|Home | About | Journals | Submit | Contact Us | Français|
Both antiretroviral therapy and the human coreceptor polymorphism CCR2-V64I slow progression of human immunodeficiency virus type 1 (HIV-1) disease. To examine the effect of V64I on disease progression in patients receiving therapy, we determined CCR2 genotypes in the Women’s Interagency HIV Study cohort. We studied 2047 HIV-1–infected women, most of whom initiated treatment during the study. No association was seen between CCR2 genotype and either disease progression or therapeutic response, suggesting that the benefits of treatment most likely overshadow the salutary effects of the V64I polymorphism.
Studies of untreated HIV-1–infected populations revealed that polymorphisms in the human genes encoding the chemokine receptors CCR5 and CCR2 can influence the rate of HIV-1 disease progression [1–3]. A variant of the gene encoding CCR2, V64I, which has been detected in all ethnic groups examined [1, 4], has been associated with delayed disease progression [1–5].
Initial investigations linking the CCR2-V64I allele to slow HIV-1 disease progression predominantly focused on white men studied before the era of HAART [1, 4]. Although HAART is also well known to prevent disease progression [6, 7], few studies have examined the influence of CCR2 genotypes on virologic and immunologic response in treated populations [5, 8, 9]. Questions remain as to whether the CCR2 polymorphism exerts a discernible, clinical effect in individuals who are receiving antiretroviral therapy (ART).
To address these issues, we determined the CCR2 genotypes of all 2606 subjects enrolled in theWomen’s Interagency HIV Study (WIHS), a multicenter, natural history study of HIV-1 infection in women in the United States . A total of 2047 WIHS participants were HIV-1 infected; of these, 70.6% received ART . The relationship of CCR2 to HIV infection status, ethnicity, transmission risk, clinical disease progression, and response to therapy was analyzed.
We studied all 2047 HIV-1–infected and 559 HIV-1–uninfected participants in the WIHS, a cohort reflecting the ethnic backgrounds and risk behaviors of groups among whom HIV-1 infection is spreading rapidly . The institutional review boards at each site and the New York State Department of Health (Albany) approved the investigation. Each woman provided informed consent. Sample collection, data analyses, and genotypic analyses have been described elsewhere [1, 10]. CCR2 genotypes were categorized as wild-type (+/+), heterozygous V64I (+/V64I), and homozygous V64I (V64I/V64I).
The association between CCR2-V64I genotype and HIV-1 infection was examined by using a logistic regression model and was reported as an OR with 95% CIs. The OR gives a measurement of the likelihood of infection among V64I heterozygotes and/or homozygotes, compared with individuals homozygous for wild-type CCR2. Analyses were performed by comparing these 3 groups overall and then by stratifying according to ethnicity and risk. Similar analyses were performed by comparing wild-type homozygotes with individuals carrying at least 1 copy of the V64I allele.
To evaluate disease progression in relation to CCR2 genotype, we studied women who were homozygous for the wild-type CCR5 allele (see Results), had CD4+ cell counts available during the study period, and did not enter the WIHS with a clinical diagnosis of AIDS. Kaplan-Meier estimates and log-rank statistics were calculated to evaluate the effect of CCR2 genotype from the time of study entry to first report of a clinical AIDS outcome. Previous studies of HIV-1 disease progression have shown that it is necessary to adjust for disease severity by accounting for each individual’s duration of infection . For a seroprevalent cohort like the WIHS, this is usually achieved by adjusting for clinical markers, such as CD4+ cell count and HIV-1 RNA load . Thus, we used a discrete time-survival model, with the time-dependent covariates of CD4+ cell count and HIV-1 RNA load used to adjust for disease severity at study entry. We similarly controlled for ART as a time-dependent covariate by stratifying study participants into 3 categories according to treatment (persons who did not receive ART [328 women], ART recipients [483 women], and HAART recipients [827 women]), as defined by National Institutes of Health guidelines. Because clinical and therapeutic data were collected longitudinally at 6-month intervals, we were also able to control for each of these potential confounders as time-varying exposure variables.
To address whether the CCR2 genotype affects response to therapy, we studied 901 women who initiated HAART and were observed for ≥3 consecutive visits after its initiation. Baseline values were defined using measurements from 1 year prior to the commencement of HAART. An individual was considered to have responded to HAART if the following criteria were satisfied at ≥1 of the follow-up visits:
For each model, the magnitude of the association between genotype and response to therapy was measured by using ORs.
After determining the CCR2 genotypes of all WIHS participants, we stratified the 2047 HIV-1–seropositive and 559 HIV-1–seronegative women according to ethnicity and transmission risk. Among the 2606 women, 86.5% carried the homozygous wild-type genotype, 12.5% were V64I heterozygotes, and 1.0% were V64I homozygotes. The frequency of the CCR2-V64I allele in the cohort was also calculated as described in table 1, and was 0.073 overall. Contrary to what has been observed in other studies [3–5, 9], the frequency of CCR2-V64I was only slightly less common among white subjects than among other ethnic groups (0.079 among African American subjects, 0.072 among Latino subjects, 0.055 among white subjects, and 0.069 among other groups).
