Although the ideal end point for
antiviral treatment in patients with
chronic hepatitis B (CHB) is loss of
HBsAg, the typical clinical end points are
HBeAg seroconversion in
HBeAg-positive patients and long-term
DNA suppression in
HBeAg-negative patients. We evaluated the long-term
antiviral response after cessation of
lamivudine treatment in CHB patients. A total of 157 patients who had discontinued
lamivudine between 1997 and 2014 were enrolled (97
HBeAg-positive and 60
HBeAg-negative CHB patients). The long-term durability of the
antiviral response (viralogical relapse; HBV
DNA ≥104 copies/ml) and the
clinical course of these patients were analyzed retrospectively. In
HBeAg-positive patients, the mean follow-up period after discontinuation was 72.3 months. The cumulative probabilities of virological relapse at 1, 12, 24, 48, 60, 96, and 120 months were 10.3%, 40.2%, 55.6%, 62.8%, 65.9%, 67.0%, and 67.0%, respectively. In
HBeAg-negative patients, the cumulative probabilities of a virological relapse at 1, 12, 24, 48, 60, 96, and 120 months were 25.0%, 35.0%, 41.7%, 43.3%, 43.3%, 46.7%, and 48.3%, respectively. Younger age (HR 1.732, 95%CI: 1.058-2.835, P = 0.02) was predictive of non-virological relapse in
HBeAg-positive patients. And achievement of undetectable HBV
DNA level within 3 months of treatment discontinuation was associated with decreased rate of virological relapse (HR 0.159, 95%CI: 0.069-0.367 P < 0.01) in
HBeAg-negative patients. Despite meeting the requirements for treatment discontinuation, approximately half of the CHB patients treated with
lamivudine relapsed. Thus, the
antiviral response is not reliably sustained after
lamivudine treatment cessation. J. Med. Virol. 89:849-856, 2017. © 2016 Wiley Periodicals, Inc.