Highlights from
Digestive Disease Week

May 20-23, 2001
Atlanta, Georgia

Does HAART Affect Liver Disease Progression in HCV/HIV Coinfected Patients?

The short answer is that the Sterling data described below suggests HAART may accelerate HCV progression if cd4 increases are good & viral suppression is good. Other studies do not address the question as directly but suggest HAART does not accelerate HCV. So, we don’t have a clear answer yet but the Sterling data suggestion that patients with undetectable HIV viral load & high cd4s are more likely to develop cirrhosis raises interest.

An unanswered question is how HAART affects the liver of the HCV/HIV coinfected person. Opinions on this are mixed. Some doctors and researchers feel HAART may increase ALT, liver inflammation, and virus activity but this may not have much long term clinical significance in making HCV progress much more quickly. Obviously, the other side is that the affect of HAART on HCV liver disease may be to accelerate progression. A small French study previously reported that patients taking HAART did not have any faster HCV progression than patients with HCV alone. This has not been confirmed, although other researchers have reported findings suggesting the same.

Although this Sterling study discussed below has flaws in its design and analysis, as most studies do, it presents interesting findings. The study is small (n=39), but Sterlings finds that patients with undetectable or low viral load and cd4s>200 are more likely to have cirrhosis than patients with lower cd4s or higher viral loads. As an explanation for this observation, Sterling suggests that superior or optimal immune restoration from HAART may adversely impact on HCV disease severity. Another possible explanation is that patients with high CD4's and undetectable viral load were more adherent, and therefore had more HIV drug exposure.

Obviously, the patients with hi cd4s & low viral loads (n=29) received treatment for HIV and were doing well with therapy, as Sterling reports their viral load was 1.3 log and cd4s were 552 on average. The patients with CD4s <200 had 2.7 log viral load and 143 cd4s (n=9). The patients with high cd4s (>200) had HIV for longer 10 vs 7.8 years. Sterling reported that comparing the patients with above or below 200 cd4s there were no differences in ALT, % with normal ALT, HCV RNA viral load, % with high HCV viral load and total HAI (Hepatitic Activity Index).

But, coinfected patients with >200 cd4 were more likely to have cirrhosis (21% vs 0%) than patients with <200 cd4s. Although the % with advanced fibrosis was similar in both groups, no case of cirrhosis was seen in those with low cd4s. Also, patients with undetectable viral load were more likely to have cirrhosis (25% vs 5%).

The authors suggest that the more advanced fibrosis seen in those with higher cd4s and negative viral load suggests that immune function may be important to HCV disease severity. Immune restoration with effective HIV therapy may adversely impact on the severity of HCV disease in patients with coinfection.

In Maribel Rodriguez’s study (see below) she finds that liver disease progression is faster in HCV/HIV coinfected patients than HCV alone infected patients despite that the HCV monoinfected had more alcohol use (97% vs 65%). She reported that these results could be explained by the higher hepatitis C viral load at baseline (5.13 log vs 5.67 log, ie 132,000 copies/ml vs 467,000 copies/ml) observed in the co-infected patients when compared to the mono-infected patients. I think it’s important to bear in mind a limitation of this study. In speaking to the author he said some patients had undetectable HIV and some patients did not. This creates a question about the findings. If the patients with undetectable viral load or high cd4s progressed more quickly, as Sterling found, and since Rodriguez did not separate the data based on that it’s hard to conclude coinfected progress more quickly. Maybe its only those with lo cd4s & hifg viral loads that progress more quickly. As well, he was concerned that HAART related elevated lipids, insulin resistance, elevated glucose, and other HIV or HAART related negative effects on the liver may worsen fibrosis. In speaking to the study authors, Rodriguez expressed concern that HAART related elevated lipids, insulin resistance, elevated glucose, and other HIV or HAART related negative effects on the liver may worsen fibrosis.

Now, if you look at the study from San Diego (abstract is below), they found that having HCV did not negatively impact on the survival of patients with well managed HIV. But this study does not report whether these patients had cirrhosis or not as the study above addresses. So, I don’t think the two studies necessarily contradict each other. These patients may have had cirrhosis but may not have died.

At the HIV Retrovirus Conference (Feb 2001), Torriani from San Diego and Spanish researchers (abstract 575) looked at the development of hepatoxicity after starting ART (antiretroviral therapy) for 94 HCV/HIV coinfected patients compared to 94 HIV infected patients. They reported hepatotoxicity was correlated with immunologic (cd4) and virologic (HIV viral load) response to ART. 37/94 (39%) of HCV/HIV coinfected vs 10/94 (11%) with HIV had >2 fold increase in ALT, leading to switches in 8 patients and cessation of ART in 2 HIV/HCV patients vs none with HIV. Hepatoxicity was seen at all time points in coinfected patients.

