In the natural history setting, heterogeneity in the rate of disease progression is characteristic of infection with HIV-1 . Determinants of this heterogeneity include age at seroconversion , plasma HIV-1 RNA concentration soon after seroconversion [3,4], viral characteristics [5–7], host genetics  and concomitant infections or neoplastic complications . The immunologic and virologic response to antiretroviral therapy measured by the ability to suppress viral replication, time to suppression, durability of suppression and level of immune reconstitution achieved with suppression, also varies at the individual  and population  levels. Factors potentially contributing to this variation include the same ones that affect the natural history of infection plus antiretroviral treatment experience, the degree of immunodeficiency, level of plasma HIV-1 RNA when therapy is initiated, antiretroviral drug resistance and adherence to the therapeutic regimen [10,12–16].
Prospective clinical trials are generally designed to determine the efficacy (treatment differences between individuals in a controlled setting) rather than the effectiveness (treatment differences among individuals in the population) of therapeutic interventions and, therefore, enroll relatively homogenous populations in terms of disease stage and treatment experience . A major strength of clinical trials is the close monitoring of therapies received that are key for the determination of their efficacy. Observational studies better represent the ‘real world’ experience and consequently can provide insight into the heterogeneity of response to highly active antiretroviral therapies that will be more relevant to clinical practice.
An increasing proportion of participants in the Multicenter AIDS Cohort Study (MACS) have initiated antiretroviral therapy since zidovudine became available in 1987. Currently, the majority of the men under treatment in the MACS are receiving highly active antiretroviral therapy (HAART) . Participants vary as to antiretroviral therapy experience and stage of immunodeficiency when HAART was initiated. The purpose of the present study was to evaluate the role of treatment experience, disease stage and genetic polymorphisms in CCR5, CCR5 promoter and CCR2 as determinants of individual responses to HAART.
Population and study design
The MACS is a prospective study of the natural and treated history of HIV-1 infection among homosexual men in the United States. A total of 4954 men were enrolled between April 1984 and March 1985 at the four study sites located in Baltimore, Chicago, Los Angeles and Pittsburgh. In 1987, enrollment was reopened and an additional 668 men, primarily African Americans, were recruited. Detailed descriptions of the MACS have been published [19,20] and only methods relevant to the present analyses are presented. The institutional review board at each of the participating centers approved the MACS study protocols and informed consent was obtained from all participants.
The participants return every 6 months for study visits which include a detailed interview, a physical examination, quality-of-life assessments and collection of blood for concomitant laboratory testing and storage in both local and national repositories. Positive enzyme-linked immunoabsorbent assays (ELISA) with confirmatory Western blots were used to determine HIV-1 seropositivity. T-lymphocyte subsets were measured by each MACS center using a standardized flow cytometry protocol [21,22]. Starting at visit 27 (April 1997–September 1997), levels of HIV-1 RNA were measured using the standard reverse transcriptase (RT)-polymerase chain reaction (PCR) assay (Roche Diagnostics, Nutley, New Jersey, USA) . Retrospective testing of stored plasma was performed for selected men using the same assay according to published methods .
Use of antiretroviral therapy was ascertained from detailed questions in the interview. Data collected on the use of each antiretroviral was summarized to define the initiation of HAART and the history of antiretroviral therapy before HAART. HAART was defined according to the US Department of Health and Human Services guidelines  as: (a) two or more nucleoside reverse transcriptase inhibitors (NRTIs) in combination with at least one protease inhibitor (PI) or one non-nucleoside reverse transcriptase inhibitor (NNRTI) (constituting 93% of initial HAART regimens); (b) one NRTI in combination with at least one PI and at least one NNRTI (5%); (c) a regimen containing ritonavir and saquinavir in combination with one NRTI and no NNRTIs (2%); and (d) an abacavir-containing regimen of three or more NRTIs without PIs or NNRTIs (< 0.5%). Combinations of zidovudine and stavudine with either a PI or NNRTI were not considered HAART.
