Overall survival analysis of HARMONi-A: PD-1 x VEGF bispecific antibody ivonescimab with chemotherapy for non-small cell lung cancer
1348. Final overall survival analysis of HARMONi-A study comparing ivonescimab plus chemotherapy to chemotherapy alone in patients with EGFR+ NSCLC progressed on EGFR-TKI treatment
Xiuning Le (The University of Texas MD Anderson Cancer Center, Houston, TX, United States) presented the final and only overall survival (OS) analysis of HARMONi-A study investigating ivonescimab with chemotherapy versus chemotherapy in patients with non-small cell lung cancer (NSCLC) with epidermal growth factor receptor (EGFR) mutations (EGFR-positive NSCLC) whose disease progressed while receiving treatment with EGFR tyrosine kinase inhibitors (EGFR-TKIs). Ivonescimab is a first-in-class bispecific antibody targeting PD-1 and vascular endothelial growth factor (VEGF). EGFR mutations occur in approximately 15% of cases of NSCLC among White people and 40 to 50% of people of Asian descent. The first line of treatment for EGFR-positive NSCLC is EGFR-TKIs, but if acquired resistance occurs, subsequent treatment options are limited. Prior studies indicate that adding PD-(L)1 blockade to chemotherapy provides no additional benefit for these patients. Adding an anti-angiogenic agent with chemotherapy and PD-(L)1 blockade is associated with progression-free survival (PFS) benefits but not overall survival (OS) benefits. 322 patients in China were randomized to receive ivonescimab with pemetrexed and carboplatin (ivonescimab + chemotherapy; n=161) or placebo with pemetrexed and carboplatin (placebo + chemotherapy; n=161). Maintenance therapy with pemetrexed was allowed in both arms. In a report from March 2023, at a median follow-up of 7.9 months, ivonescimab + chemotherapy was associated with significant improvements in PFS (HR 0.46, p<0.001). At a median follow-up of 32.5 months, the median OS was 16.8 months with ivonescimab + chemotherapy, compared to 14.1 months with placebo + chemotherapy (HR 0.74, p=0.019). OS benefits with ivonescimab + chemotherapy were observed in all patient subgroups analyzed, including in patients with and without brain metastases (HR 0.61 and 0.77, respectively) and in patients with EGFR 19Del and EGFR L858R alleles (HR 0.83 and 0.41, respectively). 62.7% of patients in the ivonescimab + chemotherapy arm and 72.0% of patients in the placebo + chemotherapy arm went on to receive subsequent anti-tumor activity, including targeted therapy (50.3% and 54.0%, respectively), chemotherapy (39.8% and 43.5%), and immune checkpoint inhibitors (12.4% and 10.6%, respectively). No new safety signals were observed, and most toxicities observed in both arms were related to chemotherapy. VEGF-related toxicities of low albumin (23.7%), proteinuria (23.7%), and hypertension (10.6%) were observed in the ivonescimab + chemotherapy arm, and the majority of cases were grade 1 or 2. Bleeding events were comparable between treatment arms. With its statistically significant survival benefit, ivonescimab with chemotherapy may potentially represent a new standard of care for patients with EGFR-positive NSCLC whose disease progressed with EGFR-TKI. A New Drug Application will be submitted to the United States Food and Drug Administration in the near future.
