Highly active antiretroviral therapy (HAART) has a major beneficial impact on HIV-related mortality and morbidity [1–7]. However, metabolic, hepatic and mitochondrial toxicity [8–10] and intercurrent disorders [11–13] often lead to treatment interruption. Chronically HIV-infected patients who discontinue HAART usually experience a viral rebound within 1 or 2 weeks, both viral load and the CD4 cell count returning to values near baseline [14–19].
Few studies have focused on patients who interrupted HAART started during primary HIV-1 infection (PHI) [20–24]. Viral replication remained low or undetectable in some patients treated during acute infection whose specific anti-HIV CD4 lymphocyte response was restored after treatment interruption [22,25–27]. This raised the possibility that transient treatment during PHI might be beneficial. However, as recently reviewed , the ability of early antiretroviral therapy to lower the virological ‘set-point’ after interruption, relative to never-treated patients, was questioned in some other studies [23,24]. The influence of patient and treatment characteristics on HIV RNA dynamics after withdrawal of HAART initiated during PHI needs to be investigated, as this would permit more accurate comparisons with the natural history of HIV infection.
The objectives of this study were to assess the influence of patient characteristics, treatment precocity (how early after infection) and the duration of the sustained virological response to HAART on HIV RNA levels after withdrawal of effective HAART started during PHI and to compare HIV RNA levels after withdrawal of effective HAART initiated during PHI with values reached during the natural history, at the same time point since infection.
Description of the PRIMO cohort
The ongoing French ANRS EP8 PRIMO cohort enrolled 393 HIV-infected patients between November 1996 and February 2003 in 66 French hospitals [29,30]. The study was approved by the local institutional review board (Paris-Cochin, France), and all the subjects gave their informed written consent. Subjects diagnosed during or soon after PHI, whether symptomatic or not, were enrolled. Recent infection was confirmed by an incomplete Western blot pattern (absence of anti-p68 and anti-p34); positive p24 antigenaemia or detectable plasma HIV RNA with a negative or weakly reactive enzyme-linked immunosorbent assay (ELISA); or an interval of less than 6 months between an authenticated negative and a positive ELISA. The date of infection was estimated as the date of symptom onset minus 15 days, the date of the incomplete Western blot minus 1 month, or the mid-point between a negative and a positive ELISA. The interval between the estimated date of infection and enrollment could not exceed 6 months (median interval in the PRIMO cohort was 40 days). A PHI was defined as symptomatic if at least one symptom related to the HIV acute viral syndrome was present. Patients had to be naive for antiretroviral drugs at enrollment. Clinical and biological examinations were planned at each follow-up visit, and included CD4 cell count and plasma HIV RNA assays.
PRIMO study population
The study population comprised 58 patients who interrupted HAART for more than 15 days after having had a sustained virological response (no viral rebound > 500 copies/ml after the first HIV RNA measurement < 500 copies/ml) until HAART interruption, irrespective of the time to HAART response (time between HAART initiation and the first HIV RNA measurement < 500 copies/ml) and the duration of the sustained virological response (time between the first HIV RNA measurement < 500 copies/ml and the interruption of HAART). HAART was defined as three nucleoside reverse transcriptase inhibitors (NRTI) including abacavir, or two NRTI combined with one or more protease inhibitors and/or one non-nucleoside reverse transcriptase inhibitor (NNRTI). All the patients started taking HAART less than 3.5 months after infection. When patients interrupted treatment more than once, only the first interruption was analysed.
HIV RNA assays in the PRIMO study population
Plasma HIV-1 RNA was quantified at each follow-up visit, using one of three assays approved in France: 74% by reverse transcriptase polymerase chain reaction (Amplicor, Roche Diagnostics, Meylan, France), 24% by bDNA (Quantiplex, Bayer Diagnostics, Eragny, France) and 2% by nucleic acid sequence-based amplification (NASBA, Organon Technika, Fresnes, France). HIV RNA values were log10-transformed. Samples below the detection limit were attributed half the value of the assay cut-off [31–33]. When analysing HIV RNA levels after stopping HAART, all values recorded between HAART interruption and either the end of follow-up (when the patient remained untreated) or treatment resumption were considered. The cut-off date for this analysis was 30 June 2003, giving a median follow-up of 35 months from enrollment. Among the 58 patients, a total of 278 HIV RNA values determined during a median period of HAART interruption of 10.2 months were available for analysis.
SEROCO study population
The French ANRS SEROCO cohort enrolled 1551 HIV-infected patients starting in 1988. As described elsewhere , 431 patients had a known date of infection and underwent serial HIV RNA assays (Amplicor, Roche Diagnostics) on sera collected between 1988 and 1995 and stored at −180°C. Of these 431 patients, patients were selected for the current study if they would have been enrolled in the PRIMO cohort had this cohort been implemented at that time, using the PRIMO inclusion criteria detailed above. None of these patients was receiving antiretroviral therapy at enrollment. The median date of enrollment was November 1989.
