Single Cell T Cell Maps of Donor Lymphocyte Infusion (DLI) Response and Resistance

Abstract

DLI can eradicate leukemic relapse after allogeneic hematopoietic stem cell transplant (allo-SCT), potentially by reversing T cell exhaustion for chronic myelogenous leukemia (CML). Critical questions remain, however, regarding the exact cellular identities and transcriptional states of those T cell subtypes mediating exhaustion, anti-leukemia responses, and DLI resistance. To map evolving phenotypic T cell states in situ at single cell resolution, we obtained single cell transcriptomes from bone marrow mononuclear cells before/after DLI from 6 responders (R's) and 6 nonresponders (NR's) with relapsed CML after T-cell depleted allo-SCT. In total, 381,462 cells derived from 43 unique patient-timepoints met our quality metrics; we curated a set of 62 distinct cell states, from T, B, NK, monocyte, and progenitor subtypes, whose identities were determined by both cell type and timepoint. Because DLI's anti-leukemic efficacy derives largely from T cell activity, we evaluated if DLI response associated with distinct T cell transcriptional phenotypes (states). We observed a marked separation of T cell clusters based on clinical outcome and a significant increase in the number of T cell clusters after DLI in matched samples after controlling for cell number (pre vs post, p-value We queried global T cell dysfunction between R's vs NR's, observing increased T cell exhaustion signatures in R-pre T cells, consistent with our previously published findings, but also detecting increased tolerance and anergy in NR-post T cells, suggesting multiple forms of T cell dysfunction in DLI resistance. Anergic and tolerant cells constituted distinct clusters, expressing reduced CD8A/CD4 expression, suggesting decreased antigen responsiveness. We will also infer novel gene regulatory networks driving T cell states and clonal expansion by integrating scRNA-seq data with associated ATAC-seq profiles and scTCR repertoires, using novel computational techniques. These findings will be updated at the meeting. Altogether, these data suggest that (1) pretreatment T cell phenotypic diversity may be important for DLI response; (2) that DLI increases such diversity differentially in R's than in NR's; and (3) even in the absence of clinical response, NR's undergo significant T cell phenotypic remodeling.