Circulating angiogenic–myeloid cytokine axis to predict benefit from first-line immune-checkpoint inhibitor combinations in metastatic clear cell renal cell carcinoma.

550 Background: Metastatic clear cell renal cell carcinoma (ccRCC) first-line treatment relies on immune checkpoint inhibitor (ICI)–based regimens, including either dual ICI or ICI plus tyrosine kinase inhibitor (TKI) combinations. Despite clinical advances, selecting the optimal regimen remains challenging. Circulating biomarkers may provide a broader and more accessible overview of the tumor immune status compared with tissue-based biomarkers. Methods: Patients with metastatic ccRCC treated with first-line ICI combinations were included (NCT04932525). A 96-soluble factors panel reflecting immune activation, checkpoint regulation, and angiogenesis was quantified at baseline using the Olink Immuno-Oncology platform. In parallel, single-cell RNA sequencing was performed on peripheral blood mononuclear cells (PBMCs) to identify major circulating immune subsets. Main endpoints included objective response rate (ORR), progression-free survival (PFS) and overall survival (OS). Results: The study prospectively included 27 patients: 15 received ICI–TKI and 12 dual ICI. Unsupervised correlation analysis of baseline cytokines identified an angiogenesis/hypoxia cluster composed of PGF, HGF and VEGFA, which were strongly intercorrelated (mean ρ = 0.65, p < 0.001). An angiogenic score, defined as the mean level of these three cytokines, was calculated. Higher scores were associated with shorter PFS and OS in the overall cohort (log-rank p = 0.048 and p = 0.009, respectively). Stratified by treatment, this score retained its negative impact on OS in the dual ICI arm (log-rank p = 0.016) but not in the ICI–TKI group, where conversely higher scores were observed in responders (Wilcoxon p = 0.04), suggesting a differential predictive value. The angiogenic cluster correlated positively with myeloid-related cytokines such as IL6 and CCL23, as well as with a circulating transcriptomic myeloid signature (ρ = 0.60, p = 0.009). Consistently, single-cell RNA analysis revealed enrichment of myeloid cells among non-responders in the dual ICI arm (χ² test, OR = 0.71, p < 0.001), but not in the ICI–TKI group, confirming the link between angiogenic-myeloid activation and resistance to dual immunotherapy. Conclusions: Circulating immune factors provide an accessible window into the systemic immune state of metastatic ccRCC. Angiogenesis- and myeloid-related cytokines formed a functional cluster associated with an immunosuppressive, myeloid-skewed profile predicting lack of benefit from dual ICI but greater response to ICI–TKI combinations. These exploratory findings suggest that circulating angiogenic and myeloid signatures may serve as non-invasive biomarkers to guide first-line treatment selection in metastatic ccRCC. Clinical trial information: NCT04932525 .