Skip to content

Not yet independently audited

This page was written from verified sources but has not yet passed the independent citation audit. Treat its claims as provisional.

Clinical trials landscape

TL;DR — The active bipolar-disorder pipeline spans new mechanisms for mania and depression, psychedelic feasibility studies, lithium comparisons, device-based treatment, metabolic interventions and biomarker validation. As of live ClinicalTrials.gov v2 API retrievals on 2026-08-31 and 2026-09-03, with all 28 records cited on this page re-fetched and re-verified on 2026-09-03, phase 3 programs included xanomeline/trospium for mania (acute plus open-label extension), lumateperone for pediatric bipolar depression, azetukalner for bipolar depression and a small brexpiprazole study, while phase 2 programs included adjunctive brenipatide for time-to-relapse, ALTO-100, ABX-002 and tightly controlled psilocybin studies (NCT06951711; NCT07140913; NCT06929273; NCT06372964; NCT07217860; NCT04569448; NCT07286175; NCT06656416; NCT06869187; NCT06706232). The frontier is not only pharmacologic: active studies test accelerated theta-burst stimulation, magnetic seizure therapy, open- and closed-loop deep-brain stimulation, vocal prediction and composite blood-marker algorithms (NCT05849402; NCT04080778; NCT06599099; NCT07127913; NCT07298278; NCT07031817). Registry status is a snapshot, not evidence of efficacy; completion, posted results and peer-reviewed outcome reporting must be tracked separately. The 2026-10-16 ClinicalTrials.gov re-fetch resolved all 38 NCT identifiers then counted condition-wide; none has posted results. That count included three log-only identifiers that were screened and not added. The 2026-10-23 re-fetch of the 35 content-cited records again found no posted results, and the page now also cites PolarUs (NCT07890805), a quality-of-life app trial, and B-HAPPI (NCT07664852), a phase 3 pramipexole trial in anhedonic bipolar depression. The azetukalner extension is active, not recruiting, and ALTO-100 has a refreshed completion estimate. The ISRCTN search additionally identifies a 206-person randomized ketogenic-diet trial and a 303-person maintenance comparison that were missing from this portfolio. Two trials have since reported in journals without registry results: bipolar-specific serial ketamine (NCT05004896) and psilocybin for bipolar II depression (NCT05065294).

Methods and snapshot boundary

The registry sweep used the ClinicalTrials.gov API v2 condition query for “Bipolar Disorder,” restricted to recruiting, not-yet-recruiting, active-not-recruiting and enrolling-by-invitation records, followed by direct /api/v2/studies/{NCT} retrieval for every ID listed below. Access date: 2026-08-31. Direct re-fetch of the 20 previously cited identifiers found no overall-status change and no posted results. Newly retained records were limited to programs that change the described pipeline (KarXT mania extension, pediatric lumateperone depression, adjunctive brenipatide relapse prevention). Incidental uses of the word “bipolar” in surgery and electrophysiology were excluded by reading title, condition and intervention. Status and enrollment are registry-reported and can change.

API source: https://clinicaltrials.gov/api/v2/studies?query.cond=Bipolar%20Disorder&filter.overallStatus=RECRUITING%7CNOT_YET_RECRUITING%7CACTIVE_NOT_RECRUITING%7CENROLLING_BY_INVITATION&pageSize=50, accessed 2026-08-31.

2026-10-09 re-fetch. All 29 identifiers previously cited by this condition were re-fetched through /api/v2/studies/{NCT}. No overall status changed and no record has posted results. Four status-verification dates advanced: NCT06951711, NCT06929273 and NCT06372964 to September 2026, and NCT07286175 to October 2026. Three records are now past their registered primary-completion date without a status update. NCT05849402 (aiTBS) lists 2026-08-29 and NCT06506019 (psilocybin neurobiology) lists 2026-09-01; both remain active, not recruiting. NCT06656416 (ALTO-100) lists August 2026 but still shows recruiting, last verified July 2025, so its registry status is stale rather than informative. A separate query for bipolar-disorder records first posted on or after 2026-08-28 returned 12 records and no records with first-posted results. Non-psychiatric "bipolar" device studies were excluded, and the bipolar-relevant registrations are listed in the tables below.

