforming donors upregulate expression of mucosal associated invariant T (MAIT) cell canonical TCRb chain TRBV28. Outperforming donors have increased expression of Tc1/Th1 and Tc17/Th17 associated cytokines: IFNg, Il-17A, Il-17F, Il23, Il22, and have reduced expression of Tc2/Th2 associated cytokine Il5. Outperforming donors also showed reduced expression of MHC class II of alpha and beta heterodimers (HLA-DQA1 and HLA-DQB1). The information collected across phenotypic and functional comparisons will be compared with clinical performance.

Phase 1 Study of SC291, a Hypoimmune, Allogeneic CD19-Directed CAR T Cell Therapy for Relapsed/Refractory B-Cell Malignancies (ARDENT) - Initial Clinical Data

ARDENT is a phase 1 study evaluating safety and tolerability of SC291, a hypoimmune (HIP), allogeneic, CD19-directed CAR T cell therapy in subjects with Non-Hodgkin's Lymphoma and Chronic Lymphocytic Leukemia (CLL). SC291 is derived from healthy donor CD4+ and CD8+ T cells that are genetically engineered to disrupt function of CD3 and HLA class I/II, overexpress CD47, and express a CD19-directed CAR. A primary obstacle for allogeneic CAR T cell approaches remains their immune recognition and rejection. Patient immune responses prevent durable allogeneic CAR T clinical responses despite the use of enhanced lymphodepletion (LD) regimens. We therefore developed hypoimmune CAR T cells that avoid immune recognition and have the potential to deliver clinical efficacy while using a standard LD chemotherapy regimen (as defined by approved autologous CAR T products). The HIP engineering approach involves the depletion of HLA class I/II to prevent adaptive immune rejection and the overexpression of CD47 to prevent innate immune rejection, which has previously demonstrated alloimmune protection in preclinical models ( Nat Biotechnol 2019;37(3):252-258; Proc Natl Acad Sci U S A 2021;118(28):e2022091118; J Exp Med. 2021;218(3):e20200839; Nat Biotechnol. 2023; doi: 10.1038/s41587-023-01784-x, and Nat Commun. 2023;14(1):2020). The initial patient, a 74-year-old male with CLL (unmutated IGHV, del(11q), BTK C481S mutation) and 3 prior lines of therapy (FCR, ibrutinib, venetoclax + rituximab), received a LD regimen of cyclophosphamide 500 mg/m 2 and fludarabine 24 mg/m 2 (daily for 3 days) followed by a starting dose of 60 million CAR+ SC291 cells (of which approximately 80% are fully HIP engineered cells). SC291 was well tolerated with no observed CRS or ICANS. The patient developed non-neutropenic fever after LD (prior to SC291 infusion). At the Day 28 visit post-SC291 infusion, the