Nationwide retrospective study of hepatitis B virological response and liver stiffness improvement in 465 patients on nucleos(t)ide analogue.

Background: Hepatitis B virus (HBV) nucleos(t)ide analog (NA) therapy reduces liver disease but requires prolonged therapy to achieve hepatitis B surface antigen (HBsAg) loss. There is limited North American real-world data using non-invasive tools for fibrosis assessment and few have compared 1 stgeneration NA or lamivudine (LAM) to tenofovir disoproxil fumarate (TDF). Aim: To assess impact of NA on virological response and fibrosis regression using liver stiffness measurement (LSM) (i.e., FibroScan® ). Methods: Retrospective, observational cohort study from the Canadian HBV Network. Data collected included demographics, NA, HBV DNA, alanine aminotransferase (ALT), and LSM. Patients were HBV monoinfected patients, treatment naïve, and received 1 NA with minimum 1 year follow-up. Results: In 465 (median 49 years, 37% female, 35% hepatitis B e antigen+ at baseline, 84% Asian, 6% White, and 9% Black). Percentage of 64 (n = 299) received TDF and 166 were LAM-treated with similar median duration of 3.9 and 3.7 years, respectively. The mean baseline LSM was 11.2 kPa (TDF) vs 8.3 kPa (LAM) (P = 0.003). At 5-year follow-up, the mean LSM was 7.0 kPa in TDF vs 6.7 kPa in LAM (P = 0.83). There was a significant difference in fibrosis regression between groups (i.e., mean -4.2 kPa change in TDF and -1.6 kPa in LAM, P < 0.05). The last available data on treatment showed that all had normal ALT, but more TDF patients were virologically suppressed (< 10 IU/mL) (n = 170/190, 89%) vs LAM-treated (n = 35/58, 60%) (P < 0.05). None cleared HBsAg. Conclusion : In this real-world North American study, approximately 5 years of NA achieves liver fibrosis regression rarely leads to HBsAg loss.