Gene expression of prostate-specific membrane antigen (FOLH1) in clear cell renal cell carcinoma predicts angiogenesis and response to tyrosine kinase inhibitors

Purpose: Combination systemic therapies (CSTs) of immuno-oncologic (IO) and VEGF-inhibiting agents (VEGFi) have become the standard of care for management of metastatic clear cell renal cell carcinoma (m-ccRCC). However, treatment outcomes vary between patients, with no established biomarkers to determine optimal CST regimens (IO/IO or IO/VEGFi). Prostate Specific Membrane Antigen (PSMA), encoded by the FOLH1 gene, is a marker of tumor neovasculature in ccRCC, the downstream target of VEGFi. We evaluated the relation between FOLH1 expression and angiogenesis, as well as clinical outcomes, in 5 m-ccRCC ST trials.

Materials and methods: using Spearman's rank correlation (SPRC) test, we assessed the correlation between FOLH1 expression and gene expression signature (GES) scores corresponding to angiogenic and immunologic features of the tumor microenvironment (TME) of m-ccRCC in our trial cohorts. Using Cox proportional hazard regression (Cox-PHR), we assessed the association between FOLH1 expression level, summarized by within-study quantiles (qFOLH1), and progression-free and overall survival (PFS, OS).

Results: Increased FOLH1 expression was significantly associated with higher TME angiogenesis GES scores (SPRC +0.5, P < 0.001), but did not consistently correlate with immune feature GES scores. Meta-analysis of PFS in the sunitinib TKI arm of trial cohorts showed an overall positive association with qFOLH1 (HR = 0.89; 95% CI = 0.85-0.94, P < 0.0001). qFOLH1 was not significantly associated with OS in the sunitinib arms of the two trials with OS data (COMPARZ, HR 0.87, 95% CI 0.71-1.07, P = 0.17; and Checkmate-214, HR 0.89, 95% CI 0.67-1.17, P = 0.70).

Conclusions: PSMA-encoding FOLH1 gene expression correlates with neoangiogenesis and predicts PFS in m-ccRCC patients treated with sunitinib TKI, suggesting that PSMA PET could be explored as a noninvasive biomarker for guiding CST choice (IO/IO or IO/VEGFi) as well as prediction of treatment response to VEGFi in m-ccRCC patients.