Chronic hepatitis B infection has a very complex natural history.1 Long-term follow-up studies show that high viral load is the major driving force for disease progression. The development of cirrhosis, hepatocellular carcinoma and liver-related mortality are positively correlated with serum viral load. Thus, the highest accumulated incidence of unfavorable outcomes occurs in those with high viral load at baseline and remaining so during the observation period.2,3 Therefore, persistent and profound suppression of viral replication is the cornerstone to prevent or slow-down disease progression in chronic hepatitis B infection and thereby to improve survival and quality of life.
For chronic hepatitis B patients without cirrhosis, many controlled clinical trials have shown that both interferons and nucleos(t)ide analogs effectively suppress serum hepatitis B virus (HBV) DNA levels, normalize liver function tests, and most importantly, improve liver histology.4,5 For those cases in which cirrhosis has already developed, especially those with hepatic decompensation, treatment with oral nucleos(t)ide analogs (but not interferon) is feasible for long-term therapy. A large, Asia-Pacific multicentre randomized controlled trial has shown that lamivudine treatment can improve the clinical outcomes in compensated cirrhotic patients, in term of reducing the development of decompensation and HCC.6 Other observations have also demonstrated that lamivudine therapy also improves hepatic functional reserve in patients with decompensated cirrhosis.7 Thus, there is considerable evidence to support the recommendation to mount antiviral therapy for HBV- related liver disease.8,9
In this issue of the Journal, Nishida and co-workers10 report their cohort study on the efficacy and drug-resistance profile of long-term lamivudine treatment in 158 Japanese patients with different stages of liver disease. They recruited 87 with uncomplicated chronic hepatitis, 28 with compensated cirrhosis, and 43 with decompensated cirrhosis. All subjects had serum HBV DNA >5 log10 copies/mL (>2 × 104 IU/mL) and alanine aminotransferase (ALT) higher than two times of the upper normal limit, or complications of hepatic insufficiency. The authors found that lamivudine reduced HBV DNA and ALT equally in all groups. Further, improvement of Child-Pugh score and its component such as serum albumin, bilirubin, prothrombin time, and ascites, were most remarkable in the group with decompensated cirrhosis.
The ‘non-randomized’ and ‘non-controlled’ design of the present study make it less favorable in term of evidence grading, but ethical concerns over long-term exposure to placebo in patients with severe chronic hepatitis B now make randomization to placebo inappropriate. Furthermore, this kind of ‘real world’ observational data still have a role, because they are closer to the scenario of day-to-day clinical practice. It is therefore important to note that the results have reasonable agreement with those of randomized controlled trials.11
While the finding that antiviral efficacy and resistant profiles of lamivudine are similar among three groups are of clinical relevance, the real meaning of the claim that ‘lamivudine treatment was most beneficial and significant in the patients with decompensated cirrhosis in terms of improvement of hepatic functional reserve’ should be explained carefully. On one hand, the liver function reserve in patients with only chronic hepatitis or compensated cirrhosis is normal or nearly normal; there is no room for further improvement. On another hand, in some decompensated cirrhotic patients with very poor liver function reserve, it is too late to get improvement. Indeed, other studies have shown that, under lamivudine therapy, fewer patients with better liver functional reserve at baseline develop unfavorable clinical outcomes than those with poor functional reserve (some of the latter fail to recover).12,13 So the timing to start antiviral therapy is not simply a matter of ‘early’ or ‘late’ and still remains a matter for further investigation, particularly among younger subjects with uncomplicated or minimal chronic hepatitis B.
Drug resistance is the major scientific obstacle for long-term nucleos(t)ide therapy. Not surprisingly in the present report, 4 years of lamivudine therapy carried a high rate of viral breakthrough, 16% at 1 year, 53% at 4 years. These rates are also in agreement with randomized controlled trials.5 Importantly, the cumulative incidence of virologic breakthrough were similar among the three groups of chronic hepatitis B liver disease. Hepatic failure developed or deteriorated in 10 patients after virologic breakthrough, and not so surprisingly nine of them had had decompensated cirrhosis at baseline. Just like in this report, genotypic analysis is not practical in daily clinical care, so close monitoring HBV DNA and ALT is of paramount importance, especially in those patients who have already developed cirrhosis and decompensation. It is a pity that this study does not report the result of adefovir dipivoxil add-on therapy in those patients who developed viral breakthrough, as this is now the standard of care in this situation.9
Importantly, the authors noticed that the strongest predictive factor for lamivudine-resistance was persistent HBV DNA at month 3. This is another piece of evidence supporting the roadmap concept. According to the report of an international workshop,14 12 and 24 weeks after starting the therapy are 2 checkpoints, at which times one can decide to continue, further monitor or modify therapy, according to complete (HBV DNA become undetectable by polymerase chain reaction (PCR)-based assay), partial (less than 2000 IU/mL but still detectable by PCR assay) or inadequate virologic response (still more than 2000 IU/mL) response, respectively. Thus clinical decisions on continuing versus modifying therapy can be made as early as week 24, allowing clinicians to tailor therapy to the needs of individual patients. More recently, even 4-week HBV DNA response has been shown to correlate near perfectly with the risk of drug resistance during 5 years of lamivudine therapy:15 the real sense of the roadmap concept is to modify or tailor therapy to optimize the efficacy and minimize or counter resistance, before the unfavorable outcomes occur. Prospective data to support this conceptual approach are still lacking, but can come from carefully conducted prospective observational studies, of the type reported by Nishida and colleagues, provided appropriate viral monitoring data are obtained.
Another important issue for long-term therapy is selection of antiviral medication. According to the newest guideline,9 entecavir or adefovir dipivoxil are preferred especially for those with cirrhosis; most such cases need virtually life-long therapy. However, to date, published long-term data mainly involve lamivudine as it was the first approved oral anti-HBV agent. As more potent and less resistant nucleos(t)ide analogs (and their combinations) become available, long-term antiviral therapy should become more effective and more cost-effective.16 This theoretical modeling also should be tested in real life prospective studies.
In summary, the present report not only confirms the efficacy of lamivudine across a broad spectrum of hepatitis B disease severity, including in hepatic decompensation, but also consolidates the general principle of long-term suppression of HBV replication by nucleos(t)ide analogs at various stages of HBV-related liver disease. In the future, randomized controlled trials as well as carefully planned and conducted observational cohort studies will play their roles in further clarifying certain practical issues, such as the timing to start antiviral therapy, the selection of antiviral agents, and the monitoring and tailoring of the treatment regimen for individual patients.