Real-world safety and efficacy study of the BCMAxCD3 bispecific antibody teclistamab for relapsed or refractory multiple myeloma
91. Real-world safety and efficacy of teclistamab for patients with heavily pretreated relapsed-refractory multiple myeloma
Danai Dima (Cleveland Clinic, Cleveland, OH, USA) presented a study of real-world safety and efficacy of teclistamab, a T cell-redirecting bispecific antibody targeting both BCMA and CD3, among patients with relapsed or refractory multiple myeloma (r/r MM). Approval of teclistamab for patients with R/R MM after four or more lines of therapy was based on results from the MajesTEC-1 clinical trial, and this multicenter retrospective study focused on patients who would have been ineligible for MajesTEC-1. 106 patients who received at least one dose of teclistamab for r/r MM were included in the study, and 88 patients met at least one exclusion criterion of MajesTEC-1. 32% of the patients were older than 70 years of age, 42% had extramedullary disease, and 53% had prior BCMA director therapy. The overall response rate (ORR) was 66%, which was comparable to the ORR of 63% in MajesTEC-1, but the complete response rate (CRR) of 29% were lower than observed in MajesTEC-1 (CRR 39%). 24.5% of patients experienced progressive disease after teclistamab, compared to 14.5% in MajesTEC-1. Among the 56 patients who had received prior BCMA-targeting treatment, responders had a numerically longer amount of time since their last BCMA-targeting treatment compared to non-responders (399 vs. 205 days). After a median follow-up of 3.8 months, 43% patients were experiencing progressive disease, and median progression-free survival (PFS) was 5.4 months. 26% of patients had died at the time of data cutoff, and 86% of the deaths were from disease progression. Multivariate analyses indicated that inferior ORR and inferior PFS were associated with poor performance status (ECOG 2 or higher) and extramedullary disease. 64% of patients experienced cytokine release syndrome (CRS), which was comparable to 72% in MajesTEC-1, and 14% of patients experienced ICANS. Incidences of severe (grade 3 or higher) CRS and ICANS were low, at 1% and 3%, respectively. Incidence of CRS was significantly higher among patients who received a condensed dosing schedule of teclistamab compared to those who experienced longer intervals between doses (51% vs. 31%, p=0.038). 27 patients experienced at least one dose delay due to adverse events, usually due to infection, and 12 patients stopped taking teclistamab due to toxicity. This study represented the largest real-world study of safety and efficacy of teclistamab for r/r MM patients, most of whom would be ineligible for MajesTEC-1 trial. Early safety and efficacy data are encouraging, even among patients with high-risk disease features, although longer follow-up is needed to determine the durability and depth of responses. Safety data from this real-world study also indicate that future clinical trials of teclistamab should explore dosing intervals to reduce risk of CRS.
Liso-cel CAR T cellular therapy for second-line treatment of large B cell lymphoma in patients not intended for stem cell transplantation
105. Lisocabtagene maraleucel as second-line therapy for relapsed/refractory large B-cell lymphoma in patients not intended for hematopoietic stem cell transplant: Final analysis of the phase 2 PILOT study
Alison Sehgal (University of Pittsburgh Medical Center, Pittsburgh, PA, USA) presented final analysis from the open-label, phase 2 PILOT study evaluating the efficacy and safety of the CD19-directed CAR-T cell therapy lisocabtagene maraleucel (liso-cel) in patients with relapsed/refractory large B-cell lymphoma (r/r LBCL) who received one prior line of treatment and were not intended for autologous hematopoietic stem cell transplantation (HSCT). Previous analyses of PILOT indicated that the primary study endpoint had been met with an overall response rate (ORR) of 80%. 61 patients were included in the liso-cel treated analysis set. None of these patients were intended for autologous HSCT, and the majority (79%) were due to an age of 70 years or older. 30% of patients had high-grade lymphoma with diffuse LBCL histology (HGBCL), 33% had double-/triple-hit disease, and 54% had refractory disease. At a median follow-up of 18.2 months, the ORR was 80%, with 54% of patients achieving a complete response (CR). Median duration of response (DOR) was 23.3 months after a median follow-up of 23.1 months. Median DOR was not reached for patients with a CR and was 2.1 months for patients with a partial response (PR). Median progression-free survival (PFS) was 9.0 months, with an 18-month PFS rate of 42%. Median overall survival (OS) was not reached, with an 18-month OS rate of 59%. 79% and 18% of patients experienced an adverse event (AE) of grade 3 or higher during the treatment-emergent (TE) and post-TE periods, respectively. Cytokine release syndrome occurred in 38% of patients (2% grade 3 or 4) and neurological events in 31% of patients (5% grade 3-4), all during the TE period. 24 deaths occurred in the study, and 20 were due to disease progression. Most deaths occurred 90 days or more after cell infusion. At 2 years, 41% (9/22) of patients had detectable trans gene (9/22). This analysis represents the longest follow-up to date of liso-cel as second-line therapy for r/r LBCL who are not intended for HSCT. The high CR rate, high DOR, and safety profiles are consistent with previous reports. These results continue to support liso-cel as second-line therapy for patients with r/r LBCL in whom HSCT is not intended.