HIV RNA assays in the SEROCO study population
To study HIV RNA kinetics during the natural history of HIV infection, all values recorded for the 116 patients were included that were obtained before initiation of HAART or before 1 January 1996, whichever came first. As in the PRIMO population, all samples below the detection limit were attributed half the value of the assay cut-off. As described in the literature, a value of 0.28 log10 copies/ml was added to serum HIV RNA values above the detection limit in order to make them comparable to plasma HIV RNA values . A total of 799 HIV RNA measurements obtained during a median period of 58.1 months since infection were available.
HIV RNA levels after HAART withdrawal were modelled for the 58 PRIMO patients and for the 116 SEROCO patients. Linear mixed-effects models were used to take into account both the correlation between measurements and changes in HIV RNA kinetics . Different time thresholds at which the slopes were allowed to change were tested and chosen using the profile likelihood method . In both cohorts, a model with three slopes was obtained, in which the first slope and the changes in the other two slopes differed significantly from 0 (P < 0.01). The intercept and the three slopes had both fixed and random effects (i.e., the intercept and slopes were specific to each patient). In the PRIMO cohort, the effect of the patient characteristics, the precocity of treatment (0 or 1 band on Western blot versus ≥ 2), and the duration of the sustained virological response to HAART on HIV RNA levels after HAART interruption was assessed by introducing interaction terms between these variables and each slope . In other words, the three slopes were allowed to change according to the value of the variable. Quantitative variables were categorized according to the median or tertile when the linearity assumption was rejected. This model allowed HIV RNA values to be estimated 12 months after stopping HAART according to the values of the variables included in the model. For continuous variables, arbitrarily fixed values were attributed to permit numerical estimations.
Before comparing HIV RNA levels after withdrawal of effective HAART with values reached during the natural history of infection, similarity between the baseline HIV RNA and CD4 cell count levels in the 58 PRIMO patients and the 116 SEROCO patients was confirmed by using restricted quadratic-spline regressions with two knots, adjusted for gender, age and time between infection and enrollment .
For the comparison with HIV RNA levels reached during the natural history of the infection, HIV RNA levels in the PRIMO patients 36 months following infection (12 months after treatment interruption) was estimated by fixing the time between infection and HAART interruption at 24 months. HIV RNA values in the SEROCO cohort were estimated at the same time since infection (36 months), after adjustment for gender and age, by giving to these variables in the model the observed values of the mean among the PRIMO patients (22.4% of women and 34.2 years, respectively).
All P values are two-tailed. Statistical analyses were carried out with SAS version 8.2 (SAS Institute, Cary, North Carolina, USA). The PROC MIXED procedure was used for linear mixed-effects models; the plot of the residual errors against time for each observation was used to check the goodness of fit of the models.
PRIMO study population (1996–2003)
The PRIMO study population comprised 58 patients (Table 1), including 13 women (22%). The median baseline plasma HIV RNA level and CD4 cell count at HAART initiation were 4.9 log10 copies/ml and 560 × 106 cells/l, respectively. The median interval between infection and HAART initiation was 45 days. Among the 44 symptomatic patients, the interval between symptom onset and HAART initiation was 27 days. Seven patients were acute seroconverters at HAART initiation (none or one Western blot band), while the other patients were early seroconverters (two or more Western blot bands). Most patients (86%) started with a protease inhibitor-containing regimen; two started with a triple NRTI regimen (including abacavir), and six started with a protease inhibitor-sparing regimen containing one NNRTI. Of note, these 58 patients did not differ from the 335 other patients for baseline characteristics.
The median duration of sustained virological responses to HAART was 17.3 months. The median increase in the CD4 cell count between initiation and interruption of HAART was 212 × 106 cells/l. The interruptions were attributable to a voluntary decision from the patient or his or her physician for 44, to treatment-related adverse events for 12 and to an intercurrent acute disease in two. The median duration of HAART interruption was 10.2 months (range, 0.9–47.5): 22 patients resumed HAART in a median time of 2.5 months after the interruption, whereas the 36 other patients remained off treatment at the end of their follow-up within a median period of 16.5 months.
Modelling of HIV RNA after treatment interruption in the PRIMO cohort
The initial increase in HIV RNA during the first 27 days after HAART interruption was estimated at 0.065 log10 copies/ml per day, followed by a milder increase between 27 days and 6.5 months (0.108 log10 copies/ml per month), and by an even milder increase of 0.014 log10 copies/ml per month thereafter (Fig. 1).
