Introduction
Hepatitis B virus (HBV), the causative factor of chronic hepatitis B (CHB), belongs to the
Hepadnaviridae family. Given that an estimated number of 650 000 people die annually from CHB and relevant diseases, including cirrhosis and hepatocellular carcinoma, chronic HBV infection remains the 10th leading cause of human death [
1–
3]. The eradication of covalently closed circular DNA (cccDNA) has been suggested to be the therapeutic goal in chronic HBV infection because the persistence of cccDNA in infected hepatocytes causes off-therapy DNA rebound and makes CHB difficult to cure [
4]. The two kinds of antiviral agents used for treatment of CHB are pegylated interferon (Peg-IFN-α) and nucleos(t)ide analogue (NA); however, they do not directly target cccDNA [
5,
6]. Although NA have been confirmed to be highly effective for treatment of CHB, long-term or even life-long NA treatment has been recommended in almost all guidelines for the management of CHB to prevent viral rebound and disease relapse [
7–
10]. However, poor compliance and increased risk of side effects are the challenges of life-long NA therapy.
The recent discovery of HBV RNA virion-like particles completes the life cycle of HBV infection [
11,
12], offering an opportunity to rethink the nature of the real viral response. Why do off-therapy HBV DNA rebound and disease relapse frequently occur after consolidation therapy? Does undetectable serum HBV DNA reflect the true virological response (VR) to antiviral therapy, particularly in patients receiving NA therapy? What is the real sustained virological response (SVR)?
In this review, suggestions on redefining VR and SVR were put forward, and the concept and potential significance of “para-functional cure” were elaborated.
Discovery and potential clinical significance of HBV RNA virion-like particle
More than 20 years ago, in addition to HBV DNA, HBV RNA was also found in the serum of chronic HBV-infected individuals [
13]. Several groups found that serum HBV RNA could be used as a new marker for monitoring the efficacies of NA and interferon therapy [
11,
14–
18], a predictor of early emergence of the YMDD mutant in patients treated with lamivudine [
19], and a useful marker for the safe discontinuation of NA treatments [
12,
20]. However, the nature of serum HBV RNA has not been sufficiently investigated until recently. First, serum HBV RNA was confirmed to be encapsidated and enveloped because it could be pulled down by hepatitis B core protein (HBcAg)-specific antibodies, and its level increased after removing the HBV envelope [
11]. Moreover, using Northern blot and 5′ rapid amplification of cDNA ends, we confirmed that serum HBV RNA was the HBV pregenomic RNA (pgRNA) and was present in the virion-like particles, which was validated by sucrose density gradient centrifugation and electron microscopy assays [
12]. Accordingly, serum HBV RNA was present in HBV RNA virion-like particles.
HBV RNA virion-like particle supplemented the HBV life cycle
With an enveloped 3.2-kb-long relaxed circular DNA (rcDNA) genome, HBV is traditionally classified as a DNA virus [
21]. As shown in Fig. 1, HBV virions enter hepatocytes through a high-affinity interaction between the myristoylated preS1 region of HBV and a functional receptor, sodium taurocholate co-transporting polypeptide [
22]. During entry, a low-affinity interaction between the major hydrophilic region of viral small S protein and heparan sulfate proteoglycan is also required for infectivity [
23,
24]. After entering into hepatocytes, rcDNA translocates into the nucleus where it is converted to cccDNA. The cccDNA persists as a minichromosome and serves as the template for the transcription of five viral mRNAs. Among them, the 3.5-kb-long pgRNA also serves as the template for the reverse transcription of HBV DNA minus strand. The binding of HBV DNA polymerase to pgRNA provides the packaging signal and subsequently initiates the assembly of viral capsid using core proteins. Next, the encapsidated pgRNA undergoes reverse transcription to produce the minus strand of viral DNA, and the plus strand is incompletely synthesized from the minus strand [
25]. The newly synthesized rcDNAs are enveloped by the viral surface proteins and released as Dane particles or re-enter the nucleus to replenish the cccDNA pool [
26,
27]. Except for rcDNA, the double-stranded linear DNA (DSL-DNA), a replicative intermediate, can also be generated when failing to translocate the RNA primer needed to prime the plus-stranded DNA synthesis [
28]. The DSL-DNA is often integrated into the host genome through recombination using host enzymes [
29]. Consequently, the integrated viral DNA fragments frequently end in the DR-1/DR-2 regions of the viral genome [
30,
31]. Theoretically, the 3.5-kb-long pgRNA can only be produced from the circular cccDNA. By contrast, because the open reading frame of the S gene with its regulatory elements remains intact in the integrated sequences, hepatitis B surface antigen (HBsAg) can also be produced from the integrated HBV DNA fragments [
24,
32,
33]. Therefore, HBsAg has two sources: cccDNA and integrated HBV DNA (Fig. 1).
