Evidence on the carcinogenicity of oral nucleos(t)ide analogues (NAs) is inconclusive and lacks data on the effects by chemical structure of the NAs in patients with chronic hepatitis B (CHB).… Click to show full abstract
Evidence on the carcinogenicity of oral nucleos(t)ide analogues (NAs) is inconclusive and lacks data on the effects by chemical structure of the NAs in patients with chronic hepatitis B (CHB). We aimed to provide definitive results on this issue using a large set of CHB patients and data on all major NA drugs. The study population consisted of 10,331 patients with CHB receiving primary NA treatment for more than 6 months, and 24,836 untreated controls followed for at least as long as the treated patients. Using the inverse‐probability‐of‐treatment‐weighted (IPTW) method, the cumulative incidence of extrahepatic cancers was compared in the treated and untreated patients and across the cyclopentane, L‐nucleoside and acyclic phosphonate categories of NAs. Analyses of individual cancers as sub‐endpoints were also performed. The cumulative incidence of overall extrahepatic malignancies did not differ between the two groups in the IPTW cohort (hazard ratio [HR] 1.002; 95% confidence interval [CI] [0.859–1.169]). Similar statistical trends were observed in analyses across the three NA chemical subsets and controls. Per‐cancer analyses indicated that NA treatment was significantly associated with increased risks of colorectal/anal cancers (HRs [95% CI], 1.538 [1.175–2.013]) and lymphoma (1.784 [1.196–2.662]). Conversely, breast cancer (HRs [95% CI], 0.669 [0.462–0.967]) and prostate cancer (0.521 [0.329–0.825]) were less prevalent in the NA‐treated group. In conclusion, prolonged NA treatment presents carcinogenic risks for colorectal/anal and lymphoid tissues in CHB patients, although it does not affect most extrahepatic organs. The protective effect of NAs on breast and prostate cancers should be confirmed.
               
Click one of the above tabs to view related content.