JAIDS Journal of Acquired Immune Deficiency Syndromes:
Epidemiology and Social Science
Comparison of Early CD4 T-Cell Count in HIV-1 Seroconverters in Côte d'Ivoire and France: The ANRS PRIMO-CI and SEROCO Cohorts
Lewden, Charlotte MD, PhD*†; Thiébaut, Rodolphe MD, PhD*†; Boufassa, Faroudy MD‡; Coulibaly, Ali MPH§; Malateste, Karen MPH*†; Seng, Rémonie MD‡; Toni, Thomas d'Aquin PhD§; Inwoley, André PharmD, PhD§; Rouzioux, Christine PharmD, PhD‖; Minga, Albert MD§; Anglaret, Xavier MD, PhD*†§; Meyer, Laurence MD, PhD‡
From the *Inserm, U897, Bordeaux, France; †Université Victor Segalen Bordeaux 2, Institut de Santé Publique, d'Epidémiologie et de Développement, Bordeaux, France; ‡Inserm, U822, Université Paris Sud 11, Assistance Publique Hôpitaux de Paris, Hôpital Bicêtre, Le Kremlin-Bicêtre, France; §Programme PACCI, Abidjan, Côte d'Ivoire; and ‖Hôpital Necker, Laboratoire de virologie, Paris, France.
Received for publication March 12, 2009; accepted July 13, 2009.
Supported by Agence Nationale de Recherches sur le Sida et les Hépatites Virales.
These results were presented partly at the 15th Conference on Retroviruses and Opportunistic Infections, February 3-6, 2008, Boston, MA.
Correspondence to: Charlotte Lewden, MD, PhD, Inserm U897, ISPED, Université de Bordeaux 2, 146 rue Léo-Saignat 33076 Bordeaux cedex, France (e-mail: firstname.lastname@example.org).
Objective: We compared CD4+ decline among untreated HIV-1-infected seroconverters living in Côte d'Ivoire (CI) and in France.
Methods: HIV-1-infected adults were enrolled in the ANRS1220 PRIMO-CI (CI, 1997-2006) and ANRSCO2 SEROCO (France, 1988-1995) cohorts. CD4+ count and percentage declines were estimated from enrollment until 24 months of seroconversion by linear random-effect models adjusted for time since seroconversion, age, gender, cell-associated HIV DNA, HIV RNA, and country.
Results: Overall 521 seroconverters (CI 148, 62% men; France 373, 77% men) were enrolled after a median of 7.6 months since seroconversion. Median follow-up duration was 12.7 months. Median age was 28 years. Median baseline CD4+ count was 472 and 560 cells per cubic millimeter, respectively. Median baseline HIV RNA was 4.4 and 4.0 log10 copies per milliliter and median HIV DNA was 3.0 and 2.8 log10 copies per 106 peripheral blood mononuclear cells, respectively. In adjusted models, CD4+ count and percentage at baseline were lower in CI than in France (P < 0.01), and the difference did not vary during follow-up (P = 0.55). Low HIV RNA and low HIV DNA at baseline were associated with higher CD4+ count at baseline.
Conclusions: CD4+ count and percentage were lower in CI than in France. These differences were already observed during early infection and remained similar after adjustment.
In HIV-infected individuals, the progressive depletion of CD4+ T cells results in a higher risk of severe morbidity.1-3 The CD4+ cell count after seroconversion varies individually4,5 and according to several factors such as age, gender, HIV transmission group,6 symptomatic primary infection,7 and viral setpoint.8,9 The absolute number of CD4+ cell per cubic millimeter is the most common marker of HIV disease progression, routinely used alone in resource-limited settings and associated with plasma HIV RNA in industrialized countries. It drives decisions on when to start antiretroviral therapy and prophylaxis of opportunistic infections. The natural history of HIV disease has long been claimed to be different in sub-Saharan African HIV-infected adults as compared with European and North American adults. There is good evidence of such a difference when it comes to the spectrum of HIV-related morbidity between regions, with active tuberculosis and invasive bacterial diseases being more frequent in sub-Saharan Africa than in industrialized countries.3,10 Though curable, these diseases may lead to death in settings with limited access to diagnostic and treatment facilities and might explain a significant death rate in HIV-infected adults with CD4+ above 200 cells per cubic millimeter in sub-Saharan Africa.11 Nevertheless, little is known about the potential link between this spectrum of morbidity or other region-specific factors and the CD4+ decrease in untreated sub-Saharan African adults. Few studies have compared the decline of CD4 cell count between high-income and low-income settings, and their results are discrepant.12-15 In this study, we aimed at exploring whether the decline in CD4+ cells differed after controlling for a set of key factors: age, gender, HIV-1 RNA, and HIV-1 DNA. We compared the decline in CD4+ before antiretroviral treatment and associated factors in 2 cohorts of HIV-1-infected adults with an estimated date of seroconversion in Côte d'Ivoire and in France.