2026-10-16 re-fetch. All 38 NCT identifiers cited anywhere in this condition resolved via the live v2 API, with no posted results. X-CEED-OLE (NCT07217860) now lists active, not recruiting, replacing this page’s enrolling-by-invitation snapshot; its record was updated on 13 October while its status-verification month remains June. ALTO-100 (NCT06656416) remains recruiting but is now verified October 2026, with estimated primary completion September 2027 and study completion October 2027; the earlier August 2026 estimate is no longer current. The bright-light record (NCT07860957) was updated on 13 October and remains completed without posted results. The date-filtered condition queries returned 74 updated records, 13 newly posted records and zero records with results first posted between 28 August and 16 October; these are raw counts before exclusion of non-psychiatric uses of “bipolar.”

2026-10-23 re-fetch. The 16 October count of 38 mixed page citations with three identifiers that, before this paragraph, appeared only in the curation log because they were screened and not added (NCT06229210, NCT06696755, NCT07789340). They are named here so the count can be audited; they are still not portfolio rows. This sweep re-fetched the 35 NCT records cited outside that log and WHATS-NEW.md. None of the 35 has a results section. Overall status was unchanged for the records already on this page. BALSAM-2 (NCT06951711) is still recruiting, now verified October 2026, last updated 2026-10-20, with estimated primary completion 2026-11-02 and enrollment 274. ALTO-100, the azetukalner extension, bright light (NCT07860957, completed, verified October 2026, no results), inhaled loxapine (NCT02525991, still no results) and the three tracked ketogenic records were re-fetched and still match the sentences above. Two records are added below. Condition searches returned 8 records updated since 17 October, 4 first posted since 17 October, 19 first posted since 28 August, and 0 with results first posted in the window. Those counts precede removal of non-psychiatric uses of “bipolar.”

Pharmacologic pipeline

Trial Phase; enrollment Status (re-fetched 2026-10-16) Intervention Scientific question
Brexpiprazole Treatment for Bipolar I Depression (NCT04569448) Phase 3; 58 Recruiting; verified Aug 2026 Brexpiprazole Antidepressant efficacy and tolerability in bipolar I depression
BALSAM-2 mania study (NCT06951711) Phase 3; 274 Recruiting; verified Oct 2026; updated 2026-10-20; primary completion estimated 2026-11-02 KarXT vs placebo Muscarinic mechanism for manic episodes
B-HAPPI anhedonic bipolar depression (NCT07664852) Phase 3; 126 Recruiting; verified Oct 2026; updated 2026-10-22; primary completion estimated 2029-12-31. First posted 2026-06-24, before this window High-dose adjunctive pramipexole vs placebo; quadruple-masked Change in Snaith-Hamilton Pleasure Scale total score at 6 weeks. Ages 18–80; bipolar I or II
Adjunctive KarXT for mania with/without mixed features (NCT07140913) Phase 3; 424 Recruiting; verified Jun 2026 Xanomeline/trospium vs placebo added to lithium, valproate or lamotrigine Adjunctive antimanic efficacy
BALSAM-3 mania open-label extension (NCT06929273) Phase 3; 450 Recruiting; verified Sep 2026 KarXT; lithium, valproate or lamotrigine permitted Long-term safety after mania/mixed-mania treatment
Lumateperone for pediatric bipolar depression (NCT06372964) Phase 3; 384 Recruiting; verified Sep 2026 Lumateperone vs placebo Efficacy and safety of a major-depressive-episode treatment in ages 10–17
X-CEED open-label extension (NCT07217860) Phase 3; 400 Active, not recruiting; verified Jun 2026; updated 13 Oct 2026 Azetukalner Longer-term exposure after bipolar-depression trials
Brenipatide adjunctive relapse prevention (NCT07286175) Phase 2; 400 Recruiting; verified Oct 2026 Brenipatide vs placebo added to standard of care Time to relapse over ≥6 months of adjunctive treatment
ALTO-100 in bipolar depression (NCT06656416) Phase 2; 200 Recruiting; verified Oct 2026; estimated primary completion Sep 2027 ALTO-100 vs placebo Novel adjunctive antidepressant mechanism
ABX-002 adjunctive bipolar depression (NCT06869187) Phase 2; 35 Active, not recruiting; verified May 2026 ABX-002 Early efficacy/safety signal
Lithium vs lamotrigine in bipolar II (NCT06184581) Phase 4; 200 Recruiting; verified Mar 2026 Lithium carbonate vs lamotrigine Pragmatic comparative treatment in bipolar II
Psilocybin in bipolar II depression and suicidality (NCT06706232) Phase 2; 10 Recruiting; verified Jul 2026 Two psilocybin doses plus therapeutic support Feasibility and safety in a population commonly excluded from psychedelic trials
Psilocybin neurobiology in treatment-resistant bipolar depression (NCT06506019) Phase 1; 30 Active, not recruiting; verified Aug 2026 Psilocybin plus functional MRI Mechanism and early safety

