Pharmaceutical companies and clinical research organizations are seeking greater clarity in an FDA draft guidance covering clinical trials with multiple endpoints.
The draft guidance urges sponsors to control error rates before launching their studies and it provides several examples of appropriate statistical methods. The document offers greater detail than the E9 statistical principles guideline from the International Council for Harmonisation, adopted by the FDA in 1998.
In a public comment, the Association of Clinical Research Organizations said the guidance could be clearer in its recommendations for statistical methods in certain situations — and the FDA should include separate sections clearly delineating the document’s scope and explaining its background.
ACRO said the guidance should include discussions on the complexity of performing interim analyses evaluating multiple endpoints, although the FDA considers this to be out of the document’s scope. If it is not to be included, ACRO suggested, the FDA should address the issue in a separate guidance.