The CD4+ T-cell count is the core marker of disease progression in HIV-1 and is the key variable considered during the decision process for such major interventions as initiation of chemoprophylaxis to prevent opportunistic diseases or prescription of antiretroviral therapy (ART).1 Another marker is the plasma HIV-1 RNA level, which can predict disease natural progression independently of CD4+ T-cell count,2 and is the key indicator used to evaluate the efficacy of ART in controlling viral replication.3,4 Policy guidelines have long recommended routine monitoring of both CD4+ T-cell counts and plasma HIV-1 RNA levels in HIV-infected patients on and off ART.
Since 2000, several studies have suggested that a third marker, the HIV-1 DNA level in peripheral blood mononuclear cells (PBMCs), may be strongly associated with disease progression, independently of plasma HIV-1 RNA level and CD4+ T-cell count.5-7 However, until now only a handful of cohort studies in Europe have demonstrated this association. In order for their findings to be convincing, they must prove replicable in diverse populations and settings.
We analyzed the association between HIV-1 DNA level in PBMCs and HIV-1 disease progression in a cohort of adults in Côte d'Ivoire, West Africa, who were recruited shortly after their estimated date of HIV-1 seroconversion.
This study was conducted in the blood donor clinic of the blood bank of Abidjan, Côte d'Ivoire. In Côte d'Ivoire, blood donors are nonpaid adult volunteers. Informed consent for HIV testing is obtained before each donation, and patients are notified that they have the option of receiving their test results at the blood donor clinic after a posttest counseling session.
Since June 1997, blood donors have been eligible to enroll in the ANRS 1220 Primo-CI cohort if they: (1) are diagnosed with HIV-1 or HIV-1/HIV-2 coinfection at blood donation; (2) tested HIV seronegative at the previous donation; (3) return to the clinic for HIV test results; and (4) are estimated to have seroconverted <36 months before the donation. We use the midpoint between the last negative and first positive HIV tests to estimate the date of seroconversion. All patients who accept to participate give written informed consent. The Primo-CI protocol was approved by the ethics committees of the national Ivorian program on AIDS and the institutional review board of the French Agency for Research on AIDS (ANRS, France).
For this study, we included in the analysis all participants in the Primo-CI cohort infected with HIV-1 alone, who entered the cohort between June 9, 1997, and February 1, 2006, and had a PBMC pellet cryopreserved at enrollment.
The procedures of the Primo-CI cohort have been previously described.8,9 In summary, patients were examined by clinicians at enrollment and every 6 months thereafter. All subjects had free access to the study clinic at any time between scheduled visits. All clinical events were reviewed by an event documentation committee. The diagnostic criteria and treatment procedures were identical to those used for the Cotrame cohort, a cohort of HIV-infected adults followed up in Abidjan during the same period, under the supervision of the same team and for which procedure has been described elsewhere.10 Cotrimoxazole prophylaxis was systematically offered to all patients at enrollment.11 In February 2001, the blood donor clinic became a center for the United Nations Joint Programme on HIV/AIDS/Côte d'Ivoire initiative for improving access to HIV drugs. From this date forward, ART was provided to all participants in the Primo-CI cohort who met World Health Organization (WHO) criteria for starting ART in resource-limited settings.12,13 All drugs, clinical assessments, hospital stays, and transport were offered free of charge.
HIV was diagnosed using 2 reactive enzyme-linked immunosorbent assays (Murex ICE 1-0-2, Abbott, North Chicago, IL; and Vironostika HIV Uni-Form II, Organon Teknika BV, Boxtel, The Netherlands). We confirmed HIV-1 infection before enrolling subjects, using 2 more enzyme-linked immunosorbent assays (Murex ICE 1-0-2 and Pepti-Lav 1-2; Pasteur Diagnostics, Marnes-la-Coquette, France). Blood samples were collected at inclusion and every 6 months thereafter. Pellets of PBMCs were separated using a Ficoll-Hypaque gradient, and aliquots of plasma and PBMCs were stored in −80°C freezers. CD4+ T-cell counts were measured for each blood sample (FACScan, Becton Dickinson, Aalst-Erembodegem, Belgium). HIV-1 RNA plasma levels were measured in the CIRBA virology laboratory, in Abidjan (Cobas Amplicor HIV-1 Monitor, version 1.5, Roche Diagnostics, Indianapolis, IN; threshold of detection 400 copies), and HIV-1 DNA levels were measured in the virology laboratory of the Necker Hospital, in Paris. We used a real-time polymerase chain reaction assay to target the long terminal repeats (LTR) gene region of HIV-1 DNA, as previously described.14 The standard curve consisting in 5-fold dilutions of 8E5 cell total DNA (containing 1 copy of HIV-1 DNA per cell) was used. The cutoff of the assay was 6 copies/polymerase chain reaction, that is, 40 copies/106 PBMCs. Total DNA in extracts was quantified using fluorescence readings at 260 nm; 1 μg of total DNA was considered to be equivalent to 15,000 cells. Results were expressed as the number of HIV-1 DNA copies/106 PBMCs.
