Chronic hepatitis B (CHB) affects more than 250 million people worldwide and causes annual mortality of nearly 1 million from cirrhosis, hepatocellular carcinoma and other diseases associated with hepatitis B virus (HBV) infection (World Health Organization, 2016; Polaris Observatory Collaborators, 2018). Nucleos(t)ide analogues (NAs) approved for CHB treatment have been demonstrated to reduce HBV disease progression, reverse liver fibrosis and decrease the risk of hepatocellular carcinoma (HCC) development. Entecavir (ETV) and tenofovir disoproxil fumarate (TDF) are clinically used as first-line nucleos(t)ide analogue (NA) antivirals for the treatment of patients with CHB, and both have reasonably improved the rates of virological suppression, biochemical and serological response. Both drugs display high genetic barriers with very minimal resistance and high rates of viral suppression (Tenney et al., 2009; Chang et al., 2010; Yang et al., 2013).

In controlled clinical trials, most but not all patients with CHB experience undetectable serum HBV DNA levels on oral antiviral therapy. Reported rates of undetectable serum HBV DNA (<300 copies/mL) range from 67% to 97% among hepatitis B e antigen–positive (HBeAg+) patients (Chang et al., 2006; Gish et al., 2007; Pan et al., 2012; Marcellin et al., 2013) and 90% to 97% among HBeAg-negative (HBeAg-) patients (Lai et al., 2006; Marcellin et al., 2013). Precisely why a proportion of patients do not achieve an undetectable serum HBV DNA level despite apparently effective antiviral treatment has not been fully explored. Both the stability of the HBV covalently closed circular DNA (cccDNA) in the hepatocyte nucleus (Allweiss and Dandri, 2017), and the resistance of HBV to NAs (Sinn et al., 2011) are believed to be some mechanisms.

Some reports suggested that persistent positive in HBV DNA level after NA therapy was associated with a higher risk of hepatocellular carcinoma (HCC) occurrence and fibrosis progression (Kim et al., 2017; Sun et al., 2020). However, it had also been reported that low-level viremia (LLV, <2,000 IU/mL) during treatment was not a predictive factor for HCC and cirrhotic complications in patients with treatment-naïve CHB and good adherence to ETV treatment (Lee et al., 2020). And Lee et al. suggested that episodic LLV among untreated patients with compensated cirrhosis did not increase the risk of disease progression compared with maintained virological response status (Lee et al., 2022). As a result, it remains unclear whether it is more beneficial to continue original NAs or to switch/add another NA in order to prevent liver-related events including fibrosis progression and HCC development in persistent viremia (PV, above the lower limit of quantification, 20 IU/ml) patients. The AASLD 2018 hepatitis B guidance suggested that persons with persistent LLV on ETV or TDF monotherapy should continue monotherapy, but the quality and certainty of this evidence was very low (Terrault et al., 2018).

Therefore, the objectives of this study were to identify factors associated with PV and investigate whether patients with PV were associated with HCC development and fibrosis progression, so as to further explore the next treatment options for patients with PV, by analyzing data collected from a well-characterized cohort of CHB patients that have been treated with ETV for 78 weeks.

In the era of NAs therapy, most patients with CHB achieve CVR and the majority experience fibrosis regression including the reversal of cirrhosis (Chang et al., 2006; Lai et al., 2006; Chang et al., 2010; Xu et al., 2015). However, despite these highly beneficial outcomes, a small number of patients still fail to achieve CVR, and the clinical significance of a PV with respect to the outcome of CHB is not clear. In this analysis of predictive factors for PV after 78 weeks of ETV treatment in a well-characterized cohort of CHB patients, we found that HBV DNA level ≥8 log₁₀ IU/mL, Anti-HBc level<3 log₁₀ IU/mL, and HBeAg seropositivity are independently predictive of failure to achieve CVR. Furthermore, of the 11 HBeAg-positive patients with HBV DNA level ≥8 log₁₀ IU/mL and Anti-HBc level<3 log₁₀ IU/mL at baseline, 9 (81.8%) had persistent positive in HBV DNA level and 0 had fibrosis progression at week 78 of treatment. Of the 90 patients with PV, only 10 (11.1%) had fibrosis progression and 0-7 (0-7.8%) were with high risk of HCC occurrence according to different HCC risk scores.

Higher levels of HBV DNA are associated with an increased risk for HCC and cirrhosis (Chen et al., 2011), but its impact on virological response is debatable. Yuen et al. reported that 100% of treatment-naïve HBV patients who received entecavir for 3 years with baseline HBV DNA <8 log₁₀ copies/mL had undetectable HBV DNA (<12 IU/mL), whereas only 75% of patients with baseline HBV DNA ≥8 log₁₀ copies/mL did (Yuen et al., 2011). Gordon et al. reported that equal virologic responsiveness between CHB patients with high viral load (HVL) (HBV DNA ≥9 log₁₀ copies/mL) and with non-HVL, 98.3% of HVL and 99.2% of non-HVL patients achieving HBV DNA <400 copies/mL by week 240 (Gordon et al., 2013). Chan et al. reported that 55% of HBeAg-positive patients with high levels of HBV DNA (mean baseline level of HBV DNA of 8.41 log₁₀ IU/mL) and normal levels of ALT treated with TDF had levels of HBV DNA <69 IU/mL at week 192 (Chan et al., 2014). In our analysis, only 77.2% of treatment-naïve CHB patients who received entecavir for 78 weeks achieve CVR, 81.5% of patients with baseline HBV DNA <8 log₁₀ IU/mL achieved CVR (HBV DNA <20 IU/mL), whilst the CVR rates were only 41.9% in patients with baseline HBV DNA ≥8 log₁₀ IU/mL. The poor efficacy of ETV may be related to previous using of lamivudine (LAM) and telbivudine (LdT), but the proportion of such patients was similar between the two groups. This finding may be attributed to the short time courses of antiviral therapy or some patients may be in immune-tolerant phase. Larger longitudinal studies are warranted to explore this factor further and its potential effect on the virological response to antiviral treatment in CHB patients.