Brexucabtagene autoleucel CAR T therapy for relapsed/refractory mantle cell lymphoma: The ZUMA-2 and ZUMA-18 trials
106. Outcomes of patients with relapsed/refractory mantle cell lymphoma (R/R MCL) treated with brexucabtagene autoleucel (brexu-cel) in ZUMA-2 and ZUMA-18, an expanded access study
Andre Goy (Hackensack Meridian Health, Hackensack, NJ, USA) reported four-year survival outcomes of the ZUMA-2 study, a phase 2 multicenter trial of CD19-targeting CAR T cell brexucabtagene autoleucel (brexu-cel) for adults with relapsed/refractory mantle cell lymphoma (r/r MCL) after two or more prior therapies. Primary analysis of ZUMA-18, an expanded access study of brexu-cel for r/r MCL was also presented. The primary objectives of ZUMA-18 were to provide access to brexu-cel for patients with r/r MCL until it was commercially available (Cohort 1) and for patients whose manufactured product did not meet commercial release specifications (Cohort 2). 23 of 27 patients enrolled in ZUMA-18 received brexu-cel. 21 patients received brexu-cel after leukapheresis and conditioning chemotherapy (Cohort 1), and the two patients in Cohort 2 received Cohort 1 treatment without leukapheresis (initial leukapheresis product used). At a median follow-up of 33.5 months, investigator-assessed overall response rate (ORR) was 87%, 13 patients (57%), including both in Cohort 2, had a complete response, 7 (30%) patients had a partial response (PR), and 2 (9%) patients had progressive disease (PD). Median duration of response (DOR) was 15 months and 20 months in patients with a CR. Median progression-free survival (PFS) was 16.1 months in all treated patients and 21.0 months in patients with a CR. Median overall survival (OS) was not reached, and 24-month OS rate was 58%. At the time of data cutoff, 61% of patients were still alive, and 39% (n=9) had died, 5 due to adverse events (AEs) and 2 due to PD. 78% of patients experienced AEs of grade 3 or higher related to brexu-cel. 87% and 70% of patients experienced cytokine release syndrome (CRS) and neurological events (NE), respectively, and no case of grade 5 CRS or NEs was reported. Regarding ZUMA-2, at a median follow-up of 47.5 months, median OS was 46.4 months, with 30 of 68 (44%) treated patients still alive. Among patients with a CR, median OS was 58.7 months. Results from ZUMA-2 and ZUMA-18 indicate that brexu-cel is associated with high clinical efficacy in patients with r/r MCL and support the continued use of brexu-cel as the standard of care for r/r MCL.
Association of an inflammatory biomarker signature with CAR-T treatment failure in patients with lymphoma
224. An inflammatory biomarker signature reproducibly predicts CAR-T treatment failure in patients with aggressive lymphoma across the Zuma trials cohorts
Sandeep Raj (Memorial Sloan Kettering Cancer Center, New York, NY, USA) presented a study of an inflammatory biomarker signature and its correlation with CAR T cell therapy failure. The objectives of the study were to understand the role of inflammation in CAR T cell treatment failure and to develop a metric to identify patients who were at risk of relapse after CAR T therapy. Previous work used a machine learning approach to develop an Inflammation Gausian Mixture Model (InflaMix) of 14 key laboratory tests and cytokine levels measured prior to CAR-T cell infusion. Patients were assigned to one of two groups, inflammatory or non-inflammatory clusters. When InflaMix was applied to multiple independent real-world cohorts, the inflammatory signature was consistently associated with poorer disease response (measured by complete response (CR) rate), progression free survival (PFS), and overall survival (OS) after CAR T therapy. The inflammatory signature was significantly associated with baseline tumor burden, metabolic tumor volume by PET-CT, elevated LDH and increased prior lines of therapy. In patients for whom the inflammatory signature was resolved prior to CAR T infusion, the CR rate was significantly higher (77/115) compared to patients for whom the inflammatory signature was not resolved (29/105 CR; p<0.001). OS and PFS were also significantly improved, suggesting the inflammatory signature is a modifiable risk factor. InflaMix was further validated in cohorts of patients from the ZUMA 1 (n=182) and ZUMA 7 (n=170) trials studying axicabtagene ciloleucel (axi-cel) for large B cell lymphoma. The inflammatory signature was significantly associated with poor CR rate in the ZUMA 1 and ZUMA 7 cohorts (HR 1.65, p=0.045 and HR 1.75, p=0.041, respectively). PFS and OS were worse among patients with the inflammatory signature compared to those with the non-inflammatory signature. Similar associations were seen within the ZUMA 5 trial (n=152) investigating axi-cel for follicular and marginal zone lymphomas. To test whether systemic inflammation mediates CAR-T dysfunction in the blood and tumor microenvironment (TME). Pre-treatment tumor biopsies from the ZUMA 1 trial indicated that the inflammatory signature was associated with a significantly lower Immunosign-21 score and a numerically lower level of activated T cells compared to the non-inflammatory signature. The inflammatory signature was also associated with diminished CAR-T cell peak expansion. Taken together, these results suggest the inflammatory signature is associated with an immunosuppressive tumor microenvironment, and it attenuates CAR T expansion. InflaMix is an inflammatory signature that is associated with CAR T cell treatment failure in LBCL and folicular lymphoma across multiple independent real-world and clinical trial cohorts. InflaMix has potential to identify patients at high risk of treatment failure, and the reported findings suggest that prophylactic anti-inflammatory therapy prior to CAR T infusion may improve long-term efficacy of CAR T cell therapy.