The PRIMO-CI ANRS 1220 prospective cohort recruited between 1997 and 2006 among blood donors of the National Blood Bank (CNTS) in Abidjan, Côte d'Ivoire those (1) diagnosed with HIV-1 infection after a blood donation; (2) HIV seronegative at the preceding blood donation; (3) who returned to the clinic to be informed of their HIV test result; and (4) for whom the delay since the estimated date of seroconversion was less than 36 months. Follow-up was scheduled every 6 months.5
The ANRS CO2 SEROCO cohort enrolled adults with a recent HIV diagnosis or a known estimated date of HIV infection in 18 hospital wards in France. Patients with a known date of infection, defined as an interval of less than 24 months between an authenticated negative enzyme-linked immunosorbent assay (ELISA) test and a positive ELISA test, or an evocative incomplete Western blot, and enrolled between 1988 and 1996 were eligible for this analysis. Follow-up was scheduled every 6 months.16
In both cohorts, written informed consent was obtained before enrolment, and the protocol was approved by an Ethic Committee. We selected for this analysis patients enrolled within 24 months of the estimated date of seroconversion for HIV-1 and with available baseline CD4+ cell count, HIV-1 RNA, and cell-associated HIV DNA values.
Definitions and Measurements
The date of seroconversion was estimated as the midpoint between the last negative and the first positive test or the date of the incomplete Western blot minus 1 month. HIV diagnosis was done by ELISA confirmed by Western blot tests. CD4 lymphocyte counts were performed by flow cytometry. In PRIMO-CI, plasma HIV RNA was measured in frozen samples using HIV-1 Cobas Amplicor Roche (400 copies/mL) until July 28, 2003 and using ANRS real-time reverse transcriptase-polymerase chain reaction (300 copies/mL) after July 28, 2003. In SEROCO, HIV RNA was measured in frozen sera using HIV-1 Cobas Amplicor Roche (400 copies/mL).17 HIV DNA in peripheral blood mononuclear cells (PBMCs) was measured in PRIMO-CI using ANRS real-time polymerase chain reaction [70 copies/(<2.5 vs ≥2.5 log10 copies per millions PBMC)]18 and in SEROCO using HIV DNA from Roche (60 copies per millions PBMC).19
The CD4 cell decline from the date of enrollment in the cohort (baseline) was estimated using 2 mixed linear random-effect models: 1 for untransformed absolute count20 and 1 for percentage. We checked the robustness of the results using squared root transformation of CD4 count. Correlation between repeated measurements within an individual was taken into account through random effects. Time since enrollment was considered as time zero because the date of seroconversion was interval censored and no viroimmunological measurements were available before enrollment. Robustness analysis was performed using estimated date of seroconversion as time zero and gave similar results (not shown).
Follow-up was censored at 24 months after the seroconversion date, death, lost-to-follow-up, initiation of combination antiretroviral therapy, or January 1996 for the SEROCO cohort only, whichever occurred the first. Models were adjusted for time since seroconversion to enrollment, age at seroconversion, gender, cell-associated HIV-1 DNA, HIV RNA, and cohort. HIV DNA and HIV RNA were categorized in 2 groups according to the threshold value of the first quartile of the distribution in the SEROCO cohort. To assess whether the effect of variables differed according to the cohort, interactions between the cohort (SEROCO or PRIMO-CI) and variables of interest were included in the model according to a stepwise ascending procedure. The effect of variables was checked on baseline CD4 and on CD4 decline, and only interaction parameters with a P value <0.05 were kept in multivariable analysis. Statistical analyses were performed using Statistical Analysis System software (SAS, version 9.1).