As previous studies have already suggested that nucleocapsid maturation may not be dependent on DNA synthesis [
34–
37], the pgRNA containing nucleocapsids may also be enveloped and released via multivesicular bodies. The discovery of HBV RNA virion-like particle provides a supplementary to the traditional HBV replication cycle (Fig. 1). The infection of hepatitis D virus (HDV) is mediated by HBV envelope proteins [
38,
39]. Theoretically, after entry into hepatocytes, the encapsidated pgRNA might restart its reverse transcription to form rcDNA and cccDNA and subsequently establish HBV infection. The infection potential of HBV RNA virion-like particles needs experimental evidences.
Para-functional cure of CHB
In recent years, studies have revealed that the half-life of cccDNA is less than two months [
40,
41]. Obviously, the replenishment of the cccDNA pool through either an extracellular or intracellular (recycling) way, or both, is essential for persistent viral infection. Thus, it is reasonable to postulate that though they do not directly act on cccDNA, via blocking rcDNA formation, NA can efficiently inhibit the replenishment of cccDNA pool [
42,
43]. Theoretically, the eventual eradication of cccDNA would be expectable after prolonged consolidation NA therapy. Indeed, a number of clinical studies have verified the significant decrease of cccDNA level in CHB patients after receiving long-term NA therapies [
44–
46].
Given that complete decay of cccDNA means the elimination of viral infection, a status nearly close to the complete cure of CHB [
47], to avoid viral DNA rebound, guaranteeing the silence or elimination of intrahepatic cccDNA before suspending NA treatment is imperative. However, restricted by invasive liver biopsy, regular detection of intrahepatic cccDNA is not feasible in clinical practice [
48,
49]. Different from the frequent occurrence of off-therapy HBV DNA rebound and disease relapse when the consistent loss of serum HBV DNA is used as a pre-requirement for NA withdrawal, relapse is quite uncommon for patients who had already reached HBsAg loss with or without anti-HBs seroconversion. Therefore, “loss of serum HBsAg” has been recommended as an ideal endpoint of therapy. Unfortunately, with the currently available anti-HBV agents, the loss of serum HBsAg is difficult to achieve. A recent study reported that the rate and durability of HBsAg seroclearance induced by NA are comparable to those developed spontaneously [
50]. As a consequence, most patients would have to receive life-long NA therapy, leading to problems such as drug resistance, adverse events, adherence, and expenses.
“Loss of serum HBsAg” has been recommended to be an effective endpoint of antiviral treatment for CHB patients, in whom antiviral treatment could be safely discontinued. On the other hand, a part of NA-treated patients could safely discontinue NA therapy after consolidation therapy, even though their serum HBsAg remains positive [
51,
52]. A recent study has demonstrated that the correlation of serum HBsAg with cccDNA is not found in CHB patients who have received NA therapy [
53]. In this study, though intrahepatic cccDNAs were undetectable in 49% (21/43) of CHB patients, only one patient achieved serum HBsAg loss. Moreover, a majority of the biopsies were negative for viral core protein and cccDNA, which strongly indicated that intrahepatic cccDNA might be silenced or exhausted after receiving long-term NA therapy.
The following question comes into mind: where does HBsAg come from in those patients whose cccDNA reservoirs were either exhausted or transcriptionally silenced? Given the fact that the integrated HBV DNA fragments were frequently detected in HBV-infected hepatocytes [
30,
32,
54], the integrated HBV DNA could act as a cellular gene to produce viral protein HBsAg [
24,
33,
55,
56]. All these observations indicated that, after a prolonged consolidation NA therapy, even after cccDNA had been eliminated or epigenetically silenced, serum HBsAg could remain at low levels in a small but certain fraction of patients. This is because serum HBsAg might originate from integrated HBV DNA fragments. The term “para-functional cure” is suggested to describe such a clinical status of potential elimination or transcriptional silencing of cccDNA defined by persistent loss of serum HBV RNA [
57]. Different from the already used terms such as “functional cure” or “clinical cure” in which HBsAg disappeared, a patient achieving “para-functional cure” could be positive for low-level serum HBsAg. However, though the status of “para-functional cure” is theoretically close to “functional cure” and might greatly lower the risk of NA off-therapy disease relapse, it must be emphasized that this is not an ideal endpoint of treatment because patients who achieve “functional cure” characterized by serum HBsAg loss would have a favorable clinical course and very low risk for hepatocellular carcinoma [
58,
59].
