Mortality vs Composite Data: Comparing SGLT2 Inhibitor Trials

In "Mortality vs Composite Data: Comparing SGLT2 Inhibitor Trials," Dr. Haumschild unpacks why similar composite results can mask very different mortality evidence.

Dr. Haumschild tackles a tricky question about weighing mortality-specific data against composite endpoint data. He notes that similar mechanisms and composite results do not guarantee identical underlying evidence. DAPA-HF, studying dapagliflozin, reduced the composite of worsening heart failure or cardiovascular death by 26%. Cardiovascular death itself was also significantly reduced, with a hazard ratio near 0.82. EMPEROR-Reduced, studying empagliflozin, reduced its composite endpoint of cardiovascular death or hospitalization by 25%. But cardiovascular mortality alone was not statistically significant in that trial, with a hazard ratio of 0.92. Dr. Haumschild is careful not to draw a direct comparison between the two agents. Without head-to-head trials, an indirect treatment comparison cannot support a superiority claim either way. That does not mean the endpoints should be ignored when evaluating the evidence together. For P&T purposes, he recommends examining several layers of data before drawing conclusions. Consistency of the heart failure hospitalization benefit across trials is one important layer. Cardiovascular mortality data across the class is another layer worth close examination. Renal outcomes and current guideline recommendations round out the full evidence picture. If a committee becomes comfortable recognizing a class effect, he says it should be named as such. At the same time, individual products can carry genuinely different evidence packages worth acknowledging. Clinical outcomes matter, but so does understanding each trial's design and population. Committees need to examine why one trial met an endpoint while another did not. That analysis separates a true class effect from a difference specific to one agent. The conclusion, he insists, should rest on evidence rather than an assumption based on mechanism alone.

In "Building an Evidence Matrix for SGLT2 Inhibitor Formulary Decisions," Dr. Haumschild closes the series with a practical framework for weighing evidence against rebate economics.