Among 254 patients enrolled in the PRIMO-CI cohort and 458 in the SEROCO cohort, 148 and 373 met the criteria for this analysis, respectively (Fig. 1). Sixty-two percent and 77% of patients were men in PRIMO-CI and SEROCO, respectively, and the median age at enrollment was 28 years in both cohorts (Table 1). The median time between the last negative and first positive tests was 6.8 months [interquartile range (IQR): 3.3-12.9], and the median time between seroconversion and enrollment was 7.6 months (IQR: 4.4-12.7), with no difference between cohorts. Median baseline CD4 counts were 472 and 560 cells per cubic millimeter, and percentages were 26% and 28%, respectively. Baseline HIV RNA and cell-associated HIV DNA were higher in Côte d'Ivoire than in France (Table 1). The median duration of follow-up after enrollment was 12.5 months (IQR: 6.2-18.6) and 13.1 months (IQR: 7.3-18.8) in PRIMO-CI and SEROCO, respectively.
In crude analysis and after adjustment for time since seroconversion to enrollment and other key factors, absolute CD4 count at baseline was lower in Côte d'Ivoire than in France (P < 0.01), and this difference did not vary (P = 0.55) during follow-up (Table 2). In both cohorts, low baseline HIV RNA and low cell-associated HIV DNA were associated with a higher baseline CD4+ count. The effect of cell-associated HIV-1 DNA on CD4+ count at enrollment was significantly different between the 2 cohorts (interaction P = 0.04 in multivariate analysis). The difference at baseline between the 2 strata of cell-associated HIV DNA (above or below 2.5 log10 copies/mL), of around 100 cells per cubic millimeter in SEROCO and 230 cells per microliter in PRIMO-CI, persisted during the whole follow-up period (Fig. 2B). The average decline in CD4+ count estimated by the model was −63 cells per year (95% confidence interval: −80 to −46) for an individual enrolled in care 6 months after seroconversion. Decline in CD4+ was faster when HIV RNA was high (−57 cells/mm3/year, P = 0.02) (Table 2, Fig. 2A). Moreover, we assessed the proportion of CD4 count variability explained by viral markers in both cohorts; HIV DNA explained 22.8% in SEROCO and 9.9% in PRIMO-CI, and HIV RNA explained 22.7% in SEROCO and 10.0% in PRIMO-CI.
The effect of age and gender on CD4+ count was significantly different between the 2 cohorts. In the French cohort only, older age was associated with lower CD4+ count at enrollment and with lower CD4+ decrease (Table 2). In the French cohort only, male gender was associated with lower CD4+ cell count at enrollment in crude analysis and after adjustment on time between seroconversion and enrollment, cohort, and age. Nevertheless, there was no longer any significant association between gender and CD4+ count after taking into account virologic markers (Table 2).
Like CD4 cell count, CD4 percentage at baseline was lower in Côte d'Ivoire than in France (P < 0.01), and the difference did not vary (P = 0.20) during follow-up (Table 3). In both cohorts, low cell-associated HIV-1 DNA was associated with a higher CD4+ percentage at enrollment but not with CD4+ percentage decline (Fig. 2D). The association of HIV RNA and CD4+ percentage did not reach statistical significance. Age was not associated with CD4+ percentage. The effect of gender differed between cohorts (interaction term P = 0.04): male gender was associated with a lower CD4+ percentage at enrollment in the French cohort only (Fig. 2C).
Comparing 2 cohorts of HIV-infected adults in Côte d'Ivoire and in France and taking into account the time since estimated seroconversion, we showed that CD4+ cell count and percentage were lower in Cote d'Ivoire than in France by around 86 CD4 cells per cubic millimeter and 4%, respectively. These differences were already observed during early infection although the rate of CD4+ decline was similar between the 2 cohorts (−63 cells/year). The difference in CD4 between the 2 settings remained similar after adjustment for viral markers.
