# Development of a weighted Alpha-Fetoprotein tumor burden score-integrated nomogram for predicting overall survival in locally ablated hepatocellular carcinoma patients

## TL;DR

This study creates a new tool to predict survival in liver cancer patients treated with local ablation, using a score based on tumor markers and other factors.

## Contribution

A novel nomogram integrating the WATS score for predicting overall survival in locally ablated hepatocellular carcinoma patients is developed and validated.

## Key findings

- WATS, age, drinking history, and prealbumin were identified as independent prognostic factors for overall survival.
- The nomogram showed strong discriminative ability and clinical utility for predicting 3-, 5-, and 8-year survival rates.
- The model effectively stratified patients into low- and high-risk groups with accurate predictive performance.

## Abstract

The Weighted Alpha-Fetoprotein Tumor Burden Score (WATS) shows promise for hepatocellular carcinoma (HCC) prognosis, but its usefulness in local ablation patients is uncertain, and no validated nomograms exist for overall survival (OS) prediction.

This retrospective study enrolled 862 HCC patients who underwent local ablation therapy at Beijing You’an Hospital between January 1, 2015 and December 31, 2022. Participants were randomly allocated into a training cohort (n=603) and validation cohort (n=259) in a 7:3 ratio. Based on the median value of the WATS score, patients were stratified into low-risk (n=431) and high-risk (n=431) groups. The Kaplan-Meier (KM) curve was used to compare the prognosis between the two groups. Potential prognostic factors were screened via least absolute shrinkage and selection operator (Lasso) regression, followed by construction of a WATS-incorporated nomogram prediction model using Cox proportional hazards regression. The SHapley Additive exPlanations (SHAP) method was employed to interpret variable contributions within the model. Model performance was evaluated via Receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). Patients were stratified into low- and high-risk groups according to the nomogram scores, and KM curves were used to compare OS differences between the two groups.

The study identified the WATS, age, history of drinking, and prealbumin as independent prognostic factors for OS, and successfully established a nomogram model for OS prediction. The ROC curves, calibration curves, and DCA all confirmed that the model possesses good discriminative ability, calibration accuracy, and clinical utility. KM curves demonstrated that the nomogram could effectively stratify patients into different risk categories with satisfactory predictive performance.

This study developed and validated a novel prognostic nomogram incorporating the WATS to assess OS in HCC patients receiving local ablation therapy. The nomogram demonstrated robust discriminative ability, enabling accurate prediction of 3-, 5-, and 8-year OS rates, thereby providing clinicians with a reliable tool for individualized risk assessment and treatment decision-making.