iconstar paper   HIV Articles  
Back grey arrow rt.gif
Has the time come to abandon efavirenz for first-line antiretroviral therapy?
  Download the PDF here
J Antimicrob Chemother
Advance Access publication 5 March 2014
Francois Raffi1*, Anton L. Pozniak2 and Mark A. Wainberg3 1Division of Infectious Diseases, Nantes University Hospital, Nantes, France; 2HIV Department, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK; 3Departments of Medicine and of Microbiology, Jewish General Hospital, McGill University, Montreal, Canada
Efavirenz has been recommended as a preferred third agent together with two nucleos(t)ides for first-line combination antiretroviral therapy (ART) for >15 years. The availability of efavirenz in a fixed-dose combination makes it very attractive. However, because of (i) adverse events associated with efavirenz, (ii) a poorer overall efficacy of efavirenz compared with newer antiretrovirals, (iii) the ranking of efavirenz as FDA Pregnancy Category D and (iv) the relatively high prevalence of transmitted drug-resistance mutations, there is a need to reconsider the role of efavirenz in first-line ART. We review the available evidence that challenges efavirenz's current position in first-line HIV treatment guidelines. Apart from its animal teratogenic potential, and moderate neuropsychiatric adverse events associated with its use, efavirenz has recently been associated with an increased risk of suicidality when compared with other antiretroviral drugs. Most importantly, efavirenz has demonstrated overall inferior efficacy to various comparator drugs, which include rilpivirine, raltegravir and dolutegravir, in antiretroviral-naive patients.
Furthermore, epidemiological data indicate that the prevalence of non-nucleoside reverse transcriptase inhibitor resistance has reached 5%-8% in various parts of the world, and minority transmitted non-nucleoside reverse transcriptase inhibitor resistance-associated mutations can have a negative impact on the outcome of first-line efavirenz-based ART. Based on considerations of efficacy, toxicity and resistance, it is time to reconsider the routine use of efavirenz in ART. This, of course, presupposes that other antiretrovirals will be available in place of efavirenz, and may not be applicable in certain developing country settings where this is not the case.
One of the major goals of antiretroviral therapy (ART) is the use of effective well-tolerated regimens that require little long-term monitoring. We need to continually re-evaluate what we consider to be a preferred regimen in the light of new data. Efavirenz, a non-nucleoside reverse transcriptase inhibitor (NNRTI) of HIV-1, has been recommended as a preferred third agent together with two nucleos(t)ides for first-line combination ART for >15 years. This choice has been based on the virological and pharmacological properties of efavirenz, such as its high in vitro potency, forgiveness in regard to missed doses, simplicity of dosing and consistent data from multiple randomized clinical studies that have demonstrated that efavirenz-containing regimens were 'unbeaten' in terms of rates of virological suppression.1 Indeed, large randomized controlled trials and cohort studies of treatment-naive patients have demonstrated potent viral suppression in efavirenz-treated patients, with a substantial proportion having HIV-1 RNA levels <50 copies/mL with up to 6 years of follow-up.2,3 In head-to-head randomized comparative studies, regimens containing efavirenz plus two nucleoside reverse transcriptase inhibitors were superior virologically to several protease inhibitor-based regimens, including indinavir,4 lopinavir/ritonavir5 and nelfinavir,6 and to triple nucleoside reverse transcriptase inhibitor-based regimens.7 In addition, efavirenz has been shown to possess comparable anti-HIV activity to nevirapine-,8 rilpivirine-9,10 and atazanavir-based regimens.11Until the spring of 2012, treatment guidelines routinely emphasized that 'To date, no regimen has proven superior to efavirenz-based regimens with respect to virologic responses.'1 In all recent guideline recommendations issued by national and international agencies as of late 2013, efavirenz-based regimens remain among the preferred options for the treatment of ART-naive patients12-15 and, in the latest WHO guidelines, as the preferred option for the initiation of ART.16 A major advantage of efavirenz is its pharmacokinetic profile, which allows it to be successfully incorporated into a once-daily, one-pill regimen with tenofovir and emtricitabine. Indeed, the availability of a once-daily, single-tablet regimen represents an important milestone with regard to convenience and potential for improved patient adherence, a factor that is the most important in ultimately determining the success or failure of any ART regimen. Three different once-daily single-tablet regimens are now available and others will soon follow.
