An analysis of two studies of the new HIV integrase inhibitor dolutegravir presented at the Eleventh International Congress on Drug Therapy in HIV Infection in Glasgow had the incidental effect of bringing into question evidence from a previous study that suggested that the NRTI (nucleoside reverse transcriptase inhibitor) drug abacavir was less potent in people starting HIV therapy with high viral loads than another NRTI drug, tenofovir.
This new evidence may have considerable importance because it could cause the compilers of treatment guidelines to revise their views on whether abacavir-containing regimens in antiretroviral combination therapy should be rated as equal to tenofovir-based ones, with a number of implications for patient choice and cost.
This year’s HIV treatment guidelines issued by the British HIV Association (BHIVA) recommended that the best NRTI choice for first-line regimens was tenofovir plus FTC (Truvada, or in a triple pill with efavirenz as Atripla). They recommended abacavir plus 3TC (Kivexa, called Epzicom in the US) as an alternative choice, but only for people starting treatment with viral loads below 100,000 copies/ml.
The authors did this after weighing up the evidence from a number of trials, but due to its size the findings from one trial predominated. In this trial, ACTG 5202 (Sax), the time to treatment failure in people taking Kivexa who started with high viral loads was significantly shorter than the time in people taking Truvada. (ACTG 5202 also compared boosted atazanavir with efavirenz, but found no difference in their efficacy.)
This recommendation in turn gave the appearance of clashing with the prescribing recommendations of the London Specialised Commissioning Group, which issued a strong recommendation to London HIV clinics in April 2011 that they should start a higher proportion of patients on Kivexa-based regimens for cost reasons (this shift was achieved and the required savings made, but in fact the majority of London patients still start on Truvada).
Findings that abacavir is as potent as tenofovir, at least in some regimens, may therefore be of some importance.
At the Glasgow conference, Joseph Eron of the University of North Carolina compared viral suppression results at 48 weeks in two trials of the new integrase inhibitor dolutegravir. In a trial called SPRING-2, presented at the International AIDS Conference in Washington this summer, dolutegravir was as potent as the first-generation integrase inhibitor raltegravir, while in another study called SINGLE, presented at the ICAAC conference in September, it proved superior to the standard-of-care drug efavirenz, producing a viral load of under 50 copies/ml in 88% of patients at 48 weeks, compared with 81% on efavirenz.
In SPRING-2, doctors chose whether to accompany the integrase inhibitors with either Kivexa or Truvada while in the SINGLE study, participants were randomised either to dolutegravir plus Kivexa/Epzicom or to Atripla.
In SINGLE, the 7% superiority of dolutegravir/abacavir/3TC over Atripla appeared to be independent of trial subjects’ initial viral loads. In people with viral loads under 100,000 copies/ml, the respective proportions with viral loads below 50 copies/ml were 90% on dolutegravir/Kivexa and 83% on Atripla. Viral load suppression rates were lower in people starting treatment with viral loads over 100,000 copies/ml but the 7% superiority margin was unchanged: 83% of dolutegravir/Kivexa versus 76% on Atripla were virologically suppressed.
The point here is that if abacavir is less potent than tenofovir in people with higher viral loads, one would expect the dolutegravir superiority margin to be smaller in those patients, and it wasn’t.
In SPRING-2, it was possible to make a more direct comparison of Kivexa and Truvada in people with high and low viral loads. In people with viral loads under 100,000 copies/ml who took Kivexa, 87% and 88% respectively on dolutegravir and raltegravir achieved undetectable viral loads, and in people who took Truvada 92% and 91% respectively.
In people starting with viral loads over 100,000 copies/ml who took Kivexa, 81% and 82% respectively on dolutegravir and raltegravir achieved undetectable viral loads, and in people who took Truvada 83% and 71% respectively. None of these differences were statistically significant.
Neither of these studies were set up specifically to compare tenofovir-based with abacavir-based regimens. However with 1655 subjects, of whom 495 (30%) had an initial viral load over 100,000 copies/ml, they may be large enough to make HIV specialists re-evaluate the relative potency of abacavir versus tenofovir.
With another meta-analysis being published recently that found no raised rates of heart attack in people taking abacavir in randomised controlled trials (Ding), in contrast to findings in cohort studies, this could also swing the balance of favour back towards the equal recommendation of abacavir and tenofovir in treatment guidelines.
Eron J et al. Dolutegravir treatment response by baseline viral load and NRTI backbone in treatment-naïve HIV-infected individuals. Eleventh International Congress on Drug Therapy in HIV Infection, Glasgow, abstract P204, 2012. See abstract here.
Sax PE et al. Abacavir-lamivudine versus tenofovir-emtricitabine for initial HIV-1 therapy. N Engl J Med 361: 2230-2240, 2009. See here for full text of study.
Ding X et al. No association of abacavir use with myocardial infarction: findings of an FDA meta-analysis. JAIDS 61(4):441-447, 2012. See abstract here.