Because CCR2-V64I is in complete negative linkage disequilibrium with CCR5-Δ32, a chemokine receptor mutation that affects HIV-1 transmission , additional analyses included only those women who were homozygous for the wild-type CCR5 allele (1940 subjects were seropositive and 513 were seronegative). The CCR2 genotypes and CCR2-V64I allelic frequencies for these 2453 WIHS participants are shown in table 1. Demographic characteristics and treatment histories of the women homozygous for the wild-type CCR5 allele are remarkably similar to those of the WIHS cohort as a whole . The frequency of the CCR2-V64I allele among these women was 0.074 overall and varied only marginally between the different ethnic groups.
We next determined whether the CCR2 polymorphisms influenced susceptibility to HIV-1 infection. HIV-1–infected and HIV-1–uninfected women were equally likely to have at least 1 copy of the CCR2-V64I allele (OR, 1.06; 95% CI, 0.79–1.41), indicating that the CCR2 genotype did not affect HIV-1 transmission (table 1). Stratification by ethnicity or transmission risk also failed to show a statistically significant effect of the V64I allele on HIV-1 transmission (table 1).
The influence of the CCR2 genotype on clinical HIV-1 disease progression was then examined in 1638 women, 80% of whom received ART during the study (see Methods). Because the CCR5Δ32 mutation also slows HIV-1 disease progression , these analyses were restricted to women who were homozygous for the CCR5 wild-type allele. Univariate and multivariate models of disease progression failed to detect any protective effect, a finding seen both when individuals homozygous or heterozygous for CCR2-V64I were analyzed as 2 separate groups and when they were combined into 1 group. Compared with individuals homozygous for wild-type CCR2, patients carrying at least 1 copy of the V64I allele had an equivalent risk of developing AIDS (OR, 1.0; 95% CI, 0.8–1.3). Kaplan-Meier estimates also failed to show any significant effect of the CCR2-V64I allele on progression to AIDS (figure 1). These results were observed for the entire WIHS population and for subsets of patients stratified by ethnicity, treatment, and transmission risk.
We investigated the effect of the CCR2 genotype on response to ART in 901 women. No significant influence of CCR2 genotype on response to HAART was found by using any of the 4 methods defining response (P > .05; table 2).
This examination of the CCR2 genotype in a large cohort of women in the United States, most of whom received ART, showed that the CCR2-V64I allele neither protected against HIV-1 acquisition nor altered the response to HAART. These data confirm and expand the findings of published smaller studies [5, 8, 9]. In contrast to other reports [3–5], however, our analysis of disease progression in a group in which 80% of subjects received ART found that the CCR2-V64I allele did not significantly delay the onset of HIV-1 disease. Post hoc analyses of power and sample size demonstrated that these studies included sufficient numbers of participants to detect differences in HIV-1 transmission and disease progression rates of as little as 5%–7%—equivalent to differences seen in other studies [3–5].
It is possible that the seroprevalent nature of the WIHS may explain our inability to detect the effect of V64I on disease progression. This effect may be evident only when individuals are observed from the time of seroconversion, a notion supported by a recent meta-analysis .
It is likely, however, that the extensive use of antiretroviral agents, including HAART, by the WIHS cohort during the follow-up period resulted in low rates of disease progression and of AIDS-defining events . Previous studies demonstrating the protective effect of the V64I allele were conducted before the era of widespread HAART use or were performed in countries where ART is not readily available [3–5]. Our results suggest that the effects of the V64I polymorphism in CCR2 are overshadowed by the power of HAART to delay or reverse disease progression.
We thank T. Moran, M. Shudt, and the Wadsworth Center Molecular Genetics Core Laboratory for oligonucleotide synthesis; S. Vermund for helpful discussions; and S. Beck for preparation of the manuscript. Data in this manuscript were collected by the Women’s Interagency HIV Study (WIHS) Collaborative Study Group with centers (principal investigators) at New York City/Bronx Consortium (Kathryn Anastos); Brooklyn, NY (Howard Minkoff); Washington DC Metropolitan Consortium (Mary Young); The Connie Wofsy Study Consortium of Northern California (Ruth Greenblatt); the Los Angeles County/Southern California Consortium (Alexandra Levine); the Chicago Consortium (Mardge Cohen); and the Data Coordinating Center (Johns Hopkins University School of Hygiene and Public Health, Baltimore, MD; Stephen J. Gange).
Financial support. National Institute of Allergy and Infectious Diseases, with supplemental funding from the National Cancer Institute, the National Institute of Child Health & Human Development, the National Institute on Drug Abuse, the National Institute of Craniofacial and Dental Research (grants U01-AI-35004, U01-AI-31834, U01-AI-34994, U01-AI-34989, U01-HD-32632, U01-AI-34993, U01-AI-42590, M01-RR00079, and M01-RR00083; and grant RO1-AI-42555 [to H.B.]), and a National Research Service Award (grant 1 F32 HD08478-01; to S.P.).
Conflict of interest. All authors: No conflict.