Previous studies have shown that rates of CVR in clinical trials of CHB patients were different between HBeAg+ and HBeAg- patients, and the former tended to have lower rates of CVR compared to the latter (Chang et al., 2006; Lai et al., 2006; Gish et al., 2007; Pan et al., 2012; Marcellin et al., 2013). Consequently, the observed associations between HBeAg-positive and a PV state is not entirely surprising. Among 875 treatment-naive chronic hepatitis B virus (HBV) mono-infected patients, 377 patients with low-level viremia (LLV; <2,000 IU/mL), Kim et al. reported that HBeAg status was the only significant factor associated with LLV (Kim et al., 2017), which coincided with our results.

Our finding that Anti-HBc level at baseline is associated independently with failure to achieve CVR on NAs therapy is novel and may relate to the intensity of liver inflammation. It remains controversial of the association between ALT levels and liver inflammation. Some studies have suggested that substantial CHB patients with normal ALT levels exhibit severe liver damage. Chao et al. reported that approximately one fifth (20.7%) of CHB patients with ALT ≤ 40 IU/L may have significant hepatic fibrosis. The corresponding proportion was 27.8% even when the newer ULN of 30 IU/L (males) and 19 IU/L (females) was applied (Chao et al., 2014).

Recent studies proposed that serum anti-HBc levels could serve as a promising marker for predicting the severity of liver inflammation and exhibited a high diagnostic accuracy in CHB patients with normal ALT (Song et al., 2015; Zhou et al., 2017). In that studies, Anti-HBc can accurately reflect the inflammation of the liver. In our analysis, the levels of ALT at baseline were comparable between the two groups, but the level of Anti-HBc in patients with PV was lower than patients in CVR (3.6 vs 3.9 log₁₀ IU/mL, P<0.001), which indicated that Anti-HBc is a more responsive indicator of liver inflammation than ALT. Of 49 patients with Anti-HBc level<3 log₁₀ IU/mL, 22 (44.9%) failed to achieve CVR, which suggested that some of patients with low levels of Anti-HBc may be in the immune tolerance phase.

Evidence has emerged that incomplete virologic suppression, particularly intrahepatic viral transcriptional activity, was associated with abnormal liver histopathology in a cross-sectional study (Wang et al., 2017). Sun et al. suggested that detectable low-level HBV DNA was associated with fibrosis progression in patients with chronic HBV infection during 78 weeks of entecavir therapy (Sun et al., 2020). But in our study, patients with PV were less likely to have a worse LSM value (0% vs 5.9%, P=0.038) and fibrosis progression (11.1% vs 21.4%, P=0.029) than patients with CVR during 78 weeks of entecavir therapy. Furthermore, Patients with PV were more likely to have a low-risk of HCC development than those with CVR. High viral load (HVL) (HBV DNA≥8 log₁₀ copies/mL), positive HBeAg, slightly liver inflammation (Anti-HBc <3 log₁₀ IU/mL), low-risk of fibrosis progression and HCC development, all these suggested that these particular PV patients may be in the immune tolerance phase (Sarin et al., 2016; European Association for the Study of the Liver, 2017; Terrault et al., 2018). Even with potent drugs like entecavir, their virological response was unsatisfactory. Hence, the initiation of antiviral therapy could be delayed or waiting for stronger and more effective drugs may be alternative choice. Further prospective studies with larger sample size and longer follow-up are needed to confirm this finding and may provide some hint to optimization of histological and longterm clinical outcomes of antiviral therapy.

Our study has several limitations. First, the definition of PV was based on a single serum HBV DNA measurement at week 78. Although HBV DNA was measured at week 0, 26, 52 and 78 in local hospitals, respectively, we selected week 0 and 78 data detected at central laboratory to ensure the consistency of the results. Second, the follow-up time was relatively short. There was a possibility that PV observed at week 78 may turn to CVR after longer follow-up. Third, the development of drug-associated mutations was not investigated. However, considering the good compliance of patients and the extremely low resistance rate of entecavir, HBV mutations are unlikely to be the main cause of PV during antiviral therapy. Finally, our study used only ETV as a first choice for treatment of naïve CHB patients. Other NAs may have differential results. Therefore, the result of present study applies only to the patients using ETV.

In conclusion, our study suggested that HBV DNA level ≥8 log₁₀ IU/mL, Anti-HBc level<3 log₁₀ IU/mL and HBeAg seropositivity at baseline contributed to persistent positivity in HBV DNA level in patients with CHB receiving 78-week antiviral treatment. The risk of HCC development and the rate of fibrosis progression in these patients were low. Maybe these particular PV patients were in the immune tolerance phase, and the initiation of antiviral therapy could be delayed or waiting for stronger and more effective drugs may be another choice. Further well-designed prospective studies with large-scale and longer follow-up are needed to confirm this finding and may provide some hint to judge true immune tolerance phase.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

The study was approved by the Ethics Committee of Peking University First Hospital and other 23 teaching hospitals. The patients/participants provided their written informed consent to participate in this study.

JL, X-QD, HZ and G-QW designed the experiments; HZ and G-QW provided the overall principle and direction of the study; X-QD and JL gathered and analyzed data, and drafted the manuscript; X-QD, Y-QL, and CZ done the laboratory examination. L-HC, Z-QZ, W-FZ, Q-HS, D-ZZ, A-LM, QX, H-LG, GZ, Y-XL, JS, S-BX and China HepB Related Fibrosis Assessment Research Group had participated in acquisition of data, revision of the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.