Chronic hepatitis B virus (HBV) infection is currently incurable. Long-term treatment with potent and safe nucleos(t)ide analogues (NAs) can reduce hepatocellular carcinoma (HCC), cirrhotic complications and liver-related mortality through substantial viral suppression. However, long-term or indefinite therapy raises several crucial issues with pros and cons. Because HBsAg seroclearance or “functional cure” is not easily achievable, a finite therapy may provide an opportunity to facilitate HBsAg seroclearance by the rejuvenation of exhausted immune cells. However, the virological relapse or surge of alanine aminotransferase (ALT) levels may increase the risk of adverse outcomes (e.g. decompensation, fibrosis progression, HCC, or liver-related mortality) before HBsAg seroclearance, which are the safety concerns of finite therapy. Little is known about whether repeated therapeutic interruption will increase the chance of drug resistance. Whereas the reduction of renal function and bone mineral density are the safety issues of infinite therapy. Lastly, the practice of “to stop” or “to continue” therapy should also consider the accessibility and affordability of the health care system. Patients who stop therapy need to be monitored very closely with frequent virological and biochemical tests in the first year, especially if they experience a virological relapse (VR) or clinical (biochemical) relapse (CR). Around 40% of NA-stoppers eventually receive retreatment. Since infinite NA therapy is unrealistic in clinical practice, the most commonly used endpoint is suppression of HBV DNA to undetectable levels with normalization of ALT (APASL HBV guideline 2016); however, this criterion for cessation of treatment is associated with various incidences of virological and clinical relapse. Recent studies suggested that decreasing HBsAg level at the end of treatment (EOT) to an appropriate cutoff value appears to be a practicable and attainable cessation criterion. In a recent systematic review, a HBsAg level < 100 IU/mL at EOT seems to be a useful marker for deciding when to discontinue NAs therapy. However, regular monitoring is required after the cessation of NAs treatment to detect early clinical relapse, and long-term outcomes need to be further evaluated. In summary, controversial issues exist for the cessation of NAs in HBeAg-negative patients and more large-scale prospective studies are needed.