We thank Prasitsumrit and Chen for their thoughtful commentary on our work and for highlighting the unmet need for effective risk stratification of hepatocellular carcinoma (HCC) beyond established populations with cirrhosis or chronic viral hepatitis 1, 2. We agree that a substantial proportion of HCC cases arise in individuals without traditional indications for surveillance, demonstrating the importance of developing pragmatic strategies to refine risk assessment in broader populations. As noted by the commentators, the PNPLA3 rs738409 (I148M) variant is one of the most consistently replicated genetic determinants of liver disease progression and HCC risk 3. Its clinical utility, however, has remained uncertain, largely due to challenges in translating genetic associations into actionable risk stratification approaches. Using data from the Singapore Chinese Health Study, an ongoing prospective cohort study of more than 63,000 Chinese Singaporeans 4, we evaluated whether this variant, when combined with readily available clinical factors (i.e., age, sex, body mass index, diabetes status), could meaningfully stratify HCC risk at the population level. We observed a dose–response increased risk of HCC by the rs738409-G allele count that was independent of established clinical factors. Such findings support the biological relevance of PNPLA3 in carcinogenesis and reinforce prior evidence from diverse populations 2, 5. We also appreciate their comment that males aged 50 years or older who carry homozygosity for the I148M variant have markedly elevated HCC risk, approaching levels that may justify surveillance. While such a finding was not intended to directly inform clinical guidelines, it demonstrates the potential value of simple genetic information for identifying high-risk subgroups within otherwise non-cirrhotic populations. The strongest association found among individuals without chronic hepatitis B infection highlights the importance of genetic and metabolic risk factors in non-viral pathways of HCC development. As noted by Prasitsumrit and Chen, the absence of baseline and longitudinal fibrosis measurements is a key limitation of our study. Fibrosis remains the strongest predictor of liver-related outcomes, and future studies integrating non-invasive fibrosis markers, including Fibrosis-4 or elastography, with genetic and cardiometabolic risk factors will be essential to clarify how these dimensions interact over time. Data from population-based cohort studies, including the UK Biobank 6, All of Us 7, Our Future Health 8 and Kadoorie Biobank, which is able to link genetic/genomic data with clinical information extracted from the electronic health records, might be useful for such integrative analyses. In summary, findings from our study 4 may have future implications for research and improved risk stratification, especially for people living with metabolic dysfunction-associated steatohepatitis 9. We showed that a single genetic variant, when combined with common clinical factors such as age, sex, body mass index or diabetes status, can substantially refine HCC risk estimation. This study may stimulate efforts to integrate genetics with clinical information and inform future precision and national strategies for HCC 10. Daniel Q. Huang: conceptualization, writing – original draft, writing – review and editing. Yen Thi-Hai Pham: writing – review and editing. Hung N. Luu: conceptualization, writing – review and editing, writing – original draft. The authors have nothing to report. This article is linked to Huang et al. papers. To view these articles, visit https://doi.org/10.1111/apt.70532 and https://doi.org/10.1111/apt.70550. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Huang et al. (Fri,) studied this question.