Unplanned antiretroviral treatment interruptions, genetic barrier, and development of resistance


  • JM Llibre,

    Corresponding author
    1. ‘Lluita contra la SIDA’ Foundation and HIV Unit, Univ Hosp Germans Trias i Pujol, Badalona, Spain
    2. Universitat Autònoma de Barcelona, Barcelona, Spain
    • Correspondence: Dr Josep M. Llibre, HIV Unit, Univ Hosp Germans Trias I Pujol, Ctra de Canyet s/n, 08916 Badalona, Spain. Tel: 934978887; fax: +34934657602; e-mail: jmllibre@flsida.org

    Search for more papers by this author
  • B Young

    1. International Association of Providers of AIDS Care, Washington, DC, WA, USA
    2. Josef Korbel School of International Studies, University of Denver, Denver, CO, USA
    Search for more papers by this author

Clinicians facing patients with antiretroviral therapy (ART) failure must undertake a thorough review of the full treatment history, and they must suspect the presence of mutations, particularly against drugs with a low genetic barrier to resistance (lamivudine, emtricitabine, nevirapine, efavirenz, rilpivirine, enfuvirtide and raltegravir) included in failing regimens despite the fact that mutations are not detected in genotypes [1]. The genetic barrier to resistance of an antiretroviral is a complex phenomenon defined as the number of mutations required to overcome drug-selective pressure, but also as the number and significance of mutations selected at failure [2].

Unplanned treatment interruptions (UTrIs) pose a difficult challenge to combination antiretroviral regimens, particularly those comprised of individual medications with heterogeneous or discordant terminal pharmacokinetics. When virus found during rebound plasma viraemia after drug termination is confronted with low levels of drugs with long plasma half-lives – well below the EC50- the situation is analogous to that of experiments performed in the laboratory to select resistant HIV-1 clones [3]. Therefore, not surprisingly, the rates of selection of M184V/I – the first mutation analysed in this scenario – are high in these episodes [4]. Many of the studies documenting resistance selection during treatment interruptions (TIs) were carried out in the ‘autovaccination’ era, and unsuccessfully tried to demonstrate the immunological benefit of those TIs [5].

In addition to nucleoside reverse transcriptase inhibitors (NRTIs), the phenomenon of resistance selection during TIs has been demonstrated to date with the first-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs) efavirenz and nevirapine [6, 7]. At least one NNRTI resistance mutation was detected in the virus of 16.4% of patients treated with an efavirenz-containing regimen with simultaneous interruption, 12.5% of patients with staggered interruption and 4.2% of patients with switched interruption, reinforcing the importance of homogeneous terminal half-lives in any combination. Therefore, because of this high rate of emergent resistance, it has been recommended that patients having TIs during a treatment using these NNRTIs should not be restarted with the same combination.

Tinago et al. have analysed the emergence of drug resistance in a cohort of patients treated with atavanavir/ritonavir (ATV/r) [8]. This is the first time that the emergence of drug resistance has been analysed for ritonavir-boosted protease inhibitors (PIs). They demonstrated that UTrIs are common in real-world practice with a broad representation in their cohort of women and injecting drug users on methadone maintenance therapy. Reassuringly, their results confirm that, in contrast to NNRTIs, ATV/r retains its activity and efficacy when reintroduced, even after several UTrIs. No patient developed primary protease mutations and only one out of 39 subjects developed enough minor mutations to confer intermediate genotypic resistance to ATV/r. Moreover, only four new NRTI mutations emerged in 88 subjects (two M184V, one K70R and one T69ins).

Boosted PIs are the paradigm of a high genetic barrier to resistance [9]. There is overwhelming evidence that PI resistance only rarely emerges among individuals experiencing treatment failure of first-line boosted PI regimens. The largest body of relevant evidence exists for ATV/r, with compiled data from the AI424-089, ArTen, Castle, ACTG A5202, ARIES, GS 216-0114 and GS 236-0103 studies for ART-naïve patients, including 2359 individuals randomly assigned to ATV/r with either abacavir/lamivudine or tenofovir/emtricitabine (plus 344 additional individuals treated with ATV boosted with cobicistat) [10-16]. In addition, brilliant analyses in two trials have clearly demonstrated that pre-therapy low-abundance HIV-1 resistance protease mutation-containing variants have no impact on initial ATV/r response or in subsequent simplification to unboosted ATV, both in B and in non-B HIV subtypes [17, 18]. Consequently, there is no evidence that pre-existing low-abundance protease mutations could emerge during those UTrIs. These studies are consistent with the analysis performed by Tinago and colleagues. Although the number of subjects in their series is limited and data must still be viewed with caution, taken together the findings of these studies support the idea that no major protease mutations emerge in individuals with UTrI on an initial triple regimen based on ATV/r. It is likely that this could be extended to other boosted PIs (mainly darunavir), but this remains to be shown.

Genotype-resistance testing of treatment-naïve chronically HIV-infected patients is cost-effective, compared with other HIV care, and recommended in contemporary treatment guidelines [19]. While it may be difficult to prove this cost-effectiveness in first-line ATV/r recipients who experienced UTrIs, Tinago and colleagues are correct in questioning the benefit of genotypic testing prior to antiretroviral reinitiation in this scenario.


Conflicts of interest

JML has received funding for research or payment for conferences or participation on advisory boards from Abbott, Boehringer Ingelheim, Bristol-Myers Squibb, Gilead Sciences, Janssen-Cilag, Merck Sharp & Dohme, and ViiV Healthcare. BY has received consulting or speaking fees from Bristol-Myers Squibb, Cerner Corporation, Gilead Sciences, GlaxoSmithKline, Merck & Co., Monogram Biosciences, and ViiV Healthcare, and research funding from Bristol-Myers Squibb Company, Cerner Corporation, Gilead Sciences, GlaxoSmithKline, Merck & Co, and ViiV Healthcare.