Circulating tumor DNA as an early predictor of durable clinical benefits from lisocabtagene maraleucel for large B-cell lymphoma
225. Circulating tumor DNA dynamics as early outcome predictors for lisocabtagene maraleucel as second-line therapy for large B-cell lymphoma from the phase 3 TRANSFORM study
Ash Alizadeh (Stanford University, Stanford, CA, USA) reported a longitudinal study of circulating tumor DNA (ctDNA) levels from patients in TRANSFORM study of lisocabtagene maraleucel (liso-cel) for primary refractory or early relapsed large B-cell lymphoma (LBCL). Prior studies indicate that ctDNA shows promise for monitoring minimal residual disease (MRD), and this exploratory study investigated the predictive value of ctDNA after second-line therapy with liso-cel for LBCL. In the TRANSFORM study, patients were randomized 1:1 for optional bridging, lymphodepletion, and liso-cel (n=92) or for standard of care immunochemotherapy followed by high dose chemotherapy and hematopoietic stem cell transplant + HSCT (n=92). Previously reported results from TRANSFORM indicate that liso-cel demonstrated superior event-free survival (EFS) over SOC (18-month EFS of 52.6% and 20.8%, respectively.) Plasma samples from both arms (n=79 for liso-cel; n=81 for SOC) were used for baseline Phase Variant Enrichment and Detection Sequencing, and samples from the liso-cel arm were used for longitudinal ctDNA assessment. Tumor-derived phase variants were identified for 85% of patients. Samples were reported of having detectable MRD when ctDNA levels exceeded a detection threshold, corresponding to 98% specificity. Baseline ctDNA levels correlated significantly with disease burden indices. Longitudinal analyses indicated that patients with a complete response (CR) to liso-cel exhibited early rapid reductions in ctDNA levels, while ctDNA levels remained relatively steady for patients with progressive or stable disease. ctDNA negativity correlated with clinical response, with 50% of patients with a CR achieving ctDNA negativity by day 15 after infusion, remissions deepened over time. Undetectable ctDNA correlated with durable event free survival (EFS), with significant EFS benefits appearing as early as day 15 (HR 3.26) and the most significant association of ctDNA and EFS benefit occurring at month 3 (HR 8.16). Among patients who achieved a CR by PET/CT scan, detectable ctDNA was associated with relapse: by month 12, all four patients with CR by PET/CT scan and detectable ctDNA had experienced relapse or death. Results suggest undetectable ctDNA strongly predicts CR and durable clinical benefits, indicated by EFS. Detectable ctDNA was associated with risk of progressive disease, suggesting ctDNA provides prognostic benefits. Similar longitudinal studies studying ctDNA dynamics in patients in the SOC arm are currently in progress.
Lisocabtagene maraleucel CAR T therapy for patients with relapsed/refractory chronic lymphocytic leukemia or small lymphocytic leukemia
330. Lisocabtagene maraleucel (liso-cel) in R/R CLL/SLL: 24-month median follow-up of TRANSCEND CLL 004
Tanya Siddiqui (City of Hope National Medical Center, Duarte, CA, USA) presented updated results of TRANSCEND CLL 004, a phase I/II single-arm trial of lisocabtagene maraleucel (liso-cel) for relapsed or refractory (r/r) chronic lymphocytic leukemia (CLL) or small lymphocytic leukemia (SLL). Primary analysis of TRANSCEND CLL 004 previously reported liso-cel was associated with rapid, deep, durable responses and a manageable safety profile. 83% patients had high risk cytogenetics and median lines of therapy was 5. With a median follow-up of 23.5 months, in the primary efficacy analysis set (PEAS) of patients who had experienced disease progression after Bruton tyrosine kinase inhibitor (BTKi) and venetoclax failure (n=50), overall response rate (ORR) was 44%, and 10 patients achieved a complete response (CR) or complete response with incomplete marrow recovery (CRi; CR/CRi rate 20%). The undetectable minimal residual disease (uMRD) rate was 64% in blood and 60% in bone marrow. Median duration of response (DOR) was 35.3 months, and DOR among patients who achieved a CR/CRi was not reached. Median progression-free survival (PFS) was 11.9 months, and median overall survival (OS) was 30.3 months. Efficacy outcomes in the full study population (n=88), with a CR/CRi rate of 19.3%, median DOR of 35.3 months, median PFS of 17.9 months, and median OS of 43.2 months. No new safety signals were observed, with cytokine release syndrome (CRS) and cytopenias as the most common adverse events. 85% of patients experienced CRS of any grade, and no cases of CRS of grade 4-5 were observed. Neurological events (NE) occurred in 45% of patients, and one patient (1%) experienced a grade 4 NE. In the context of this longer follow-up, liso-cel continues to demonstrate durable complete responses, high uMRD rates, and manageable safety profile for a broad patient population, especially among patients with r/r CLL/SLL that progressed after BTKi and venetoclax treatment, meeting unmet needs for a critical patient population.
