C-reactive protein kinetics superior to radiographic response as a surrogate endpoint for survival in patients with advanced renal cell carcinoma.

333 Background: As cytokines and targeted agents against advanced renal cell carcinoma (aRCC) are considered to achieve high stable disease (SD) rate rather than objective response (OR) by radiographic measurement, we often face the therapeutic dilemma in deciding whether to continue the ongoing treatment and when to change it. Therefore, other valid surrogate endpoints have been desired. We have previously demonstrated C-reactive protein (CRP) kinetics could predict prognosis of pts with aRCC (Eur Urol. 2009:1145-53). Methods: This study was performed on 56 pts with aRCC (metastatic: 54, unresectable: 2) enrolled in a phase II trial of interferon-alpha, cimetidine, COX-2 inhibitor and renin-angiotensin-system inhibitor (I-CCA; Cancer Sci. in press). CRP levels were measured at pretreatment, thereafter almost every visit. Pts were divided into 3 groups according to CRP kinetics. Pts whose pretreatment CRP levels were < 5 mg/l were assigned to nonelevated group. Pts whose pretreatment CRP levels were > 5 mg/l but normalized (< 5 mg/l) at least one time during I-CCA were assigned to normalized group. Pts whose CRP level never decreased to normal level were assigned to non-normalized group. Radiographic response was assessed by WHO criteria; survivals were estimated by Kaplan–Meier method and prognostic factors were assessed by Cox's proportional hazard model. Results: Median follow-up was 26 mo. An OR and clinical benefit rate to I-CCA were 20 and 64%, respectively. The median progression-free and overall survival (OS) was 12 and 45 mo, respectively. The median OS was 74, 83 and 13 mo in in non-elevated (n=26), normalized (n=16) and non-normalized (n=14) group, respectively (p<0.0001). Of the 25 pts achieving SD, CRP kinetics was independent prognostic factor for OS (p<0.0001). Of the pts whose pretreatment CRP was elevated, all pts achieving OR had CRP normalization and multivariate analysis revealed CRP normalization was independent prognostic factor for OS (p=0.0008), whereas achieving OR was not (p=0.19). Conclusions: CRP kinetics compares favorably with objective response to systemic therapy as a valid surrogate endpoint for survival. No significant financial relationships to disclose.