patient had a partial response (per iwCLL 2018 criteria), along with significant (99%) B-cell reduction and improvements in platelet levels. Analysis of cellular kinetics is ongoing. SC291 is a mixture of T cells, in which portions are partially or fully HIP-engineered. As such, we evaluated the initial patient's immune response against SC291 subpopulations using a panel of previously described in vitro assays. Blood samples from the patient were taken pre-treatment (Day -5) and at the Day 13 and Day 28 visits following SC291 treatment and were assessed for immune evasion. The patient's PBMCs were sorted into CD3+ T cells and CD3-CD56+ NK cells and incubated with the following subpopulations sorted from SC291 drug product (DP): (i) fully HIP engineered (CD19 HIP CAR T cells), (ii) HLA I/II negative T cells without CD47 overexpression (DKO), and (iii) HLA I/II positive CAR T cells overexpressing CD47 (WT CD19 CAR T cells). Patient serum from Day -5 and the Day 28 visit were used for donor-specific antibody analyses, complement-dependent cytotoxicity (CDC), and antibody-dependent cellular cytotoxicity (ADCC) against the DP subpopulations. At Day -5, T cell EliSpot assays showed no immune responses against any of the SC291 sorted subpopulations, but by Days 13 and 28, these assays produced high and increasing IFN-g spot frequencies against the WT CD19 CAR T cells. Similar results were shown with T cell cytotoxicity assays (Fig. 1B). In contrast, no T cell activation and no T cell killing was observed with DKO and CD19 HIP CAR T cells at any time point. As expected, patient NK cells vigorously killed DKO T cells at all time points but spared WT CD19 CAR T and CD19 HIP CAR T cells (Fig. 1B). No antibodies against any sorted subpopulation were detected at Day -5, but IgG antibody binding of the WT CD19 CAR T population was observed at Day 28. These antibodies did not bind to DKO or CD19 HIP CAR T cells, and there was no CDC or ADCC response against these subpopulations. Interestingly, with the LD regimen used, we observed a 95% and 99% reduction of CD19+ cells on Days 0 and 28, respectively, while T cells rebounded from an 88% reduction on day 0 to pretreatment levels on Day 28 (Fig. 1A). These data demonstrate the preliminary safety and tolerability of SC291 in the initial treated patient. Immune assays demonstrated that the CD19 HIP CAR T cell subpopulation effectively evades the host adaptive and innate immune responses and could overcome the allogeneic barrier in humans. Additional data from the ARDENT study will be presented at the time of the conference.

