Studying phenotypic states of antitumor human T cells

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

Following my recent appointment as a Stadtman tenure-track Investigator at the National Cancer Institute, Center for Cancer Research, I have now acquired key reagents and the necessary equipment and technology for carrying out my research. We have now begun acquiring data for multiple studies understanding the phenotypic states of antitumor T cells from cancer patients. 1) We have completed a study focused on identifying cell surface protein expression markers of antitumor CD4-positive T cells in patients with metastatic solid tumors. Comprehensive analysis of 45 antitumor CD4-positive TCR-clonotypes demonstrated uniquely high cell surface protein expression of ADGRG1, CD86, and CD57. In all samples containing tumor neoantigen-reactive CD4-positive TIL clonotypes, ADGRG1, and CD86 based cell surface enrichment of known reactive TCR-clonotypes were 11-fold and 9-fold higher than bulk CD4-positive TIL, respectively. ADGRG1-based cell surface isolation successfully identified antitumor, neoantigen-specific CD4-positive TCR clones in four (of 7) prospective epithelial cancer patients. ADGRG1 and CD86 based cell-surface enrichment captured tumor-organoid reactive TCRs as well as private neoantigen, tumor driver neoantigen-specific TCR clones. Thus, ADGRG1 and CD86 appear to be effective single surface protein markers for enrichment of antitumor CD4-positive T cells from human solid tumors. This study is currently being prepared for submission as a peer-reviewed research article to a journal in the upcoming months (Hitscherich and Dinerman et al, in preparation for submission). 2) We have completed a study where we have comprehensively evaluated the effect of systemic vaccination on the phenotypes of antitumor T cells during adoptive cell therapies (ACT). Most antitumor CD8 T cells in patients exhibit dysfunctional phenotypes, limiting the efficacy of adoptive cell transfer (ACT) using tumor-infiltrating lymphocytes (TIL). We show using both murine and human T cell models that concurrent neoantigen vaccination substantially enhances the antitumor activity of ACT products dominated by dysfunctional T cells and is contingent on host antigen-presenting cell participation. Vaccination remodels the anti-inflammatory state of the tumor microenvironment, promoting the expansion and trafficking of TSL to tumors and lymphoid organs. Mechanistically, vaccination does not rescue dysfunctional T cells directly; instead, it selectively amplifies low-frequency TSL (as little as 0.1% in the infusion product) to mediate tumor control. Analysis of dysfunctional TIL-ACT products in a metastatic melanoma patient corroborated these findings, with expansion and persistence of antitumor-TIL observed only after concurrent vaccination and coinciding with clinical tumor regression. These results suggest that concurrent vaccination can unlock the therapeutic potential of rare stem-like T cells within otherwise dysfunctional ACT products. This study is currently being prepared for submission as a peer-reviewed research article to a journal in the upcoming months (Krishna et al, in preparation for submission). 3) We are actively investigating physiologic site-specific differences in the frequency and phenotypes of antitumor T cells in patients with advanced solid tumors. We have accumulated data for a study profiling the frequency, specificity, and phenotype of antitumor T cells within regional, draining lymph nodes (rDLN) in patients with metastatic cancer. Paired resection of 8 lymph nodes and metastatic lesions from colorectal cancer patients enabled in vitro functional assays to validate and isolate tumor-specific CD8-positive TCRs derived from either rDLNs or TILs. We subsequently leveraged scRNA-seq and CITE-seq to characterize the transcriptional states of these validated, tumor-reactive CD8-positive T-cells, with the goal of determining whether rDLNs in metastatic solid epithelial cancers serve as an untapped reservoir of less dysfunctional, tumor-reactive CD8-positive clonotypes that can be leveraged for ACT. Tumor-reactive lymph-node-derived CD8-positive T cells were phenotypically distinct from those found intratumorally with less expression of TIL-exhaustion markers (TOX, ENTPD1, PDCD1) and higher expression of stemness markers (TCF7, KLF2, and SELL (CD62L)) at both transcriptional and protein levels. We have performed in vivo murine ACT experiments showing that rDLN-derived T cells are superior in mediating tumor regressions compared to TIL-derived T cell ACT. We are now in the final analysis stage of comparison of antitumor T cell frequency and states between blood-, lymphnode- and tumor-derived antitumor T cells from the same patient. This study is currently being prepared for submission as a peer-reviewed research article to a journal in the upcoming months (Dulemba and Dinerman et al, in preparation for submission). 4) We are actively accumulating data on genetic engineering of T cell stemness factors onto dysfunctional antitumor T cells to enhance cell therapies. We are conducting a research study aimed at evaluating the phenotypic effect of overexpressing KLF2 gene construct within neoantigen-specific dysfunctional TIL utilized for our TIL-ACT in our clinical trials obtained from 5 metastatic epithelial cancer patients. KLF2-overexpression did not significantly change the proportion of neoantigen-specific TIL, although we noted that KLF2-engineered TIL proliferated less well than control. Single cell transcriptomic sequencing of the KLF2-engineered TIL samples indicated distinct transcriptional states relative to mock-engineered TIL. In 5 in vitro expanded TIL from metastatic CRC patients, We further confirmed these data by flow cytometric interrogation of a second CRC patient neoantigen TIL for cell surface CD39/CD69 status which showed between 5-10-fold induction of CD39-CD69- stem-like T cell state in both neoantigen-specific and irrelevant TIL due to KLF2-overexpression. KLF2-enforced stemness state was also observed in every neoantigen-specific TIL that we evaluated with concomitant reduction of terminally differentiated exhausted CD39-positive CD69-positive TIL state. These encouraging preliminary results suggest that genetic engineering of key T cell stemness factors can reprogram exhausted human TIL to phenotypically less-dysfunctional stemlike states. Further experiments are ongoing.