AZT PERINATAL HIV INDICATION PHASE III PROTOCOL: OBJECTIONS

AZT PERINATAL HIV INDICATION PHASE III PROTOCOL: OBJECTIONS to the proposed trial were expressed by the FDA Antiviral Drugs Advisory Committee at its Sept. 17 meeting. The advisory panel voted not to recommend a go-ahead for Protocol 076 as presented. The panel offered a number of modifications to make the protocol acceptable, including a request for a pilot study. The Antiviral Committee meeting was the third review of an AZT extension since the committee's inception at the beginning of this year. The protocol, under development by the National Institute of Allergy and Infectious Diseases' AIDS Clinical Trial Group (ACTG) since 1988, was designed as a Phase III randomized, placebo-controlled clinical to evaluate the safety, efficacy and tolerance of oral and I.V. AZT (Burroughs Wellcome's zidovudine, Retrovir) in HIV-infected women and their infants. A Phase I trial is being conducted on asymptomatic HIV-infected women at the University of California at Los Angeles and the University of Miami. The study would examine the use of oral and I.V. AZT through the latter stages of pregnancy (36 weeks gestation or later), delivery and through six weeks of early infancy as a means of reducing the risk of HIV transmission from mother to child. The Centers for Disease Control estimates that one of every three or four babies born to HIV-infected women are themselves HIV-positive. The protocol aimed at a reduction in the rate of perinatal transmission from 30% to 20%, based on having a sample size of 631 women, of whom 563 would have to be evaluable. Under the protocol, women pregnant 36 weeks or more would receive an initial I.V. dose of AZT followed by oral therapy at 200 mg/kg five times a day until they began labor, at which time they would receive 2 mg/kg/hour until they delivered. The baby would then receive 2 mg/kg oral AZT every six hours from birth until two weeks followed by a dosage of 3 mg/kg every six hours from three weeks until termination of the study period at six weeks. Women taking AZT prior to the study would be excluded as would those with CD4 cell counts between 200-500 cells/mm. Major modifications to the protocol suggested by panel members were the initiation of therapy prior to 36 weeks, the addition of toxicology studies in the study's early stages and a demand for a pilot study. Fred Valentine, MD, New York University Medical Center, called for pushing the time for initiating AZT therapy back to the second trimester with a concomitant increase in the study sample size. Deborah Cotton, MD, Harvard School of Public Health, and David Feigal, MD, University of California/San Diego Medical Center, concurred with a second trimester therapy start-date. John Graybill, MD, Audie Murphy Memorial Veterans Hospital (San Antonio), suggested that toxicology studies should be added "up front," to make up for the lack of adequate animal data. Eight of the nine committee members present at the meeting agreed that existing animal safety data are "inadequate" and that existing data are not enough "to support the [protocol's] underlying hypothesis that zidovudine can prevent the transmission of HIV from mother to fetus." Graybill said the protocol should add these studies early on to catch any potential toxicity problems in the babies before large numbers of patients became involved. Further, he wanted long-term follow-up observations on the babies added to the protocol to examine potential neurodevelopmental problems from perinatal exposure to AZT. Richard O'Brien, MD, Centers for Disease Control, Division of TB Control, concurred with Graybill that the protocol should "emphasize toxicity." Donald Abrams, MD, San Francisco General Hospital AIDS Activities Division, called for a pilot study on the protocol. He said the pilot should include women already taking AZT. Harvard's Cotton agreed, saying simply: "Pilot it," and suggested that women with ARC and AIDS also be included. The decision not to recommend the ACTG protocol as proposed was unusual for two reasons. First, the subject matter was not typical of many FDA advisory committee meetings. It is not common for an advisory committee to conduct an open review of a protocol. Secondly, the extent of disagreement between NIAID and FDA over the protocol was reminiscent of past disputes between FDA and the National Cancer Institute over cancer drug development. FDA's Paul Beninger, MD, and Michael Ussery, PhD, both of the Division of Antiviral Drug Products, and statistician Lawrence Hauptman, PhD, Division of Biometrics, each made presentations opposing the protocol. Among numerous objections to the protocol as presented, Beninger said it was "unclear" when transmission of HIV occurs and how differing "patient patterns" affect the rate of transmission. He also criticized the protocol's exclusion criteria as making the potential patient pool "substantially less."