Trial Termination and Drug Misclassification in Sequential Adaptive Clinical Trials

Sequential adaptive clinical trials allow for early termination of drug testing for benefit or futility at interim analysis points. Early stopping allows the trial sponsor to mitigate investment risks on ineffective drugs and to shorten the development timeline of effective drugs, hence reducing expenditures and expediting patients’ access to these new therapies. However, this new flexibility may translate into a higher drug misclassification rate (i.e., false positives and false negatives). We examine the nature and implications of wrongly terminating the development of an effective candidate drug, which may lead to unrecoverable expenses and unfulfilled patient needs. To this end, we build a simulation model of a phase 3 sequential adaptive trial and focus on the continuation or termination decision at one of the planned interim analysis points based on the feedback from the drug-testing process. This feedback’s accuracy depends on the interim sample size and the candidate drug’s true efficacy. We examine the effects of imperfect information and the conditions that lead to drug misclassification by conducting an extensive Monte Carlo–style sensitivity analysis. Contrary to the literature’s focus on false positives, our results suggest that false negatives can be more likely. Based on our analysis, we provide important insights for trial sponsors, investigators, and other stakeholders on the causes and potential impact of false negatives.