Prognostic value of soluble P‐selectin levels in colorectal cancer

Measurement of soluble (s) P-selectin levels has been proposed as a diagnostic tool for monitoring the clinical course of human neoplasms. Thus, our study was aimed at analyzing the role of sP-selectin in association with clinicopathological variables in 181 patients with primary (n149) or metastatic (n32) colorectal cancer (CRC), 34 patients with benign diseases and 181 control subjects. The results obtained showed that sP-selectin levels were higher in patients with CRC compared either to patients with benign disease (p 0.006) or controls (p 0.003). No differences were observed between the latter and patients with benign diseases. Increased median sP-selectin levels were significantly associated with the presence of distant metastasis (68.2 ng/ml vs. 48.6 ng/ml, p 0.002). Of interest, carcinoembryonic antigen (CEA) levels were independently associated to sP-selectin (regression coefficient 0.28, p< 0.002). Cox’s proportional hazards survival analysis of primary CRC patients demonstrated that beside the stage of disease sP-selectin levels had an independent prognostic role in predicting recurrent disease (HR 2.22, p 0.019) and mortality from CRC (HR 3.44, p 0.017). These results suggest that measurement of plasma sP-selectin might represent a prognostic indicator in the management of patients with CRC. © 2004 Wiley-Liss, Inc.