The new-mechanism pipeline should be read against a high attrition baseline. Network meta-analyses can establish average efficacy for completed interventions, but they cannot validate registry optimism or substitute for posted results (Kishi 2021, PMID 33177610). Mixed-feature and rapid-cycling eligibility also matters because broad diagnostic labels can conceal populations with different benefit-risk profiles (Roosen 2022, PMID 35545157).

Neuromodulation and device trials

Trial Enrollment Status (re-fetched 2026-10-09) Intervention / comparator Key design issue
Accelerated iTBS for treatment-resistant bipolar II depression (NCT05849402) 60 Active, not recruiting Active vs sham aiTBS Blinding integrity and durability
Magnetic seizure therapy in bipolar and treatment-resistant depression (NCT04080778) 60 Recruiting MST vs ECT Noninferiority, cognitive burden and diagnosis-specific effects
Deep-brain stimulation for treatment-resistant bipolar depression (NCT06599099) 10 Recruiting Medtronic Percept RC; blinded discontinuation Invasiveness, tiny sample and within-person signal
Closed-loop DBS for treatment-resistant bipolar depression (NCT07127913) 10 Recruiting Stimulation-on/off/active control Personalized control signal and safety
FEAST versus right-unilateral and bitemporal ECT (NCT07857122) 108 Not yet recruiting; first posted 2026-10-05 Focal electrically administered seizure therapy vs two ECT montages; triple-masked Mixed unipolar and bipolar I depressive episodes; primary outcome is day-5 quality of life (Q-LES-Q-SF), not depression severity
tACS plus cognitive training in veterans (NCT07845409) 138 Not yet recruiting; first posted 2026-09-28 Active vs sham transcranial alternating current stimulation plus cognitive training, and a no-intervention arm Primary outcomes are cognitive-control tasks; suicide-related outcomes are listed among the conditions

These studies occupy different development stages. Sham-controlled noninvasive stimulation can estimate short-term symptomatic effects, whereas implanted-device studies at n=10 primarily establish feasibility and individualized physiology. Device novelty should not be confused with comparative effectiveness.

Biomarker and prediction trials

Trial Enrollment Status Measurement Intended context of use
Vocal biomarkers for impending depressive/manic episodes (NCT07298278) 170 Not yet recruiting; verified Dec 2025 Voice interviews, under-mattress sleep sensor, smartwatch activity/sleep/temperature Prospective episode anticipation
Composite blood-marker algorithm plus MDQ (NCT07031817) 623 Recruiting; verified Oct 2025 Venous blood panel plus standardized psychiatric assessment Diagnostic support
Prediction on recurrence of manic and depressive episodes (NCT05828056) 100 Recruiting; directly verified 2026-08-30 Prospective clinical and digital measures Recurrence prediction
mHealth algorithms signaling upcoming recurrence (NCT06204705) 200 Recruiting; directly verified 2026-08-30 Mobile-health estimates and alerts Prospective episode warning

The vocal study directly addresses a major gap in retrospective digital-phenotyping work: prospectively defined episodes and a time-to-warning endpoint. Existing reviews show substantial methodological heterogeneity and overfitting, so external validation and false-alert burden will be more decision-relevant than a high internal AUC (Saccaro 2021, PMID 34488086; Ortiz 2021, PMID 34706433).

