Clinical trials landscape¶
TL;DR — The registered landscape is fragmented across mechanisms, feasibility studies, caregiver interventions and small experimental trials. A live ClinicalTrials.gov v2 query on 2026-09-02 returned both AN-specific and mixed-eating-disorder studies; registration does not imply evidence of benefit. Small planned samples and terminated pilots are common — three neuromodulation or neurofeedback registrations in the table below terminated at 1, 2 and 10 participants, and a formally designed olanzapine feasibility study in young people recruited 20 of a planned 55 (Filiz 2026, PMID 42116676). The field needs harmonized recovery outcomes, longer follow-up and transparent reporting of concurrent nutritional/psychological care.
Selected live records¶
| NCT | Status on 2026-09-02 | Intervention/question | Enrollment (E = estimated, A = actual) |
|---|---|---|---|
| NCT07169747 | Recruiting | Two psilocybin doses with psychological support in young adults | 40 (E) |
| NCT06624397 | Recruiting | New program for carers of people with AN | 50 (E) |
| NCT06752304 | Recruiting | Longitudinal severe/enduring eating-disorder study | 800 (E) |
| NCT06144905 | Recruiting | Norwegian microbiota study in AN | 180 (E) |
| NCT05056597 | Recruiting | Incentive processing/learning in AN and bulimia nervosa | 252 (E) |
| NCT06332963 | Recruiting | Interoceptive mechanisms of body-image disturbance | 102 (E) |
| NCT05477537 | Recruiting | Adapted physical activity in treatment | 220 (E) |
| NCT05368844 | Terminated | TMS in AN | 10 (A) |
| NCT02535780 | Terminated | Transcranial treatments in eating disorders | 1 (A) |
| NCT05834816 | Terminated | Stress and neurofeedback in AN | 2 (A) |
| NCT04378101 | Completed | Eating Disorders Genetics Initiative | 17,991 (A) |
| NCT01184443 | Terminated | Olanzapine in children/adolescents | 38 (A) |
| NCT01280799 | Completed | Family treatment for adolescents | 26 (A) |
Registrations behind the trials cited elsewhere in this knowledge base, all resolved live on 2026-09-02:
| NCT | Status | Trial | Enrollment |
|---|---|---|---|
| NCT02488109 | Completed | StRONG higher- vs lower-calorie refeeding (Garber 2021, PMID 33074282; Golden 2021, PMID 33753542) | 120 (A) |
| NCT00149786 | Completed | FBT vs adolescent-focused individual therapy (Lock 2010, PMID 20921118; Le Grange 2014, PMID 25440306) | 120 (A) |
| NCT04661514 | Completed | Phase 1 psilocybin in AN (Peck 2023, PMID 37488291; commentary Majić 2023, PMID 37488290) | 16 (A) |
| NCT01476540 | Completed | Subcallosal cingulate DBS (Lipsman 2013, PMID 23473846; Lipsman 2017, PMID 28238701) | 15 (A) |
All NCT identifiers and statuses above were re-fetched individually from the ClinicalTrials.gov v2 API on 2026-09-02 and every one resolved with the status and enrollment shown. Some records include mixed diagnoses; readers must inspect eligibility before treating them as AN trials.
The termination pattern is the most informative thing in the first table. Three of the four terminated studies are neuromodulation or neurofeedback trials that stopped at 10, 2 and 1 participant, and the fourth is an olanzapine trial in children and adolescents that stopped at 38. Those are recruitment failures, not negative results, and they leave no publishable evidence behind. The same pattern recurred in the OPEN olanzapine feasibility study, which pre-screened 52 young people, found 35 eligible, recruited 20 and retained 15 on drug for ≥16 weeks — and concluded that the recruitment target and adherence rate were the binding constraints on a future RCT (Filiz 2026, PMID 42116676). Feasibility, not equipoise, is what is stopping trials in this field.
