Pimavanserin Fails in Autism

Placebo worked just as well as a drug designed to do better

STUDY: Bugarski-Kirola D et al, Journal of the American Academy of Child & Adolescent Psychiatry 2026;

STUDY TYPE: Randomized, double-blind, placebo-controlled phase 2 trial

FUNDING: Acadia Pharmaceuticals Inc.

Background

Irritability affects up to 80% of autistic children. Only two medications, risperidone and aripiprazole, are FDA-approved for it, and both are antipsychotics with metabolic and movement problems. This trial looked at whether pimavanserin, which has a safer side effect profile and little dopamine blockade, is effective.

Pimavenserin is approved for adults with Parkinson’s psychosis as Nuplazid, but recent attempts to expand its indications into depression and schizophrenia (where it was tested for negative symptoms) have failed.

The Study

• 237 children and adolescents ages 5 to 17 with autism and marked irritability, enrolled across 57 sites in 8 countries.
• Randomized 1:1:1 to low-dose pimavanserin, high-dose pimavanserin, or placebo for 6 weeks.
• Primary outcome: caregiver-rated irritability scale (ABC-I).

All three groups improved. Placebo dropped about 9.6 points on the ABC-I; both pimavanserin groups dropped about 11.2 points. The difference, roughly 1.6 points, was not statistically meaningful (Cohen’s d = 0.17). No secondary endpoints favored pimavanserin. Responder rates at week 6 were 54% on placebo, 59% on low-dose, and 58% on high-dose: essentially the same.

Side Effects

Side effect rates were nearly identical across groups, around 50%. No serious events occurred in either pimavanserin group. The most common complaints were upper respiratory infections, headache, and somnolence.

Limitations

The low dose did not reach target plasma exposure. The primary outcome relied on caregiver ratings, which may have enhanced placebo effect as they are especially prone to placebo-by-proxy effects in pediatric trials.

Practice Implications

1. This is a well-run, adequately powered negative trial. Pimavanserin doesn’t work here.
2. Before reaching for an antipsychotic in autism, make sure the severity of the problem warrants the risks of the intervention. If it is not severe, consider psychotherapy and environmental change first.
3. Clonidine and guanfacine are safer options, and a few supplements have evidence from small RCTs for irritability in autism, like omega-3, coenzyme Q10, sulforaphane, palmitoylethanolamide, and micronutrients.
4. Learn more options in our Carlat interview with Josh Feder.