ALT to qHBsAg ratio predicts long-term HBsAg seroclearance after entecavir cessation in Chinese patients with chronic hepatitis B.

ABSTRACT

BACKGROUND & AIMS: Factors predicting HBsAg seroclearance after treatment cessation, irrespective of nucleos(t)ide analogue (NA) resumption, have important clinical implications. We evaluated predictors of long-term HBsAg seroclearance after entecavir cessation. METHODS: This study followed-up Chinese patients with chronic hepatitis B from two previous studies of entecavir cessation. All patients were non-cirrhotic, HBeAg-negative, with undetectable HBV DNA (<20 IU/ml) at end-of-treatment (EOT). They were monitored closely for 48 weeks with regular HBV DNA, quantitative HBsAg (qHBsAg) and alanine aminotransferase (ALT) measurements. Entecavir was resumed at HBV DNA >2,000 IU/ml, irrespective of ALT levels. After the initial 48 weeks, patients were assessed every 6 months, regardless of entecavir resumption, to monitor for HBsAg seroclearance. RESULTS: A total of 194 patients (63.4% male, mean age 49.9 years, on entecavir for a median of 47.2 months) were recruited; 94 (48.5%) and 158 (81.4%) patients had EOT qHBsAg <100 IU/ml and <1,000 IU/ml, respectively; 151 (77.8%) patients were eventually resumed on entecavir. After follow-up for a median of 70.7 (51.0-118.2) months, 28 (14.4%) patients had HBsAg seroclearance. qHBsAg levels at weeks 36 and 48 after EOT independently predicted HBsAg seroclearance (both p <0.01), whereas qHBsAg from EOT to week 24 only trended towards statistical significance. The ratio of ALT/qHBsAg at all time points from EOT to week 48 independently predicted HBsAg seroclearance (hazard ratios ranging from 1.003-1.028, all p <0.01) with excellent diagnostic performance (area under the receiver-operating characteristic curve 0.799-0.933, negative predictive value >90% at different time points), regardless of whether entecavir was resumed. CONCLUSIONS: The ALT/qHBsAg ratio after entecavir cessation predicts HBsAg seroclearance, even in patients who were resumed on treatment. Its use may mitigate the risk of severe hepatitis flares in patients managed by observation without treatment resumption. IMPACT AND IMPLICATIONS Current predictors of HBsAg seroclearance after finite nucleos(t)ide analogue (NA) therapy have suboptimal predictive value. We demonstrated that the ALT/qHBsAg ratio may be able to reflect the balance between host control and virological activity. The ALT/qHBsAg ratio at different time points from end-of-treatment till week 48 independently and accurately predicted HBsAg seroclearance in patients who have stopped entecavir. The ALT/qHBsAg ratio may be utilized by clinicians for patient selection and retreatment decisions in finite NA therapy.