Can we stop nucleoside analogues before HBsAg loss?

Most of the current guidelines and the existing data suggest that long-term therapy with nucleos(t)ide analogue(s) [NA(s)] may be stopped in carefully selected chronic hepatitis B patients who remain HBsAg positive. In particular, NA(s) may be discontinued in such patients without pre-existing cirrhosis who achieved long-term on-therapy virological remission (>12 months of HBeAg seroconversion and HBV DNA undetectability for initially HBeAg-positive cases; >= 3 years of HBV DNA undetectability for HBeAg-negative cases) and are expected to remain under close follow-up after NA(s) discontinuation. The majority of patients will develop post-NA(s) virological relapses and a proportion of them will have biochemical relapses and occasionally flares, but prompt retreatment can reintroduce remission. No reliable predictor(s) of post-NA(s) relapses have been identified so far. HBsAg loss develops in a progressively increasing proportion of chronic hepatitis B patients who discontinue NA(s) with HBsAg loss rates being higher in Caucasian patients with HBeAg-negative chronic hepatitis B. Follow-up at least every 3 months for the first year seems to be appropriate for all chronic hepatitis B patients who discontinue NA(s), while HBeAg-negative patients need to be followed more closely (monthly) during the first 3 months. Predefined criteria for retreatment are quite important, and the best candidates for retreatment are probably the patients with persistent (>= 3 months) liver disease activity and those with severe flares.