What I would add is that 13 is an awfully small number for a sample (the placebo) and would bias the p value to be insignificant.  Since you are comparing a sample of 35 (not so small) with this tiny placebo group, the difference is again biased towards non-significance.  I would put a lot more weight on the eGFR statistical significance on the RVX sample group alone, than on the difference with the placebo.  

This suggests to me that the Phase 2 testing was sub-optimal, at least as far as CKD testing is concerned.  Perhaps RVX did not think it (CKD improvements) would turn out to be so great and devoted a lot less attention to it?