Unrevealing the natural course of the so-called “inactive HBsAg or HBV carrier state”

When hepatitis B surface antigen (HBsAg), originally named “Australia antigen,” was discovered in the 1960s [1, 2], its link with viral hepatitis could not be recognized for some time, since most individuals carrying this antigen had little if any evidence of underlying hepatitis. Then, following the identification of its association with viral hepatitis in 1967 and 1968 [3, 4], the glossary was modified from Australia antigen to Hepatitis Associated, Serum Hepatitis, Hepatitis B antigen, and finally to HBsAg. However, regardless of the terms used, all evidence gathered over the years has clearly supported the view that many HBsAg-positive individuals harbor normal or nearly normal liver enzymes (alanine/aspartate aminotrasferase, ALT/AST) and have little if any liver damage [5–8]. Therefore the terms “healthy,” “asymptomatic,” and “inactive” carriers of HBsAg, or of hepatitis B virus (HBV) have been introduced, to distinguish such individuals from patients with chronic viral hepatitis B. In support of this concept were also the findings of serological testing for hepatitis B e antigen (HBeAg) and its corresponding antibody (anti-HBe) by sensitive radioimmunossay (RIA) methods [9] and the immunohistochemical observations on the expression of HBV proteins in the liver. Thus, serum HBeAg and liver hepatitis B core antigen (HBcAg), two markers of HBV replication, were reported to be negative in these carriers [10] while liver histology was essentially normal but with numerous “ground-glass” hepatocytes expressing abundant hepatitis B surface proteins [11–13] from viral sequences integrated into their genome [12]; however, HBcAg expression was clearly negative [14, 15]. The wide acceptance of the view that HBsAg-positive persons who are HBeAg-negative and anti-HBe positive have inactive HBV infection without associated liver disease dominated worldwide for many years and subsequently led to doubts on whether HBeAg-negative chronic hepatitis B (CHB) described from the early 1980s [16] was really existent or not and if it could actually be attributed to HBV replication. Moreover, in clinical practice HBsAg-positive individuals, labeled as healthy or inactive “carriers,” attracted little interest and only few investigators worldwide took their time to follow-up such carriers prospectively. However, with the introduction in clinical research of molecular hybridization and polymerase chain reaction (PCR) techniques, HBV cloning and sequencing, and the discovery of replication-competent HBeAg-negative HBV mutants [17], it was soon realized that HBeAg-negativity/anti-HBe-positivity with normal ALT levels is neither synonymous with absence of HBV replication and of HBV-induced liver damage nor it represents a permanent, life-lasting condition. Thus the term “inactive HBsAg carrier state” rather than “inactive HBV carrier” has been widely applied and is still recommended in recent hepatitis B guidelines [18–20].

Currently, it is generally perceived that the inactive HBsAg carrier state represents the third phase in the natural course of chronic HBV infection. It follows a previous, immunologically active, phase of HBeAg-positivity with replicating HBV and increased ALT activity, during which HBeAg clears and anti-HBe develops [21, 22]. This phase may remain stably inactive and even finish in resolution of HBV infection with clearance of HBsAg and development of anti-HBs. On the other hand, HBV replication can also be resumed either by reversion to the previous HBeAg positive phase, or much more frequently, by progression to a fourth phase of HBV reactivation while HBeAg-negativity with anti-HBe-positivity is maintained. During this phase of reactivation of HBV infection, known as HBeAg-negative CHB, HBV replication induces liver damage either persistently or intermittently [16]. Therefore, meaningful and clinically relevant criteria are necessary for the diagnosis of the “inactive HBsAg carrier state,” its differentiation from HBeAg-negative CHB and for understanding its short- and long-term course and prognosis.

Currently accepted criteria [19, 20] for the diagnosis of the inactive HBsAg-carrier state may appear to several clinicians as rather stringent, at least compared to those adopted in the past. In addition to the prerequisite for persistent HBsAg-positivity with HBeAg-negativity/anti-HBe-positivity, they include ALT values persistently <30 IU/l for men and <19 IU/l for women (≤X 0.75 upper limit of normal values/ULN) and serum HBV DNA levels <10,000 copies or <2,000 IU/ml. Even so it is further advocated that such HBsAg carriers are tested for ALT every 3 months during the first year to verify that they are truly in the inactive carrier state and then every 6–12 months. More frequent HBV DNA testing is also recommended if ALT/AST increase above the ULN [19, 20]

In most of hitherto published studies on the natural course of the inactive HBsAg carrier state, the diagnostic criteria used have been rather loose both in terms of ALT activity and HBV DNA levels, requiring values <X 1.25 ULN and of <100,000 copies/ml, respectively, or non-detectability of HBV DNA by molecular hybridization assays. In these studies loss of the state of inactivity of HBV infection and transition to CHB has been found to vary greatly ranging from 10% to >30%, with annual frequencies ranging from 0.9 % to 2% [5, 7, 8, 23–29] and with rates of development of cirrhosis and HCC also greatly varying being highest in Asia and lowest in Europe, Canada, and Alaska [8, 18, 30–33].

