Studies Look at Liver Fibrosis in HIV-HCV and HIV-HBV Coinfected People

By Liz Highleyman

Over years or decades, chronic hepatitis B virus (HBV) or hepatitis C virus (HCV) infection can progress to advanced liver disease including cirrhosis and liver cancer. Some evidence indicates that this process may happen faster in HIV positive people -- especially those with low CD4 cell counts -- but data have been inconsistent.

Several studies presented at the recent 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009) last month in Montreal shed further light on fibrosis in HIV-HBV and HIV-HCV coinfected individuals.

More Fibrosis in HIV-HCV Coinfected Hemophiliacs

Margaret Ragni from the University of Pittsburgh and colleagues compared liver disease progression in 92 HIV-HCV coinfected and 128 HCV monoinfected hemophiliac men at 34 U.S. treatment centers (participants in the HHH Hemophilia Cohort Study) who underwent liver biopsies.

Among the 220 men with available biopsy findings, 86 (39%) had some degree of fibrosis (Ishak scores > 3 or Metavir scores > 2). However, HIV-HCV coinfected participants were significantly more likely to have fibrosis (47%) than HCV monoinfected individuals (32%; P = 0.03), despite being about the same age (43 vs 40 years), being infected with HCV for a similar duration (38 vs 36 years), drinking similar amounts of alcohol, and being about equally likely to receive hepatitis C treatment (48% vs 41%). After controlling for HIV status, older age --but not race, alcohol consumption, or lack of hepatitis C treatment -- predicted worse fibrosis.

Average Ishak fibrosis scores were 2.3 in the coinfected group versus 1.9 in the HCV monoinfected group (Metavir 1.6 vs 1.3), a difference that was marginally significant (P = 0.06 for Ishak, 0.045 for Metavir). HIV-HCV coinfected participants also had higher average AST levels and lower average platelet counts than HCV monoinfected men.

The investigators also measured levels of various cytokines to test the hypothesis that HIV might accelerate liver disease progression through up-regulation of cytokines that promote fibrosis. Coinfected men were more likely than HCV monoinfected men to have high TFG-beta-2 levels, but less likely to have elevated interferon-gamma (IFN-gamma). Levels of interleukin-6, interleukin-10, and cytokine promoter genotypes did not differ according to HIV serostatus.

"Nearly half of HIV-HCV coinfected hemophiliac men have Ishak > 3 fibrosis, a 1.5-fold higher prevalence than in HCV monoinfected men," the investigators concluded. "Cytokine dysregulation may contribute to fibrosis progression among those with HIV-HCV coinfection, but other mechanisms likely affect fibrosis score."

No Fibrosis Difference in Texas Study

In another analysis of the relationship between fibrosis and HIV infection, Mamta Jain and colleagues from the University of Texas Southwestern Medical Center in Dallas performed a retrospective cross-sectional analysis of liver biopsies from 180 HIV-HCV coinfected and 407 HCV monoinfected patients between 1998 and 2008. The HIV positive participants had well-preserved immune function, with a median CD4 count of 459 cells/mm3 and 70% with undetectable HIV viral load.

This study classified biopsy samples using the Batts-Ludwig scale (mild fibrosis = stage 0, 1, or 0-1; moderate = 1-2, 2, or 2-3; severe = 3-4 or 4). Here, HIV-HCV coinfected and HCV monoinfected participants had a similar distribution of mild (39% vs 34%), moderate (41% vs 49%), and severe fibrosis (18% vs 16%). Liver steatosis, or fat accumulation, was slightly less common among HIV negative individuals (genotype distribution did not differ, with about 90% having genotype 1).

"The distribution of fibrosis stage was similar in those with HCV compared to those with HIV-HCV, which may be due to high CD4 cell counts and viral suppression of HIV in the coinfected population," the researchers concluded; however, they added that the coinfected patients were younger (median 43 vs 47 years) and "faster progression of disease remains a possibility."

In a multivariate analysis adjusting for potential confounding factors, older age, steatosis, and elevated AST were independent risk factors for moderate-to-severe versus mild fibrosis, but HIV infection itself was not, leading the investigators to suggest that "perhaps other factors besides HIV-infection may play a significant role in the development of fibrosis."

Effect of HIV-HBV versus HIV-HCV Coinfection

Paula Tuma and colleagues from Hospital Carlos III in Madrid, Spain, looked at liver disease outcomes in HIV positive individuals coinfected with HBV or HCV. This longitudinal retrospective study included all 27 HIV-HBV coinfected patients seen at their center who had undetectable HBV DNA while being treated with HAART that had anti-HBV activity. [The nucleotide/nucleoside analog drugs lamivudine (3TC; Epivir), emtricitabine (Emtriva), and tenofovir (Viread, also in the Truvada and Atripla combination pills) have dual activity against both HIV and HBV.] In addition, 46 patients with HIV only and 148 HIV-HCV coinfected patients -- also treated with HAART -- were used as comparison groups.

The median age was 47 years in the HIV-HBV group and 44 years in the other 2 groups. Overall, patients had well-controlled HIV with a median CD4 count of 528 cells/mm3 and HIV RNA < 50 copies/mL. At baseline, median ALT levels tended to be higher in HIV-HCV coinfected compared with HIV-HBV coinfected or HIV monoinfected patients, but the difference did not reach statistical significance.

The investigators used transient elastometry (FibroScan), a method that estimates fibrosis based on liver "stiffness." Changes in estimated fibrosis were compared between the first and last examinations, done 2.0 to 3.6 years apart. Baseline liver stiffness was 6.9, 5.6, and 6.8 kiloPascals (kPa) in the HIV-HBV, HIV monoinfected, and HIV-HCV group, respectively, but again this was not a significant difference.