A Fixed-duration treatment with the bispecific antibody, glofitamab for heavily pre-treated patients with relapsed or refractory large B-cell lymphoma
433. Glofitamab monotherapy in relapsed or refractory large B-cell lymphoma: Extended follow-up from a pivotal phase II study and subgroup analyses in patients with prior chimeric antigen receptor T-cell therapy and by baseline total metabolic tumor volume
Martin Hutchings (Rigshospitalet and University of Copenhagen, Copenhagen, Denmark) presented follow-up and subgroup analyses of a phase II study of the CD20- and CD3-targeting bispecific antibody glofitamab for relapsed/refractory large B-cell lymphoma (r/r LBCL). Obinutuzumab was given 1 week prior to glofitimab to reduce the tumor burden and mitigate the risk of CRS. Glofitimab was administered in step-up doses during the first 21-day cycle, and the target dose (30 mg) was administered on the first day of cycles 2 through 12, for a fixed duration of 8.3 months. 155 patients were enrolled in the study, and 154 patients received one or more dose of glofitimab. The patient population was heavily pre-treated, with a median of 3 prior therapies and 33.1% of patients receiving prior CAR T therapy. At a median follow-up of 32 months, the overall response rate (ORR) was 52%, complete response rate (CRR) was 40%, and median duration of complete response (DoCR) was 26.9 months. Response rates of the population of patients who had received prior CAR T therapy (n=52) were similar, with an ORR of 50%, CRR of 37%, and DoCR of 22.0 months. Among patients who had achieved an early CR by Cycle 3 (n=44), median progression-free survival (PFS) of 33.1 months and a 24-month PFS rate of 63.5%. The 24-month OS rate was 73.4%, suggesting that most patients who reached a CR by cycle 3 remained progression-free and alive for more than two years. Among patients who had achieved a CR by the end of 12 cycles of treatment (EOT; n=45), median PFS was 24.0 months, and 18-month PFS rate was 66.6%. 18-month OS rate was 80.7%. Higher baseline total tumor metabolic volume (TMTV) was associated with lower PFS. 24-month PFS for patients with baseline TMTV at or above the median TMTV (128.7 mL; n=72) was 11.8 months, compared to 41.6 months among patients (n=72) with baseline TMTV below the median. No new adverse events were reported, with cytokine release syndrome (CRS) occurring in 64% of patients, and most cases were grade 1 or 2 and occurred during cycle 1 or 2 of treatment. Higher baseline TMTV was associated with increased risk of experiencing CRS of grade 2 or higher. Taken together, the data indicate that fixed-duration glofitamab is safe, with most adverse events occurring early in treatment, and it can produce deep and enduring responses in heavily pre-treated populations of patients with r/r LBCL.
Memory-like natural killer cells exhibit enhanced metabolic activity and cytotoxicity against acute myeloid leukemia
466. WU-NK-101 (W-NK), a memory-like (ML) NK cell, intrinsically overcomes factors restricting adoptive cell therapy (ACT) in acute myeloid leukemia (AML)
Sergio Rutella (Nottingham Trent University, Nottingham, United Kingdom) reported a study characterizing WU-NK-101, a cytokine-reprogrammed, expanded, cryopreserved memory-like (ML) natural killer (NK) cell using scRNA-seq, multidimensional flow and mass spectrometry. ML-NK cells are generated from conventional NK cells (cNK) activated by IL12/15/18 to enhance fitness, persistence, and function, however, 1 donor to 1 recipient cells product hinders the therapeutic development. W-NK is derived from healthy human peripheral blood (PBMC) cNK cells with scalable manufacturability. Compared to adoptive cell therapy, NK cells are not restricted to a single antigen target, thereby mitigating antigen escape. Prior studies have indicated that cytokine-induced ML NK cells is associated with changes in T cells and dendritic cells in the tumor microenvironment in r/r AML. Bulk RNA sequencing indicated that expression of genes involved in metabolic gene programs such as glycolysis and amino acid metabolism is enriched in WU-NK-101 cells compared to cNK cells. This finding was confirmed by mass spectrometry analysis of WU-NK-101 cell lysates, which indicated a protein profile enriched in cell division pathways and metabolic pathways, including oxidative phosphorylation. In comparison to cNK cells, WU-NK-101 cells exhibited higher levels of proliferation markers (Ki67), costimulatory and activating receptors including IFN-gamma and IL-2/STAT5 signaling, cytotoxic effector protein, and lower level of inhibitory receptions and cellular maturation markers. WU-NK-101 also expressed high levels of CXCR3 and CXCR4, with homing to bone marrow. When co-cultured with HL-60 leukemia cells, WU-NK-101 cells exhibited higher cytotoxicity, and no off-target cytotoxicity. In contrast to cNK cells, Cytotoxicity of WU-NK-101 cells was not affected by hypoxic conditions and only marginally impacted in conditions mimicking an immunosuppressive environment. In a mouse xenograft model of AML, WU-NK-101, as one infusion and as three infusions, efficiently controlled tumor burden in mouse models of AML in a dose dependent manner. Data suggest that in preclinical models of AML, WU-NK-101 cells have enhanced effector functions, metabolic activity, robust homing to BM, to maintain effective anti-tumor activity even in a nutrient-depleted, immunosuppressive tumor microenvironment. A clinical trial of WU-NK-101 for patients with r/r AML is currently recruiting patients (NCT#05470140).