Trends in the use of HAART by this cohort reflect the timing of drug availability and treatment recommendations in the United States ; HAART became available in 1995, 10 to 11 years after study entry. The collection of adherence data was initiated at visit 30 (October 1998–March 1999); adherence levels were validated with HIV-1 RNA levels. Using these data, 78% of antiretroviral therapy users were shown to be 100% compliant to their regimens . In addition, less than 18% of those initiating HAART in the MACS subsequently discontinued using a HAART regimen, although many changed the composition of the regimen.
A nested cohort study was used to examine the individual responses to HAART. The visit at which HAART use was first reported was denoted as the index visit. Initiation of HAART was defined as the midpoint between the index visit and prior visit. Participants reporting HAART, who contributed at least five out of 10 visits during the 5 years prior to the index visit and who had CD4 cell measurements before, at and after the index visit comprised the cohort for the analysis.
Classification of prior antiretroviral therapy use
Antiretrovirals used during the 5 years (10 visits) prior to the index visit were compared with the antiretroviral therapies contained in the initial HAART regimen. Men were classified as either having or not having had prior experience with any of the antiretrovirals subsequently contained in their initial HAART regimen. Those who had prior experience were further divided into those who had reported antiretroviral therapy use at each visit during the past 5 years (`consistent users'), and those for whom antiretroviral therapy had not been consistently reported (`intermittent users'). Among the intermittent users, the median percentage of visits at which antiretroviral therapy was reported during the 5 years prior to HAART was 50%, with interquartile range 22 to 80%. Thus, the three comparative groups of prior antiretroviral therapy experience were treatment naive, intermittently experienced and consistently experienced. Figure 1 shows the algorithm used to define these exposure groups.
Other potential predictors
Age at the time of first HAART report (< 45 or ≥ 45 years) and race (white or non-white) were assessed for their effects on response to the initial HAART regimen after adjusting for prior antiretroviral therapy use. The CCR5, CCR5 promoter and CCR2 genotypic polymorphisms were examined as independent predictors of immunologic and virologic responses. Each of the host genotypes was classified as either protective or non-protective according to the published effects of these genes on the natural progression of HIV-1 [25–28]. The protective genotypes were defined for CCR5 by CCR5 −+/Δ32, for the CCR5 promoter by G/G or G/A, and for CCR2 by CCR2 −+/64I or 64I/64I. The non-protective genotype for CCR5 was defined by +/+, for the CCR5 promoter by A/A and for CCR2 by +/+.
Outcomes: short- and long-term changes in CD4 cell counts and HIV-1 RNA levels
Changes in CD4 cell counts and suppression of plasma HIV-1 RNA levels to below the limit of assay detection (< 400 copies/ml) were used to measure the short- and long-term effects of HAART. The short-term CD4 cell response was defined as the CD4 cell count change from baseline (obtained within 6 months prior to HAART initiation) to the first visit (approximately 3 months) after initiating HAART. The long-term CD4 cell response was determined by the annual linear cell count change from baseline, during the period between 3 and 33 months after HAART initiation (i.e., after but not including this first visit).
Short-term HIV-1 RNA response was defined as whether or not plasma HIV-1 RNA was reduced to below 400 copies/ml, approximately 3 months after HAART initiation. Long-term HIV-1 RNA response was achieved by the maintenance of HIV-1 RNA levels below the limit of detection. Maintenance was satisfied if every HIV-1 RNA measurement following the first undetectable level was also undetectable; the first undetectable level could occur at any time after HAART initiation, however at least two undetectable measurements were required for qualification as maintenance.
At least two visits were required to measure the long-term responses. However, 75% of the participants contributed at least four viral load measurements and CD4 cell counts following initiation; 50% contributed at least five such measurements.
