Efficacy and safety of memantine in children with autism spectrum disorder: Results from three phase 2 multicenter studies.

The randomized trial enrolled only children who had already responded to open-label memantine, so it does not show whether memantine helps children who have not already improved. The loss-of-response criterion was based on a 10-point change on the Social Responsiveness Scale, but a minimum clinically important difference for this scale in autism has not been formally established. The enriched responder design and caregiver-rated outcomes make placebo and expectation effects likely, and the authors say these may have obscured a treatment effect. Many children were taking concomitant medications and supplements, including risperidone, melatonin, multivitamins, ibuprofen, paracetamol, and loratadine, so the contribution of memantine alone cannot be separated from those treatments. The long-term extension was terminated early by the sponsor, so its efficacy outcomes were not fully evaluated. The study used DSM-IV-TR categories autistic disorder, Asperger’s disorder, and PDD-NOS, not current autism spectrum disorder terminology. The controlled trial was phase 2 and the open-label efficacy analyses were exploratory, so the findings are not confirmatory evidence of efficacy or safety.

Funding was provided by Forest Research Institute, the sponsor, and writing support was funded by Allergan plc. The sponsor made the administrative decision to terminate MEM-MD-69 prematurely.

Do not read the large score improvements in the open-label lead-in study as proof that memantine works. Those improvements were not compared with placebo, and in the randomized withdrawal trial children who stopped memantine lost response at the same rate as children who switched to placebo.