Phase II study of the CD20- and CD3- targeting bispecific antibody mosunetuzumab in combination with CD79b-targeting antibody drug conjugate polatuzumab vedotin for B-cell non-Hodgkin lymphoma
613. Mosunetuzumab plus polatuzumab vedotin demonstrates a favorable safety profile and efficacy in patients (pts) with relapsed or refractory (R/R) large B-cell lymphoma (LBCL): Primary analysis of a phase Ib/II study
Elizabeth Lihua Budde (City of Hope National Medical Center, Duarte, CA, USA) presented primary analysis from a Phase II expansion cohort of a study of mosunetuzumab (mosun), a CD20xCD3 T cell engaging bispecific antibody, in combination with CD79b-targeting antibody drug conjugate polatuzumab vedotin (Pola) in patients with R/R B-cell non-Hodgkin lymphoma (B-NHL). 98 patients in the dose expansion cohort received a fixed duration of mosunetuzumab and Pola (M-Pola), and patient hospitalization was not mandatory during treatment administration. Patients had received a median of 2 prior lines of therapy. 35.7% (n=35) had received prior CAR T therapy, and of those patients, 26 had disease refractory to CAR T-cell therapy. 57% of patients had primary refractory disease. After 23.9 months of follow-up, investigator-assessed overall response rate (ORR) and complete response rate (CRR) were 63.3% and 51.0%, respectively and ORR and CRR by independent review were 59.2% and 45.9%, respectively. Clinically meaningful ORRs were observed across all patient subgroups, including prior CAR T cell therapy. Responses were durable, with a median duration of response (DoR) of 20.8 months and a 24-month event-free rate of 49.7%. Duration of complete response (DoCR) was not reached, and the 24-month event-free rate among patients with a CR was 60.8%. The median progression-free survival and overall survival (OS) were 11.4 months and 23.3 months, respectively. Among patients who had received prior CAR T cell therapy (n=35), best ORR was 57.1%, median DoR was not reached, CRR was 40%, and the median DoCR was not reached. Median PFS and OS was 9.6 months and 15.2 months, respectively. Adverse events (AEs) of grade 3 or 4 were observed in 55.1% of patients, and treatment-related AEs (TRAEs) of grade 3 or 4 were observed in 34.7% of patients. Grade 5 AEs occurred in 3.1% of patients, and none were TRAEs. Cytokine release syndrome (CRS) occurred in 18.4% of patients, and 3.1% of cases were grade 3. All cases of CRS were confined to cycle 1 of mosun, during step-up dosing. ICANS occurred in 5.1% of patients (2.0% grade 3 or 4). The median time to B cell recovery was 12.4 months from completion of therapy (n=27). These results indicate that fixed-duration administration of M-Pola induces durable responses in patients with R/R LBCL, including those who received prior CAR T cell therapy. Administration of M-Pola on an outpatient basis was associated with a manageable safety profile, and CRS incidence was low. SUNMO, a phase III study of M-pola as second-line + treatment of transplant-ineligible patients with R/R LBCL is currently enrolling patients.
A novel mechanism of PDL1 blockade resistance in B-cell acute lymphoblastic leukemia: differentiation of leukemic-specific CD4+ CD40L-high IL-10-low T cells into IL-10-high, Tr1-like state.
