J Antimicrob Chemother 2011; 66: 1869 – 1877 doi:10.1093/jac/dkr208 Advance Access publication 1 June 2011
Comparative effectiveness of continuing a virologically effective first-line boosted protease inhibitor combination or of switching to a three-drug regimen containing either efavirenz, nevirapine or abacavir T. Bommenel 1,2, O. Launay 3, J. L. Meynard 4, J. Gilquin 5,6, C. Katlama 1,2,7, A. S. Lascaux 8, A. Mahamat 9, V. Martinez 10, C. Pradier 11, E. Rouveix 12, A. Simon 13, D. Costagliola 1,2,7 and S. Abgrall 1,2,14* on behalf of FHDH-ANRS CO4† 1
*Corresponding author. Tel: +33-1-42-16-42-60; Fax: +33-1-42-16-42-61; E-mail:
[email protected] †Members are listed in the Acknowledgements and at http://www.ccde.fr/
Received 31 January 2011; returned 17 March 2011; revised 26 April 2011; accepted 28 April 2011 Objectives: To compare virological effectiveness in patients who continued on a virologically successful first-line boosted protease inhibitor (PI)-containing combination antiretroviral therapy (cART) regimen or who switched to a PI-free cART including efavirenz, nevirapine or abacavir. Methods: From the French Hospital Database on HIV, we selected 439 patients with undetectable viral load (VL) on a first-line boosted PI-containing cART regimen who switched to a PI-free combination including efavirenz, nevirapine or abacavir. Each of these patients was matched with three patients who continued to take their first-line cART regimen, on the basis of gender, age, CD4 cell count, VL, date of cART initiation and the duration of VL undetectability. Time to virological failure (VF) was analysed with Kaplan – Meier curves and Cox models. Results: The 12 month probabilities of VF were 3.7% and 5.7% in non-switch and switch patients, respectively, and 3.9%, 7.2% and 9.0% in patients switching to efavirenz-, nevirapine- and abacavir-containing cART, respectively. After adjustment, only patients switching to abacavir-containing cART had a higher risk of VF than non-switch patients (adjusted hazard ratio, 1.99; 95% confidence interval, 1.05 –3.79). Conclusions: Switching from a virologically successful first-line boosted PI-containing cART regimen to a nonnucleoside reverse transcriptase inhibitor-containing cART regimen containing either efavirenz or nevirapine is virologically safe, while switching to abacavir-containing cART should be avoided. Keywords: HIV/AIDS, switch, virological effectiveness, boosted protease inhibitor, cohort study
Introduction Potent protease inhibitor (PI)-containing combination antiretroviral therapy (cART) drastically reduces the morbidity and mortality associated with HIV infection,1 while reducing the viral load (VL) below the detection limit and increasing the CD4 cell count.2,3 Such regimens sometimes need to be replaced,
however, mainly because of PI toxicity (hepatic or gastrointestinal disorders, lipodystrophy, hyperlipidaemia, diabetes mellitus and an excess of cardiovascular events independent of metabolic changes), drug/drug interactions or for simplification.4 – 6 Although some non-nucleoside reverse transcriptase inhibitors (NNRTIs) have been reported to affect cholesterol, triglyceride and blood glucose levels,4,7 NNRTIs are not associated with an
# The Author 2011. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail:
[email protected]
1869
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
INSERM U943, Paris F-75013, France; 2UPMC Universite´ Paris 06, UMR_S 943, Paris F-75013, France; 3Universite´ Paris Descartes, AP-HP, Hoˆpital Cochin, Paris F-75014, France; 4AP-HP, Hoˆpital Saint-Antoine, Service des maladies infectieuses et Tropicales, Paris F-75012, France; 5AP-HP, Hoˆpital Hotel-Dieu, unite´ de the´rapeutique en immuno-infectiologie, Paris F-75004, France; 6AP-HP, Hoˆpital Necker, service de maladies infectieuses et tropicales, Paris F-75015, France; 7AP-HP, Groupe hospitalier Pitie´-Salpe´trie`re, Service des maladies infectieuses et Tropicales, Paris F-75013, France; 8AP-HP, Hoˆpital Henri-Mondor, Service d’immunologie clinique, Cre´teil F-94000, France; 9 Centre Hospitalier Andre´e Rosemon, Service des maladies infectieuses et tropicales, Cayenne F-97300, France; 10AP-HP, Hoˆpital Antoine Be´cle`re, service de me´decine interne, Clamart F-92141, France; 11De´partement de Sante´ Publique, Hoˆpital de l’Archet 1, Nice F-06202, France; 12Hoˆpital Ambroise Pare´, Service de me´decine interne, Boulogne F-92100, France; 13AP-HP, Groupe hospitalier Pitie´-Salpe´trie`re, Service de me´decine interne, Paris F-75013, France; 14AP-HP, Hoˆpital Avicenne, Service des maladies infectieuses et tropicales, Bobigny F-93000, France
Bommenel et al.
