The HIV Prevention Trials Network study HPTN 052 yielded a major breakthrough in HIV prevention last year when early results of the ongoing study revealed that anti-retroviral treatment can reduce HIV transmission by an incredible 96 percent. The study, which includes nearly 1,800 HIV-discordant couples from nine different countries, was named as Science magazine’s 2011 breakthrough of the year. VIDD biostatistician Dr. Ying Chen is the study’s lead statistician. Now, Chen and other VIDD statisticians, including Drs. Benoit Masse, Lei Wang, Deborah Donnell and Thomas Fleming, and San-San Ou and Xin Li, have published an article describing important statistical considerations for HPTN 052.
In a complex study such as this one, several statistical issues are paramount, namely trial design, sample size, trial monitoring and data analysis. HPTN 052 is a two-arm randomized Phase 3 study in which one group of HIV-infected partners were started on ART right away, while the “delayed” group started ART when CD4+ cell counts dropped below 250 or the participant developed an AIDS-defining illness. The study’s primary endpoint is HIV infection of the uninfected partner with a virus that can be genetically tied to their infected partner. Because of this design, the statisticians assumed that the observed effects on HIV transmission might change over time, as more people in the delayed group started ART, and needed to adjust their calculations accordingly. Even though the study researchers decided to offer everyone in the trial immediate ART once the early results were analyzed, the time to initiate ART between the two arms was still different enough that the study can continue to assess whether immediate or later ART is more effective at preventing HIV infection in the long-term.
Chen YQ, Masse B, Wang L, Ou SS, Li X, Donnell D, McCauley M, Gamble T, Ribauldo HJ, Cohen MS, Fleming TR. Statistical considerations for the HPTN 052 Study to evaluate the effectiveness of early versus delayed antiretroviral strategies to prevent the sexual transmission of HIV-1 in serodiscordant couples. Contemp Clin Trials. 2012 Nov;33(6):1280-6.