Phase 1 Dose Escalation Study of Pembrolizumab in Combination with Rituximab and Lenalidomide for Relapsed/Refractory Large B-Cell or Follicular Lymphoma Following Progression after Anti-CD19 Chimeric Antigen Receptor (CAR) T-Cell Therapy

Introduction: While autologous CD19-targeting chimeric antigen receptor T-cells (CAR T) can induce high complete response (CR) rates in relapsed and refractory (r/r) large B-cell lymphoma (LBCL) and follicular lymphoma (FL), over 50% of patients experience progressive disease. Survival following CAR T failure is dismal, with only 1 in 5 patients living >2 years. Multiple studies indicated that resistance to CAR T therapy may be driven by CAR T cell intrinsic and/or tumor intrinsic mechanisms. Here, we evaluated the combination of PD-1 blockade with rituximab (R) and lenalidomide (len) with the rationale that it would restore the function of exhausted T cells, enhance CAR T function with len and target the tumor and its immunosuppressive microenvironment. Methods: In this single center, phase I dose-escalation study, eligible patients included adults with r/r LBCL or FL who progressed following CD19 CAR T. Treatment consisted of R (375 mg/m2/day) on days 1, 8, and 15 of cycle 1 and day 1 of cycle 2. Pembrolizumab (200 mg/day) was administered on day 1 of each 21-day cycle with the cycles repeated for up to 2 years (those who attain an investigator-determined CR could stop pembrolizumab after 24 weeks). A standard 3+3 dose escalation of len was pursued to determine the recommended phase 2 dose (RP2D), starting at 2.5 mg and escalated up to 10 mg on days 1-14 of each cycle for up to 12 cycles. The primary endpoint was overall response rate (ORR). Secondary endpoints included safety, CR rate, progression-free survival (PFS), and overall survival (OS). Exploratory objectives sought to determine the effect of combination therapy on peripheral blood T cells and evaluate baseline tumor characteristics that associated with clinical outcomes. Results: Fourteen patients initiated study treatment, 13 had LBCL (high grade B-cell lymphoma n=4; non-GCB DLBCL n=4; GCB DLBCL n=3; transformed FL n=2), 1 had FL. 12 progressed following axicabtagene ciloleucel, 1 progressed after lisocabtagene maraleucel, and 1 following an investigational allogeneic CAR T. The median time from CAR (day 0) to enrollment was 171 days (range 60-1134). The median time to progression following prior CAR T was 104 days (range 22-280). Median age was 57.5 years (range 41-73), 64% were male, all had advanced stage disease (stage III n=1, stage IV n=13), 79% had high-risk IPI, and 79% had an ECOG performance status of 1. Patients had a median of 3 prior lines of therapy (range 2-5), 86% were refractory to their last therapy. Four were treated at dose level 1 (2.5 mg len), 7 at dose level 2 (5 mg len), and 3 at dose level 3 (10 mg len). Five DLTs were observed, 3 at dose level 2 (3 gr 4 neutropenia > 7 days) and 2 at dose level 3 (1 gr 4 neutropenia > 7 days and 1 gr 3 febrile neutropenia). The RP2D was determined to be 2.5 mg of len. The most common all grade treatment-emergent non-hematologic adverse events (TEAEs) included rash (n=6, 43%), hyperglycemia (n=6, 43%), arthralgia/myalgia (n=5, 36%), diarrhea (n=4, 29%), and AST/ALT elevation (n=3, 21%). Grade 3 or higher TEAEs included 79% neutropenia (gr3 n=2; gr4 n=9), 29% lung infection (gr3 n=3; gr5 n=1), 21% febrile neutropenia (gr3 n=3), 21% lymphopenia (gr3 n=3), 21% anemia (gr3 n=3), and 14% thrombocytopenia (gr3 n=1; gr 4 n=1). With a median follow up of 37.5 months, the most common reason for discontinuation was progressive disease (n=7). Four patients (all LBCL) responded with all achieving a CR (ORR and CR rates were 29%) and all 4 patients remain in CR beyond 2 years, Figure 1. Median time to CR was 39.5 days. Median PFS was 2.3 months (95% CI: 1.6 mos-NR). Six patients died due to progressive lymphoma. Two patients died as a result of treatment-related AEs, 1 due to COVID19 infection, and 1 due to pneumonia/respiratory failure. Median OS was 21.3 months (95% CI:4.7 mos-NR). Conclusion: In these heavily pre-treated patients who progressed following CAR T-cell therapy, neutropenia was the dose limiting toxicity that prevented escalation of lenalidomide beyond 2.5 mg when combined with pembrolizumab and R. Four durable CRs were observed, none were observed at the RP2D. Risk mitigating strategies to reduce the incidence of febrile or grade 4 neutropenia and biomarker exploration to identify those likely to achieve a durable response are needed to explore this combination further.

