I remember Don McCaffrey giving an explanation for why the Crestor group did better than the Lipitor group...and it had little to do with the statins having a differential impact, and more to do with the Crestor group having a higher incidence of diabetes than the Lipitor group.  Maybe my memory is all messed up.

In any case, to this day I am not sure why they went after plaque regression as an end point in the first place.  Impressive result (if proven) but as we know, the medical profession is more interested in MACE reductions than plaque regressions.