Methods Patients The FHDH is a large prospective cohort study of HIV-infected patients aged ≥15 years and is managed in 68 French university hospitals. The only enrolment criteria are documented HIV-1 or HIV-2 infection and written informed consent. Trained research assistants use the French Ministry of Health DMI2 software to collect clinical and biological data prospectively at inclusion and at each visit or hospital admission for an HIV-related clinical event or a new treatment prescription, or at least every 6 months, on standardized forms. For this study, we selected 9208 antiretroviral-naive HIV-1-infected patients starting a first-line PI-cART regimen (at least three antiretroviral drugs with at least two NRTIs plus a ritonavir-boosted PI) between 1 January 1998 and 31 December 2008, and who had at least two subsequent follow-up visits. Of these, 1174 patients were excluded because their regimen included abacavir or an NNRTI. We subsequently selected 6011 patients whose plasma VL (pVL) remained undetectable (,500 copies/mL) for ≥6 months on their first-line regimen. The date of virological failure was the date of the first of two consecutive pVL values ≥500 copies/mL recorded ≥1 month apart. The cut-off date for database analysis was 31 December 2008. Exposed patients were patients who switched from their first virologically successful PI-containing regimen to cART composed of two NRTIs plus efavirenz, nevirapine or abacavir. The NRTI backbone could also be modified at the switch. We selected 439 patients with no virological failure between their first undetectable pVL on cART and the switch date. Each exposed patient was matched with three unexposed patients who were randomly chosen with incident density sampling among the 3173 patients who had not switched to another cART regimen and
1870
who had the same duration (+6 months) of pVL undetectability as the exposed patients on their first-line PI-cART regimen.20 To approach the design of a randomized trial, unexposed patients were also matched with exposed patients on the basis of sex, age (+5 years), CD4 cell count (,350 and ≥350 cells/mm3) and VL (,4 log10, 4– 5 log10 and ≥5 log10 copies/mL) at the date of PI-cART initiation and the date of cART initiation (+6 months).21 The index date was the date of the switch for exposed patients and the matching date for unexposed patients, i.e. the date at which the length of virological rebound-free follow-up on first-line cART was the same as in the matched exposed patient prior to the switch. A total of 1261 unexposed patients were selected.
Statistical analysis Time to first subsequent virological failure was analysed by means of Kaplan– Meier life-table methods and Cox proportional hazard models. First, unexposed (PI-cART) and exposed (switch-cART) patients were compared, regardless of the drugs in the switch regimen. We then compared four groups of patients: (i) those who kept their first PI-cART; and those who switched to regimens containing (ii) efavirenz (efavirenz-cART), (iii) nevirapine (nevirapine-cART) or (iv) abacavir (abacavir-cART). Follow-up data were censored at the date of any subsequent change in cART. Thus, follow-up ended at the date of virological failure, the last follow-up visit, death or 31 December 2008, whichever occurred first. Potential confounding variables considered in the analyses were: the HIV transmission group; the NRTI backbone and PI in the first-line PI-cART regimen; and, at the index date, AIDS status, the CD4 cell count, the NRTI backbone, the calendar period, time since enrolment in the database, time since PI-cART initiation, and time between undetectability and the switch. Each variable associated with virological failure in the univariate model (P, 0.20) was included in a multivariable model designed to evaluate the impact of the switch, and then the precise switch regimen, on the risk of virological failure, after adjustment for other risk factors. Each model was stratified by the matched groups (exposed/matched unexposed patients). The SAS statistical package version 9.1 was used for all analyses (SAS Institute, Cary, NC, USA).
Results Matching ensured that the 439 switch patients and the 1261 non-switch patients were similar with regard to sex (68% male), age (both medians 37 years), the CD4 count stratum (82% ,350 cells/mm3), VL at cART initiation (both medians 5.07 log10 copies/mL) and the calendar period of cART initiation (1998 –2003 in 67% of cases; Table 1). However, the median absolute CD4 counts were slightly different, with 192 cells/mm3 in switch patients and 167 cells/mm3 in non-switch patients (P,0.01, Wilcoxon rank sum test). In this population of 1700 patients (439 switch patients plus 1261 non-switch patients), the ritonavir-boosted PI included in the first PI-cART regimen was lopinavir in 1048 cases (61.7%), indinavir in 391 (23.0%), saquinavir in 118 (6.9%), fosamprenavir in 71 (4.2%), atazanavir in 69 (4.1%) and darunavir in 3 (0.1%). The NRTI backbone was mainly zidovudine/lamivudine (1131 patients, 66.5%). At the index date, 196 patients (44.7% of the switchers) switched to efavirenz-cART, 123 (28.0%) to nevirapine-cART and 120 (27.3%) to abacavir-cART (Table 1). CD4 counts were significantly higher (P,0.01) in patients who switched to a PI-free cART regimen {median [interquartile range (IQR)] 413 cells/mm3 (IQR 285–578)} than in non-switchers [350 cells/mm3
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
independent cardiovascular risk,5 nor was cumulative exposure for .1 year or past exposure to abacavir.8 – 12 Clinical trials show that efavirenz and nevirapine, when combined with two nucleoside reverse transcriptase inhibitors (NRTIs), have similar virological efficacy in antiretroviral-naive patients.13,14 This is despite the fact that non-inferiority was not demonstrated in these studies when considering the primary composite endpoint (i.e. disease progression, virological failure and therapy change) efficacy outcome.13,14 Triple NRTI therapy including abacavir is less potent.15,16 Simplification of PI-containing regimens is virologically safe, at least in patients who do not harbour archived resistance mutations.17 However, randomized trials have limited power to detect differences in the risk of virological rebound between different switch strategies after a first-line PI-containing cART (PI-cART) regimen. Cohort studies suggest that switching to a nevirapine- or abacavircontaining regimen from a virologically successful first-line PI-cART regimen is associated with more virological failures than switching to an efavirenz-containing regimen,18,19 but none of these studies compared switch regimens with ongoing first-line PI-cART. Here, to compare the efficacy of PI-free switch regimens with that of continued first-line PI-cART in patients with no acquired resistance mutations, we compared virological outcomes among patients enrolled in the French Hospital Database on HIV (FHDH) who had sustained undetectable VL on a first PI-cART regimen and who either continued the same treatment or switched to a PI-free regimen. We also compared virological outcomes according to whether the switch regimen contained efavirenz, nevirapine or abacavir.
