Controversy still exists regarding which is the best antiretroviral therapy (ART) to start with. We all agree that the regimen must be potent, safe, well tolerated, convenient for the individual, and must demonstrate both efficacy and durability.
Antela and colleagues attempted to address the question by conducting a prospective, randomized, open-label study in ART-naive patients comparing two backbone nucleoside regimens (AZT [zidovudine, Retrovir]/3TC [lamivudine, Epivir)] vs. d4T [stavudine, Zerit]/ddI [didanosine, Videx]) plus either nevirapine (NVP, Viramune), efavirenz (EFV, Sustiva) or boosted indinavir (IDV, Crixivan). In addition to routine labs, CD4 counts and HIV viral load, a genotype was obtained at baseline and in patients developing virologic failure (HIV RNA more than 200 copies/mL).
Sixty-nine patients were enrolled with a median age of 37 years; 77 percent were male; 33 percent had an AIDS diagnosis; 35 percent were former intravenous drug users (IVDUs). Median baseline viral load and CD4 cell counts were 4.92 log10 copies/mL and 180 cells/µL. At baseline, a mutation in the reverse transcriptase (RT) gene was found in two patients (M41L and V118I) and 28 percent of the patients had a mutation in the protease inhibitor (PI) gene (17 percent M36I, 7 percent L10I, 4 percent K20M).
Patients were randomized as follows:
| ||AZT + 3TC plus||d4T + ddI plus|
In an intention-to-treat (M=F) analysis after nine months of follow-up, 74 percent of the patients had a viral load below 200 copies/mL, without significant differences when comparing d4T/ddI vs. AZT/3TC (73 percent vs. 83 percent), nevirapine vs. efavirenz (75 percent vs. 76 percent) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) vs. indinavir/ritonavir (RTV or r, Norvir) (76 percent vs. 85 percent).
The median increase in CD4 count was 184 cells/µL, significantly greater with d4T/ddI compared to AZT/3TC (249 vs. 116 cells/µL, p=0.006), without differences between nevirapine and efavirenz (178 vs. 192 cells/µL) or NNRTIs and indinavir/ritonavir (184 vs. 201 cells/µL). In 13 patients failure was due to adverse events, which occurred more frequently with indinavir/ritonavir. Two patients had virologic rebound, with a major new mutation (M184V) in one. The investigators concluded that virologic efficacy was similar between arms and that immunological recovery was significantly better in the patients receiving d4T/ddI. Tolerability was a concern only for patients receiving indinavir/ritonavir, 23 percent of whom discontinued due to adverse events which has been shown in previous studies.
This study is too underpowered to be able to come to any conclusions regarding virologic efficacy. The difference in CD4 recovery is consistent with other studies comparing non-AZT nucleosides with AZT, which may reflect the myelosuppressive effect of AZT. CD4 percentages were not available nor were absolute white blood counts. Another major problem with this study is that it is no longer clinically applicable given the warnings for potential liver toxicity and lactic acidosis when d4T is given in combination with ddI.
More importantly, ACTG 384, an adequately powered 900-subject study, demonstrated the virologic inferiority and increased toxicity of d4T/ddI compared with AZT/3TC in combination with efavirenz. This current study is unable to address the issues of whether a NNRTI-based regimen vs. a PI-containing regimen is a better first line option and its results add nothing to what we already know.