February 4, 1999
| Those who reached PCR < 400 copies/ml | |||
| OBS | LOCF | NC=F | |
| EFV/ZDV/3TC | 98% | 86% | 71% |
| EFV/IDV | 83% | 66% | 53% |
| IDV/ZDV/3TC | 86% | 61% | 48% |
Statistically, there is a difference (p<0.05) between EFV/ZDV/3TC over IDV/ZDV/3TC, in all comparisons. The data have been criticized in that the IDV/ZDV/3TC group experienced much greater dropout (due to treatment side effects) making this group look worse in its achievement of viral load below 500 than the EFV/ZDV/3TC group. A 48% response rate (NC=F) is also lower than what has been reported in other studies with indinavir. But, raising the bar further to compare these three treatment groups in their ability to suppress viral load below 50 copies/ml:
| OBS | LOCF | NC=F | |
| EFV/ZDV/3TC | 90% | 68% | 65% |
| EFV/IDV | 74% | 55% | 47% |
| IDV/ZDV/3TC | 79% | 50% | 44% |
Again, there was statistical superiority of the first group over thae last group (p<0.05).
Further analysis of a subgroup of study participants who exhibited viral burdens in excess of 100,000 copies/ml at baseline continued to show superiority of the ZDV/3TC/EFV arm, reassuring the potency of this regimen even in patients with much higher risk of disease progression. Adverse events were common in the EFZ arms of the study but were of low grade (I or II) in severity. Hallucinations or somnolence was observed in half of those on the ZDV/3TC/EFV arm, though these resolved within a two week time frame, and rash was seen in about 30% of EFV treated individuals.