Dihexa for Bipolar Disorder: Lithium Neuroprotection, BDNF/HGF, Caplyta Lumateperone, Cariprazine & the 2026 UK Review

Around 1% of UK adults — approximately 660,000 people — live with bipolar disorder, making it the fourth most common mental health condition in the UK on Mental Health Foundation and NHS England Adult Psychiatric Morbidity Survey 2023-24 denominators. Bipolar disorder is the prototypical recurrent affective illness: it almost always begins in the late teens and early twenties, it carries a suicide risk 10-30 times that of the general population, and the most recent NCISH National Confidential Inquiry data show 1,491 patient suicides with a primary bipolar diagnosis — 8% of all patient suicides — with the 2019-2022 average 19% higher than the 2015-2018 average. 2025-2026 has been a consequential window in bipolar drug development: the FDA approved Intra-Cellular Therapies / Johnson & Johnson’s Caplyta (lumateperone) in November 2025 as adjunctive therapy for major depressive disorder, building on its existing US indication for bipolar I and II depression as monotherapy and adjunct with lithium or valproate; the MHRA reauthorised semi-sodium valproate in September 2025 for the treatment of mania when lithium is not tolerated or contraindicated; and cariprazine (Vraylar/Reagila) continues to expand its bipolar evidence base through paediatric Phase 4 trials and a D3-imaging mechanistic Phase 4 programme. At the same time, lithium — the 70-year-old gold standard — has accumulated a substantial neuroprotective and disease-modifying evidence base: a 2012 meta-analysis by Hajek and colleagues showed that smaller hippocampal volumes in bipolar disorder are masked by lithium exposure, the 2016 Kakhki voxel-based morphometry study showed bilateral hippocampal volume increases in elderly bipolar patients on long-term lithium, and a 2025 translational mechanisms review consolidated the GSK-3β-inhibition / BDNF-upregulation / tau-dephosphorylation / amyloid-β42-reduction case. Add the often-overlooked Russo et al. 2010 finding that serum hepatocyte growth factor (HGF) is decreased in individuals with bipolar disorder and normalises with zinc and antioxidant therapy, and the question that brings traffic to this page becomes legible: where, mechanistically, does a synaptogenic HGF/c-Met peptide like Dihexa sit in a disorder where lithium already raises BDNF, increases hippocampal volume, and is the closest thing psychiatry has to a neuroprotective gold standard? A rigorous 2026 UK evidence review.

Bipolar Disorder in the UK in 2026: Scale, Demographics & the Treatment Gap

The Mental Health Foundation places bipolar disorder as the fourth most common UK mental health condition, affecting approximately 1% of UK adults. The Adult Psychiatric Morbidity Survey 2023-24 from NHS England Digital provides the most recent denominators on English prevalence and treatment access. The diagnosed UK population is somewhere around 660,000 adults, with substantial under-diagnosis: median delay from first symptoms to formal bipolar diagnosis remains in the order of 6-10 years across UK studies, with mis-diagnosis as unipolar major depressive disorder being the dominant pattern (the bipolar II phenotype is particularly liable to be missed because hypomanic episodes are subjectively pleasant and rarely lead to help-seeking).

Onset is sharply concentrated in the late teens and early twenties, with bipolar I tending to present slightly earlier than bipolar II. There is no strong sex difference in prevalence, although rapid cycling and bipolar II are over-represented in women. Comorbidity is the rule rather than the exception: anxiety disorders, alcohol use disorder, substance use, ADHD, eating disorders and cardiometabolic disease accumulate across the life course, and the 15-20 year mortality gap that is documented in schizophrenia is also present in bipolar disorder, driven roughly half by suicide and half by cardiovascular disease.

The treatment gap is dimensional rather than absolute. NICE CG185 maintenance lithium produces decisive episode-prevention benefit at the population level, but real-world adherence is poor — the long-running Hayes et al. UK retrospective cohort of 2,649 people with bipolar disorder followed for 5 years documented high relapse rates and the practical difficulty of sustained maintenance. Bipolar depression, which dominates the long-run symptom burden in both bipolar I and bipolar II, has only four NICE-supported pharmacological options (quetiapine, olanzapine-fluoxetine combination, lurasidone and lamotrigine), and the bipolar-specific cognitive impairment that persists between episodes — verbal memory, executive function, attention and processing speed deficits — has no licensed pharmacological treatment at all in the UK in 2026. That dimensional gap drives search interest in unlicensed neurotrophic and synaptogenic agents — Dihexa, Semax, Selank, NSI-189, cerebrolysin, sarcosine, N-acetylcysteine, BDNF-axis modulators and adjacent compounds — from patients who have stabilised on lithium but want to recover cognitive sharpness or lift residual depressive symptoms.

The Neurobiology of Bipolar Disorder: BDNF, Synaptic Plasticity, GSK-3β & the Mitochondrial Story

Bipolar disorder is not one biology. The clinically useful frame is to think of it as alternations between three states — mania/hypomania, depression, and euthymia — layered on a recurrent vulnerability that is at root a problem of neuroplasticity and cell resilience. The convergent finding across post-mortem, MRI and serum biomarker work over the past two decades is that bipolar disorder is fundamentally a disorder of impaired synaptic plasticity, reduced neurotrophic support, mitochondrial dysfunction, and increased oxidative stress, with these processes amplifying with each successive mood episode (the neuroprogression hypothesis associated with Berk, Kapczinski and colleagues).