The need for a reappraisal of efavirenz
Despite its virological potency and efficacy, it has long been recognized that efavirenz has important limitations. One of these is potential teratogenicity, while the second, more serious, limitation involves CNS and psychiatric side effects, which may commonly resolve within several weeks. First, efavirenz was found to cause major CNS congenital abnormalities in a number of non-human primates, an unfortunate occurrence of possibly limited significance, since most other antiretroviral drugs have not been tested in monkeys.1 Despite the fact that new meta-analyses of the use of efavirenz during pregnancy are reassuring,17 the data cannot exclude a potential 2-fold to 3-fold increase in the rates of neural tube birth defects as a result of first-trimester exposure to efavirenz. Indeed, several cases of neural tube defects in the newborns of mothers exposed to efavirenz during the first trimester of pregnancy have been reported.18,19 A recent 16 year retrospective analysis of the French perinatal cohort showed that first-trimester exposure of pregnant women to efavirenz was associated with an adjusted OR of CNS birth defects of 3.2 (95% CI: 1.1-9.1; P = 0.03).20 However the overall risk-benefit profile has led the British HIV Association 2012 guidelines committee to recommend that efavirenz, if initiated in women of childbearing potential, be continued in the event that conception occurs while on the drug,21 and the WHO 2013 guidelines give the same recommendation.16 Because efavirenz is classified as FDA Pregnancy Category D, meaning the evidence is based on studies in primates and observations in humans, the US Department of Health and Human Services guidelines advocate that alternative treatment options that do not include efavirenz should be strongly considered in sexually active women who are not taking effective contraception.15 As the risk of neural tube defects is restricted to the first 6 weeks of pregnancy, i.e. before pregnancy is usually recognized, there is still a debate on the use of efavirenz in women of childbearing potential. The most important clinical issue with efavirenz is that it frequently causes CNS side effects, including dizziness, abnormal dreams, headaches and depression. Although these adverse events can sometimes lead to treatment discontinuation, they are often self-limited and resolve spontaneously after a few days to weeks. Even in people who have tolerated efavirenz for years, changing to other drugs results in improvements in mood.22 A recent meta-analysis of four large randomized clinical studies in ART-naive patients, of which one was double-blinded, suggested that the adjusted risk of suicidality (tendency to commit suicide) was significantly higher in efavirenz than non-efavirenz recipients, with a hazard ratio of 2.28.23 These data, if confirmed, raise the question of the risk-benefit profile of efavirenz, taking also into account the fact that neurocognitive function improves less in people taking efavirenz.24
Phase 3 studies
In addition to the foregoing, three independent randomized clinical studies have shown that efavirenz is not non-inferior to comparator agents that have been used in first-line therapy (Table 1).25-27 The STaR study was a multicentre, international, randomized, open-label phase 3b 96 week study that evaluated two single-tablet regimens, tenofovir/emtricitabine/efavirenz and tenofovir/emtricitabine/rilpivirine, in first-line ART in 786 HIV-1 antiretroviral-naive adults. Randomization involved stratification on the basis of baseline HIV-1 RNA (≤ or >100 000 copies/mL). The proportions of patients achieving virological suppression by time to loss of virological response (in a missing-equals-failure analysis) were 80% and 85% at week 48 and 73% and 79% at week 96 for the efavirenz-based regimen and rilpivirine-based regimen, respectively.27,28 The 95% CIs for the differences were 0.6%-11.2% at week 48 (P = 0.03) and 0.7%-12.6% at week 96 (P = 0.03) and indicated the superiority of tenofovir/emtricitabine/rilpivirine compared with tenofovir/emtricitabine/efavirenz. In the subgroup of patients with baseline HIV-1 RNA ≤100 000 copies/mL (65% of the total study population), there was a statistically significant difference in virological suppression at week 96 favouring tenofovir/emtricitabine/rilpivirine (79% versus 71%, 95% CI for difference: 0.2%-15.1%; P = 0.048 by snapshot analysis).27 In the subgroup of patients with a baseline HIV-1 RNA ≥100 000 copies/mL, tenofovir/emtricitabine/rilpivirine was non-inferior to tenofovir/emtricitabine/efavirenz (76% versus 75%, 95% CI for difference: -8.7% to 11.6%; P = 0.78 by snapshot analysis).27 Tenofovir/emtricitabine/rilpivirine was also better tolerated than tenofovir/emtricitabine/efavirenz with significantly fewer nervous system and psychiatric adverse events (P < 0.001) and significantly fewer discontinuations due to adverse events (3% versus 11%, respectively; Table 2).