JAC
Switching from PI to EFV, NVP or ABC
Table 1. Characteristics of the patients according to whether or not they switched to a PI-free regimen after a first virologically effective first-line PI-containing regimen Characteristics
Switch-cART (n¼439)
PI-cART (n¼1261)
Male sex [n (%)]
296 (67.4)
861 (68.3)
Transmission group [n (%)] men who have sex with men intravenous drug users heterosexual transmission other
140 (31.9) 12 (2.7) 252 (57.4) 35 (8.0)
379 (30.1) 90 (7.1) 678 (53.8) 114 (9.0)
Age (years) [median (IQR)]
0.61 ,0.01
37 (31 –44)
37 (31 –44)
192 (85 –313) 5.07 (4.42– 5.55)
167 (60 –294) 5.07 (4.46–5.53)
310 (70.6) 63 (14.4) 37 (8.4) 29 (6.6)
821 (65.1) 225 (17.9) 106 (8.4) 109 (8.6)
267 (60.8) 116 (26.4) 28 (6.4) 28 (6.4)
781 (61.9) 275 (21.8) 90 (7.2) 115 (9.1)
300 (68.3) 139 (31.7)
838 (66.5) 423 (33.5)
413 (285 –578) 112 (25.5)
350 (211– 512) 350 (27.8)
272 (62.0) 128 (29.1) 11 (2.5) 28 (6.4)
821 (65.1) 225 (17.9) 106 (8.4) 109 (8.6)
0.62 ,0.01 0.90 0.11
0.07
0.42
At the index datea CD4 count (cells/mm3) [median (IQR)] prior AIDS diagnosis [n (%)] NRTI backbone [n (%)] zidovudine/lamivudine tenofovir/emtricitabine stavudine/lamivudine other boosted PI [n (%)] lopinavir indinavir saquinavir other efavirenz nevirapine abacavir
196 (44.7) 123 (28.0) 120 (27.3)
Calendar period 1998 –2003 2004 –2008 Time between database enrolment and switch (months) [median (IQR)] Time between first cART and switch (months) [median (IQR)] Time between undetectability and switch (months) [median (IQR)]
157 (35.8) 282 (64.2) 18 (11 –32) 13 (8– 20) 10 (5– 10)
,0.01 0.31 0.30
781 (61.9) 275 (21.8) 90 (7.2) 116 (9.1)
0.01 534 (42.3) 727 (57.7) 14 (6–29) 9 (3–16) 5 (0–13)
,0.01 ,0.01 ,0.01
cART, combination antiretroviral therapy; PI, protease inhibitor; IQR, interquartile range; pVL, plasma viral load; NRTI, nucleoside reverse transcriptase inhibitor. a Date of switch for patients switching to PI-free cART (efavirenz-, nevirapine- or abacavir-cART) and matching date for non-switch patients (i.e. the date corresponding to the same virological rebound-free follow-up on first-line cART as in the matched exposed patient before the switch).
1871
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
At first-line PI-cART CD4 count (cells/mm3) [median (IQR)] pVL (log10 copies/mL) [median (IQR)] NRTI backbone [n (%)] zidovudine/lamivudine tenofovir/emtricitabine stavudine/lamivudine other boosted PI [n (%)] lopinavir indinavir saquinavir other calendar period 1998 –2003 2004 –2008
P value
Proportion without virological rebound
Bommenel et al.
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
No switch Switch
P = 0.61 0
12
24
36
48
60
Time since the date of switch (months)
Proportion without virological rebound
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
1008 323
810 265
627 208
456 157
248 70
No switch Switch to EFV Switch to NVP Switch to ABC
P = 0.26 0
12
Patients at risk No switch 1261 Switch to EFV 196 Switch to NVP 123 Switch to ABC 120
1008 131 95 95
24 36 48 Time since the date of switch (months) 810 101 84 83
627 71 65 71
456 48 49 58
60
248 19 14 35
Figure 1. Kaplan– Meier curves for the probability of virological rebound-free survival after the index date, comparing patients who did not switch and patients who switched to PI-free cART, whatever the switch regimen (upper graph) and according to the switch regimen (lower graph). EFV, efavirenz; NVP, nevirapine; ABC, abacavir.
(IQR 211–512)]. Median (IQR) CD4 counts at the switch date were 386 cells/mm3 (259 –529), 431 cells/mm3 (293–594) and 438 cells/mm3 (293 –596) in patients who switched to efavirenz-, nevirapine- or abacavir-containing cART, respectively (P,0.01). Median follow-up after the index date was 35 months (IQR 14 –55) for the whole population, 34 months (IQR 11 –54) for the switchers and 36 months (IQR 15–56) for the non-switchers. A virological failure occurred in 153 patients (40 exposed and 113 unexposed) during a total follow-up of 5215.16 person-years. The Kaplan–Meier 12 month probability of virological rebound was 5.7% (95% CI, 3.8%–8.6%) for patients who switched to PI-free cART and 3.7% (95% CI, 2.8%–5.0%) for patients who did not switch. There was no difference in the rate of virological rebound between these two groups of patients at 60 months (P¼0.61, log-rank test). Among the 439 patients who switched
1872
to PI-free cART, the 12 month probability of virological rebound was 3.9% (95% CI, 1.8%–8.4%) in those who switched to efavirenz-cART, 7.2% (95% CI, 3.7%–13.9%) in those who switched to nevirapine-cART and 9.0% (95% CI, 4.9%–16.1%) in those who switched to abacavir-cART; the differences between these subgroups were not significant (P¼0.26; Figure 1). After adjustment for the PI included in the first cART regimen, as well as the CD4 count and AIDS status at the date of the switch, switching was not associated with a higher risk of virological failure [crude hazard ratio (HR), 1.20; 95% CI, 0.81 – 1.77; adjusted HR (aHR), 1.19; 95% CI, 0.80 – 1.76, compared with non-switch]. However, after adjustment, switching to abacavir-containing cART was associated with a roughly 2-fold increase in the risk of virological failure (aHR for abacavir-cART versus non-switch, 1.99; 95% CI, 1.05–3.79), while switching to efavirenz- or nevirapine-containing regimens
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
Patients at risk No switch 1261 switch 439
Switching from PI to EFV, NVP or ABC
was not associated with an increased risk of virological failure (aHR, 0.82; 95% CI, 0.41 –1.65; and aHR, 0.96; 95% CI, 0.44 – 2.07 for efavirenz and nevirapine versus no switch, respectively). Boosted lopinavir in the first cART regimen (aHR, 1.75; 95% CI, 1.03– 3.03), a higher CD4 count at the index date (aHR, 1.81; 95% CI, 1.09–3.01 for CD4 ≥350 cells/mm3 versus ,350 cells/ mm3) and AIDS diagnosis before the index date (aHR, 1.55; 95% CI, 0.94 –2.55) were associated with a higher risk of virological rebound (Table 2).
