Summaries from the American Association for the Study of Liver Diseases Conference
Synopsis: The American Association for the Study of Liver Diseases (AASLD) met in Chicago, IL, on Nov. 4-6, 1998. Abstracts are published in Hepatology (1998;28:Supplement 1). Historically, this is the pre-eminent meeting for presentation of data on the epidemiology, pathogenesis, treatment, and resistance issues relating to all hepatitis viruses. The predominant forces in the field remain the pathologists and hepatologists, although infectious disease specialists are beginning to make a small mark. It is no surprise to learn that as treatments improve, the key clinical management issues relate to the realm of infectious diseases: viral load, resistance, etc. A total of 2440 abstracts were published and 1721 accepted for presentation, so the summary that follows touches only on selected teaching points. —Stephen L. Sacks, MD, FRCP
Several key features of basic management paradigms of hepatitis B and C shifted significantly, or at least began to shift, during this important meeting. Hepatitis C is a common and important disease, soon to manifest itself in large numbers emanating from transient injection drug use (IDU) during the 60s and 70s. Sexual transmission of HCV was also confirmed here as a common source of transmission. In fact, it ranks only second in frequency to IDU. Other routes not always included in the list include cocaine snorting, endoscopy, and medical use of needles in countries not adherent to strict rules of sterilization (e.g., patients treated for schistosomiasis are at extremely high risk).
Dozens of studies examined the combination of interferon alpha and ribavirin for the management of hepatitis C. In fact, there is no question that this combination significantly increases the proportion of patients responding to treatment compared with interferon alpha alone. The combination also increases the proportion who escape relapse when therapy is withdrawn. Viral loads seem to decrease within several hours of the first interferon injection and then rebound a bit during the skipped day before the next injection, then, in a biphasic curve, head quickly to undetectable in responding patients. Prolonged half-life interferons, such as pegylated interferons, may obviate the need for continuous interruption of the antiviral effects that may, in turn, increase response rates.
Previous data have shown that HCV PCR RNA should be negative within 12 weeks after commencement of interferon alone. Patients not fully responsive (to interferon alone) by this point are very unlikely to respond—ever. Interestingly, the decision to pull therapy should probably be delayed in patients on combination therapy until five or six months. This is not because ribavirin slows the antiviral response, but rather because it pushes more patients with high viral loads into the category of potential responders and they may take longer to reach baseline—often doing so during the second three months of therapy. Patients on combination who have not responded by six months are unlikely to benefit from continuing treatment. A number of hepatology magic rules then intervene that are less well grounded on data. Many would consider an ALT response to the normal range as favorable, even in the absence of an antiviral response. These individuals will often retain ALT normality for some time after treatment withdrawal, even in the face of renewed viral replication. Other studies suggest that ALT should not be used to make treatment withdrawal decisions.
Combination therapy is more effective than simple alpha interferon therapy in relapsers (respond to interferon but relapse on withdrawal) but is also more effective (in decreasing orders of magnitude of efficacy) in treatment-naïve patients and even to some extent, in nonresponders. Should a nonresponsive patient be encouraged to face another year of interferon side effects and treatment costs for a low but measurable increase in the chances of recovery? This is not settled. Treatment still needs to be individualized. Ironically, the best responders are those with the least aggressive liver disease.
An exciting set of findings, however, demonstrated the relative consensus of data suggesting that type II and type III genotypes of HCV respond more quickly and more often. It appears appropriate to discontinue combination therapy in patients with these genotypes at six months without altering response rates, thereby saving a full six months’ of drug expense and toxicity. It is clear, then, that a pretreatment genotype will save a significant amount of money and morbidity in non-type-I patients. Patients with genotype 1 HCV can be pulled from treatment at six months if they are failing combination therapy, as defined by at least a continuing positive PCR RNA, since they will continue to fail. If ALTs have normalized and viral load is still measurable, you may have a hard time stopping therapy, but there is a paucity of data to support continuation. Patients with genotype I HCV who have a complete loss of viral load at 5-6 months should be continued on combination therapy for a full 48 weeks. Since type 1 (a and b) is the most common in the United States, it is important to discriminate genotype in understandng the plethora of published and presented data in this field.
