Gray zone of hepatitis B virus infection

Abstract

A substantial proportion of patients with chronic hepatitis B (CHB) do not fall into any of the defined phases and are considered to be in the “gray zone” or “indeterminate phase.” Most of the current clinical practice guidelines have no recommendations for antiviral treatment for them. However, the gray zone CHB patients with significant hepatitis B virus levels (>2000 IU/mL) and persistently normal alanine aminotransferase (ALT) levels have a significantly high risk of hepatic inflammation, fibrosis, and hepatocellular carcinoma. The molecular, clinical, and economic data that we have reviewed collectively in this article provide support for simplification of treatment initiation strategies that incorporate broader treatment of adult patients with CHB in the gray zone (hepatitis B virus [HBV] DNA ≥2000 IU/mL), regardless of ALT levels.

DEFINITION OF GRAY ZONE OR INDETERMINATE PHASE CHRONIC HEPATITIS B

Most of the current clinical practice guidelines define the natural history of chronic hepatitis B (CHB) into four phases [Figure 1]:

Figure 1.

Re-appraised natural course of CHB patients. CHB = chronic hepatitis B

• Immune-tolerant (or hepatitis B e antigen [HBeAg]-positive chronic hepatitis B virus [HBV] infection) phase

• Immune-active (or HBeAg-positive CHB) phase

• Inactive hepatitis (or immune-control or HBeAg-negative chronic HBV infection) phase

• Reactivation (or immune-escape or HBeAg-negative CHB) phase

These guidelines define immune-active and reactivation hepatitis phases by significant liver histologic disease through liver biopsy or elevations in serum alanine aminotransferase (ALT) levels, as well as the presence of significant HBV viremia (>2000 IU/mL with HBeAg-negative and >20,000 with HBeAg-positive).[1,2,3]

Liver biopsy is very rarely performed in practice because of the dynamic nature of liver histology and the risks and costs of repeat liver biopsy, especially in the regions of high CHB prevalence. Therefore, in the vast majority of cases, ALT levels are used to define the phase of CHB.

However, these definitions are only conceptual and are unclear and inconsistent across the guidelines. For example, the notion of immune-tolerant phase patients is that they are young, HBeAg positive, and have high serum HBV DNA levels and normal ALT levels. Nonetheless, there are no consistent criteria for their age and lower cutoff level of HBV DNA across the guidelines [Table 1]. The only consistency is that they have normal ALT levels and are not indicated for the antiviral treatment. However, the ambiguity in the definition of immune-tolerant phase has caused serious confusion and controversy in the treatment of the patients.

Table 1.

Definition of immune-tolerant phase CHB by guidelines

Guidelines	Age criteria	HBV DNA	ALT	Treatment
APASL 2016	None	>2×7 log10 IU/mL	Normal	No
EASL 2017	None	>7 log10 IU/mL	Normal	No (may be considered if age >30 years [B2])
AASLD 2018	None	>6 log10 IU/mL	Normal or mildly elevated	No

AASLD=American Association for the Study of Liver Diseases, ALT=alanine aminotransferase, APSL=Asia-Pacific Association for the Study of the Liver, EASL, European Association for the Study of the Liver, CHB=chronic hepatitis B, HBV=hepatitis B virus

In fact, in many cases, the changes in serum HBV DNA levels precede the elevation in ALT levels [Figure 1]. In HBeAg-positive CHB patients, the viral load at the initial phase of infection is above 1 × 9 log₁₀IU/mL in most cases, which decreases slowly before the increase in ALT levels. Among HBeAg-negative, inactive hepatitis phase patients, some show gradual increase in viral load followed by ALT elevation after several months or years later.

Therefore, a substantial proportion of patients with CHB do not fall into any of the defined phases and remain undefined. Patients are considered to be in the “gray zone” or “indeterminate phase” when they do not meet the standard criteria for any of the clinical phases. For example, middle-aged, HBeAg-positive or -negative CHB patients with serum HBV DNA between 5 and 6 log₁₀IU/mL do not fall into any of the CHB phases and are classified as gray zone if they have normal ALT levels. The HBeAg-negative patients with elevated ALT levels and very low HBV DNA levels (<2000 IU/mL) may also be regarded in the gray zone. But the elevation of ALT in such patients may be caused by factors other than HBV. Consequently, the patients in gray zone are not homogeneous, but heterogeneous [A and D in Figure 2]. Therefore, in this review, only the patients with significant HBV viremia (>2000 IU/mL) and persistently normal ALT levels will be focused [D in Figure 2].

Figure 2.