The registered portfolio in aggregate¶
Counting rather than sampling changes the picture. A ClinicalTrials.gov v2 count query for interventional studies with condition "anorexia nervosa", run live on 2026-09-02, returned:
| Overall status | Interventional records |
|---|---|
| Completed | 159 |
| Recruiting | 55 |
| Terminated | 23 |
| Not yet recruiting | 22 |
| Active, not recruiting | 14 |
| Withdrawn | 12 |
| Enrolling by invitation | 8 |
| Unknown status | 44 |
| Total | 337 |
Thirty-five records — 23 terminated plus 12 withdrawn — represent trials that stopped early or never started, about 10% of the 337-record portfolio. The 44 unknown-status records are a separate reporting limitation and should not be treated as completed studies.
The drug pipeline is the striking part. Filtering the same query to phase 2 and phase 3 returns 49 records for the entire history of the register, and inspection of the first 20 by relevance shows what they contain: bone-directed agents (teriparatide, denosumab, romosozumab, rhIGF-1, growth hormone, parathyroid hormone), hormonal studies (oestrogen), repurposed psychotropics (olanzapine, quetiapine, mirtazapine, naltrexone), psilocybin, omega-3 supplementation and neurofeedback. Actual enrolments in that set include 2 (denosumab, terminated), 9 (oral naltrexone in paediatric eating disorders, terminated), 12, 15, 21, 21, 23, 23, 24 (quetiapine, terminated), 28, 30, 34, 77 and 90; the largest current phase 2 record is a 170-participant olanzapine-versus-mirtazapine comparison that is enrolling by invitation, and one MDMA-assisted-psychotherapy multi-site study was withdrawn at 0 participants (all figures from the live ClinicalTrials.gov v2 API, 2026-09-02).
Among the 55 currently recruiting interventional studies, most carry phase "NA" — they are behavioural, device or service interventions. Six records carry a phase designation: naltrexone neuroimaging (NCT05509257, early phase 1), focused-ultrasound capsulotomy (NCT07113665, phase 1), metreleptin (NCT06305182, phase 2), transdermal oestrogen (NCT03875378, phase 2), and two psilocybin studies (NCT06399263 and NCT07169747, phase 2). This is a thin, predominantly repurposing and physiological-endpoint pipeline, not an absent one (live ClinicalTrials.gov v2 query, 2026-09-02).
Themes among currently recruiting studies¶
All records below were retrieved from the live ClinicalTrials.gov v2 API on 2026-09-02 with the status and estimated enrolment shown.
| Theme | Examples (NCT, estimated enrolment) |
|---|---|
| Non-invasive neuromodulation | NCT05788042 enhanced neurostimulation (70); NCT05918835 rTMS and food choice (72); NCT06942858 deep TMS augmenting CBT (110); NCT06286930 cerebellar tDCS (15); NCT07017322 cranial electrical stimulation for mealtime anxiety (30); NCT05554172 non-invasive vagus nerve stimulation (30) |
| Invasive neuromodulation | NCT06112067 combined-target DBS for treatment-refractory mental disorders (18) |
| Delivery and setting | NCT06759402 family-based telemedicine vs inpatient treatment (200); NCT05563649 online guided self-help FBT for adolescents (200); NCT06851273 guided self-help implementation (90) |
| Carers | NCT06624397 new carer programme (50); NCT07734779 online FBT-DBT caregiver intervention (100) |
| Novel psychotherapies | NCT06497101 temperament-based therapy with support (120); NCT05763849 interoceptive exposure for adolescents with low-weight eating disorders (120); NCT06208605 personalizing treatment (320) |
| Metabolic/microbiome | NCT06540703 ketogenic diet and brain response (90); NCT05290285 amino acids vs placebo (92); NCT06043154 microbial dysbiosis (60) |
| Clinical practice questions | NCT06085092 open vs blind weighing in adolescents and young adults (70) |
Two of these are worth flagging for what they test. NCT06759402 puts family-based telemedicine against inpatient treatment — the setting question that DAISIES could not answer in adults, asked in a form patients may actually accept. NCT06085092 randomizes open versus blind weighing, a decision made thousands of times a day in eating-disorder services on no randomized evidence at all.
Bias in what reaches publication¶
The portfolio counts above describe what is registered; a systematic audit describes what survives to the literature. All AN trials listed on ClinicalTrials.gov between 2000 and 2018 were examined for non-completion, non-publication, non-prospective registration and non-replication. Of 201 trials, only 101 were completed; of those, only 41 were published; of the 41 published, only 8 showed evidence of prospective registration, and only 7 had their primary findings replicated (Murray 2020, PMID 31638722). Roughly one registered AN trial in five reaches publication, and roughly one in twenty-five is both prospectively registered and published.