ALLO-647 for Lymphodepletion in the Allogeneic CAR T Setting: Safety Experience with ALLO-501/501A in Patients (Pts) with Relapsed/Refractory (r/r) Large B-Cell and Follicular Lymphomas

Background: While breakthroughs in hematologic malignancies with autologous CAR T-cell therapies have been met with growing clinical interest due to impressive outcomes, limitations in logistics/manufacturing, quality consistency, and product availability persist. Allogeneic CAR T-cell therapies may circumvent such challenges by providing an off-the-shelf therapeutic option derived from healthy donors. For allogeneic CAR T cells to be successful, there must be a safe and effective way to control host lymphocyte rejection of allogeneic CAR T cells (allo-rejection). ALLO-501 and ALLO-501A are allogeneic anti-CD19 CAR T-cell products that use Cellectis technologies' TALEN® gene editing to disrupt both the TCRα constant ( TRAC) and CD52 genes. CD52 disruption specifically permits use of ALLO-647, an anti-CD52 antibody, for the transient and selective depletion of host lymphocytes that enables ALLO-501 and ALLO-501A to proliferate after infusion without rapid allo-rejection. Updated phase 1 data for ALLO-501 (ALPHA; NCT03939026) and ALLO-501A (ALPHA2; NCT04416984) showed that administration of anti-CD19 allogeneic CAR T product following use of lymphodepletion that includes ALLO-647 plus fludarabine and cyclophosphamide provided durable responses and an acceptable safety profile in CAR T-cell-naive pts with r/r large B-cell lymphoma (LBCL; Locke FL, et al. ASCO 2023; #2517). Herein, we provide safety results of ALLO-647 in pts with r/r LBCL and follicular lymphoma (FL). Methods: Pts with r/r LBCL and FL enrolled in ALPHA and ALPHA2 studies received a 3 to 5-day lymphodepletion regimen consisting of fludarabine 30 mg/m 2 and cyclophosphamide 300-500 mg/m 2 (FC) and 39, 60, or 90 mg of ALLO-647 in divided doses. ALLO-501/ALLO-501A were administered after the completion of lymphodepletion. Incidence rates of grade ≥3 cytopenias (neutropenia, thrombocytopenia, anemia and pancytopenia) were assessed at 3 timepoints (Study Day 28, Day 56, and Month 4). Pts underwent weekly cytomegalovirus (CMV) monitoring. Leukocyte reconstitution was evaluated following lymphodepletion and CAR T-cell infusion. Results: As of April 20, 2023, 87 pts with r/r LBCL (n=61) and FL (n=26) were treated with ALLO-647 and included in the analysis (median age, 64 years; median number of prior regimens, 3). In total, 11 (13%), 39 (45%), and 37 (43%) pts received 39, 60, and 90 mg ALLO-647, respectively. Among the LBCL pts, 33 were CAR T-cell-naive and treated with product manufactured using the Phase 2 process. Median follow-up was 29.5 months (range, 6.8-47.5). The most common treatment-emergent adverse events (TEAEs), assessed from the time of first dose of lymphodepletion, included neutropenia (79%), anemia (61%), infusion-related reactions (IRRs, 55%), and thrombocytopenia (53%). The most common grade ≥3 TEAEs included neutropenia (74%), anemia (38%), and thrombocytopenia (38%). The proportion of pts experiencing grade ≥3 cytopenias decreased over time from Day 28 (29%) to Day 56 (20%) to Month 4 (15%), which was consistent across all lymphoma pt subgroups. ALLO-647 IRRs were typically low grade, except for 5 (6%) grade 3 events; all IRRs were managed with supportive care measures. After treatment with ALLO-501/501A, 20 pts (23%) experienced cytokine release syndrome events, which were low grade except for 1 (1%) grade 3 event. No GvHD, grade ≥3 immune effector cell-associated neurotoxicity syndrome events, or progressive multifocal leukoencephalopathy were reported. Any grade and grade ≥3 infections were reported in 50 pts (58%) and 18 pts (21%), respectively. Two pts had (2%) fatal infectious events (COVID-19 pneumonia and pneumonia, n=1 each) as previously reported. The most common infection was CMV reactivation (any grade, n=22 [25%]; grade 3, n=8 [9%]). Following lymphodepletion, the median time to absolute neutrophil count and absolute lymphocyte count recovery to grade <4 was 7 days and 28 days, respectively. T-cell counts exhibited a steady increase over time, with the majority of pts achieving full T-cell recovery (return to baseline levels) by Month 6. Detailed pharmacokinetics and pharmacodynamics data will be presented. Conclusions: These data suggest allogeneic CAR T-cell products administered following lymphodepletion consisting of FC and ALLO-647 can provide a safe and tolerable alternative to autologous CAR T-cell therapy.

Expansion, Persistence and Pharmacodynamic Profile of ADI-001, a First-in-Class Allogeneic CD20-Targeted CAR Gamma Delta T Cell Therapy, in Patients with Relapsed/Refractory Aggressive B-Cell Non-Hodgkin's Lymphoma

