A preoperative risk score based on early recurrence for estimating outcomes after resection of hepatocellular carcinoma in the non-cirrhotic liver.

BACKGROUND: Liver resection is the mainstay treatment option for patients with hepatocellular carcinoma in the non-cirrhotic liver (NCL-HCC), but almost half of these patients will experience a recurrence within five years of surgery. Therefore, we aimed to develop a rationale-based risk evaluation tool to assist surgeons in recurrence-related treatment planning for NCL-HCC.

METHODS: We analyzed single-center data from 263 patients who underwent liver resection for NCL-HCC. Using machine learning modeling, we first determined an optimal cut-off point to discriminate early versus late relapses based on time to recurrence. We then constructed a risk score based on preoperative variables to forecast outcomes according to recurrence-free survival.

RESULTS: We computed an optimal cut-off point for early recurrence at 12 months post-surgery. We identified macroscopic vascular invasion, multifocal tumor, and spontaneous tumor rupture as predictor variables of outcomes associated with early recurrence and integrated them into a scoring system. We thus stratified, with high concordance, three groups of patients on a graduated scale of recurrence-related survival.

CONCLUSION: We constructed a preoperative risk score to estimate outcomes after liver resection in NCL-HCC patients. Hence, this score makes it possible to rationally stratify patients based on recurrence risk assessment for better treatment planning.