Conceptual definition of gray zone chronic hepatitis B

RISK OF HISTOLOGIC DISEASE AND HEPATOCELLULAR CARCINOMA IN GRAY ZONE CHB PATIENTS

Most practice guidelines first define the patients who definitely need antiviral treatment, that is, those in immune-active and reactivation hepatitis phases with ALT elevation. Antiviral treatment is generally not recommended in all other patients, including those in immune-tolerant, inactive hepatitis phase and gray zone CHB. For example, without ALT level elevation, HBeAg-negative patients with rising HBV DNA levels >100,000 IU/mL may not qualify for treatment.[1,2,3] Such recommendations are based on the notion that the risk of hepatocellular carcinoma (HCC) and liver disease progression is very low with normal ALT levels.[4,5]

However, studies have consistently shown that ALT is not a reliable indicator for histologic disease. Over 70% of gray zone CHB patients with moderate HBV DNA and normal ALT levels had significant histologic disease,[6] and moderate HBV DNA (5–7 log₁₀IU/mL) was a risk factor for significant inflammation in patients with CHB and normal ALT, in the absence of significant fibrosis.[7]

RISK OF HCC IN GRAY ZONE CHB

The community-based prospective REVEAL cohort study from Taiwan suggested that the risk of HCC increases with increasing levels of baseline serum HBV DNA levels up to 10⁶ copies/mL (about 5 log₁₀IU/mL), irrespective of serum ALT levels and HBeAg status, in CHB patients.[8] It is important to note that the REVEAL cohort primarily consisted of HBeAg-negative patients (85%) and the highest HBV DNA tier analyzed was 10⁶ copies/mL (approximately 5 log₁₀IU/mL). Therefore, the association between very high HBV DNA levels (especially >6 log₁₀IU/mL) and HCC risk has remained unclear. Accordingly, antiviral treatment of the patients with high HBV DNA and normal ALT levels (gray zone) is controversial.

Recently, we have analyzed the association between a broad range of serum HBV DNA levels and long-term HCC risk in a total of 6949 HBeAg-positive and HBeAg-negative, noncirrhotic, treatment-naïve CHB patients, who are not generally indicated for antiviral therapy by current practice guidelines because of no significant ALT level elevation.[9] We found that the association between HBV DNA levels and HCC risk is not linear, but parabolic in these patients [Figure 3]. The HCC risk was highest at moderate HBV DNA levels (6.3 log₁₀IU/mL), while decreasing at higher and lower HBV DNA levels. Very high HBV DNA levels (>8 log₁₀IU/mL) showed the lowest HCC risk, which was not significantly different from that of very low HBV DNA levels (≤4 log₁₀IU/mL). Specifically, the adjusted HCC risk in patients with baseline HBV DNA levels of 6–7 log₁₀IU/mL was about 5 times higher than the risk observed in those with viral load >8 log₁₀IU/mL. This association was consistent across all age groups, and neither HBeAg status nor ALT levels were predictive of HCC.

Figure 3.

Nonlinear parabolic association between baseline serum HBV DNA levels and HCC risk during long-term follow-up in CHB patients. CHB = chronic hepatitis B, HBV = hepatitis B virus, HCC = hepatocellular carcinoma

ASSOCIATION BETWEEN BASELINE HBV VIRAL LOAD AND ON-TREATMENT HCC RISK

Following the initiation of antiviral treatment, serum HBV DNA levels rapidly become undetectable in patients with CHB. As a result, baseline HBV viral load has not been considered a significant predictor of on-treatment HCC risk.[10,11,12] Nevertheless, our recent findings revealed that pre-treatment baseline HBV DNA level is a significant and independent factor, that is inversely associated with HCC risk during treatment with entecavir (ETV) or tenofovir disoproxil fumarate (TDF) in 2073 HBeAg-positive, noncirrhotic CHB patients.[13] On-treatment HCC risk increased incrementally with decreasing baseline HBV DNA levels in the range of ≥5.00 log₁₀IU/mL in those patients. Compared to baseline HBV DNA ≥8.0 log₁₀IU/mL, the adjusted hazard ratios (HRs) for HCC risk for baseline HBV DNA 7.00–7.99 log₁₀IU/mL, 6.00–6.99 log₁₀IU/mL, and 5.00–5.99 log₁₀IU/mL were 2.48 (P = 0.03), 3.69 (P = 0.002), and 6.10 (P < 0.001), respectively. These data indicate that early initiation of antiviral treatment with a high viral load (≥8.00 log₁₀IU/mL), before HCC risk becomes irreversible, may maintain the lowest risk of HCC in HBeAg-positive patients.

REDUCTION IN HCC RISK BY ANTIVIRAL TREATMENT IN PATIENTS WITH GRAY ZONE CHB

Our recent study demonstrated that compared to untreated patients with the same ranges of baseline HBV DNA levels, antiviral treatment significantly reduced HCC risk in patients with moderate viral load (5.00–7.99 log₁₀IU/mL), but HCC risk did not decrease to the level of patients who initiated antiviral treatment with high viral load (≥8.0 log₁₀IU/mL) [Figures 4 and 5].[13]

Figure 4.

