Cessation of Nucleos(t)ide Analogue Therapy after HBeAg Seroconversion Might Be Associated with High Relapse Rates and Fatal Outcomes

Introduction

Hepatitis B e-antigen (HBeAg) seroconversion is considered as a very important parameter in assessing the efficacy of nucleos(t)ide analogues (NA) in the treatment of chronic hepatitis B (HBV) infection. However, high relapse rates make it difficult to decide whether the NA treatment should be stopped or continued after NA-induced HBeAg seroconversion. The recommendations of 3 major international scientific hepatological associations - Asian, American and European are also contrasting. In addition, treatment cessation has been linked to high relapse rates in Asian patients. It is difficult to ascertain whether these immunological and clinical outcomes would be similar in the Caucasian population. The Belgian HBV reimbursement criteria suggests cessation of NA after 6 months therapy following HBeAg seroconversion.

Aim

• To ascertain the relapse rates, identify the predictors of relapse and evaluate the clinical outcomes after stopping the NA treatment in a nationwide study.

Method

Study design

• Nationwide observational cohort study
• Chronic HBV-infected HBeAg positive patients with NA-induced HBeAg seroconversion across 18 centers in Belgium
• Out of the total of 2090 patients, 533 received NA treatment and 115 had HBeAg seroconversion
• HBeAg seroconversion was defined as loss of HBeAg or HBsAg and appearance of anti-HBeAg and anti-HBs antibodies on 2 occasions at least 1 month apart
• Consolidation therapy was defined as time between HBeAg seroconversion and cessation of therapy
• Data on demographics, biochemistry, virology, treatment modality, fibrosis stage and disease severity was analyzed
• Histological assessment of fibrosis was done in 92.5% using the Metavir score on pre-treatment liver biopsies
• Viral load, serological and biochemical parameters were estimated

Endpoints

• Virologic relapse defined as a single elevation of HBV DNA >2000 IU/ml
• Combined relapse defined as ALT levels >2 x upper limit of normal (ULN) of ALT (40 IU/ml)
• Relapse was considered persistent on detection on 2 or more occasions consecutively, 3 months apart
• Relapse was considered clinically significant if persistent or if immediate pretreatment was recommended by the physician
• Sustained response was defined as persistent HBV DNA <2000 IU/ml after stopping the treatment
• Appearance of HBeAg was considered as HBeAg reversion
• Both HBsAg loss and HBeAg reversion had to be confirmed on 2 consecutive samples, 1 month apart
• Presence of variceal bleeding, ascites and/or encephalopathy was considered as hepatic decompensation
• Diagnosis of hepatocellular carcinoma and cirrhosis according to international guidelines
• Predictors of relapse were determined using Cox analysis

Results

• 533 out of the 2090 HBeAg-positive patients fulfilled the NA treatment reimbursement criteria and subsequently started NA treatment
• 356 out of 533 had a HBV monoinfection and were not taking immunosuppressive drugs
• 32% had confirmed HBeAg seroconversion after a median treatment duration of 17.7 months
• 9 patients were excluded due to HBsAg loss or LTFU immediately after stopping the treatment
• The final cohort comprised of 98 subjects, out of which 75% were males and 68% were Caucasian
• Follow up for a median duration of 5.8 years
• Treatment was stopped in 62 and continued in 36 patients
• 39% were cirrhotic in continuous treatment group as compared to 18% in the NA stop group (p=0.028)
• The continuous treatment group had a shorter follow-up duration of 2.1 years vs 5.3 years; p<0.001 after HBeAg seroconversion
• The time to HBeAg seroconversion was similar in both the groups (p=0.916)
• The relapse rate was 48.3% in the NA stop group
• Predictors of clinically significant relapse were higher gamma-glutamyl transferase (gamma-GT) levels at both treatment initiation (HR 1.004; p=0.001 per unit increment) and HBeAg seroconversion (HR 1.006; p=0.013 per unit increment)
• Out of the 14 cirrhotic patients in the continuous treatment group, 3 had an episode of ascites but all of them recovered
• In the NA stop group there were 2 liver-related deaths, out of which one had a severe flare
• There were no relapses or severe hepatic outcomes in the continuous treatment group

Conclusion

• Cessation of nucleos(t)ide analogue therapy in Caucasian patients after HBeAg seroconversion led to relapse in more than 50% of the patients with fatal outcomes
• Gamma-GT levels at the initiation of treatment and at HBeAg seroconversion were significant predictors of clinical relapse
• This real world study suggests continuation of NA therapy after HBeAg seroconversion until HBsAg loss.

Aliment Pharmacol Ther. 2018; 47:1170-80.