Discussion
showed a lower virological failure rate in patients switching to efavirenz from a regimen including indinavir, nelfinavir or ritonavir.26 However, our study included patients treated with newer, more potent and better-tolerated boosted PIs, and Kaplan –Meier curves showed that the risk of virological failure in patients switching to efavirenz was similar to that in non-switchers. Our data are in keeping with the results of a cohort study showing higher failure rates in patients switching to abacavir than in patients switching to efavirenz,19 but not with those of cohorts showing higher rates of virological failure with nevirapine than with efavirenz.18,19,23 The efficacy of the switch regimen in these latter studies could have been undermined by the acquisition of resistance mutations to the NRTI background, compromising subsequent virological outcome: most patients had a history of NRTI exposure before their first PI-containing regimen and very few patients received a combination including a boosted PI. Ours is the first cohort study in which only newer boosted PIs were used and in which the patients were antiretroviral naive before the first PI-cART. In this study, switching to nevirapine was as virologically safe as switching to efavirenz. Nevirapine may be associated with inferior virological outcomes compared with efavirenz in NNRTInaive patients with extensive resistance to other drug classes.27,28 In addition to being free of acquired resistance mutations before the PI-cART, our patients had not experienced virological failure during their first-line cART regimen (median 11 months), probably reflecting better initial adherence than in the overall database population. Differences in the intrinsic virological efficacy of nevirapine and efavirenz may be less pronounced in patients with fully susceptible viruses and good adherence, as observed in a clinical trial.13,29 In a previous study of switch regimens,19 we found that switching to abacavir rather than to efavirenz was associated with a significantly higher risk of virological failure only in antiretroviral-experienced patients. However, with an HR of 1.29 and a 95% CI of 0.88– 1.89, we could not rule out a difference in naive patients too. Here, the aHR for virological failure was 2.43 (95% CI, 0.93–6.34) for switch abacavir-cART compared with switch efavirenz-cART. A cohort study showed similar virological outcomes in antiretroviral-naive patients switching from PI-cART to efavirenz-cART or abacavir-cART, but only 38 patients switched to abacavir and only 3 of them experienced virological failure, yielding an aHR for virological failure of 1.56 (95% CI, 0.24 –10.09) compared with efavirenz, again ruling out firm conclusions.18 Higher CD4 counts at the switch date were associated with a higher risk of virological failure, for reasons that are unclear.30 CD4 counts at the switch date were different across the three switch groups: they were the highest in the abacavir-cART arm, probably owing to prescribers’ knowledge of the differences in the intrinsic virological potency of the drugs and to the need for a well-tolerated switch treatment in these successfully treated patients. Also, patients with higher CD4 counts might have better subjective health and, thus, adhere less strictly than patients with lower counts. Lopinavir was associated with a higher rate of virological failure than other PIs, mainly including indinavir, which is much less well tolerated. It is possible that patients switching from less-well-tolerated treatments, such as those containing boosted indinavir, had better subsequent
1873
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
This study shows that patients switching from a virologically effective first-line boosted PI-cART regimen to a cART regimen containing efavirenz or nevirapine have similar virological outcomes to patients who continue on the first-line regimen, whereas patients switching to a regimen containing abacavir have an increased risk of virological failure. In cohort studies, differences in virological outcome associated with different regimens do not necessarily reflect differences in efficacy, as bias due to residual confounding factors cannot be excluded despite adjustments for factors known to be associated with outcome.22,23 However, by matching patients who switched and patients who did not switch for factors known to influence virological outcome, such as age, sex, the CD4 count and VL at cART initiation, the duration of VL undetectability, and the date of first cART, we tried to mimic the design of a randomized study. We also took into account residual confounding factors that might have influenced the decision to prescribe a specific switch regimen, such as the drugs included in the initial cART regimen, and the CD4 count and AIDS status at the index date. In addition, we adopted an intention-to-treat approach, as used in randomized studies. Sensitivity analyses in which follow-up was censored at the date of any further change in the cART in the switch and non-switch populations gave similar results, as did analyses focusing separately on the first year after the index date or the subsequent follow-up period (data not shown). Analysis also gave similar results when the date of virological failure after the index date was defined as the date of the first of two consecutive pVL values ≥500 copies/mL, whatever the delay between the two detectable VL (data not shown). The FHDH, owing to its very large size and extensive follow-up, allows rare events such as virological rebound to be studied in selected patients treated with potent regimens and gives the opportunity to assess clinical practice in everyday life. Although clinical trials of treatment simplification in patients receiving PI-cART have given reassuring results,17 outcomes tend to be discordant when the third drug included in the regimen (efavirenz, nevirapine or abacavir) is taken into account.17,24,25 Abacavir-containing regimens are frequently compromised by cross-resistance between abacavir and other NRTIs. More than 50% of the patients included in such trials had previously received single- or dual-NRTI therapy, and ≥90% of them received an unboosted PI, facilitating the acquisition of archived resistance mutations.17,24 We found no difference in the virological failure rate between patients who switched to PI-free cART and patients who remained on their first-line PI-cART regimen, whereas another cohort study
JAC
Bommenel et al.
