bioRxivCancer BiologyPreprint
Integrative analysis of TCGA transcriptomic states and DepMap dependencies prioritizes candidate vulnerabilities in immune-cold microsatellite-stable colorectal cancer
Abstract
Microsatellite-stable/microsatellite instability-low colorectal cancer (MSS/MSI-L CRC) is generally resistant to immune checkpoint blockade, but the biological states underlying this resistance are heterogeneous. We integrated TCGA COAD/READ patient transcriptomic profiles, MSIsensor-based MSS/MSI-L classification, curated immune and stromal module scoring, focused differential expression and DepMap CRISPR dependency data to prioritize candidate vulnerabilities in immune-cold MSS CRC. Among 494 MSS/MSI-L tumours, 218 were classified as MSS intermediate, 102 as MSS immune-cold, 91 as MSS hot/inflamed and 83 as MSS barrier-high. MSS immune-cold tumours showed lower cytotoxic, IFN{gamma}-chemokine and antigen-presentation programmes than MSS hot/inflamed tumours, including reduced NKG7, CD8A, CXCL9, CXCL10 and LAG3 expression. MSS barrier-high tumours showed enrichment of stromal and extracellular-matrix programmes, including COL1A1, COL1A2 and COL3A1. Integration with DepMap CRISPR gene-effect data from 1208 cancer models, including 63 colorectal cancer models, separated tumour-cell-intrinsic dependencies from patient-derived microenvironmental signatures. Candidate target classes included ERBB2, VEGFA, PIK3CB, ATR/WEE1/CHEK1, HDAC1/HDAC3/BRD4 and BCL2L1/MCL1, while collagen genes were interpreted as stromal-barrier markers rather than tumour-cell dependencies. ERBB2 expression was higher in MSS immune-cold than MSS hot/inflamed tumours and further elevated in MSS barrier-high tumours, supporting ERBB2 as a candidate subset-associated signal that requires orthogonal HER2 validation. These findings support a stratified therapeutic framework for immune-cold, barrier-high and intermediate MSS CRC.
- organism: human
- assay: bulk RNA-seq
- research focus: differential expression
- research focus: gene regulatory network inference
- approach: statistical model
- approach: machine learning (classical)
- Citations
- 0
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Molecular and functional profiling unravels targetable vulnerabilities in colorectal cancer
Efstathios‐Iason Vlachavas, Konstantinos Voutetakis, … Stefan Wiemann·bioRxiv·Apr 17, 2024
- Aberrant NOTUM+ Program Induced in LGR5+ Crypt Base Columnar Cells Maintains an Immunosuppressive Niche in Colorectal Cancer
Julian Chua, Parnika Sakthivel, … Arshad Ayyaz·bioRxiv·Mar 10, 2025
- ACB1801 enhances tumor immunogenicity by targeting glycolysis/ferroptosis vulnerability and activating STAT1-signaling to overcome anti-PD-1 resistance in MSS colorectal cancer
Ruize Gao, Kris Van Moer, … Bassam Janji·bioRxiv·Feb 25, 2026
- Single cell spatial transcriptomics identifies coordinated cellular programs associated with good prognosis in microsatellite stable colorectal cancer
Victoria T. Karlsen, Umair Majid, … Frode Lars Jahnsen·bioRxiv·Mar 31, 2025
- Multi-omics Analysis Identifying CD58 Plays Important Role on Prognosis of Colorectal Cancer
Xuan Wang·bioRxiv·Jun 19, 2022
- Reovirus sensitizes microsatellite stable colorectal cancer to anti-PD-1 treatment via cross-talk in innate and adaptive immune systems
T. Augustine, P. John, … S. Goel·bioRxiv·Sep 3, 2021