Table 2 shows estimated HIV RNA values 12 months after stopping HAART, according to the patients’ characteristics at HAART initiation, the precocity of HAART, and the duration of sustained virological responses to HAART. Twelve months after stopping HAART, women had lower viral load than men, but they also had lower baseline viral loads (4.41 versus 5.06 log10 copies/ml; P = 0.05), as already described in the natural history of HIV-1 infection [40,41]. Viral load during HAART interruption was not associated with age or with the symptomatic nature of PHI. Viral load after stopping HAART was strongly associated with baseline viral load, the baseline CD4 cell count and the CD4 cell count at HAART interruption. The precocity of HAART initiation did not influence viral load 12 months after stopping HAART (4.13 and 3.95 log10 copies/ml, respectively, in patients with acute and early-phase infection, as defined above; P = 0.74). Similar results were obtained when the acute versus early-phase comparison was restricted to a more homogeneous population consisting only of men who had a symptomatic PHI (4.15 log10 copies/ml in six acute-phase patients and 4.32 log10 copies/ml in 39 early-phase patients; P = 0.78). The precocity of treatment also had no influence on viral load after stopping HAART, when defined by the time since infection or since PHI symptom onset. Viral load after HAART interruption was not associated with the time taken to respond to HAART, or with the duration of sustained virological responses.
As variables associated with viral load after HAART interruption were also associated with baseline viral load, an adjustment was made for baseline viral load in order to assess the independent role of the other variables. Women still tended to have lower viral load than men after stopping HAART (difference of 0.81 log10 copies/ml; P = 0.07). Viral load after HAART interruption was no longer significantly associated with the baseline CD4 cell count (P = 0.12) but was still associated with the CD4 cell count at HAART interruption, independently of viral load at HAART initiation (P = 0.02). In all these adjusted models, viral load after stopping HAART remained strongly associated with baseline viral load (P < 0.01).
Estimated viral load 36 months after infection, and 12 months after treatment interruption was 3.95 log10 copies/ml [95% confidence interval (CI), 3.57–4.32].
SEROCO study population (1988–1995)
The SEROCO study population comprised 116 seroconverters (Table 3), including 33 women (28%). Median age at enrollment was 27 years. No difference in the baseline viral load or CD4 cell count was found between the 116 SEROCO patients and the 58 PRIMO patients after adjustment for gender, age and the time between infection and enrollment: −0.06 log10 copies/ml (P = 0.74) and −18 × 106 cells/l CD4 cells (P = 0.76) for the PRIMO patients compared with the SEROCO patients. This allowed HIV RNA kinetics to be compared in the two populations.
Modelling of HIV RNA kinetics during the natural history of infection (SEROCO cohort)
After modelling HIV RNA kinetics in the SEROCO cohort, with the same gender and age distribution as in the PRIMO patients, there was an initial decrease in HIV RNA lasting 3.5 months (0.364 log10 copies/ml per month), followed by a milder decrease between 3.5 and 35.1 months (0.007 log10 copies/ml per month); thereafter HIV RNA increased (0.012 log10 copies/ml per month) (Fig. 1). The model inferred that viral load was 4.11 log10 copies/ml (95% CI: 3.93, 4.30) 36 months after infection.
Consequently, the estimated viral load 12 months after withdrawal of HAART started during PHI was slightly lower than the value reached at the same point during the natural history (36 months) (−0.16 log10 copies/ml; 95% CI, −0.58 to 0.25).
Most previous studies of viral load after HAART withdrawal have examined changes during the first months only [14,15,17,42,43]. The large range in duration of treatment interruption in the PRIMO cohort enabled us to describe viral load dynamics during a longer period. Estimated viral load was 3.95 log10 copies/ml 12 months after withdrawal of effective HAART started during PHI. This estimation is lower than in two previous studies that investigated viral load level 48 weeks after stopping HAART initiated during PHI: 4.25 log10 copies/ml by Fidler et al.  and 4.20 log10 copies/ml by Markowitz et al. . The proportion of women in these studies was lower than in our study, 4% and 7%, respectively, compared with 22% in the PRIMO cohort. Since women were found to have lower HIV RNA 12 months after stopping HAART than men, this lower value after stopping HAART in our study may reflect the higher proportion of women. Out of the 58 PRIMO patients, 22 patients resumed HAART, whereas the 36 other patients remained off treatment at the end of their follow-up. The estimated HIV RNA after stopping HAART was higher among these 22 patients than in the 36 off-treatment patients (3.67 versus 3.03 log10 copies/ml 1 month after interruption; 4.23 versus 3.82 log10 copies/ml 12 months after interruption). To estimate the HIV RNA level 12 months after stopping HAART, the model takes into account both the overall history of viral load of each individual between the date of the interruption to the last measurement available during the interruption, and the measurements still available 12 months after stopping HAART among the 58 initial patients. The model may have underestimated the viral load level if the viral loads recorded 12 months after stopping HAART (and thereafter) were lower than viral loads not recorded at this time point in those patients who resumed HAART before this date. This potential underestimation would lead to an overestimation of the difference in HIV RNA 36 months after infection between the PRIMO patients and the SEROCO patients.
