Dolutegravir superior to standard dose efavirenz in WHO analysis

Polly Clayden, HIV i-Base

A systematic review and meta-analysis, conducted to inform the new World Health Organization (WHO) Consolidated Guidelines, found dolutegravir superior to standard dose efavirenz for both viral suppression and discontinuation rates. [1]

The analysis, published in online in the Lancet HIV, 6 September 2016, also showed low dose efavirenz to be superior to standard dose for discontinuation rates and CD4 count gains.

The investigators wrote: “A research question posed by WHO in anticipation of the guideline development was how INSTIs compared with efavirenz, and to this end our results suggest a clear hierarchy within the INSTI class with dolutegravir being the most efficacious, followed by raltegravir then elvitegravir.” They suggest that although there are several reasons beyond safety and efficacy for WHO to continue to recommend standard dose efavirenz as the preferred first-line drug in the recent guidelines, these results signal the potential for future changes.

For the systematic review and network meta-analysis, the investigators searched MEDLINE, Embase, and the Cochrane register of Controlled trials for randomised clinical trials of antiretroviral regimens in treatment-naive adults and adolescents (aged 12 years and above) with HIV, published up to 5 July 2015.

In this analysis 3TC and FTC were considered to be interchangeable. ART regimens with one, two or four drugs (with the exception of boosted regimens) were not eligible. Regimens were defined according to their third drug with the other two NRTIs considered as the treatment backbone.

The primary outcomes were: viral suppression, mortality, AIDS defining illnesses, discontinuations, discontinuations due to adverse events, and serious adverse events. Secondary outcomes included mean change in CD4. The investigators used GRADE to rate the overall quality of the evidence

The investigators found 5865 citations, they selected 513 of these for full text review and included 126 articles associated with 71 trials in the analysis. The final network of eligible comparisons – including both head-to head and indirect – between treatments included 34 032 patients randomised to 161 treatment groups.

In the assessment of viral suppression (using data from 70 trials, including 31 404 participants receiving 16 third drugs), the analysis revealed dolutegravir to be significantly better than efavirenz at 48 and 96 weeks: the odds ratio (OR) for viral suppression was 1·87 (95% CI 1·34–2·64) with dolutegravir and 1.90 (95% CI 1.40-2.59) at these time points respectively. Raltegravir was the only other third drug that was statistically superior to efavirenz: OR 1·40 (CI 95% 1·02-2.59) and 1.45 (1.07-1.95) at 48 and 96 weeks respectively.

The investigators noted that ritonavir-boosted lopinavir “fared worst” and was inferior to standard dose efavirenz and all INSTI.

The investigators also performed a random-effects network meta-analysis for discontinuations due to adverse events. This showed that dolutegravir had the most protective effect relative to efavirenz: OR 0·26 (95% CrI 0·14–0·47). Low dose efavirenz followed: OR 0·39 (95% Crl 0·16-0·92). They noted that although there was no statistical difference between dolutegravir and low dose efavirenz, their estimations suggested higher rates of discontinuations with the latter drug.

At 48 weeks the mean difference in CD4 count was about 20 cells/mm3 with all three INSTI compared with standard dose efavirenz. Low dose efavirenz was also superior to standard dose with a mean difference of approximately 25 cells/mm3.

Due to insufficient data, the investigators were unable to make any conclusions about mortality, AIDS defining illnesses (both low event rates) or serious adverse events.

In an accompanying commentary, [2] Anton Pozniak and Andrew Hill note that low-income and some middle income-countries will be able to access generic dolutegravir in the not-too-distant future at very low prices through voluntary licensing. But in other middle-income and all high-income countries, the patents on new antiretrovirals will remain for at least another 10 years, keeping the prices high.

The authors argue that in the WHO meta-analysis, the most common endpoints used to define viral suppression classified virological failures as discontinuation of their randomised treatment for any reason. But in most clinical trials with these endpoints only a minority of failures are truly virological. Most people have undetectable viral when they discontinue treatment, doing so because of adverse events or other reasons and can be switched onto alternative treatments to sustain long-term virological suppression, they write.

They explain that in the SINGLE trial, the virological failure rate was actually slightly higher for dolutegravir (10%) than efavirenz (7%). There were more discontinuations for adverse events or other reasons in the efavirenz group (30%) than the dolutegravir group (18%) and a small non-significant risk resistance between the groups.

The authors go on to ask whether dolutegravir – which is significantly more expensive than generic efavirenz in most middle-income and high-income countries – is worth the additional cost. They note that one option, as proposed by the International Antiviral Society-USA treatment guidelines panel, would be to start people on low-cost generics and only switch to more expensive integrase inhibitors for adverse events. “Difficult decisions will need to be taken if we are to achieve the UNAIDS targets for antiretroviral treatment coverage”, they conclude.


It’s no big surprise that dolutegravir fared well in the WHO systematic review and network meta-analysis!

And the first generic version of dolutegravir is on its way: FDA tentative approval of the Aurobindo single was recently granted and it will be available to generic accessible countries for a per person annual cost of about US $44 under an agreement with ViiV and the Clinton Health Access Initiative (CHAI). [3] And dolutegravir-based fixed dose combinations are not far behind.

Hill and Pozniak and their group have looked extensively at prices of antiretrovirals in countries where voluntary licenses are not permitted such as in South America, South Asia and Eastern Europe. [4] They continue this discussion in the Journal of Virus Eradication, arguing for scale up of low-cost generics as patents expire in these regions in order to achieve 90-90-90 targets. [5]

Meanwhile a group of six non-profit organisations, led by the International Treatment Preparedness Coalition are working to make essential HIV medicines more affordable in middle-income countries. [6] They are supporting governments to issue selective compulsory licenses, which allow to generic companies produce the patented product without the consent of originator. They are also challenging undeserved patents.

Mechanisms to ensure fair pricing across middle-income countries need to be much improved to ensure equitable and sustainable access, particularly to improved treatments as they are recommended.


  1. Kanters S et al. Comparative efficacy and safety of first-line antiretroviral therapy for the treatment of HIV infection: a systematic review and network meta-analysis. The Lancet HIV. Published online 6 September 2016.
  2. Pozniak AL and Hill AM. First-line integrase inhibitors for HIV – prices versus benefits. The Lancet HIV. Published online 6 September 2016.
  3. UNAIDS. Press release. Three new agreements announced with the potential to expand access to innovative HIV treatment in low- and middle-income countries. 30 November 2015.
  4. Hill AM and L Pozniak AL. How can we achieve universal access to low-cost treatment for HIV? Journal of Virus Eradication 2016 (in press).
  5. Gotham D et al. Differences in antiretroviral drug prices between countries within and outside sub-Saharan Africa. 21st International AIDS Conference. 18-22 July 2016. Durban South Africa. Poster abstract TUPEE624. (abstract) (poster)
  6. Make medicines affordable.

Links to other websites are current at date of posting but not maintained.