Use of quantitative hepatitis B surface antigen with hepatitis B virus DNA in clinical practice†
Potential conflict of interest: Grace L.-H. Wong has served as an advisory committee member for Otsuka, and a speaker for Echosens. Henry L.-Y. Chan is a consultant for Abbott, Bristol-Myer Squibb, Gilead, Merck, Novartis, and Roche; has received lecture honoraria from Abbott, Bristol-Myers Squibb, Echosens, Gilead, Glaxo-Smith-Kline, Merck, Novartis, and Roche; and has received an unrestricted grant from Roche for hepatitis research.
Hepatitis B virus (HBV) DNA is the key prognostic factor determining both disease activity and risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B viral infection. It is also the key viral factor in deciding when to start antiviral therapy and to monitor the response to therapy. Quantitative hepatitis B surface antigen (HBsAg) reflects the amount and the transcriptional activity of covalently closed circular DNA inside hepatocytes.1 Therefore, quantitative HBsAg provides information concerning disease activity over and above an estimation of viral replication.
The enzyme immunoassays of HBsAg that are in widespread use do not quantify total circulating protein, nor do they distinguish between the different HBsAg proteins. HBsAg quantification detects all three forms of circulating HBsAg, namely the virion-associated HBsAg, subviral particles and HBsAg produced from integrated sequence. The two commercially available HBsAg quantification assays, the Architect QT assay (Abbott Laboratories, Wiesbaden, Germany) and the Elecsys HBsAg II Quant assay (Roche Diagnostic, Indianapolis, IN), give a measure of total HBsAg and have good correlation with each other.2 The clinical use of measuring quantitative HBsAg can be summarized below.
Diagnosis of the True Inactive HBV Carrier
Patients in the inactive HBV carrier state do not need frequent follow-up or medical treatment. Therefore, the recognition of the inactive HBV carrier state is important. Based on natural history cohorts with long-term follow-up, the inactive HBV carrier state is often defined by persistently normal alanine aminotransferase levels,a low HBV DNA level (<2,000 IU/mL) in a hepatitis B e antigen (HBeAg)-negative patient with no or minimal liver injury.3 Recent data show that HBeAg-negative patients with HBV DNA between 2,000 and 20,000 IU/mL with persistently normal alanine aminotransferase levels can also have a very good prognosis without the need of antiviral therapy.4 However, HBV DNA may fluctuate with time; some patients who have low HBV DNA levels at one time may have viral and biochemical reactivation at a later time. The serum HBsAg level, which tends to change very slowly with time and remain at a low level among inactive carriers, is a useful adjunct to HBV DNA, thereby assisting the identification of true inactive HBV carriers.5 Numerous studies have shown that inactive HBV carriers usually have serum HBsAg levels <1,000 IU/mL.1 In European HBeAg-negative patients with HBV DNA levels <2,000 IU/mL, HBsAg levels <1,000 IU/mL can predict inactive disease in the subsequent year of follow-up.6 In Asian patients, HBsAg <100 IU/mL has been shown to increase the chance of subsequent spontaneous seroclearance of HBsAg.7
While we would expect that the true inactive HBV carrier state would also indicate a low HCC risk, this remains to be confirmed. Based on long-term follow-up studies in Taiwan, HBV DNA <2,000 IU/mL is associated with a lower HCC risk than higher HBV DNA levels. However, HBV DNA levels <2,000 IU/mL did not correlate with the risk of HCC risk.8 The exact reason is uncertain but may be related to fluctuating HBV DNA during follow-up. On the other hand, among patients with HBV DNA <2,000 IU/mL, those who have HBsAg <1,000 IU/mL have a significantly lower HCC risk (<100 per 100,000 person-years) than those with HBsAg >1,000 IU/mL. Hence the definition of a true inactive carrier probably needs a low HBV DNA (<2,000 IU/mL) plus a low HBsAg level (<1,000 IU/mL) (Table 1).
