Background: HIV-1 infection is characterized by an inverted CD4/CD8 T-cell ratio, but the distribution of inversions over time after seroconversion and whether delay of inversion is associated with a favorable prognosis are not known.
Methods: T-cell counts and clinical outcomes among men in the Multicenter AIDS Cohort Study who had incident HIV-1 infection before December 31, 1995 were analyzed by Kaplan-Meier and Cox proportional hazards methods. Results were also analyzed by time-dependent multivariate methods to adjust for CD4 lymphocyte counts, viral loads, age, race, and polymorphisms in host chemokine receptor genes (CCR5-Δ32 and CCR2-64I).
Results: Among 424 cases whose date of seroconversion was known to within ±4.5 months, 317, 52, and 55 inverted their CD4/CD8 ratio within less than 1, 1 to 2, and more than 2 years of seroconversion, respectively. Longer time to inversion was significantly associated with longer time to AIDS, even after adjusting for CD4 lymphocyte count and viral load at the first seropositive visit and over the first 3 seropositive visits. Of the 6 seroconverters who had more than 500 CD4 lymphocytes 10 years after seroconversion without receiving highly active antiretroviral therapy, 5 took more than 2 years to invert their CD4/CD8 ratio.
Conclusions: Time from HIV-1 seroconversion to inversion of the CD4/CD8 ratio independently predicted time to AIDS. Early measurements of the CD4/CD8 ratio until inversion occurs may identify people likely to become long-term nonprogressors or slow progressors, thus facilitating detailed studies of the mechanism of HIV-1 disease progression.
From the Departments of *Molecular Microbiology and Immunology, †Epidemiology, and ‡Environmental Health Sciences, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; §Department of Epidemiology, University of California School of Public Health, Los Angeles, CA; ∥Department of Pediatrics, Feinberg School of Medicine, Northwestern University, Chicago, IL; ¶Department of Microbiology, University of Pittsburgh Graduate School of Public Health, Pittsburgh, PA; and #Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD.
Received for publication August 10, 2005; accepted April 11, 2006.
Reprints: Joseph B. Margolick, MD, PhD, Department of Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, 615 E. Wolfe St, Room E5153, Baltimore, MD 21205 (e-mail: firstname.lastname@example.org).
HIV-1 infection is characterized by a decline in CD4 lymphocyte counts and an increase in CD8 lymphocyte counts, leading to inversion of the CD4/CD8 ratio from the normal value of more than 1.6 to a value of less than 1.0. In most cases, inversion of this ratio occurs soon after HIV-1 seroconversion. Anecdotally, some seroconverters have been observed to maintain ratios more than 1 for more than one year after seroconversion, but no studies have examined this phenomenon systematically. Therefore, the purposes of this study were (1) to characterize the time from HIV-1 seroconversion to inversion of the CD4/CD8 ratio, and (2) to determine whether this interval is associated with the rate of progression of HIV-1 disease. For these purposes, we studied the large group of seroconverters in the Multicenter AIDS Cohort Study (MACS) who had well-defined times of seroconversion as well as follow-up before the advent of highly active antiretroviral therapy (HAART). We also assessed whether any association between the time to CD4/CD8 ratio inversion and the rate of HIV-1 disease progression was influenced by the CD4 lymphocyte counts, viral loads, and selected genetic characteristics of the study population.
