Abstract NG14: Dissecting immune microenvironment of T-cell acute lymphoblastic leukemia

Abstract T-cell acute lymphoblastic leukemia (T-ALL) represents a particularly aggressive subtype of leukemia with no targeted therapies or immune interventions. Currently, in response to intensive chemotherapy, ~25% of pediatric and 50% of adult patients undergo relapse and succumb to this therapy-resistant disease. Patients that relapse have poor outcomes, with <10% surviving long-term. Large-scale sequencing efforts focused at elucidating the genetic makeup of acute lymphoblastic leukemia, have not been able to identify targeted treatment strategies or predict relapse. Therefore, there is an urgent clinical demand for identifying T-ALL vulnerabilities and therapeutic approaches. Targeting the immunosuppressive tumor microenvironment has revolutionized the treatment of solid tumors. Despite its success, immunotherapy has not improved T-ALL patient outcomes. This is partly due to a lack of understanding of which immune populations interact with leukemia. Additionally, while cancer heterogeneity correlates with drug resistance, poor prognosis, and patient mortality, the impact of leukemic heterogeneity on a patient’s immune recognition of leukemic antigens is unknown. To examine leukemic heterogeneity and leukemia-associated immune landscape, we coupled Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE-seq), a cutting-edge technique that combines highly multiplexed protein marker detection with unbiased transcriptome profiling of thousands of single cells, with 5'TCR-seq, to track clonal T-ALL expansion, as well as healthy TCR repertoire. We have assembled a broad CITE-seq antibody panel that encompasses 38 markers associated with different lineage subsets and cellular states, such as T-cell activation/exhaustion markers and lymphocyte recruitment. 36,714 cells from T-ALL (n=9), early T-cell precursor (ETP) (n=1), and mixed phenotype (MP) (n=2) adult patients at disease diagnosis, and healthy BM donors. To generate a comprehensive reference atlas of hematopoietic development across the human BM and thymus, we took advantage of previously published scRNA-seq datasets. As expected, the transcriptional profiles characteristic of leukemic cells were highly represented within early hematopoietic and thymic progenitors. T-ALL is a clonal malignancy: within one patient, all leukemic cells share identical TCR rearrangements. We took advantage of this unique feature to identify leukemic clones. Leukemic cells from 5 out of 9 TALL samples harbored clonal TCRβ rearrangements, suggesting that transformation occurred in a T cell progenitor that had already undergone TCRβ rearrangement in these samples. Interestingly, in 4 T-ALL samples, we observed subclonal TCRα rearrangements at varying frequencies, suggesting ongoing TCRα locus rearrangement after transformation. In agreement with our TCR analysis, subpopulations of leukemic cells from each sample mapped along the continuum of bone marrow (HSPCs) to thymus T cell development subsets (early thymic progenitors (ETP), double negative (DN), double positive (DP), single positive CD4 or CD8 T-cells) on our reference map. Next, we quantified the leukemic cell types per sample and found that samples could be differentiated into two major subgroups based on the frequency of immature (HSPC to ETP) versus mature (DP to mature T cell stage) mapping cells. In addition to cell type prediction, gene expression profiles of the majority of cells from samples designated as immature by HSPC/ETP cell frequency, but not those designated as mature, scored above the predicted threshold for the expression of an ETP cell signature. SCENIC analysis indicated that cells from T-ALL samples that scored as immature, were associated with early hematopoietic (Pu.1/SPI1), myeloid (CEBPB, CEBPD) or erythroid transcription factor activity (GFI1B), while mature samples showed FOS, JUN, and CUX1 activity. Interestingly, cells from both mature and immature samples displayed activity of transcription factors involved in MYC regulation, including MAZ and LEF1. Finally, all T-ALL samples harbored both immature and mature cell types, albeit at differing frequencies, suggesting that, similar to AML, subclonal populations of T-ALL cells may be organized into a developmental hierarchy. Our analysis of BM immune composition in leukemic patients revealed a significant loss of differentiated immune subsets, including mature myeloid cells, dendritic cells (DCs), and plasmacytoid dendritic cells (pDCs). On the other hand, we detected a significant increase in the abundance of CD4 memory, CD8, CD8 memory T and NK cells, indicating a substantial remodeling of T and NK populations in response to T-ALL presence. In addition to changes in abundance, gene set scoring using the AUCell algorithm demonstrated that single cells from the leukemic BM and healthy BM are differentially enriched for the expression of genes involved in Interferon Alpha (IFNA), previously linked to T cell response, and Tumor Necrosis Factor Alpha (TNFA) signaling, respectively. Therefore, both the frequency and gene expression signatures of mature immune subsets are altered in the T-ALL-associated BM microenvironment. Anti-cancer immune responses are intimately linked to T cell functional status. Thus, we leveraged our cell type prediction algorithm and TCR-seq data to subset nonleukemic T cells from our single cell dataset for further analysis. Using subtype and functional markers, along with cluster partitions (Appendix 2B), we annotated eight distinct T cell clusters, including two CD4 T cell clusters, one invariant T cell cluster, marked by expression of KLRB1, and comprising both CD4 and CD8 expressing cells, and five additional clusters dominated by CD8-expressing T cells. These additional five clusters are functionally distinct by the absence of granzyme expression (naive CD8), GZMK+GZMBA+GZMB- (cytotoxic/pre-exhausted), GZMK+GZMA-LAG3+HAVCR2+ non-cycling (exhausted) versus cycling (cycling exhausted), and GZMK-GZMA+GZMB+ cytotoxic T cells. Our T cell scoring analysis revealed an expansion of cytotoxic T cells in leukemia patients compared to healthy donors. Another advantage of our dual CITE-seq+TCR-seq approach is the ability to identify clonally expanded nonmalignant T cells. Strikingly, the TCR repertoire of endogenous T cells in the leukemic patients was skewed toward oligoclonal TCR use when compared with normal donor T cells. Moreover, oligoclonal TCR use reflected the presence of an expanded population of cytotoxic CD8+ T cells. Collectively, our preliminary data suggests an induction of immune response directed against leukemic clones in T-ALL patients. Our preliminary data underscored an unexpected level of intratumoral and intertumoral heterogeneity of malignant cells in T-ALL, whereas samples shared a common enrichment of non-malignant T cells in the BM. To further explore the immune landscape and malignant cell heterogeneity in T-ALL, we are currently analysing an additional 9 T-ALL and 3 healthy BM samples. Our proposed studies represent the first comprehensive mapping of leukemic hierarchy and immune system in primary human T-cell acute leukemia. We believe that this work will uncover novel immune subpopulations and cellular interactions, which could be targeted to enhance treatment response and improve patient outcomes. Citation Format: Mark Gower, Minerva Fernandez, Andrea Aruda, Mark Minden, Johann Hitzler, Anastasia Tikhonova. Dissecting immune microenvironment of T-cell acute lymphoblastic leukemia. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr NG14.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.