Main Session
Sep
28
PQA 03 - Digital Health Innovation and Informatics, Patient Safety & Quality, and Radiation and Cancer Biology
Presenter(s)
Sai Kumar Samala, MD, MBBS - University of California Los Angeles, Los Angeles, CA
S. K. Samala1, P. C. Mallepaddi1, N. T. Tra1, P. M. Q. Mai1, K. Koushki1, E. Han2, L. W. Cheung1, G. Vijay1, and S. Krishnan1; 1The University of Texas Health Science Center at Houston, Houston, TX, 2Rice University, Houston, TX
Purpose/Objective(s):
Discoidin domain receptor 1 (DDR1) is a collagen-activated receptor tyrosine kinase that regulates multiple pro-tumorigenic signaling pathways. It also contributes to collagen architecture, which may create an immunosuppressive tumor microenvironment. We hypothesize that DDR1 expression is associated with immune exclusion phenotypes in cancers and that DDR1 represents a targetable mechanism for converting immunologically cold tumors into immunologically hot ones.Materials/Methods:
We performed integrative pan-cancer analysis with publicly available data. TIMER 3.0 was used to quantify correlations between DDR1 gene expression and immune infiltration (e.g., CD8? T cells, Tregs, MDSCs, and CAFs), as well as with ESTIMATE/TIDE-derived immune estimation and exclusion scores, and T-cell dysfunction scores. The UCSC Xena platform was used to assess DDR1 mRNA levels across TCGA pan-cancer samples stratified based on Thorsson’s immune subtypes. CPTAC proteomics was employed to validate protein expression across various tumor and normal tissues. The mRNA levels of DDR1, extracellular matrix components (COL1A1, COL1A2, COL3A1), integrins (ITGAV, ITGB1), and focal adhesion kinase (FAK) were analyzed for correlation using CPTAC and TIMER 3.0.Results:
Although the strength and direction of associations varied between deconvolution algorithms and cancer types, multi-algorithm immune deconvolution in TIMER 3.0 showed that higher DDR1 expression was generally associated with reduced CD8? T-cell infiltration across several tumor types, with particularly consistent inverse correlations in pancreatic, lung, and breast cancers. DDR1 levels were positively associated with MDSC and T-cell exclusion scores (TIDE) and inversely associated with ESTIMATE immune scores across various tumors, consistent with an immune-excluded phenotype. Immune subtype analysis demonstrated that DDR1 expression was highest in Immunologically Quiet tumors compared with all other immune subtypes. CPTAC data show generally higher DDR1 protein levels in tumors versus normal tissues across multiple cancer types. Purity-adjusted DDR1 transcript levels were positively correlated with PTK2 and ITGB1/ITGAV and showed a modest correlation with fibrillar collagens across several tumor types. This is consistent with coordinated transcriptional regulation of a collagen–integrin–FAK extracellular matrix–adhesion program.Conclusion:
The analyses suggest DDR1 as a marker of the immune-excluded microenvironment. DDR1-collagen-integrin interactions represent a potential mechanistic link between matrix organization and T-cell exclusion. Further investigations are underway to explore DDR1 as a therapeutic target for enhancing immune infiltration and immunotherapy responses in tumors with increased DDR1-collagen axis signatures.