A separate app trial is not a prediction study. PolarUs (NCT07890805) was first posted on 2026-10-22. It is randomized, not yet recruiting, phase not applicable, with anticipated enrollment 504, start November 2026 and primary completion December 2028, verified October 2026. The primary outcome is the Brief Quality of Life in Bipolar Disorder scale at enrolment and every 4 weeks across 12 weeks. Eligibility is a DSM-5 diagnosis of bipolar I, bipolar II, cyclothymic disorder, other specified bipolar disorder or unspecified bipolar disorder, age 18 or older, no prior PolarUs use, and a smartphone; residence is Canada or the United States. The record’s masking field is double-blind, limited to the investigator and the outcomes assessor, and the protocol states that participants cannot be fully blinded. A quality-of-life endpoint does not answer whether an app prevents syndromal relapse.

The composite blood-marker study is larger than many discovery cohorts, but size alone does not establish clinical utility. It must show incremental discrimination beyond the MDQ and structured clinical assessment, report calibration and predefine how discordant results should be used (Hu 2023, PMID 36347076).

Metabolic, circadian and lifestyle interventions

Trial Enrollment Status Arms Principal outcome domain
Ketogenic-diet mechanism study (NCT06081426) 107 Recruiting; verified Dec 2025 Ketogenic, non-ketogenic and no-diet arms Symptoms, metabolism and neurobiology
Healthy Lifestyles for Bipolar Disorder (NCT06188754) 300 Recruiting; verified Jul 2026 Time-restricted eating vs Mediterranean diet Feasibility, metabolic and mood outcomes

Diet studies face predictable adherence, blinding and co-intervention problems. They are important because cardiovascular mortality is elevated in bipolar disorder, but mechanistic or weight changes must not be described as mood efficacy unless the prespecified mood endpoint supports it (Biazus 2023, PMID 37491460; Paljärvi 2024, PMID 38826056).

Non-US registry additions

A direct ISRCTN XML API search for condition:bipolar with lastEdited between 28 August and 16 October 2026 returned nine records. Three change the portfolio described here. Registry DOIs were resolved through Crossref; access date: 2026-10-16.

Trial Registry design and sample Current snapshot Endpoint and boundary
ENERGISE-BD (ISRCTN14945909; doi:10.1186/isrctn14945909) UK; randomized, registry-labelled single-blind, ketogenic diet versus NHS Eatwell guidance; target 206 Recruiting; registered 3 Aug 2026; anticipated first enrolment 14 Oct 2026 Primary: ≥6-point PHQ-9 improvement at 12 weeks; follow-up at 24 weeks. A depression-response trial, not only a metabolic mechanism study; no efficacy result is reported
COMBINER (ISRCTN17178652; doi:10.1186/isrctn17178652) UK; randomized single-blind lithium plus quetiapine versus either monotherapy; target 303 Recruiting; estimated completion 31 Jul 2029 Primary: time to first new mood episode over 24 months; polarity-specific recurrence is secondary. Not restricted to first-episode illness
CRiB2 (ISRCTN10362331; doi:10.1186/isrctn10362331) UK; assessor-blinded cognitive remediation plus usual care versus usual care; actual enrolment 222 (target 250) No longer recruiting; recorded completion 31 Jan 2026 Primary: FAST everyday functioning at 25 weeks. The retrieved record links a protocol; outcome reporting remains a follow-up task

Sources: ENERGISE-BD, COMBINER, CRiB2, retrieved through the registry’s /api/query/format/who and /api/trial/ISRCTN10362331/format/who endpoints. The WHO ICTRP portal was also queried, but its small, COVID-heavy result set did not support a comprehensive portfolio comparison. CENTRAL could not be searched (Cochrane Library HTTP 403).

