| |
Hepatitis C's effect on HIV disease remains elusive despite three new studies; Coinfection Death Rate
|
| |
| |
Crabb, Charlene
AIDS: Volume 20(3) 14 February 2006 p N1-N2
Received 9 September, 2005
Accepted 23 September, 2005
The effect of hepatitis C virus (HCV) on HIV-1 disease progression is anything but clear. Some research has found evidence of faster HIV-1 disease progression, or at least poorer CD4 cell count response, in coinfected individuals receiving highly active antiretroviral therapy (HAART). Other studies have not.
Recently three independent research teams in Europe and the United States published new results that continue the controversy. However, researchers from the groups told AIDS that the latest findings, as well as those from earlier ones, might not be as disparate as they seem.
I think every group sees more death events in coinfected individuals,says Jurgen Rockstroh of University Hospital Bonn in Germany, lead author of one of the new studies (J Infect Dis 2005, 192: 992-1002). It's only when you start separating AIDS-related, liver-related, and non-HIV related deaths that you start to see differences. If we are more clear on the death definition, maybe the results of the various trials aren't really that different.
With HAART extending the lives of HIV-infected individuals, HCV-related liver disease has emerged as an increasingly common cause of death among HIV patients. Liver disease causes 17% to 45% of in-hospital deaths among HIV-infected individuals in the West. In the United States, 15% to 30% of people infected with HIV are also infected with HCV, a virus largely associated with intravenous drug use in industrialized countries. Estimates of HCV coinfection in Europe range from more than 50% in Spain and Italy to less than 15% in Germany and Denmark.
Rockstroh and his co-authors evaluated data from 5957 participants with known HCV status in the EuroSIDA Study Group, which includes HIV-1 patients at 89 centers across Europe, as well as Israel and Argentina. One-third of the individuals were HCV seropositive. During follow-up, the patients experienced 917 new AIDS-defining illnesses, and there were 819 deaths, including 109 related to liver disease and 462 that were not HIV-related, such as suicide or drug overdoses.
When the researchers lumped together data on AIDS-defining illness and deaths by any cause, they found a significantly higher occurrence among coinfected individuals compared to individuals infected with HIV alone (incidence rate ratio, or IRR, 1.44; P < 0.0001). But that difference did not hold up when the analyses were adjusted for baseline factors including CD4 cell count and age (IRR, 0.97; P = 0.72), or time-dependent factors including initiation of HAART (IRR, 1.06; P = 0.50).
However, when the researchers looked at the incidence of AIDS-related illnesses and the different types of death separately, they found a significantly higher risk of death among HCV-positive individuals (IRR, 1.41; P = 0.0024 baseline adjusted; IRR, 1.80; P < 0.0001 time-dependent adjusted). The incidence of liver disease-related death was more than 10-fold greater among coinfected individuals (IRR, 11.71, P = 0.0001). But, surprisingly, they had a significantly lower incidence of AIDS-defining illnesses after adjustment.
In addition, among HCV-infected and HCV-noninfected individuals who initiated HAART, there was no difference in the percentage whose HIV RNA load dropped below 500 copies/ml, nor the time it took to reach that level. Similarly, the two groups of patients regained their CD4 cell count at comparable rates, whether measured as an increase of more than 50% or an increase of 50 cell/μl from baseline.
Rockstroh and his colleagues concluded that HCV coinfection markedly increased the risk of liver disease-related deaths, but it does not affect HIV-1 disease progress or an individual's virologic and immunologic response to HAART.
Meanwhile, another large study concluded that HCV coinfection leads to an increased risk of death (J Acquir Immune Defic Syndr 2005, 39:613-619). Veterans Health Administration researchers in Palo Alto, California, analyzed the data on deaths among 12 216 HIV-positive military veterans who received HAART. A total of 4668 patients, or 38%, were coinfected with HCV. During a follow-up period averaging 3.5 years, 2087 patients died. The researchers did not analyze for different causes of death or AIDS-related illness.
Proportionally more deaths occurred among coinfected veterans than those infected with HIV alone (22.2% versus 13.9%). Compared to their HIV-only counterparts, coinfected patients who received HAART had a higher risk of death (hazard ratio, 1.38). HCV-positive individuals who never received the therapy were also more likely to die (hazard ratio, 1.56).
The risk factors for death among the veterans are in exactly the same range as those reported in the 'any death' category of the EuroSIDA study, says Lisa Backus, lead author of the VA study. It's somewhere in that 1.4 to 1.8 range, she adds. It just depends on what you adjust for.
Backus says that some of the confusion between the VA and EuroSIDA studies, as well as others, lies in combining AIDS-related illnesses with death as an endpoint for analysis. The 1993 [Centers for Disease Control and Prevention's (CDC)] list of AIDS-defining illnesses is predominantly what gay men were manifesting more than 10 years ago, Backus says. This is a different epidemic now. AIDS-related opportunistic infections that an injection drug user might get aren't necessarily on the list, so it becomes a competing issue.
Like the EuroSIDA researchers, Backus and her team found a similar virologic response to HAART among HCV-coinfected and HIV-only individuals. About 80% of both groups of patients achieved an undetectable viral load at least once, and about 38% of individuals in both groups maintained good control of HIV replication. However, CD4 cell response was lower among the coinfected patients, who gained an average of 199 cells/μl compared to 239 cells/μl for HCV uninfected patients.
The third study, conducted by researchers at Chelsea and Westminster Hospital in London, found a significantly higher risk of progressing to an AIDS-related illness or a CD4 cell count below 200 cells/μl (hazard ratio, 1.52) among coinfected patients than HIV-only patients. However, CD4 cell count decreased at similar rates for both groups of patients. The study involved 2049 people who received HAART, 1467 of them were tested for HCV and 85 were positive for the virus (Clin Infect Dis 2005; 41:906-911).
Justin Stebbing, lead author of the Chelsea and Westminster study, suggests that discrepancies between the trio of new studies, as well as earlier studies, may be due to differences in demographics of populations studied, inclusion criteria and clinical endpoints. A cumulative meta-analysis would help sort the issue out, Stebbing says. The conflicting results of the studies done so far, he says, probably mean there isn't a massive effect of hepatitis C on HIV infection.
|
|
| |
| |
|
|
|
