New oral anticoagulants and hepatocellular carcinoma among patients with HBV, HCV, and alcoholic liver diseases: a nationwide cohort study.

BACKGROUND

Coagulation pathways are increasingly recognized as an influencer of tumor development. Experimental evidence suggests that commonly used anticoagulants, including warfarin and non-vitamin K oral anticoagulants (NOACs), may exhibit anti-tumor effects.

OBJECTIVES

To examine the association between NOAC use and the risk of hepatocellular carcinoma (HCC) among patients with chronic liver diseases due to hepatitis B virus (HBV), hepatitis C virus (HCV) or alcoholic liver disease (ALD).

DESIGN

Retrospective population-based cohort study.

SETTING

Taiwan National Health Insurance Research Database, 2012-2021.

PARTICIPANTS

Adults with HBV, HCV, or ALD who initiated NOAC or warfarin therapy for at least 28 consecutive days and had no prior cancer diagnosis.

INTERVENTIONS: Use of rivaroxaban, dabigatran, apixaban, or edoxaban vs. warfarin.

MEASUREMENTS: The primary outcome was new-onset HCC, ascertained via national catastrophic illness certification. Adjusted hazard ratios (aHRs) were estimated using Cox proportional hazards models with propensity score matching.

RESULTS

Among 29,332 matched patients, including 14,666 NOAC users and 14,666 warfarin users, NOAC use was associated with a significantly lower risk of HCC (aHR, 0.47; 95% CI 0.26-0.85; p = 0.013), which remained significant in the time-dependent model. The risk reduction was evident in patients with HCV infection and alcoholic liver disease, but not in those with HBV infection. Among individual NOACs, rivaroxaban and dabigatran were associated with a significantly lower risk of HCC.

LIMITATIONS: Potential residual confounding; unavailable laboratory-based indicators of liver disease severity and viral activity; lack of dosage and adherence analyses.

CONCLUSIONS

NOAC prescription was associated with a reduced risk of HCC compared with warfarin, although associations may vary across liver disease etiologies. Prospective studies are warranted to validate these findings.