Abstract
INTRODUCTION: Gaining insights into the molecular features associated with T cell exhaustion (TEX) can offer fresh perspectives on predicting treatment responses, and we aim to investigate TEX-related tumor associated macrophages (TAM) subset to deeply understand underlying mechanisms of immune exhaustion. METHODS: We performed pan-cancer single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics RNA sequencing (stRNA-seq) analyses to investigate the subtype of TEX-associated TAMs, exploring its spatial distribution characteristics in context of immunotherapy. The pan-cancer scRNA-seq and RNA-seq datasets were incorporated to develop the STMN2+ Macrophage Signature (STMN2.SIG), which predicts immunotherapy response based on integrative machine learning techniques. Comprehensive scRNA-seq analysis, with in vitro experiments, investigated the mechanisms by which STMN2+ TAMs influence tumor progression and immune exhaustion. RESULTS: A macrophage subset, STMN2+ TAMs, and an epithelial subtype, S phase Sympathoblasts were identified as TEX-related cellular subpopulations. A higher proportion of STMN2+ TAMs was observed in non-responders compared to responders in pan-cancer immunotherapy landscape. Pan-cancer STMN2.SIG performed well in predicting immunotherapy response in pan-cancer cohorts, potentially linked to intercellular interactions between STMN2+ TAMs and CD8+ Tex cells. stRNA-seq analysis confirmed that interactions and cellular distances between STMN2+ TAMs and CD8+ Tex cells impact therapy efficacy. In a co-culture system, silencing BCAP31 on TAMs drives CD8+ T cells toward an effector state in NB. And BCAP31 on TAMs is associated with modulation of JAK2-STAT3 pathway in tumor cells. CONCLUSION: Our study provides pan-cancer STMN2.SIG as an outperforming approach for patient selection of immunotherapy, and advances our understanding of TAM biology and suggests potential therapeutic strategies for downregulation of BCAP31 in TAMs.