HIV Medications & HCV at Dallas AASLD Conference: More Research Needed
    Reported by Jules Levin

At the AASLD liver conference in Dallas 2 studies (abstracts below) were reported on regarding HIV medications for the person with HCV. Nyingi Kemmer, a pathologist from Texas performed biopsies on coinfected persons after starting HAART and found abnormalities (see abstract below). He conducted biopsies following consultations with HIV doctors about these cases. Several HIV docs I talked with in Dallas were skeptical of these findings. But this is often the case where researchers are initialy skeptical of new research. In addition, several doctor/researchers in HIV & hepatitis I have spoken with do not think HIV meds cause long term harm to HCV progression in HCV/HIV coinfected individuals. The pathologist at Dallas suggested PIs may be responsible for these abnormalities but of course further research needs to be done, and there is no information on if these abnormalities have real clinical affect, and affect longer term HCV progression. Opinions by thought leaders are mixed on whether or not HIV meds worsen liver condition & progression of HCV. I think that the affect of HIV meds on HCV progression may be different for different individuals based on their circumstances. 

You may recall Mark Sulkowski's study, published I think in JAMA around the turn of 2000, showing elevated LFTs with protease inhibitors in a small percentage of coinfected individuals. You can read about this study on the NATAP web site. In particular, individuals taking ritonavir experienced more severe incidence of elevated LFTs. We don't know what the long-term clinical significance of this is. Does it have a negative impact on progression of HCV in coinfected individuals? And is there an affect in non-HCV HIV infected individuals over the long term? I don't think we know, and opinions are mixed among a handful of expert researchers/doctors I've spoken with. Most tend to think there is no negative impact. Some say they don't know.  Another small study from French research group found PI HAART stopped fibrosis progression in coinfected individuals. Several researchers I've spoken with don't believe these findings. They think PI HAART therapy will  not halt HCV progression. Several studies show HIV accelerates HCV progression 3-5 fold. Most researchers/doctors I've spoken with accept this notion, but but one noted researcher I spoke with believes this fear is overblown. A minority do not believe HCV progresses as quickly in HCV/HIV coinfected individuals as several studies suggest.

What is important I think is that this area receive attention by researchers. We don't have clear answers to this question. It's possible that for some individuals treating HCV before HIV may be beneficial.

The second study looks at the affect of NNRTIs and study results show that the overall rate of Grade 3-8 hepatotoxicity with NNRTIs is low; but the rate of Grade 1-2 LFT elevations ranges from 33% for nevirapine to 27% for efavirenz (see the table below. When hepatitis is present, Grade 1-2 LFT elevation is higher than when hepatitis is not present (48-28%; see table below. The rate of severe hepatotoxicity among patients coinfected with HBV and HCV was 2/53 (3.8%), while the rate among patients known to be HBV and HCV negative was 8/178 (4.5%). Of note, Ron Palmon from NYU reported there were no statistically significant differences between the 3 NNRTIs but all 3 cases of grade 4 hepatotoxicity in the study were associated with the use of nevirapine. In fact, 92% of the study participants (n=275) in this retrospective analysis were male. In Glasgow Ian Sanne reported 9.4% grade 4 LFT elevations in the nevirapine arm of the South African study, but he reported that the high grade 4 incidence rate was driven by women. He felt women may be affected differently than men.

There are 3 potential models that have been proposed:

(1 ) HIV meds don't worsen liver progression in HCV infected individuals

(2 ) HIV meds do worsen progression

(3 ) HAART makes fibrosis progression stop or slow as suggested in small French study (opinions are mixed on the viability of this study).

The point is we don't have an answer to this question. Following are the abstracts from the authors.

A DISTINCTIVE HISTOLOGIC PATTERN OF LIVER INJURY IN HIV-POSITIVE PATIENTS ON HAART: A POSSIBLE HEPATOTOXIC EFFECT OF PROTEASE INHIBITORS.
    Nyingi M Kemmer, Claudia P. Molina, John E. Fuchs, Emil P. Miskovsky, David M. Asmuth, Michael J. Borucki, Steven A. Weinman, Univ. of Texas Medical Branch, Galveston, TX; Brian West, New York Univ. New York, NY.

Liver disease is common in patients with HIV infection. Recently, HAART (highly active anti-retroviral therapy), and specifically Ritonavir, have been implicated as causing hepatocellular damage with marked transaminase elevations (JAMA 2000;283:74-80). Following the introduction of HAART, we observed a distinctive histologic pattern of liver injury in biopsies from HIV patients that we had not seen previously in this population. We term this pattern BKPF for the three observed components: hepatocyte Ballooning with Kupffer cell activation and Pericellular Fibrosis in zone 3. It is often associated with central-to-central bridging fibrosis, without portal inflammation, iron overload or evidence of immune liver disease or infection other than HBV or HCV. Withdrawal of HAART led to return of transaminases towards normal in several of these patients.

