Prognostic significance of serum S100B protein in high-risk surgically resected melanoma patients participating in Intergroup Trial ECOG 1694.

Abstract

We evaluated adjuvant trial E1694 to more precisely define the prognostic significance of serum S100B in patients with high-risk resected melanoma. Sera from 670 E1694 patients banked at baseline and three additional time points were tested for S100B protein using chemiluminescence. S100B testing results showed that the higher the S100B level is, the higher the risk of relapse and death, regardless of the cutoff value. Univariate analysis showed that baseline S100B > or = 0.15 microg/L is significantly correlated with overall survival (OS; P = .01). Multivariate analysis was performed adjusting for significant prognostic factors (ulceration and lymph node status) and treatment. Baseline S100B was a significant prognostic factor for survival (hazard ratio = 1.39; 95% CI, 1.01 to 1.92; P = .043). S100B values measured at later time points over 1 year were also demonstrated to be significant prognostic factors for relapse-free survival (RFS) and OS. Lower S100B values at baseline and during follow-up were associated with longer survival. A changing S100B from low at baseline to high on follow-up seemed to be associated with the worst RFS and OS. For patients with high-risk surgically resected melanoma, a high baseline or increasing serum S100B is an independent prognostic marker of risk for mortality that may allow us to refine the application of adjuvant therapy in the future.