Table 2. Relative hazards of virological rebound at the index date in patients with undetectable plasma viral load while on a first-line PI-containing regimen, according to whether or not they switched to an efavirenz (EFV)-, nevirapine (NVP)- or abacavir (ABC)-containing regimen (1700 patients, 153 events); multivariable Cox proportional hazards analysis Univariate analysis
Switch status no switch switch to EFV-cART switch to NVP-cART switch to ABC-cART
At first-line cART initiation NRTI backbone zidovudine/lamivudine other associations boosted PI lopinavir other
Rebound (n)
HR
1261 196 123 120
113 11 11 18
1.00 0.87 1.07 1.81
519 930 251
37 101 15
1.00 1.12 0.71
1131 569
103 50
1.00 0.78
1048 652
95 58
1.00 0.66
1238 462
107 46
1.00 1.45
(95% CI)
P value
HR
(95% CI)
0.30
0.20 1.00 0.82 0.96 1.99
(0.44– 1.75) (0.47– 2.13) (0.97– 3.38)
P value
(0.41– 1.65) (0.44– 2.07) (1.05– 3.79)
0.38 (0.62– 2.03) (0.33– 1.52) 0.27 (0.51– 1.21) 0.12
0.04 1.00 0.57
(0.40– 1.11)
(0.33– 0.97)
a
At the index date prior AIDS diagnosis no yes CD4 cell count(/mm3) ,350 ≥350
0.12
58 95
1.00 1.66
1093 607
108 45
1.00 0.98
691 1009
97 56
1.00 0.96
Time since database enrolment 0 –8 months 9 –15 months 16– 28 months ≥29 months
485 351 407 457
65 35 27 26
1.00 0.75 0.69 0.54
Time between first cART and index date 0 –4 months 5 –10 months 11– 17 months ≥18 months
361 434 447 458
50 51 30 22
1.00 0.99 0.60 0.46
Time between VL undetectability and index date 0 –7 months 8 –15 months ≥16 months
824 459 417
101 36 16
1 0.85 0.36
Calendar period 1998 –2003 2004 –2008
(0.90– 2.35)
(0.94– 2.55)
0.04 792 908
NRTI backbone zidovudine/lamivudine other
0.08 1.00 1.55
0.02 1.00 1.81
(1.02– 2.70)
(1.09– 3.01)
0.92 [0.64– 1.49] 0.93 (0.43– 2.16) 0.30 (0.43– 1.29) (0.35– 1.34) (0.28– 1.04) 0.44 (0.58– 1,69) (0.27– 1.29) (0.13– 1.65) 0.39 (0.38– 1.87) (0.08– 1.64)
EFV, efavirenz; NVP, nevirapine; ABC, abacavir; cART, combination antiretroviral therapy; PI, protease inhibitor; IQR, interquartile range; VL, viral load; NRTI, nucleoside reverse transcriptase inhibitor. a Date of switch for patients switching to PI-free cART (efavirenz-, nevirapine- or abacavir-cART) and matching date for non-switch patients (i.e. the date corresponding to the same virological rebound-free follow-up on first-line cART as in the matched exposed patient before the switch).
1874
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
Transmission group men who have sex with men heterosexuals other
N
Multivariate analysis
Switching from PI to EFV, NVP or ABC
adherence and, thus, lower rates of virological failure than patients switching from a better-tolerated PI. In conclusion, patients with a virologically effective first-line cART regimen including a boosted PI can safely switch to an NNRTI-containing regimen with efavirenz or nevirapine, but not with abacavir.
Acknowledgements FHDH investigators
F. Bricaire (Corevih Ile de France Centre, Site Investigator); S. Herson (Corevih Ile de France Centre, Site Investigator); N. Desplanque (Corevih Ile de France Centre, Site Investigator); P. M. Girard (Corevih Ile de France Centre, Site Investigator); M. C. Meyohas (Corevih Ile de France Centre, Site Investigator); O. Picard (Corevih Ile de France Centre, Site Investigator); J. Cadranel (Corevih Ile de France Centre, Site Investigator); C. Mayaud (Corevih Ile de France Centre, Site Investigator); J. P. Clauvel (Corevih Ile de France Est, Site Investigator); J. M. Decazes (Corevih Ile de France Est, Site Investigator); L. Gerard (Corevih Ile de France Est, Site Investigator); J. M. Molina (Corevih Ile de France Est, Site Investigator); M. Diemer (Corevih Ile de France Est, Site Investigator); P. Sellier (Corevih Ile de France Est, Site Investigator); P. Honore´ (Corevih Ile de France Est, Site Investigator); V. Jeantils (Corevih Ile de France Est, Site Investigator); S. Tassi (Corevih Ile de France Est, Site Investigator); D. Mechali (Corevih Ile de France Est, Site Investigator); B. Taverne (Corevih Ile de France Est, Site Investigator); E. Bouvet (Corevih Ile de France Nord, Site Investigator); B. Crickx (Corevih Ile de France Nord, Site Investigator); J. L. Ecobichon (Corevih Ile de France Nord, Site Investigator); C. Picard-Dahan (Corevih Ile de France Nord, Site Investigator); P. Yeni (Corevih Ile de France Nord, Site Investigator); H. Berthe´ (Corevih Ile de France Ouest, Site Investigator); C. Dupont (Corevih Ile de France Ouest, Site Investigator); C. Chandemerle (Corevih Ile de France Ouest, Site Investigator); E. Mortier (Corevih Ile de France Ouest, Site Investigator); D. Tisne-Dessus (Corevih Ile de France Sud, Site Investigator); L. Weiss (Corevih Ile de France Sud, Site Investigator); D. Salmon (Corevih Ile de France Sud, Site Investigator); I. Auperin (Corevih Ile de France Sud, Site Investigator); L. Roudie`re (Corevih Ile de France Sud, Site Investigator); R. Fior (Corevih Ile de France Sud, Site Investigator); J. F. Delfraissy (Corevih Ile de France Sud, Site Investigator); C. Goujard (Corevih Ile de France Sud, Site Investigator); C. Jung (Corevih Ile de France Sud, Site Investigator); P. Lesprit (Corevih Ile de France Sud, Site Investigator); D. Vittecoq (Corevih Ile de France Sud, Site Investigator); P. Fraisse (Corevih Alsace, Site Investigator); D. Rey (Corevih Alsace, Site Investigator); G. Beck-Wirth (Corevih Alsace, Site Investigator); J. P. Stahl (Corevih de l’Arc Alpin, Site Investigator); P. Lecercq (Corevih de l’Arc Alpin, Site Investigator); F. Gourdon (Corevih Auvergne-Loire, Site Investigator); H. Laurichesse (Corevih Auvergne-Loire, Site Investigator); A. Fresard (Corevih Auvergne-Loire, Site Investigator); F. Lucht (Corevih