Resistance to Dolutegravir—A Chink in the Armor?

(See the Major Article by Wijting et al, on pages 688–97 and the Major Article by Pham et al, on pages 698–706.)

The advent of integrase strand- transfer inhibitors (INSTIs) has transformed antiretroviral therapy (ART) of human immunodeficiency virus type 1 (HIV-1) infection. The potency, safety, tolerability, and dosing convenience of these drugs—particularly when coformulated as single-tablet regimens—have led ART guidelines panels of the Department of Health and Human Services and the International Antiviral Society–USA to recommend INSTI-containing regimens as preferred first-line regimens for most patients with HIV-1 infection [1, 2]. On a global scale, the World Health Organization and the US President’s Emergency Plan for AIDS Relief are recommending regimens that include tenofovir (as the disoproxil fumarate or alafenamide formulation), together with lamivudine and dolutegravir, as a first-line regimen to replace efavirenz-based regimens, given the increasing prevalence of transmitted resistance to nonnucleoside reverse transcriptase inhibitors in resource-limited settings.

As with other classes of antiretroviral drugs, treatment failure of the INSTIs raltegravir and elvitegravir can result in the emergence of INSTI resistance [3–5]. Because raltegravir and elvitegravir share many resistance mutations, resistance to one drug frequently results in cross- resistance to the other. By contrast, resistance to dolutegravir and bictegravir has not been reported in clinical trials when these drugs are used as part of initial triple-drug ART [6–9] and has been observed only rarely in treatment-experienced patients receiving a dolutegravir-containing regimen [10]. Combinations of INSTI resistance mutations selected by prolonged exposure to raltegravir or elvitegravir in the setting of treatment failure can, however, result in cross-resistance to dolutegravir and bictegravir [11, 12].

The apparent high genetic barrier of dolutegravir to resistance has prompted comparisons to ritonavir- or cobicistat-boosted protease inhibitors and encouraged exploration of dolutegravir in 2-drug regimens, in combination with rilpivirine (now approved as a fixed-dose combination) [13] or lamivudine (now in phase 3 clinical trials) [14]. By contrast, exploratory trials of dolutegravir monotherapy have not gone so well. High rates of virologic failure have been reported by several groups, accompanied by the emergence of virus carrying a variety of INSTI resistance mutations.

A pair of articles in this issue of The Journal of Infectious Diseases provide an in-depth analysis of dolutegravir resistance arising in one such trial, the Dolutegravir as Maintenance Monotherapy for HIV (DOMONO) study [15]. That study randomly assigned participants with virologic suppression by a combination ART regimen to switch to dolutegravir monotherapy (the immediate switch group) or continue the combination regimen for 24 weeks followed by a switch to dolutegravir monotherapy (the delayed switch group). Although the study demonstrated noninferiority of dolutegravir monotherapy as compared to continued combination ART at 24 weeks (the primary end point), additional virologic failures accumulated during continued follow-up, resulting in termination of the study.

At the time of the initial report, 8 participants had experienced virologic failure (6 in the immediate switch group and 2 in the delayed switch group), and genotypic resistance data were available for 6 participants. Virus from 3 participants showed emergence of INSTI resistance mutations (R263K, N155H, and S230R, respectively); virus from the other 3 participants remained wild type. In the report by Wijting et al in this issue of the Journal [16], genotypic data are provided on virus from 2 additional participants with virologic failure, 1 of whom had virus with the E92Q and N155H substitutions (virus from the other participant was wild type). Thus, of 10 participants with virologic failure in the DOMONO study, INSTI resistance mutations were documented in 4, virus remained wild type in 4 (1 of whom had virologic failure at week 4, due presumably to nonadherence), and HIV sequencing was unsuccessful in 2. When considered along with other dolutegravir monotherapy switch studies (some of which included INSTI-experienced patients), the risks of virologic failure and selection of INSTI resistance are clearly unacceptable; this strategy should be abandoned.

The R263K substitution in integrase was first identified as a dolutegravir resistance mutation by in vitro selection experiments in the laboratory of the late Mark Wainberg [17] and subsequently found in virus from 4 participants with virologic failure in a clinical trial of dolutegravir in treatment-experienced patients [10]. By contrast, the E92Q and N155H substitutions are typically associated with resistance to raltegravir and elvitegravir. The article by Pham et al in this issue of The Journal reports for the first time the phenotypic effects of the S230R mutation [18]. Introduction of S230R by site-directed mutagenesis into a wild-type background resulted in substantial reductions in integrase enzyme efficiency and replication capacity and a nearly 4-fold reduction in dolutegravir susceptibility. A similar insusceptibility to cabotegravir was observed, along with more-modest shifts in susceptibility to raltegravir and elvitegravir. As noted by the authors, these effects are similar to the impact of the R263K mutation.

Wijting et al also report that one of the viruses in which no resistance mutations were observed in integrase was found to have 2 mutations (GGGGG→GGGAGC) in the highly conserved poly-purine tract of the 3’-long terminal repeat (referred to as the 3ʹ-polypurine tract [3ʹ-PPT]), a region important for binding of the intasome. A previous study by another group found that serial in vitro passage in the presence of dolutegravir selected for a dolutegravir-resistant variant that lacked resistance mutations in integrase but was mutated in the 3ʹ-PPT (GGGGGG→GGACTdel) [19]. No data are provided by Wijting et al on the phenotypic effects of the 3ʹ-PPT mutations they observed, which differ from those in the earlier report. It is therefore difficult to know the extent to which this new 3ʹ-PPT mutation contributed to failure of dolutegravir monotherapy.

An interesting observation from the DOMONO study is that most virologic failures occurred later in the trial (after 24 weeks). Wijting et al speculate that stochastic activation of a minority population of latently infected CD4⁺ T cells harboring infectious proviruses carrying preexisting INSTI resistance mutations accounted for these late failures. Confirmation of this intriguing hypothesis would require demonstration of these mutations in intact proviral sequences obtained at study entry (or prior to dolutegravir failure). Even if confirmed, it remains challenging to determine whether these mutations are a cause or consequence of dolutegravir failure. Dolutegravir trough concentrations would still exceed, by a substantial margin, the protein-adjusted 50% inhibitory concentrations of the mutants observed in this study, given the modest (≤4-fold) change in susceptibility associated with these mutations. The late virologic failure of 2 participants whose virus lacked mutations in integrase or the 3ʹ-PPT remains unexplained.

Despite these unanswered questions, one point is clear: the apparently high genetic barrier to resistance of dolutegravir may be breached when the drug is given as monotherapy. The findings by Wijting et al raise a cautionary warning against the headlong switch to dolutegravir-based regimens among patients currently receiving efavirenz-based ART in resource-limited settings, unless virus load testing can be performed at the time of switching, to confirm virologic suppression. Viremic patients in whom long-term exposure to a failing regimen has resulted in undetected emergence of tenofovir and lamivudine (and emtricitabine) resistance could end up receiving unintended dolutegravir monotherapy, with predictable adverse outcomes.

Notes

Financial support. This work was supported by the National Institute of Allergy and Infectious Diseases (grants UM1 AI068636 and UM1 AI106701).

Potential conflicts of interest. The author is a consultant to and has received honoraria and/or research support from Gilead, GlaxoSmithKline, Janssen, Merck, and ViiV. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References