Redefining the virological response of CHB patients receiving NA therapy
To assess the VR for NA therapy, monitoring the dynamic change of a patient’s serum HBV DNA levels during therapy is important. Nowadays, VR has been defined as undetectable serum HBV DNA during therapy, and achieving VR (serum HBV DNA below the lower limit of detection by a sensitive PCR-based assay) is one of the prerequisites for discontinuation of NA therapy in clinical practice guidelines [
7–
10]. However, off-therapy virological rebound and hepatitis relapse still occur at high frequency, even after prolonged therapy to consolidate the constant suppression of viral DNA replication. Even worse, disease progression occasionally occurs to patients whose serum HBV DNA remains undetectable. As mentioned above, the presence and transcription of cccDNA were not directly affected by NA. Thus, the detection of serum HBV DNA would not comprehensively reflect the activity of cccDNA in patients under NA therapy. Therefore, understanding why frequent off-therapy viral rebound and CHB disease relapse occur is not difficult when only serum HBV DNA loss is adopted as a prerequisite for discontinuation of NA therapy.
As shown in Fig. 1, HBV RNA virion-like particle is confirmed to be an additional by-product of HBV viral replication. Theoretically, the formation of rcDNA, but not the production of HBV RNA virion-like particle, is directly blocked by NA [
12,
20]. Therefore, under NA treatment, serum HBV RNA may better reflect the activity of intrahepatic cccDNA than serum HBV DNA [
20,
57,
60]. Based on the discoveries mentioned above, instead of solely measuring serum HBV DNA, simultaneously monitoring serum HBV DNA and HBV RNA for CHB patients under NA therapy may better reflect whether true VR is being achieved or not. Accordingly, VR should be redefined as persistent loss (below the lower limit of detection by a sensitive PCR-based assay) of HBV DNA and HBV RNA.
Suggestion on the safe discontinuation of NA treatment
According to current clinical practice with loss of serum HBsAg as treatment goal for CHB management, only 2%–13% of CHB patients can attain functional cure with the current antiviral therapy roadmap [
7–
10]. The finding means that most patients will have to receive life-long NA therapy, and the cost will be a considerable burden to patients and society. As serum HBV RNA is a potential indicator of cccDNA activity, the loss of serum HBV RNA might indicate exhausting or transcription silencing of cccDNA reservoirs [
57,
60]. Moreover, serum HBV RNA can be used as a potential marker for the safe discontinuation of NA treatments [
12,
20]. Therefore, as shown in Table 1, in addition to serum HBsAg level, serum HBV RNA should be monitored simultaneously during consolidation NA therapy. In HBeAg-positive CHB patients without liver cirrhosis, safe discontinuation of NA therapy might be recommended to those who have reached HBeAg seroconversion and have persistently undetectable serum HBV DNA and HBV RNA during consolidation therapy and persistently normal ALT levels. In HBeAg-negative CHB patients without liver cirrhosis, discontinuation of NA therapy might be done to those with persistently undetectable serum HBV DNA and HBV RNA and persistently normal ALT levels. However, serum HBsAg could be positive in HBeAg-positive and HBeAg-negative patients at low levels. As low HBsAg is correlated with low risk of off-therapy viral DNA rebound and disease relapse, and the chance of HBsAg loss would be increased in CHB patients with HBsAg<1500 IU/mL at the time of switching from NA to Peg-IFN-α therapy [
61,
62], HBsAg level less than 1500 IU/mL is suggested in Table 1.
Recently, HBcrAg and empty virions have also been reported to be potential surrogate markers used to reflect intrahepatic cccDNA activity [
63,
64]. How to integrate these new markers and HBsAg to monitor VR, predict disease progression, and guide safe discontinuation of NA treatment need to be studied in the future.
Conclusions
In this review, the life cycle of HBV replication has been supplemented with HBV RNA virion-like particles, and the definition of HBV DNA-based VR has been modified by combined monitoring of serum HBV DNA and HBV RNA. Based on the suggested novel concept of “para-functional cure,” “persistent loss of serum HBV RNA” has been put forward as an indicator for safe discontinuation of NA therapy to patients who finished consolidation NA therapy. However, a multi-centered and large-scale cohort study should be conducted to testify the feasibility of safe withdrawal among CHB patients receiving consolidation NA therapy in the future.
Higher Education Press and Springer-Verlag GmbH Germany