Nucleoside reverse transcriptase inhibitors (NRTIs) were the first antiretroviral (ARV) drugs to be marketed, starting in 1987.1 Dual NRTI therapy became available in the early 1990s and was subsequently recommended by national and international guidelines.2,3 New classes of potent drugs (protease inhibitors [PIs] and nonnucleoside reverse transcriptase inhibitors [NNRTIs]) started to become available in 1996 and were shown to be associated with significantly increased virologic, immunologic, and clinical efficacy when combined with NRTIs.4,5 Those potent combination antiretroviral therapies (cARTs), consisting of at least 2 NRTIs plus 1 PI or NNRTI, were rapidly recommended for initial treatment of HIV infection in European and American guidelines.6,7 In France, where medical care is free for all HIV-infected patients, dual NRTI regimens represented 10% of all ARV prescribed regimens in late 2000.8 The aim of this study was to determine the characteristics of patients enrolled in the French Hospital Database on HIV (FHDH-ANRS CO4) who were still receiving dual NRTI therapy in 1998 to 2002, the period during which a dual NRTI regimen was no longer recommended for initial therapy but a switch to cART was not recommended for patients with stable disease while on 2 NRTIs.9 We also examined factors associated with a subsequent switch to cART or with discontinuation of all ARV drugs.
For this study, we selected from the FHDH HIV-1-infected patients aged at least 16 years who received dual NRTI therapy for at least 6 months between January 1, 1998, and December 31, 2002, whatever the start date of dual ARV therapy. The date of study inclusion was the date of the first hospital visit or admission, after a follow-up of at least 6 months on dual NRTI therapy during the period 1998 to 2002. We excluded patients previously enrolled in blind clinical trials of ARV therapy, patients with no known CD4 cell count within 6 months before inclusion, and pregnant women. We also excluded patients who were switched to an ARV regimen that did not comply with contemporary French guidelines (single-agent PI, NNRTI, or NRTI therapy or dual-agent therapy with a NRTI plus a PI or a NNRTI or a PI and a NNRTI).
Three groups of patients were defined on the basis of ARV treatment after the inclusion date: (1) patients who switched to cART (at least 3 ARV drugs or 2 ritonavir-boosted PIs), (2) patients who discontinued ARV therapy for at least 6 months, and (3) patients treated with a dual NRTI regimen throughout the study period. Follow-up data were censored on the date of interruption of dual NRTI therapy, a switch to cART, death, the last follow-up visit, or December 31, 2003, whichever occurred first.
Kaplan-Meier life tables were used to evaluate the time taken to switch from dual NRTI therapy to cART or to ARV drug discontinuation. The 3-year probabilities of switching to cART or discontinuing ARV drugs were estimated by using a competing-risk approach.10 Factors associated with switching to cART and factors associated with ARV drug interruption were identified by using univariate and multivariable Cox proportional hazards models. The univariate model included gender, age, place of birth (ethnic origin), region of care, HIV transmission group, calendar year of inclusion in the study, duration of dual NRTI therapy before baseline, type of dual NRTI therapy, CD4 cell count at baseline, AIDS status at baseline and during follow-up (time-dependent variable), CD4 cell count at the last date of follow-up, and plasma viral load (pVL) during the follow-up period (time-dependent variable). All variables associated with the risk of switching to cART or of interrupting ARV drugs in the univariate model (P < 0.20) were included in the multivariable models. Statistical analyses were performed using SAS statistical package version 9.1 (SAS Institute, Cary, NC).
