A randomized trial (ISS 901) of switching to didanosine versus continued zidovudine after the diagnosis of AIDS.

Although the efficacy of switching from zidovudine (AZT) to didanosine (ddI) has already been evaluated in controlled studies, prior investigations were not specifically designed to evaluate this issue in patients with acquired immune deficiency syndrome (AIDS). This open, randomized, multicenter study (ISS 901) was designed to evaluate the clinical benefit in patients with AIDS of switching to ddI after 6-18 months of AZT and no major intolerance. Patients were randomized to continue AZT, maintaining the current dosage at randomization (n = 79), or to receive ddI (n = 80) at the dosage of 375 mg and 250 mg b.i.d. for body weight > 60 and < or = 60 kg, respectively. Primary efficacy measures were survival and time to new AIDS-defining events, analyzed by the intent-to-treat approach. The two groups were comparable for baseline characteristics, follow-up (15 months), and time spent on allocated treatment. At the end of the study, 104 patients (48 AZT, 56 ddI) had died and 90 had at least one new AIDS-defining event (44 AZT, 46 ddI). Kaplan-Meier estimates of survival and of time to first new AIDS-defining event showed no differences between the treatment groups. No differences were detected in other efficacy measurements (p24 antigenemia, CD4+ count, Karnofsky score, and body weight), occurrence of severe toxicities, and treatment modifications. Pancreatitis occurred only in ddI-treated patients (6%). In our population of patients with advanced disease, switching from AZT to ddI did not produce apparent benefits, suggesting that application of this strategy earlier in the course of human immunodeficiency virus disease should be considered.