MATERIALS AND METHODS
Population, Laboratory Methods, and Study Design
The Multicenter AIDS Cohort Study is a prospective study of the natural and treated history of HIV-1 infection among gay and bisexual men in the United States. A total of 4954 men were enrolled between April 1984 and March 1985 at the 4 study sites, located in Baltimore-Washington, Chicago, Los Angeles, and Pittsburgh. From 1987 to 1990, an additional 668 men, primarily African Americans, were recruited. Detailed descriptions of the MACS have been published,1,2 and only methods relevant to the present analyses are given here. 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 laboratory testing and storage in both local and national repositories. Enzyme-linked immunosorbent assays with confirmatory Western blots are performed at each study visit to determine HIV-1 seropositivity. T-lymphocyte subsets are measured by each MACS center using a standardized flow cytometry protocol3,4 along with automated complete blood counts and differentials. Plasma concentrations of HIV-1 RNA are measured using the reverse transcriptase-polymerase chain reaction assay (Roche Diagnostics, Nutley, NJ),5 and measurements for this study were performed as described6 on stored specimens, with a lower limit of detection of 400 HIV-1 RNA copies/mL. Genetic testing of most MACS seroconverters has been performed and has contributed to the identification of polymorphisms that affect the rate of progression of HIV-1 disease.7 In this study, we analyzed polymorphisms in genes for CCR5 and CCR2. The institutional review boards at each of the participating centers approved the MACS study protocols, and informed consent was obtained from all participants.
For the present analysis, data were included only through December 31, 1995, when HAART began to be used within the MACS, to exclude the effect of HAART on study outcomes and covariates. As of that date, 526 seroconverters had been identified within the MACS cohort. The present study included the 424 men (81%) who had an interval of 9 months or less between their last seronegative and first seropositive visits (so that the date of seroconversion, taken as the midpoint of this interval, was known to within ±4.5 months).
CD4 and CD8 lymphocyte percentages and absolute counts were evaluated at the first seropositive visit and at each follow-up visit. CD4/CD8 inversion was defined as a CD4/CD8 ratio of less than 1.0. For those who had CD4/CD8 inversion at the first HIV-seropositive study visit, time from seroconversion to the inversion was taken as 3 months. A Kaplan-Meier estimator was used to describe the time from seroconversion to (a) the first inversion of the CD4/CD8 ratio and (b) the development of clinically defined AIDS8 within groups with similar times (<1, 1-2, and >2 years) from seroconversion to inversion of the CD4/CD8 ratio. Seven men, who developed AIDS within 2 years of seroconversion, were excluded from the latter analysis. The log-rank test was used to evaluate the significance of differences between the comparison groups.
Cox proportional hazards models were used to evaluate whether the time to CD4/CD8 ratio inversion was associated with the time to the development of AIDS. Multivariate models were used to adjust for CD4 lymphocyte counts, HIV-1 viral load, age, race, and CCR2 and CCR5 polymorphisms (64I and Δ32, respectively). Because the predictive value of the CD4 lymphocyte count and viral load is enhanced over time (with more recent measurements exerting greater influence than the values at baseline), these 2 variables were analyzed both at the first seropositive visit and as time-dependent covariates (at the first, second, and third seropositive visits) in the Cox models. Both CD4 lymphocyte count and viral load were categorized into quartiles, and the bottom quartiles of CD4 lymphocyte count and viral load were used as reference groups in the models. To analyze the relationship between the preseroconversion CD4/CD8 ratio and the above outcomes, the last CD4/CD8 ratio before seroconversion for each seroconverter was entered into the models. P values reported are 2-sided. All analyses were performed with SAS statistical software (version 8.1; SAS Institute Inc, Cary, NC).
In these analyses, we used a separate category of "missing" for study participants who did not have the variable under analysis, to test for any biases due to nonrandomly distributed missing data. No such biases were found.
Description of the Study Population
Descriptive characteristics of the 424 men contributing to this study are presented in Table 1. Eighty-five percent were white. The mean age at seroconversion was 34.1 years. The mean CD4 lymphocyte count at the last seronegative visit was 975 cells/μL, and for CD8 lymphocytes, this number was 692 cells/μL. At the first seropositive visit, 2.8% of the men had undetectable (<400 copies/mL) HIV-1 viral load, and for those whose HIV-1 viral load was detectable, the mean value was 4.4 log10 copies/mL. Distributions of enrollment center, year of seroconversion, interval between last seronegative and first seropositive visits, and CCR2 and CCR5 genotypes are also presented in Table 1. As of December 31, 1995, 182 men (42.9%) had developed clinically defined AIDS.