Many current studies will still use interferon alpha, alone, in various doses. Expectations and predictors for treatment response are different for those using interferon alone compared with combination therapy. Ribavirin is generally well-tolerated, but causes a significant anemia in many, requiring drug withdrawal or dose reductions in some cases. If insurance is able to pay and the situation calls for this, erythropoetin is worth trying in patients who need treatment but cannot tolerate their hemoglobin drop. The cost of erythropoetin, however, is often prohibitive. The highest risk patients for fibrosis and progression to cirrhosis and hepatocellular carcinoma (which requires cirrhosis in HCV) are males, older than 40, with existing severe liver disease. One study suggests that the gamma GT to ALT ratio is inversely proportional to the treatment response rate. African Americans clearly do not respond well to interferon and respond poorly to combination therapy. Accordingly, it may not be justified to use the current drug regimens in most African American patients. Japanese data suggest that interferon somehow prevents some hepatocellular HCC even when it fails to resolve hepatitis.
New interferons are on the horizon, including pegylated interferons (given less often with prolonged half-lives) and consensus interferons engineered to mimic all the human interferon subtypes. Early data suggest that both are more effective than interferon alpha alone. Accordingly, combinations of one of these new interferons with ribavirin offer hope of even better response rates, perhaps driving expectations in naïve patients more than 50% in the near term. In counseling your patients, know that most (80-90%) patients with undetectable RNA by PCR six months after treatment withdrawal will sustain that response over the long-term. The much studied amantadine has no mechanistic rationale for effectiveness in HCV. Clinical studies confirm the absence of activity as a single agent or as a mixture with interferon alpha. Furthermore, prednisone withdrawal works just as poorly in HCV as has been shown previously with HBV.
Clinical management of patients with HCV requires knowledge of many psychosocial factors. Fatigue is a common symptom and has been thought to result from changes in central neurotransmission (Hepatology 1997; 25:492-494). Jones and Yurdaydin were not able to show a correlation between fatigue in HCV and decreased physical activity. Depression is a common cofactor in HCV. One study demonstrated psychological disorders in two of three HCV patients. Severity of fatigue, however, is independent of the presence of depression or anxiety, or surprisingly, the severity of underlying liver disease. The degree of dysfunction induced by fatigue, however, correlates well with underlying depression. Depression is often worsened by interferon and suicide is also a potential risk. Nevertheless, Jones and Yurdaydin treated HCV patients with serious psychiatric disease and found that treatment increased morbidity only modestly and, with close psychiatric observation, was manageable. Another study of fatigue showed correlations with age older than 40, female gender, heavy drinkers, and those with extensive liver fibrosis. Fatigue was associated with arthralgia, myalgia, paresthesia, and pruritis, but not with cryoglobulinemia, genotype, or histological activity scores.
What about long-term prognosis? Looking at all-comers, Seeff and colleagues (Seeff LB, et al. N Engl J Med 1992;327:1906-1911) followed up on their classical study showing no change in mortality overall by following patients for an additional five years. Again, there was no difference in overall mortality, although there appeared to be a difference in liver-related mortality identified in the follow-up. In understanding the Seeff data, it is important to note that the so-called benign outcome of patients with HCV in these studies refers to all persons infected, including those who spontaneously clear the virus and those who never develop abnormal liver function tests. These are not the patients presenting to our offices for treatment who are already selected to some extent. Unfortunately, we still have no reliable prognostic indicators to predict outcomes for the patients who present for treatment with elevated ALTs and positive HCV PVR RNA. In a study from Argentina, the majority of patients (post-transfusion-related HCV) developed a slowly progressive disease with cirrhosis in more than 30%. All HCC followed cirrhosis in this study. There is no question that only a small proportion of the overall population of infected persons with HCV will not have a benign outcome. Unfortunately, the predictive factors are general and statistical and cannot be translated to the individual case where clearly, some risk of serious sequelae exists.