The BDNF story. Grande and colleagues consolidated the case for BDNF as a central mediator of the neuroplasticity changes in bipolar disorder. The state-marker pattern is reproducible across multiple meta-analyses: serum BDNF is decreased during depressive episodes, decreased during manic episodes, and returns to normal during euthymia. This is the closest thing bipolar psychiatry has to a state-marker biomarker. The 2021 BJPsych Open Cambridge study of newly diagnosed bipolar patients and unaffected first-degree relatives showed that the BDNF deficit is partly present in unaffected relatives, suggesting at least a trait-marker component. The 2023 systematic review on BDNF as biomarker for bipolar disorder concluded that BDNF cannot yet serve as a stand-alone diagnostic biomarker but does index disease activity and treatment response.

The GSK-3β story. Glycogen synthase kinase-3β is one of the most consequential single enzymes in psychiatric biology. Lithium is a direct, potent inhibitor of GSK-3β; valproate inhibits GSK-3β indirectly via HDAC inhibition and other mechanisms; lamotrigine modulates voltage-gated sodium channels but also engages GSK-3β-adjacent biology. The 2013 PMC review of lithium and neuroprotection mapped the cascade: GSK-3β inhibition reduces tau phosphorylation, reduces amyloid-β42 production, increases β-catenin signalling, and upregulates BDNF expression — a cluster of effects that overlap substantially with what is wanted in Alzheimer’s disease and vascular cognitive impairment. The 2025 translational mechanisms review extended this to cytoskeletal and proteomic effects.

The hippocampal volume story. Bipolar disorder, like recurrent major depression, is associated with smaller hippocampal volumes on structural MRI. The Hajek et al. 2012 meta-analysis in Journal of Psychiatry & Neuroscience showed that this hippocampal volume reduction is masked when lithium-exposed patients are included in the sample — i.e. lithium exposure increases (or preserves) hippocampal volume in bipolar patients. The Kakhki et al. 2016 voxel-based morphometry study in Translational Psychiatry in elderly bipolar patients showed bilateral hippocampal volume increases on long-term lithium with verbal memory improvement over 4 years. A 2007 longitudinal MRI study showed bilateral hippocampal volume increases after long-term lithium treatment in patients with bipolar disorder.

The mitochondrial and oxidative-stress story. Bipolar disorder is associated with reduced complex I activity, increased lactate, and disrupted mitochondrial calcium handling, particularly in the BPD-mitochondrial subgroup identified by Kato and colleagues. Antioxidant deficits include reduced glutathione and increased lipid peroxidation. N-acetylcysteine (NAC), as a glutathione precursor, has small-trial positive signals in bipolar depression and is sometimes used off-label as an adjunct. The Russo 2010 finding (below) that serum HGF is decreased in bipolar disorder and normalises with zinc and antioxidant therapy sits at the intersection of the neurotrophic and oxidative-stress stories.

Where Does HGF/c-Met Sit in Bipolar Disorder Biology?

The single most direct biological link between Dihexa’s claimed mechanism and bipolar disorder is the Russo et al. 2010 finding published in Nutrition and Metabolic Insights: serum hepatocyte growth factor is decreased in individuals with bipolar disorder relative to controls, and the deficit normalises with zinc and antioxidant therapy. This finding has been somewhat overlooked — it sits in a nutrition journal rather than a high-impact psychiatry journal — but it is the most direct evidence in the bipolar literature for HGF involvement in the disorder.

The implications matter for any reader thinking about a synaptogenic HGF/c-Met intervention. If HGF is constitutively low in bipolar disorder, then potentiating HGF/c-Met signalling pharmacologically is a mechanistically coherent corrective intervention — in a way that it is not, for example, in schizophrenia, where the CommonMind Consortium found HGF mRNA upregulated in post-mortem dorsolateral prefrontal cortex. The directionality argument that makes synaptogenic agents theoretically risky in psychotic illness (where the dominant pathology is excessive synapse elimination) is, in bipolar disorder, more nuanced: bipolar depression involves reduced synaptic density on most accounts, but mania may involve the opposite.

The HGF biology more broadly is consistent with neurotrophic, neuroprotective and pro-regenerative effects in the CNS. HGF promotes neuronal survival, supports axonal regeneration, modulates microglial phenotype, and is secreted by mesenchymal stem cells as part of their neurotrophic profile. The same biology that we cover in detail on the Dihexa vs BDNF and mechanism of action pages applies here: HGF/c-Met and BDNF/TrkB are convergent rather than redundant neurotrophic axes, and the polarity of pharmacological intervention is, for bipolar disorder, less ambiguous than for schizophrenia.

Critical caveat. The Russo 2010 finding is a single study with a modest sample, in a low-impact journal, that has not been replicated to our knowledge. The MET schizophrenia GWAS signal does not have a clear bipolar parallel. The 2021 BJPsych Open Cambridge study focused on BDNF rather than HGF. The HGF/c-Met case in bipolar disorder is real but thin. Anyone telling you it is decisively established — in either direction — is over-interpreting the evidence.

BDNF, Val66Met & the Lithium-Valproate-Lamotrigine Convergence

BDNF, the canonical synaptic-plasticity neurotrophin, is the single most-studied biomarker in bipolar disorder. The clinical pattern is reproducible: serum BDNF is reduced during acute mood episodes (both depressive and manic) and returns to normal in euthymia. This is the dynamic that Grande and colleagues framed as the BDNF-mediator-of-neuroplasticity case for bipolar disorder. The 2021 newly-diagnosed-bipolar BJPsych Open paper extended the picture by showing partial BDNF reduction in unaffected first-degree relatives, suggesting at least a trait component.

