An important Korean study recruited 133 children (mean age 10.6 years) with immune-tolerant CHB over more than 10 years reported that a low rate (1.7/100 patient-years) of conversion to early immune clearance phase. The estimated transition rate from immune tolerance to clearance increased dramatically with age: it was 4.6%, 7.1%, and 28.0% for patients aged 6, 6-12, >12 years, respectively [8]. Another Korean study of 90 adult HBeAg CHB patients revealed spontaneous HBeAg seroconversion in patients with HBeAg-positive CHB was rare (1/90; 1.6%) within 6 months; in contrast biochemical deterioration occurred up to 18.9% [9]. But this may be different from other parts of the world as Korea is one of the endemic areas of genotype C HBV infection, which is wellknown to be associated with delayed HBeAg seroconversion [10,11]. In a study of 240 Taiwanese CHB patients (of genotype B and C HBV) who were HBeAg positive and normal ALT at baseline, 85% patients had spontaneous HBeAg seroconversion at age 15 to 39 years [12].

The Taiwanese study described previously showed a small yet non-negligible risk of progression from immune-tolerant phase liver cirrhosis, which occurred in 13 patients (5.4%) after a median follow-up of 10.5 years [12]. The key risk factors included older age of HBeAg seroconversion and hepatitis relapse associated [12]. A small retrospective histological series of 40 immune-tolerant CHB patients showed 50% of them had no fibrosis at all; while the other 50% had only F1 fibrosis [13]. This implies liver fibrosis progression is uncommon in immune-tolerant CHB patients.

Subsequently another Hong Kong study involved 453 Chinese HBeAg-positive patients revealed that age >35 years and highnormal ALT>0.5xULN predicted advanced liver fibrosis defined by liver stiffness measurement (LSM) by transient elastography [14,15]. Nonetheless this so-called “high-normal” ALT would be defined as abnormal according to the current AASLD criteria and hence those patients were not truly of immune tolerance. Yet in patients with ALT≤0.5xULN, 6.8% still had advanced fibrosis, the probability would be even high in the subgroup of patients above 35 years old (15%). As LSM is least affected if serum ALT is normal [16,17], these observations supported what the latest EASL guidelines point out not all patients in the immune-tolerant phase would have minimal fibrosis [6].

Histology series in Asian-Americans with positive HBeAg, normal ALT (AASLD criteria) and high HBV DNA (mean 7.7 logs copies/mL) demonstrated up to one-quarter of patients had significant fibrosis of F2-3 [18]. A prospective cohort involved paired transient elastography examinations at a 3-year interval, liver fibrosis progression happened in 3.1% among immune-tolerant patients. Liver fibrosis progressed at similar rate (3.3%) if the immune-active patients received antiviral treatment, but higher risk of such patients did not receive antiviral treatment (13.3%) [19].

HBeAg positivity has known to be a risk factor of hepatocellular carcinoma (HCC) [20]. HCC incidence was 1,169, 324 and 39 per 100,000 person-years among subjects who were positive for both hepatitis B surface antigen (HBsAg) and HBeAg, positive for HBsAg only, and negative for both, respectively [20]. However, HBeAg positivity was not equal to immune tolerance as serum ALT was not analyzed; the information of any subsequent immune clearance or HBeAg seroconversion was also not provided. Subsequently a few HCC risk scores also include HBeAg positivity as one of the factors contributing to the scores. For example, REACH-B score identified gender, age, serum ALT level, HBeAg status, and serum HBV DNA level as the factors increasing the HCC risk in non-cirrhotic CHB patients [21]. But similar to that landmark study, this HBeAg positivity does not necessarily reflecting immune tolerance; instead it reflects immune clearance as patients with elevated ALT and older age are two other risk factors of HCC. Other common HCC risk scores namely CU-HCC score [22,23], GAG-HCC score [24] and LSM-HCC score [25] do not include HBeAg positivity as a contributing factor, instead they put much more weight and emphasis to the presence of cirrhosis or high LSM. This also indirectly reflect patients in immunetolerant phase are not at increased risk of HCC.

More direct evidence had not been available until very recently proving the above concept could have been wrong. A Korean historical cohort study of 413 HBeAg-positive CHB patients with normal ALT (AASLD criteria), high HBV DNA levels (≥20,000 IU/mL) and no evidence of cirrhosis were compared another cohort of 1,497 CHB patients in immune-clearance phase treated with NA [26]. During the long-term follow up, the 10-year estimated cumulative incidences of HCC was significantly higher in the untreated immune-tolerant patients than the treated patients in immune-clearance phase (12.7% vs. 6.1%; P =0.001) [26]. This observation with different conclusion from the above studies may imply either these initially immune-tolerant patients had evolved over time; or the NA treatment reduced the risk of HCC so much that the risk would be even lower than untreated immune-tolerant patients.