Hepatitis B is more treatable now. The good news: most patients respond to lamivudine at 100 mg daily. This drug is well tolerated and effective. A small proportion of patients will develop antiHBe and lose HBeAg with treatment and in about half of those, seroconversion will persist. Rarely, a patient will seroconvert also to anti-HBs. It appears that worrisome elevated ALTs on treatment or at treatment withdrawal are just as common in placebo recipients. However, some ALT rise is expected 12 weeks after treatment is stopped. Treatment reduces the DNA and ALTs of HBeAg-negative patients also (precore mutants and others). To qualify for treatment, patients should have a measurable HBV DNA by a non-PCR test. These are relatively insensitive and a positive result apparently reflects a significant viral load worthy of treatment. To my knowledge, it is not yet clear that those with lower viral loads should not also receive treatment but this has not been adequately studied. It is clear that responders to lamivudine do well, benefiting from reduced ALTs and improved histologies. However, be wary of data showing glowing histology rates, since most patients who fail to respond to treatments drop out and move on, leaving the success stories to be rebiopsied happily at the conclusion of studies. We see few histologies on the true failures. This has been the model for both HBV and HCV treatment protocols.
The bad news is resistance. A Chinese patient population treated over a two-year period saw a 61% incidence of resistance to lamivudine. Others have found the one-year rate to be 15% and the two-year rate to be about 30%. HBV develops resistance to lamivudine in the same place as HIV, the YMDD sequence of the reverse transcriptase. Such strains may be less virulent in vitro and patients with breakthroughs of resistance often (but not always) maintain lower HBV DNAs and ALTs quantitatively than they did prior to therapy. Does this justify maintenance of therapy in the face of resistance? This is not proven and is reminiscent of the black magic of HCV treatment. Don’t forget, it took years to demonstrate that HIV resistance was clinically meaningful. Lamivudine and lamivudine plus HBIG have been effective in reducing the incidence of fibrosing cholestatic hepatitis (FCH) following transplantation for HBV. However, resistance rates are even higher here and with HIV, as one would expect. Resistance often results in liver failure or FCH in this setting.
New drugs are desperately needed. Until AASLD, hope was held for famciclovir 500 mg tid as an alternative or as a synergistic agent in combination. Once again demonstrating the importance of the properly controlled, randomized trial, several abstracts at this meeting showed that famciclovir is an inferior drug to lamivudine in clinical trials. DNA levels are reduced but only partially overall. A large proportion of patients do not ever respond and among those who do, resistance is also possible. In any event, famciclovir has not been shown to reverse clinical disease and probably has no future in the management of HBV.
But, there are important new drugs. Adefovir dipivoxil (Gilead Sciences) and lobucavir (Bristol-Myers Squibb) are in clinical trials. Phase II studies of either were positive with good reductions of viral load (approximately 3-4 logs) good clinical responsiveness, minimal side effects, etc. Phase III’s are now in progress. Preclinical studies of BMS 200, 475—a new agent—support an incredible potency with more than a 7 log drop in animals. Lamivudine-resistant strains are susceptible in vitro to both adefovir and lobucavir.
It is inherently obvious that resistance is a function of several factors in HBV that are also important for HIV. The bottom line is the degree of reduction of viral load. Nothing, to date, is capable of lowering viral loads far enough as a single agent. Resistance to lobucavir and adefovir has not been seen yet in clinical specimens, but the numbers of treated patients are small. Perhaps their increased potencies over lamivudine will keep them from the funeral pyre of single agent treatment. BMS 200, 475 perhaps will be potent enough on its own to avoid the need for combination. It is clear that interferon and lamivudine (8 weeks lamivudine followed by combination) are no better (and may be worse) than either agent alone. Other combinations will come slowly. None have been tested to date, although limited plans are underway. Lamivudine needs to find itself a good drug for combination or it risks resistance rates that parallel its use as a single agent in HIV. Whether the potency of the drug or its combination will determine treatments of the future is unclear. For now, the hepatologists are just beginning to address long-term or life-long drug administration as the obvious solution. Triggers for stopping are being identified, such as anti-HBe seroconversion, but this is risky as seroconversions can revert. On the lighter side, it would appear that a possible source of HBV infection in endemic areas is the common bedbug (Cimex Lectularius L.) found in beds and movie theater seats in the Phillipines, for example.