CD4 cell responses were assessed according to prior antiretroviral therapy categories, overall and stratified by baseline CD4 levels (< 200, 200–400, > 400 × 106 cells/l). The effect of prior antiretroviral therapy on the short-term CD4 cell response was tested using the non-parametric Wilcoxon rank sum statistic. The long-term CD4 cell response was estimated using multivariate random effects linear models in which the slopes and intercepts were allowed to vary randomly for each participant. This model adjusts for the within-subject correlation encountered in the analysis of longitudinal data. The random effects models were implemented using the SAS mixed procedure (SAS Institute, Cary, North Carolina, USA). The effects of age and race on short- and long-term CD4 cell responses were assessed with multivariate linear regression and random effects models, respectively, adjusted for prior therapy and stratified by baseline CD4 level. For the CCR5, CCR5 promoter and CCR2 polymorphisms, the short- and long-term CD4 cell responses were assessed with the same models as age and race, but only according to baseline CD4 cell count. Each of these host characteristics was analyzed separately.
HIV-1 RNA responses also were assessed according to prior antiretroviral therapy categories both overall and stratified by baseline CD4 levels. Differences in HIV-1 RNA response between the prior therapy categories were compared using χ2 tests. The effects of age, race and the genetic polymorphisms were assessed with separate logistic regression models, stratified by baseline CD4 cell counts. Age and race were also adjusted for prior therapy categories.
Between October 1995 and September 1999, a total of 645 MACS participants initiated HAART. As shown in Figure 2, men who initiated HAART had significantly lower baseline CD4 cell counts than those who did not initiate HAART. The median CD4 cell count among the initiators significantly increased up to July 1998 (P < 0.01). Although only a small number of individuals subsequently started HAART in the last two periods, these participants did so at lower CD4 cell counts. The overall increase of CD4 cell count in the non-HAART initiators was due to the fact that those with lower CD4 cell counts were selected to begin HAART (i.e., a consequence of the selection by indication).
The reversal of declining CD4 cell counts following the initiation of HAART is documented in Figure 3. The distribution of individual CD4 cell count slopes over a 5-year span before and up to 4 years after starting HAART are displayed for 497 of the 645 men who had sufficient data to estimate the CD4 cell slope both before and after HAART initiation. Prior to HAART, a median change of −44 × 106 CD4 cells/l per year was observed which increased significantly to +61 × 106 CD4 cells/l per year after HAART (P < 0.001). As shown, the immunologic response was heterogeneous in this group of HAART users.
Among the 645 HAART initiators, 397 participants satisfied the inclusion criteria for this study. These individuals had a median baseline CD4 cell count of 278 × 106 cells/l. Of these 397 men, 392 also had plasma HIV-1 RNA measurements; approximately 10% of the 392 had HIV-1 RNA levels below 400 copies/ml before HAART.
Clinical progression was uncommon among the men included in this analysis. A total of 316 men were free of AIDS (i.e., no opportunistic infections or malignancies) when they initiated HAART. Among these participants, seven (2.2%) either developed AIDS or died during the period of the study. None of the remaining 81 men with a clinical diagnosis of AIDS prior to starting HAART died during the study. It must be noted, however, that men were required to have CD4 cell counts measured at the time of initiation for inclusion in the analysis. This requirement may have excluded the men with advanced immunodeficiency who no longer had CD4 cell counts determined routinely. Among those included in the analysis, the median CD4 cell counts at the time of HAART initiation were 114, 299 and 585 × 106 cells/l for the < 200, 200–400 and > 400 CD4 groups, respectively.
Short-term response to HAART
As shown in Table 1, the overall short-term increases in CD4 cell counts were significantly (P < 0.05) larger for the naive (median = 82 × 106 cells/l) compared with the intermittently (56 × 106 cells/l) and consistently experienced users (29 × 106 cells/l). As treatment history may be driven by stage of disease, baseline CD4 cell count was used to control for this selection by indication. This adjustment also was necessary because those initiating earliest had the lowest CD4 cell counts (Figure 2) and contributed more data following initiation. The treatment-naive users significantly benefited from their initial HAART regimen in all baseline CD4 cell strata (P < 0.01). However, for the two experienced user groups, significant CD4 cell response was only observed among those starting with ≤ 400 × 106 CD4 cells/l. The naive users’ response was significantly higher than the consistently experienced users (P < 0.05) in the lowest and highest baseline CD4 cell count strata. In the highest stratum, those who were intermittently experienced also had a better CD4 cell response than the consistently experienced group. However, for those with a baseline CD4 cell count between 200 and 400 × 106 cells/l, the magnitude of short-term CD4 cell response was indistinguishable among the therapy categories.