Auvergne-Loire, Site Investigator); C. Bazin (Corevih Basse-Normandie, Site Investigator); R. Verdon (Corevih Basse-Normandie, Site Investigator); P. Chavanet (Corevih Bourgogne, Site Investigator); C. Arvieux (Corevih Bretagne, Site Investigator); C. Michelet (Corevih Bretagne, Site Investigator); P. Choutet (Corevih Centre, Site Investigator); A. Goudeau (Corevih Centre, Site Investigator); M. F. Maıˆtre (Corevih Centre, Site Investigator); B. Hoen (Corevih Franche-Comte´, Site Investigator); P. Elinger (Corevih Franche-Comte´, Site Investigator); J. P. Faller (Corevih Franche-Comte´, Site Investigator); F. Borsa-Lebas (Corevih Haute-Normandie, Site Investigator); F. Caron (Corevih Haute-Normandie, Site Investigator); J. P. Daure`s (Corevih Languedoc-Roussillon, Site Investigator); T. May (Corevih Lorraine, Site Investigator); C. Rabaud (Corevih Lorraine, Site Investigator); J. L. Berger (Corevih Lorraine, Site Investigator); G. Re´my (Corevih Lorraine, Site Investigator); E. Arlet-Suau (Corevih de Midi-Pyre´ne´es, Site Investigator); L. Cuzin (Corevih de Midi-Pyre´ne´es, Site Investigator); P. Massip (Corevih de Midi-Pyre´ne´es, Site Investigator); M. F. Thiercelin Legrand (Corevih de Midi-Pyre´ne´es, Site Investigator); G. Pontonnier (Corevih de Midi-Pyre´ne´es, Site Investigator); Y. Yasdanpanah (Corevih Nord-Pas de Calais, Site Investigator); P. Dellamonica (Corevih PACA Est, Site Investigator); P. Pugliese (Corevih PACA Est, Site Investigator); K. Aleksandrowicz (Corevih PACA Est, Site Investigator); D. Quinsat (Corevih PACA Est, Site Investigator); I. Ravaux (Corevih PACA Ouest, Site Investigator); J. P. Delmont (Corevih PACA Ouest, Site Investigator); J. Moreau (Corevih PACA Ouest, Site Investigator); J. A. Gastaut (Corevih PACA Ouest, Site Investigator); F. Retornaz (Corevih PACA Ouest, Site Investigator); J. Soubeyrand (Corevih PACA Ouest, Site
1875
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
S. Abgrall (U943 INSERM et UPMC, Statistical Analysis Center, Member of Scientific Committee); F. Barin (Corevih Centre, Member of Scientific Committee); M. Bentata (Corevih Ile de France Est, Member of Scientific Committee); E. Billaud (Corevih Pays de la Loire, Site Investigator, Member of Scientific Committee); F. Boue´ (Corevih Ile de France Sud, Site Investigator, Member of Scientific Committee); C. Burty (Corevih Lorraine, Member of Scientific Committee); A. Cabie´ (Corevih de Martinique, Site Investigator, Member of Scientific Committee); D. Costagliola (U943 INSERM et UPMC, Director of Statistical Analysis Center, Member of Scientific Committee); L. Cotte (Corevih de la Valle´e du Rhoˆne, Site Investigator, Member of Scientific Committee); P. de Truchis (Corevih Ile de France Ouest, Site Investigator, Member of Scientific Committee); X. Duval (Corevih Ile de France Nord, Member of Scientific Committee); C. Duvivier (Corevih Ile de France Sud, Member of Scientific Committee); P. Enel (Corevih PACA Ouest, Member of Scientific Committee); J. Gasnault (Corevih Ile de France Sud, Member of Scientific Committee); C. Gaud (Corevih de La Re´union, Site Investigator, Member of Scientific Committee); J. Gilquin (Corevih Ile de France Sud, Site Investigator, Member of Scientific Committee); S. Grabar (U943 INSERM et UPMC, Statistical Analysis Center, Member of Scientific Committee); C. Katlama (Corevih Ile de France Centre, Site Investigator, Member of Scientific Committee); M. A. Khuong (Corevih Ile de France Est, Member of Scientific Committee); J. M. Lang (Corevih Alsace, Site Investigator, Member of Scientific Committee); A. S. Lascaux (Corevih Ile de France Sud, Member of Scientific Committee); O. Launay (Corevih Ile de France Sud, Member of Scientific Committee); A. Mahamat (Corevih Guyane, Member of Scientific Committee); M. Mary-Krause (U943 INSERM et UPMC, Statistical Analysis Center, Member of Scientific Committee); S. Matheron (Corevih Ile de France Nord, Site Investigator, Member of Scientific Committee); J. L. Meynard (Corevih Ile de France Centre, Site Investigator, Member of Scientific Committee); J. Pavie (Corevih Ile de France Est, Member of Scientific Committee); G. Pialoux (Corevih Ile de France Centre, Site Investigator, Member of Scientific Committee); F. Pilorge´ (AIDES, Member of Scientific Committee); I. Poizot-Martin (Corevih PACA Ouest, Site Investigator, Member of Scientific Committee); C. Pradier (Corevih PACA Est, Site Investigator, Member of Scientific Committee); J. Reynes (Corevih Languedoc-Roussillon, Site Investigator, Member of Scientific Committee); E. Rouveix (Corevih Ile de France Ouest, Member of Scientific Committee); A. Simon (Corevih Ile de France Centre, Site Investigator, Member of Scientific Committee); P. Tattevin (Corevih Bretagne, Member of Scientific Committee); H. Tissot-Dupont (Corevih PACA Ouest, Site Investigator, Member of Scientific Committee); J. P. Viard (Corevih Ile de France Sud, Site Investigator, Member of Scientific Committee); N. Viget (Corevih Nord-Pas de Calais, Site Investigator, Member of Scientific Committee); M. Brosseau (French Ministry of Health, DMI2 Coordinating Center); V. Salomon (French Ministry of Health, DMI2 Coordinating Center); N. Jacquemet (Technical Hospitalization Information Agency, DMI2 Coordinating Center); M. Guiguet (U943 INSERM et UPMC, Statistical Analysis Center); E. Lanoy (U943 INSERM et UPMC, Statistical Analysis Center); L. Lie`vre (U943 INSERM et UPMC, Statistical Analysis Center); H. Selinger-Leneman (U943 INSERM et UPMC, Statistical Analysis Center); J. M. Lacombe (INSERM Transfert, Statistical Analysis Center); V. Potard (INSERM Transfert, Statistical Analysis Center);
JAC
Bommenel et al.