Although analyses were performed among HIV seroconverters, the delay between the estimated date of seroconversion and enrollment in the cohort (defining the availability of the first measurements of the markers) prevented a valid estimate of baseline CD4 value, that is, before HIV infection. Hence, estimated differences between baseline values represent differences established very early after primary infection. It would be interesting to confirm these results with different datasets comparing African and non-African HIV-infected subjects prospectively followed since before seroconversion. We censored the follow-up at 2 years after seroconversion to reasonably exclude selection bias that may result from informative lost-to-follow-up due to AIDS or deaths.21
Comparison With Other Studies
Studies comparing the decline in CD4 cell count after seroconversion between high-income and low-income settings are scarce. Our data are in agreement with data from other cohorts of seroconverters. In Brazil, CD4 cell counts are lower than in the North American Multicenter AIDS Cohort Study (MACS) studies with the same time after seroconversion.12 Among individuals followed in France, those originating from sub-Saharan Africa had a lower CD4+ after seroconversion than non-Africans.13
Data from prevalent studies are more difficult to interpret. For a same CD4+ percentage, individuals from Côte d'Ivoire and Burkina Faso had a higher CD4+ count than individuals from France.14 In a study comparing Ethiopian and Dutch individuals, CD4+ decrease was lower in Ethiopians for a same CD4+ level.15
Meanings and Implications
Our results indicate that the differences in CD4 levels between the 2 settings are established early during HIV infection. This may be the consequence of pathophysiological mechanisms such as higher cell activation resulting in faster decrease in CD4 cell count in the first weeks of HIV infection.22,23 This phenomenon has been attributed to be more frequent in HIV-infected adults living in Africa who are exposed to various infections.24 Several other factors may also play a role in these differences between both settings. The level of CD4 may have been lower in African adults before seroconversion.25 To our knowledge, there is no evidence of differential evolution according the HIV subtype as long as the recombinant CRF02_AG predominant in Cote d'Ivoire is concerned.26 Moreover, several nonmeasurable individual or environmental characteristics may influence CD4 count or evolution.
CD4+ variability was partly explained by differences in HIV DNA and in HIV RNA in both settings. The percentage of variability explained by these viral markers was higher in our study, which considered early viral measurements, than in studies considering patients at later stages of HIV infection.27 The differences in viral markers that we observed between the 2 cohorts may be related to technical reasons, sera or plasma samples for HIV RNA, and Roche or real-time polymerase chain reaction for HIV DNA.
Previous studies showed that cell-associated HIV DNA after seroconversion was associated with spontaneous disease progression, independently of HIV RNA and CD4+.19,28 We found here that the difference at baseline in CD4 cells per cubic millimeter according to the level of HIV DNA persisted in both settings during the whole follow-up period because there was no association between HIV DNA level at baseline and the rate of CD4+ decline. This also argues for an early role of HIV DNA on the disease course. Alternatively, HIV DNA levels may also have an effect on CD4 decrease that could be demonstrated during a longer follow-up period.
In our study, age was associated with CD4+ cell count only in France. This cannot be explained by the distribution of ages that did not differ between cohorts, and this may be due to a higher statistical power in the French cohort or to an unmeasured factor. The lowest decline in older patients in the Seroco cohort found in our analysis should be interpreted carefully because of the limited duration of follow-up. Of note, increasing age has been described to be associated with a more rapid decline in the CD4+ cell count only when it fell below 500 cells per cubic millimeter.29 The role of HIV on the changes in the immune system induced by age remains to be clarified.30,31
The association between male gender and lower baseline CD4 count in the SEROCO cohort was present in univariate analysis and was not evidenced when virologic markers were taken into account. In fact, men had higher HIV RNA and HIV DNA in our study. Higher HIV RNA in men has been previously reported in other settings.32 Concerning the percentage analysis, the association of male gender and low CD4 persisted in adjusted analysis.
Other predictors of the decrease in CD4 cell count have been identified. For instance, a study performed in Abidjan among adults whose date of seroconversion was unknown showed the role of body mass index.33 Nevertheless, this study was performed among individuals with a more advanced disease than the patients in our study.
