Abstract 771: Multi-omics profiling can predict for relapse and response to brexucabtagene autoleucel CAR T-cell therapy in patients with mantle cell lymphoma

Abstract

Abstract Background Due to excellent clinical response, brexucabtagene autoleucel (BA), an anti-CD19 chimeric antigen receptor T (CAR T) cell therapy, was approved by FDA to treat patients with MCL. However, outcomes of BA-relapsed patients are very poor. Therefore, in this study, we performed multi-omics profiling for BA-treated patients and explore the potential mechanisms of response/resistance to BA therapy in MCL. Methods We performed single cell transcriptomic profiling for longitudinally collected primary patient samples (n = 39) from 15 patients (1 refractory, 14 CR and 5 relapsed). Cytokine multiplex analysis of 80 analytes were performed for longitudinal patient plasma samples (n = 80) for 18 patients (17 CR, 5 relapsed and 1 refractory). 15 patients had data available for both analyses. Functional studies were performed to further validate the findings from multi-omic analysis. Results Using single cell RNA sequencing, we demonstrated that the fraction of CD4/CD8 cytotoxic T cells among total immune cells were significantly reduced at relapse compared to those at pretreatment or at BA remission. Furthermore, T cell exhaustion at relapse was demonstrated by upregulated immune checkpoint molecules TIGIT, CD96, and LAG3 in CD4/CD8 cytotoxic T cells. Ex vivo stimulation induced T cell activation to produce IFNγ and IL-2 in the samples collected at remission, but not at relapse. These data showed an acquisition of exhausted T cell state at relapse. In contrast, a significant elevation in myeloid cells was detected in samples at relapse compared to those at pretreatment or at BA remission. Myeloid-derived suppressive cells were enriched at relapse. Together, these data may indicate tumor immune escape at relapse by inducing a state of T cell exhaustion and expanded MDSCs. To further understand the tumor microenvironment, we performed longitudinal profiling of cytokines in patient plasma. A panel of soluble checkpoint molecules including PD-L1, PD-L2, TIM3, and LAG3 were significantly decreased at remission, compared to baseline and resurrected back to baseline at relapse. Moreover, soluble IL2 receptor (sIL2R) was significantly decreased at remission but was elevated at relapse at levels even higher than baseline. Ex vivo stimulation with IL-2 plus sIL2R enhanced cell growth compared to IL-2, or sIL2R alone in the sample collected at pretreatment, but diminished cell growth in samples collected at relapse. This suggests that sIL2R may play a suppressive function for T cell expansion and activation. Conclusion This study is the first study reporting on exploring potential mechanisms of resistance to BA therapy in patients with MCL. We demonstrate a reprogrammed tumor-immune landscape inducing T cell exhaustion, enrichment of suppressive MDSC, and elevated sIL-2R and check point molecules leading to resistance and relapses after BA therapy. Citation Format: Vivian Changying Jiang, Dapeng Hao, Preetesh Jain, Yijing Li, Qingsong Cai, Yang Liu, Yuxuan Che, Jovanny Vargas, Yixin Yao, Lei Nie, Enyu Dai, Guangchun Han, Ruiping Wang, Kunal Rai, Andrew Futreal, Christopher Flowers, Linghua Wang, Michael Wang. Multi-omics profiling can predict for relapse and response to brexucabtagene autoleucel CAR T-cell therapy in patients with mantle cell lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 771.