43
Background: Serum C-reactive protein (CRP), a marker of inflammation, appears to demonstrate a prognostic impact in small studies of metastatic prostate cancer (MPC). We conducted a meta-analysis to better quantitate its prognostic impact. Methods: A systematic review was conducted to identify publications and presentations exploring the association of baseline serum CRP and overall survival (OS) in MPC, both castration-sensitive and castration-resistant. The hazard ratio (HR) and 95% confidence interval (CI) with P values were pooled. HRs were selectively extracted from multivariable analyses. Estimates of HRs were weighted and pooled using the generic inverse variance and random-effect models. Heterogeneity among trials was assessed using Cochrane’s Q statistic and the I
2
statistic was used to quantify inconsistency among trials. The assumption of homogeneity was considered invalid for P<0.1. All statistical tests were two-sided, and P<0.05 was considered significant Results: Six studies comprising 659 evaluable patients were eligible. The first authors were Prins (N=119), Pond (N=112), Ito (N=80), Nakashima (N=126), Beer (N=160), and McArdle (N=62). The Nakashima and McArdle studies evaluated castration-sensitive men, while the remaining four studies evaluated castration-resistant men receiving docetaxel-based chemotherapy. The primary endpoint was OS except the McArdle study, which used cancer specific survival. All HRs were derived from multivariate analyses except the Pond study, which was a bivariate analysis. Men with higher CRP had significantly worse OS than those with lower CRP (HR = 1.42, P<0.001, 95% CI: 1.17 to 1.73; I
2
= 72.6%, p = 0.003). In trials of castration-sensitive men receiving hormonal therapy, high CRP yielded a HR = 1.92 (P=0.005, 95% CI: 1.22 to 3.03; I
2
= 0). In castration-resistant men receiving chemotherapy, high CRP yielded HR=1.35 (p=0.003, 95% CI: 1.11 to 1.65; I
2
= 79%), P for subgroup difference=0.20. Conclusions: This meta-analysis suggests a strong prognostic impact for CRP on OS in men with both castration-sensitive and castration-resistant prostate cancer. Prospective validation is justified, given the affordability, ready availability and large dynamic range of CRP.