Dynamic change of the systemic immune inflammation index predicts the prognosis of patients with hepatocellular carcinoma after curative resection

Authors:
Publisher website
Google Scholar
Abstract
Background: The aim of the study was to determine the utility of the dynamic change and serial monitoring of the systemic immune inflammation index (SII), which was based on the numbers of patients’ lymphocytes (L), platelets (P), neutrophils (N) and defined as P*N/L, for predicting prognosis of patients with hepatocellular carcinoma (HCC) after curative resection. Methods: We conducted a prospective study of 163 patients with HCC who underwent curative resection at Zhongshan Hospital from January 2012 to May 2013. SII was calculated using data acquired before and approximately 1 month after surgery. An optimal cutoff value stratified patients into groups with high or low SII. Patients were classified into unfavorable and favorable groups using the dynamic change of the SII. Two groups that were further divided into four categories within the entire cohort and the low-risk subgroups were serially monitored for ≥6 months. Prognostic values of the SII and other factors were determined using the Kaplan-Meier method, the Cox proportional hazards model, and the receiver operating characteristics (ROC) curve. Results: The favorable group was likely to have cirrhosis, and the unfavorable group was likely to have larger tumors and a higher recurrence rate. Multivariate analysis revealed that tumor size and dynamic change of the SII were independent risk factors for early recurrence. Moreover, the predictive value of the SII was retained in α-fetoprotein (AFP)-negative and HBeAg-negative-HBV-DNA <2000 IU/mL subgroups. Further, the serial changes of the SII for recurrence and no recurrence groups were statistically significant. Conclusions: The dynamic change and serial monitoring of the SII represent new indicators for predicting the early recurrence of HCC determining advance optimal therapy in advance.

This publication has 38 references indexed in Scilit: