Raltegravir has the edge over two protease inhibitors in three-drug comparison study

In a head-to-head comparison of three drugs presented at the 21st Conference on Retroviruses and Opportunistic Infections (CROI) on Wednesday, the integrase inhibitor raltegravir (Isentress) proved to be superior, in terms of the overall likelihood of treatment failure, when compared to the two ritonavir-boosted protease inhibitors atazanavir (Reyataz) and darunavir (Prezista). All three drugs were taken alongside tenofovir and FTC (co-formulated in Truvada).

The superiority of raltegravir was not due to the likelihood of virological failure, which was not very different for the three drugs, but to discontinuations of the other two drugs, mainly due to gastrointestinal symptoms (both drugs) and to jaundice related to atazanavir. When virological failure and discontinuations were taken together as an overall measure of treatment failure, darunavir was also superior to atazanavir.

The study

In the ACTG A5257 study, 1809 people living with HIV, who had not taken HIV treatment before, were randomised into three groups which received Truvada plus ether: raltegravir; atazanavir boosted with ritonavir; or darunavir boosted with ritonavir, and their progress was monitored for 96 weeks. Their mean age was 37, a quarter were women, 34% were non-Hispanic white, 42% non-Hispanic black, and 22% Hispanic. Their mean viral load was 40,000 copies/ml before treatment, though 31% had a viral load over 100,000 copies/ml, and the average CD4 count at baseline was 308 cells/mm³.

Ninety-two per cent completed the study, with similar numbers in the three groups, although it should be noted that switching to one of the other drugs in the study for reasons of toxicity was allowed.

Results

At 96 weeks, viral load was below 50 copies/ml in 88% of participants who had started treatment with atazanavir, 94% who had started with raltegravir and 89% for those who had started with darunavir. These rates were not statistically different within the pre-determined definition of superiority in the study, although one audience member pointed out that the proportion with a viral load over 50 copies/ml in the raltegravir arm was half the proportion in participants taking the other two drugs.

‘Virological failure’ was defined as two viral load results over 1000 copies/ml in the first six months, or over 200 copies/ml during the rest of the study. By this measure, raltegravir came out slightly better than darunavir, with a 5.6% absolute difference in the rate of virological failure, and this was statistically significant.

However, in terms of discontinuations due to toxicity, atazanavir had a clear disadvantage compared to the other two drugs. Its rate of toxicity discontinuation (16% at week 96) was 13% worse than raltegravir and 9.2% worse than darunavir. Darunavir was 3.6% worse than raltegravir. All these differences were statistically significant. Only 1% of participants taking raltegravir (8 people) discontinued due to toxicity.

The main reason for the atazanavir results being worse for toxicity was that 47 participants discontinued due to the jaundice that is a common side-effect of the drug. This, caused by a build-up of the waste compound bilirubin in the body, is normally considered harmless, but can cause distress about appearance. In addition, there were 25 discontinuations due to gastrointestinal side-effects in people taking atazanavir and 14 in people taking darunavir, compared to only two in people taking raltegravir.

When the two measures – virological failure and toxicity failure – were combined into an overall definition of treatment failure, raltegravir was shown to be 15% better than atazanavir and 7.5% better than darunavir, and darunavir was shown to be 7.5% better than atazanavir. If switching drugs due to toxicity was included in the definition of treatment failure, then the overall efficacy of the drugs at 96 weeks was 80% for raltegravir, 73% for darunavir/ritonavir and 63% for atazanavir/ritonavir. These differences were all significant within the predetermined definition of superiority in the study.

The only area where raltegravir was at a slight disadvantage was that 18 (3%) of participants taking it acquired drug resistance compared to 1.5% of people taking atazanavir and 0.8% taking darunavir. Eleven people had resistance specifically to raltegravir compared to three to atazanavir and none to darunavir.

Participants taking darunavir had slightly lower CD4 rises compared to the other two drugs (256 cells/mm³ compared to 284 on atazanavir and 288 on raltegravir), though this was probably of no clinical significance. As expected, participants taking the two protease inhibitors had greater increases in their cholesterol and triglycerides compared to people taking raltegravir, but there was no change in their HDL-to-LDL ratio (their ‘good’ versus ‘bad’ cholesterol). Analyses of other side-effects are ongoing.

Implications for guidelines

This study is important because it may lead to changes in recommendations for first-and second-line drugs in treatment guidelines.

At present, in the US treatment guidelines, the three drugs are equally recommended for use as the third drug alongside Truvada as drugs of first choice. The non-nucleoside reverse transcriptase inhibitor (NNRTI) drug efavirenz (Sustiva, Stocrin) is also recommended (and is co-formulated with Truvada in the combination pill Atripla); however in one study, over a long time period (five years), raltegravir has already proven to perform significantly better than efavirenz in terms of its ability to maintain continued viral suppression.

The British HIV Association (BHIVA) guidelines have the same recommendations for first-line drugs, except that they add in the four-drug combination pill Stribild (Truvada plus elvitegravir/cobicistat).

In short, raltegravir was shown to be superior to both atazanavir/ritonavir and darunavir/ritonavir; and darunavir to atazanavir, when considering virological and toxicity failure together in this study, and this finding is likely to be taken into account the next time treatment guidelines are revised.