Utilization of Therapy-Naïve T Cells to Enhance the Efficacy of CAR-T Cells in the Treatment of B-NHL Patients

Introduction Lymphoma treatment has been revolutionized by chimeric antigen receptor (CAR) T cell treatment. The success of this therapy hinges on the quality of the patients' T cells, as these are utilized to create a personalized treatment. CAR T-cell therapies are not yet approved as first-line treatment for B-cell non-Hodgkin's lymphoma (B-NHL), meaning the majority of patients undergo CAR T-cell treatment after two lines of previous therapy. Any preceding chemotherapy can impair T cell quality, thus impacting the source material to manufacture CAR T cells. We propose that T cells that have not been exposed to chemotherapy (“pre-therapy”) possess superior functional requirements for adoptive T cell therapies, which could lead to improved treatment efficacy. Methods We performed a multimodal analysis of patients´ T cells, including flow cytometry, epigenetics, single-cell transcriptomics, as well as functional studies. Blood samples were collected from B-NHL patients at initial diagnosis and before starting treatment (pre-therapy, n=14) and from B-NHL patients who had received two or more chemotherapy regimens (post-therapy, n=15) at the time of T-cell apheresis. Results Epigenomic analysis revealed differential methylation at over 10,000 individual CpG sites between the two patient cohorts, including an epigenetic signature in T cells of chemotherapy-exposed patients that is associated with CAR T cell treatment failure. In line with these findings, flow cytometric analysis revealed a shift towards a less naïve, increasingly exhausted phenotype (e.g. increased PD-1 expression) in the T cells of previously treated patients, which were also characterized by reduced in vitro proliferation. Single-cell RNA sequencing confirmed the phenotypic results and identified cohort-specific differences in gene expression. Notably, pretreated patients exhibited an upregulation of genes associated with reduced T cell functionality, such as SP140, LGALS1, and RGS9, and downregulation of genes critical for T cell development and function, including HIVEP2, PBX1 and TGFA. T cells from patients exposed to chemotherapy exhibited a gene expression that overlaps with established gene signatures correlating with T cell exhaustion, apoptosis, hypoxia, dysfunction and reduced persistence - traits associated with failure of adoptive T cell therapy. To ascertain whether these descriptive findings also translate into functional differences, we manufactured CAR-T cells utilizing T cells from pre-therapy versus post-therapy patients. Flow cytometric analysis of the final T cell products revealed a reduction of the previously observed differences in the two patient groups, indicating a convergence in the expression of phenotypic and exhaustion-associated markers after lentiviral transduction and in vitro culture. However, the epigenomic distinction remained evident in the generated CAR-T products and the CAR-T cells derived from untreated patients exhibited significantly superior cytotoxicity against the DLBCL cell line SUDHL10 in successive repetitive stimulations. Conclusion In conclusion, our comprehensive multi-omic analyses underline a decisive negative impact of prior chemotherapy on the quality of the T cells, triggering significant phenotypic, epigenetic and functional differences. Functional studies confirmed superior CAR T cell cytotoxicity against target cells if pre-therapy T cells serve as the source for CAR T cell manufacturing. This discovery holds significant implications not only for CAR T-cell therapy but also for the application of bispecific T-cell engagers (BiTEs), potentially enhancing the efficacy of both T cell-based treatments when used as a first-line therapy, as well as supporting ongoing investigations into CAR T cell treatment as a first-line therapy (Zuma-23).