As shown in Table 2, a significantly (P < 0.01) greater proportion of the treatment-naive users (72%) had HIV-1 RNA levels below 400 copies/ml after 3 months of HAART use, compared with the intermittently (43%) and consistently experienced users (46%). Although more of the naive users achieved short-term undetectable RNA levels across each baseline CD4 cell strata, these differences were not always statistically significant. Overall, the short-term CD4 cell and HIV-1 RNA responses in the two experienced groups were statistically indistinguishable. Furthermore, after adjusting for the short-term HIV-1 RNA response, prior treatment experience no longer significantly affected the short-term CD4 cell response, with the exception of those with > 400 × 106 CD4 cells/l at baseline. In this group, the consistently experienced users still exhibited a significantly worse response compared with the naive group.
After accounting for prior antiretroviral therapy experience, the effects by race were neither significant nor consistent across the CD4 cell strata. However, among men with > 400 × 106 CD4 cells/l, older age significantly (P = 0.044) decreased the short-term CD4 cell response such that men younger than 45 years had an average gain of 97 more cells after initiation than did older men. Although older men had an increased odds of having undetectable HIV-1 RNA levels in the CD4 cell strata, these differences were minimal and statistically not significant.
Long-term response to HAART
The overall, long-term mean CD4 cell changes per year were 47, 32, and 36 × 106 cells/l for the treatment-naive, intermittently experienced and consistently experienced users, respectively (Table 3). However, these long-term gains in CD4 cell count were only significant among the naive and intermittently experienced users starting HAART with ≤ 400 × 106 CD4 cells/l and among the consistently experienced users with baseline CD4 cell count < 200 × 106 cells/l. There were no significant differences in long-term CD4 cell count change by prior antiretroviral therapy experience.
As shown in Table 4, the proportion maintaining plasma HIV-1 RNA concentration to < 400 copies/ml was consistently higher in the therapy-naive group, although these differences were not always statistically significant. With the exception of those with CD4 cell count > 400 × 106 cells/l, the two experienced groups did not differ in the proportion maintaining undetectable HIV-1 RNA levels, the average being 30% among those with baseline CD4 cells < 200 × 106 cells/l and 41% among those with CD4 cells between 200 and 400 × 106 cells/l.
As in the short-term, race had no further effect on long-term CD4 cell response. Although more of the white men had a sustained HIV-1 RNA response (odds ratios between 1.3–2.8 across CD4 categories), the differences were not statistically significant. Unlike the short-term response, age did not significantly affect long-term HIV-1 RNA nor CD4 cell response. In fact, among those starting with > 400 × 106 CD4 cells/l, older men continued to increase their CD4 cells over time by an average of 50.4 × 106 cells/l per year more than younger men (P = 0.07).
Effect of host genotype on short- and long-term responses to HAART
For the short-term CD4 cell response (Table 5), the differences between participants carrying protective versus non-protective polymorphisms were inconsistent. Overall, there was a lack of significant findings supporting the hypothesis that having a protective genetic marker for disease progression is associated with better CD4 cell response to HAART in the short term. However, for those with baseline CD4 cell count > 400 × 106 cells/l and having the protective CCR5 genotype, the short-term CD4 cell response was 14 times higher than among those with the corresponding non-protective polymorphisms (P = 0.009). For the long-term CD4 cell response (Table 6), those with the Δ32 CCR5 genotype had consistently better CD4 cell response than those who were homozygous wild type, although these differences were not statistically significant (P > 0.05). Being heterozygous for the Δ32 deletion also was associated with better short-term virologic response only (odds ratios ranging from 2.5 to 3.3 across CD4 cell strata). There were no significant associations (P > 0.20) between the other genetic polymorphisms with either short- or long-term virologic response.