Introduction ADI-001 is a first-in-class allogeneic gamma delta (γδ) CAR T cell therapy targeting the B-cell antigen, CD20. Expansion and persistence of cell therapy products and release of functional cytokines have historically correlated with patient outcomes. Here we report cellular kinetics and pharmacodynamic correlates from a phase 1, multicenter, open-label, dose escalation study to evaluate ADI-001 in R/R B cell NHL. Methods Cellular kinetics of ADI-001 were measured using three orthogonal methods, including quantitative SNP profiling of cell product (AlloCell), flow-cytometry for CAR+ Vδ1 γδ T cells, and droplet digital PCR (ddPCR) quantification of CAR transgene copies. Using these methods, expansion of ADI-001 was assessed in the peripheral blood for 24 evaluable patients across four dose levels in association with a phase 1 dose-escalation trial (NCT04735471). Relationships between ADI-001 cellular kinetics and radiographic clinical responses were also examined. Serum biomarkers related to host immune cell recovery during lymphodepletion and cytokine release were monitored for pharmacodynamic purposes. Humoral immunogenicity was assessed by Luminex™-based anti-HLA antibody screening. Other correlative characteristics were also evaluated, including degree of HLA mismatching between patient and ADI-001 product in relation to response and/or ADI-001 expansion and persistence. Results As of the May 2023 cutoff for reporting, ADI-001 was detected at all dose levels using ddPCR and flow cytometry to quantify CAR transgene copies and CAR+ Vδ1 γδ T cells, respectively. Treatment at the highest dose level, achieved a notable mean C max of 201,666 copies/µg or 483 CAR+ cells/µl, and mean day 28 persistent exposure of 16,553 copies/µg or 21 CAR+ cells/µL. Additional measures of ADI-001 exposure (AlloCell) further demonstrated a robust dose-dependent expansion profile of ADI-001 in the peripheral blood. Subjects with a BOR of CR or PR were observed to have a mean peak of 180,107 CAR copies/µg versus a mean peak of 20,950 CAR copies/µg in subjects with a BOR of SD or PD. Additionally, as expected following lymphodepletion, a transient increase in the homeostatic cytokine IL-15, coinciding with ADI-001 expansion and host-mediated immune cell recovery, was observed. Using high-resolution HLA typing, the degree of HLA mismatch between allogeneic ADI-001 product and patients did not associate with degree of cellular expansion, day 28 persistence, or response. Finally, emergence of donor-specific anti-drug antibodies following treatment was not observed. Conclusions Using three orthogonal measures of exposure, we show for the first time robust dose-dependent expansion and persistence of an allogeneic CAR γδ T cell therapy, ADI-001, irrespective of extensive degree of HLA mismatch. ADI-001 C max and D28 persistent exposure were comparable to, or exceeded, those demonstrated by alternative allogeneic CAR T therapies and approved autologous CD19 CAR T therapies. Favorable clinical responses were also associated with higher mean and median peak of ADI-001 cells.

Hypoimmune, Allogeneic CD22-Directed CAR T Cells That Evade Innate and Adaptive Immune Rejection for the Treatment of Large B Cell Lymphoma Patients That Are Relapsed/Refractory to CD19-Directed CAR T Cell Therapy