The treatment-response side of the story is equally consistent. Preclinical studies show that lithium upregulates BDNF expression in hippocampus and prefrontal cortex, that valproate enhances BDNF in the same regions, that lamotrigine elevates hippocampal BDNF in animal models, and that memantine and dextromethorphan exert neurogenic effects via BDNF/TrkB signalling. The clinical implication is that the entire NICE CG185 first-line pharmacopoeia — lithium, valproate, lamotrigine — converges on BDNF upregulation as a downstream mechanism, alongside their direct ion-channel and enzyme-inhibition effects.

The BDNF Val66Met polymorphism is moderately studied in bipolar disorder, with smaller and less reproducible effects than in schizophrenia, where Mas et al. 2021 documented a clean Met-allele dose-dependent negative-symptom association. In bipolar disorder, the Met allele has been associated in some studies with poorer treatment response to lithium, with increased rapid cycling, and with worse cognitive performance in euthymia, but the literature is genuinely mixed.

What this means for a synaptogenic agent. If the entire first-line NICE CG185 mood-stabiliser pharmacopoeia is already engaging the BDNF axis, then the marginal benefit of adding a separate synaptogenic peptide depends on (a) whether the BDNF-mediated synaptogenic capacity is fully saturated by lithium / valproate / lamotrigine at therapeutic doses, and (b) whether the HGF/c-Met axis adds something orthogonal to BDNF/TrkB. The honest preclinical answer is that we do not know — this question has not been asked in any bipolar-relevant animal model. The Dihexa vs BDNF page covers the upstream pharmacology in more detail. See also the depression and mood review for the broader BDNF-and-affective-disorder argument that bipolar inherits.

Lithium as the Gold-Standard Neuroprotective Psychotropic — The 2026 Update

Lithium occupies a singular place in bipolar pharmacology and in neuropsychiatry more generally. It is the only psychotropic with a robust, replicated, mechanistically-coherent neuroprotective signal in living human patients, and it remains the NICE CG185 first-line maintenance treatment of choice in bipolar disorder in 2026. The mechanistic and structural-MRI case has matured substantially since the original lithium-and-suicide work of Schou and Baastrup in the 1960s and 1970s.

The structural MRI case. Multiple imaging studies and meta-analyses converge on the same picture. Smaller hippocampal volumes in bipolar disorder are masked by lithium exposure (Hajek et al. 2012 meta-analysis). Bilateral hippocampal volume increases after long-term lithium treatment in patients with bipolar disorder (Yucel et al. 2007 longitudinal MRI study). Voxel-based morphometry in elderly bipolar patients shows bilateral hippocampal volume increases with verbal memory improvement on long-term lithium (Kakhki et al. 2016). Short- and long-term lithium treatment is associated with increased hippocampal and amygdala volume and cortical thickness. The 2025 ENIGMA bipolar working group cross-sectional imaging signal continues to track lithium exposure.

The mechanistic case. The 2025 lithium-induced neuroprotection translational review consolidated the cellular and proteomic mechanisms. The central engine is GSK-3β inhibition, with downstream effects including (a) reduced tau phosphorylation, (b) reduced amyloid-β42 production, (c) increased β-catenin/Wnt signalling, (d) BDNF upregulation, (e) increased Bcl-2 / decreased Bax (anti-apoptotic), (f) cytoskeletal stabilisation, (g) mitochondrial biogenesis, and (h) reduced oxidative stress. These are the same mechanisms invoked in Alzheimer’s research and in MCI and brain aging, which is why several Phase 2 trials of low-dose lithium in mild cognitive impairment and prodromal Alzheimer’s have shown signals of slowing cognitive decline and reducing CSF phospho-tau (Forlenza and colleagues, particularly).

The amnestic-MCI signal. Low-dose lithium has been shown to slow decline in cognitive performance and reduce phospho-tau in patients with amnestic mild cognitive impairment in a small Phase 2 trial. Lithium is on the long-list of repurposed agents in MCI prevention pipelines, although the cardiovascular and renal monitoring burden makes it a non-trivial choice in primary prevention.

Practical UK lithium prescribing in 2026. NICE CG185 places lithium as first-line maintenance after a single manic episode in bipolar I (where the recurrence risk over 12 months is >50% without prophylaxis) and after two depressive episodes in bipolar II. CG185 specifies that lithium should not be started in primary care for people who have not taken lithium before, except under shared-care arrangements. Practical UK shared-care prescribing (NHS Nottinghamshire APC and equivalents) involves baseline U&Es, eGFR, TFTs, weight, ECG and pregnancy testing where appropriate; plasma lithium 12 hours post-dose at week 1 and weekly until stable in the target range (0.6-0.8 mmol/L for prophylaxis, 0.8-1.0 mmol/L for treatment of acute mania, with the 1.0-1.2 mmol/L upper range only when needed for treatment-resistance and never above 1.2 mmol/L); thereafter every 3 months for the first year and every 6 months after, with renal and thyroid monitoring twice yearly. A multi-centre UK audit across three NHS trusts found that many patients are managed below the NICE lower limit, with monitoring intervals that often fall short of guideline. Anyone reading this page who is on lithium and is not having regular U&Es, eGFR and TFTs should raise that with their GP at the next opportunity.

2026 News Context: What Has Just Changed in Bipolar Disorder

Three news stories matter for any 2026 UK bipolar readership.