Differences in definitions and inclusion criteria as well as in the duration of follow-up, geography, HBV genotypes, modes, and age at acquisition of HBV infection, gender, and age at entry in the various follow-up studies appear to account, to a large extent, for these significantly differing and even contrasting findings. Therefore, in order to reach some valid conclusions of general acceptance, it is of prime importance to recognize differences and similarities between the various cohorts of persons included in the reported studies. In this context the study of Chu and Liaw [34] published in this issue of Hepatology International represents an important contribution towards unrevealing some of the possible variables that may determine significant variability in the long-term natural course of the inactive HBsAg carrier state and not only from country to country and from population to population but also in the same population and country.

A good number of HBsAg carriers positive for anti-HBe, with normal ALT values <30 U/l persisting for more than 2 years was entered in the study and followed up for 25 years (1980–2005). The overall cumulative relapse rate of this cohort of persons during the long period of follow-up was found to be around 20% with more than 85% of the relapses occurring in all age groups within the first 10 years of the study. This is in keeping with the findings of a previous study conducted by the same research group in the general population of Taiwan, showing that, following spontaneous HBeAg seroconversion, the cumulative probability of relapse with development of HBeAg-negative CHB is highest during the first years, reaching levels of 14, 18, and 22% at years 3, 5, and 10, respectively [26]. Clearly these findings in an area of high HBV endemicity, where the vast majority of chronic HBV infections occur in neonates, are quite different from those reported in several studies of inactive HBsAg carriers identified among blood donors in geographical areas of intermediate or low HBV endemicity, where horizontal transmission among children predominates (7;8;20;28) and HBeAg seroconversion seems to occur earlier than in the East [20].

In the study of Chu and Liaw [34], interesting and challenging differences in the yearly relapse rates and the cumulative relapse rates by the age of patients at entry were also observed. Thus, in the youngest age group of <30 years, the overall annual relapse rate was 0.89% and the 10 year cumulative relapse was 10% compared to 2.14 and 23%, respectively in the age group of 40–49 years (p < 0.001). Since their age at HBeAg seroconversion was unknown, it is difficult to explain why in all age groups despite their significantly differing annual relapse rates the great majority of relapses similarly occurred during the first years of follow-up; unless it is accepted that HBeAg seroconversion had occurred much later in the older than in the younger age groups and that, as the authors suggest, HBeAg seroconversion at younger age carries a lower risk for spontaneous reactivation. However, such and other possibilities and the documentation of other predictors and determinants of reactivation as mode and age at HBV transmission, HBV genotypes, viraemia levels at entry, and during the study and ranking of ALT levels with the “normal” range, must be properly evaluated in future prospective studies.

Finally in agreement with previous studies from different geographical areas of the world, a clear-cut gender effect on the frequency of HBV reactivation was again observed by Chu and Liaw [34], with males being 2.4 times more likely to develop spontaneous relapses compared to females, thus indicating the universality and strength of this biological factor.

Despite several rising questions and of the actual difficulties encountered in the design and execution of very long-term follow-up studies of cohorts of persons in the inactive HBsAg carrier state, a number of clear-cut messages and conclusions can be derived from the present and many previous studies.

First, for the diagnosis of the inactive carrier state repeated assays of ALT activity and HBV DNA levels over an initial period of 1 to preferably 2 years with stringent criteria of biochemical inactivity are required to exclude cases of HBeAg-negative CHB in transient remission. For gray-zone individuals in terms of ALT activity and with HBV DNA levels >2,000 IU/ml, the liver histology may give the answer.

For persons in the inactive HBsAg carrier state, defined on the basis of stringent criteria and repeated ALT and PCR HBV DNA assays, the risk for relapse and development of CHB is clearly lower compared to the respective risk of the inactive HBsAg carrier state diagnosed on the looser criteria of inactivity of past years. Nevertheless, the exact size of this risk over lengthy periods of observation, including the risk for development of progressive liver disease and HCC, still remains unknown. However, in designing future follow-up cohort studies it should always be remembered that loss during follow-up is a real problem that may be quite high particularly if the follow-period is very long and that this loss rarely occurs at random.

For the inactive HBsAg carrier state diagnosed on the basis mainly of ALT values considered as “normal” or nearly normal when lower than x1.25-2ULN and on negativity of serum HBV DNA tested by molecular hybridization assays or of <100,000 copies/ml, there is a considerable but variable risk for subsequent relapse and development of HBeAg-negative CHB and its sequellae. The risk is higher in vertically than horizontally transmitted HBV infection, in males than in females, in older than younger ages, is probably higher in HBV genotypes B, C, D compared to A, and it also appears to be influenced by co-morbid conditions and several other host and environmental factors.

Finally, the large series of cohorts of inactive Chinese HBsAg/anti-HBe carriers followed up for lengthy periods of time, clearly show that the so-called inactive carrier state cannot be generally viewed as an innocent long-lasting condition of good prognosis and that regular follow-up with appropriate clinical, biochemical and virological monitoring is important with the aim of prompt diagnosis of virological relapses and early initiation of antiviral treatment.