Thursday, July 17, 2008

Made what some are saying is a “landmark decision not to proceed with the PAVE 100 HIV vaccine trial of the NIAID VRC product. This followed weeks and months of committee meetings and recommendation such as the AIDS Vaccine Research Subcommittee (AVRS) who voted to proceed and the Strategic Working Group (SWG) who voted not to proceed. The issue was that the Merck STEP trial failed and actually had some safety issues. The PAVE trial is somewhat different (3 DNA primes and an Ad5 boost containing env) from the STEP trial (3 Ad5s without env), yet the preliminary immunological data from PAVE 100 was in my mind (some such as Gary Nabel disagreed) not much better than those of STEP. I received many e-mails from people arguing for and against. The decision was mine alone and I decided to reject the proposed PAVE trial of 2400 people since it was powered to do immunological correlates on a subgroup who might seem to benefit even if the endpoint of the entire group was not “effective”, i.e. decrease in viral set point after infection. I argue that if the trial failed, I am not interested in the immunological correlates in a subgroup since this has no statistical value. There was much activity in the Press about my decision and the feedback was overwhelming that I had made the right choice. I encouraged Larry Corey of HVTN and Scott Hammer the PI of PAVE to come back to me with a “leaner, meaner” protocol that is less expensive, fewer people, less samples to ask if the product works. If it does, then we will go back and do immune correlates.