1. FDA approval of Caplyta (lumateperone) as adjunctive MDD therapy, November 2025. Intra-Cellular Therapies / Johnson & Johnson’s lumateperone was approved by the FDA in November 2025 as adjunctive therapy with antidepressants for the treatment of major depressive disorder in adults. This is a substantial extension of the lumateperone label: the drug had previously been approved for schizophrenia (2019) and for bipolar I and II depression both as monotherapy and as adjunct with lithium or valproate (2021). Lumateperone’s pharmacological profile is unusual: it is a serotonin (5-HT2A antagonist), dopamine (D2 phosphorylation modulator with low intrinsic D2 occupancy at clinical doses) and glutamate (NMDA receptor modulator) triple-mechanism agent, with a relatively clean metabolic and movement-disorder side effect profile compared with most atypical antipsychotics. The 2026 UK position is that lumateperone has no MHRA approval as of May 2026 and Caplyta has not been considered by NICE. UK clinicians cannot prescribe it on the NHS for any indication. Whether MHRA submission and NICE appraisal follow in the 2027-2028 window will be a meaningful piece of UK bipolar news to watch.

2. UK MHRA reauthorisation of semi-sodium valproate for mania, September 2025. Semi-sodium valproate (the equimolar mixture of sodium valproate and valproic acid, branded Depakote, with marginally better gastrointestinal tolerance than plain sodium valproate) received UK marketing authorisation in September 2025 for the treatment of mania when lithium is not tolerated or is contraindicated. This is meaningful: the MHRA Pregnancy Prevention Programme on valproate is stringent (valproate must not be used in women or girls of childbearing potential unless they meet the Pregnancy Prevention Programme criteria, and from January 2024 may not be initiated in male or female patients under 55 unless two specialists independently document no alternative effective treatment), but the reauthorisation gives UK clinicians a second mood-stabiliser option when lithium is not viable. CG185 remains the operative guideline.

3. The NCISH 2025 annual report and the UCL bipolar suicide analysis. The 2025 National Confidential Inquiry into Suicide and Safety in Mental Health report from the University of Manchester documented 1,491 patient suicides with a primary diagnosis of bipolar disorder, 8% of all patient suicides, averaging 136 per year, with the 2019-2022 average 19% higher than the 2015-2018 average. The accompanying UCL analysis highlighted lithium under-use, missed bipolar diagnosis (and consequent inappropriate antidepressant monotherapy), insufficient CBT/psychoeducation access, and post-discharge care gaps as the main modifiable factors. For UK bipolar care, this is the single most important 2025-2026 data point: the suicide trajectory is going in the wrong direction, the modifiable drivers are largely structural, and unlicensed neurotrophic peptides are nowhere on the priority list compared with lithium adherence, bipolar diagnostic detection, CBT access and post-discharge follow-up.

Beyond Lithium, Valproate & Lamotrigine: The 2024-2026 Bipolar Pipeline

Caplyta (lumateperone) — Intra-Cellular Therapies / Johnson & Johnson

Lumateperone is the most consequential single bipolar drug development of the past decade. The drug’s pharmacology — serotonin 5-HT2A antagonism, indirect dopamine D2 modulation via phosphorylation, NMDA receptor modulation — is unusual enough to fit awkwardly in the ‘atypical antipsychotic’ class. Approved by the FDA for schizophrenia in 2019, for bipolar I and II depression in 2021 (both as monotherapy and as adjunct with lithium or valproate), and extended to adjunctive MDD in November 2025. The clinical profile is favourable: low metabolic side effects, low movement-disorder side effects, low prolactin, no QT signal of concern at standard doses, no clinically important manic switch signal in the bipolar II depression studies. The UK position: no MHRA approval as of May 2026; not on the NHS; not in NICE CG185.

Cariprazine (Vraylar in US, Reagila in Europe) — AbbVie / Allergan / Gedeon Richter

Cariprazine is a D3-preferential D3/D2 partial agonist that has been progressively building a bipolar evidence base since its 2015 US approval for schizophrenia and bipolar I mania. The D3 receptor preference is mechanistically distinctive: D3 is concentrated in mesolimbic circuits, has been implicated in motivational and reward processing, and may explain the drug’s relative efficacy for the negative-symptom dimension in schizophrenia and the bipolar depression signal. The 2021 PMC review mapped cariprazine’s bipolar pharmacology. The 2018 CANMAT/ISBD guidelines recommend cariprazine as first-line for acute mania and second-line for bipolar I depression. A Phase 4 D3 imaging trial is exploring the brain-circuit mechanism of action in bipolar depression. A large Phase 4 observational study is tracking real-world disease activity changes in bipolar I disorder on oral cariprazine. The paediatric Phase 4 programme enrolling ~380 patients aged 10-17 with bipolar I across ~60 sites worldwide will be a meaningful 2026-2027 readout. UK position: cariprazine is MHRA-approved for adult schizophrenia (Reagila) and is on the NICE TA pathway; bipolar use in the UK is currently off-label.

Esketamine (Spravato) and racemic ketamine in bipolar depression

Esketamine and intravenous racemic ketamine are the dominant rapid-acting antidepressant story of the past decade, but their bipolar depression evidence base lags behind their unipolar depression evidence base. Salloum et al. 2024 reported real-world clinical experience in 38 patients with bipolar depression on intranasal esketamine: 39% response rate, 13% remission rate after acute series, mean depression score reduction of 38.3%, and importantly no manic switch during the acute series phase. A large 2025 retrospective cohort study of 2,126 patients on esketamine for bipolar depression in real-world settings showed significantly lower suicide-related outcomes at 1-7 days through 1 year. The 2024 systematic review and meta-analysis of IV ketamine in bipolar depression consolidated the rapid-acting antidepressant signal with acceptable safety. The mechanism — NMDA receptor modulation, downstream BDNF upregulation and synaptic plasticity — is the same axis we cover for treatment-resistant depression on the depression and mood page. UK position: esketamine (Spravato) is MHRA-approved and available on the NHS for treatment-resistant unipolar depression in specialist services; bipolar use is off-label.

