Prognostic Models for Small Hepatocellular Carcinoma Using Inflammatory Indices and Machine Learning: A Propensity
1Hepatobiliary & Hydatid Disease Department, Digestive & Vascular Surgery Center, First Affiliated Hospital of Xinjiang Medical University, State Key Laboratory of Pathogenesis, Prevention and Treatment of High Incidence Diseases in Central Asia, Urumqi, People's Republic of China.
International Journal of General Medicine
|June 11, 2026
Summary
This study developed prognostic models for hepatocellular carcinoma (HCC) patients, integrating tumor size and inflammation markers. The LASSO-Cox model accurately predicts early recurrence and long-term survival after resection.
Area of Science:
- Hepatobiliary Surgery
- Oncology
- Biostatistics
Background:
- Prognosis for small hepatocellular carcinoma (HCC) post-resection is variable, with early recurrence posing a significant challenge.
- Existing models lack integration of inflammatory biomarkers for predicting both early recurrence and long-term survival.
- This study focuses on developing and validating prognostic models, particularly for the small HCC subgroup (≤3 cm).
Purpose of the Study:
- To develop and validate prognostic models for overall survival (OS) and recurrence-free survival (RFS) in HCC patients post-resection.
- To assess model performance within the small HCC subgroup.
- To integrate tumor size with inflammatory and liver function markers for refined risk stratification.
Main Methods:
- Retrospective analysis of hepatectomy patients.
- Propensity score matching (PSM) to balance baseline characteristics between small and non-small HCC groups.
- Development and comparison of traditional nomogram, risk score, and machine learning models (LASSO-Cox, Random Forest, XGBoost) using time-dependent AUC.
Main Results:
- Post-PSM, 165 patients (90 small HCC, 75 non-small HCC) were analyzed; small HCC showed better OS and RFS.
- Independent predictors for OS included tumor size, SII, and AAR; for RFS, tumor size, PAR, NLR, and GPR.
- The LASSO-Cox model demonstrated superior performance for predicting early recurrence (2-year RFS AUC = 0.727) and long-term survival (5-year OS AUC = 0.698).
Conclusions:
- A comprehensive prognostic framework integrating tumor size, inflammation (SII, NLR), liver function (AAR, GPR), and nutritional-coagulation status (PAR) was established for small HCC.
- The LASSO-Cox model is recommended for predicting early recurrence and long-term survival.
- Refined risk stratification using the LASSO-Cox model can guide postoperative management for HCC patients.