Objective: To determine the frequency of BKPF in patients with HIV undergoing liver biopsy, and investigate its relationship to the use of specific medications. Method: Liver biopsies of 110 adult patients with HIV infection on HAART, performed in 1997 and 1998 at the University of Texas Medical Branch because of elevated transaminases, were reviewed by one pathologist without knowledge of the clinical data. The morphologic findings were then correlated with patient age and gender, liver enzymes, CD4 count, quantitative HIV and HCV RNA, HCV Ab positivity, HBsAg and detailed history of medication exposure during the two months preceding the biopsy. Medication history was derived from patient records and the pharmacy database.

Results: Of the 110 patients, BKPF was present in 28 (26%). A minority of these patients had other features of liver injury in addition, such as steatosis, cytoplasmic cholestasis, ductular proliferation, bile duct injury, Mallory bodies, 'giant mitochondria' or giant cell transformation. Among the 82 non-BKPF patients a minority had activated Kupffer cells, pericellular fibrosis or steatosis, but none had ballooning degeneration of hepatocytes. Both groups were nearly identical in age, gender, HCV Ab positivity, HIV RNA, HCV RNA, CD4 count, bilirubin, alkaline phosphatase and ALT. However, BKPF patients were on a greater number of drugs than non-BKPF patients, and notably more new protease inhibitors. No one specific medication correlated with the presence of BKPF.

Conclusion: BKPF is a frequent finding in HIV patients with markedly elevated transaminases who are taking multiple medications and in particular HAART. We hypothesize that this is a toxic injury, possibly related to the use of protease inhibitors.

HEPATOTOXICITY ASSOCIATED WITH NON-NUCLEOSIDE REVERSE TRANSCRIPTASE INHIBITORS FOR THE TREATMENT OF HUMAN IMMUNODEFICIENCY VIRUS AND THE EFFECT OF HEPATITIS B OR C VIRUS INFECTION.
    Ron Palmon, New York Univ. School of Medicine, New York, NY; Rose Tirelli, Liberty Medical, LLP, New York, NY; Jim F. Braun, St Vincent's Medical Ctr, New York, NY; Stacy Kreiswirth, Liberty Medical, LLP, New York, NY; Alex F. McMeeking, New York Univ. School of Medicine, New York, NY; Michael Mullen, Cabrini Medical Ctr, New York, NY; Karen Weisz, Liberty Medical, LLP, New York, NY; Douglas T. Dieterich, New York Univ, New York, NY

Antiretroviral drugs for the treatment of human immunodeficiency virus (HIV) have been associated with hepatotoxicity. This is of particular concern in patients who are coinfected with hepatitis B virus (HBV) or hepatitis C virus (HCV). We conducted a retrospective study of hepatotoxicity associated with a specific class of antiretroviral drugs: non-nucleoside reverse transcriptase inhibitors (NNRTI's). Our population consisted of 275 patients in an urban HIV practice. HBV and HCV status was available for 175 patients, of which 24 (13.7%) were coinfected with HCV, and 20 (11.4%) were coinfected with HBV. We included all patients who were treated with delavirdine, nevirapine, or efavirenz for which laboratory data was available. Patients who were treated with more than one of the above drugs at different times were included once for each course of treatment. We examined the rate of hepatotoxicity as defined by grade 0-4 change in serum alanine aminotransferase or aspartate aminotransferase. Severe (grade 3 or 4) hepatotoxicity was noted in 10 (2.8%) out of 357 total courses of treatment with a NNRTI. The average time to the development of severe hepatotoxicity was 210 days. The rate of severe hepatotoxicity was 2/55 (3.6%) with delavirdine, 3/126 (2.4%) with efavirenz, and 5/176 (2.8%) with nevirapine. There were no statistically significant differences among the three. Of note, all 3 cases of grade 4 hepatotoxicity found in our study were associated with the use of nevirapine. The rate of severe hepatotoxicity among patients coinfected with HBV or HCV was 2/53 (3.8%), while the rate among patients known to be HBV and HCV negative was 8/178 (4.5%). There was a statistically significant (p=0.033) increase in grade 1-2 hepatotoxicity in coinfected patients. These results suggest that the overall rate of severe hepatotoxicity with NNRTI's is low. However, patients who receive nevirapine may be at a higher risk for grade 4 hepatotoxicity. The risk of severe hepatotoxicity does not appear to be increased in patients with HBV or HCV, suggesting that it is safe to use these drugs in this population. This is distinct from another class of antiretrovirals, protease inhibitors, which have been associated with an increased risk of severe hepatotoxicity in coinfected patients.