Viral load after HAART interruption was independently associated with viral load at HAART initiation and with the CD4 cell count at HAART interruption. The wide range of durations of sustained virological responses to HAART among PRIMO patients also enabled us to investigate the potential effect of this parameter on viral load after HAART interruption, as this duration was not a selection criterion. This cannot be done in clinical trials, where treatment periods are generally fixed. We found that the duration of sustained virological responses to HAART did not influence viral load values 12 months after HAART interruption, whether this duration was analysed as a continuous or categorized variable.
We found no epidemiological evidence that very early HAART initiation (during acute seroconversion) led to lower viral load after HAART interruption relative to ‘early’ HAART initiation (two or more Western blot bands). The number of acute seroconverters included in this analysis may have been inadequate to show a significant difference, but the other two measures of treatment precocity (i.e., time since infection and time since symptom onset during PHI) yielded the same results. Lacabaratz-Porret et al.  also showed that HIV-specific CD4 and CD8 T cell responses were similar whether HAART was started before or after full seroconversion. Together, these results indicate that very early treatment of HIV infection may not be more beneficial than early treatment.
After adjustment for gender and age, we found that viral load 36 months after infection (12 months after HAART interruption) was only 0.16 log10 copies/ml lower than in untreated patients studied at the same time point. This was a non-randomized comparison, but it is important to note that no difference in baseline viral load or CD4 cell counts was found between the two study populations after adjustment for gender, age and the time between infection and enrollment. Importantly, levels of proviral HIV-1 DNA were also similar. In natural history, HIV RNA levels 36 months after infection were estimated after excluding all measurements (n = 15) recorded after initiation of a multidrug therapy occurring before 1996, but not the 53 measurements recorded within the first year following an initiation of a monotherapy. This may have led to an underestimation of the viral load level 36 months following infection. However, two alternative strategies provided basically the same results. After removing all the measurements recorded within the first year following the initiation of a monotherapy, the viral load level 36 months after infection was 4.11 log10 copies/ml. Alternatively, after correcting the transient effect of monotherapy described in the literature [45–48] by adding 0.4 log10 copies/ml to measurements recorded within 3 months following initiation of monotherapy, and 0.2 log10 copies/ml to measurements recorded within 3–9 months, the estimation of viral load level was 4.13 log10 copies/ml. Although based on observational data, our results, therefore, do not support the notion that transient HAART initiated during PHI can lower the virological set-point that would have been reached had these patients not been treated.
Our results in no way challenge the major beneficial impact of HAART on mortality and morbidity [49,50] and on CD4 cell recovery [38,51–54]. It is important to note that this study was strictly designed to examine the ability of early transient HAART to achieve a lower viral load after treatment interruption than during the natural history of the infection. The possible impact of early transient treatment on clinical, immunological and virological parameters, and time off therapy, needs to be addressed in a randomized study with long-term follow-up comparing early treatment with delayed treatment. The risk–benefit ratio of early short-term treatment should also be considered in the light of possible toxicity and antiretroviral resistance, and the reduction in infectiousness.
We thank all the patients participating in the PRIMO and SEROCO cohorts; the physicians of the PRIMO and SEROCO Group (see Appendix); Nabil Saichi, Myriam Gharib, Ilham Iraqui, Samia Hendou, Soraya Boucherit and Ahlem Chaib for data monitoring; and David Young for editing the manuscript. We also thank Rodolphe Thiebaut for statistical advice.
Sponsorship: the PRIMO and SEROCO cohorts are supported by the Agence Nationale de Recherches sur le Sida. The PRIMO cohort received a grant from Bristol-Meyers-Squibb.