Among the 52,981 patients who received at least 1 ARV drug for at least 6 months during the study period, the treatment consisted of 2 NRTIs in 6832 (12.9%), of whom 5222 patients were included in this study. Dual NRTI therapy was usually started before or during 1998. At the date of start of the dual NRTI therapy, the median CD4 count was 357 cells/mm3 (interquartile range [IQR]: 263 to 476 cells/mm3) (missing values for 1007 patients [19.3%]), and the median pVL was 13,788 copies/mL (IQR: 1534 to 54,000 copies/mL) (missing values for 2045 patients [39.2%]). At inclusion in the study, the most frequently prescribed regimens were zidovudine/lamivudine (1464 cases [28.0%]), stavudine/lamivudine (1265 cases [24.2%]), stavudine/didanosine (907 cases [17.4%]), zidovudine/didanosine (790 cases [15.1%]), and zidovudine/zalcitabine (725 cases [13.9%]). The place of birth was France in 4807 cases (92.1%) and sub-Saharan Africa or Haiti in 284 cases (5.4%). The hospitals caring for the patients were located in the Paris area in 2437 cases (46.7%) and in the Provence Alpes Côte d'Azur region (southern France) in 1119 cases (21.4%). The characteristics of the patients are shown in Table 1 according to their treatment after inclusion. During a follow-up of 9054 person-years (PY), 2956 patients were switched to cART (incidence rate [IR] = 32.7/100 PY, 95% confidence interval [CI]: 30.7 to 33.7), 675 discontinued their ARV therapy (IR = 7.5/100 PY, 95% CI: 7.0 to 8.0), and 1591 (30.5%) received dual NRTI therapy throughout follow-up. The median duration of dual NRTI therapy after inclusion was 18.6 months (IQR: 5.8 to 47.9 months). Four hundred seventy-four patients discontinued the dual NRTI regimen at inclusion in the study (these patients had received at least 6 months of dual NRTI therapy during 1998 to 2002). The 3-year probability of switching to cART was 55.2% (95% CI: 53.8 to 56.7). The 3-year probability of discontinuing all ARV therapy was 10.9% (95% CI: 10.1 to 11.8).
Factors Associated With Switching to cART or With ARV Discontinuation
In univariate analysis, age, ethnic origin, and region of care were not associated with switching to cART or with ARV discontinuation. After adjustment, the likelihood of switching to cART was lower in women than in men and lower in intravenous drug users than in homosexual men (Table 2). Compared with patients who were AIDS-free at inclusion, patients who had an AIDS-defining event (ADE) before inclusion or during follow-up were 1.41 times more likely to switch to cART. Patients with a pVL <500 copies/mL on at least 1 occasion during follow-up had a 30% lower likelihood of switching to cART than patients whose pVL always remained >500 copies/mL, whereas patients with lower CD4 cell values at inclusion or at the end of follow-up had a higher likelihood of switching to cART. After adjustment, the likelihood of discontinuing all ARV drugs was higher among women than among men and among intravenous drug users and heterosexual patients than among homosexual men (see Table 2). Compared with patients who were AIDS-free at inclusion, ARV drug discontinuation was not more likely among patients who had an ADE before inclusion or during follow-up. Patients with at least 1 pVL value <500 copies/mL during follow-up had a 37% lower likelihood of discontinuing all ARV drugs than patients whose pVL always remained >500 copies/mL. Patients with lower CD4 cell counts at inclusion or at the end of follow-up had a lower likelihood of discontinuing all ARV drugs. Ongoing zidovudine/zalcitabine or didanosine/stavudine therapy at inclusion was more likely than zidovudine/lamivudine to be switched to cART or to be interrupted.
Among 5222 patients enrolled in the FHDH who received dual NRTI therapy for at least 6 months during the period 1998 to 2002, 1591 (30.5%) patients received the same ARV strategy throughout this period. The 3-year probabilities of switching to cART and of discontinuing all ARV drugs were 55.2% and 10.9%, respectively. Factors associated with a change in the dual NRTI strategy were gender, HIV transmission group, nature of the dual NRTI regimen, AIDS status, CD4 cell count, and pVL; place of birth (ethnic origin) and place of care were not associated with either probability.