Our results have several implications. First, there is a need to further study acute HIV infection in Africa to better understand the mechanisms that lead to a lower CD4+ count observed at an early stage of HIV infection. Second, these results argue for the need of an early HIV diagnosis, particularly in settings such as Africa, to follow patients before their need of prophylactic and antiretroviral treatments. Of note, cotrimoxazole prophylaxis is recommended in Africa when CD4 cell count is <350 cells per cubic millimeter34 and is prescribed when CD4 cell count is <500 cells per cubic millimeter in countries like Côte d'Ivoire. Nevertheless, case management of HIV infection often starts at a symptomatic stage or when CD4+ cell count is below 200 cells per cubic millimeter.35 Moreover, the monitoring of CD4+ before antiretroviral treatment in Africa should be at least as frequent than in the North.36
The CD4+ count and percentage measured at an early stage of HIV infection, lower in Côte d'Ivoire than in France, argue for additional efforts toward an earlier diagnosis of HIV infection. Moreover, the early role of HIV DNA and HIV RNA is confirmed and may lead to use both these markers in the decision-making regarding case management of HIV-infected adults.
2. Mellors JW, Munoz 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.
3. Anglaret X, Messou E, Oaussa T, et al. Pattern of bacterial disease in a cohort of HIV-1 infected adults receiving cotrimoxazole prophylaxis, Abidjan, Côte d'Ivoire. AIDS. 2003;17:575-584.
4. van Asten L, Danisman F, Otto SA, et al. Pre-seroconversion immune status predicts the rate of CD4 T cell decline following HIV infection. AIDS. 2004;18:1885-1893.
5. Salamon R, Marimoutou C, Ekra D, et al. Clinical and biological evolution of HIV-1 seroconverters in Abidjan, Côte d'Ivoire, 1997-2000. J Acquir Immune Defic Syndr. 2002;29:149-157.
6. CASCADE Collaboration. Differences in CD4 cell counts at seroconversion and decline among 5739 HIV-1-infected individuals with well-estimated dates of seroconversion. J Acquir Immune Defic Syndr. 2003;34:76-83.
7. Madec Y, Boufassa F, Porter K, et al. Spontaneous control of viral load and CD4 cell count progression among HIV-1 seroconverters. AIDS. 2005;19:2001-2007.
8. Touloumi G, Hatzakis A, Rosenberg PS, et al. Effects of age at seroconversion and baseline HIV RNA level on the loss of CD4+ cells among persons with hemophilia. Multicenter Hemophilia Cohort Study. AIDS. 1998;12:1691-1697.
9. Katzenstein TL, Pedersen C, Nielsen C, et al. Longitudinal serum HIV RNA quantification: correlation to viral phenotype at seroconversion and clinical outcome. AIDS. 1996;10:167-173.
10. 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, Cote d'Ivoire. Bull World Health Organ. 2007;85:116-123.
11. Badri M, S DL, Wood R. Short-term risk of AIDS or death in people infected with HIV-1 before antiretroviral therapy in South Africa: a longitudinal study. Lancet. 2006;368:1254-1259.
12. Djomand G, Duerr A, Faulhaber JC, et al. Viral load and CD4 count dynamics after HIV-1 seroconversion in homosexual and bisexual men in Rio de Janeiro, Brazil. J Acquir Immune Defic Syndr. 2006;43:401-404.
13. Meyer L, Chaix ML, Nagy Z, et al. Low CD4 T cell counts in sub-Saharan African subjects enrolled at the time of HIV-1 primary infection in the French ANRS PRIMO cohort. Presented at: Proceedings of the 4th IAS conference on HIV pathogenis, treatment and prevention; July 22-25, 2007; Sydney, Australia; 2007. TUPEB004.
14. Anglaret X, Diagbouga S, Mortier E, et al. CD4+ T-lymphocyte counts in HIV infection: are European standards applicable to African patients? J Acquir Immune Defic Syndr Hum Retrovirol. 1997;14:361-367.
15. Mekonnen Y, Geskus RB, Hendriks JC, et al. Low CD4 T cell counts before HIV-1 seroconversion do not affect disease progression in Ethiopian factory workers. J Infect Dis. 2005;192:739-748.