AXS-05 (dextromethorphan-bupropion / Auvelity)

AXS-05 combines dextromethorphan (NMDA receptor antagonist, sigma-1 receptor agonist) with bupropion (CYP2D6 inhibitor that boosts dextromethorphan plasma levels and adds dopamine-noradrenaline reuptake inhibition). FDA-approved for major depressive disorder in August 2022. The mechanism engages BDNF/TrkB signalling and the rapid-acting NMDA-glutamate axis. Bipolar development has been slower but is in progress. UK position: not MHRA-approved; not on the NHS.

KarXT (Cobenfy) bipolar I mania Phase 4 trial

The xanomeline-trospium muscarinic combination that received FDA approval for schizophrenia in September 2024 has a Phase 4 commitment to a bipolar I mania trial that we cover in detail on the schizophrenia review. The 2026 UK MHRA submission and launch for Cobenfy in schizophrenia is the proximal news; bipolar I mania data are downstream.

Dihexa Specifically in Bipolar Disorder: What the Evidence Actually Is

This is the section that does most of the cognitive work on this page. The structure mirrors the rest of the rolling 2026 series.

What Dihexa is. Dihexa (PNB-0408, N-hexanoic-Tyr-Ile-(6) aminohexanoic amide) is a small-molecule angiotensin IV analogue developed in the laboratories of Joseph Harding and John Wright at Washington State University. The Benoist 2014 JPET paper is the canonical mechanistic paper: Dihexa stabilises hepatocyte growth factor (HGF) and potentiates signalling through the Met receptor tyrosine kinase, with downstream MAPK/ERK and PI3K/Akt effects, synaptogenesis on cortical and hippocampal neurons, and reversal of scopolamine-induced spatial memory deficits in the Morris water maze. Full mechanism of action and Dihexa vs BDNF coverage on the dedicated pages.

What Dihexa is not, in 2026. Dihexa is not a licensed medicine anywhere in the world for any indication. There is no Investigational New Drug application, no published Phase 1, no published clinical-trial registration, and no published human PK or safety data for Dihexa itself in any therapeutic area, let alone bipolar disorder. The clinical analogue in the same chemical family is fosgonimeton (ATH-1017) from Athira Pharma, which has been studied in Alzheimer’s disease (LIFT-AD), dementia with Lewy bodies (SHAPE) and Parkinson’s disease dementia. The LIFT-AD Phase 2/3 Alzheimer’s trial missed its primary endpoint in September 2024; Athira paused further development and is exploring partnering options. None of those Athira programmes has been studied in bipolar disorder, bipolar depression, mania or any related condition.

What the preclinical literature actually covers. The published Dihexa animal literature is dominated by Alzheimer’s-relevant cognitive paradigms (spatial water maze, passive avoidance, scopolamine reversal), Parkinson’s-relevant 6-OHDA models, and angiotensin IV / IRAP receptor pharmacology. There is no published Dihexa work in canonical bipolar preclinical models — amphetamine sensitisation, sleep deprivation-induced manic-like behaviour, ouabain-induced manic-like behaviour, lithium reversal paradigms, valproate reversal paradigms, dark-light box, sucrose preference in chronic mild stress, or any of the rodent endophenotype panels used in bipolar drug discovery.

What the inference chain therefore actually is. To get from current Dihexa evidence to “Dihexa for bipolar disorder” you must chain together: angiotensin IV peptides → HGF/c-Met activation → synaptogenesis (Benoist 2014) + decreased serum HGF in bipolar disorder normalising with zinc/antioxidant therapy (Russo 2010) + BDNF as state-marker mediator of neuroplasticity in bipolar disorder (Grande 2010, BJPsych Open 2021) + lithium-induced GSK-3β-BDNF axis + hippocampal volume gains with long-term lithium (Hajek 2012, Yucel 2007, Kakhki 2016). Each link is real. The composite is suggestive. No link is a controlled clinical trial of Dihexa in any bipolar disorder population.

The polarity problem in bipolar disorder. Unlike schizophrenia, where the directionality of any synaptogenic intervention is fundamentally ambiguous (because cortical pathology is dominated by excessive synapse elimination), the bipolar disorder directionality is somewhat clearer in bipolar depression and euthymia — both phases involve reduced BDNF, reduced HGF (Russo 2010), reduced hippocampal volume in the absence of lithium, and a synaptogenic intervention is mechanistically coherent with what is wanted. The ambiguity in bipolar disorder is in the manic and hypomanic phase: there is reasonable theoretical concern that potentiating synaptic plasticity on top of an already-elevated mood state could trigger or worsen a manic switch. This concern has not been tested for Dihexa specifically — there are no published manic-switch data — but it is the most important theoretical safety concern for any reader thinking about Dihexa in a bipolar context. The depression and mood page covers analogous concerns for treatment-resistant unipolar depression in patients who may carry an undetected bipolar diathesis.

Safety: Manic Switch, Lithium/Valproate Polypharmacy & the Pregnancy Prevention Programme

Practical safety considerations for any reader with bipolar disorder, bipolar I, bipolar II, schizoaffective disorder bipolar type, or undiagnosed bipolar diathesis who is contemplating Dihexa fall into five categories.