Most of the patients who remained on dual NRTI therapy during the study period had stable disease, with low viremia (<500 copies/mL at least once in 65% of cases) and no pVL >500 copies/mL during the follow-up period for 765 patients (48%), a relatively satisfactory CD4 count (>200 cells/mm3 in 95% of cases at baseline and in 89% at the last follow-up visit), and no clinical progression during follow-up, as in other comparable cohort studies.11 Dual NRTI therapy in ARV-naive patients can maintain low viral replication or an undetectable pVL, particularly if introduced when the CD4 count is >200 to 350 cells/mm3 and viremia is low.12,13 Despite guidelines recommending first-line use of cART as soon as the first PI and NNRTI drugs became available, many clinicians thought that dual NRTI therapy was an acceptable alternative for patients diagnosed and treated early in the course of HIV infection.11 Moreover, until recently, French national guidelines did not recommend a switch to cART for patients with stable disease while on dual NRTI therapy.14 Subsequent studies showed that this regimen often failed to maintain viral suppression, leading to NRTI resistance mutations, and more potent treatment strategies were thus proposed.11
Other studies showed significant differences in access to standard-of-care management depending on gender, risk group, ethnic origin, and health insurance status, with men (mainly homosexual men) being more likely to receive cART and women, blacks, intravenous drug users (especially current drug injectors), recently incarcerated patients, uninsured or private/self-paid insured patients, and poorly educated patients being less likely to receive cART, whatever the CD4 cell count or pVL.15-17 We were unable to adjust the analysis for indicators of socioeconomic status, such as education and income, that have been shown to be associated with gender and to influence access to HAART and clinical outcome18 or to assess adherence data that are not recorded in the FHDH database. Possible reasons for the less frequent switch to cART in women than in men include a lack of time because of child care activities19 and a higher incidence of adverse effects of drugs in women,20 leading physicians or women to postpone cART. Women were also more likely than men to discontinue dual NRTI therapy. It is possible that pregnancies were undernotified in the FHDH database and that some women who received dual NRTI therapy solely to prevent vertical transmission were enrolled. Access to ARV treatment in industrialized countries is influenced by ethnic origin,21 but this was not the case in our study population. Free access to HIV/AIDS care in France (at least until recently) may explain this discrepancy with other studies.
Delayed access to care has been shown as a strong prognostic factor of mortality;22 however, cohort studies failed to show any difference in midterm clinical outcome between patients with suboptimal therapy and patients with cART.20 Several genetic, immunologic, and virologic factors have been implicated in slower disease progression in some patients.23-26 A small number of patients (long-term nonprogressors) have a relatively normal CD4 cell count and remain free of AIDS for many years.27 Other patients have stable disease (CD4 count >300 cells/mm3 and pVL <10,000 copies/mL) despite suboptimal therapy with regimens such as 2 NRTIs and share virologic characteristics with long-term nonprogressors.25 The high rate of interruption of dual NRTI therapy soon after inclusion in the study (26% during the first 6 months) shows that clinicians gradually adopted contemporary guidelines after 1998. Before 1998, it was recommended to start treatment as soon as the CD4 count fell to <500 cells/mm3 and the pVL rose to >10,000 to 30,000 copies/mL, although more conservative CD4 cell count-based guidelines were subsequently adopted.3,6,28
In conclusion, until recent years, some patients (especially women and intravenous drug users) were still receiving dual NRTI therapy. Indeed, until 2004, French guidelines did not recommend a switch to cART for patients with stable disease on dual NRTI therapy, and after 2004, no recommendations were stated for these patients.14 Because of the high risk of NRTI resistance mutations in case of no durable virologic control with cross-resistance to all drugs of the class, these regimens have to be gradually replaced by cART or simply discontinued when clinical or biologic status permits.
The authors thank all the participants in the FHDH, especially the research assistants, without whom this work would not have been possible.
1. Fischl MA, Richman DD, Grieco MH, et al. The efficacy of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex-a double-blind, placebo-controlled trial. N Engl J Med
2. Delta Coordinating Committee. A randomised double-blind controlled trial comparing combinations of zidovudine plus didanosine or zalcitabine with zidovudine alone in HIV infected individuals. Lancet
3. Dormont J et le groupe d'experts sur l'infection par le VIH. Traitements antirétroviraux. In: Prise en charge thérapeutique des personnes atteintes par le VIH. France. 1996:185-220.