Baseline data were defined as measurements taken on the date of inclusion in the cohort. Data were censored either at time of death or on December 31, 2006, for those patients who were still alive. For patients whose last contact with the study team was before December 31, 2006, we used tracing procedures up to June 30, 2007. Patients who were found alive or had died in 2007 were considered to be alive on December 31, 2006. The remaining patients were considered to be lost to follow-up, and their data were censored at the date of their last contact with the study team. Finally, data on patients who started ART before December 31, 2006, but did not reach any disease progression outcome were censored at the date of ART initiation.
Spearman correlation test was used to estimate the association between baseline HIV-1 DNA, HIV-1 RNA, and CD4+ T-cell count. We considered 3 outcomes: death, first severe morbidity event, and first CD4+ T-cell count <200/mm3. Severe morbidity events were defined as all WHO stage 3 or 4 events.12 We estimated the probability of remaining free of each outcome separately, and all 3 outcomes combined using the Kaplan-Meier method. We analyzed the association of the baseline HIV-1 DNA level in PBMCs with each outcome and with at least 1 of the 3 outcomes, using multivariate Cox proportional hazard regression models. Baseline and follow-up CD4 count was included as a time-dependent variable. All other dependent variables were baseline variables associated with the outcome in univariate analysis (P < 0.25).
Of the 254 HIV-1-positive patients included in the Primo-CI cohort during the study period, 200 patients had available baseline PBMC samples and were included in the study. Sixty-two percent of subjects were men, 65% were single, 53% had a steady source of income, and 75% had attained higher than primary school level education. Table 1 shows the other baseline characteristics of the cohort. The HIV-1 DNA level was found to be significantly correlated with baseline HIV-1 RNA level (Spearman test: R = 0.48, P < 0.001) and inversely correlated with baseline CD4+ T-cell count (Spearman test: R = −0.44, P < 0.001).
During follow-up, 10 patients died, 25 had at least one WHO stage 3 or 4 event, 41 had a CD4+ T-cell count decline to <200/mm3, and 57 had at least 1 of these 3 outcomes. In the 25 patients with at least one stage 3 or 4 event, the first was tuberculosis (n = 12), oral candidiasis (n = 7), or an invasive bacterial disease (n = 6).
During follow-up, 62 participants started ART, including 43 of the 100 (43%) patients with baseline HIV-1 DNA >3 log10 copies/106 PBMCs and 19 of the 100 (19%) patients with baseline HIV-1 DNA ≤3 log10 copies/106. Among the 100 patients with baseline HIV-1 DNA >3 log10 copies/106 PBMCs, 20% started ART at >200 CD4/mm3 and 24% started ART before being diagnosed with a WHO stage 3 or 4 clinical event. In the 100 patients with baseline HIV-1 DNA ≤3 log10 copies/106, these percentages were 6% and 12%, respectively.
Figure 1 shows the overall probability of remaining free of any outcome over time. The 5-year probabilities of survival, of remaining free of WHO stage 3 or 4 event, of not having a CD4+ T-cell count below 200/mm3, and of remaining free of all 3 outcomes were estimated at 0.92 [95% confidence interval (CI): 0.84 to 0.96], 0.81 (95% CI: 0.72 to 0.87), 0.64 (95% CI: 0.54 to 0.73), and 0.56 (95% CI: 0.46 to 0.65), respectively.
