Clonazepam repurposing in ARID1B patients through conventional RCT and N-of-1 trials: an experimental strategy for orphan disease development

Background Clinical trials for rare disorders have unique challenges due to low prevalence, patient phenotype variability and high expectations. These challenges are highlighted by our study on clonazepam in ARID1B patients, a common cause of intellectual disability. Previous studies on Arid1bhaploinsufficient mice showed positive effects of clonazepam on various cognitive aspects. Methods This study used a randomised, double-blinded, placebo-controlled, two-way crossover study (RCT), followed by an N-of-1 design. In the crossover study, ARID1B patients received clonazepam (max 0.5 mg, two times per day) or a placebo for 22 days with a 3-week washout period. Assessments included safety, tolerability, pharmacokinetics, pharmacodynamics on neurocognitive tasks, behaviour and cognitive function. During phase I of the N-of-1 trial the optimal dosage and individual treatment goals were determined. Phase II evaluated the treatment effect. This phase was composed of three periods: an open-label period with placebo (4 weeks), followed by a double-blinded period (6 weeks), followed by an open-label period in which the patient received clonazepam (4 weeks). Results In the clonazepam group (n=16, 15 completing both periods), seven (44%) reported improvement on Clinician Global Impression of Improvement versus two (13%) on placebo. 13 (87%) showed’no change’ after placebo (two (13%) on clonazepam), while seven (44%) on clonazepam reported deterioration, often linked to side effects (n=6), suggesting potential benefit from lower dosing. Three N-of-1 trials with RCT responders saw two patients improve on clonazepam during double-blinding, but clinical evaluation deemed the improvements insufficient. Conclusions Our approach shows the feasibility and strength of combining conventional RCT and N-of-1 studies for therapeutic studies in populations with intellectual disabilities, distinguishing real treatment effects from expectation bias. Our findings suggest that clonazepam has no additional therapeutic value in ARID1B patients.