1. Theoretical manic switch risk. The most important theoretical safety concern. Synaptogenic agents that engage the BDNF/TrkB and HGF/c-Met axes are mechanistically capable of amplifying activity-dependent plasticity in cortical and limbic circuits. In bipolar disorder, where manic and hypomanic episodes are characterised by exactly this kind of cortico-limbic disinhibition, the theoretical risk of triggering or worsening a manic switch is non-trivial. Esketamine, ketamine, AXS-05, antidepressants generally, and corticosteroids all carry documented manic-switch warnings in bipolar populations. There is no published evidence that Dihexa precipitates mania. There is also no published evidence that it does not. The honest scientific position is uncertainty, and the honest clinical position is ‘not while not on a mood stabiliser’.

2. Lithium drug-drug interactions. Lithium has a narrow therapeutic window (target plasma 0.6-0.8 mmol/L for prophylaxis; toxicity from 1.5 mmol/L upwards, with risk of permanent renal and cerebellar damage above 2.0 mmol/L). Most lithium drug interactions are pharmacokinetic and renal: NSAIDs, ACE inhibitors, angiotensin-receptor blockers and diuretics all elevate lithium levels by reducing renal clearance. There are no published Dihexa-lithium pharmacokinetic data. Given Dihexa’s parent structure as an angiotensin IV analogue, there is at least theoretical concern about interaction with the renin-angiotensin system, although Dihexa is not an AT1 or AT2 receptor ligand. Any UK reader on lithium contemplating self-administered Dihexa is taking on completely uncharacterised interaction risk.

3. Valproate and the MHRA Pregnancy Prevention Programme. Sodium valproate and semi-sodium valproate are first-rank teratogens (neural tube defects, congenital cardiac malformations, lower IQ in exposed children) and the MHRA Pregnancy Prevention Programme is the most stringent UK Yellow Card pharmacovigilance programme of its kind. Any woman or girl of childbearing potential on valproate must meet the PPP criteria including documented dual contraception. There are no Dihexa teratogenicity data in any species. Self-administered Dihexa during preconception or pregnancy is contraindicated by absence of evidence.

4. Antipsychotic polypharmacy. Many bipolar patients are on at least one atypical antipsychotic (quetiapine, lurasidone, olanzapine, cariprazine, aripiprazole, lumateperone if private prescription). The same QT, metabolic and CYP-interaction concerns apply as cover on the schizophrenia page: Dihexa has no published QT data, no published metabolic data, no published CYP interaction data. The pharmacokinetic uncertainty stack is genuinely large.

5. Lamotrigine and the rash window. Lamotrigine is a NICE CG185 option for bipolar maintenance, with the well-known serious rash risk (Stevens-Johnson syndrome, toxic epidermal necrolysis) in the dose-titration window. Dihexa-lamotrigine interactions are uncharacterised. Any reader in the lamotrigine titration phase should not be adding unlicensed peptides.

See also the Dihexa side effects & risks page and the dosage discussion — both written for the general nootropic / cognitive-enhancement reader rather than for patients with bipolar disorder, where the risk-benefit calculation is meaningfully different, the theoretical manic-switch concern is genuine, and the lithium / valproate / lamotrigine / antipsychotic polypharmacy creates a large uncharacterised interaction surface.

NICE CG185 & the Current UK Standard of Care for Bipolar Disorder in 2026

NICE CG185 (Bipolar disorder: assessment and management) remains the operative guidance, covering bipolar I, bipolar II, mixed affective and rapid cycling presentations, in children, young people and adults. The headline recommendations:

For acute mania. Antipsychotic (oral haloperidol, olanzapine, quetiapine or risperidone) as first-line; lithium or valproate as alternative or addition where antipsychotic alone is insufficient. Carbamazepine in limited circumstances. ECT for severe, treatment-resistant or pregnancy-related mania. The September 2025 MHRA reauthorisation of semi-sodium valproate for mania extends the practical options.

For acute bipolar depression. Quetiapine, olanzapine, olanzapine-fluoxetine combination, lurasidone, or lamotrigine. Antidepressant monotherapy is not recommended (risk of manic switch and rapid cycling acceleration); antidepressants are used cautiously in combination with a mood stabiliser. CBT and family-focused therapy are first-line non-pharmacological options.

For maintenance / relapse prevention. Lithium first-line. Valproate second-line (subject to the Pregnancy Prevention Programme). Olanzapine, quetiapine, aripiprazole or risperidone where lithium and valproate are not options. Lamotrigine particularly for bipolar II maintenance and for depressive relapse prevention. Psychoeducation, structured psychological intervention and an agreed crisis plan throughout.

For children and young people. CG185 specifies CAMHS-led care, with broadly the same pharmacological framework but with greater caution around antipsychotic and mood-stabiliser side effects in developing brains. The paediatric cariprazine Phase 4 programme will be a meaningful 2026-2027 evidence read.

For physical health and ECT. Annual physical health checks including cardiovascular risk, metabolic monitoring (weight, blood glucose, lipids), thyroid function (particularly on lithium), renal function (particularly on lithium), and ECG where appropriate. ECT for severe, prolonged or treatment-resistant manic, mixed or depressive episodes, including in pregnancy where it is sometimes the safest option.

Notably absent from CG185: any pharmacotherapy specifically for the bipolar-cognitive-impairment dimension that persists into euthymia. CBT and psychoeducation help with relapse prevention, social rhythm therapy helps with mood-cycle regularity, but the residual verbal-memory, executive-function and attention deficits that many euthymic bipolar patients describe remain pharmacologically unaddressed in 2026. That gap is what drives search interest in unlicensed neurotrophic peptides.