The same ISRCTN query run on 2026-10-23, with lastEdited through 23 October, returned 10 records. The three trials above were re-fetched and still match this table: ENERGISE-BD recruiting, target 206, enrolment still anticipated and results enrolment 0, and still labelled single-blind without a statement of who is masked; COMBINER recruiting, target 303, completion 31 July 2029; CRiB2 no longer recruiting, actual enrolment 222, completion 31 January 2026. The new identifier, ISRCTN15962574, is an online parenting trial (target 284, actual enrolment 303, no longer recruiting, completion 31 January 2027) whose primary outcome is the child’s Strengths and Difficulties Questionnaire at 24 weeks. It was screened and is not part of the mood-efficacy pipeline.

Suicide-focused psychosocial trial

The registry lists a 130-participant recruiting trial of four psychosocial approaches for suicidal behavior in bipolar disorder: interpersonal and social rhythm therapy, bipolar-specific CBT, mindfulness-based stress reduction, and psychoeducation/medication understanding (NCT02604277; status verified January 2026). Because the arms are active psychosocial strategies, interpretation should distinguish comparative benefit from absolute change over time.

Suicide outcomes require adequate event counts and careful composite definitions. A prior 519-participant randomized trial of lithium augmentation after a recent suicide-related event stopped for futility and found no reduction in the composite outcome (HR 1.10, 95% CI 0.77-1.55), illustrating why observational protection signals cannot predetermine trial results (Katz 2022, PMID 34787653).

Completed evidence that shapes trial design

Design lesson Quantitative example Implication
Rare events produce imprecision Lithium meta-analysis: 48 RCTs, 6,674 participants; suicide OR 0.13, 95% CI 0.03-0.66 vs placebo (Cipriani 2013, PMID 23814104) Event definitions and follow-up length dominate power
Composite outcomes can dilute a specific effect VA trial: HR 1.10, 95% CI 0.77-1.55 for repeat suicide-related events (Katz 2022, PMID 34787653) Report each component as well as the composite
Digital models need participant-level separation Mood forecast: 15,975 observations from only 84 people, next-day R² 0.51 (Busk 2020, PMID 32234702) Observation count is not participant count
Diagnostic classification remains difficult Smartphone study AUC 0.62 for bipolar vs major depression (Langholm 2023, PMID 38129571) Compare against a strong clinical baseline
Actigraphy effects are heterogeneous Mood-disorder meta-analysis: 38 studies, n=3,758 (Tazawa 2019, PMID 31060012) Harmonize devices and state definitions
Safety can be long-latency Lithium toxicity synthesis screened 5,988 abstracts and included 385 studies (McKnight 2012, PMID 22265699) Trials need registry or cohort follow-up for renal/endocrine outcomes

How to read registry records

Field What it means What it does not mean
Recruiting Site reports current recruitment Participants are necessarily enrolling at every listed site
Active, not recruiting Study continues without open recruitment Primary analysis is complete
Enrollment Planned or actual count, as labeled in the record An analyzable final sample
Phase Regulatory development label Methodological quality
Primary outcome Sponsor-declared endpoint Result or clinical importance
Last update Registry record activity Independent verification

Portfolio gaps

  • Acute mania has major drug development, but fewer trials test service-level approaches to sleep restoration, financial/sexual risk and family-supported crisis planning.
  • Bipolar II depression remains underrepresented relative to bipolar I despite dedicated lithium-versus-lamotrigine and psilocybin feasibility trials (NCT06184581; NCT06706232), and the first bipolar II-only psilocybin report is open-label with n=14 (Downey 2026, PMID 42765891; NCT05065294). Pediatric bipolar depression now has a 384-participant phase 3 lumateperone record (NCT06372964); that does not fill the adult bipolar II gap.
  • The active-record sweep, re-run on 2026-09-03 and checked again on 2026-10-23, found prospective prediction and alert studies (NCT05828056; NCT06204705; NCT07298278), but no retrieved protocol clearly randomized an alert-plus-response pathway with hospitalization prevention as the clinical-utility endpoint. The nearest record remains a transitional-care study whose primary outcomes are feasibility and acceptability (NCT07213492; still recruiting; verified January 2026; last update 2026-10-06; primary completion 2027-05-01; n=10) alongside a probiotic relapse trial (NCT03349528). PolarUs (NCT07890805) randomizes an app, with quality of life rather than hospitalization as the primary outcome. This is an evidence gap in the retrieved portfolio, not a claim that no such protocol can exist outside the registries queried.
  • Invasive neuromodulation samples are necessarily small; transparent stopping, adverse-event and explant criteria are critical (NCT06599099; NCT07127913).
  • Diet trials create an opportunity to link metabolic and mood outcomes, but causal interpretation requires measured adherence and prespecified mediation (NCT06081426; NCT06188754).
  • Patient-prioritized outcomes such as cognition, sexual function, work and treatment burden are not visible in most brief registry titles and require full-outcome review.