1. Gulick RM. Assessing the benefits of antiretroviral therapy.Ann Intern Med
2. Plana M, Garcia F, Gallart T, Tortajada C, Soriano A, Palou E, et al.Immunological benefits of antiretroviral therapy in very early stages of asymptomatic chronic HIV-1 infection.AIDS
3. Autran B, Carcelain G, Li TS,
Blanc C, Mathez D, Tubiana R, et al.Positive effects of combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease.Science
4. Angel JB, Kumar A, Parato K,
Filion LG, Diaz-Mitoma F, Daftarian P, et al.Improvement in cell-mediated immune function during potent anti-human immunodeficiency virus therapy with ritonavir plus saquinavir.J Infect Dis
5. Lederman MM, Connick E, Landay A,
Kuritzkes DR, Spritzler J, St Clair M, et al.Immunologic responses associated with 12 weeks of combination antiretroviral therapy consisting of zidovudine, lamivudine, and ritonavir: results of AIDS Clinical Trials Group Protocol 315.J Infect Dis
6. Gorochov G, Neumann AU, Kereveur A,
Parizot C, Li T, Katlama C et al.Perturbation of CD4+ and CD8+ T-cell repertoires during progression to AIDS and regulation of the CD4+ repertoire during antiviral therapy.Nat Med
7. Pakker NG, Notermans DW, de Boer RJ,
Roos MT, de Wolf F, Hill A, et al.Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection: a composite of redistribution and proliferation.Nat Med
8. Tsiodras S, Mantzoros C, Hammer S, Samore M. Effects of protease inhibitors on hyperglycemia, hyperlipidemia, and lipodystrophy: a 5-year cohort study.Arch Intern Med
9. Carr A, Miller J, Law M, Cooper DA. A syndrome of lipoatrophy, lactic acidaemia and liver dysfunction associated with HIV nucleoside analogue therapy: contribution to protease inhibitor-related lipodystrophy syndrome.AIDS
10. Saint-Marc T, Partisani M, Poizot-Martin I,
Bruno F, Rouviere O, Land JM, et al.A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy.AIDS
11. Taffe P, Rickenbach M, Hirschel B,
Opravil M, Furrer H, Janin P, et al.Impact of occasional short interruptions of HAART on the progression of HIV infection: results from a cohort study.AIDS
12. Thorner A, Rosenberg E. Early versus delayed antiretroviral therapy in patients with HIV infection: a review of the current guidelines from an immunological perspective.Drugs
13. Bassetti S, Battegay M, Furrer H,
Rickenbach M, Flepp M, Kaiser L, et al.Why is highly active antiretroviral therapy (HAART) not prescribed or discontinued? Swiss HIV Cohort Study.J Acquir Immune Defic Syndr
14. Harrigan PR, Whaley M, Montaner JS. Rate of HIV-1 RNA rebound upon stopping antiretroviral therapy.AIDS
15. Garcia F, Plana M, Vidal C,
Cruceta A, O'Brien WA, Pantaleo G, et al.Dynamics of viral load rebound and immunological changes after stopping effective antiretroviral therapy.AIDS
16. Davey RT, Jr, Bhat N, Yoder C,
Chun, TW, Metcalf, JA, Dewar, R, et al.HIV-1 and T cell dynamics after interruption of highly active antiretroviral therapy (HAART) in patients with a history of sustained viral suppression.Proc Natl Acad Sci USA
17. Neumann AU, Tubiana R, Calvez V, Robert C, Li TS, Agut H, et al.HIV-1 rebound during interruption of highly active antiretroviral therapy has no deleterious effect on reinitiated treatment. Comet Study Group.AIDS
18. Hatano H, Vogel S, Yoder C,
Metcalf JA, Dewar R, Davey Jr RT, et al.Pre-HAART HIV burden approximates post-HAART viral levels following interruption of therapy in patients with sustained viral suppression.AIDS
19. Oxenius A, Price DA, Gunthard HF,
Dawson SJ, Fagard C, Perrin L, et al.Stimulation of HIV-specific cellular immunity by structured treatment interruption fails to enhance viral control in chronic HIV infection.Proc Natl Acad Sci USA
20. Lafeuillade A, Poggi C, Hittinger G, Counillon E, Emilie D. Predictors of plasma human immunodeficiency virus type 1 RNA control after discontinuation of highly active antiretroviral therapy initiated at acute infection combined with structured treatment interruptions and immune-based therapies.J Infect Dis
