Platelet × CRP Multiplier Value as an Indicator of Poor Prognosis in Patients With Resectable Pancreatic Cancer.

OBJECTIVE: Thrombocytosis in patients with various cancers has been considered a parameter for poor prognosis; however, its contribution to pancreatic cancer remains controversial.
METHODS: Potential preoperative prognostic parameters (platelets, neutrophils, lymphocytes, the platelet-lymphocyte ratio, the neutrophil-lymphocyte ratio, the serum C-reactive protein [CRP], and carbohydrate antigen 19-9) were retrospectively analyzed in 95 patients with pancreatic cancer. Cutoff values were defined according to receiver operating characteristic curve analysis, and median survival times (MSTs) were compared.
RESULTS: Median survival times (days) significantly differed according to platelet count (high [552] vs low [735], P = 0.017), CRP (high [471] vs low [750], P = 0.001), and carbohydrate antigen 19-9 level (high [639] vs low [765], P = 0.021), whereas there was no difference in the platelet-lymphocyte ratio and the neutrophil-lymphocyte ratio. Multivariate analysis identified thrombocytosis (hazard ratio, 2.015) and CRP level (hazard ratio, 1.771) as independent prognostic factors. The combinatory effects of platelets and the inflammatory response using a platelet × CRP multiplier value could effectively distinguished the MSTs (days) of patients with pancreatic cancer (high [482] vs low [812], P < 0.001).
CONCLUSIONS: Thrombocytosis and CRP influenced pancreatic cancer patient prognosis. Platelet × CRP multiplier is assumed as a useful parameter that reflects the contribution of activated platelets to cancer progression.