A multicenter clinical trial comparing TDF versus TDF and emtricitabine combination therapy for four years in 162 patients with immune-tolerant CHB revealed low virologic response rates – 0% to 5% of HBeAg seroconversion, and 0% HBsAg seroclearance [27]. A extension follow-up study for another 4 years of a subgroup of 20 Chinese patients from this trial showed a high rate of virologic and clinical relapse after stopping NA in these immune-tolerant patients [28]. All patients had virologic relapse (HBV DNA >2,000 IU/mL) at week 4, 10 (50%) patients had clinical relapse (HBV DNA >2,000 IU/mL and ALT >2xULN); among the 9 untreated patients, 7 remained immune tolerant and 2 had HBeAg seroconversion subsequently [28]. These observations imply treating patients with immune-tolerant CHB did not alter its natural history much; once the antiviral treatment is stopped, it would be back to square one of high serum HBV DNA.

Because of the above observations from the landmark clinical trial, the latest AASLD guidelines recommend against antiviral therapy for adults with immune-tolerant CHB [3]. The rationale behind is mainly the lack of studies demonstrating antiviral is beneficial in reducing rates of HCC, cirrhosis and liver-related death in patients with immune-tolerant CHB; hence potential harm, including cost, antiviral drug side effects and development of resistance, outweighs its benefits [3]. AASLD suggests that ALT levels be tested at least every 6 months for adults with immune-tolerant CHB to monitor for potential transition to immune-active or -inactive CHB [3].

EASL guidelines have a slightly different view on that; delayed HBeAg seroconversion beyond 30 years old is already a treatment indication, as patients with HBeAg-positive chronic HBV infection (i.e. persistently normal ALT and high HBV DNA levels) may consider treatment if they are older than 30 years regardless of the severity of liver histological lesions [6]. In contrast, such patients younger than 30 years and do not fulfill any of the treatment indications otherwise should be followed at least every 3–6 months.

Antiviral treatment is recommended when HBeAg-positive patients are already out of immune-tolerant phase, namely with high serum ALT > 2x ULN and HBV DNA >20,000 IU/mL, or serum ALT >1-2x ULN and HBV DNA >2,000 IU/mL with moderate/severe inflammation or significant liver fibrosis. Certainly, HBeAg-positive patients with compensated liver cirrhosis with detectable HBV DNA (while APASL guidelines set HBV DNA >2,000 IU/mL or elevated ALT with detectable HBV DNA), or decompensated liver cirrhosis with detectable HBV DNA [3,4,6].

Apart from the conventional serum ALT and HBV DNA levels, EASL guidelines also put the following factors into consideration before starting antiviral treatment for patients at this phase – HCC risk, reactivation and transmission of HBV; and extrahepatic manifestation [6].

When we are uncertain if an HBeAg-positive patient is really still of immune tolerance, liver biopsy considered to detect advanced liver fibrosis for antiviral therapy, or alternatively with noninvasive assessments such as transient elastography and/or serum tests as first screening tool [29-32]. APASL guidelines 2015 recommend that noninvasive methods for the estimation of the extent of fibrosis are useful in selecting patients for liver biopsy. Patients with the suggestion of significant fibrosis by noninvasive markers (mean liver stiffness >8 kPa (by Fibroscan) or aspartate aminotransferase (AST)-to-platelet ratio index (APRI) >1.5) should be considered for liver biopsy [4].

World Health Organization (WHO) issued her very first specific guidelines for the prevention, care and treatment of persons with CHB in March 2015 [33]. With a global view and coverage (with Africa also included), WHO guidelines provide very practical recommendations with special consideration in resource-limited settings. For example, APRI is recommended as the preferred noninvasive test to assess for the presence of cirrhosis at a cutoff of 2 for adult patients in resource-limited settings [33]. Other more advanced non-invasive tests namely transient elastography or FibroTest may be preferred in settings that are more resourceful [33].

One of the recent breakthrough in CHB is the risk and prevention of maternal-to-child transmission (MTCT) of HBV. Multiple observational studies establish the risk of MTCT of HBV in HBeAgpositive mothers despite HBV vaccination to infants, as very often they have very high serum viral load (6 to 9 log10 IU/mL) [34,35]. This led to the subsequent prospective study of 118 Taiwan HBeAgpositive pregnant women with serum HBV DNA ≥7.5 log10 IU/mL [36]. This was the first study to prove that antiviral treatment with TDF at third trimester to one month post-partum reduce the HBsAg positivity rate of infant compared to no antiviral treatment versus no medication (1.54% vs. 10.71%; P =0.048). A multicenter randomized placebo-controlled trial in China provided the level I evidence of the prevention of MTCT of HBV with antiviral treatment at third trimester [37]. With a slightly lower maternal serum HBV DNA (<200,000 IU/mL) as the inclusion criterion, the perprotocol analysis of the primary outcome, infants with serum HBV DNA >20 IU/mL or HBsAg positivity at 28 weeks of age was 0% in the TDF arm and 6.82% in the placebo arm [37].

With these new data from two important studies, the latest international guidelines recommend HBeAg-positive mothers with HBV DNA >106 IU/mL [6] or 200,000 IU/mL [3] should receive oral antiviral treatment during the last trimester, together with hepatitis B immunoglobulin (HBIG) and HBV vaccination for the newborn. The area of uncertainty now leaves when to stop antiviral treatment. The guidelines recommend to stop immediately or up to 3 months post-partum if the mothers do not meet treatment criteria otherwise [3,6]. Nonetheless it would be important to closely monitor the mother as virologic relapse is a rule rather than exception; biochemical flare also occurs but luckily no hepatic decompensation has been observed [36,37].