Chemokine Receptor Co-expression Analysis in External Clinical Melanoma Cohort scRNA-seq
Overview
This repository contains the data analysis code to validate findings in a study of chemokine receptor (CR) co-expression by CD8+ T cells in the setting of dual checkpoint blockade, in collaboration with Reshef Lab. Combined checkpoint inhibitors are transformative cancer therapies, but their efficacy is limited, and severe immune-related adverse events (IRAEs) are common. A central challenge is identifying T cells that effectively infiltrate and destroy tumors without causing off-target damage. Here, we show that the co-expression landscape of chemokine receptors CXCR3, CCR5, and CXCR6 on CD8+ T cells defines functionally distinct subsets with divergent roles in tumor immunity and off-target inflammation in both mice and humans.
Study Design
In a mouse model of melanoma treated with dual checkpoint blockade, a triple-positive (CXCR3+CCR5+CXCR6+) T-cell subset is essential for tumor control and its genetic signature correlates with response in patients. Conversely, a subset co-expressing CCR5 and CXCR6 (R6R5) is associated with migration to the liver, a site of IRAEs. This chemokine receptor "code" helps separate therapeutic efficacy from toxicity, providing a new framework for developing safer, more effective immunotherapies. We analyzed external scRNA-seq of CD8+ T cells from a clinical cohort of metastatic melanoma patients treated with immune checkpoint blockade (Pozniak Cell 2024; https://www.sciencedirect.com/science/article/pii/S0092867423013223) for co-expression of CR markers. After batch correction, we examine continuous phenotypic differentiation with respect to CR co-expression patterns using diffusion mapping and pseudotime calculation, and we compare CR expression and co-expression signature between cclinical groups using pseudobulked expression across samples.
Figures
- Fig 1f: Diffusion maps of CD8+ T cell differentiation (TCF7, TOX) and CR marker expression (CCR7, CXCR3, CCR5, CXCR6)
- Fig 1g: Pseudotemporal expression patterns of CR markers
- Extended Data 2a: Pseudotime rank sorted expression of CR markers
- Extended Data 2b: Sliding window expression of CR markers across pseudotime
- Extended Data 2c: Distribution of sliding window normalized CR marker expression
- Fig 4d: CR co-expression signature comparisons between ICB responders and non-responders (MWU test)
- Extended Data 6d: CR marker expression comparisons between ICB responders and non-responders (MWU test)
Data Access
Data analyzed for this work is publically available as described in the original publication (Pozniak Cell 2024; https://www.sciencedirect.com/science/article/pii/S0092867423013223) under access code EGAD00001009291 in the European Genome-phenome Archive.
Acknowlegement
This work was made possible by the collaboration between the Azizi Lab and Reshef Lab, as well as our external collaborators.
Citation