That funnel compounds every limitation described elsewhere in this knowledge base. A network meta-analysis screening 14,003 reports to find 16 trials (Solmi 2021, PMID 33600749) is working downstream of a 20% publication rate, and it cannot recover what the other 80% would have shown. It is also the strongest available argument that "no evidence of benefit" statements in adult AN should be read as statements about a heavily filtered literature.
Portfolio by development stage¶
| Stage | Typical design | Main inferential limit |
|---|---|---|
| Mechanism/observational | Imaging, interoception, microbiome, biomarkers | Association and state confounding |
| Feasibility | Small open intervention | No efficacy estimate |
| Pilot randomized | Small comparator study | Wide intervals and selective population |
| Definitive comparative | Adequate RCT with core outcomes | Rare in adult AN: the only adequately powered drug trial randomized 152 outpatients across five sites, and the largest psychotherapy trial randomized 242 (Attia 2019, PMID 30654643; Zipfel 2014, PMID 24131861) |
| Post-trial follow-up | Relapse, function, harms | Attrition and intervening care |
Recruitment and retention¶
ANTOP screened 727 adults to randomize 242; 22% were lost by end of treatment and 30% by 12-month follow-up, and by the 5-year assessment 154 of 242 (64%) remained (Zipfel 2014, PMID 24131861; Herzog 2022, PMID 35294860). The adult network meta-analysis screened 14,003 reports to find 16 randomized trials, of which 13 contributed 1,047 patients — a yield that describes the size of the entire adult randomized evidence base better than any single trial can (Solmi 2021, PMID 33600749). High burden, ambivalence, narrow eligibility and medical transitions all affect recruitment. Reports should publish screened, eligible, randomized, treated and analyzed denominators.
Two recent trials show what adequately reported attempts look like when they fail to find an effect. A multi-site phase II RCT randomized 61 female inpatients to four weeks of intranasal oxytocin or placebo during nutritional rehabilitation and found no difference on eating-disorder psychopathology or BMI at any timepoint (Maguire 2024, PMID 38520886). A pooled network meta-analysis of all patient-level DBS data available anywhere assembled 36 patients across 11 studies, and its authors' own conclusion was that blinded on/off periods are needed before any target can be recommended (Shaffer 2023, PMID 36724522). Neither is a failure of reporting; both are the field's actual scale.
Outcomes needed¶
| Domain | Minimum reporting |
|---|---|
| Nutrition/weight | Trajectory and developmentally appropriate metric, not endpoint alone |
| Core symptoms | Validated ED psychopathology measure |
| Medical safety | Instability, electrolytes, ECG/adverse events as relevant |
| Function | School/work, relationships and quality of life |
| Durability | Relapse/readmission and follow-up ≥12 months. The rationale is quantitative: pooled recovery across eating disorders rises from 42% before two years to 67% at ten years or longer, so a trial reporting only end-of-treatment status measures the flattest part of the curve (Solmi 2024, PMID 38214616) |
| Acceptability | Completion, reasons for withdrawal, patient experience |
| Co-intervention | Nutrition, psychotherapy, medication and rescue care |
Open questions¶
- Can registry-based follow-up capture rare mortality and suicide outcomes for treatment comparisons? Danish register studies have now produced a severity index validated against cause-specific mortality and an involuntary-treatment mortality estimate, so the data infrastructure exists — but neither compares treatments, and as of September 2026 no target-trial emulation of AN treatment using registry-linked mortality has been published (Larsen 2025, PMID 40670905; Bager 2026, PMID 42383339).
- Which outcome set permits synthesis across adult psychotherapy and biological trials?
- What happens to the 80% of registered AN trials that never publish? Of 201 registered 2000–2018, 101 completed, 41 published, 8 prospectively registered and 7 replicated (Murray 2020, PMID 31638722).
- Why is the drug-development pipeline so thin? Forty-nine phase 2/3 interventional records exist across the register's entire history, most of them physiological-endpoint or repurposing studies, with actual enrolments frequently under 30 (live ClinicalTrials.gov v2 query, 2026-09-02).
