Single-Cell Atlas of Diffuse Large B-Cell Lymphoma Reveals Integrated Malignant B Cell and Tumor Microenvironment Characteristics Associated with Clinical Response

Background:

Diffuse large B-cell lymphoma (DLBCL) is the most common aggressive B-cell lymphoma with high clinical and biological heterogeneity. Despite current effective immunochemotherapy, up to 40% of patients do not respond or develop refractory disease. DLBCL is characterized by two major cell-of-origin (COO) subtypes, germinal center B cell-like (GCB) and activated B cell-like (ABC), with recent research uncovering additional molecular subgroups based on genomic alterations or tumor microenvironment (TME) features. We recently reported (Baaklini et al. ASH 2023) a multimodal single-cell atlas of 62 DLBCL cases, integrating single-cell transcriptomes, immune repertoire and genetic features, that we have leveraged to decrypt recurrent B cell states and TME ecosystems while identifying clinically relevant prognostic biomarkers. Here we report further analyses of the single-cell atlas dataset focused on TME composition and differentiation trajectories, and the validation of our main findings through cell-state deconvolution of bulk RNA-seq datasets and single-cell spatial transcriptomics analyses.

Methods:

The DLBCL single-cell atlas was generated with 5'-end single-cell RNA, B-cell receptor (BCR) and T-cell receptor sequencing on 63 DLBCL lymph node biopsies obtained from the real-world, clinically-annotated, multicentric CeVi collection and included 43 patient samples at diagnosis (ABC = 17, GCB = 11, Unclassified = 6, Not Evaluated = 9; 21 of which did not achieve event-free survival at 24 months (EFS24) after R-chemotherapy regimens), 11 relapsed/refractory DLBCL and 8 DLBCL transformed from indolent lymphoma. Methods for overall cell annotation and malignant B cell states identification were reported previously (Baaklini et al. ASH 2023). For the analysis of the TME and the focus on T cells, we identified clonally expanded T-cell subsets based on their TCR sequences, and used trajectory inference algorithms to track CD8 T cell branched differentiation. Validation on public, clinically annotated data (Schmitz et al. NEJM 2018) was performed by gene signature scoring and reference-based deconvolution. Single-cell spatial transcriptomics was performed on a subseries of 20 DLBCL cases laid out as tissue microarrays, using two distinct technologies with off-the-shelf 1000-plex and 377-plex panels.

Results:

We previously reported the identification of 5 conserved malignant B cell transcriptional archetypes (Arch.1-5) that co-exist in DLBCL samples: Arch.1 characterized by plasma cell identity, Arch.2 by MYC and NFKB pathways, Arch.3 by MYC pathway only, Arch.4 by quiescence markers, and Arch.5 by GC markers. Integrated analysis of the TME revealed three major ecosystems based on the relative proportions of non-malignant T, NK and myeloid cell subsets, including one enriched for effector and cytotoxic T cells. TCR clonal analysis revealed that the largest clonotypes were enriched in a subset of terminally differentiated hyperactivated CD8 effector T cells. Differentiation trajectory analysis of CD8 T cells revealed that patients achieving event-free survival at 24 months (EFS24) accumulated terminally differentiated hyperactivated CD8 effector T cells, while CD8 T cell differentiation was blocked at an intermediate effector stage in patients not achieving EFS24. We validated those findings by cell-type specific gene signature scoring and reference-based deconvolution of a public bulk RNA-seq dataset. Integrating proportions of malignant B cell archetypes and TME subsets, we defined 4 groups of samples with distinct event-free survival differing notably in their proportions of Arch.4 cells and terminally differentiated hyperactivated CD8 effector T cells, which we also validated by reference-based deconvolution of public bulk RNA-seq. Finally, single-cell spatial transcriptomics mapped distinct cellular neighborhoods of malignant B cells, non-malignant immune cells, and stromal cells.

Conclusion:

In conclusion, fine analysis of our comprehensive single-cell atlas of DLBCL revealed integrated malignant and TME characteristics associated with clinical response, that were also validated on other datasets and technologies. Those findings bring novel biomarkers for stratifying non-responsive DLBCL patients and guiding treatment decisions, as well as for designing novel therapies targeting the malignant B cell – TME crosstalk.