Two studies presented on Monday at the 40th ICAAC raised some difficult questions about the role of nelfinavir in antiretroviral therapy.
Nelfinavir is one of the most frequently prescribed drugs for first-line therapy in both the UK and the US, partly because it seems fairly well tolerated, and partly because its resistance profile offers opportunities for second-line protease inhibitor therapy.
However, questions have been raised in recent months about the potency of nelfinavir relative to other PIs and even to NNRTIs. At ICAAC, 40 week results of a head to head study of nelfinavir and lopinavir/ritonavir (ABT-378/r) were presented that suggest lopinavir/ritonavir-based HAART to be significantly more potent. Six hundred and fifty four patients were randomised to received d4T/3TC with either nelfinavir three times daily or lopinavir/ritonavir. After 40 weeks, an intent to treat analysis showed that 79% of the LPV group and 64% of the NFV group had viral load below 400 copies, and 70% of the LPV group and 54% of the NFV group had viral load below 50 copies.
Arguably this study was not a fair comparison, because lopinavir/ritonavir is a boosted PI (co-dosed with a small amount of ritonavir), whilst nelfinavir was dosed three times daily, which results in a less favourable pharmacokinetic performance than twice daily dosing. The study protocol switched to twice daily dosing midway through recruitment, but we don't know what proportion of time on drug as a proportion of all those treated can be regarded as bid nelfinavir treatment.
This problem also complicates interpretation of the COMBINE study, a head to head comparison of nelfinavir and nevirapine, once again in treatment-naïve individuals. Thirty-six week results of this open label study showed that 66% of the nevirapine group had viral load below 20 copies, compared to only 38% of the nelfinavir group, a highly significant difference. Roche (manufacturers of nelfinavir) argue that nelfinavir's poor performance is largely attributable to drop outs rather than virological failure, and that discontinuations due to intolerance were not substantially greater in the nelfinavir arm. So why did people drop out of the nelfinavir arm more frequently? Roche argue that it's because they didn't want to take nelfinavir three times a day when they knew that it was standard practice to dose twice daily.
Relying on trials designed several years ago when making treatment decisions now is always perilous in the field of HIV, but these studies are nevertheless a cause for concern, regarding both the potency of nelfinavir and the use of lopinavir/ritonavir. Is it really sensible to use a drug with proven activity against multi-PI resistant virus as first-line therapy? Some would regard this as wreckless. On the other hand, is it sensible to use a drug when significant questions arise about its potency? The performance of nelfinavir in the small sub-set of COMBINE participants with baseline viral load above 100,000 copies was even poorer than across the trial population as a whole, but at present we have no evidence that the discontinuation rate was especially high in this sub-set.
Further clarification of these data are likely at the Glasgow Drug Therapy in HIV Infection conference next month.