According to the World Health Organization (WHO), approximately 296 million people globally suffer from chronic hepatitis B (CHB). Despite the advent of new therapies, long-term treatment with nucleos(t)ide analogs (NAs) remains the primary strategy for managing CHB patients. NAs effectively suppress hepatitis B virus (HBV) DNA and reduce the risk of hepatocellular carcinoma (HCC) and cirrhosis. However, stopping NAs carries risks, including hepatitis flares, liver decompensation, and death. A recent study published in JOH investigated the incidence and clinical outcomes of hepatitis flares following NA discontinuation and identified risk factors to help clinicians assess high-risk patients and determine when to resume antiviral therapy.Hepatitis flares are common and potentially harmful after stopping NAs, making it essential to identify predictive factors for risk stratification and post-discontinuation follow-up. This multicenter cohort study involved CHB patients with viral suppression who were HBeAg-negative at the time of stopping NAs. Hepatitis flares were defined by ALT levels reaching ≥5x, 10x, or 20x the upper limit of normal (ULN). Researchers conducted a multivariate Cox regression analysis, using retreatment, HBsAg clearance, and loss to follow-up as endpoints. A subgroup analysis further explored HBV DNA levels within 12 weeks as a time-dependent covariate.
The study enrolled 1,552 participants, predominantly male and Asian. At the end of treatment (EOT), the median age was 54, and the median HBsAg level was 2.76 log10 IU/mL. A total of 1,311 patients had ALT and/or HBV DNA tests within three months, and 1,107 patients had tests within 6 to 12 months.
Following NA discontinuation, 350 participants experienced a hepatitis flare (ALT≥5x ULN), with 70% of these flares occurring within the first year. The cumulative one-year incidence rates for ALT≥5x ULN, ALT≥10x ULN, and ALT≥20x ULN were 18.6%, 10.2%, and 3.4%, respectively. The severity of flares decreased over time, but severe cases could still occur beyond the first year. Thirteen patients experienced liver decompensation following hepatitis flares, and three patients died.
Timing and Clinical Outcomes of Hepatitis Flares
In the first three months after the flare, 29% of patients experienced severe hepatitis flares (ALT≥20x ULN), with 91% of these severe cases occurring after stopping tenofovir disoproxil fumarate (TDF). Even one year after the flare, the median ALT peak level remained at 9.06 (6.34–12.2)x ULN, indicating significant clinical relevance. Compared to early flares, patients with late flares (more than one year after stopping NAs) showed a slight increase in HBV DNA levels prior to the flare.
Thirteen patients developed liver decompensation, five of whom had a history of cirrhosis, and four showed signs of advanced fibrosis. Nine patients exhibited clinical symptoms, and among them, seven had a history of cirrhosis or advanced fibrosis. Three patients died due to liver-related complications.
After the hepatitis flares, 13 patients achieved HBsAg clearance, including two patients who cleared HBsAg following severe flares (ALT≥20x ULN). The median time from flare to HBsAg clearance was 134 days, with eight patients achieving clearance within 12 months of stopping NAs.
Predictive Factors for Hepatitis Flares
In a multivariate Cox proportional hazards model, researchers assessed various factors affecting hepatitis flares. The analysis revealed that older age, male gender, TDF treatment, and higher HBsAg levels at EOT were predictors of hepatitis flares with ALT≥5x ULN.
TDF treatment was also identified as an independent risk factor for ALT≥20x ULN flares. Additionally, patients with a history of NA treatment were at higher risk of severe flares (ALT≥10x ULN), particularly those treated with TDF. Elevated HBV DNA levels (>5 log10 IU/mL) were also a strong predictor of severe hepatitis flares (ALT≥20x ULN).
In a subgroup analysis, 86 patients had HBV DNA levels above 5 log10 IU/mL. Among these, 37 patients restarted treatment before ALT elevation, while 30 of the remaining 49 patients developed flares after their HBV DNA levels increased. Among patients who experienced flares within the first year after stopping NAs, 44% developed ALT elevation within three months if no intervention was initiated when HBV DNA levels exceeded 5 log10 IU/mL, compared to 29% when levels were just over 4 log10 IU/mL.
The study concluded that hepatitis flares are common after stopping NAs, especially within the first year, and may lead to liver decompensation and death. Older age, male gender, higher HBsAg levels at the end of treatment, and TDF treatment are associated with an increased risk of flares. If HBV DNA levels exceed 5 log10 IU/mL within the first 12 weeks, close monitoring and retreatment should be considered.