Practical Realities: What Actually Works in 2026

For a UK adult with diagnosed bipolar disorder in 2026, the evidence-based offer reads:

1. NICE CG185 maintenance lithium for bipolar I and recurrent bipolar II, with target plasma 0.6-0.8 mmol/L, U&Es / eGFR / TFTs twice yearly minimum, and shared-care primary-secondary care prescribing. The single most evidence-based intervention in bipolar disorder.
2. Antipsychotic for acute mania: Olanzapine, quetiapine, risperidone or haloperidol; semi-sodium valproate for mania where lithium is not viable, subject to MHRA PPP.
3. Bipolar depression pharmacology: Quetiapine, lurasidone, olanzapine-fluoxetine combination, or lamotrigine per CG185. Cautious antidepressant addition only on top of a mood stabiliser. Off-label use of Caplyta (lumateperone) on private prescription where MHRA approval is not in place.
4. Esketamine / ketamine for treatment-resistant bipolar depression in specialist services, off-label for the bipolar indication but with growing real-world evidence base.
5. Cognitive behavioural therapy for bipolar disorder (CBT-BD), psychoeducation and family-focused therapy through NHS Talking Therapies (formerly IAPT) and secondary care psychology services.
6. Social rhythm therapy and self-management. Sleep regularity, light exposure, alcohol moderation and stimulant avoidance are the high-leverage modifiable factors for relapse prevention.
7. Annual physical health review and metabolic monitoring, per NG185 / NG191. The cardiovascular and metabolic burden in bipolar disorder is comparable to schizophrenia.
8. Clinical-trial enrolment for bipolar depression, bipolar cognition, treatment-resistant bipolar and paediatric bipolar via NIHR Be Part of Research.
9. Peer support and family support: Bipolar UK (eCommunity, peer support, helpline), Mind, Rethink Mental Illness.
10. Crisis pathway: Your local NHS Community Mental Health Team, the Samaritans 24-hour helpline 116 123, or 999 / A&E in immediate danger.

Dihexa does not enter this list. Not because it has been tried and failed in bipolar disorder — it has never been tried in a controlled trial in any bipolar population — but because the licensed evidence-based offer is more developed than the unlicensed peptide option in every dimension that matters, the theoretical manic-switch concern is genuine, and lithium itself already provides a substantial neurotrophic and neuroprotective benefit that is closer in mechanism to what synaptogenic peptides claim to do than is often appreciated.

What Would Need to Happen for Dihexa to Become a Real Bipolar Option?

The same checklist used on the Huntington’s, schizophrenia, PTSD, FTD and vascular dementia pages applies here, with three bipolar-specific additions.

1. Published preclinical work in canonical bipolar models — amphetamine sensitisation, sleep-deprivation manic-like behaviour, ouabain-induced manic-like behaviour, lithium-reversal paradigms, valproate-reversal paradigms. None published to date for Dihexa.
2. Mechanistic preclinical work on manic-switch risk in cortico-limbic circuits — does Dihexa amplify or attenuate amphetamine-induced hyperactivity? Does it interact with the prefrontal-cortex / nucleus accumbens / amygdala circuit in models of mood elevation? Open question.
3. Replication of the Russo 2010 decreased-serum-HGF finding in bipolar disorder in a larger, modern cohort with euthymia stratification, would meaningfully strengthen the mechanistic case. The original finding has not been replicated to our knowledge.
4. Phase 1 healthy-volunteer PK/PD data — Dihexa has none. Fosgonimeton has Phase 1 PK from the Athira programme.
5. Phase 2 proof-of-concept in bipolar cognitive impairment in euthymia, not in acute mania or acute depression — the mechanistic case is cleanest for the residual / cognitive dimensions, not for acute mood episodes.
6. Phase 3 with a recognised neurocognitive endpoint (such as MATRICS Consensus Cognitive Battery) and a mood-stabiliser-adjunctive design — replacing mood stabilisers on the basis of a synaptogenic peptide is not a defensible design and would not pass any data and safety monitoring board.
7. MHRA and NICE pathway, with the Pregnancy Prevention considerations, before NHS uptake.

None of the above exists for Dihexa in bipolar disorder in 2026. By contrast, all of the above (or its equivalent for the bipolar indication) did exist for lumateperone by 2021, which is why Caplyta is FDA-approved for bipolar I and II depression and Dihexa is not approved for anything anywhere in the world.

The Bottom Line: A Plausible Mechanism, a Severe Disorder, and Zero Human Dihexa Data

Bipolar disorder is a real and treatable condition. The BDNF state-marker biology, the Russo 2010 decreased-HGF finding, and the lithium-as-GSK-3β-inhibitor-BDNF-upregulator mechanism genuinely overlap with Dihexa’s claimed synaptogenic mechanism. The HGF/c-Met biology is more directly aligned with bipolar pathology (where HGF is reduced) than it is with schizophrenia pathology (where HGF mRNA is upregulated in DLPFC). And yet none of this is a controlled clinical trial of Dihexa in any bipolar population, the theoretical manic-switch risk in a patient with synaptogenic potentiation on top of an already-vulnerable mood-regulation circuit is non-trivial, and the standard of care — lithium — is itself one of the most robustly neuroprotective psychotropics ever discovered.