597. A specialized CD4+ T-cell subset protects residual leukemic cells from immune surveillance, enabling relapse
Sean Tracy (University of Minnesota, Saint Paul, MN, USA) presented a study of the role of CD4+ T cells in anti-leukemia immunity using a mouse model of B cell acute lymphoblastic leukemia (B-ALL). Previous studies indicate that persisting minimal residual disease (MRD) in the absence of PD-L1 blockade is associated with polyclonal expansion of PD1+ TIM3+ FOXP3- CD4+ T cells that expresses high level of suppressive cytokine IL10, predicting relapse. Multiomic analyses of CD4+ T cells of mice challenged with Ph+ B-ALL identified a class of CD4+ TIM3+ FOXP3- T cells that expressed high levels of IL-10, indicative of a T-regulatory type 1 (Tr1) identity, producing proliferative and differentiation signal towards mutation-harboring hematopoietic stem cells (HSC), and creating an immunosuppressive niche. Successful therapy with nilotinib and PD-L1 blockade led to a marked decrease in T cell receptor diversity of Tr1 cells, suggesting expansion of a specific dominant clonotype (DC) of Tr1, which expressed lower levels of IL-10 and higher levels of CD40L, instead. In order to understand the mechanism of Tr1s in protecting leukemic cells from immune attack, the group made a transgenic mouse (HV1) that expressed a fixed TCR with a sequence derived from DC Tr1 cells, and HV1 cells were adoptively transferred to naïve recipient mice which were challenged with the LM138 leukemia cell line. Transfer of HV1 cells prolonged survival of mice with leukemia, but mice experienced relapse at later time points suggesting the anti-leukemic activity of HV1 cells is not maintained in the long term. Single cell analyses of HV1 cells indicate that naïve HV1 cells are mostly homogenous but by day 17 after transfer, they expand and differentiate into T helper 1- (Th1-) or T-follicular helper- (Tfh-) states by Day 9 after cell transfer. Chromatin at the Foxp3 locus was rarely accessible during HV1 expansion and differentiation. Both Th1 and Tfh cells have high levels of IL-10 chromatin accessibility, suggesting both types of cells adopt an IL-10-high phenotype. HV1 cells did not develop features of exhaustion, as measured by tumor necrosis factor and interferon gamma expression. Flow cytometry analysis indicated that HV1 cells expressed CD40L early after activation, but over time, IL-10 expression was upregulated and CD40L expression was lost. An independent study recently reported that antigen presentation by mutated hematopoeitic stem cells induces differentiation of CD4+ T cells to a Tr1-like state. These data support a novel mechanism underlying resistance to immune checkpoint blockade, in which leukemia-specific CD4+ CD40L-high IL-10-low T cells that originally correlate with eradication of MRD differentiate over time to adopt an IL-10-high, Tr1-like state, eliciting an immunosuppressive environment and protecting leukemia cells from immune surveillance.
CD8 T cell profiles of patients treated with blinatumomab for B cell precursor acute lymphoblastic leukemia.
600. Single-cell transcriptomic profiling of T cells from blinatumomab-treated patients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) reveals circulating CD8 T cell subsets
Francesco Corrado (Humanitas University, Pieve Emanuele, Milan, Italy) presented a study using cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) to identify pre-treatment T cell features associated with response to blinatumomab (blina), a bispecific CD3xCD19 T cell engager for B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Peripheral blood of 13 patients receiving blina for BCP-ALL (10 responders and 3 non-responders) was collected before treatment start and every 7 days during treatment. 12 distinct CD3 positive T cell clusters were defined, and this presentation focused on CD8-positive cells. Baseline CD8 T cells from responders were enriched with Granzyme K (GZMK) positive, effector and memory-like cell types and exhibited lower levels of CD8 naïve T cells. CD8 T cells from responders were enriched for cytotoxicity, while CD8 T cells from non-responders expressed higher levels of genes associated with T cell exhaustion and dysfunction. Serum proteomic analyses indicated a shared immune inhibitory proteomic profile, including upregulated levels of IL-12 receptor subunit beta 1, possibly reflecting differences in tumor microenvironment immune composition. These results identify cell sets and gene expression profiles involved in T cell function in the presence of the bispecific T cell engager blinotumomab. Future studies include confirming these results in an independent patient cohort and carrying out similar cellular analyses of peripheral monocytes from patients treated with blinotumomab.
Updated results of ZUMA-12: Axi-cel as first-line therapy for patients with high-risk large B cell lymphoma
894. 3-year analysis of ZUMA-12: A phase 2 Study of axicabtagene ciloleucel (axi-cel) as first-line therapy in patients with high-risk large B-cell lymphoma (LBCL)
Julio C. Chavez (Moffitt Cancer Center, Tampa, FL, USA) reported updated efficacy and safety outcomes from ZUMA-12, a phase II single-arm study of CD19-targeting CAR T cell axicabtagene ciloleucel (axi-cel) as part of first-line treatment in patients with high-risk large B-cell lymphoma (LBCL), defined as a) with MYC and BCL2 and or BCL6 rearrangement (double- or triple-hit histology) or b) International Prognostic Index score = 3 plus positive interim PET per Lugano classification (Deauville score 4/5) after 2 cycles of an anti-CD20 monoclonal antibody and anthracycline-containing regimen. Axi-cel has been approved for treatment of early relapsed or refractory LBCL. Non-chemotherapy bridging could be administered. 40 patients with high-risk LBCL received lymphodepleting chemotherapy followed by axi-cel infusion; data from 37 patients were evaluable for response, and safety data were reported for all patients treated with axi-cel (n=40). Previously reported primary efficacy analysis (n=37) showed a overall response rate (ORR) was 89%, and complete response (CR) rate was 78%. Since the primary efficacy analysis, 8 patients progressed from a partial response to a CR, and 1 patient progressed from stable disease to CR. At this report with a median follow-up of 40.9 months, ORR was 92%, and CR rate was 86%. Responses were ongoing in 73% of response-evaluable patients. Median duration of response (DoR) was not reached, with 3-year DoR rates of 81.8% among the total efficacy evaluable patient population and 84.4% among patients who achieved CR. Event-free survival (EFS) was not reached, with 3-year EFS rates of 73.0% among the total efficacy evaluable patient population and 84.4% among patients who achieved CR. Median progression-free survival (PFS) and overall survival (OS) were not reached in efficacy evaluable patients, and among patients who achieved a CR, 3-year PFS and OS rates were 84.4% and 90.6%, respectively. A total of 8 deaths occurred in ZUMA-12, and 5 deaths were due to progressive disease. No new cases of cytokine release syndrome (CRS) or neurologic events were reported. No significant differences in CAR T cell expansion were observed between responders, patients with relapse, and non-responders. Final analysis from ZUMA-12 indicates that axi-cel demonstrates a high rate of durable responses among patients with untreated high-risk LBCL. To determine whether axi-cel specifically benefits patients with high-risk LBCL exposed to fewer prior therapies, the phase III ZUMA-23 trial is comparing first-line axi-cel to standard of care chemoimmunotherapy in patients with high-risk LBCL.
