Psychiatry · Clinical trials
Anorexia Nervosa Clinical Trials
Approved therapies, the pivotal-trial endpoints they were judged on, the magnitude of benefit — and the drugs that failed their endpoints, and why.
Indication overview
About anorexia nervosa — and why its trials are hard
Anorexia Nervosa (AN) is a serious psychiatric illness marked by restriction of energy intake, intense fear of weight gain, and disturbed body image, carrying one of the highest mortality rates in psychiatry. Critically, there is NO FDA-approved medication for anorexia nervosa. Treatment centers on nutritional rehabilitation and psychotherapy (notably family-based treatment in adolescents and CBT), with medications used only off-label and without robust efficacy for the core disorder. Olanzapine is the most studied off-label agent: a randomized trial (Attia et al., Am J Psychiatry 2019) showed a modest increase in BMI versus placebo but no benefit on core psychological symptoms such as body-image disturbance or obsessional cognitions. Fluoxetine failed to prevent relapse after weight restoration in a rigorous trial (Walsh et al., JAMA 2006). Other agents (dronabinol, antipsychotics broadly) have shown only marginal or inconsistent effects. The field represents a stark unmet need: repeated pharmacologic failures reflect incomplete understanding of AN's neurobiology, high dropout, and the confounding effects of starvation on drug response.
Where trials have failed
Drugs that missed their endpoint — and what contributed
The most instructive lessons in anorexia nervosa development come from programmes that failed the endpoint that mattered.
| Drug / trial | Endpoint outcome | What contributed |
|---|---|---|
| Fluoxetine — Walsh et al., JAMA 2006 (relapse prevention after weight restoration) | Failed — no benefit over placebo in preventing relapse or improving outcomes in weight-restored patients | High dropout; serotonergic function may be impaired in low-weight states; no signal for core AN psychopathology |
| Olanzapine — Attia et al., Am J Psychiatry 2019 (RCT in adult outpatients) | Modest weight/BMI gain but no effect on core psychological symptoms; remains off-label, not FDA-approved for AN | Effect limited to weight; no improvement in body-image or obsessional features; metabolic side effects and adherence concerns |
| Dronabinol — Andries et al., Int J Eat Disord 2014 (crossover RCT) | Small statistically significant weight gain but not clinically transformative; not pursued to approval | Small sample, short duration, and modest effect insufficient to change core illness; regulatory path not advanced |
Choosing the right endpoint
Primary endpoints that matter in anorexia nervosa trials
- Body Mass Index (BMI) / weight restoration — Primary physiological target, but weight gain alone does not equal recovery
- Eating Disorder Examination (EDE) score — Assesses core cognitive symptoms — restraint, shape/weight concern — where drugs consistently fail
- Relapse rate after weight restoration — Key durability endpoint; fluoxetine failed here (Walsh 2006)
- Treatment retention/dropout — Very high dropout confounds AN trials and limits interpretability
How iNGENū runs anorexia nervosa trials
Physician-led design, built for FDA submission
Endpoint & biomarker strategy
Board-certified specialists design endpoints and patient selection aligned to current FDA guidance for this indication.
FDA-ready data
Built to ICH-GCP and 21 CFR 312.120, with direct FDA submission — data accepted by the FDA, EMA and MHRA.
Faster, lower-cost delivery
~4-week ethics via the TGA CTN scheme, up to 43.5% R&D rebate, and 80–90% below US CRO cost.
Request a fixed-milestone proposal and a tailored endpoint & feasibility summary for this indication.
Frequently asked questions
Anorexia Nervosa clinical trials — FAQs
Is there an FDA-approved drug for anorexia nervosa?
Does olanzapine treat anorexia?
Why have so many drugs failed in anorexia?
Related
Explore more
Ready to discuss your anorexia nervosa trial?
Talk to our physician-led team about an FDA-ready, cost-efficient trial design.
Request a proposal