First in-human study of in vivo CAR myeloid cell therapy for solid tumors
1342. First-in-human dose escalation study to investigate the safety, pharmacokinetics, pharmacodynamics and initial efficacy of mRNA-LNP MT-302 in vivo CAR therapy in solid tumors
Rasha Cosman (The Kinghorn Cancer Centre, St. Vincent’s Hospital, Darlinghurst, Australia) presented results from a phase 1 trial of of MT-302, a TROP2-targeting mRNA lipid nanoparticle, part of the CREATE platform for in vivo CAR myeloid cell therapy, for advanced solid tumors. 27 patients representing 11 different tumor types participated in the dose escalation study, receiving MT-302 every two weeks. The patient population was heavily pre-treated, with 48% of patients receiving four or more prior systemic therapies, 41% of patients had received prior immunotherapy, and 7.4% had received prior TROP2-targeting therapy. 81% of patients had at least three sites of disease. MT-302 was associated with a manageable safety profile. 51.9% of patients experienced low-grade cytokine release syndrome (CRS), and no cases of CRS of grade 3 or higher were observed. One patient who received the highest dose level of MT-302 (0.15 mg/kg) experienced grade 4 immune effector cell-associated neurotoxicity syndrome (ICANS), and treatment was discontinued. The maximum tolerated dose was found to be 0.10 mg/kg. Patients tolerated repeat dosing, receiving an average of 4 doses of MT-302. Levels of proinflammatory cytokines TNF-alpha and IL-1 beta did not increase throughout the course of treatment, further reinforcing the two-week dosing schedule. Among the 18 patients with target lesion assessments, one patient with hormone receptor-positive breast cancer achieved a confirmed partial response (cPR), and 7 patients achieved stable disease. The patient with the cPR remained in treatment for 16 months, receiving 20 doses of MT-302. CAR mRNA-positive myeloid cells were detected in peripheral blood after MT-302 treatment, and CAR-positive myeloid cells were trafficked to tumors in the vicinity of tumor cells. Tumor-associated macrophages exhibited increased expression of pro-inflammatory genes like CXCL9 and downregulation of immunosuppressive M2-like markers, including SPP1, indicating a pro-inflammatory tumor microenvironment. MT-302 was also associated with increased levels of T cell-attracting chemokines CXCL10 and CCL-2 in peripheral blood and the tumor as well as dose-dependent increases in IFN gamma, and increased infiltration of T cells to the tumor. MT-302 is safe, well-tolerated and associated with preliminary signals of clinical benefit, including pro-inflammatory remodeling of the tumor microenvironment. These results provide proof-of-mechanism of in vivo myeloid CAR cells for solid tumors and support the further development of next-generation mRNA constructs for in vivo CAR therapy
Blockade of semaphorin 4D enhances immune checkpoint inhibition for metastatic melanoma
22. Neoadjuvant pepinemab enhances immune checkpoint blockade in metastatic melanoma characterized by biomarkers of TME reprogramming including tertiary lymphoid structures
Crystal Mallow (Vaccinex, Inc., Rochester, NY, United States) presented results from a phase 2 biomarker integrated clinical trial investigating neoadjuvant pepinemab in combination with nivolumab and/or ipilimumab in patients with resectable metastatic melanoma. Pepinemab is a monoclonal antibody targeting semaphorin 4D (Sema4D), a driver of myeloid immunosuppression and resistance to immune checkpoint inhibitors. Sema4D is expressed on tumor cells and interacts with suppressive myeloid-derived suppressor cells (MDSCs) to inhibit T cell function. 32 patients with resectable metastatic melanoma received two preoperative doses of pepinemab + nivolumab + ipilimumab (n=8), pepinemab + nivolumab (n=8), pepinemab + ipilimumab (n=8), or nivolumab (n=8) followed by definitive intent surgery, and 6 patients received surgery alone. Samples from patients who received preoperative nivolumab and ipilimumab were used as synthetic controls. Multiomic biomarker analysis of resected on-treatment lymph nodes indicated pepinemab treatment was associated with decreased levels of Sema4D-expressing tumor cells at the periphery of the tumor bed. The myeloid landscape of on-treatment lymph nodes had shifted from immunosuppressive to immune-activating, especially with pepinemab + nivolumab + ipilimumab, and in pepinemab + nivolumab + ipilimumab-treated tumors, activated antigen presenting cells were located throughout the tumor instead of being restricted to the tumor periphery, as observed in nivolumab-treated tumors. Neoadjuvant pepinemab was also associated with increased infiltration of CD4 and CD8 T cells into tumors and increased infiltration of B cell aggregates, especially with the triple therapy regimen of pepinemab + nivolumab + ipilimumab. Triple therapy was also associated with increased levels of mature lymphoid aggregates, characterized by specific zones of immune cells and germinal center markers. Increased mature lymphoid aggregates correlated with enhanced recurrence-free survival (p=0.01). The major pathologic response (mPR) rate was 75% with pepinemab + nivolumab + ipilimumab, 37.5% with pepinemab + nivolumab, 37.5% with nivolumab, and 12.5% with pepinemab + ipilimumab, and pepinemab + nivolumab + ipilimumab was also associated with enhanced recurrent-free survival compared to the other treatments. These data support the hypothesis that Sema4D promotes myeloid cell-mediated immunosuppression in the TME, which can be released through blockade of Sema4D with pepinemab, thus enhancing the efficacy of neoadjuvant immune checkpoint blockade.
