P625: single-cell rna sequencing reveals the heterogeneity of tumor cells and immune microenvironment in richter syndrome and accelerated cll
Y. Sha,Y. Miao,S. Qin,R. Jiang,W. Wu,Y. Xia,L. Wang,L. Fan,W. Xu,H. Jin,H. Zhu,J. Li
DOI: https://doi.org/10.1097/01.hs9.0000845388.91550.a4
2022-01-01
HemaSphere
Abstract:Background: Richter Syndrome (RS) is defined as development of aggressive lymphoma arising from chronic lymphocytic leukemia/Small lymphocytic leukemia (CLL/SLL). Accelerated CLL/SLL (aCLL/SLL) is pathologically defined as a highly proliferative form of CLL/SLL (either expanded proliferation center or Ki-67>40%). The underlying mechanism behind transformation from indolent CLL/SLL to aggressive DLBCL remains largely unknown. The biological difference between RS DLBCL and aCLL/SLL has not been explored yet. Aims: To explore intra- and inter-tumor and immune microenvironment (TME) heterogeneity of RS and aCLL/SLL. To disclose the molecular mechanism underlying transformation from CLL/SLL to DLBCL. Methods: Single-cell RNA sequencing (scRNA-seq) of paired lymph node (LN) and peripheral blood (PB) was done among 2 RS (P1, P2) and 2 aCLL/SLL (P3, P4) patients (pathologically diagnosed in LN, Fig. A) and bioinformatics analysis was conducted. Results: Identification of cell type in 8 samples showed great heterogeneity (Fig. B). Single-cell trajectory analysis was done using “monocle” inferred tumor development and differentiation of each patient. Monoclonal B cells with high Ki-67 expression and high proportion of G2M phase cells were enriched in the end of pseudotime trajectories (Fig. C). Further analysis of CD5, CD20 and CD23 expression alongside pseudotime trajectories were also conducted, indicating that our trajectories could depict the transformation from indolent CLL/SLL to aggressive DLBCL. Therefore, cell states were clustered according to the trajectories. P1 and P2 were clustered as seven groups respectively (P1A-G, P2A-G) and P3 were clustered as three groups (P3A-G). Notably, although minor proportion of proliferating cells were enriched in the end of P4 pseudotime trajectories, clusters could not be split in P4 patients. All cell states groups were mapped to UMAP analysis and the proportion of G2M phase was highest in state G of RS patients. Single-sample GSEA (ssGSEA) analysis was done to disclose the pathway activation in transformation process and found that pathways including cell cycle, TCA cycle, spliceosome, MYC targets, unfolded protein response, epigenetics, DNA repair, glycolysis and oxidative phosphorylation were gradually activated alongside trajectories from state A-G in RS. Moreover, pathways including NOTCH, P53, PI3K-AKT, Antigen processing and presentation, apoptosis, hypoxia and etc. were found significantly upregulated in RS (Fig. E). Further analysis of immune microenvironment of RS and aCLL/SLL showed that exhaustion T cells were significantly enriched in RS lymph node compared with aCLL/SLL, indicating the immune surveillance escape mediated by exhaustion T cells might prompt DLBCL transformation. PDCD1 (PD-1), CTLA4, LAG3, HAVCR2 (TIM3), TIGIT expression were significantly upregulated in exhaustion T cells of RS patients, suggesting potential treatment target of RS for immune checkpoint inhibitors (Fig. F). Image:Summary/Conclusion: Our study was the first one using scRNA-seq to disclose the tumor heterogeneity and immune microenvironment of RS and aCLL/SLL in single-cell dimension. Evolution trajectories from indolent CLL/SLL to aggressive DLBCL were constructed and pathway activation in the process of transformation were further explored. Moreover, immune microenvironment analysis of RS and aCLL/SLL showed that exhaustion T cells were significantly enriched in RS lymph node and immune checkpoints including PDCD1 (PD-1), CTLA4, LAG3, HAVCR2 (TIM3), TIGIT were significantly upregulated in exhaustion T cells of RS patients, suggesting potential treatment target of RS.