Maturity matrix for this snapshot

Program Current maturity Evidence needed next
Xanomeline/trospium for mania Two acute phase 3 records plus an open-label extension Completed primary outcomes, adverse-event discontinuation and controlled maintenance data
Azetukalner for bipolar depression Phase 3 open-label extension Parent-trial placebo-controlled efficacy plus switch and long-term safety
Brexpiprazole for bipolar I depression Small phase 3 registry record Precision around efficacy and tolerability in a larger representative sample
Brenipatide adjunctive relapse prevention Phase 2, n=400, time-to-relapse primary Prespecified relapse definition, adjunctive standard-of-care composition and adverse-event profile
Lumateperone pediatric bipolar depression Phase 3, ages 10–17 Developmental safety, switch surveillance and whether adult bipolar-depression effects transport
ALTO-100 Phase 2 Replication and validity of any enrichment strategy
ABX-002 Phase 2, active not recruiting Posted results and peer-reviewed report
Psilocybin Phase 1/2, n=10-30; one bipolar II open-label report (n=14) Acute mania/hypomania surveillance, durability and controlled efficacy
aiTBS / MST Sham or active-comparator device trials Blinding, cognitive outcomes and longer follow-up
DBS Feasibility samples of 10 Reproducible target/biomarker logic and multi-site safety
Vocal prediction Prospective multimodal cohort Locked alert threshold and external validation
Blood-marker algorithm Diagnostic-validation study Calibration, incremental value and clinical-utility trial
Diet / timed eating Behavioral-mechanism trials Adherence measurement, metabolic mediation and mood specificity

Trial-tracking fields to preserve

For every future sweep, retain the prior status, status-verification date, enrollment type (anticipated or actual), primary completion date, results-posting state, publication link and reason for any termination. A status transition is evidence about trial conduct, not treatment effect. Direct record endpoints used here follow the reproducible form https://clinicaltrials.gov/api/v2/studies/NCT######## and were fetched on 2026-08-31.

Small signals at the experimental edge

A 60-person bipolar trial randomized omega-3 fatty acids 2 g/day or placebo for two months and reported lower depression scores, TNF-α, IL-6 and hsCRP in the active arm (all P<.001), but the abstract does not provide between-group effect sizes or confidence intervals (Eslahi 2023, PMID 37213459). It is hypothesis-supporting, not sufficient for treatment ranking.

Ketogenic-diet evidence remains earlier. A 2024 review found human psychiatric studies generally small, short, often uncontrolled, heterogeneous and affected by high dropout or unmeasured adherence, despite plausible metabolic and anti-inflammatory mechanisms (Chrysafi 2024, PMID 38892480). A ClinicalTrials.gov v2 query on 2026-09-03 shows that registered bipolar-specific ketogenic trials now exist — a randomized open-label first-episode study (NCT06221852; n=50; recruiting; primary outcomes are brain redox/creatine-kinase metabolites, insulin resistance and symptom change), a single-arm bipolar-depression study (NCT07121894; n=30; recruiting) and an adolescent study (NCT06920940; n=80; recruiting) alongside the mechanistic study already tracked here (NCT06081426). The 2026-10-16 ISRCTN search broadens that ClinicalTrials.gov-only snapshot: ENERGISE-BD is registered as a single-blind randomized trial with a depression-response endpoint (ISRCTN14945909; target 206; see table above). It remains recruiting, so it adds a controlled efficacy design rather than a completed efficacy result. The earlier statement that none of the retrieved ketogenic trials was blinded should not be generalized beyond the ClinicalTrials.gov records.