Table 1. HBV DNA and HBsAg Levels in Different Settings1
|Natural history|| || || |
| Inactive hepatitis||<2,000 IU/mL||<1,000 IU/mL||6|
| Low risk of HCC||<2,000 IU/mL||<1,000 IU/mL||8|
|Treatment response to PEG-IFN|| || || |
| HBeAg-positive|| || || |
| Good||<20,000 IU/mL at week 24||<1,500 IU/mL at week 24||9|
| Poor||High, no clear cutoff||>20,000 IU/mL at week 24||10, 11|
| ||High, no clear cutoff||No decline at week 12 or 24||12|
| HBeAg-negative|| || || |
| Good||Unclear||>1 log decline at week 48||13|
| ||Unclear||>10% decline at week 24||14|
| Poor||≤2 log decline at week 12||No decline at week 12||15|
Treatment Response to Pegylated Interferon
Pegylated interferon (PEG-IFN) is a recommended first-line therapy for chronic hepatitis B. However, the use of PEG-IFN is limited by its relatively low efficacy (∼30%-40% sustained response), side effects, and high cost. It is therefore desirable to predict which patients will fail therapy so that PEG-IFN can be stopped. Several studies have evaluated the use of serum HBV DNA to predict response to PEG-IFN. In general, a high serum HBV DNA during treatment predicts nonresponse, although the exact cutoff HBV DNA level and the timing of evaluation to be used as a stopping rule are controversial.17 In contrast, the absolute levels as well as the kinetics of serum HBsAg are predictive of treatment response to PEG-IFN.
Based on the phase 3 studies of peginterferon alfa-2a (with or without lamivudine in combination) in HBeAg-positive patients, HBsAg >20,000 IU/mL at week 12 or 24 predicts treatment failure.10, 11, 16 Among patients on PEG-IFN monotherapy, HBsAg >20,000 IU/mL together with HBV DNA >100,000 copies/mL (∼20,000 IU/mL) can further increase the negative predictive value for HBeAg seroconversion (i.e., failure to seroconvert) to 91% and predicts failure to achieve a combined response (HBeAg seroconversion and HBV DNA <10,000 copies/mL) in 98% of patients.3 A more vigorous early decline in HBsAg level starting at week 4 of peginterferon alfa-2b treatment was observed in sustained responders (those with HBeAg loss and HBV DNA <2,000 IU/mL at 6 months posttreatment), whereas absence of such an HBsAg decline at weeks 12 and 24 had negative predictive values of 97% and 92%, respectively.4
When PEG-IFN is administered to HBeAg-negative patients, the most discriminatory and best validated stopping rule has been described in genotype D HBV-infected patients who fail to achieve both a decline in serum HBsAg and a <2 log reduction in HBV DNA at week 12, but more studies are required before this rule can be generalized to patients infected by other HBV genotypes.5 More generally, an HBsAg decline (either by 10% or by l log) at week 12, week 24 or the end of PEG-IFN therapy was also found predictive of sustained response (HBV DNA <2,000 IU/mL or HBsAg loss) in HBeAg-negative patients.13, 14 HBsAg <10 IU/mL at the end of 48-week treatment can predict 52% HBsAg loss 3 years posttreatment.6
Response-Guided Therapy to PEG-IFN
In the future, determining when to stop PEG-IFN might be guided by the on-treatment responses of HBsAg and HBV DNA.17 As more data on extended administration of PEG-IFN to patients with HBV infection emerge,18 the use of on-treatment measurement of quantitative HBsAg as an aid to select patients for extension of PEG-IFN beyond 1 year should be examined. Similarly, more data are needed to determine how to fit serial measures of HBV DNA into the algorithm.
Quantitative measurement of HBsAg is an indispensable tool in the clinical assessment and management of chronic hepatitis B viral infection. Quantitative serum HBsAg should be used together with HBV DNA in the diagnosis of the true inactive carrier state and in monitoring the clinical response to PEG-IFN (Table 1).