BKC,

So great to have you chime in. You have a great point about the possibility of the BETonMACE trial being stopped early for efficacy if the statistics are strong enough. Definitely a possibility. I recall Don saying previously that BETonMACE woudn't be stopped early. However, if the interim analysis finds an overwhelming RRR and strong enough statistics to warrant recommending to Resverlogix and the Clinical Steering Committee an early stoppage for efficacy, are we sure they would turn down the recommendation?   

We've dissected that "Projected primary MACE rate still 8.0 per 100 patient years" statement before. By itself, it is ambiguous. However, when I emailed Clayton he responded "To answer your questions, yes the actual events we are observing in the blinded analysis from BETonMACE is close to 8 per 100 patient years as we had planned. As you know this trial remains blinded and therefore, we do not have any access to who is on placebo or who is on drug, so these numbers are derived from all patients." Based on that clarification, the 8 per 100 patient years is an average of all patients in the trial. There are many combinations (some promising, some not so much) of event rates for the placebo and apabetalone groups that would lead to an 8 per 100 patient year average.

If I understand your calculations correctly, you are assuming a 9.5 events per 100 patient years in the placebo group and a 6.5 events per 100 patient years in the apabetalone group. This would be consistent with an 8 events per 100 patient year average for all patients and an a 31.6% RRR. Statistics isn't my strength, so I'll take your word on the p=0.007 and p<0.001 projections assuming the placebo event rate to be 9.5 and 9.9 events per 100 patient years, respectively. 

One very important distinction here. Clinical trials often report the event rates in two ways. One way is the events per 100 patient years. This way is simple to understand. For an 8 per 100 patient year rate, one can think of this as an annual (12 month) event rate of 8%. However, another way that event rates are expressed is to state the event rates at the median follow up period of the trial, which for cardiovascular outcomes trial is going to be at a point much longer that 12 months out. An example is the EXAMINE (Alogliptin) trial that Resverlogix and the Clinical Steering Committee used in part to model BETonMACE. EXAMINE had a median follow up period of 1.46 years and at that point they observed 11.8% vs. 11.3% event rates for 3-point MACE in the two treatment groups. However, if this were expressed as events per 100 patient years, it would be approximately 8.08 vs. 7.74 (annual event rate of ~8%) for 3-point MACE.

I am a little skeptical of the 9.5% annual MACE rate shown on slide 16 for ACS patients with diabetes. In no ACS/diabetes CVOT trial that I've seen has the annual event rate exceeded the ~8% observed in EXAMINE (described above). I agree that the low HDL requirement of BETonMACE should push the event rate higher than that observed in previous ACS/diabetes CVOT trials. However, it is not clear to me from those references on slide 16 how Resverlogix arrived at those numbers shown on slide 16. Perhaps slide 16 includes not just 3-point MACE but other "soft" MACE in those calculations. If anyone knows of a diabetes CVOT trial that had an annual event rate this high, please enlighten me. 

I recall Resverlogix back in the beginning of the BETonMACE trial highlighting a projected event rate of approximately 10.5% (inspired by EXAMINE) at 18 months (1.5 years) as the reasoning for targeting 250 events with 2400 patients. At a median follow up period of 18 months (1.5 years), then 2400 patients X 10.5% is ~250 events. In other words, 3600 patient years was their projected goal with their assumptions. Perhaps this 10.5% event rate at 18 months was a projected average for all patients and assumed 1) that the low-HDL requirement would bump up the ~11.5% observed at 18 months in EXAMINE to a greater number in placebo patients; and 2) apabetalone would greatly decrease (30% RRR) that event rate relative to the placebo group. Note, that if 10.5% average event rate at 18 months is expressed as an annual event rate (12 months), then this works out to an average annual event rate of about 7 events per 100 patient years. By this logic, if BETonMACE is showing an average of 8 per 100 patient years as reported, then overall patients are experiencing a MACE rate greater than projected. However, the discrepency in these numbers could also be in the way the annual (12 month) event rate is being calculated. I fully admit I could be making some false assumptions here.

75% of the projected need of 3600 patient years is 2700 patient years. By my most overly conservative estimate, BETonMACE should be at about 2600 patient years by the end of Q2 2018. If truly the observed event rate is 8 per 100 patient years, then BETonMACE should be at about 206 MACE events by end of Q2 2018. And if that assumption is true then the SSRA is any day now. 

All in my opinion. Do your own due diligence.

BearDownAZ