NNRTI: Efavirenz

Nelfinavir (750 mg tid) plus efavirenz (600 mg qd) were administered to 32 antiretroviral naïve and 30 NRTI experienced (but PI and NNRTI naïve) patients with HIV RNA more than 10,000/mL and CD4 more than 50 cells/mm3. At 24 weeks, in an intent-to-treat analysis, 64% of naïve and 41% of NRTI experienced patients had a viral load less than 50 copies/mL. (Abstract I-102.)

Of 59 patients receiving efavirenz + indinavir, at 84 weeks, 74% had viral loads of less than 50 copies/mL and a mean increase of 350 CD4+ T cells/mm3 at the time of their last assay. Contrary to results from other studies, for some inexplicable reason, no CNS toxicity was observed by these investigators. (Abstract I-104.)

Four hundred fifty PI, NNRTI, and 3TC naïve patients (85% were completely antiretroviral naïve) with viral loads of more than 10,000 copies/mL and CD4 counts more than 50/mm3 were randomized to receive one of three regimens: efavirenz + indinavir, efavirenz + ZDV + 3TC, or indinavir + ZDV + 3TC. The somewhat startling 24-week results reported in Geneva continued to hold at the 36-week analysis. In the intent-to-treat analysis, the proportions of patients whose plasma HIV RNA was reduced to less than 50 copies/mL were, respectively, 43%, 64%, and 44%. Rashes and, especially, CNS side-effects (mostly insomnia, anxiety, and dysphoria) were commonly attributed to efavirenz. Only 22% discontinued the drug as a consequence, however. This study demonstrates that efavirenz + ZDV + 3TC is a potent regimen in a largely antiretroviral naïve patient population. The apparent superiority over the comparable indinavir-based regimen is puzzling and is confounded by the high, premature discontinuation rate in this group (41% vs 25-30%) and by the discrepancy with the much better results obtained with this regimen in other studies. (Abstract I-103.)


One hundred twenty-five Brazilian patients with CD4 count less than 300/mm3 were randomized to receive GM-CSF (sargramostim) or placebo in addition to their standard antiretroviral therapy, which consisted of ZDV alone or with a second NRTI. The GM-CSF was administered subcutaneously in a dose of only 125 mg/m2 twice weekly. At six months, the median changes in viral load were -0.51 log10 in the GM-CSF recipients and -0.01 log10 in those assigned placebo (P = 0.02). In addition, 80% of GM-CSF recipients had a more than 30% increase in CD4 count, compared to only 58% in the placebo group (P = 0.027). The incidence of opportunistic infections in the former group was 40% and 63% in the placebo group; this difference was not, however, statistically significant. (Abstract I-243.) Coadministration of GM-CSF does not affect indinavir pharmacokinetics. (Abstract A-72.) The significant reduction in viral load associated with GM-CSF administration is consistent with a recent finding that this cytokine reduces CCR-5 expression and HIV-1 (R5 strain) infectivity of macrophages in vitro (AIDS Res Hum Retroviruses 1998;14:129).

Interferon alpha-n3

The antiretroviral activity of interferon alpha-n3, derived from human leukocytes, had a modest antiretroviral effect, comparable to that of ZDV, in patients with baseline CD4 count of more than 400/mm3. (Abstract I-100.)

Management of Antiretroviral Treatment Failure

The combination of abacavir, efavirenz, and adefovir dipivoxil (plus L-carnitine) was administered to four heavily pretreated patients with viral loads of more than 100,000 copies/mL. All had previously received ZDV, d4T, 3TC, nevirapine, saquinavir, indinavir, nelfinavir, and ritonavir. None, however, had detectable mutations at the 103, 106, or 181 positions of the RT gene associated with NNRTI resistance. The mean log reduction at two months was 0.7 log10. (Abstract I-200.)

In contrast to the more dismal experience of others, 49% of 29 saquinavir hard gel experienced patients (for a median of 20 months), whose therapy was changed to indinavir plus at least one new NRTI, experienced a reduction of viral load to less than 50 copies/mL at six months. Neither the number of preexisting PI mutations, the viral load at baseline, nor the duration of exposure to saquinavir correlated significantly with virologic response. Baseline resistance to 3TC was, however, a predictor of a poor virologic response. (Abstract I-82.)

Arguments for Early Aggressive Treatment

In the opening symposium, the importance of the role of HIV-specific CD4 T cell response to the immunologic control of HIV infection was discussed. A study of a small number of patients with acute HIV infection found that early treatment was associated with restoration or improvement in the CD4 response to antigens of HIV.

Treatment of primary HIV-1 infection with ZDV + 3TC + indinavir was associated with acceleration of the appearance of specific neutralizing antibodies directed at autologous virus when compared to a cohort of untreated patients. (Abstract I-190.)

In a study using an assay with a lower limit of detection of 5 copies/mL, results were found that further support the value of early treatment. Eight of 10 subjects with primary infection who were given ZDV, 3TC, and indinavir achieved a negative result with this assay, while none of those with chronic infection and CD4 count below 500/mm3 did so. While approximately one-half of those with chronic infection and CD4 cell count above 500/mm3 did reach a viral load of less than 5 copies/mL, none were able to maintain this level of viral suppression. (Abstract S-102.)

An analysis of the INCAS study at the Geneva conference demonstrated that, at least with nevirapine-based regimens, it is necessary to achieve viral load suppression below the limits of detection of ultrasensitive assay (< 20 copies/mL) in order to achieve long-term viral suppression.

These findings, taken together, provide a strong argument for not delaying the institution of antiretroviral therapy. Early treatment may also prevent the viral divergence and emergence of SI strain predominance associated with subsequent disease progression. (Abstract S100.)