4. Collier AC, Coombs RW, Schoenfield DA, et al. Treatment of human immunodeficiency virus infection with saquinavir, zidovudine and zalcitabine. N Engl J Med
5. Palella FJ Jr, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med
6. Dormont J et le groupe d'experts sur l'infection par le VIH. Traitement initial. In: Stratégies d'utilisation des antirétroviraux dans l'infection par le VIH. France. 1997:11-15.
7. Panel on Clinical Practices for Treatment of HIV Infection. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. Ann Intern Med
9. Delfraissy JF et le groupe d'experts sur l'infection par le VIH. Suivi d'un patient sous traitement antiretroviral. In: Prise en charge thérapeutique des personnes infectées par le VIH. France. 2002:47-62.
10. Satagopan JM, Ben-Porat L, Berwick M, et al. A note on competing risks in survival data analysis. Br J Cancer
11. Manfredi R, Calza L, Chiodo F. Dual nucleoside analogue treatment in the era of highly active antiretroviral therapy (HAART): a single-centre cross-sectional survey. J Antimicrob Chemother
12. Flandre P, on behalf of the Delta Coordinating Committee. Patients with HIV-1 RNA below 1000 copies/mL after 48 weeks on dual nucleoside analogue combination therapies. AIDS
13. Ungsedhapand C, Srasuebkul P, Cardiello P, et al. Three-year durability of dual-nucleoside versus triple-nucleoside therapy in a Thai population with HIV infection. J Acquir Immune Defic Syndr
14. Delfraissy JF et le groupe d'experts sur l'infection par le VIH. Suivi d'un patient sous traitement antiretroviral. In: Prise en charge thérapeutique des personnes infectées par le VIH. France. 2004:59-77.
15. Ghani AC, Donnelly CA, Anderson RM. Patterns of antiretroviral use in the United States of America: analysis of three observational databases. HIV Med
16. Cunningham WE, Markson LE, Andersen RM, et al. Prevalence and predictors of highly active antiretroviral therapy use in patients with HIV infection in the United States. J Acquir Immune Defic Syndr
17. Celentano DD, Vlahov D, Cohn S, et al. Self-reported antiretroviral therapy in injection drug users. JAMA
18. Wood E, Montaner JSG, Chan K, et al. Socioeconomic status, access to triple therapy, and survival from HIV-disease since 1996. AIDS
19. Moore J, Schuman P, Schoenbaum E, et al. Severe adverse life events and depressive symptoms among women with, or at risk for, HIV infection in four cities in the United States of America. AIDS
20. Murri R, Cozzi Lepri A, Phillips AN, et al. Access to antiretroviral treatment, incidence of sustained therapy interruptions, and risk of clinical events according to sex. Evidence from the I.Co.N.A. study. J Acquir Immune Defic Syndr
21. Moore RD, Stanton D, Gopalan R, et al. Racial differences in the use of drug therapy for HIV disease in an urban community. N Engl J Med
22. Lanoy E, Mary-Krause M, Tattevin P, et al. Frequency, determinants and consequences of delayed access to care for HIV infection in France. Antivir Ther
23. Ioannidis JP, Rosenberg PS, Goedert JJ, et al. Effects of CCR5-Δ32, CCR2-64I, and SDF-1 3′A alleles on HIV-1 disease progression: an international meta-analysis of individual-patient data. Ann Intern Med
24. Wei X, Liang C, Gotte M, et al. Negative effect of the M184 mutation in HIV-1 reverse transcriptase on initiation of viral DNA synthesis. Virology
25. Mologni D, Citterio P, Menzaghi B, et al. Vpr and HIV-1 disease progression: R77Q mutation is associated with long-term control of HIV-1 infection in different groups of patients. AIDS
26. Vieillard V, Costagliola D, Simon A, et al. Specific adaptive humoral response against a gp41 motif inhibits CD4 T-cell sensitivity to NK lysis during HIV-1 infection. AIDS
27. Rodes B, Toro C, Paxinos E, et al. Differences in disease progression in a cohort of long-term non-progressors after more than 16 years of HIV-1 infection. AIDS
28. Delfraissy JF et le groupe d'experts sur l'infection par le VIH. Traitement antiretroviral. In: Prise en charge thérapeutique des personnes infectées par le VIH. France. 1999:51-84.