Limited sustained remission after nucleos(t)ide analogue withdrawal: Results from a large, global, multi-ethnic cohort of patients with chronic hepatitis B (RETRACT-B study).

BACKGROUND:Complete viral suppression with nucleos(t)ide analogues (NAs) has led to a profound reduction in hepatocellular carcinoma and mortality among chronic hepatitis B (CHB) patients. Finite therapy yields higher rates of functional cure however, initial HBV DNA and ALT elevations are almost certain after treatment interruption. We aimed to analyze off-treatment outcomes beyond 12 months after NA cessation. METHODS:Well-suppressed CHB patients who were HBeAg negative at NA cessation and remained off-treatment without HBsAg loss at 12 months were included (n=945). HBV DNA and ALT fluctuations were allowed within the first 12 months. We used Kaplan-Meier methods to analyze outcomes beyond 12 months. Sustained remission was defined as HBV DNA <2,000 IU/mL and ALT <2x ULN, and an ALT flare as ALT ≥5x ULN. RESULTS:Cumulative probability of sustained remission was 29.7%, virological relapse was 65.2% with a mean peak HBV DNA of 5.0±1.5 log10 IU/mL, an ALT flare was 15.6% with a median peak ALT x ULN of 8.3 (5.7-11.3), HBsAg loss was 9.9% and retreatment was 34.9% at 48 months after NA cessation. A single occurrence of virological relapse or an ALT flare within the first 12 months off-treatment were associated with significantly lower rates of sustained remission beyond 12 months. CONCLUSION:Despite allowing for HBV DNA and ALT fluctuations within the first 12 months off-treatment, most patients without HBsAg loss did not maintain a sustained response thereafter. The best candidates for NA withdrawal are patients with low HBsAg levels at NA cessation, and those without profound or recurrent virological and biochemical relapses in the first off-treatment year.