Figure 2 shows the probability of remaining free of all 3 outcomes over time, depending on baseline PBMC HIV-1 DNA level (2A), baseline CD4 count (2B), and baseline plasma HIV-1 RNA level (2C). The 5-year probability of remaining free of any outcome was 0.74 in patients with baseline HIV-1 DNA ≤3.0 log10 copies/106 PBMCs and 0.35 in patients with baseline DNA >3.0 log10 copies/106 PBMCs (P < 0.001) (Fig. 2A); 0.64 in patients with baseline CD4 count >350/mm3 and 0.31 in those with baseline CD4 count ≤350/mm3 (P < 0.001) (Fig. 2B); and 0.61 in patients with baseline plasma HIV-1 RNA ≤5.0 log10 copies/mL and 0.36 in those with HIV-1 RNA >5.0 log10 copies/mL (P = 0.055) (Fig. 2C).
Table 2 shows the results of the multivariate analysis on the association between disease progression and patient characteristics, both for each outcome separately and for all outcomes combined. Patients with baseline HIV-1 DNA levels >3 log10 copies/106 PBMCs were 6.97 times more likely to die, 2.35 times more likely to attain a CD4+ T-cell count <200/mm3, and 2.17 times more likely to reach any of the 3 outcomes than patients with lower HIV-1 DNA levels. The HIV-1 DNA level had the strongest independent association with all outcomes. Follow-up CD4+ T-cell counts were also associated with the combined outcomes, but the association was weaker. Interestingly, plasma HIV-1 RNA levels were weakly associated with outcomes after controlling for HIV-1 DNA level.
Within this cohort of West African adults who seroconverted to HIV-1 a median of 9 months before enrollment, baseline HIV-1 DNA level in PBMCs was strongly associated with HIV-1 disease progression, independently of plasma viral load and CD4+ T-cell count. Patients with baseline HIV-1 DNA level >3 log10 copies/106 PBMCs had a 7.0 increase in risk of death and a 2.4 increase in risk of CD4+ T-cell count decline to <200/mm3, compared to patients with HIV-1 DNA <3 log10 copies/106 PBMCs, after adjusting for other characteristics.
A similar independent association between early PBMC HIV-1 DNA level and HIV-1 disease progression has been reported elsewhere for European patients infected with subtype B HIV-1 viruses.5-7 Our study demonstrates that these findings can be replicated in African adults harboring different HIV-1 subtypes and immunologic and virologic characteristics. Compared with subjects in the European studies, our patients had higher baseline PBMC HIV-1 DNA levels and lower baseline CD4+ T-cell counts.15 In addition, all were infected with non-B HIV-1 strains, mainly CRF02_AG.16-19 Despite these differences, our results are consistent with previous studies, showing the same large relative risk of progression for patients with high HIV-1 DNA levels, the same robustness in the results when using separate or combined progression outcomes, and the same overwhelming prognostic importance of HIV-1 DNA over plasma HIV-1 RNA. In fact, the association of HIV-1 RNA with progression outcomes was not significant after adjusting for DNA levels.
In our study, 31% of participants started ART before the study ended. Some subjects started ART before having reached one or several criteria used to define disease progression. Time at risk was censored at ART initiation in these patients. This may have led to an informative bias, assuming that the progression of HIV-1 disease could have been different in these patients if they had not started ART, as compared with the progression of patients who did not start ART. This bias would lead to underestimate the disease progression more deeply for subjects whose baseline HIV-1 DNA level was >3 log10 copies/106 PBMCs because a larger proportion of patients with high HIV-1 DNA levels started ART. Therefore, we may have underestimated the hazard ratio of disease progression in patients with high baseline HIV-1 DNA levels, relative to patients with low baseline HIV-1 DNA.
PBMC HIV-1 DNA may reflect the cellular reservoir of HIV-1, as opposed to plasma HIV-1 RNA that reflects HIV-1 replication.20 The strong association between baseline PBMC HIV-1 DNA and disease progression may reflect the prognostic importance of the number of HIV-1-infected cells during acute infection. CD4+ T-cell counts and plasma HIV RNA levels have proven to be reliable and consistent markers and are currently the major indicators in the clinical standard of care. However, HIV-1 DNA levels could act as an additional marker to help distinguish those among patients with similar CD4 counts who might benefit from different therapeutic decisions. In our study, HIV-1 DNA levels were strongly associated with disease progression, even after adjusting for time-updated CD4+ T-cell counts. This suggests that HIV-1 DNA levels could help identify individuals at risk of disease progression within a specific CD4 stratum, thereby informing decisions to start ART earlier in selected individuals. Further trials of early ART initiation should measure participants' PBMC HIV-1 DNA to address this question.