21. Markowitz M, Vesanen M, Tenner-Racz K,
Cao Y, Binley JM, Talal A, Hurley A, et al.The effect of commencing combination antiretroviral therapy soon after human immunodeficiency virus type 1 infection on viral replication and antiviral immune responses.J Infect Dis
22. Rosenberg ES, Altfeld M, Poon SH,
Phillips MN, Wilkes BM, Eldridge RL, et al.Immune control of HIV-1 after early treatment of acute infection.Nature
23. Markowitz M, Jin X, Hurley A,
Simon V, Ramratnam B, Louie M, et al.Discontinuation of antiretroviral therapy commenced early during the course of human immunodeficiency virus type 1 infection, with or without adjunctive vaccination.J Infect Dis
24. Fidler S, Oxenius A, Brady M,
Clarke J, Cropley I, Babiker A, et al.Virological and immunological effects of short-course antiretroviral therapy in primary HIV infection.AIDS
25. Ho DD. Time to hit HIV, early and hard.N Engl J Med
26. Girard PM, Schneider V, Dehee A,
Mariot P, Jacomet C, Delphin N, et al.Treatment interruption after one year of triple nucleoside analogue therapy for primary HIV infection.AIDS
27. Lisziewicz J, Rosenberg E, Lieberman J,
Jessen H, Lopalco L, Siliciano R, et al.Control of HIV despite the discontinuation of antiretroviral therapy.N Engl J Med
28. Smith DE, Walker BD, Cooper DA, Rosenberg ES, Kaldor JM. Is antiretroviral treatment of primary HIV infection clinically justified on the basis of current evidence?AIDS
29. Dalod M, Dupuis M, Deschemin JC,
Goujard C, Deveau C, Meyer L, et al.Weak anti-HIV CD8(+) T-cell effector activity in HIV primary infection.J Clin Invest
30. Goujard C, Boufassa F, Deveau C, Laskri D, Meyer L. Incidence of clinical lipodystrophy in HIV-infected patients treated during primary infection.AIDS
31. Hughes JP. Mixed effects models with censored data with application to HIV RNA levels.Biometrics
32. Fitzgerald AP, DeGruttola VG, Vaida F. Modelling HIV viral rebound using non-linear mixed effects models.Stat Med
33. Lynn HS. Maximum likelihood inference for left-censored HIV RNA data.Stat Med
34. Hubert JB, Burgard M, Dussaix E,
Tamalet C, Deveau C, Le Chenadec J, et al.Natural history of serum HIV-1 RNA levels in 330 patients with a known date of infection. The SEROCO Study Group.AIDS
35. Griffith BP, Rigsby MO, Garner RB, Gordon MM, Chacko TM. Comparison of the Amplicor HIV-1 monitor test and the nucleic acid sequence-based amplification assay for quantitation of human immunodeficiency virus RNA in plasma, serum, and plasma subjected to freeze-thaw cycles.J Clin Microbiol
36. Laird NM, Ware JH. Random-effects models for longitudinal data.Biometrics
37. Binquet C, Chene G, Jacqmin-Gadda H,
Journot V, Saves M, Lacoste D, Dabis F, et al.Modeling changes in CD4-positive T-lymphocyte counts after the start of highly active antiretroviral therapy and the relation with risk of opportunistic infections: the Aquitaine Cohort, 1996–1997.Am J Epidemiol
38. Le Moing V, Thiebaut R, Chene G,
Leport C, Cailleton V, Michelet C, et al.Predictors of long-term increase in CD4(+) cell counts in human immunodeficiency virus-infected patients receiving a protease inhibitor-containing antiretroviral regimen.J Infect Dis
39. Greenland S. Dose–response and trend analysis in epidemiology: alternatives to categorical analysis.Epidemiology
40. Farzadegan H, Hoover DR, Astemborski J,
Lyles CM, Margolick JB, Markham RB, et al.Sex differences in HIV-1 viral load and progression to AIDS.Lancet
41. Sterling TR, Lyles CM, Vlahov D, Astemborski J, Margolick JB, Quinn TC. Sex differences in longitudinal human immunodeficiency virus type 1 RNA levels among seroconverters.J Infect Dis
42. Fischer M, Hafner R, Schneider C,
Trkola A, Joos B, Joller H, et al.HIV RNA in plasma rebounds within days during structured treatment interruptions.AIDS
43. Ruiz L, Martinez-Picado J, Romeu J,