In this report, it has been documented that individuals with low CD4 cell counts were selected to initiate HAART. Differences between those who had initiated and those who were eligible, but did not initiate, attenuated over time up to July 1998 (Figure 2). However, during the last two periods in 1999, those selected to receive HAART had significantly lower CD4 cell counts.
Overall, CD4 cells increased following the use of HAART as shown in this study and by others [15,29,30]. However, this response was not uniform. History of antiretroviral therapy use before HAART initiation partially explains the heterogeneity whereby those who were treatment naive demonstrated the greatest CD4 cell response within a short time after starting therapy. Although one may speculate that these individuals may have been more immune-compromised prior to therapy and that this was related to subsequent response, their significantly larger response was observed across all baseline CD4 cell groups. However among experienced users, CD4 cell count at the time of initiating HAART modified subsequent response, such that those with lower CD4 cell counts prior to HAART demonstrated the best short-term response. Most of these treatment effects on immune reconstitution may be attributed to differences in HIV-1 RNA suppression, except among those initiating HAART with high CD4 cell counts. In this group, the CD4 cell response remained significantly worse among consistently experienced users when HIV-1 RNA was added to the model. The effect of age on short-term CD4 cell response also was only observed among those with > 400 × 106 CD4 cells/l at baseline. Only those with lower baseline CD4 cell counts had significant CD4 cell gains beyond 3 months after starting HAART.
It is necessary to adjust for CD4 cell level prior to starting HAART when evaluating independent determinants of immunologic and virologic response. Participants with lower CD4 cell counts generally initiated HAART earlier, had greater increases in CD4 cells and had the longest follow-up times. In addition, those with a history of prior therapy were more likely to be at a later stage of HIV disease; overall, the median baseline CD4 cell count was 266 × 106 cells/l for those experienced with their HAART regimen (259 and 298 × 106 cells/l for the intermittently and consistently experienced groups, respectively) and 318 × 106 cells/l for the treatment-naive. This selection by indication has also been shown in other studies .
Previous studies have suggested that lower CD4 cell counts predict less robust increases in CD4 cell levels subsequent to HAART initiation [12–15] but did not address the effect modification by prior therapy. Many of the participants in these studies also had lower baseline cell counts at initiation than men included in our analysis. In this study, only among the naive group did those with the highest CD4 cell count have a slightly better response. In contrast, among the experienced users, the best CD4 cell response was observed in those with the lowest CD4 cell count. Another study of the MACS also showed that overall, those with lower CD4 cells at initiation realized greater increases in CD4 cell count after 2 years .
Treatment experience also impacted the durability of suppression of viral replication. A greater proportion of the treatment-naive men demonstrated HIV-1 RNA levels below 400 copies/ml after short- and long-term HAART use. These results are consistent with the inability to suppress virus that has become resistant to antiretroviral therapy as a response to either long-term or inconsistent use of prescribed regimens. Consistent with other studies [12–15], decreases in plasma HIV-1 RNA copy number were more common in those with higher baseline CD4 cell counts, but only among the naive and intermittently experienced groups. In the consistently experienced group, the number of CD4 cells at initiation did not affect viral response.
Other host characteristics such as age and genetics may play a greater role in the immediate immune reconstitution following use of HAART. Whereas the viral response to HAART was similar across age groups, the immediate host biologic response, defined by the change in CD4 cells within 3 months of initiation, was affected by age. This was particularly observed among those with higher CD4 cells at baseline, such that older men had the lowest T-cell response, consistent with the effect of age on thymic-reconstitution of immune cells [32–36]. Similarly, those with the Δ32 CCR5 genotype exhibited greater immediate CD4 cell response, but again, only among those starting with > 400 × 106 CD4 cells/l. The CCR5 genotypic positive influence on CD4 cell response is consistent with those presented by Valdez et al.  and Workman et al. .
Other studies have shown that race influences CD4 cell depletion  and survival in advanced HIV-1 infection . In this study, race did not significantly affect response, however only 13, 18 and 16% of the naive, intermittently experienced and consistently experienced users, respectively, were non-white. Therefore there may have been insufficient numbers to detect small differences due to race.
