Tumor Microenvironment Differences between Diagnostic and Relapsed Classic Hodgkin Lymphoma Revealed By Single Cell Transcriptome Sequencing
Bibliographic record
Abstract
Introduction: The composition and function of the tumor microenvironment (TME) in classic Hodgkin lymphoma (cHL), in particular in relapsed disease, are not fully described. This knowledge gap makes it challenging to develop effective and ideally tailored immunotherapies for cHL patients. Our recent imaging mass cytometry (IMC) study (Aoki et al, JCO 2024) revealed unique features in the spatially resolved TME according to relapse status (early-relapse vs late-relapse), including differences in B-cell abundance. However, the full extent of heterogeneity in the normal B-cell compartment is still unknown. Therefore, we aimed to characterize B-cell populations and related immune cell interactions using single cell RNA sequencing (scRNA-seq) with the goal to elucidate TME correlates of treatment failure in relapsed cHL. Methods: We performed 10X Chromium 5' scRNA-seq on 11 paired cell suspension samples of diagnostic and relapsed cHL, along with 4 reactive lymph node (RLN) control samples. To enhance representation of immune and lineage-related genes for accurate description of TME biology, we performed hybrid-capture sequencing (CapID+) on the same scRNA-seq libraries. Additionally, we validated our results at the protein level in intact tissue using immunohistochemistry (IHC) and IMC on a tissue microarray (TMA) representing 71 independent diagnostic/relapse sample pairs. For correlative analyses, patients were classified as early-relapse if disease progressed within 12 months after initial diagnosis or if their disease was refractory to first-line ABVD/ABVD-like treatment. Results: Across all diagnostic/relapsed cHL and RLN samples, we obtained a total of 49,843 transcriptomes after quality filtering, of which 21,718 (HL:20038; RLN:1680) were found to be expressing canonical B cell markers. Unsupervised clustering on B cells identified 10 subclusters including 1 naïve-B cell cluster, 1 memory B-cell cluster, 2 activated B-cell clusters, 2 atypical B-cell clusters, 1 cycling B-cell cluster, 1 germinal center B-cell (GCB) cluster, and 1 plasma cell cluster. The naïve B-cell cluster, C1, showed statistically significant enrichment (P < 0.05) in early-relapse samples compared to either diagnostic or late-relapse samples. In contrast, the memory B-cell cluster, C2, showed an opposite enrichment where early-relapse samples showed a statistically significantly decreased proportion compared to either diagnostic or late-relapse samples (P < 0.05). C1 was characterized by high expression of migration markers (CCR6+, CXCR5+) and regulatory markers (LGASL9+) along with naïve B-cell markers (IGHD+, IGHM+, IL4R+, SELL+). C2 demonstrated high expression of regulatory markers (ITGAM+, IL2RA+) as well as memory B-cell markers (IGHA2+, IGHG1+ CD27+ TNFRSF13B+). We validated the enrichment of unswitched naïve B cells in early-relapse samples using IgD IHC compared to diagnostic (P<0.05) and late-relapse samples (P < 0.05). Using in silico cell-to-cell interaction analysis (CellChat), the naïve B-cell cluster, C1, was predicted to interact with CD4+ LAG3+ Tregs uniquely in early-relapse samples through the Galectin-9 - TIM-3 axis (P < 0.05). To validate the cell-cell interaction results from scRNA-seq at the protein level and gain a more comprehensive understanding of spatial relationships of non-malignant B cells, we reanalyzed IMC data published by our group (Aoki et al, JCO 2024). CXCR5 and Galectin-9 positivity was used to define naïve B cells consistent with C1 co-expression patterns by scRNA-seq data. We found that CXCR5+ Galectin-9+ naïve B cells were in close proximity to HRS cells in early-relapse samples compared to diagnostic and late-relapse samples (P<0.05). Additionally, CXCR5+ Galectin-9+ naïve B cells were in closer spatial proximity of all Tregs and TIM-3+ CD4+ T cells in early-relapse samples compared to diagnostic and late-relapse samples. Conclusion: Our single cell studies of relapsed cHL revealed unique naive and memory B cell populations specific to relapse status. Our cell-cell interaction and spatial analyses highlight future potential for immunotherapeutic targeting of B cell and Treg subsets contributing to immunosuppression in early-relapse cHL.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".