Paredes R, Zayat MK, Marfil S, et al.Structured treatment interruption in chronically HIV-1 infected patients after long-term viral suppression.AIDS
44. Lacabaratz-Porret C, Urrutia A, Doisne JM,
Goujard C, Deveau C, Dalod M, et al.Impact of antiretroviral therapy and changes in virus load on human immunodeficiency virus (HIV)-specific T cell responses in primary HIV infection.J Infect Dis
45. O'Brien WA, Hartigan PM, Martin D,
Esinhart J, Hill A, Benoit S, et al.Changes in plasma HIV-1 RNA and CD4+ lymphocyte counts and the risk of progression to AIDS. Veterans Affairs Cooperative Study Group on AIDS.N Engl J Med
46. Welles SL, Jackson JB, Yen-Lieberman B,
Demeter L, Japour AJ, Smeaton LM, et al.Prognostic value of plasma human immunodeficiency virus type 1 (HIV-1) RNA levels in patients with advanced HIV-1 disease and with little or no prior zidovudine therapy. AIDS Clinical Trials Group Protocol 116A/116B/117 Team.J Infect Dis
47. Katzenstein DA, Hammer SM, Hughes MD,
Gundacker H, Jackson JB, Fiscus S, et al.The relation of virologic and immunologic markers to clinical outcomes after nucleoside therapy in HIV-infected adults with 200 to 500 CD4 cells per cubic millimeter. AIDS Clinical Trials Group Study 175 Virology Study Team.N Engl J Med
48. Delta Coordinating Committee and Delta Virology Committee. HIV-1 RNA response to antiretroviral treatment in 1280 participants in the Delta trial: an extended virology study.AIDS
49. Egger M, May M, Chene G,
Phillips AN, Ledergerber B, Dabis F, et al.Prognosis of HIV-1-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies.Lancet
50. Palella FJ Jr, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al.Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators.N Engl J Med
51. Mocroft A, Devereux H, Kinloch-de-Loes S,
Wilson D, Madge S, Youle M, et al.Immunological, virological and clinical response to highly active antiretroviral therapy treatment regimens in a complete clinic population. Royal Free Centre for HIV Medicine.AIDS
52. Grabar S, Le Moing V, Goujard C,
Leport MD, Kazatchkine D, Costagliola L, et al.Clinical outcome of patients with HIV-1 infection according to immunologic and virologic response after 6 months of highly active antiretroviral therapy.Ann Intern Med
53. Li TS, Tubiana R, Katlama C, Calvez V, Ait Mohand H, Autran B. Long-lasting recovery in CD4 T-cell function and viral-load reduction after highly active antiretroviral therapy in advanced HIV-1 disease.Lancet
54. Opravil M, Ledergerber B, Furrer H,
Hirschel B, Imhof A, Gallant S, et al.Clinical efficacy of early initiation of HAART in patients with asymptomatic HIV infection and CD4 cell count > 350 × 106/l.AIDS
The PRIMO Cohort Study Group: C. Caulin, E. Badsi, M. Bendenoun, J. Cervoni, V. Vincent (Hôpital Lariboisière; Paris); J. M. Decazes, D. Ponscarme, D. Sereni, C. Lascoux, P. Morel, F. Timsit (Hôpital St Louis; Paris); J. F. Delfraissy, C. Goujard, Y. Quertainmont (Hôpital Bicêtre; Le Kremlin Bicêtre, Paris); C. Katlama, H. Ait Mohand, S. Herson, A. Simon, E. Capitaine, J. Dagron (Hôpital Pitié-Salpétrière; Paris); J. L. Vildé, C. Jestin, C. Jadand P. Yeni, E. Bouvet, I. Fournier, P. Campa, S. Abgrall (Hôpital Bichat; Paris); D. Sicard, D. Salmon, G. Spiridon (Hôpital Cochin; Paris); B. Dupont, J. P. Viard (Hôpital Necker; Paris); M. Kazatchkine, P. Castiel, D. Batisse (HEGP; Paris); W. Rosembaum, L. Slama, P. Mariot (Hôpital Tenon; Paris); E. Oksenhendler, L. Gérard, (Hôpital St Louis; Paris); P. M. Girard, D. Samanon-Bollens, O. Picard (Hôpital St Antoine; Paris); G. Huchon, A. Compagnucci (Hôpital Hôtel-Dieu; Paris); M. Gayraud, L. Bodard (IMM; Paris); C. Miodovski (Paris); J. Beylot, P. Morlat, D. Malvy, M. Bonarek, F. Bonnet (Hôpital St André; Bordeaux); J. M. Ragnaud, I. Raymond (Hôpital Pellegrin; Bordeaux); A. P. Blanc, T. Allègre (Aix en Provence); E. Rouveix, D. Régnier, S. Morelon, C. Dupont (Hôpital