16. Seng R, Goujard C, Desquilbet L, et al. Rapid CD4+ cell decrease after transient cART initiated during primary HIV infection (ANRS PRIMO and SEROCO cohorts). J Acquir Immune Defic Syndr. 2008;49:251-258.
17. Hubert JB, Burgard M, Dussaix E, et al. Natural history of serum HIV-1 RNA levels in 330 patients with a known date of infection. The SEROCO Study Group. AIDS. 2000;14:123-131.
18. 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. 2007;140:222-227.
19. Rouzioux C, Hubert JB, 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. 2005;192:46-55.
20. Jacqmin Gadda H, Sibillot S, Proust C, et al. Robustness of the linear mixed model to misspecified error distribution. Comput Stat Data Anal. 2007;51:5142-5144.
21. Drylewicz J, Matheron S, Lazaro E, et al. Comparison of viro-immunological marker changes between HIV-1 and HIV-2-infected patients in France. AIDS. 2008;22:457-468.
22. Schacker TW, Hughes JP, Shea T, et al. Biological and virologic characteristics of primary HIV infection. Ann Intern Med. 1998;128:613-620.
23. Deeks SG, Kitchen CM, Liu L, et al. Immune activation set point during early HIV infection predicts subsequent CD4+ T-cell changes independent of viral load. Blood. 2004;104:942-947.
24. Eggena MP, Barugahare B, Okello M, et al. T cell activation in HIV-seropositive Ugandans: differential associations with viral load, CD4+ T cell depletion, and coinfection. J Infect Dis. 2005;191:694-701.
25. Tsegaye A, Messele T, Tilahun T, et al. Immunohematological reference ranges for adult Ethiopians. Clin Diagn Lab Immunol. 1999;6:410-414.
26. Taylor BS, Sobieszczyk ME, McCutchan FE, et al. The challenge of HIV-1 subtype diversity. N Engl J Med. 2008;358:1590-1602.
27. Rodriguez B, Sethi AK, Cheruvu VK, et al. Predictive value of plasma HIV RNA level on rate of CD4 T-cell decline in untreated HIV infection. JAMA. 2006;296:1498-1506.
28. Minga AK, Anglaret X, d' Aquin Toni T, et al. HIV-1 DNA in peripheral blood mononuclear cells is strongly associated with HIV-1 disease progression in recently infected West African adults. J Acquir Immune Defic Syndr. 2008;48:350-354.
29. Markers as time-dependent covariates in relative risk regression. MAP Workshop (Multi-cohort Analysis Project). Stat Med. 1993;12:2087-2098.
30. Linton PJ, Dorshkind K. Age-related changes in lymphocyte development and function. Nat Immunol. 2004;5:133-139.
31. Effros RB, Fletcher CV, Gebo K, et al. Aging and infectious diseases: workshop on HIV infection and aging: what is known and future research directions. Clin Infect Dis. 2008;47:542-553.
32. Touloumi G, Pantazis N, Babiker AG, et al. Differences in HIV RNA levels before the initiation of antiretroviral therapy among 1864 individuals with known HIV-1 seroconversion dates. AIDS. 2004;18:1697-1705.
33. Duvignac J, Anglaret X, Kpozehouen A, et al. CD4+ T-lymphocytes natural decrease in HAART-naive HIV-infected adults in Abidjan. HIV Clin Trials. 2008;9:26-35.
34. World Health Organization. Guidelines on co-trimoxazole prophylaxis for HIV-related infections among children, adolescents and adults. Recommendations for a public health approach. 2006. Available at: http://www.who.int/hiv/pub/guidelines/ctxguidelines.pdf
. Accessed December 4, 2007.
35. Toure S, Kouadio B, Seyler C, et al. Rapid scaling-up of antiretroviral therapy in 10,000 adults in Cote d'Ivoire: 2-year outcomes and determinants. AIDS. 2008;22:873-882.
36. Hallett TB, Gregson S, Dube S, et al. The impact of monitoring HIV patients prior to treatment in resource-poor settings: insights from mathematical modelling. PLoS Med. 2008;5:e53.
Africa; CD4 cell count; France; HIV infection; resource-limited setting; seroconversion
© 2010 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.