Granulocytic myeloid-derived suppressor cells as a predictive biomarker of CAR T cell failure
1015. High levels of circulating granulocytic myeloid-derived suppressor cells (G-MDSCs) predict failure of CD19-targeting CAR-T cell therapy
Federica Sorà (Università Cattolica del Sacro Cuore, Rome, Italy) reported a study of the predictive role of circulating myeloid-derived suppressor cells (MDSCs) of monocytic (M-MDSC) and granulocytic (G-MDSC) origin in CAR T therapy. MDSCs are known to promote immunosuppression and tumor progression in hematological malignancy, lessening the anti-tumor immune response and the efficacy of CAR-T therapy. Prior studies indicate that high baseline levels of M-MDSCs are associated with lower CAR T cell expansion and poor prognosis for large B cell lymphoma (LBCL). Blood samples were collected from 45 patients receiving CAR-T cell therapy for high-grade B cell lymphoma or acute lymphoblastic leukemia before (day –5) and after lymphodepletion (day 0), and levels of circulating G-MDSCs were quantified. Median G-MDSC counts on day –5 and day 0 were 5.8 and 15.6 cells/ul, respectively. At a median follow-up of 188 days, superior progression-free survival (PFS) was observed among patients with less than 24.6 G-MDSC/ul (low G-MDSC) on day 0 compared to patients with 24.6 G-MDSC/ul or higher (high G-MDSC), with 3-year PFS rates of 73% and 17%, respectively (OR 4.25, p=0.003). A similar pattern was observed in patients with diffuse large B cell lymphoma (DLBCL), low G-MDSC was associated with significantly higher PFS compared to high G-MDSC, with 12-month PFS rates of 64% and 17%, respectively (p=0.047). Low G-MDSC was associated with significantly higher levels of CAR T expansion at day 14 (p=0.004). Results indicate that high circulating levels of G-MDSCs before CAR T infusion correlate with better CAR T expansion and higher PFS rates. Although these results need to be confirmed and validated in larger studies, G-MDSCs may represent a promising predictive biomarker of failure of CAR T therapy and a potential point of intervention to improve CAR T efficacy.
Phase I study of anitocabtagene autoleucel, a novel BCMA-targeting CAR T cell therapy for relapsed or refractory multiple myeloma
1023. Phase 1 Study of CART-ddBCMA for the treatment of patients with relapsed and/or refractory multiple myeloma: Results from at least 1-year follow-up in all patients
Matthew J. Frigault (Harvard Medical School, Massachusetts General Hospital Cancer Center, Boston, MA, USA) reported results from a phase I, first in-human study of anitocabtagene autoleucel (anito-cel)/CART-ddBCMA, an autologous anti-B cell maturation antigen (BCMA) CAR T cell in patients with relapsed or refractory (r/r) multiple myeloma (MM). Anito-cel utilizes a D-domain binder, a unique synthetic binding domain that is highly compact and stable, which results in a high proportion of CAR+ cells and high levels of CAR expression, potentially promoting more efficient MM cell killing. 40 patients enrolled, 38 received CART-ddBCMA; 2 patients who were not dosed did have cell product manufactured but were ineligible for cell infusion due to medical complications. Two dose levels (DL1 of 100 +/- 20% x 106 or DL2 of 300 +/- 20% x 106 were evaluated. The patient population was a median age of 66 years old and had received a median of 4 prior lines of therapy. All patients were triple-refractory, and 26 (68%) were penta-refractory. 63% of patients had high-risk features, including 9 patients (24%) with high tumor burden of more than 60% bone marrow plasma cells, 13 (34%) with extramedullary disease (EMD), and 11 (29%) with high-risk cytogenetics. At a median follow-up of 26.5 months, the overall response rate (ORR) was 100%, and 29 patients (76%) achieved a complete response (CR) or stringent CR (sCR). Median progression-free survival (PFS) was not met and estimated 24-month PFS was 56%. 89% of patients were minimal residual disease (MRD) negative. Among the 13 patients with EMD, median PFS was not reached, and estimated 24-month PFS was 57.5%. At approximately 2 years of follow-up, the median PFS had not been reached. No new safety signals were observed. No cases of cytokine release syndrome (CRS) of grade 3 or higher occurred at DL1, and one case occurred at DL2. No atypical neurotoxicities were observed, and only one grade 5 adverse event occurred, but it was not related to the study drug. Based on these results, a dose of 115 +/- 10 x 106 cells was recommended for the phase 2 study. Based on this phase I study, anito-cel is well-tolerated, and it is associated with a high ORR and durable responses, specifically in high-risk populations of patients with r/r MM. The phase II IMMagine-1 trial of anito-cel for r/r MM is currently enrolling patients.
