Correlation of plasma and serum vascular endothelial growth factor levels with platelet count in colorectal cancer: Clinical evidence of platelet scavenging?

Most studies measuring circulating vascular endothelial growth factor (VEGF ) have sampled serum rather than plasma. There has been much debate whether the collection of sera (which causes the activation of platelets and VEGF release) is a true reflection of tumor angiogenic activity or whether plate lets act as scavengers of VEGF. Addressing this issue, we measured serum an d plasma VEGF, before and after colorectal resection, with reference to pla telet counts, Serum and plasma samples were collected from 116 colorectal c ancer (CRC) and 116 control patients. Ninety CRC and 32 benign resections w ere performed. Both plasma and serum VEGF were significantly higher in CRC patients (18.5 and 327 pg/ml, respectively) compared with controls (9.0 and 151.5 pg/ml, respectively; P < 0.0001), Paired serum and plasma VEGF measu rements correlated in both CRC (r = 0.56) and control patients (r = 0.73; P < 0.0001), Serum and plasma VEGF levels correlated with platelet count in CRC patients (r = 0.58 and 0.44, respectively) but not in controls, Plasma and serum VEGF levels, and VEGF concentration per platelet, increased with advancing disease stage. The correlation of serum and plasma VEGF with plat elet counts in CRC but not in benign disease may be attributable to the sca venging of VEGF from the tumor source by platelets, with plasma levels refl ecting free circulating VEGF in equilibrium with platelet levels. VEGF leve ls in citrated plasma are low and lie close to the limits of ELISA sensitiv ity. We recommend that a standardized measurement of serum VEGF-normalized by the patient's platelet count to give a value of serum VEGF per platelet- be adopted.