Ketamine literature demonstrates denominator leakage. A real-world treatment-resistant-depression synthesis pooled 79 studies (n=2,665) and estimated response 45%±10% and remission 30%±5.9%, but it did not provide a bipolar-specific pooled estimate (Alnefeesi 2022, PMID 35688035). The bipolar-specific Ket-BD trial, which used a midazolam active placebo, now partly closes that gap for short-term efficacy (Orsini 2026, PMID 42684694). Likewise, a 24-trial comparison of racemic ketamine and esketamine combined unipolar and bipolar depression; racemic ketamine showed larger pooled response (RR 3.01 versus 1.38) and remission (RR 3.70 versus 1.47), but formulation and route were confounded with study populations (Bahji 2021, PMID 33022440).

Trials that reported or surfaced since the last sweep

Trial Registry state at 2026-10-09 What was reported
Ket-BD, adjunctive IV ketamine vs midazolam, treatment-resistant bipolar I/II depression (NCT05004896) Completed; primary completion 2025-11-14; registry enrollment 71; no posted results n=68 randomized; day-14 MADRS −7.3 vs midazolam (95% CI −12.0 to −2.5); no mania, hypomania, psychosis or suicide attempts (Orsini 2026, PMID 42684694). See bipolar depression.
Psilocybin therapy for bipolar II depression (NCT05065294) Active, not recruiting; last verified Sep 2024; primary completion 2024-08-01; no posted results Open-label, n=14. MADRS −12.7 after 10 mg, with four participants (28.5%) remitting; nine went on to 25 mg, after which MADRS fell −18.6 at day 21. Three notable psychiatric events (suicidal ideation or hypomania) resolved with support (Downey 2026, PMID 42765891)
Midday bright-light therapy for bipolar depression (NCT07860957) Completed; start 2021-04-01, completion 2026-06-11; first posted 2026-10-06, i.e. registered after completion; n=37; no posted results Randomized, double-masked; primary outcome is week-4 SIGH-SAD remission. No peer-reviewed report was found in PubMed in this sweep

The psilocybin registry record has not been updated since September 2024 even though the trial is now published. Its status field therefore says nothing about the trial's real conduct. The bright-light record was registered only after completion, so its prespecified outcome cannot be checked against a prospective registration.

One small psilocybin-assisted psychotherapy trial admitted major depression and bipolar II depression: 30 participants were randomized to immediate or delayed treatment, 29 reached week 2, and the immediate arm showed a MADRS effect size g=1.07 (P<.01) without serious adverse events (Rosenblat 2024, PMID 38359838). The mixed diagnosis, wait-list design and tiny sample leave polarity-switch and durability risks unresolved. A bipolar II-only open-label dose-escalation study has since reported large within-group MADRS improvement in 14 participants, but three had suicidal ideation or hypomania that needed support. Without a control arm, the design cannot separate drug effect from expectancy or the psychotherapy delivered around dosing (Downey 2026, PMID 42765891).

Open questions

  • Will muscarinic treatment improve mania without an offsetting cholinergic tolerability burden in phase 3, and is longer-term exposure tolerable in the open-label extension (NCT06951711; NCT07140913; NCT06929273)?
  • Can adjunctive brenipatide delay syndromal relapse on standard of care, and with what adverse-event profile (NCT07286175)?
  • Can azetukalner or ALTO-100 produce durable bipolar-depression benefit with acceptable switch risk (NCT07217860; NCT06656416)?
  • Does lumateperone separate from placebo on pediatric bipolar-depression ratings without a developmental safety signal that adult trials cannot detect (NCT06372964)?
  • Can vocal and wearable measures provide calibrated, actionable warning rather than retrospective separation (NCT07298278; Saccaro 2021, PMID 34488086)?
  • Which treatment-resistant bipolar-depression phenotype justifies invasive stimulation, and how should n-of-1 signals be combined across participants (NCT06599099; NCT07127913)?
  • Can metabolic interventions improve both cardiometabolic risk and mood without conflating adherence effects with diet mechanism (NCT06081426; NCT06188754)?
  • How often do completed bipolar trials post timely registry results and publish all prespecified outcomes?