The STARTMRK trial was a randomized double-blind study comparing the use of tenofovir plus emtricitabine plus either raltegravir or efavirenz in 566 treatment-naive patients. Raltegravir was dosed twice daily, while patients on the efavirenz arm of the study received active efavirenz as part of their night-time regimen and a placebo in the morning.26 Raltegravir was shown to be superior to efavirenz at week 192 of treatment, despite the fact that efavirenz possesses a far longer plasma half-life than raltegravir. Thus, it can be argued that any degree of non-adherence in the STARTMRK trial should have favoured the efavirenz arm and the demonstrated superiority of raltegravir over efavirenz could be seen as more impressive than the data of the intent-to-treat analysis would seem to indicate [proportions with HIV RNA <50 copies/mL at week 240 were 71% and 61% for raltegravir and efavirenz, respectively (95% CI: 1.7%-17.3%)].26 Significantly fewer drug-related clinical adverse events occurred in patients on raltegravir (P < 0.0001),30-33 and there was a benefit of raltegravir over efavirenz in regard to lipid profile.26
The SINGLE study was a multicentre, international, randomized, double-blind phase 3 96 week study that evaluated two once-daily regimens, i.e. a tenofovir/emtricitabine/efavirenz single-tablet regimen versus abacavir/lamivudine/dolutegravir in 833 HIV-1 ART-naive adults. At weeks 48 and 96, the proportion of patients having HIV-1 RNA <50 copies/mL was significantly higher in the abacavir/lamivudine/dolutegravir group than in the tenofovir/emtricitabine/efavirenz group (88% versus 81%; P = 0.003). Rates of discontinuation for adverse events were also lower in the abacavir/lamivudine/dolutegravir group than in the tenofovir/emtricitabine/efavirenz arm (2% versus 10%).25
Epidemiological data on resistance
A major disadvantage of currently available NNRTIs is the prevalence of NNRTI resistance mutations in antiretroviral-naive patients and the low genetic barrier of NNRTIs for the development of drug resistance.34 Indeed, treatment failure of first-line NNRTI-based regimens can have serious consequences, including a further accumulation of NNRTI and nucleoside reverse transcriptase inhibitor resistance mutations, which can result in cross-resistance to second-generation NNRTIs (e.g. etravirine and rilpivirine) and diminished effectiveness of the nucleoside 'backbone' of subsequent treatment regimens, respectively. Other potential consequences are suboptimal immunological recovery and increased morbidity and mortality associated with lack of virological control, especially in patients presenting with advanced HIV infection. In randomized first-line clinical studies conducted in 2001-2004 in the USA, in which efavirenz-containing regimens were compared with a non-efavirenz-containing regimen, the prevalence of baseline NNRTI resistance was 5% and the risk of virological failure for subjects with baseline NNRTI resistance was higher than for subjects without such resistance (hazard ratio = 2.27, 95% CI: 1.15-4.49; P = 0.018).35 Detection of low-level NNRTI resistance by either bulk sequencing or more ultrasensitive analyses is significantly associated with the likelihood of virological failure involving first-line efavirenz- or nevirapine-based therapy.36 In the EuroCoord-CHAIN joint project, the risk of virological failure was doubled in patients with transmitted NNRTI drug resistance.37 More recently, a meta-analysis of 985 participants from 10 studies showed that the presence of a minority HIV-1 NNRTI drug resistance mutation before starting ART was associated with a 2-fold to 3-fold higher risk of virological failure in treatment-naive patients initiating an NNRTI-based regimen.38 In a European case-control study of 260 patients, minority NNRTI resistance mutations at baseline were associated with an adjusted 2.4-fold (95% CI: 1.12-5.18; P = 0.024) increased risk of virological rebound of HIV-1 RNA to >200 copies/mL in patients on first-line NNRTI-based regimens.39This concern in regard to an increased prevalence of transmitted drug resistance to NNRTIs has led to the recommendation in high-income countries that HIV genotypic resistance testing be performed before initiating therapy with an NNRTI, a practice which may not always be possible in low-income countries where resources for resistance testing are limited. Although the prevalence of detectable NNRTI resistance mutations in ART-naive patients is stable or declining in most regions of the world (i.e. 5%-8%)40,41 these numbers do not take into consideration the presence of minority NNRTI resistance mutations.39,42 These considerations all suggest that the use of either nevirapine or efavirenz in first-line therapy may no longer be a preferred therapeutic option, either in high- or low-income countries. Of course, however, a decision no longer to recommend efavirenz in resource-limited settings will only be ethical if safer and more effective antiretrovirals are available for use as third agents in combination ART.
In conclusion, efavirenz today remains one of the commonly used components of prescribed antiretroviral regimens. We acknowledge that cost, availability and acceptability are issues that drive much antiretroviral use, but we should continually re-evaluate what we recommend and challenge current paradigms. Treatment advances have now resulted in safer and even superior alternatives to efavirenz. Considerations of efficacy, toxicity and resistance suggest that efavirenz should be reconsidered for use in first-line therapy, as has already happened for multiple other drugs, e.g. stavudine, zalcitabine and nelfinavir.
This should not only happen in high-income countries but ideally also in low-income settings, if alternative drugs are available, and this recommendation should be reflected in the treatment guidelines of the WHO and both governmental and non-governmental organizations, especially since the monitoring of drug toxicities and transmitted resistance is often not practicable in resource-limited areas. In order for the success of ART to become a reality in all countries, we need always to use the safest and most effective antiretroviral combinations. Assuming that alternative therapies are available, this means that the risk-benefit profile of efavirenz as a first-line agent should now be reconsidered by guidelines committees.

  iconpaperstack View Older Articles   Back to Top   www.natap.org