Phase I dose escalation study of a tri-specific BCMA-targeting T cell engager for relapsed/refractory multiple myeloma
1012. Results from the completed dose escalation portion of the phase 1 study of HPN217, a half-life extended tri-specific T cell activating construct (TriTAC®) targeting B cell maturation antigen (BCMA) for relapsed/refractory multiple myeloma (MM)
Sumit Madan (Banner MD Anderson Cancer Center, Gilbert, AZ, USA) reported the completed dose escalation portion of HPN217-3001, a first in-human Phase I study of HPN217, a half-life-extended BCMA-targeting T cell engager for relapsed or refractory (r/r) multiple myeloma (MM). HPN217 contains 3 binding domains, anti-BCMA for MM cell binding, anti-CD3 for T cell engagement, and anti-albumin for half-life extension. The absence of an Fc domain eliminates Fc receptor binding, minimizing T cell activation in the absence of target cells and improving the therapeutic window. 97 patients were enrolled across 15 dose escalation cohorts. Of the two highest target doses, 19 patients were enrolled in the 12 mg target dose cohort and 33 patients in the 24 mg target dose cohort. 20 (21%) of all patients in the study had received prior BCMA-targeting treatment, and 88% had disease refractory to the last line of therapy. Cytokine release syndrome (CRS) occurred in 29 (30%) patients, and all cases were of grade 1 or 2, except for two cases of grade 3 CRS in the 24 mg dose cohort. The most common grade 3-4 toxicity is cytopenia and transaminitis. Dose-limiting toxicities (DLT) were observed in 2 patients at the 2.86 mg fixed dose escalation cohort with grade 3-4 transaminitis. There was no DLT in the step dose escalation cohort. ICANS occurred in 3 patients, and all cases were grade 1. One treatment-related death from traumatic subdural hematoma. Overall response rate (ORR) at doses of 2.15 to 6 mg was 39%; ORR of the 12 mg and 24 mg target dose cohorts were 63% and 45%, respectively, and 53% of patients in the 12 mg target dose cohort achieved a very good partial response or better. 22 of 38 responders remain on treatment, and 7 responders had previously received BCMA-targeting therapy. Median duration of response was 20.5 months, and median time to first response was 1.2 months. These results indicate that the 12 mg and 24 mg doses of HPN217 are well-tolerated and associated with early and durable clinical responses in patients with r/r MM. Based on safety and efficacy data, step-up dosing to a target dose 12 mg will be used for future clinical development.
Inhibition of epigenetic modifier EZH2 improves efficacy of CD19-targeting CAR T cell therapy in mouse models of lymphoma
1018. Inhibition of EZH2 improves CART19 immunotherapy by reprogramming lymphoma tumor cells and enhancing T-cell functionality
Patrizia Porazzi (University of Pennsylvania, Philadelphia, PA, USA) presented a study of the biological role of epigenetic modulator EZH2 in patient response to CD19-targeting CAR T cell (CART19) therapy. EZH2 is a histone methyltransferase that represses gene expression and is known to play a role in lymphomagenesis and immune evasion. In a study of 85 patients receiving CD19-targeting T cell therapy for lymphoma, patients with activating mutations in EZH2 (n=14) had a lower complete response rate (CRR) compared to patients with wild-type EZH2 (n=71; CRR 14.3% vs. 42.3%, p=0.048). Pre-treatment of a diffuse large B cell lymphoma (DLBCL) cell line with EZH2 inhibitor tazemetostat significantly increased cell killing by CART19 in vitro, compared to untreated cells. The addition of tazemetostat also significantly increased CART19 production of inflammatory cytokines and enhanced long-term CART19 expansion and cytotoxicity. This system was tested in vivo, where immunodeficient mice were implanted with the DLBCL cell line, treated with vehicle or tazemetostat, and infused with CART19. Tazemetostat in combination with CART19 exhibited better tumor control compared to CART19 alone and improved survival. At 20 days post-CART19 infusion, tazemetostat significantly increased CART19 expansion compared to treatment with vehicle, and increased the levels of naïve and early-memory CD4 T cells were found in peripheral blood. Tumors from mice that received prolonged (2 week) pre-treatment with tazemetostat prior to CART19 exhibited increased T cell infiltration at the tumor site, and tumor-infiltrating CAR T cells from tazemetostat-treated mice exhibited higher levels of activation and cytotoxicity. Bulk RNA sequencing of tumor cells from tazemetostat-treated mice indicated increased B cell activation and apoptosis. These results suggest that inhibiting epigenetic modifier EZH2 reprograms anti-lymphoma immunity, making tumor cells more susceptible to CART19 killing and reducing differentiation of CART19 cells. This approach has potential to increase long-term efficacy of CD19-targeting CAR T therapy in patients with DLBCL.