Early termination of clinical trials with prolonged observation of individual patients: a case study

Area of Science:

• Clinical Trials Methodology
• Cardiovascular Research

Background:

• Placebo-controlled clinical trials are essential for evaluating new anticoagulant therapies.
• Acute myocardial infarction (AMI) necessitates effective anticoagulant treatments.

Purpose of the Study:

• To evaluate optimal stopping rules for anticoagulant trials post-AMI using computer simulations.
• To define criteria for early trial termination based on interim analyses.

Main Methods:

• Computer simulations were employed to model a placebo-controlled clinical trial.
• Five interim analyses were incorporated into the simulation design.
• The logrank test was used to determine one-sided P-values for hypothesis testing.

Main Results:

• Specific P-value thresholds were identified for rejecting the null hypothesis at interim analyses (0.005, 0.005, 0.005, 0.014, 0.023) and final evaluation (0.032).
• Early termination for futility was defined by P-value thresholds (0.95, 0.88, 0.81, 0.74, 0.67).
• The proposed stopping rules maintain a total alpha of 0.05 and power comparable to fixed-sample testing.

Conclusions:

• The study provides evidence-based stopping rules for anticoagulant trials in AMI patients.
• These rules facilitate efficient trial conduct without compromising statistical integrity.
• Implementing these rules can optimize resource allocation in clinical research.