Assessing risk scores for predicting hepatocellular carcinoma in Thai patients with chronic hepatitis B

Area of Science:

• Hepatology and Gastroenterology
• Oncology
• Epidemiology

Background:

• Chronic Hepatitis B (CHB) infection is a leading cause of hepatocellular carcinoma (HCC), particularly in Asia.
• Existing HCC risk prediction models are often validated in specific patient groups (treated or untreated).
• Limited data exists on the performance of these models in mixed cohorts of treated and untreated CHB patients.

Purpose of the Study:

• To analyze and validate various HCC risk prediction scores in a mixed cohort of treated and untreated CHB patients.
• To assess the accuracy and comparability of different HCC risk scores.
• To identify potential differences in predictive performance between treated and untreated CHB patient subgroups.

Main Methods:

• A cohort of 2208 CHB patients in a Thai HCC surveillance program was analyzed.
• Baseline clinical and radiological data were used to calculate multiple HCC risk scores.
• HCC development was tracked over a 5.9-year follow-up period, with 0.9% developing HCC.

Main Results:

• Several HCC risk scores demonstrated accurate and comparable overall predictive performance at 5 years.
• Areas under the receiver operating characteristic curves (AUROCs) ranged from 0.70 to 0.80.
• Subgroup analysis indicated superior HCC risk prediction in treated CHB patients compared to untreated patients.

Conclusions:

• Current HCC risk scores are generally accurate for CHB patients but show variability in performance.
• The predictive utility of these scores is diminished in CHB patients not meeting criteria for antiviral therapy.
• Clinicians should exercise caution when applying HCC risk scores to untreated CHB patient populations.