Further information about the combination of atazanavir/ritonavir was presented this week at the Second International AIDS Society Conference on HIV Pathogenesis and Treatment in Paris, suggesting that even when boosted with ritonavir, atazanavir is more lipid friendly than lopinavir.
Results from the Bristol Myers Squibb-sponsored 045 study show that after 24 weeks, PI-experienced individuals are just as likely to have undetectable viral load on atazanavir/ritonavir (300/100mg) as on lopinavir/ritonavir. 64% of the atazanavir group and 62% of the lopinavir group had viral load below 400 copies at week 24, compared with 44% of those treated with atazanavir/saquinavir.
045 is a randomised study in which patients who have experienced failure of at least two prior regimens, and of at least one agent from each of the three drug classes, are randomised to either atazanavir/ritonavir (n=120), atazanavir/saquinavir (400/1200mg qd) (n=115) or lopinavir/ritonavir (n=123). Patients took these agents with their existing nucleoside analogue backbone for two weeks, and then switched to tenofovir and another new nucleoside analogue.
Patients were well-matched at baseline, with a median of around 2.5 years prior PI exposure and five years nucleoside analogue exposure in each group. At baseline around 90% of atazanavir/ritonavir and lopinavir/ritonavir recipients were sensitive to the drug they were allocated, and around two-thirds had fewer than four primary protease mutations. Nearly half had less than four primary nucleoside analogue mutations, implying that a large minority had the opportunity to assemble a dual nucleoside analogue backbone that was active.
The study may not thus have provided the toughest test of the efficacy of atazanavir in heavily pre-treated patients, and week 2 results comparing the virologic effects of the protease inhibitor were not presented. An analysis of response according to genotypic and phenotypic susceptibility scores similar to that performed in the TORO studies of T-20 would be a useful adjunct to any report on studies in heavily pre-treated patients.
After 24 weeks 64% of the atazanavir/ritonavir group, 42% of the atazanavir/saquinavir group and 62% of the lopinavir/ritonavir group had viral load below 400 copies. 39% of the atazanavir/ritonavir group, 23% of the atazanavir/saquinavir group and 42% of the lopinavir/ritonavir group had viral load below 50 copies/ml.
Excluding patients on lipid lowering therapy on entry to the study, lipid changes were as follows:
- Total cholesterol (-8%, -9%, +3%) p<0.0001 favouring atz
- LDL cholesterol (-10%, -11%, -4%)
- HDL cholesterol (-7%, -1%, no change)
- Triglycerides (-2%, -14%, +31%) p<0.0001 favouring atz
During the study, seven atazanavir/ritonavir-treated patients began protocol-mandated lipid-lowering therapy, compared to 15 lopinavir/ritonavir treated patients (p=0.005).
Overall, the incidence of grade 2-4 adverse events did not differ between the groups, but diarrhoea was more common among lopinavir-treated patients (13 cases vs 8 among all atazanavir treated patients), whilst jaundice affected only the atazanavir-treated patients (9 cases). No patients withdrew due to jaundice, and although 45% of the atazanavir/ritonavir group experienced grade 3 or 4 bilirubin elevations, none withdrew because of it. Jaundice on atazanavir is caused by elevations in unconjugated bilirubin.
Further follow-up and analysis from this study will hopefully include not only response according to baseline resistance, but an analysis of response according to inhibitory quotient (the ratio of plasma trough drug levels to IC90 (the concentration needed to inhibit 90% of virus replication)).
One of the big unanswered questions about atazanavir is whether high enough drug levels can be achieved with ritonavir boosting to overcome higher numbers of protease mutations. The rationale presented for dosing atazanavir once daily with ritonavir as 300mg/100mg is to ensure a sustained plasma level above 1000ng/dL - around fifty times the median trough level - compared to a plasma level closer to 100ng/dL when atazanavir is dosed alone. Knowing whether these levels are being achieved will encourage confidence in atazanavir/ritonavir.