Boosted saquinavir as good as Kaletra, Gemini study shows

The 11^th European AIDS Conference in Madrid heard that saquinavir boosted by ritonavir (SQV/r – Invirase) is essentially equivalent in terms of potency to boosted lopinavir (LPV/r - Kaletra) in patients new to HIV therapy.

After nearly a year on therapy (48 weeks) the proportion of patients with an undetectable viral load (below 50copies/ml) was 64.7% on SQV/r and 63.5% on LPV/r.

Viral load suppression by the two drugs was also almost identical, with SQV/r achieving a 3.39 log₁₀ reduction in viral load and LPV/r a 3.36₁₀ log reduction (about 2500- and 2300-fold respectively). The probability of this result being due to chance was 0.0058 or one in 172.

Lead investigator, Professor Sharon Walmsley of the University of Toronto, was presenting the results of the GEMINI study, which randomised 337 antiretroviral-naïve patients to receive either SQV/r plus tenofovir/emtricitabine (TDF/FTC -Truvada) or LPV/r plus TDF/FTC.

All patients who started the study had CD4 counts below 350 cells/mm³ and viral loads over 10,000 copies/ml. Four out of five were men, with a median age of 38 years. They were in advanced HIV infection: the median viral load at the start of the study was high, at 160,000 copies/ml and the median CD4 count low, at 138 cells/mm³. Sixty-three per cent had viral loads above 100,000 copies/ml and 40% had CD4 counts below 100 cells/mm³.

One hundred and sixty-seven patients started the study on SQV/r and 170 on LPV/r but the 48-week analysis was of 263 patients, 128 on SQV/r and 135 on LPV/r. About 20 patients in each arm were excluded from the final results because of protocol violations (i.e. they should never have been in the study in the first place) and 30 either changed treatment due to failure or toxicity or were lost to follow-up.

In all, 23% of patients discontinued SQV/r treatment and 21% LPV/r. Of these, 5% versus 8% were due to safety concerns, i.e. adverse lab results, and 3% versus 7% were due to adverse events, i.e. symptoms. The proportion who discontinued due to incomplete viral suppression was 9% in the SQV/r arm versus 3% in the LPV/r arm. There were four deaths, three of them in the SQV/r arm.

The proportion with undetectable viral loads at week 48 was about 5% lower than halfway through the study at week 24, when the proportion with viral loads under 50 had been 69.9% on SQV/r and 69.0% on LPV/r.

The increase in CD4 count was 127 at week 24 and 178 at week 48 for patients on SQV/r; it was 134 at weeks 24 and 204 at week 48 for patients on LPV/r. The differences between the two drugs were not statistically significant.

There were eleven failures due to incomplete viral suppression on SQV/r and five on LPV/r. Five of the eleven SQV failures had at least one HIV protease inhibitor resistance mutation at baseline compared with none of the three LPV failures; five of the patients on SQV and four of those on LPV developed the M184V FTC resistance mutation; and one patient in each arm failed virologically with apparently wild-type, non-resistant HIV.

Only one patient, taking saquinavir, developed a new protease inhibitor resistance mutation (in addition to M184V) while on failing therapy, and they were one of the ones with primary PI resistance. Six patients on SQV/r had ‘documented poor adherence’, Walmsley said, versus two on LPV/r.

In terms of side effects, the study tended to favour SQV/r. Seventeen per cent of patients on SQV/r complained of diarrhoea versus 27% on LPV/r.

In terms of lipids, the percentage of patients with total cholesterol above the upper limit of normal went up in both groups, from 14% to 31% on SQV/r and 10% to 39% in patients on LPV/r. However at week 48 34% of patients on SQV/r versus 24% on LPV/r had raised levels of ‘bad’ LDL-cholesterol.

For triglycerides, the proportion with abnormal levels (two times the upper limit of normal) went down in patients on SQV/r from 1.9% to 1.4% but up from 2.4% to 9% in patients on LPV/r.

“We urgently need more HIV treatment options with favourable lipid profiles, which is why I welcome the results from this study,” Sharon Walmsley commented.