Funding FHDH is supported by Institut National de la Sante´ et de la Recherche Me´dicale (INSERM), Agence Nationale de la Recherche sur le SIDA (ANRS) and the French Ministry of Health.
Transparency declarations S. Abgrall has received support for conference attendance from pharmaceutical companies including VIHHealthcare, Abbott, Janssen-Cilag, Gilead and Boehringer Ingelheim. J. L. Meynard has received travel grants, consultancy fees, honoraria from various pharmaceutical companies including Bristol-Myers-Squibb, Abbott, Janssen-Cilag, Viiv Healthcare. D. Costagliola has received travel grants, consultancy fees, honoraria or study grants from various pharmaceutical companies including Bristol-Myers-Squibb, Abbott, Janssen-Cilag, Viiv Healthcare, Gilead Sciences, Boehringer Ingelheim, Merck-Sharp & Dohme-Chibret and Roche. C. Katlama has received travel grants, consultancy fees, honoraria or study grants from various pharmaceutical companies including Bristol-Myers-Squibb, Abbott, Janssen-Cilag, Viiv Healthcare, Gilead Sciences, Boehringer Ingelheim, Merck-Sharp & Dohme-Chibret. A. Simon has received travel grants, consultancy fees, honoraria or study grants from various pharmaceutical companies including Bristol-Myers-Squibb, Abbott, Janssen-Cilag, Viiv Healthcare, Gilead Sciences, Boehringer Ingelheim. E. Rouveix has received travel grants, consultancy fees, honoraria or study grants from various pharmaceutical companies including Bristol-Myers-Squibb, Abbott, Janssen-Cilag, Viiv Healthcare, Gilead Sciences, Boehringer Ingelheim. C. Pradier has received travel grants, consultancy fees, honoraria or study grants from various pharmaceutical companies including Abbott, Merck-Sharp & Dohme-Chibret. J. Gilquin has received travel grants, consultancy fees, honoraria or study grants from various pharmaceutical companies including Bristol-Myers-Squibb, Abbott, Janssen-Cilag, Viiv Healthcare, Gilead Sciences, Boehringer Ingelheim, Merck-Sharp & Dohme-Chibret and Roche. O. Launay has received travel grants from various pharmaceutical companies including Abbott, Viiv Healthcare, Gilead Sciences. V. Martinez has received consultancy fees or honoraria from pharmaceutical companies including Abbott, Boehringer-Ingelheim, Bristol-Myers-Squibb, Gilead Sciences, GlaxoSmithKline, Janssen-Cilag, Merck-Sharp and Dohme-Chibret and Viiv Healthcare. A. Mahamat, A. S. Lascaux and T. Bommenel have none to declare.
1876
References 1 Palella FJ Jr, Baker RK, Moorman AC et al. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr 2006; 43: 27–34. 2 Ortiz R, Dejesus E, Khanlou H et al. Efficacy and safety of once-daily darunavir/ritonavir versus lopinavir/ritonavir in treatment-naı¨ve HIV-1-infected patients at week 48. AIDS 2008; 22: 1389 –97. 3 Molina JM, Andrade-Villanueva J, Echevarria J et al. Once-daily atazanavir/ritonavir compared with twice-daily lopinavir/ritonavir, each in combination with tenofovir and emtricitabine, for management of antiretroviral-naı¨ve HIV-1-infected patients: 96-week efficacy and safety results of the CASTLE study. J Acquir Immune Defic Syndr 2010; 53: 323–32. 4 Carr A, Samaras K, Thorisdottir A et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia, and diabetes mellitus: a cohort study. Lancet 1999; 353: 2093 –9. 5 Mary-Krause M, Cotte L, Simon A et al. Increased risk of myocardial infarction with duration of protease inhibitor therapy in HIV-infected men. AIDS 2003; 17: 2479– 86. 6 Friis-Moller N, Reiss P, Sabin CA et al. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med 2007; 356: 1723 –35. 7 Penzak SR, Chuck SK. Hyperlipidemia associated with HIV protease inhibitor use: pathophysiology, prevalence, risk factors and treatment. Scand J Infect Dis 2000; 32: 111–23. 8 Lang S, Mary-Krause M, Cotte L et al. Impact of individual antiretroviral drugs on the risk of myocardial infarction in HIV-infected patients: a case– control study nested within the FHDH ANRS Cohort CO4. Arch Intern Med 2010; 170: 1228 –38. 9 Sabin CA, Worm SW, Weber R et al. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients enrolled in the D:A:D study: a multi-cohort collaboration. Lancet 2008; 371: 1417 –26. 10 Strategies for Management of Anti-Retroviral Therapy/INSIGHT; DAD Study Groups. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients. AIDS 2008; 22: F17 –24. 11 Brothers CH, Hernandez JE, Cutrell AG et al. Risk of myocardial infarction and abacavir therapy: no increased risk across 52 GlaxoSmithKline-sponsored clinical trials in adult subjects. J Acquir Immune Defic Syndr 2009; 51: 20–8. 12 Ribaudo HJ, Benson CA, Zheng Y et al. No risk of myocardial infarction associated with initial antiretroviral treatment containing abacavir: short and long-term results from ACTG A5001/ALLRT. Clin Infect Dis 2011; 52: 929–40. 13 Van Leth F, Phanuphak P, Ruxrungtham K et al. Comparison first-line antiretroviral therapy with regimens including nevirapine, efavirenz, or both drugs, plus stavudine and lamivudine: a randomised open-label trial, the 2NN Study. Lancet 2004; 363: 1253– 63. 