Distributions of Times From Seroconversion to Inversion of the CD4/CD8 Ratio and to AIDS-Related Events
The cumulative incidence of inversion of the CD4/CD8 ratio is shown in Figure 1. Among the 424 men studied, 240 (56.6%) had already inverted their CD4/CD8 ratio by the first seropositive study visit. Because these men were assigned a time to inversion of 3 months, the median time from estimated date of HIV-1 infection to inversion was 3 months (95% confidence interval (CI), 3-4 months). Among the 184 men (43.4%) who had CD4/CD8 ratios more than or equal to 1.0 at their first HIV-1-positive visit, the median time from estimated date of seroconversion to inversion of the CD4/CD8 ratio was 14 months (95% CI, 12-18 months). There were 154 men (36.3%) whose ratios were still more than 1 at the second seropositive visit (ie, about 9 months after seroconversion), and 107 men (25.2%) maintained a CD4/CD8 ratio more than 1 for more than one year after seroconversion (Table 1). Sixteen men who were followed for more than 2 years did not invert their ratios within the study period. Subgroups of men with differing times to ratio inversion were similar with regard to enrollment center and CCR2 or CCR5 genotype. However, there was a trend toward greater preseroconversion CD4 lymphocyte counts and CD4/CD8 ratios in those who took longer to invert or did not invert this ratio (Table 1).
Association Between CD4/CD8 Ratio Inversion and Time to AIDS
To achieve an unbiased analysis of the relationship between time to inversion of the CD4/CD8 ratio and time to AIDS, we eliminated from this analysis the 7 men who developed AIDS within 2 years of seroconversion. Among the remaining 417 men, an inverse relationship was observed between time to inversion of the CD4/CD8 ratio and the rate of progression to AIDS (Fig. 2). Specifically, the median time from seroconversion to AIDS in seroconverters who had CD4/CD8 ratio inversion within one year after seroconversion (n = 310) was 96 months (95% CI, 87-104 months), and the corresponding figure for those who had CD4/CD8 ratio inversion 1 to 2 years after seroconversion (n = 52) was 122 months (95% CI, 99-∞ months). For those who had CD4/CD8 inversion 2 or more years after seroconversion (n = 55), the median AIDS-free survival was not reached, but the lower bound of the 95% CI of median time-to-AIDS in this group was 126 months.
The associations between time to CD4/CD8 inversion and time to the development of AIDS, derived from proportional hazards models with baseline data, are presented in Table 2. The univariate analysis indicated that higher CD4 lymphocyte count and lower HIV-1 viral load at the first seropositive visit, as well as longer time to inversion of the CD4/CD8 ratio, were significantly associated with development of AIDS. Viral load and time to inversion of CD4/CD8 ratio were significant in the multivariate analysis. Similar results were obtained using univariate and multivariate proportional hazards analysis in which both CD4 lymphocyte counts, and viral load levels were treated as time-dependent covariates over the first 3 seropositive visits (Table 3). Again, longer time to CD4/CD8 inversion was independently associated with slower development of AIDS, with a relative risk of 0.52 for those who did not have ratio inversion by 2 years after seroconversion.
In all of these analyses, no significant differences in time to CD4/CD8 ratio inversion were observed because of CCR2 genotypes (homozygous wildtype, heterozygous 64I, and 64I/64I) and CCR5 genotypes (homozygous wildtype and heterozygous CCR5-Δ32) (see Table 1; P > 0.05). Heterozygosity for CCR2-64I or CCR5-Δ32 was associated with longer time to AIDS in the baseline analysis (Table 2) but not in the time-dependent analysis (Table 3).
