A blueprint for tumor-infiltrating B cells across human cancers.
Jiaqiang MaYingcheng WuLifeng MaXupeng YangTiancheng ZhangQiang GaoTeng LiKe GaoXia ShenJian LinYamin ChenXiaoshan LiuYuting FuXixi GuZechuan ChenShan JiangDong-Ning RaoJiaomeng PanShu ZhangJian ZhouChen HuangXianjun YuJia FanGuoji GuoXiaoming ZhangQiang GaoPublished in: Science (New York, N.Y.) (2024)
B lymphocytes are essential mediators of humoral immunity and play multiple roles in human cancer. To decode the functions of tumor-infiltrating B cells, we generated a B cell blueprint encompassing single-cell transcriptome, B cell-receptor repertoire, and chromatin accessibility data across 20 different cancer types (477 samples, 269 patients). B cells harbored extraordinary heterogeneity and comprised 15 subsets, which could be grouped into two independent developmental paths (extrafollicular versus germinal center). Tumor types grouped into the extrafollicular pathway were linked with worse clinical outcomes and resistance to immunotherapy. The dysfunctional extrafollicular program was associated with glutamine-derived metabolites through epigenetic-metabolic cross-talk, which promoted a T cell-driven immunosuppressive program. These data suggest an intratumor B cell balance between extrafollicular and germinal-center responses and suggest that humoral immunity could possibly be harnessed for B cell-targeting immunotherapy.
Keyphrases
- single cell
- endothelial cells
- papillary thyroid
- immune response
- rna seq
- gene expression
- end stage renal disease
- electronic health record
- squamous cell
- quality improvement
- induced pluripotent stem cells
- peripheral blood
- genome wide
- dna methylation
- chronic kidney disease
- newly diagnosed
- pluripotent stem cells
- big data
- childhood cancer
- dna damage
- prognostic factors
- ms ms
- peritoneal dialysis
- high throughput
- lymph node metastasis
- machine learning
- cancer therapy
- drug delivery
- patient reported outcomes
- patient reported
- artificial intelligence