Despite the success of CD19CAR therapy against large B-cell lymphoma, up to 60% of CD19-directed CAR T cell treated patients relapse after, or are refractory to, treatment. Of the relapsed/refractory population a significant subset of patients, around 30%, have CD19-negative tumor and a short median survival prognosis - necessitating the implementation of rapid intervention strategies (Atilla, Transl Oncol 2022; Spiegel, Blood 2021). Sana is developing “off-the shelf” allogeneic CD22-directed CAR T cells that may enable rapid intervention and induce clinically meaningful activity for these CD19 CAR T cell relapsed/refractory patients. Here we describe methods to create hypoimmune (HIP), allogeneic CD22-directed CAR T cells using a clinically validated CD22CAR that: 1) support adaptive immune cell evasion via B2M and CIITA gene disruption, 2) promote persistence and innate immune cell evasion by overexpression of CD47, and 3) mitigate GvHD via TRAC gene disruption, (Hu, Nat Commun 2023). HIP, allogeneic CD22-directed CAR T cells (CD22 HIP CAR T cells) were derived from healthy donor CD8 + and CD4 + T cells that were gene-edited to disrupt the B2M, CIITA, and TRAC (TCR) genes and transduced to overexpress CD47 and express a second-generation, clinically validated CD22-directed CAR. Gene-edited Mock T cells and unedited, allogeneic CD47-CD22CAR T cells were generated as donor-matched controls. The CD22 HIP CAR T cells were evaluated for immune evasive activity using a previously described in vitro xCelligence cytotoxicity assay (Hu, Nat Commun 2023). The T cells were further evaluated in vitro for dose-dependent short-term and long-term cytolytic activity analysis via luciferase or IncuCyte assay, respectively. Correlate cytokine responses were performed by meso-scale discovery analysis and T cell expansion by IncuCyte classification. Systemic NALM and RAJI parental or CD19 antigen knockout tumor models (IV injection) were executed in a NSG or NSG-SGM3 mouse background. xCelligence analysis confirmed that CD22 HIP CAR T cells were protected from NK and macrophage recognition; whereas Mock control T cells that lack HLA expression succumbed to NK cell and macrophage killing within 5-hours of the 50-hour assay due to the absence of CD47 protection. Furthermore, CD22 HIP CAR T cells elicited cytotoxicity and robust cytokine responses (e.g. IFNγ, IL-2) in a dose-dependent manner comparable to allogeneic CD47-CD22CAR T cells when cultured with CD22 +/CD19 + or CD22 +/CD19 - NALM, RAJI, or K562 targets. The CD22 HIP CAR T cell and allogeneic CD47-CD22CAR T cell control produced low level IFNγ (<400pg/mL) and Granzyme B (<100pg/mL) when cultured alone or with CD22 - tumor targets. These basal levels of IFNγ and Granzyme B did not have long-term cytolytic effects on CD22 - tumor cells, nor was any significant level of IL-2 produced by CD22 HIP CAR T cells cultured alone, or in the presence of CD22-negative tumor targets. Indeed, CD22 HIP CAR T cell proliferation was dependent upon CD22 + tumor co-culture. NALM and RAJI CD19 knockout in vivo tumor models demonstrated that CD22 HIP CAR T cells significantly reduced flux (Day 21 AUC vs Mock: NALM and RAJI (p=0.0001)) and prolonged survival (Median vs Mock: p<0.004). Furthermore, CD22 HIP CAR T cells elicited comparable, dose-dependent antitumor activity relative to allogeneic CD47-CD22CAR T cell controls in a systemic NALM tumor model. Our data provide evidence that CD22 HIP CAR T cells elicit robust antitumor activity in an antigen and dose-dependent manner. Furthermore, these studies demonstrate that CD22 HIP CAR T cells are agnostic to CD19 expression and selectively elicit pharmacologic activity against, or in response to, CD22 + tumor. Results also demonstrate that the HIP approach to generate allogeneic CAR T cells that evade innate and adaptive immune rejection can be expanded as a platform to generate CAR T cells against target antigens alternative to CD19. Collectively, the data suggest that CD22 HIP CAR T cells display a combination of antigen-specific pharmacologic activity and immune evasion that support their progression into human clinical studies for the treatment of CD19 CAR T cell-refractory patients.

Preliminary Results of Nathali-01: A First-in-Human Phase I/IIa Study of UCART20x22, a Dual Allogeneic CAR-T Cell Product Targeting CD20 and CD22, in Relapsed or Refractory (R/R) Non-Hodgkin Lymphoma (NHL)

