Nomogram based on inflammation-related markers for predicting survival of patients undergoing hepatectomy for hepatocellular carcinoma

Abstract

BACKGROUND Previous nomograms for hepatocellular carcinoma (HCC) did not include the neutrophil-to-lymphocyte ratio (NLR) or platelet-to-lymphocyte ratio (PLR). Thisstudy aimed to establish an effective nomogram capable of estimating theassociation between preoperative inflammatory factors and overall survival (OS)of HCC patients after hepatectomy.AIMTo analyse the factors affecting the prognosis of HCC and establish a nomogram.METHODSA total of 626 HCC patients (410 training set patients from the First AffiliatedHospital of Anhui Medical University and 216 validation set patients from theFirst Affiliated Hospital of University of Science and Technology of China)underwent hepatectomy from January 2014 to December 2017 and were followedup every 3–6 mo. The nomogram was based on OS-related independent riskfactors identified by Cox regression analysis. The C-index, calibration curve, andarea under the curve (AUC) were used to evaluate the nomogram’s accuracy.RESULTSThe 1-, 2 and 3-year OS rates were 79.0%, 68.0% and 45.4% in the training cohort(median OS = 34 mo) and 92.1%, 73.9% and 51.2% in the validation cohort(median OS = 38 mo). Higher α-fetoprotein [hazard ratio (HR) = 1.812, 95%confidence interval (CI): 1.343–2.444], NLR (HR = 2.480, 95%CI: 1.856–3.312) andPLR (HR = 1.974, 95%CI: 1.490–2.616), tumour size ≥ 5 cm (HR = 1.323, 95%CI:1.002–1.747), and poor differentiation (HR = 3.207, 95%CI: 1.944–5.290) weresignificantly associated with shortened OS. The developed nomogram integratingthese variables showed good reliability in both the training (C-index = 0.71) andvalidation cohorts (C-index = 0.75). For predicting 1-, 2 and 3-year OS, thenomogram had AUCs of 0.781, 0.743 and 0.706 in the training cohort and 0.789,0.815 and 0.813 in the validation cohort. The nomogram was more accurate inpredicting prognosis than the AJCC TNM staging system.CONCLUSIONThe prognostic nomogram combining pathological characteristics and inflammationindicators could provide a more accurate individualized risk estimate forthe OS of HCC patients with hepatectomy