First in-human study of ERVE-4 TCR T cells for metastatic clear cell renal cell carcinoma
506. Immune cell composition analysis of ERVE-4 T-cell receptor gene therapy infusion products in metastatic ccRCC: Insights from a first-in-human trial
Rosa Nadal Rios (Fred Hutchinson Cancer Center, Seattle, WA, United States) presented a phase 1 dose escalation study to evaluate the safety and efficacy of adoptive T cell receptor (TCR)-T cell therapy targeting ERVE-4 in patients with metastatic clear cell renal cell carcinoma (ccRCC). ERVE-4 is an endogenous retrovirus expressed by most ccRCC tumors. ERVE-4 expression is tumor-specific, and an ERVE-4 derived peptide, CT-RCC1, is immunogenic. A cytotoxic T cell clone targeting CT-RCC1 presented on HLA-A*11 was isolated from a patient with metastatic ccRCC, and T cells transduced with the TCR from this clone has been shown to kill tumors in vitro and in vivo. 17 HLA-A*11:01-positive patients with metastatic ccRCC underwent lymphodepletion and received a one-time infusion of ERVE-4 TCR transduced T cells (ERVE-4 T cells). Patients received IL-2 after infusion to support T cell expansion. Four dose levels were used in the study, ranging from 1 million to 50 million ERVE-4 T cells per kilogram of body weight. Toxicities were manageable, no dose-limiting toxicities were observed, and most adverse events were associated with lymphodepletion or the low-dose IL-2 regimen. No clinically meaningful benefits were observed in the patient population. One patient achieved a partial response, and most patients (n=10) experienced progressive disease. Median progression-free survival was 60 days, and most patients received systemic treatment and/or metastasis-directed therapy for progressive disease after ERV-4 T cell infusion. ERVE-4 T cells were detected in peripheral blood at moderate levels, even at the highest dose levels, and peak concentration occurred 7 days after infusion. Elevated levels of cytokines including IFN-gamma and IL-6 were observed in peripheral blood after infusion, and this elevation was transient. ERVE-4 T cell infusion products were characterized by flow cytometry to provide insights to the mechanisms underlying their limited clinical activity. Infusion products consisted of cell populations that were predominantly T effector memory cells with low levels of T central memory cells, and high approximately 20% and 50% of CD3 T cells in infusion products expressed PD-1 and TIM3, respectively, indicating a terminally exhausted phenotype. Although ERVE-4 T cells had limited clinical efficacy, this study represents the first trial of an adoptive cellular therapy targeting an ERVE-4 product. Manufacturing of cells was feasible, and treatment administration was safe. The limited clinical activity was likely due to the predominance of terminally differentiated exhausted T cells. Future directions include improving TCR T cell engineering and manufacturing processes to enhance the proliferative capacity of the cells and developing ERVE-4-targeting CAR T cells.
