 |
|
|
| |
HCV Coinfection May Affect CD4 Response to HAART
|
| |
| |
|
|
| |
| |
 |
|
| |
| |
Reported by Jules Levin
"Virologic and immunologic response to antiretroviral therapy among HIV patients co-infected with hepatitis C virus (HCV) - analysis of a 6,401 HIV patient cohort" poster B3300
Cowling B.J. (1), Donnelly C.A. (1), Ghani A.C. (1), Rode R.A. (2), Da Silva B. (2),
Cameron D.W. (3), Cooper C.L. (3), Kwong G.P.S. (1), Van Sighem A.I. (4), de Wolf F. (1,4), Anderson R.M. (1)
(1) Department of Infectious Disease Epidemiology, Imperial College London, UK; (2) Abbott Laboratories, Abbott Park, Illinois, USA;
(3) Division of Infectious Diseases, Ottawa Hospital, Ottawa, Canada; (4) HIV Monitoring Foundation, Amsterdam, Netherlands.
BRIEF SUMMARY
This study examines over 6,000 patients in 3 large databases. HCV/HIV coinfection did not impaire viral suppression but did impaire cd4 increase in the time frame the study looked at: 3-12 months after starting HAART. STUDY AUTHORS CONCLUDED: While HIV-HCV co-infection does not seem to affect the time to viral suppression, CD4+ T-cell count recovery is attenuated in this patient population in the time frame evaluated.
THIS STUDY LOOKS AT:
Outcomes of interest are the patients' short-term virologic response and medium-term immunologic response including the affect of HCV coinfection. Short-term virologic response is defined as the time from initiation of ART to plasma HIV RNA viral load suppression (<500 copies/ml).
Medium-term immunologic response is defined as the mean CD4+ T-cell count between 3 and 12 months after ART initiation.
Of the 6,401 patients meeting the inclusion criteria, 947 (15%) were co-infected with HCV. The time to reaching <500 copies/ml HIV viral load was not impaired by having HCV coinfection: The time to viral suppression was not significantly associated with HCV co-infection status. However, having 'late-stage' HIV disease (AIDS, <200 cd4, >100,000 HIV viral load) did impair viral response to HAART.
Patients with no insurance/self-pay, their viral response to HAART was slower. Patients with private insurance had better viral response to HAART.
HCV status is significantly associated with lower mean CD4+ T-cell count over the period of 3 to 12 months after ART initiation.
Having 'late-stage' HIV disease impaired cd4 response and so did having category 1 insurance: Medicare, Medicaid, HMO, POS, Ryan White, other, compared to caregory 2 insurance-- Private, PPO, Clinical Study.
Background
With the advent of highly active antiretroviral therapy (HAART), there has been a significant reduction in HIV-associated morbidity and mortality in HIV-infected patients. As a result, concomitant infections such as hepatitis C virus (HCV) are increasingly contributing to observed morbidity and mortality in these patients.
In this study, we explore the effect of HIV-HCV co-infection on response to antiretroviral therapy (ART), using a larger patient cohort than those used in previous studies.
Data were analysed from 3 sources:
1. ATHENA database, which records information on 8,496 treated HIV patients in the Netherlands with 2,674 patients of known HCV status initiating ART between January 1998 and June 2003.
2. HIV Insight database, (including data from the CDC HIV Outpatient study) which records information on 13,929 HIV patients at treatment centres throughout the U.S. with 1,397 patients of known HCV status initiating ART between January 1998 and June 2003.
3. Plum Data Mining LLC database, which reports information on 5,000 HIV patients at treatment centres in the U.S. with 2,330 patients of known HCV status initiating ART between January 1998 and June 2003.
Data were standardised to include basic demographic and behavioural information, laboratory data (including CD4+ T-cell counts and viral load measurements), antiretroviral medication usage and clinical outcomes.
