Abstract
In patients with cancer circulating vascular endothelial growth factor (VEGF) may be tumor-derived and have prognostic significance. Activated platelets may also be a source of VEGF, releasing it in serum formation. Debate exists as to whether serum or plasma VEGF (S-VEGF, P-VEGF) is the most appropriate surrogate marker of tumor angiogenesis. As healing wounds produce VEGF that can spill over into the circulation, we aimed to investigate the potential confounding effects of cancer surgery on both perioperative S-VEGF and P-VEGF levels and to evaluate their relationship with platelet count. S-VEGF, P-VEGF and platelet counts were measured in 23 patients undergoing esophageal cancer resection. Samples were taken preoperatively and six weeks following surgery. Seven patients were also sampled on postoperative days 1, 5 and 10. VEGF was assayed using a commercial enzyme linked immunosorbent assay. S-VEGF and P-VEGF both rose after surgery (S-VEGF; day 5: 1017 [446–1224] pg/mL and day 10: 1231 [626–2046] pg/mL versus pre-op: 329 [189–599] pg/mL. P-VEGF; day 1: 55 [46–104] pg/mL and day 10: 58 [20–154] pg/mL versus pre-op: 23 [13–46] pg/mL), falling towards pre-operative levels by six weeks. Platelet count correlated with S-VEGF (rho=0.281; p<0.05, Spearman's rank) and P-VEGF (rho=0.330; p<0.01, Spearman's rank). Platelets may contribute to VEGF levels in plasma as well as in serum. The effects of surgery on S-VEGF or P-VEGF levels are mainly transient. Care must be exercised when interpreting circulating VEGF levels in the early postoperative period.