After a median 32 months of follow-up, HIV-HCV coinfected patients had significantly higher median ALT (55 IU/L) than HIV monoinfected (34 IU/L; P = 0.01) or HIV-HBV coinfected (30 IU/L; P = 0.02) participants. Median liver stiffness was similar in the HIV-HCV and HIV-HBV coinfected groups (7.7 vs 7.4 kPa), but significantly lower in the HIV monoinfected group (5.4 kPa; P = 0.02).

About twice as many HIV-HCV coinfected patients experienced liver disease progression of at least 1 Metavir stage compared with HIV-HBV coinfected patients, who in turn were about twice as likely to progress as the HIV monoinfected participants (26.4%, 14.8%, and 6.5%, respectively). But median liver stiffness increased by 0.6 and 0.05 kPa in the HIV-HCV and HIV monoinfected groups, while slightly decreasing by 0.2 kPa in the HIV-HBV coinfected group (P = 0.03).

In a multivariate analysis adjusting for age, sex, baseline ALT, and baseline liver stiffness, HIV-HCV coinfected patients showed significantly more fibrosis progression than those with HIV-HBV coinfection (P < 0.05), with an even greater disparity compared with the HIV monoinfected group. However, the difference between the HIV-HBV coinfected and HIV monoinfected groups did not reach statistical significance (P = 0.3).

Based on these findings, the researchers concluded, "Liver fibrosis progression in HIV patients is only associated with HCV coinfection in the HAART era. The use of anti-HBV active HAART regimens has halted and/or even reversed liver fibrosis in HIV-HBV coinfected patients." However, given the greater likelihood of progression in the HIV-HBV patients and the small numbers in these 2 groups, a deleterious effect of HIV-HBV coinfection cannot be ruled out.

Spanish GRAFIHCO Study

Juan Pineda from Hospital Universitario de Valme in Seville, Spain, and colleagues also used the transient elastometry in a cross-sectional analysis of liver fibrosis among 1310 HIV-HCV coinfected patients at 95 institutions throughout Spain participating in the GRAFIHCO study (individuals who were triply infected with HBV were excluded). Using this non-invasive method, the researchers noted, allowed them to include more patients than typical biopsy-based studies.

Out of the 1310 patients, 524 (40%) had a liver stiffness > 9 kPa, indicating "significant" fibrosis, while 255 (19%) had liver stiffness > 14 kPa, indicating cirrhosis. Factors independently associated with significant fibrosis were older age (adjusted odds ratio [OR] 1.04; P 0.002), heavy alcohol consumption (> 50 g/day during the past 5 years)(OR 1.58; P = 0.013), and duration of HCV infection (OR 1.03; P = 0.023).

Having a CD4 count below 200 cell/mm3 was also marginally associated with greater fibrosis (OR 1.67; P = 0.053), which the researchers said supports earlier initiation of antiretroviral therapy in this setting. On the other hand, having HCV genotype 4 was marginally associated with less fibrosis (OR 0.66; P = 0.066), a finding they said "will deserve further investigation."

Benefits of Anti-HCV Treatment

Finally, Jose Del Valle and colleagues -- members of the same research collaboration as the study above -- looked at the effect of hepatitis C treatment on liver fibrosis as estimated by transient elastometry. Here, the investigators said that the non-invasive method might allow for easier monitoring of treatment outcomes compared with repeated sequential biopsies.

This prospective analysis included 50 HIV-HCV coinfected and 30 HCV monoinfected patients who started treatment with pegylated interferon alfa-2a (Pegasys) or alfa-2b (PegIntron) plus ribavirin at 2 Spanish hospitals; overall, 59% had hard-to-treat HCV genotypes 1 or 4.

All participants underwent transient elastometry at baseline and 24 weeks after the scheduled end of treatment (the point at which sustained virological response, or SVR, is determined). After completion of treatment, liver stiffness decreased from a median 7.6 kPa at baseline to 6.6 kPa (P = 0.001), with a median decrease of 0.50 kPa.

In a multivariate model that included genotype, type of pegylated interferon, and length of anti-HCV therapy, only higher baseline liver stiffness (P < 0.001) and achievement of SVR (P = 0.005) were significant predictors of decreased liver stiffness after treatment.

"Liver stiffness decreases significantly in HCV-infected patients treated with pegylated interferon and ribavirin, especially in those with higher liver stiffness prior to treatment and those who achieved sustained virological response," the investigators concluded. "Transient elastometry can be a useful tool for monitoring changes in liver fibrosis in HCV-infected patients receiving antiviral therapy."

3/13/09

References

M Ragni, M Nalesnik, L Qin, and others. Cytokines and Liver Fibrosis in HIV/HCV Co-infection. 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 830.

M Jain, E Neak, M Limerick, and others. Prevalence of Fibrosis Similar in Those with HIV/HCV Compared to HCV Alone. CROI 2009. Abstract 831.

P Tuma, L Martin-Carbonero, P Barreiro, and others. Liver Fibrosis Progression in HIV/HBV-co-infected Patients in the HAART Era. CROI 2009. Abstract 832.

J Pineda, C Tural, E Ortega, and others. Prevalence and Factors Associated with Significant Liver Fibrosis Measured by Transient Elastometry in HIV/HCV-co-infected Patients. CROI 2009. Abstract 824.

J Del Valle, A Camacho, J Mira, and others. Evolution of Liver Stiffness Evaluated by Transient Elastometry in Patients Treated with Pegylated Interferon + Ribavirin. CROI 2009. Abstract 826.