Association between baseline HBV DNA levels and on-treatment HCC risk in HBeAg-positive, noncirrhotic, adult patients with CHB[13]. All panels in this figure were made by Lim based on data in Kim et al.[9] and Choi et al[13]. CHB = chronic hepatitis B, HBeAg = hepatitis B e antigen, HBV = hepatitis B virus, HCC = hepatocellular carcinoma

Figure 5.

Reduction in HCC risk by antiviral treatment according to the baseline HBV viral loads in HBeAg-positive, noncirrhotic, adult patients with CHB[13]. CHB = chronic hepatitis B, HBeAg = hepatitis B e antigen, HBV = hepatitis B virus, HCC = hepatocellular carcinoma

Compared to untreated HBeAg-positive, noncirrhotic adult CHB patients with normal ALT levels (n = 2643), nucleos(t)ide analog (NA) treatment in HBeAg-positive, noncirrhotic, adult CHB patients (n = 2073) reduced HCC incidence in patients with moderate baseline viral load (5.00–7.99 log₁₀IU/mL), but HCC risk did not decrease to the same extent in patients with high baseline viral load (≥8.0 log₁₀IU/mL).

MECHANISM OF HEPATOCARCINOGENESIS IN GRAY ZONE CHB

Although the exact oncogenic mechanism of HBV has not been completely determined, there are several mechanisms that may account for the nonlinear parabolic association between baseline HBV DNA levels and HCC risk in untreated and treated CHB patients. Firstly, most HBeAg-positive CHB patients exhibit very high HBV DNA levels (≥8 log₁₀IU/mL) during the initial immune-tolerant phase of infection, as fully infected hepatocytes can generate 10⁹–10¹⁰ viruses per milliliter of serum.[14] If the infection is benign in an immune-tolerant phase host, it would be expected to persist throughout the course of the disease. Therefore, the risk of HCC may be relatively low in HBeAg-positive CHB patients who have very high HBV DNA levels (≥8.0 log₁₀IU/mL). In contrast, low-level cytotoxic T lymphocytes induce a low yet persistent immune-mediated elimination of HBV-infected hepatocytes.[15,16] This process can result in the clonal emergence of hepatocytes, a major risk factor for HCC, that produce less virus, a gradual decrease in serum HBV DNA levels, and exhibits resistance to HBV infection.[14,17,18] Consequently, a decrease in HBV DNA titers (<8 log₁₀IU/mL) may indicate the expansion of clonal hepatocytes resistant to HBV entry and a subsequent increase in HCC risk.[19,20,21,22] Accordingly, a decrease in viral load in CHB patients may reflect progressive liver damage and increased HCC risk. Correspondingly, a serum HBV viral load of 5.00–7.00 log₁₀IU/mL was found to be a risk factor for significant hepatic inflammation in patients with CHB, despite normal ALT levels and the absence of significant fibrosis. Secondly, during the earliest phase of CHB, HBV viral DNA integrates into the host genome, which can result in chromosomal damage and instability, the functional loss of tumor suppressor genes, and/or the activation of tumor-promoting genes associated with HCC.[19,20] Moreover, a recent study demonstrated that high viremia levels (>20,000 IU/mL) correlated with increased frequencies of HBV genome integration into the host chromosome in HBeAg-negative patients.[23] Lastly, in CHB patients with normal ALT levels and no significant fibrosis, moderate serum HBV DNA levels (5–7 log₁₀IU/mL) have been identified as a risk factor for hepatic inflammation,[6,7] which is itself a risk factor for HCC.

These potential mechanisms of hepatocarcinogenesis highlight the benefit of starting antiviral treatment as early as possible, even with persistently normal ALT levels, before cellular inflammation and HBV DNA integrations result in an irreversible accumulation of HCC risk. Indeed, antiviral treatment has been shown to reduce both HBV DNA integration and hepatocyte clonal expansion.[24,25]

IMPACT OF ANTIVIRAL TREATMENT ON HBV DNA INTEGRATION AND CLONAL HEPATOCYTE EXPANSION

The intermediary form of HBV before integration into the host DNA contains double-stranded linear DNA (dslDNA). As dslDNA is formed via reverse transcription of HBV RNA, NAs should reduce the formation of dslDNA and DNA integration into the host genome. While data on antiviral treatment and HBV DNA integration are limited, some evidence suggests NA therapy reduces HBV DNA integration. Analysis of treated and untreated liver biopsies has demonstrated that treatment is associated with reduction in viral load, integrations, and chromosomal translocations.[24,25,26]

IMPACT AND COST-EFFECTIVENESS OF TREATING GRAY ZONE CHB

The current first-line NAs, ETV, TDF, and tenofovir alafenamide (TAF), have favorable and well-described long-term safety profiles, with minimal to no resistance, even in heavily treatment-experienced patients with pre-existing resistance,[27,28,29,30,31] and are associated with high adherence rates.[32,33] While long-term treatment does not eliminate HCC risk, it can reduce liver disease progression, improve necroinflammation and fibrosis, and even reverse cirrhosis.

