C-reactive protein kinetics to predict recurrence of high-risk renal cell carcinoma after radical surgery

Background With new options in adjuvant settings, clinical biomarkers to predict recurrence after radical surgery for high-risk renal cell carcinoma (hrRCC) are in need but are scarcely investigated. We aimed to verify the predictive value of perioperative C-reactive protein (CRP) kinetics on hrRCC recurrence. Methods We retrospectively evaluated 154 patients who underwent radical surgery for hrRCC (>= pT3 and/or N1-2 and M0) at two institutions. Patients were classified into Normal (< 0.5) and High (>= 0.5) according to their preoperative serum CRP (mg/dL). The High group were further classified into Normalized (< 0.5 at post) or Non-normalized (>= 0.5 at post), and recurrence-free survival (RFS) was compared between groups. Factors for RFS were further analysed, and Harrell's concordance index (C-index) for the accuracy of predicting RFS was compared with and without the addition of CRP-related variables to pre-existing models. Results The RFS was significantly shorter in the High (n = 72, 46.8%) compared to the Normal (n = 82, 53.2%) group (9.7 vs. 66.7 months, p < 0.001). Within the High group, Non-normalized (n = 27, 17.5%) patients showed a significantly shorter RFS compared to the Normalized (n = 45, 29.2%) group (6.2 vs. 20.3, p = 0.009). In the multivariable stepwise analysis, CRP kinetics (hazard ratio 2.15, p = 0.029) effectively predicted RFS while baseline CRP fell short of significance. Higher C-index improvement was observed with CRP non-normalization than the baseline value when added to factors in the Karakiewicz and University of California Los Angeles Integrated Staging System models. Conclusions CRP kinetics effectively predicted RCC recurrence after surgery and may aid in decision-making for adjuvant systemic therapy.