Elucidating the role of FAS in bystander killing by CAR T cells
210. Mechanisms and impact of bystander killing by CAR T cells
Joanna Chorazeczewski (University of Chicago, Chicago, IL, United States) presented a translational study investigating the association FAS expression with the cancer-killing activity of CAR T cells. Although CD-19-targeting CAR T therapy has improved outcomes in patients with relapsed/refractory diffuse large B cell lymphoma (r/r DLBCL), relapse is common, and approximately 50% of patients fail to have a long-term response to CAR T therapy. T cell-mediated killing occurs through the granzyme/perforin pathway and through the binding of FAS to the FASL receptor pathway. Previous studies indicate that high levels of FAS expression correlate with favorable CAR T survival outcomes, even in CD19-low DLBCL. This association is specific to CAR T cell therapy; similar associations of high FAS and survival are not observed with chemotherapy. Other studies indicate that a high death receptor signature also correlates with improved event-free survival with CD19-targeting axicabtagene ciloleucel for refractory large B cell lymphoma. In order to characterize the dependence of CAR T cell killing on FAS expression, murine lymphoma cell lines were co-cultured with CD19-targeting CAR T cells. Bystander killing of CD19-negative cells was only observed when CAR T cells were co-cultured with CD19-positive and CD19-negative cells. Incubating CD19-targeting CAR T cells with co-cultures of CD19-positive and FAS-negative CD-19 negative cells indicated FAS expression was also required for bystander killing in vitro. A mouse xenograft system of human cell line in immunodeficient NSG mice exhibited similar patterns in vivo: optimal levels of bystander killing of CD19-negative cells required the presence of CD19-positive cells and FAS. These results were also verified in a mixed tumor system in syngeneic murine system. In a bioluminescence experiment, mice were injected with two mixed tumors and treated with CD19-targeting CAR T cells. The tumor in one flank of the mouse consisted of FAS-positive CD19-negative cells expressing luciferase and FAS-positive CD19-positive cells not expressing luciferase. The tumor in the other flank of the mouse consisted of FAS-negative CD19-negative cells expressing luciferase and FAS-positive CD19-positive cells not expressing luciferase. Killing of CD19-negative tumor cells was monitored by loss of bioluminescence. FAS-dependent and FAS-independent killing was observed in both flanks of the mice, but two weeks after CAR T cell administration, the remaining tumor cells were FAS-negative CD19-negative cells, suggesting FAS-dependent killing dominates in vivo. CAR T cell bystander killing was enhanced with brinapant, an activator of the apoptotic pathway , in the murine syngeneic system, indicating the FAS/FASL pathway could manipulated pharmaceutically. Mice overexpressing a non-cleavable allele of FASL also exhibited increased CAR T cell bystander killing, indicating the FAS/FASL pathway could also be manipulated genetically. Results from this study indicate that FAS is necessary for CAR T cell bystander killing of CD19-low or CD-19 negative cells. This study also suggests that manipulation of FAS/FASL interactions could potentially enhance the efficacy of CD19-targeting T cells for CD19-low or -heterogenous DLBCL.
Updated results of the IGNYTE trial: oncolytic virus RP1 plus nivolumab for melanoma that progressed with PD-1 blockade
1327. Biomarker and updated clinical data for RP1 plus nivolumab in anti–PD-1–failed melanoma from the IGNYTE trial demonstrate reversal of mechanisms of resistance to immune checkpoint blockade
Trisha Wise-Draper (University of Cincinnati, Cincinnati, OH, United States) presented updated biomarker and clinical data from the IGNYTE trial of vusolimogene oderoarepvec (RP1), a herpes simplex virus type 1-based oncolytic virus expressing GM-CSF and GALV-GP-R, in combination with nivolumab for advanced melanoma that progressed during treatment with anti-PD-1 therapy. 140 patients were treated with RP1 and nivolumab every two weeks for eight cycles, followed by nivolumab every 4 weeks for 21 cycles. 55.7% of the patient population was PD-L1 negative, and 65.7% had primary resistance to anti-PD-1 therapy. The objective response rate (ORR) in all patients was 33.6%. Among patients with primary resistance to anti-PD-1 therapy, the ORR was 34.8%, and the ORR was 24.4% in patients with PD-L1 negative disease. The median duration of response for the entire patient population was 24.8 months. Tumor biopsies taken from the same lesion pre-treatment and 43 days after the first dose of RP1 indicated that RP1 + nivolumab improved CD8 T cell infiltration and PD-L1 expression in the tumor. Intratumoral levels of activated (PD-1 positive) CD8 T cells and CD68-positive macrophages also increased. CD8 T cell infiltration and PD-L1 expression increased in tumors regardless of the type and duration of prior treatments. 57% of all lesions exhibited increased levels of PD-L1 expression, and 47% of lesions exhibited increased infiltration with CD8 T cells. RP1 + nivolumab also enhanced the IFN gamma signature in responders compared to pre-treatment levels. Further characterization of T cell clones in three patients indicated RP1 + nivolumab was associated with expansion of HSV-1- and melanoma-specific T cell clones, indicating a systemic anti-tumor immune response. The safety profile of RP1 + nivolumab was consistent with previous reports. RP1 in combination with nivolumab is associated with a clinically meaningful response rate and durable response rates in a patient population that is difficult to treat. RP1 combined with nivolumab upregulates gene signatures associated with responsiveness to PD-1 blockade, providing promising evidence that RP1 can overcome multiple mechanisms of resistance to PD-1 blockade.