Alaskan investigators introduced alphafetoprotein screening (902) q 6 months with ultrasound on those above 15 mg/mL for native patients with HBV. They increased five-year survival from hepatocellular carcinoma (HCC) from 0% to 45% using this paradigm. The problem with AFP is its relatively poor positive predictive value. CT scan appears more sensitive than ultrasound in detecting HCC.
There are new hepatitis viruses, too. Hepatitis G is still being studied and its vertical transmission pathway idenitifed. Parvovirus can cause an acute fulminant hepatitis that is rarely fatal, fully reversible, and should not be transplanted. Transfusion-transmitted virus (TTV) is a newly discovered DNA virus (Biochem Biophys Res Comm 1997;241:92) that has four distinct serotypes identified so far. Every study of this agent suggested that no significant liver disease results from it, much like HGV. The role for these viruses as possible cofactors continues to be debated and studied.
HIV-HCV coinfected patients do significantly worse than HCV patients without HIV. Alcohol abuse continues to be a significant cofactor for HIV and for all HCV patients. Be careful of hepatotoxicity in HIV-infected patients on HAART. Coinfections with HCV and/or HBV are common and may complicate therapy. Certain protease inhibitors were again found to be a problem here, associated with cholestasis, especially in coinfected patients. Some investigators did not see a correlation with protease inhibitors and disease severity. Others found that protease inhibitors improved liver function as a secondary effect of reducing HIV loads. Interferon alone does not induce a long-term benefit in HIV-coinfected patients, regardless of dose. Lamivudine does lower HBV DNA in patients coinfected with HBV, but resistance is common in this group. It appears safe to combine HAART with ribavirin and interferon without altering HIV load, despite ribavirin’s potential for antagonism with some antiretrovirals. Patients with CD4 counts of more than 300 appear to respond to therapy as controls in a preliminary study of interferon alone. HCV is also a significant cofactor in morbidity and mortality of kidney transplant patients. Severe cholestatic hepatitis was associated with genotype 1. Coinfection with hepatitis A virus was suggested to carry a 35% mortality rate in HCV-infected individuals recently (N Engl J Med 1998;338:286). This was not confirmed in a Swiss study of more than 5000 patients, although there would be little reason not to immunize HCV-infected patients against both HBV and HAV, if indicated.
In summary, combination therapy appears to be the treatment of choice for HCV. The situation still requires careful patient selection and counseling. Not all patients will want to be treated. Most will want treatment regardless of reassurance about the low overall risk of death from liver disease in patients exposed to HCV, because prognosis is often not predictable and treatment may be more effective in advance of fibrosis. Many will suffer fatigue and depression and, for many, antivirals will not change that reality. Much of our disease management approach is based on studies dating back two decades when interferon alpha was used alone and disease was measured by the ALT response. It is important to distinguish the changing rules in assessing treatment studies from the past.
1. Poynard T, et al. Lancet 1998;352:1426-1432.
2. Marcellin P, et al. Ann Intern Med 1997;127:875-881.
3. Reichard O, et al. Lancet 1998;351:83-87.
4. Nevens F, et al. Gastroenterology 1997;113(4): 1258-1263.
5. Lai CL, et al. Hepatology 1997;25(1):241-244.
6. Markowitz JS, et al. Hepatology 1998;28(2):585-589.
7. Mutimer D, et al. J Hepatol 1998;Jun 28(6):923-929.
8. Lai CL, et al. N Engl J Med 1998;339(2):61-68.
9. James JS. AIDS Treat News 1998;293:6-7.