In the study below from a group in Jacksonville they compare HCV/HIV coinfected patients who receive HAART or no HAART and look at biopsy. They concluded that HAART may play a role in slowing HCV disease progression as the French research group found. But this group did not analyze by response to HAART as the Sterling group did. The HAI inflammation activity score in the HIV/HCV patients with HAART was slightly less than patients without HAART. They also reported HCV RNA level and HAI inflammation activity scores were slightly decreased in HIV/HCV patients on HAART, but there was no difference in HAI fibrosis scores when compared with the control HIV/HCV and HCV patients, suggesting that HAART may play a role in slowing HCV disease progression. They also reported that after being on HAART for 2 years or more ALT (hepatoxicity) was more likely to increase.

     Richard K. Sterling, Robert Orenstein, Melissa J. Contos, Velimir A. Luketic, Arun J. Sanyal, A S. Mills, Mitchell L. Shiffman, MCV/VCU, Richmond, VA

Sterling ABSTRACT:

HCV and HIV infections are global health concerns. Due to shared modes of transmission, the prevalence of HCV-HIV coinfection is high. Until recently, the clinical manifestations of HCV were overshadowed by the morbidity of HIV. However, with the introduction of effective anti-HIV therapy, HCV has become a significant cause of illness in coinfection. Previous studies have been limited to pts with either abnormal ALT or advanced histopathology. As such, the true spectrum of liver disease has not been defined.

Aim: To describe the spectrum of liver disease and determine the significance of the immune system on the severity of liver disease in pts with HCV-HIV coinfection.

Methods: A cross sectional analysis of HCV-HIV pts evaluated at MCV/VCU from 1997 to the present was performed. All pts were positive for HCV-RNA (Amplicor, Roche) and HIV antibodies and negative for hepatitis B surface antigen and other causes of liver disease. Histologic Activity Index (HAI) was assessed by Knodell Score.

Results: 39 pts were evaluated: mean age 40 yrs; 77% male and 18% Caucasian. Comparing HIV-RNA + vs HIV-RNA - pts, + pts had lower CD4 counts (p=.02), a higher % with CD4 <200 (36 vs 9;p=.05), a higher % with normal (nl) ALT, and a trend toward less advanced fibrosis (bridging fibrosis [BF] and cirrhosis [cx]). No differences in HCV-RNA or % > 106 were seen. Pts with low (<200) vs high (>200) CD4 counts had similar ALT, HCV-RNA, and HAI. Although no significant differences in % BF/Cx were observed, no cases of cx were seen in pts with CD4 counts < 200ml/ml.

Conclusions: The spectrum of liver disease in coinfection includes a significant proportion of pts with nl ALT (59%) and BF/Cx (39%). The more advanced fibrosis seen in those with higher CD4 counts and negative HIV-RNA suggests that immune function may be important in HCV severity and that immune restoration from HIV therapy may impact on the severity of HCV.

     Deanna L. Oliver, Robin C. Hilsabeck, Lina Rossetti, Edward Barber, Wm. Christopher Mathews, Tarek I. Hassanein, Univ of CA, San Diego, San Diego, CA

The survival of HIV-infected patients has been improving since the introduction of highly active anti-retroviral therapies (HAART). HCV co-infection is prevalent in HIV patients who contracted the disease through IVDU. Current data suggests that co-infection with HCV reduces survival in patients with HIV. Examination of the impact of HCV on survival of well-managed HIV patients was the aim of this study. Methods: 1261 HIV patients seen over a 10-year period were included in this study. Mean age of the sample was 37.3 ± 8.3 years. 84% were male, and 16% were female. 286 patients were co-infected with HCV. To assess the influence of HCV co-infection on survival of patients with HIV, Kaplan Meier and Cox regression analyses stratified by year of entry were used to examine both the unadjusted and the adjusted survival distributions. Variables that were significantly associated with survival or HCV infection were entered as covariates in a multivariate Cox regression. Results: The log rank test for equality of survival was not significant (c2 = 1.32, p = 0.25), HCV status does not significantly impact survival of HIV patients. The multivariate Cox regression adjusting for baseline CD4 count, baseline plasma HIV viral load, race, age, and gender, resulted in a hazard ratio of 1.85 (p = 0.07). Thus, even when accounting for these significant covariates, HCV status did not predict survival in HIV patients. Conclusion: HCV status did not have a significant impact on survival of well-managed HIV-infected patients.

     Maribel Rodriguez, Fundacion Gastroenterologica de Diego, San Juan, PR; Jose F. Rodriguez, Univ of Puerto Rico, San Juan, PR

Background: HIV co-infection in HCV-infected patients is common and is associated to a greater risk of progression to end stage liver disease.