14 Van Leth F, Andrews S, Grinsztejn B et al. The effect of baseline CD4 cell count and HIV-1 viral load on the efficacy and safety of nevirapine or efavirenz-based first-line HAART. AIDS 2005; 19: 463–71. 15 Gulick RM, Ribaudo HJ, Shikuma CM et al. Triple-nucleoside regimens versus efavirenz-containing regimens for the initial treatment of HIV infection. N Engl J Med 2004; 350: 1850– 61. 16 Sax PE, Tierney C, Collier AC et al. Abacavir– lamivudine versus tenofovir–emtricitabine for initial HIV-1 therapy. N Engl J Med 2009; 361: 2230– 40. 17 Martinez E, Arnaiz JA, Podzamczer D et al. Substitution of nevirapine, efavirenz, or abacavir for protease inhibitors in patients with
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
Investigator); A. Galinier (Corevih PACA Ouest, Site Investigator); J. M. Ruiz (Corevih PACA Ouest, Site Investigator); T. Allegre (Corevih PACA Ouest, Site Investigator); P. A. Blanc (Corevih PACA Ouest, Site Investigator); D. Bonnet-Montchardon (Corevih PACA Ouest, Site Investigator); G. Lepeu (Corevih PACA Ouest, Site Investigator); P. Granet-Brunello (Corevih PACA Ouest, Site Investigator); J. P. Esterni (Corevih PACA Ouest, Site Investigator); L. Pelissier (Corevih PACA Ouest, Site Investigator); R. Cohen-Valensi (Corevih PACA Ouest, Site Investigator); M. Nezri (Corevih PACA Ouest, Site Investigator); S. Chapadaud (Corevih PACA Ouest, Site Investigator); A. Laffeuillade (Corevih PACA Ouest, Site Investigator); F. Raffi (Corevih Pays de la Loire, Site Investigator); A. Boibieux (Corevih de la Valle´e du Rhoˆne, Site Investigator); D. Peyramond (Corevih de la Valle´e du Rhoˆne, Site Investigator); J. M. Livrozet (Corevih de la Valle´e du Rhoˆne, Site Investigator); J. L. Touraine (Corevih de la Valle´e du Rhoˆne, Site Investigator); C. Trepo (Corevih de la Valle´e du Rhoˆne, Site Investigator); M. Strobel (Corevih de Guadeloupe, Site Investigator); F. Bissuel (Corevih de Guadeloupe, Site Investigator); R. Pradinaud (Corevih de Guyane, Site Investigator); M. Sobesky (Corevih de Guyane, Site Investigator); and M. Contant (Corevih de La Re´union, Site Investigator).
Switching from PI to EFV, NVP or ABC
human immunodeficiency virus infection. N Engl J Med 2003; 349: 1036– 46. 18 Chiesa E, Bini T, Adorni F et al. Simplification of protease inhibitor-containing regimens with efavirenz, nevirapine or abacavir: safety and efficacy outcomes. Antivir Ther 2003; 8: 27 –35. 19 Abgrall S, Yeni PG, Bouchaud O et al. Switch from a first virologically effective protease inhibitor-containing efavirenz, nevirapine or abacavir. AIDS 2006; 20: 2099– 106. 20 Robins JM, Gail MH, Lubin JH. More on ‘biased selection of controls for case–control analyses of cohort studies’. Biometrics 1986; 42: 293–9. 21 Hernan MA, Alonso A, Logan R et al. Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease. Epidemiology 2008; 19: 766–79.
23 Phillips AN, Pradier C, Lazzarin A et al. Viral load outcome of non-nucleoside reverse transcriptase inhibitor regimens for 2203 mainly antiretroviral-experienced patients. AIDS 2001; 15: 2385– 95. 24 Opravil M, Hirschel B, Lazzarin A et al. A randomized trial of simplified maintenance therapy with abacavir, lamivudine, and zidovudine in human immunodeficiency virus infection. J Infect Dis 2002; 185: 1251– 60.
25 Bucher HC, Kofler A, Nuesch R et al. Meta-analysis of randomized controlled trials of simplified versus continued protease inhibitor-based antiretroviral therapy in HIV-1-infected patients. AIDS 2003; 17: 2451– 59. 26 Hirschel B, Flepp M, Bucher HC et al. Switching from protease inhibitors to efavirenz: differences in efficacy and tolerance among risk groups: a case– control study from the Swiss HIV Cohort. AIDS 2002; 16: 381– 5. 27 Abgrall S, Yeni PG, Bouchaud O et al. Comparative biological and clinical outcomes after a switch from a virologically unsuccessful first protease inhibitor-containing antiretroviral combination to a 3-drug regimen containing efavirenz, nevirapine or abacavir. CID 2007; 44: 120–7. 28 Bannister WP, Ruiz L, Cozzi-Lepri A et al. Comparison of genotypic resistance profiles and virological response between patients starting nevirapine and efavirenz in EuroSIDA. AIDS 2008; 22: 367–76. 29 Van den Berg-Wolf M, Hullsiek KH, Peng G et al. Virologic, immunologic, clinical, safety, and resistance outcomes from a long-term comparison of efavirenz-based versus nevirapine-based antiretroviral regimens as initial therapy in HIV-1-infected persons. HIV Clin Trials 2008; 9: 324– 36. 30 Wolbers M, Opravil M, von Wyl V et al. Predictors of optimal viral suppression in patients switched to abacavir, lamivudine, and zidovudine: the Swiss HIV Cohort Study. AIDS 2007; 21: 2201– 7.
1877
Downloaded from http://jac.oxfordjournals.org/ by guest on December 28, 2015
22 Pocock SJ, Elbourne DR. Randomized trials or observational tribulations? N Engl J Med 2000; 342: 1907–9.
JAC
