The multi-crossover model for classifying patients as responders to a given treatment

Area of Science:

• Clinical Pharmacology
• Gastroenterology
• Biostatistics

Background:

• Accurate classification of individual therapy responders is crucial for treatment optimization.
• The multi-crossover (MCO) model is a potential method for classifying responders.
• Non-ulcer dyspepsia (NUD) presents challenges in identifying effective treatments.

Purpose of the Study:

• To evaluate the validity and strength of the MCO model for classifying therapy responders.
• To optimize the MCO model's procedure for accurate responder classification.
• To assess the efficacy of ranitidine in NUD patients using the MCO model.

Main Methods:

• A 6-week double-blind MCO trial involving 115 NUD patients and alternating ranitidine/placebo treatments.
• Calculation of an individual effect score (X score) based on symptom changes relative to placebo.
• Reclassification of responders and unclassifiables through cross-tabulation of efficacy and adverse effects, followed by a placebo-controlled relapse study.

Main Results:

• Eighty-five percent of initial MCO responders and 62% of unclassifiables were reclassified as responders.
• Reclassified responders showed significantly higher relapse rates and shorter time to relapse compared to MCO unclassifiables.
• The MCO model demonstrated reliability in classifying responders, with potential for improved accuracy by including patients with an X score of 3.

Conclusions:

• The MCO model is a reliable method for classifying individual therapy responders.
• The study validates the MCO model's strength in identifying patients who benefit from treatment.
• Modifying the MCO model's responder definition (including X score of 3) can optimize classification accuracy.