A. Paré; Boulogne); J. M. Livrozet, F. Jeanblanc, P. Chiarello (Hôpital E. Herriot; Lyon); F. Raffi, V. Reliquet, E. Billaud, J. L. Esnault (Hôpital Hôtel-Dieu; Nantes); P. Henon, G. Beck-Wirth, C. Beck (Mulhouse); R. Thomas, F. Souala, C. Bouvier (Hôpital Pontchaillou; Rennes); A. Cabié, S. Abel (Fort de France); P. Massip, M. Obadia (Hôpital Purpan; Toulouse); F. Saint-Dizier (Hôpital Ducuing; Toulouse); J. Reynes, V. Baillat, V. Lemoing (Montpellier); C. Bazin, M. Six, R. Verdon (Caen); H. Gallais, I. Ravaux, C. Tomei (Hôpital La Conception; Marseille); J. C. Messmer, B. Delmas, M. Saada (Perpignan); A. Devidas, P. Chevojon, P. Kousignian (Corbeil); J. Achard, P. Fialaire, J.M. Chennebault (Angers); A. Sobel, P. Lesprit, A.S. Lascaux (Hôpital H. Mondor; Créteil); D. Merrien (Compiegne); J. Beytout, C. Jacomet (Hôpital Hôtel Dieu; Clermont-Ferrand); P. Galanaud, F. Boué, J. Polo de Veto (Hôpital A. Béclère; Clamart); J. P. Cassuto, C. Sohn, E. Rosenthal (Hôpital L'Archet; Nice); B. Hoen, C. Drobacheff (Hôpital St Jacques; Besançon); P. Choutet, P. Nau, F. Bastides (Tours); Y. Mouton, A. Cheret (Tourcoing); B. Ponge, L. Fournier (Melun); P. Dellamonica, S. Chaillou (Hôpital L'Archet; Nice); G. Dien, C. Daniel, C. Devaurs (St-Brieuc); P. Canton, L. Boyer (Nancy); C. Trepo, C. Augustin-Normand (Hôpital Hôtel-Dieu; Lyon); J. Laffay, A. Greder Belan (Hôpital A. Mignot; Le Chesnay); Ph. Vinceneux, M. Bloch (Hôpital L. Mourier; Colombes); P. Chavanet, M. Buisson (Dijon); P. Lagarde, F. David (Lagny); O. Bletry, D. Zucman (Hôpital Foch; Suresnes); C Peronne, J. Salomon, P. de Truchis (Hôpital R. Poincaré; Garches); M. Bentata, F. Rouges (Hôpital Avicenne; Bobigny); M. Chousterman, V. Garray (Hôpital Intercommunal; Creteil); A. Regnier (Vichy); J. J. Girard (Loches); P. Moreau, O. Vaillant (Lorient); F. Grihon (Noyon); D. Houlbert (Alençon); N. Plaisance (Colmar); F. Caron, Y. Debab (Rouen); F.Trémolières, V. Perronne (Mantes la Jolie); E. Brottier, L. Faba (La Rochelle); R. Armero, E. Counillon (Frejus); G. Guermonprez, A. Dulioust (Briis s/Forges); P. Boudon, D. Malbec (Aulnay S/Bois); O. Patey, C. Semaille (Villeneuve St Georges); J. Deville, G. Remy, I. Beguinot (Reims). The members of the Scientific Committee of the PRIMO cohort are M. L. Chaix, J. F. Delfraissy, C. Deveau, C. Goujard, M. Harzic, L. Meyer, I. Pellegrin, C. Rouzioux, M. Sinet, A. Venet.
The SEROCO Cohort Study Group: C. Rouzioux, M. Bary, M. Burgard (ACCTES, Paris); D. Séréni, J.-M. Molina, C. Lascoux (Hôpital Saint Louis, Paris); J. F. Delfraissy, P. Lebras, C. Goujard, Y. Quertainmont, S. Poirier (Hôpital de Bicêtre, Le Kremlin-Bicêtre, Paris); J. L. Vildé, C. Leport (Hôpital Bichat, Paris); M. Kazatchkine, L. Weiss, C. Piketti, M. Buisson (Hôpital Européen Georges Pompidou, Paris); B. Dupont, J. P. Viard, A. Maignan, M. Azar (Hôpital Necker, Paris); D. Sicard, D. Salmon-Ceron, L. Guillevin, C. Bernasconi (Hôpital Cochin, Paris); S. Herson, A. Simon-Coutellier, F. Bricaire, C. Katlama (Hôpital Pitié, Salpétrière, Paris); G. Pialoux (Hôpital Tenon, Paris); J. J. Lefrère, P. M. Girard, M. C. Meyohas, J. Lerable, (Institut National de Transfusion Sanguine et Hôpital Saint Antoine, Paris); F. Boué, A. Lévy (Hôpital Antoine Béclère, Clamart); P. Dellamonica, I. Perbost, V. Mondain, M. Quaranta (Hôpital L'Archet, Nice); J.-A. Krivitsky, B. Jarousse, P. Cohen (Hôpital Avicenne, Bobigny); J. P. Cassuto, M. Quaranta (Hôpital Cimiez, Nice); H. Gallais, J. Gallais (Hôpital de la Conception, Marseille); A. Sobel, A. S. Lascaux, C. Jung (Hôpital Henri Mondor, Créteil); D. Vittecoq, L. Escaut, C. Boliot (Hôpital Paul Brousse, Villejuif); J. A. Gastaut, I. Poizot-Martin, G. Fabre (Hôpital Ste Marguerite, Marseille). The members of the Scientific Committee of the SEROCO cohort are: F. Boufassa, J. F. Delfraissy, H. Gallais, L. Meyer, M. C. Meyohas, I. Poizot-Martin, C. Rouzioux, D. Séréni, J. L. Vildé.