1. Mellors JW, Munoz A, Giorgi JV, et al. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med
2. Mellors J, Rinaldo CJ, Gupta P, et al. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science
3. Hammer S, Squires K, Hughes M, 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. AIDS Clinical Trials Group 320 Study Team. N Engl J Med
4. Gulick R, Mellors J, Havlir D, et al. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N Engl J Med
5. Kostrikis L, Touloumi G, Karanicolas R, et al. Quantitation of human immunodeficiency virus type 1 DNA forms with the second template switch in peripheral blood cells predicts disease progression independently of plasma RNA load. J Virol
6. Rouzioux C, Hubert J, Burgard M, et al. Early levels of HIV-1 DNA in peripheral blood mononuclear cells are predictive of disease progression independently of HIV-1 RNA levels and CD4+ T cell counts. J Infect Dis
7. Goujard C, Bonarek M, Meyer L, et al. CD4 cell count and HIV DNA level are independent predictors of disease progression after primary HIV type 1 infection in untreated patients. Clin Infect Dis
8. Salamon R, Marimoutou C, Ekra D, et al. Clinical and biological evolution of HIV-1 seroconverters in Abidjan, Cote d'Ivoire, 1997-2000. J Acquir Immune Defic Syndr
9. Minga A, Danel C, Abo Y, et al. Progression to WHO criteria for antiretroviral therapy in a 7-year cohort of adult HIV-1 seroconverters in Abidjan, Côte d'Ivoire. Bull World Health Organ
10. Anglaret X, Messou E, Ouassa T, et al. Pattern of bacterial diseases in a cohort of HIV-1 infected adults receiving cotrimoxazole prophylaxis in Abidjan, Côte d'Ivoire. AIDS
11. Mermin J, Lule J, Ekwaru J, et al. Effect of co-trimoxazole prophylaxis on morbidity, mortality, CD4-cell count, and viral load in HIV infection in rural Uganda. Lancet
13. World Health Organization. Antiretroviral therapy for HIV infection in adults and adolescents in resource-limited settings: towards universal access: recommendations for a public health approach (2006 revision). Available at: http://www.who.int/hiv/pub/guidelines/WHO
14. Schvachsa N, Turk G, Burgard M, et al. Examination of real-time PCR for HIV-1 RNA and DNA quantitation in patients infected with HIV-1 BF intersubtype recombinant variants. J Virol Methods
15. Lewden C, Thiebaut R, Boufassa F, et al. Comparison in Early CD4 T Cell Count Evolution in HIV-1 Seroconverters in Côte d'Ivoire and France: The ANRS PRIMO-CI and SEROCO Cohorts: Proceedings of the 15th Conference on Retroviruses and Opportunistic Infections (CROI), Boston, MA, USA, February 3-6 2008
. Poster 691, Abstract K229.
16. Toni TD, Adje-Toure C, Vidal N, et al. Presence of CRF09_cpx and complex CRF02_AG/CRF09_cpx recombinant HIV type 1 strains in Côte d'Ivoire, West Africa. AIDS Res Hum Retroviruses
17. Toni T, Recordon-Pinson P, Minga A, et al. Presence of key drug resistance mutations in isolates from untreated patients of Abidjan, Cote d'Ivoire: ANRS 1257 study. AIDS Res Hum Retroviruses
18. Toni T, Masquelier B, Bonard D, et al. Primary HIV-1 drug resistance in Abidjan (Cote d'Ivoire): a genotypic and phenotypic study. AIDS
19. Toni T, Masquelier B, Minga A, et al. HIV-1 antiretroviral drug resistance in recently infected patients in Abidjan, Côte d'Ivoire: a 4-year survey, 2002-2006. AIDS Res Hum Retroviruses
20. Cao Y, Qin L, Zhang L, et al. Virologic and immunologic characterization of long-term survivors of human immunodeficiency virus type 1 infection. N Engl J Med