- Will the currently recruiting telemedicine-versus-inpatient trial succeed where an adult inpatient-versus-day-patient trial failed to recruit (NCT06759402, retrieved 2026-09-02; İnce 2025, PMID 39943786)?
- Why are small experimental studies terminated, and are reasons published? Four registrations in the table above stopped at 1, 2, 10 and 38 participants, and a formally designed feasibility study missed its target by more than half (Filiz 2026, PMID 42116676); no synthesis of termination reasons in AN trials was located in this audit.
Related pages¶
- Neuromodulation and experimental therapy — early biological interventions.
- Pharmacotherapy — medication evidence.
- Mortality and long-term outcome — endpoints beyond acute weight.
References¶
- Zipfel S, et al. ANTOP randomized controlled trial. Lancet. 2014. PMID 24131861.
- Solmi M, et al. Outcomes in people with eating disorders. World Psychiatry. 2024. PMID 38214616.
- Majić T, et al. Psilocybin for the treatment of anorexia nervosa. Nat Med. 2023;29:1906-1907. Commentary on Peck 2023. PMID 37488290.
- Attia E, et al. Olanzapine versus placebo in adult outpatients with anorexia nervosa. Am J Psychiatry. 2019. PMID 30654643.
- Peck SK, et al. Psilocybin therapy for females with anorexia nervosa: a phase 1, open-label feasibility study. Nat Med. 2023;29:1947-1953. PMID 37488291.
- Filiz ES, et al. Olanzapine for young people with anorexia nervosa: the OPEN open-label feasibility study. Health Technol Assess. 2026;30:1-17. PMID 42116676.
- Maguire S, et al. A phase II randomised controlled trial of intranasal oxytocin in anorexia nervosa. Psychoneuroendocrinology. 2024;164:107032. PMID 38520886.
- Shaffer A, et al. Efficacy of deep brain stimulation for the treatment of anorexia nervosa: a systematic review and network meta-analysis of patient-level data. Neurosurg Focus. 2023;54:E5. PMID 36724522.
- Solmi M, et al. Comparative efficacy and acceptability of psychological interventions for adult outpatients with anorexia nervosa. Lancet Psychiatry. 2021. PMID 33600749.
- Herzog W, et al. ANTOP 5-year follow-up. Lancet Psychiatry. 2022. PMID 35294860.
- Golden NH, et al. Higher-calorie refeeding in anorexia nervosa: 1-year outcomes. Pediatrics. 2021;147:e2020037135. PMID 33753542.
- Larsen JT, et al. Register-based severity index for anorexia nervosa in Denmark. Eur Psychiatry. 2025;68:e104. PMID 40670905.
- Bager L, et al. Excess mortality among patients with anorexia nervosa treated involuntarily. Int J Eat Disord. 2026. PMID 42383339.
- Murray SB, et al. Registration, reporting, and replication in clinical trials: the case of anorexia nervosa. Int J Eat Disord. 2020;53:138-142. PMID 31638722.
- İnce B, et al. Stepping into day treatment approach versus inpatient treatment for adults with anorexia nervosa: the DAISIES RCT. Health Technol Assess. 2025;29:1-37. PMID 39943786.
- Lock J, et al. Randomized clinical trial comparing family-based treatment with adolescent-focused individual therapy. Arch Gen Psychiatry. 2010;67:1025-1032. PMID 20921118. (NCT00149786)
- Le Grange D, et al. Relapse from remission at two- to four-year follow-up in two treatments for adolescent anorexia nervosa. J Am Acad Child Adolesc Psychiatry. 2014;53:1162-1167. PMID 25440306.
- Garber AK, et al. Short-term outcomes of the Study of Refeeding to Optimize Inpatient Gains. JAMA Pediatr. 2021;175:19-27. PMID 33074282. (NCT02488109)
- Lipsman N, et al. Subcallosal cingulate deep brain stimulation for treatment-refractory anorexia nervosa: a phase 1 pilot trial. Lancet. 2013;381:1361-1370. PMID 23473846. (NCT01476540)
- Lipsman N, et al. Deep brain stimulation of the subcallosal cingulate for treatment-refractory anorexia nervosa: 1-year follow-up of an open-label trial. Lancet Psychiatry. 2017;4:285-294. PMID 28238701.