Methods
Patients are considered for analysis if they have had at least one serological HCV test and initiated ART between January 1998 and June 2003. Patients who have ever tested serologically positive for HCV are compared to those who have only tested serologically negative.
Outcomes of interest are the patients' short-term virologic response and medium-term immunologic response. Short-term virologic response is defined as the time from initiation of ART to plasma HIV RNA viral load suppression (<500 copies/ml). Patients are excluded from analysis if they have no viral load measurements following initiation of ART, or if they have viral load <500 copies/ml at baseline. Accelerated life regression models accounting for interval-censoring and assuming a lognormal distribution are used to analyse the effect of HCV serological status on short-term virological response, adjusting for baseline covariates.
Medium-term immunologic response is defined as the mean CD4+ T-cell count between 3 and 12 months after ART initiation. Patients are excluded from analysis if they do not have at least one CD4+ T-cell count in that period. Patients are categorised into three groups of mean counts (<=200; 201-500; >500 cells/mm3), and modelled using ordinal regression with a probit link function, adjusting for baseline covariates.
RESULTS
TIME TO VIRAL SUPPRESSION
Of the 6,401 patients meeting the inclusion criteria, 947 (15%) were co-infected with HCV.
A subset of 5,893 patients (92%) had at least one viral load measurement after ART initiation, and either unknown baseline viral load, or baseline viral load >500 copies/ml at baseline.
The time to viral suppression was not significantly associated with HCV co-infection status (acceleration factor: 1.10; 95% C.I.: 0.95-1.29; p=0.22).
However, year of ART initiation, baseline HIV disease stage and medical insurance group were significantly associated with the time to viral suppression. Patients with no insurance/self-pay or category 1 insurance, their viral response to HAART was slower. Patients with private insurance (categoery 2) had better viral response to HAART.
INSURANCE
"Category 1" insurance includes: Medicare, Medicaid, HMO, POS, Ryan White, Other.
"Category 2" insurance includes: Private, PPO, Clinical Study.
BASELINE HIV DISEASE defined: whether patient had 'early' or 'late' HIV disease-
"Baseline HIV disease stage" is a hierarchical binary classification:
1. 'late' if patient has an AIDS diagnosis before ART initiation;
2. 'early' ('late') if patient has CD4+ T-cell count >=200 (<200) cells/mm3 within 0-90 days before ART initiation;
3. 'early' ('late') if patient has HIV-RNA plasma viral load <=100,000 (>100,000) copies/ml within 0-90 days before ART initiation;
4. 'late' if patient has CD4+ T-cell count <200 cells/mm3 within 0-180 days after ART initiation.
Of note, among the group of non-HCV infected this study found that 20-30% started ART with late stage HIV disease. Among those with HCV/HIV coinfection 3-10% had baseline late-stage HIV disease when they started ART.
The following variables were found to be non-significant in regression analyses of both short-term virologic response and medium term immunologic response: age, gender, race, risk group, marital status, employment status and indicators of smoking, alcohol use and illicit drug use.
MEDIUM TERM IMMUNE RESPONSE (3-12 months)
A subset of 4,807 patients (75%) had at least one CD4+ T-cell count between 3 and 12 months after ART initiation.
HCV status is significantly associated with lower mean CD4+ T-cell count over the period of 3 to 12 months after ART initiation (odds ratio: 1.21; 95% C.I.: 1.10-1.33; p<0.001).
Year of ART initiation, baseline HIV disease stage and medical insurance group were also significantly associated with medium-term immunologic response. Having 'late-stage' HIV disease impaired cd4 response and so did having category 1 insurance.
Acknowledgements
We gratefully acknowledge the help of the physicians and patients who contributed data. The ATHENA database is supported by a grant from the Dutch Health Minister. The HIV Insight database includes data from the CDC HIV Outpatient Study and data collection is funded by Cerner Corporation. The AIDS Healthcare Foundation provides the data recorded in the Plum Data Mining LLC database. This work was supported by Abbott Laboratories.
|
| |
|
|
|
|
|