Given that age, CCR5 genotype and baseline CD4 cell count play a role in the immunologic, more so than viral, response, we propose a model for these relationships (Figure 4). Prior antiretroviral therapy experience affects the ability of HAART to suppress HIV-1, but host characteristics such as age, genetics and level of immune competence have a more direct role in the reconstitution of the immune system. For example, a younger individual who has more CD4 cells may regenerate naive cells more quickly than older individuals once virus is suppressed. The main effect of prior HIV-1 therapeutic experience is to modify the ability of HAART to suppress virus. Additional factors that may affect virologic response to HAART include the utilization modalities of the drugs themselves. For example, there have been temporal changes in the medications constituting the initial HAART regimen (decreased use of saquinavir and increased use of nevirapine). However, with the tremendous heterogeneity in the regimen defining HAART, and thus the small numbers using specific regimens, we have presented the overall, robust responses to HAART. Other factors, including hepatitis infection, modify the immune reconstitution following HAART. Indeed, a recent report documents that HIV-1-hepatitis C virus co-infected patients show a lesser CD4 response to HAART than patients who are not co-infected with hepatitis C virus . In our cohort, the role of hepatitis C virus would be minimal due to the low rate of co-infection .
These results suggest that multiple factors are important in determining the CD4 cell response and durability of suppression of viral replication after initiation of HAART. Further studies including the effects of adherence, level of viral replication prior to therapy, initial prevalence of resistance and selection of resistant mutants are underway in this cohort. Longer follow-up of the cohort will permit assessing the durability and clinical relevance of these initial changes.
1. Muñoz A, Sabin CA, Phillips AN. The incubation period of AIDS. AIDS 1997, 11 (suppl A): S69 –S76.
2. Muñoz A, Xu J. Models for the incubation of AIDS and variations according to age and period. Stat Med 1996, 15: 2459 –2473.
3. Lyles RH, Muñoz A, Yamashita TE. et al
. Natural history of human immunodeficiency virus type 1 viremia after seroconversion and proximal to AIDS in a large cohort of homosexual men. J Infect Dis 2000, 181: 872 –880.
4. Mellors JW, Muñoz A, Giorgi JV. et al
. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1
infection. Ann Intern Med 1997, 126: 946 –954.
5. Shankarappa R, Gupta P, Learn GH Jr. et al
. Evolution of human immunodeficiency virus type 1 envelope sequences in infected individuals with differing disease progression profiles. Virology 1998, 241: 251 –259.
6. Wolinsky SM, Korber BTM, Neumann AU. et al
. Adaptive evolution of human immunodeficiency virus-type 1 during the natural course of infection. Science 1996, 272: 537 –542.
7. Deacon NJ, Tsykin A, Solomon A. et al
. Genomic structure of an attenuated quasi species of HIV-1
from a blood transfusion donor and recipients. Science 1995, 270: 988 –991.
8. Kaslow RA, Carrington M, Apple R. et al
. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1
infection. Nat Med 1996, 2: 405 –411.
9. Donovan RM, Bush CE, Markowitz NP, Baxa DM, Saravolatz LD. Changes in virus load markers during AIDS-associated opportunistic diseases in human immunodeficiency virus-infected persons. J Infect Dis 1996, 174: 401 –403.
10. Hirschel B, Opravil M. The year in review: antiretroviral treatment. AIDS 1999, 13 (suppl A): S177 –S187.
11. Tarwater PM, Margolick JB, Jin J, et al
. Maintenance of increased CD4 T cell counts up to 3 ½ years after initiation of potent antiretroviral therapy.J Acquir Immune Defic Syndr
2001, [provisionally accepted for publication].
12. Ledergerber B, Egger M, Opravil M. et al
. Clinical progression and virological failure on highly active antiretroviral therapy
patients: a prospective cohort study. Lancet 1999, 353: 863 –868.
13. Clough LA, D'Agata E, Raffanti S, Haas DW. Factors that predict incomplete virological response to protease inhibitor-based antiretroviral therapy. Clin Infect Dis 1999, 29: 75 –81.