References

  1. Kishi T, et al. Mood stabilizers and/or antipsychotics for bipolar disorder in the maintenance phase: systematic review and network meta-analysis. Mol Psychiatry. 2021;26:4146-4157. PMID 33177610
  2. Roosen L, et al. Evidence-based treatment strategies for rapid cycling bipolar disorder: a systematic review. J Affect Disord. 2022;311:69-77. PMID 35545157
  3. Saccaro LF, et al. Portable technologies for digital phenotyping of bipolar disorder: A systematic review. J Affect Disord. 2021;295:323-338. PMID 34488086
  4. Ortiz A, et al. Apps and gaps in bipolar disorder: electronic monitoring for episode prediction. J Affect Disord. 2021;295:1190-1200. PMID 34706433
  5. Hu X, et al. Biomarkers and detection methods of bipolar disorder. Biosens Bioelectron. 2023;220:114842. PMID 36347076
  6. Biazus TB, et al. All-cause and cause-specific mortality among people with bipolar disorder: a systematic review and meta-analysis. Mol Psychiatry. 2023;28:2508-2524. PMID 37491460
  7. Paljärvi T, et al. Cardiovascular mortality in bipolar disorder: Population-based cohort study. Acta Psychiatr Scand. 2024;150:56-64. PMID 38826056
  8. Katz IR, et al. Lithium Treatment in the Prevention of Repeat Suicide-Related Outcomes in Veterans With Major Depression or Bipolar Disorder. JAMA Psychiatry. 2022;79:24-32. PMID 34787653
  9. Cipriani A, et al. Lithium in the prevention of suicide in mood disorders: updated systematic review and meta-analysis. BMJ. 2013;346:f3646. PMID 23814104
  10. Busk J, et al. Forecasting Mood in Bipolar Disorder From Smartphone Self-assessments. JMIR Mhealth Uhealth. 2020;8:e15028. PMID 32234702
  11. Langholm C, et al. Classifying and clustering mood disorder patients using smartphone data. NPJ Digit Med. 2023;6:238. PMID 38129571
  12. Tazawa Y, et al. Actigraphy for evaluation of mood disorders: A systematic review and meta-analysis. J Affect Disord. 2019;253:257-269. PMID 31060012
  13. McKnight RF, et al. Lithium toxicity profile: a systematic review and meta-analysis. Lancet. 2012;379:721-728. PMID 22265699
  14. Eslahi H, et al. Effects of omega-3 fatty acids on inflammatory cytokines and depression status in bipolar disorder: a randomized double-blind controlled trial. J Res Med Sci. 2023;28:36. PMID 37213459
  15. Chrysafi M, et al. Potential effects of the ketogenic diet in prevention and co-treatment of psychiatric disorders. Nutrients. 2024;16:1546. PMID 38892480
  16. Alnefeesi Y, et al. Real-world effectiveness of ketamine in treatment-resistant depression: a systematic review and meta-analysis. J Psychiatr Res. 2022;151:693–709. PMID 35688035
  17. Bahji A, et al. Comparative efficacy of racemic ketamine and esketamine for depression: a systematic review and meta-analysis. J Affect Disord. 2021;278:542–555. PMID 33022440
  18. Rosenblat JD, et al. Psilocybin-assisted psychotherapy for treatment-resistant depression: a randomized clinical trial evaluating repeated doses. Med (N Y). 2024;5:190–200.e5. PMID 38359838
  19. Orsini DK, et al. Serial ketamine infusions for treatment-resistant bipolar depression: a randomized clinical trial. JAMA Psychiatry. 2026:e262658. PMID 42684694
  20. Downey AE, et al. An open-label, dose-escalation trial of psilocybin-assisted therapy for bipolar II depression. J Clin Psychiatry. 2026;87(4):25m16227. PMID 42765891