Adaptive designs have been identified as methods for improving the efficiency of the clinical trial process but there are concerns that some techniques can lead to bias. These concerns were highlighted in a recent paper in the Journal of the American Medical Association.
To help companies and regulatory agencies navigate this emerging area the US Food and Drug Administration (FDA) has drafted guidance which discusses the clinical, statistical and regulatory aspects of various adaptive design studies.
Speaking to Outsourcing-Pharma, Jim Higgins, executive vice president (EVP) and director of biometrics at Health Decisions, said the document is balanced and contains a lot of practical information.
This could help to reduce confusion and allow pharma and contract research organisations (CRO) to embrace certain acceptable methods. Well understood adaptive designs covered by the draft include group sequential methods, which Higgins believes are “nicely clarified”.
Debate
Higgins explained that adaptive design is a well debated topic in the statistics and clinical research communities and he anticipates this will lead to a “robust discussion” of the draft.
The only comment on the draft so far has come from William Rosenberger, professor and chair, department of statistics, George Mason University, Virginia, US, who said the “document is remarkably narrow for such an important topic”.
Higgins believes the document’s scope is deliberate, with the FDA choosing to focus on the better understood basics before branching out to other areas in the future.
The comment period closes on June 1.