Background: UCART20x22 is a dual targeted allogeneic CAR T-cell product candidate generated from healthy donor T-cells. Donor-derived T-cells are transduced with a bicistronic lentiviral vector to express CARs targeting CD20 and CD22. TALEN ® gene editing technology is used to inactivate the TRAC and CD52 genes to minimize graft-versus-host disease (GvHD) and allow for the use of alemtuzumab (CLLS52, an anti-CD52 monoclonal antibody) in the lymphodepletion (LD) regimen, respectively. UCART20x22 previously demonstrated robust and sustained activity in vitro and in vivo (Aranda-Orgilles B, et al. Cancer Immunol Res 2023). Here we present the initial preliminary results from the NatHaLi-01 trial (NCT05607420), the first-in-human Phase 1/2a dose-finding and expansion study evaluating UCART20x22 in R/R B-cell NHL. Methods: The primary endpoints are safety, tolerability, and determining the recommended phase 2 dose of UCART20x22. Additional endpoints include anti-lymphoma activity and cellular kinetics of UCART20x22. Key eligibility criteria are age 18-80y, lymphoma expression of CD20 and/or CD22 by local assessment, and ≥ 2 prior treatment regimens including autologous CD19 CAR T-cell therapy (CAR19), if eligible. After LD with FCA (fludarabine 30 mg/m 2 × 3d, cyclophosphamide 0.5g/m 2 × 3d, CLLS52 12 mg on D1, 24 mg on D2, D3), a single infusion of UCART20x22 is administered at a flat dose level ([DL1] 50 x 10 6 cells; [DL2] 150 x 10 6 cells; and [DL3] 450 x 10 6 cells). Results: As of 01 July 2023, 3 pts were enrolled and treated at DL1. Pt 1 is a 76yo female with double-expressor diffuse large B-cell lymphoma (DLBCL) relapsed after R-CHOP, radiation therapy, and polatuzumab vedotin with bendamustine/rituximab (BR) and was considered ineligible for CAR19 due to recent high dose bendamustine. Pt 2 is a 65yo female with R/R triple-hit lymphoma transformed from follicular lymphoma who was previously treated with radiation therapy, BR, dose-adjusted R-EPOCH, and two separate CAR19 treatments. Pt 3 is an 18yo female with R/R DLBCL transformed from marginal zone lymphoma who previously received chemoimmunotherapy, venetoclax, ibrutinib, BR, CAR19, obinutuzumab, glofitamab, tafasitamab, lenalidomide, and an experimental epigenetic modifier. Cytokine release syndrome (CRS) occurred in all patients. Pt 1 had G1 CRS and received tocilizumab (toci) x3 and dexamethasone (dex) x1. Pt 2 had G2 CRS and received toci and dex x1. Pt 3 had G1 CRS and received toci x1. All pts' CRS resolved with treatment. No immune effector cell associated neurotoxicity (ICANS) or GvHD was observed. There were no UCART20x22 dose limiting toxicities (DLTs), and there was 1 CLLS52 SAE and 1 DLT. The SAE was G3 CMV reactivation and was managed with IV antivirals. The DLT was in Pt 2, who entered the study with G1/2 cytopenias and a hypocellular bone marrow and had prolonged bone marrow aplasia after treatment thought to be due to cumulative chemotherapy exposure. This pt proceeded to allogeneic hematopoietic stem cell transplant (HSCT) from a matched donor while in complete remission from UCART20x22 treatment. Responses were assessed per Lugano criteria beginning on D28. At D28, all 3 pts responded with 2 CRs and 1 PR. Host lymphocytes remained suppressed through D28 for all pts. UCART20x22 expansion was observed by flow cytometry in 3/3 pts and was initially detected on D7 followed by peaks between D9-14. Peak expansion ranged from 16-682 cells/μL and was predominantly CD8 T-cells. Ferritin levels and increases in serum cytokines, including IFNγ, TNFα, IL-2, and IL-6, correlated closely with UCART20x22 expansion. Summary: Treatment with UCART20x22 at DL1 has been safe and well-tolerated, with no UCART20x22 DLTs or ICANS, and CRS was <G3 in all pts. There was 1 DLT of prolonged bone marrow aplasia in a pt with impaired hematopoietic reserve at study entry. UCART20x22 was active at DL1, with all 3 pts responding, including 2 CRs in pts who had previously received CAR19. UCART20x22 expansion was observed in all pts. These initial data support the continued study of UCART20x22 in R/R NHL, and updated results will be presented.

Long Term Follow-up after Treatment with Allogeneic Off-the-Shelf CAR T Cell Therapy for Relapsed or Refractory B-Cell Malignancies

