BackgroundImmune checkpoint inhibitors (ICIs) is the main treatment option for patients with metastatic renal cell carcinoma (mRCC); however, significant heterogeneity in response is commonly observed. This study aimed to evaluate the ability of C-reactive protein (CRP) to predict overall survival (OS) and progression-free survival (PFS) in patients with mRCC treated with immunotherapy.Patients and MethodsData from patients with mRCC treated with atezolizumab plus bevacizumab in the IMmotion150 and IMmotion151 trials were pooled. Cox proportional regression was used to model prognostic associations. The relative importance of CRP against International Metastatic RCC Database Consortium (IMDC) factors was confirmed using machine learning.ResultsCRPs were available from 527 patients (mean[range] CRP, 6.3[0.21–340]mg/L). Elevated CRP was significantly associated with worse OS (HR[95%CI], 1.71[1.54–1.90], p<0.001) and PFS (1.27[1.18–1.35], p<0.001). CRP was the most prognostic factor for survival within the available clinicopathological data. The prognostic performance of CRP was superior to IMDC model for OS (CRP c=0.76, IMDC c=0.67, p<0.001) and PFS (CRP OS c=0.62, IMDC c=0.59, p=0.03). Predicted 2-year OS probabilities for patients with CRP values of 0.5, 5, 40, and 150 mg/L were 96%, 73%, 42%, and 23%, respectively.ConclusionsCRP is a powerful prognostic marker for survival, and its prognostic value was superior to the IMDC risk model. This study highlights that CRP could be implemented as stratification factor for mRCC immunotherapy trials and potentially as an easy-to-use prognostic tool in the clinic.