A comparison of two dosing regimens of zidovudine in Thai adults with early symptomatic HIV infection: Conducting clinical HIV trials in South-East Asia.

Aim: To compare the clinical and immunological efficacy, and tolerance of two dosage regimens of zidovudine (ZDV) in an adult Thai population with early symptomatic human immunodeficiency virus (HIV) disease and to identify important clinical issues associated with conducting HIV trials in South-East Asia. Method(s): HIV-infected Thai adults, with early symptomatic HIV disease and CD4 lymphocyte counts less than 400/mm3, who were managed in the infectious diseases clinics at two university teaching hospitals in Bangkok, Thailand, were enrolled in a randomised, open-label, dose-regimen comparison trial of ZDV. Two oral ZDV dosing regimens: regimen A, 100 mg tid+200 mg nocte (ZDV-A) vs regimen B, 250 mg bid (ZDV-B) were compared. The main outcome measures were: 1. Clinical efficacy: rate of progression to acquired immunodeficiency syndrome (AIDS) or death. 2. Immunologic efficacy: changes in CD4 lymphocyte numbers compared to baseline; rate of decline of CD4 lymphocyte numbers to less than 100/mm3. 3. Toxicity, as defined by clinical symptomatology and laboratory parameters. Result(s): Two hundred and four patients were enrolled (103 ZDV-A; 101 ZDV-B) of whom 195 were followed beyond baseline. Patients were typical of those encountered with HIV in Thailand: mean age 33 years; 89% male; 88% heterosexual HIV acquisition; mean baseline CD4 lymphocyte count 241/mm3. Follow-up while on therapy was comparable for the two groups (mean +/-SD): 533+/-236 days (ZDV-A) vs 592+/-210 days (ZDV-B). One hundred and eleven patients (57%; 51 ZDV-A; 60 ZDV-B) were treated for at least 22 months (669+/-30 days). Clinical and immunological outcomes for ZDV-A and ZDV-B, including rate of progression to AIDS or death, development of non-AIDS-defining opportunistic infections, mean changes in CD4 lymphocyte numbers/mm3, difference in area under the CD4:time distribution curve and difference in the rate of decline of CD4 lymphocyte numbers to less than 100/mm3, were not significantly differe