CAR T cells provide benefit in patients with relapsed/refractory (R/R) hematologic malignancies. However, limitations of autologous products include cost, production complexity, quality of T cells, and life-threatening toxicity. To overcome these limitations, we developed allogeneic, “off-the-shelf” CD19-specific CAR T cells by transducing our institutional second-generation CD28-containing CAR (19-28z) into Epstein-Barr Virus (EBV)-specific cytotoxic lymphocytes (19-28z CAR EBV-CTLs). Patients with R/R B-cell malignancies were stratified into 3 treatment cohorts: cohort 1 (disease recurrence after allogeneic or autologous hematopoietic cell transplant (HCT)), cohort 2 (19-28z CAR EBV-CTLs as consolidative therapy after autologous HCT), or cohort 3 (19-28z CAR EBV-CTLs as consolidative therapy after allogeneic HCT). The objectives were determining dose limiting toxicity (DLT; primary endpoint) and optimal dose for multiple infusions. Sixteen patients were treated with 19-28z CAR EBV-CTLs with 8 in cohort 1 (B-ALL n=6; Burkitt lymphoma n=1; CLL n=1), 6 in cohort 2 (PMBCL n=2; DLBCL n=3; marginal zone lymphoma n=1), and 2 in cohort 3 (DLBCL n=2). Median age at treatment was 30 years (range 1-71 years). All patients were heavily pretreated, receiving a median of 5 lines of therapy (including any HCT received; range 3-11 lines of therapy) prior to 19-28z CAR EBV-CTLs. Dosing on this protocol was based off T cells/kg: median dose infused was 3x10 6 T cells/kg (range 1-10x10 6 T cells/kg). Twelve patients received multiple doses (overall median 2.5; range 1-3) with 3x10 6 T cells/kg determined to be the optimal dose enabling multiple treatments per manufactured cell line. Products had variable transduction efficiencies with a median of 28.7% (range 7.4-41%), and median 19-28z CAR EBV-CTL dose was 0.7x10 6/kg (range 0.05-2.7x10 6/kg). 19-28z CAR EBV-CTL sources were primary HCT donors (n=4) and third-party donors (n=12). HLA matching between patients and donors was 10/10 (n=3), 6/10 (n=2), 5/10 (n=1), 4/10 (n=3), and 2/10 (n=7). Ten of the patients were EBV-seropositive, four were seronegative, and two were serostatus unknown. Severe cytokine release syndrome (CRS) or neurotoxicity did not occur post-infusion, and no DLT was noted in the trial. Six patients developed diffuse skin rash with 3 patients' rashes deemed to be graft-versus-host disease (n=1 HLA-matched HCT donor; n=2 third-party donor), which resolved with topical steroids (n = 3). Median follow-up was 53.3 months (range 4.2-135 months) with 4 deaths due to disease progression. Overall survival of all patients was 81% at 12 months and 74% at 24 months ( Figure 1). In summation, we successfully treated 16 patients with an allogeneic, “off-the-shelf” 19-28z CAR EBV-CTL product without significant toxicity. The manufacturing platform effectively overcomes many of the limitations of autologous CAR T cell products including cost, need for patient-specific manufacturing, and life-threatening toxicity. CAR EBV-CTLs also broaden the applicability of CAR T cell therapy for patients with insufficient circulating T cells to generate autologous products. This trial determined an optimal manufacturing dose of allogeneic CAR T cells to enable repeated dosing. In the cohort of heavily pre-treated patients, long-term outcomes following multiple infusions of 19-28z CAR EBV-CTLs were excellent. Allogeneic, “off-the-shelf” CAR EBV-CTLs provide a readily available therapy for patients with limited therapeutic options, and this trial demonstrates the safety of CAR T cell infusions as consolidation after HCT. This trial was registered at www.clinicaltrials.gov as NCT01430390. Figure 1: Overall survival following 19-28z CAR EBV-CTLs of all patients by disease (A) and donor source (B).

Constitutive Turbodomains enhance expansion and antitumor activity of allogeneic BCMA CAR T cells in preclinical models.

The magnitude of CAR T cell expansion has been associated with clinical efficacy. Although cytokines can augment CAR T cell proliferation, systemically administered cytokines can result in toxicities. To gain the benefits of cytokine signaling while mitigating toxicities, we designed constitutively active synthetic cytokine receptor chimeras (constitutive Turbodomains) that signal in a CAR T cell-specific manner. The modular design of Turbodomains enables diverse cytokine signaling outputs from a single homodimeric receptor chimera and allows multiplexing of different cytokine signals. Turbodomains containing an IL-2/15Rβ-derived signaling domain closely mimicked IL-15 signaling and enhanced CAR T cell potency. Allogeneic TurboCAR T cells targeting BCMA showed no evidence of aberrant proliferation yet displayed enhanced expansion and antitumor activity, prolonging survival and preventing extramedullary relapses in mouse models. These results illustrate the potential of constitutive Turbodomains to achieve selective potentiation of CAR T cells and demonstrate the safety and efficacy of allogeneic BCMA TurboCAR T cells, supporting clinical evaluation in multiple myeloma.