"Ha ha. Nice try. You can't have an interim analysis without an interim analysis."

I should clarify. The six prior DSMB reports that found no safety or efficacy concerns are technically interim analyses. These are looking for obvious safety concerns/side effects/toxicity/unanticipated adverse events. These periodic checks would also identify any overwhelming efficacy concerns (harmful or beneficial) that may warrant stoppage of the study. This stoppage would only likely occur if the stats were super super super strong. One caveat is that long-term safety data is considered very valuable to make sure any early benefits seen in a study won’t be temporary. So favorable interim results must really be considered overwhelming evidence for the DSMB to advise stoppage. Since this is such a relatively short cardiovascular outcomes trial, I must emphasize overwhelming again.

The distinction between these periodic DMSB safety/effficacy checks and a pre-specified futility and/or sample-size re-estimation analysis is that the SSRA is a pre-specified statistical test that will be used to inform and potentially adapt the trial design. Resverlogix and the Clinical Steering Committee included the pre-specified SSRA in their original BETonMACE trial design with the intention of using the info to potentially adapt the trial. These types of pre-specified statistical tests of interim data often come with a statistical penalty for the final trial statistical analysis (those statisticians in the group here can chime in), so it is not advisable to do these too often. But the benefit of doing this SSRA at a late point in the trial is that this adaptive trial design allows changes to be made to correct for false/uncertain assumptions at the trial design stage for treatment effect, control event rate and subject discontinuation rate.

Some might say that a boring SSRA that states "continue on, no changes" would be pointless. However, I would find it very valuable to know with certainty that a pre-specified SSRA isn't going to 1) increase sample size; 2) increase cost; 3) increase trial duration. Importantly, this would validate DM's recent statements of the trial being fully enrolled with topline around end of year. Furthermore, we have had zero indication of how many events to date there are. The futility analysis that was supposed to happen at 50% of events (125 events) was skipped. Therefore, an SSRA announcement at 75% of events (188 events) would time stamp for the first time where BETonMACE is for adjudicated 3-point MACE events. It is interesting that the treatment duration is now beyond 104 weeks. Instead of increasing patient years by adding more patients, this would increase patient years by extending duration of current patients without increasing sample size. By not requiring patients to stop dosing at 104 weeks, the patient year accumulation will occur at a higher rate and MACE occurence should happen faster than if these patients had to stop dosing.

BearDownAZ