| |
ACTG 5142, Efavirenz v Kaletra v Kaletra+Efavirenz Class Sparing Study. Initial Treatment for HIV Infection - An Embarrassment of Riches EDITORIAL
|
| |
| |
NEJM
May 15, 2008
Bernard Hirschel, M.D., and Alexandra Calmy, M.D.
From the Department of Infectious Diseases, Geneva University Hospital, Geneva.
Drugs that are used to treat patients with human immunodeficiency virus (HIV) infection are classified according to their target. The first ones to be developed were nucleoside reverse-transcriptase inhibitors (NRTIs), which lead to premature termination of the nascent DNA chain, and nonnucleoside reverse-transcriptase inhibitors (NNRTIs), which bind and inhibit reverse transcriptase. The viral protease inhibitors were next. NRTIs, NNRTIs, and protease inhibitors remain the staples of highly active antiretroviral therapy, but other targets, such as the CCR5 receptor, the fusion peptide, and viral integrase, have recently yielded promising molecules.
At this time, eradication of HIV is impossible. Rebound inevitably follows cessation of therapy, and therapy must therefore be lifelong. With more than 20 drugs to choose from, there is an embarrassment of riches. Possible combinations are almost endless, as are the possibilities of side effects, either beneficial or damaging drug interactions, and the development of viral resistance.
Early in the antiretroviral-therapy era, the combination of indinavir (a protease inhibitor) and zidovudine and lamivudine (both NRTIs) predominated as the reference treatment. In 1999, the NNRTI efavirenz, in combination with zidovudine and lamivudine, proved to be more effective in diminishing the plasma concentration of HIV type 1 (HIV-1) RNA (the "viral load") than the reference treatment.1 Indinavir has since been largely replaced by atazanavir or lopinavir combined with a small dose of ritonavir to boost absorption and plasma levels.
Current guidelines recommend initiating antiretroviral therapy with two NRTIs in combination with either an NNRTI or a protease inhibitor.2 So the first question is, Which NRTIs and which protease inhibitor do we choose? And the second question is, Which is better, an NNRTI or a protease inhibitor? Phase 4 studies that compare treatment strategies are desirable, but they are difficult to do. In a rapidly moving field such as HIV therapy, what is the "reference treatment"? Trials have to be large and continue for a long time, and patients may vote with their feet and refuse to continue with a therapy that they judge, rightly or wrongly, to be inferior to the latest miracle drug. And large trials that continue for a long time are expensive. Drug companies have little to gain, and much to lose, from comparing one of their already marketed drugs with another that may be better. The National Institutes of Health, through the Clinical Trials Network, have very properly undertaken trials such as the Strategies for Management of Antiretroviral Therapy (SMART; ClinicalTrials.gov number, NCT00027352 [ClinicalTrials.gov] ),3 which showed that intermittent treatment was inferior to continuous treatment for patients with HIV infection.
In this issue of the Journal, Riddler et al.4 report on the AIDS Clinical Trials Group Study A5142, which compared three drug combinations in the initial therapy of 753 patients with HIV infection: efavirenz plus two NRTIs (efavirenz group), lopinavir-ritonavir plus two NRTIs (lopinavir-ritonavir group), and lopinavir-ritonavir plus efavirenz (NRTI-sparing group). As previously noted, the first two regimens were popular and widely prescribed. The third is theoretically attractive, since it avoids the use of NRTIs, which are suspected of contributing to side effects. An uncontrolled study of 86 patients showed that this combination would be effective, although it was not well tolerated: after 48 weeks, 24% of patients either discontinued the study regimen because of adverse events or were lost to follow-up.5 A study by Boyd et al. looked at efavirenz with ritonavir-boosted indinavir as an NRTI-sparing option, with similar conclusions.6
The results of the study by Riddler et al. are difficult to put in a nutshell. We want regimens that win in all categories: suppression of HIV-1 RNA, an increase in the CD4 cell count, a lack of emergence of resistance, low toxicity, and simplicity. However, the study by Riddler et al. yields a split decision. When the regimens were ranked according to suppression of HIV-1 RNA, the efavirenz group had the best results, closely followed by the NRTI-sparing group and the lopinavir-ritonavir group, although the difference between the efavirenz group and the NRTI-sparing group was not significant. When the regimens were ranked according to the emergence of drug resistance, the winner was the lopinavir-ritonavir group, followed by the efavirenz group and the NRTI-sparing group, and again the difference between the lopinavir-ritonavir group and the efavirenz group was not significant. Finally, as measured by the proportion of patients who discontinued or changed their treatment, all three groups had similar rates of adverse events.
Patients who participate in clinical trials differ from the majority who do not participate - one reason why clinical practice often cannot reproduce published results. Efavirenz causes side effects involving the central nervous system, including sleep disturbances with vivid dreams, dizziness, and daytime drowsiness.7 Such symptoms are frequent and troublesome early on; they largely disappear after a few weeks of therapy. Nonetheless, in all studies we are aware of, a sizable percentage of patients discontinued efavirenz because of these effects; the proportion was particularly high among patients who acquired HIV through illicit drug use, partly because efavirenz interferes with methadone. We are struck by the fact that Riddler et al. did not record much of this type of discontinuation in their study. This suggests that their patients were greatly motivated to continue their prescribed regimen, perhaps through their repeated and close contact with the investigators - a type of Hawthorne effect8 that is difficult to duplicate in routine practice.
Another problem with the study relates to the NRTIs that were administered in the efavirenz group and the lopinavir-ritonavir group. All patients received lamivudine, but the second NRTI was zidovudine (which was assigned to 42% of patients), extended-release stavudine (24%), or tenofovir (34%). NRTIs differ in both side effects9 and efficacy.10 Since the study started, the formulation in the lopinavir-ritonavir capsule has been replaced by tablets that produce a more consistent plasma drug level11 and are perhaps associated with less diarrhea and nausea. Extended-release stavudine has never been marketed because of pancreatic toxicity.12 Tenofovir and emtricitabine (a drug that was not used in the study) have become the reference NRTI combination. In summary, these reservations cast doubt on the future applicability of the study's findings - doubts that will not be easily resolved by further studies.
Nonetheless, on the basis of this study, it seems that efavirenz plus two NRTIs is hard to beat. In addition to the stated results, one has to consider the low pill burden, since brand-name formulations contain efavirenz, emtricitabine, and tenofovir for a one-pill daily regimen, and the fact that in most countries, efavirenz costs less than lopinavir-ritonavir. These data should challenge the 40% of clinicians who start antiretroviral treatment with a protease inhibitor and should reassure those who, in resource-limited settings, must use combinations of NRTIs and NNRTIs because they are cheaper.
Will new drugs dethrone efavirenz? Etravirine (an NNRTI),13 raltegravir (an integrase inhibitor),14 and maraviroc (a CCR5 inhibitor)15 are targeted to patients with drug-resistant virus. But because of their excellent pharmacokinetics and initially favorable side-effect profiles, these drugs have a potential for earlier use16 and in a few years may even be successfully combined.
Dr. Hirschel reports receiving consulting and lecture fees from Merck, serving on advisory boards for Merck and Tibotec, and receiving travel grants from Bristol-Myers Squibb, GlaxoSmithKline, and Roche. No other potential conflict of interest relevant to this article was reported.
|
|
| |
| |
|
|
|
