Prognostic Impact of Increased Perioperative Platelet Count in Gastric Cancer Patients

Abstract

BackgroundSolid tumors are a common cause of secondary (reactive) thrombocytosis, a paraneoplastic syndrome that is also a prognostic factor for various cancers. However, cutoff values for platelet count specific to gastric cancer and their prognostic roles are unknown. MethodsWe retrospectively analyzed records of 4643 patients with gastric cancer who underwent radical surgery from 2007 to 2010. The minimum P-value approach was used with the log-rank test to determine the optimal prognosis predicting threshold for preoperative platelet count. Change in perioperative platelet count over time was evaluated with a generalized estimating equation. Clinicopathologic features and prognostic significance were analyzed according to platelet count. ResultsThrombocytosis prevalence (platelet count ≥40 × 104/μL) was 1.6% (75 of 4643 patients). The platelet count cutoff value with the lowest P-value was ≥25.5 × 104/μL, and patients with higher platelet count had more advanced disease. Multivariate analysis showed that cutoff value was an independent prognostic factor of overall survival (hazard ratio 1.19, 95% confidence interval 1.03-1.37, P = 0.017). Patients who underwent adjuvant chemotherapy (n = 1300) and had a greater than 10% increase in platelet count at postoperative 1 y compared to before surgery had significantly poorer overall survival (hazard ratio 1.65, 95% confidence interval 1.11-2.45, P = 0.013). ConclusionsElevated preoperative platelet count (≥25.5 × 104/μL) and increased platelet count (≥10%) at postoperative 1 y in an adjuvant chemotherapy group were unfavorable prognostic factors. Platelet count could be a cost-effective biomarker for screening and monitoring patients with unfavorable survival outcomes.