Methods: We evaluated 219 patients referred to our clinic for HCV treatment from the metropolitan area of San Juan during 1998-2000. These patients were divided as follows: HCV non-HIV = 84 pts and HCV/HIV co-infected = 135 pts. All patients underwent laboratory assessments including ALT, HCV PCR, HCV genotype, and liver biopsy. Detailed medical history was also obtained (risk factors for infection, time of diagnosis, and alcohol use). Statistical analysis was performed using Statview(SAS Institute). The Kolmogorov-Smirnov test was used to test for normalcy of distribution. Student's t-test was used to compare the aforementioned variables between both groups. Chi-square (Fisher's Exact p-value) was also used to compare nominal variables.

Results and Discussion: HCV-non-HIV and HIV-HCV co-infected patients have similar risk factors of HCV infection and the gender distribution was also comparable. There were no significant differences in ALT mean values (48 vs. 43; p>0.1), HCV genotype 1 (77% vs. 79%; p>0.1), and histology among these groups (Knodell score: 10.2 vs. 10.4; p>0.1). Both groups showed a high percentage of advanced liver disease (48% vs. 50%) despite the fact that HIV-HCV co-infected patients had a shorter time of diagnosis to HCV infection (60 mo. vs 7 mo.; p=0.0001) and they were younger than HCV mono-infected patients (43 y vs. 48 y; p=0.0007). Thus, progression of liver disease is faster in HIV/HCV co-infected patients than HCV mono-infected patients, even though the latter pool of patients used more alcohol than the co-infected population (97% vs. 65%; p<0.0001). These results could be explained by the higher hepatitis C viremia observed in the co-infected patients when compared to the mono-infected patients (5.68 log vs. 5.12 log; p=0.0002). A multivariate analysis is still in progress and will be also presented.

Conclusions: HIV-HCV co-infected patients have a faster progression to severe fibrosis and cirrhosis than HCV-mono infected patients.

      Abhijit Roychowdhury, Div of GI, Univ of Florida Health Science Ctr, Jacksonville, FL; Louis R. Lambiase, Carmela Monteiro, Laura Cubbedge, Joseph Allen, Jianjun Li, Same as above

Background: The highly active antiretroviral therapy (HAART) has significantly modified the prognosis of HIV infection. However, the role of HAART in the progression of hepatitis C in patients with HIV/HCV coinfection has not been well established.

Aims: To investigate the impact of HAART on HCV disease in patients with HIV/HCV.

Methods: A total of 109 patients, including 59 patients with HIV/HCV and 50 patients with HCV, were investigated. There were 75% of HIV/HCV patients received HAART. In order to compare short and long term effects of HAART in HIV/HCV patients with and without HAART and HCV alone, four groups were divided up as following: Group A (N=50): HCV alone; Group B (N=9): HIV/HCV without HAART; Group C (N=24): HIV/HCV using HAART less than 2 years; Group D (N=26) HIV/HCV using HAART for more than 2 years. Each group had a 50% alcohol use. Serology including HIV RNA, HCV RNA and ALT was analyzed. Liver biopsies were performed in 44 HCV and 43 HIV/HCV. A modified Knodells histological activity index (HAI) was used to define the grade of inflammation and extent of fibrosis.

Results: Serology: There was a significant decrease in HIV RNA levels in patients with HAART (HIV RNA log10: Group C 3.31±0.313 and Group D 2.69±0.24) compared to patients without HAART (3.89±0.29) [P< 0.05]. A slight decrease but no significant difference in HCV RNA levels in HIV/HCV with HAART was noted as compared with patient without HAART (HCV RNA log10: Group A 5.92±0.25; Group B 5.97±0.29; Group C 5.64±0.23 and Group D 5.29±0.23). There was a significant elevation of ALT levels in patients of Group D (67%, 17/26) compared to patients in Group C (42%, 10/24) and Group B (33%, 3/9) [P<0.05]. Histology: The HAI inflammation activity score in the HIV/HCV patients with HAART was slightly less than patients without HAART (mean score: Group A: 7.51±0.47; Group B: 6.61±0.78; Group C: 5.67±1.43; and Group D 5.42±0.58; P=N.S). No significant difference in HAI fibrosis score was detected among the four groups (mean score: Group A: 2.46±0.24; Group B: 2.33± 0.42; Group C: 2.88 ±0.41 and Group D: 2.84±0.34).

Conclusion: 1). HCV RNA level and HAI inflammation activity scores were slightly decreased in HIV/HCV patients on HAART, but there was no difference in HAI fibrosis scores when compared with the control HIV/HCV and HCV patients, suggesting that HAART may play a role in slowing HCV disease progression. Further evaluation is needed. 2). Hepatic cytolysis is more frequent among patients treated with long term HAART, suggesting increasing the risk of hepatoxicity in long-term use.


to top

All DDW Articles >