However, CHB antiviral treatment indications and guidelines are largely based on evidence from randomized controlled trials (RCTs) that were not designed to demonstrate the impact of these treatments on long-term outcomes, such as HCC, in a real-world setting. As such, they may fail to address important groups of patients and result in withholding of therapy in patients who may benefit. Moreover, the current concept of CHB disease phases and corresponding treatment guidelines based on HBeAg, HBV DNA, and ALT are complex and do not represent the whole spectrum of CHB patients observed in clinical practice. Consequently, many patients currently ineligible for NAs, according to current guidelines, remain at risk of adverse clinical events, particularly HCC. An analysis of treatment-naïve CHB patients showed that among those who developed HCC, 64%, 46%, and 34% did so outside of treatment guideline recommendations from the Asian Pacific Association for the Study of the Liver (APASL), the American Association for the Study of Liver Diseases (AASLD), and the European Association for the Study of the Liver (EASL), respectively.[34]

Evidence for the treatment of “gray zone” patients is provided by an analysis from the TORCH-B study, a randomized, double-blind, placebo-controlled study examining the treatment of patients with HBV DNA > 2000 IU/mL, ALT 40–80 U/L, and no cirrhosis (79% HBeAg negative).[26] During 3 years of follow-up, the placebo group showed a significantly higher proportion of progression in fibrosis stage compared to the TDF group (47% vs. 26%, P = 0.013).

Several analyses have estimated that the costs associated with expanded CHB therapy may be offset by reduced expenditure needed for future consequences of disease progression from untreated CHB. Our study reported that starting treatment in the immune-tolerant phase was more cost effective than delaying until the immune-active phase, especially with increasing HCC risk, decreasing drug costs, and considering productivity loss.[35] Another study of ours compared current guidelines with an extended indication (all patients with HBV DNA ≥ 2000 IU/mL and any ALT). The base case was maintaining current eligibility requirements and treatment levels through 2035. Another scenario was treating all non-cirrhotic individuals with a viral load above 2000 IU/mL, regardless of the HBeAg status and ALT levels, as well as all cirrhotic patients regardless of the viral load. The last scenario had the greatest impact and was estimated to avert 9800 cases of decompensated cirrhosis and 43,300 cases of HCC and save 37,000 lives until 2035 in Korea. This approach was estimated to be highly cost effective.

CONCLUSIONS

Current international CHB guidelines are complicated and only recommend antiviral treatment in subsets of patients with elevated ALT levels as well as significant serum HBV DNA levels.[1,2,3,36] The strict and complex nature of these guidelines may contribute to the mere 2.2% of global CHB patients who received antiviral treatment in 2019.[37] This is in contrast to hepatitis C or human immunodeficiency virus (HIV), where recent guidelines have promoted early treatment of all infected patients regardless of end-organ damage. However, evidence from multiple studies, as discussed in this review, demonstrates that patients currently ineligible for antiviral treatment may have liver damage and remain at risk for disease progression and HCC. Therefore, it is time to consider adoption of expanded treatment strategies to reduce these risks.

The data that we have reviewed collectively provide support for simplification of treatment initiation strategies that incorporate broader treatment of adult patients with HBV DNA ≥2000 IU/mL, regardless of ALT levels. Furthermore, a strategy to treat all noncirrhotic patients with HBV DNA ≥2000 IU/mL, regardless of ALT levels, would first define the patients who may remain untreated with minimal long-term risk of disease progression and HCC, namely, true “inactive carriers.” Compared to traditional guideline recommendations, this strategy would incorporate treatment of (1) all cirrhotic patients, (2) gray zone patients with viremia and normal ALT, and (3) immune-tolerant phase patients. The benefits of an expanded approach outweigh the costs and any risks associated with early treatment.

The potential safety implications of expanding initiation of long-term antiviral treatment should, of course, be considered. However, accumulating clinical experience suggests a minimal risk of side effects associated with current antiviral treatment options, and this must, therefore, be balanced against the risk of disease progression and HCC in untreated CHB patients. Finally, and importantly, the current research to achieve a functional cure for CHB makes us optimistic that any concerns related to life-long NA therapy are likely to be time limited.

In conclusion, the available molecular, clinical, and economic data provide a strong rationale for earlier initiation of antiviral treatment in CHB patients to reduce the risk of disease progression and HCC. Adoption of such a simplified “treat to prevent” approach could save countless lives in a cost-effective manner.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.