14. Wit FW, van Leeuwen R, Weverling GJ. et al
. Outcome and predictors of failure of highly active antiretroviral therapy
: One-year follow-up of a cohort of human immunodeficiency virus type 1-infected persons. J Infect Dis 1999, 179: 790 –798.
15. Casado JL, Perez-Elías MJ, Antela A. et al
. Predictors of long-term response to protease inhibitor therapy in a cohort of HIV-infected patients. AIDS 1998, 12: F131 –F135.
16. DeGruttola V, Dix L, D'Aquila R. et al
. The relation between baseline HIV drug resistance and response to antiretroviral therapy: re-analysis of retrospective and prospective studies using a standardized data analysis plan. Antivir Ther 2000, 5 (1): 41 –48.
17. Muñoz A, Gange SJ, Jacobson LP. Distinguishing efficacy, individual effectiveness and population effectiveness of therapies. AIDS 2000, 14: 754 –756.
18. Detels R, Muñoz A, McFarlane G. et al
. Effectiveness of potent antiretroviral therapy on time to AIDS and death in men with known HIV infection duration. JAMA 1998, 280: 1497 –1503.
19. Kaslow RA, Ostrow DG, Detels R, Phair JP, Polk BF, Rinaldo CR Jr. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. Am J Epidemiol 1987, 126: 310 –318.
20. Dudley J, Jin S, Hoover D, Metz S, Thackeray R, Chmiel J. The Multicenter AIDS Cohort Study: Retention after 9½ years. Am J Epidemiol 1995, 142: 323 –330.
21. Giorgi JV, Cheng HL, Margolick JB. et al
. Quality control in the flow cytometric measurement of T-lymphocyte subsets: The Multicenter AIDS Cohort Study experience. Clin Immunol Immunopathol 1990, 55: 173 –186.
22. Schenker EL, Hultin LE, Bauer KD, Ferbas J, Margolick JB, Giorgi JV. Evaluation of a dual-color flow cytometry immunophenotyping panel in a multicenter quality assurance program. Cytometry 1993, 14: 307 –317.
23. Centers for Disease Control and Prevention. Report of the NIH panel to define principles of therapy of HIV infection and guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. MMWR CDC Surveill Summ 1998, 47 (RR-5): 1 –82.
24. Kleeberger CA, Phair J, Strathdee SA, Detels R, Kingsley L, Jacobson LP. Determinants of heterogeneous adherence to HIV-antiretroviral therapies in the Multicenter AIDS Cohort Study. J Acquir Immune Defic Syndr 2001, 26: 82 –92.
25. Dean M, Carrington M, Winkler C. et al
. Genetic restriction of HIV-1
infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Science 1996, 273: 1856 –1862.
26. Huang Y, Paxton WA, Wolinsky SM. et al
. The role of a mutant CCR
5 allele in HIV-1
transmission and disease progression. Nat Med 1996, 2: 1240 –1243.
27. Smith MW, Dean M, Carrington M. et al
. Contrasting genetic influence of CCR
2 and CCR
5 variants on HIV-1
infection and disease progression. Science 1997, 277: 959 –965.
28. Martin MP, Dean M, Smith MW. et al
. Genetic acceleration of AIDS progression by a promoter variant of CCR
5. Science 1998, 282: 1907 –1911.
29. Schapiro JM, Winters MA, Stewart F. et al
. The effect of high-dose saquinavir on viral load and CD4+ T-cell counts in HIV-infected patients. Ann Intern Med 1996, 124: 1039 –1050.
30. Hammer SM, Squires KE, Hughes MD. et al
. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med 1997, 337: 725 –733.
31. Ahdieh L, Gange S, Greenblatt R. et al
. Selection by indication of potent antiretroviral therapy usage in a large cohort of HIV-infected women. Am J Epidemiol 2000, 152: 923 –933.
32. Mackall CI, Gress RE. Thymic aging and T-cell regeneration. Immunol Rev 1997, 160: 91 –102.
33. Haynes BF, Hale LP, Weinhold KJ. et al
. Analysis of the adult thymus in reconstitution of T lymphocytes in HIV-1
infection. J Clin Invest 1999, 103: 453 –460.
