Sample Site Impacts RNA Biomarkers for Renal Cell Carcinoma

Immunotherapy (ICIs) remains a mainstay for treatment of advanced clear-cell renal cell carcinoma (ccRCC). Biomarker analyses have demonstrated that gene expression profiles are associated with regimen-specific outcomes. These transcriptomic analyses used mixed sample cohorts (primary and metastatic tumor specimens) and it is unknown whether the clinical relevance of transcriptomic signatures is impacted by tissue site. We evaluated data for 1132 patients with metastatic ccRCC treated with ICI in prior studies (IMmotion151 and CheckMate-009, -010, and -025). We identified significant and reproducible differences in gene expression by tissue site. We tested the association between previously described molecular tissue clusters (MTCs) by tissue site (MTC1-primary and MTC1-metastasis) and progression-free survival (PFS) and objective response to systemic therapy. In IMmotion151, MTC2-metastasis was significantly associated with better PFS on sunitinib (hazard ratio [HR] 3.39, 95% confidence interval [CI] 1.32-8.69; p = 0.01) in comparison to MTC2-primary (HR 0.95, 95% CI 0.65-1.38; p = 0.80; p_interaction = 0.02). Evaluation of known RNA signatures in the CheckMate trials revealed that JAVELIN-metastasis was associated with better PFS on ICI (HR 0.77, 95% CI 0.62-0.97; p = 0.03) in comparison to JAVELIN-primary (HR 1.04, 95% CI 0.91-1.19; p = 0.56; p_interaction = 0.02). These results indicate that tissue site may be a relevant confounder in biomarker analyses.