In Reply: We agree with Drs Arribas and Pulido
that insufficient statistical power may limit the conclusions drawn from a
comparative clinical study and their value for treatment recommendations.
However, we believe that the 2 studies1,2 they
cited deserved to be cautiously mentioned in our recommendations because,
despite limitations, they bring scientific information pertinent for discussing
therapy with nevirapine in treatment-naive patients. In the COMBINE study,2,3 for instance, in which 142 patients
were enrolled, the proportion of patients with plasma HIV-1 RNA level below
detection limits after 1 year of therapy was 75% in the nevirapine-containing
arm and 60% in the nelfinavir-containing arm,3 allowing
the authors to conclude that "zidovudine/lamivudine/nevirapine is at least
as effective as zidovudine/lamivudine/nelfinavir as first-line therapy."
Yeni PG, Hammer SM, . Underpowered Clinical Trials of Antiretroviral Treatment—Reply. JAMA. 2002;288(17):2120-2121. doi:10.1001/jama.288.17.2118