 |
|
|
| |
Do HCV Genotype and Load Affect Mortality With HIV?
|
| |
| |
11th European AIDS Conference
October 24-27, 2007
Madrid
Mark Mascolini
Neither hepatitis C virus (HCV) genotype nor HCV load affected the risk of death from any cause in people coinfected with HIV, according to results of a 2263-person EuroSIDA cohort analysis [1]. In the same cohort the risk of liver-related death did not vary by HCV genotype, HCV load, or HCV clearance status, but statistical power was weaker for this second analysis because it included fewer deaths.
An earlier EuroSIDA study disclosed no higher risk of progression to AIDS or death in HIV-infected people with HCV infection established by HCV antibodies alone [2]. Neither did HCV coinfection influence virologic or CD4 responses to antiretrovirals. But HIV/HCV-coinfected people did have a higher risk of death from liver disease.
To see whether HCV load or genotype affects liver-related death or death from any cause in EuroSIDA, HCV/HIV expert Jurgen Rockstroh studied all 1677 cohort members with a known HCV genotype. The analysis also included 586 patients with an HCV load below 615 IU/mL but still positive for HCV antibodies--in other words these people had cleared their HCV infection.
From these two coinfected groups, 340 (15%) died during 9048 person-years of follow-up, and 91 of them died from liver disease. After statistical adjustment for an array of potentially confounding variables, the all-cause death rate in people with hard-to-treat genotype 1 did not differ from death rates among people with HCV genotypes 2, 3, or 4, or from the rate in people who had cleared HCV. Adjusted death incidence rate ratios measured 1.13 per 100 person-years for genotype 2, 1.09 for genotype 3, 1.04 for genotype 4, and 0.80 for HCV clearance compared with a reference of 1.00 for genotype 1.
EuroSIDA statisticians found that for every 10-fold higher HCV load, the risk of all-cause mortality rose 23% (P = 0.01). But after adjustment for confounding variables the all-cause death risk was only 6% higher with every 10-fold higher HCV load, and this increase lacked statistical significance.
A univariate statistical analysis that did not account for possible confounding factors and included only people with a detectable HCV load confirmed significantly faster progression to death through 84 months in people with an HCV load above 500,000 copies than in those with lower loads (P = 0.0022). This analysis started with 667 people with a sub-500,000 load on their first HCV load test and 705 with a load above 500,000 on their first test. At the end of follow-up respective numbers being tracked in each group were 151 and 181. At that point about 27% in the high viral load group versus 15% in the low viral load group had died. (from jules: higher HCV viral load could be a surrogate for longer duration of HCV-infection or other factors such as alcohol use and ART-related hepatotoxicity).
Rockstroh did not find a difference in liver-related death rates based on HCV genotype or HCV clearance in an analysis that factored in possibly confounding variables. But statistical power in this analysis was weaker than in the all-cause death analysis because fewer people died from liver disease.
The list on confounding factors in these analyses included gender, risk exposure group, region of Europe, coinfection with hepatitis B virus (HBV), race, AIDS diagnosis, age, CD4 count at genotyping, treatment regimen, and date of HCV genotype testing. But Rockstroh observed that statisticians could not factor in other potentially important variables, such as alcohol consumption, because EuroSIDA does not collect data on those factors. He also agreed with a questioner who noted that it generally takes HCV a long time to kill someone, so longer analyses may begin to show differences in liver-related deaths based on genotype or HCV load.
A separate 2263-person EuroSIDA analysis found that three quarters of cohort members with detectable HCV antibody have active HCV replication [3]. Detectable HCV viremia proved more likely in injecting drug users and in people without chronic HBV infection. Just over half of EuroSIDA members with active HCV replication had HCV genotype 1, which correlated with higher HCV loads and male gender.
References
1. Rockstroh JK, Mocroft A, Reiss P, et al. Does hepatitis C virus (HCV) genotype or viral load predict the risk of all-cause mortality of HIV/HCV coinfection? European AIDS Conference. October 24-27, 2007. Madrid. Abstract P8/2.
2. Rockstroh JK, Mocroft A, Soriano V, et al. Influence of hepatitis C virus infection on HIV-1 disease progression and response to highly active antiretroviral therapy. J Infect Dis. 2005;192:992-1002.
3. Soriano V, Mocroft A, Ledergerber G, et al. Epidemiological and virological characteristics of chronic hepatitis C (HCV) within the EuroSIDA observational cohort. European AIDS Conference. October 24-27, 2007. Madrid. Abstract P8/1.
|
| |
|
|
|
|
|