34. Zhang L, Lewin SR, Markowitz M. et al
. Measuring recent thymic emigrants in blood of normal and HIV-1
-infected individuals before and after effective therapy. J Exp Med 1999, 190: 725 –732.
35. Jamieson BD, Douek DC, Killian S. et al
. Generation of functional thymocytes in the human adult. Immunity 1999, 10: 569 –575.
36. Douek DC, McFarland RD, Keiser PH. et al
. Changes in thymic function with age and during the treatment of HIV infection. Nature 1998, 396: 690 –695.
37. Valdez H, Purvis SF, Lederman MM, Fillingame M, Zimmerman PA. Association of the CCR5
Δ32 mutation with improved response to antiretroviral therapy. JAMA 1999, 282: 734. 734.
38. Workman C, Whittaker W, Forrester J, Dyer W, Sullivan J. Association of theCCR5Δ32 mutation with improved response to antiretroviral therapy commenced in primary HIV infection.Seventh Conference on Retroviruses and Opportunistic Infections
. San Francisco, January 2000 [abstract 570].
39. Easterbrook PJ, Farzadegan H, Hoover DR. et al
. Racial differences in rate of CD4 decline in HIV-1
-infected homosexual men. AIDS 1996, 10: 1147 –1155.
40. Apolonio EG, Hoover DR, He Y. et al
. Prognostic factors in human immunodeficiency virus-positive patients with a CD4+ lymphocyte count < 50/μl. J Infect Dis 1995, 171: 829 –836.
41. Greub B, Ledergerber B, Battegay M. et al
. Clinical progression, survival and immune recovery during antiretroviral therapy in patients with HIV-1
and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet 2000, 356: 1800 –1805.
42. Donahue JG, Nelson KE, Muñoz A. et al
. Antibody to hepatitis C virus among cardiac surgery patients, homosexual men, and intravenous drug users in Baltimore, Maryland. Am J Epidemiol 1991, 134: 1206 –1211.
The Multicenter AIDS Cohort Study (MACS) includes the following: Baltimore: The Johns Hopkins University School of Hygiene and Public Health: Joseph B. Margolick, Principal Investigator; Haroutune Armenian, Homayoon Farzadegan, Nancy Kass, Justin McArthur, Steffanie Strathdee, Ellen Taylor. Chicago: Howard Brown Health Center and Northwestern University Medical School: John P. Phair, Principal Investigator; Joan S. Chmiel, Bruce Cohen, Maurice O'Gorman, Daina Variakojis, Steven M. Wolinsky. Los Angeles: University of California, UCLA Schools of Public Health and Medicine: Roger Detels and Beth Jamieson, Principal Investigators; Barbara R. Visscher, Co-Principal Investigator; Eric G. Bing, John L. Fahey, John Ferbas, Otoniel Martínez-Maza, Eric N. Miller, Hal Morgenstern, Parunag Nishanian, John Oishi, Paul Satz, Elyse Singer, Jeremy Taylor, Harry Vinters, Dorothy Wiley, Stephen Young. Pittsburgh: University of Pittsburgh, Graduate School of Public Health: Charles R. Rinaldo, Principal Investigator; Lawrence Kingsley, Co-Principal Investigator; James T. Becker, Phalguni Gupta, Lawrence Kingsley, John Mellors, Sharon Riddler, Anthony Silvestre. Data Coordinating Center: The Johns Hopkins University School of Hygiene and Public Health: Alvaro Muñoz, Principal Investigator; Lisa P. Jacobson, Co-Principal Investigator; Linda Ahdieh, Stephen Cole, Stephen Gange, Cynthia Kleeberger, Steven Piantadosi, Ellen Smit, Sol Su, Patrick Tarwater, Traci E. Yamashita. NIH: National Institute of Allergy and Infectious Diseases: Paolo Miotti, Epidemiology Branch Chief; Carolyn Williams, Project Officer. National Cancer Institute: Sandra Melnick.