Association Between Inversion of the CD4/CD8 Ratio and Long-Term Nonprogression
To evaluate whether time to inversion of the CD4/CD8 ratio was related to identification of men destined to be long-term nonprogressors, we reviewed follow-up data on the study group to identify men who had CD4 lymphocyte counts of 500 cells/μL or more at 10 years or more after seroconversion without receiving HAART. There were 6 men who met these criteria, and 5 of them maintained a CD4/CD8 ratio of 1.0 or more for more than 2 years after seroconversion (Table 4). Thus, inversion of the CD4/CD8 ratio more than 2 years after seroconversion was significantly associated with becoming a long-term nonprogressor (5/55 vs 1/369; odds ratio, 36.8; 95% CI, 4.2-321.3; P < 0.01 by Fisher exact test). Other markers of HIV-1 disease progression were less strongly associated with nonprogression: only 2 of the 6 had undetectable viral loads (<400 copies/mL) at 2 years after seroconversion, and 2 other men had viral loads of 8900 and 21,890 at this time. Participant 2 had viral loads ranging from 5399 to 31098 over the first 10 years of infection, with maintenance of high CD4 lymphocyte counts.
These data suggested that maintenance of a CD4/CD8 ratio 1.0 or more for more than 2 years after HIV-1 seroconversion may be a useful marker for identifying a population that could be studied prospectively with a meaningful likelihood of having slow progression or nonprogression of HIV-1 disease. To test this hypothesis, we performed a similar analysis on the members of the MACS cohort who were HIV seropositive at study entry. Of those who had a CD4/CD8 ratio of more than 1 at 1.5 to 2 years after study entry (n = 440), 49 (11.1%) were long-term nonprogressors as defined above (ie, maintenance of CD4 lymphocyte counts >500 cells/μL for at least 10 years in the absence of HAART). This was similar to the 8.9% observed among the seroconverters who had ratios of more than 1 for more than 2 years after seroconversion. In contrast, only 20 of 1079 men whose CD4/CD8 ratios were less than 1 at 1.5 to 2 years after study entry went on to become long-term nonprogressors by this definition.
The CD4/CD8 ratio has long been recognized as a predictor of progression of HIV-1 disease, although other predictors are more useful clinically, including CD4 lymphocyte count and percentage9 and plasma viral load.5 It has also been reported that this ratio can predict responses to HAART.10 To our knowledge, this is the first study to examine the distribution of intervals from HIV-1 seroconversion to inversion of this ratio and whether this interval is associated with time to AIDS. We found that in a large cohort of men who have sex with men, the interval to inversion of the CD4/CD8 ratio not only had predictive value for the time to AIDS but was an independent predictor after taking into account CD4 lymphocyte count and viral load in the first 2 years after HIV-1 seroconversion. Whether this is true of other HIV risk groups, such as women or children, remains to be shown.
In HIV-uninfected people, the CD4/CD8 ratio has a genetic component,11 although it is not known what specific genes are involved. It has been proposed that people genetically programmed to have a high CD4/CD8 ratio may be relatively resistant to progression of HIV-1 infection,12 but experimental data to support this hypothesis are lacking.13,14 In the present study, there was a significant trend toward higher preseroconversion CD4/CD8 ratios in those with delayed inversion of the ratio, which in turn was significantly associated with delayed progression to AIDS. Therefore, this question may deserve further study, although in the present study preseroconversion CD4/CD8 ratio was not significantly associated with time to AIDS. A recent finding that a region of chromosome 11 was associated with the CD4/CD8 ratio may help address this question.15
Another possible explanation for the relationship between ratio inversion and progression of HIV-1 disease is that changes in the CD4/CD8 ratio after HIV-1 infection, including variations in time to inversion, reflect variations in immune responses to the virus, such as altered compartmentalization of lymphocytes between peripheral blood and lymphoid tissues.16,17 We did not find that the time to inversion of the ratio was influenced by polymorphisms in 2 genes known to affect the rate of HIV-1 disease progression, CCR5 and CCR2, but it is likely that the study size was underpowered to find such an association.
The importance of these findings in the current era is 2-fold. First, this work establishes a potential correlate of immune protection that complements CD4 cell counts and plasma HIV-1 RNA concentrations. This may be important for vaccine initiatives. Second, these results establish a potentially improved phenotype for identifying future slower progressors and long-term nonprogressors. This may facilitate pathogenesis studies to define the early events that determine the rate at which HIV-1 disease progresses. In such studies, it is critical to analyze the full spectrum of immune responses to HIV, but often, these studies have been hampered by restriction to stored samples, small numbers of available cells, and/or infrequent sampling. All of these limitations could be best addressed by prospective studies that were enriched for people with a high likelihood of much slower than average progression of HIV-1 disease.
