The following quote attributed to Kausic Ray:

"The compelling signals for a large treatment effect suggest a trial of between 4,000 and 5,000 people with 400 to 500 events in a similar population would be adequately powered to test this approach in reducing CVD. A follow-up study is now needed."

indicates he feels that as a cardiovascular drug, apabetalone needs another trial.  I think he is likely right because the trial results are a bit complicated to interpret so far.  It is not simply that apabetalone has a lower overall efficacy (about 20% RRR for MACE rather than the expected 30% or better).  In fact, in half the patients (those with baseline LDL values below the median value), the drug worked exactly as expected.  It achieved a HR of 0.6 (meaning a 40% RRR).  In the other half of patients, the drug did not work. Not a case of "the drug did not work as well." A case of "the drug did not work at all."  So, this is not just a story of overall reduced efficacy, it is a story of amazing efficacy in patients with relatively low LDL values and zero efficacy (at the current dose) in patients with relatively high LDL values.  Two completely different outcomes of the same drug within diabetic cardiovascular patients. RVX was fortunate (luck favours the well-prepared) enough to pre-specifiy two subgroups defined by a very good biomarker to identify the distinct drug effects.  Since both LDL groups contain half the trial patients (over 1000 with 500 on treatment), small sample size effects are unlikely to explain the results here.  Something is going on...

So, some new hypotheses need to be generated as to why high LDL values seem to preclude benefit from apabetalone at this dose.  And you can bet that the next trial will not be identical to this one in ways other than just size  - it will take into account everything that this trial reveals.  The patient group might be defined differently, the drug dose may be different, the standard-of-care will inevitably be different and the endpoint components are likely to be different. So, there is work to be done on the CVD front and I think it is unlikely that the breakthrough application is for apabetalone as a CVD drug. 

One thing to note is that switching out stroke events for Hospitalization for CHF would have made BoM successful because of the high number of CHF events and the strong RRR that apabetalone had on CHF.  But, and this is a big but, it would not have made much difference to the high LDL group.  Apabetalone did provide a nominal benefit for CHF in this group, but it was cancelled out by nominal detrimental effects in other MACE categories and the overall RRR for the high LDL group would still be around zero.  Using CHF as a MACE component would just have made the already very good 40% RRR for the low LDL group even better!  So, RVX still needs to figure out why the high LDL group are non-responders (remember that over half the MACE occurred in this group). 

Jupe