The authors thank Shaoguang Chen, Xiuhong Li, and Dr Lisa Jacobson for the expert statistical assistance.
Data in this manuscript were collected by the Multi-center AIDS Cohort Study (MACS) with centers (Principal Investigators) located at The Johns Hopkins Bloomberg School of Public Health (Joseph Margolick); Howard Brown Health Center and Northwestern University of California, Los Angeles (Roger Detels); University of Pittsburgh (Charles Rinaldo); and Data Analysis Center (Lisa Jacobson). The MACS is funded by the National Institute of Allergy and Infectious Diseases, with additional supplemental funding from the National Cancer Institute and the National Heart, Lung, and Blood Institute: UO1-AI-35042, 5-M01-RR-00052 (GCRC), UO1-AI-35043, UO1-AI-37984, UO1-AI-35039, UO1-AI-35040, UO1-AI-37613, and UO1-AI-35041.
1. Kaslow RA, Ostrow DG, Detels R, et al. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. Am J Epidemiol
2. Dudley J, Jim S, Hoover D, et al. The Multicenter AIDS Cohort Study: retention after 9 1/2 years. Am J Epidemiol
3. Giorgi JV, Cheng H-L, Margolick JB, et al. Quality control in the flow cytometric measurement of T-lymphocyte subsets: the Multicenter Aids Cohort study (MACS) experience. Clin Immunol Immunopathol
4. Schenker EL, Hultin LE, Bauer KD, et al. Evaluation of a dual-color flow cytometry immunophenotyping panel in a multicenter quality assurance program. Cytometry
5. Mellors J, Muñoz A, Giorgi J, et al. Plasma viral load and CD4+
lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med
6. Lyles RH, Muñoz A, Yamashita T, 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
7. O'Brien SJ, Moore JP. The effect of genetic variation in chemokines and their receptors on HIV transmission and progression to AIDS. Immunol Rev
8. CDC. Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. MMWR
9. Taylor JMG, Fahey JL, Detels R, et al. CD4 percentage, CD4 number and CD4:CD8 ratio in HIV infection: which to choose and how to use. J Acquir Immune Defic Syndr
10. Bonnet F, Thiebaut R, Chene G, et al. Determinants of clinical progression in antiretroviral-naive HIV-infected patients starting highly active antiretroviral therapy. Aquitaine Cohort, France, 1996-2002. HIV Med
11. Amadori A, Zamarchi R, De Silvestro G, et al. Genetic control of the CD4/CD8 T-cell ratio in humans. Nat Med
12. Amadori A, Zamarchi R, Chieco-Bianchi L. CD4:CD8 ratio and HIV infection: the 'tap-and-drain' hypothesis. Immunol Today
13. Taylor JMG, Sy JP, Visscher B, et al. CD4+
T-cell number at the time of acquired immunodeficiency syndrome. Am J Epidemiol
14. Bostik P, Villinger F, Ansari AA, et al. Pre-infection CD4+
ratio and HIV infection. Immunol Today
15. Evans DM, Zhu G, Duffy DL, et al. A major quantitative trait locus for CD4-CD8 ratio is located on chromosome 11. Genes Immun
16. Rosenberg YJ, Anderson AO, Pabst R. HIV-induced decline in blood CD4/CD8 ratios: viral killing or altered lymphocyte trafficking? Immunol Today
17. Cajigas A, Suhrland M, Harris C, et al. Correlation of the ratio of CD4+
cells in lymph node fine needle aspiration biopsies with HIV clinical status. A preliminary study. Acta Cytol
Keywords:© 2006 Lippincott Williams & Wilkins, Inc.
CD4/CD8 ratio; HIV-1 disease progression; long-term nonprogressor