The 2026 honest answer for someone living with bipolar disorder in the UK is: NICE CG185 lithium first-line maintenance with proper plasma-level and renal-thyroid monitoring; antipsychotic for acute mania; semi-sodium valproate for mania where lithium is not viable (under MHRA Pregnancy Prevention Programme); quetiapine / olanzapine-fluoxetine / lurasidone / lamotrigine / off-label Caplyta for bipolar depression; CBT-BD, psychoeducation, family-focused therapy and social rhythm therapy throughout; esketamine / ketamine for treatment-resistant bipolar depression in specialist services; annual physical health review; clinical-trial enrolment for bipolar cognitive impairment and treatment-resistance via NIHR Be Part of Research; and Bipolar UK, Mind, the Rethink helpline, and your local Community Mental Health Team for peer and crisis support. Unlicensed synaptogenic peptides like Dihexa belong nowhere in that hierarchy in 2026.

Frequently Asked Questions: Dihexa & Bipolar Disorder

Can Dihexa help bipolar disorder, bipolar depression or mania?

There is no published controlled clinical trial of Dihexa in bipolar I, bipolar II, bipolar depression, mania, mixed affective episode, rapid cycling, schizoaffective disorder bipolar type, or bipolar cognitive impairment. The mechanistic case via HGF/c-Met, the Russo 2010 decreased-serum-HGF finding, BDNF as bipolar state-marker, and lithium’s GSK-3β-BDNF axis is real but inferential. Self-experimenting with an unlicensed research peptide while symptomatic, or in a patient with undetected bipolar diathesis, carries a non-trivial theoretical risk of triggering manic switch and is not a sensible plan.

What is Caplyta (lumateperone) and is it available in the UK for bipolar disorder?

Caplyta (lumateperone) is a serotonin / dopamine / glutamate triple-mechanism agent from Intra-Cellular Therapies / Johnson & Johnson, FDA-approved for schizophrenia (2019), for bipolar I and II depression both as monotherapy and as adjunct with lithium or valproate (2021), and most recently extended in November 2025 to adjunctive therapy for major depressive disorder. The UK position is that lumateperone has no MHRA approval for any indication as of May 2026 and is not available on the NHS. Private prescription is possible in limited circumstances. NICE has not appraised the drug.

What happened with the UK semi-sodium valproate reauthorisation in September 2025?

Semi-sodium valproate received UK marketing authorisation in September 2025 for the treatment of mania when lithium is not tolerated or contraindicated. The MHRA Pregnancy Prevention Programme remains stringent: valproate must not be used in women or girls of childbearing potential unless they meet PPP criteria, and from January 2024 may not be initiated in male or female patients under 55 unless two specialists independently document no alternative effective treatment. The reauthorisation gives UK clinicians a second mania mood stabiliser option but the PPP constraints are real.

Is lithium really neuroprotective in bipolar disorder?

Yes, on the structural-MRI evidence. The Hajek 2012 meta-analysis showed that smaller hippocampal volumes in bipolar disorder are masked by lithium exposure. The Kakhki 2016 voxel-based morphometry study in elderly bipolar patients showed bilateral hippocampal volume increases on long-term lithium with verbal memory improvement over 4 years. The Yucel 2007 longitudinal study showed bilateral hippocampal volume increases after long-term lithium. The 2025 lithium-induced neuroprotection translational review consolidated the GSK-3β-BDNF-tau-amyloid-β42 mechanistic case. Low-dose lithium also has a small Phase 2 signal in amnestic MCI for slowing cognitive decline.

What does NICE CG185 say about bipolar treatment?

CG185 places lithium first-line for bipolar maintenance, antipsychotic first-line for acute mania, and quetiapine / olanzapine-fluoxetine / lurasidone / lamotrigine for bipolar depression. Antidepressant monotherapy is not recommended (risk of manic switch). Lithium should not be started in primary care for people who have not taken it before, except under shared-care arrangements. Dihexa is not mentioned anywhere in CG185.

Can Dihexa cause manic switch in bipolar disorder?

There are no published manic-switch data for Dihexa specifically. The theoretical concern is mechanistically genuine: synaptogenic agents that potentiate BDNF/TrkB and HGF/c-Met signalling could in principle amplify cortico-limbic activity-dependent plasticity in a way that triggers or worsens mania, the same way antidepressants, esketamine, AXS-05 and corticosteroids can do so. The honest position is uncertainty; the honest clinical position is ‘not while not on a mood stabiliser’.

Where can I find UK bipolar support?

Your GP and your local NHS Community Mental Health Team for diagnosis, prescribing and ongoing care. Bipolar UK (eCommunity, peer support, helpline) for peer and family support specifically. Mind and Rethink Mental Illness for broader mental-health information and family support. Samaritans 24-hour helpline 116 123 in crisis; 999 or A&E for immediate danger. The NIHR Be Part of Research portal for clinical-trial enrolment.

Sources & Citations

• NICE CG185 — Bipolar disorder: assessment and management.
• FDA approval of Caplyta (lumateperone) as adjunctive therapy for major depressive disorder (Johnson & Johnson, November 2025).
• MedCheck: FDA Approves Caplyta as Depression Adjunct — Psychiatric News (December 2025).
• FDA Approves Caplyta as Adjunctive Therapy for Major Depressive Disorder Based on Positive Phase III Results.
• Caplyta (lumateperone) prescribing information — Drugs.com.
• Grande I et al. The role of BDNF as a mediator of neuroplasticity in bipolar disorder. Psychiatry Investigation (2010).
• Brain-derived neurotrophic factor levels in newly diagnosed patients with bipolar disorder, their unaffected first-degree relatives and healthy controls. BJPsych Open (2021).
• Can Brain-Derived Neurotrophic Factor Be Considered a Biomarker for Bipolar Disorder? An Analysis of the Current Evidence (2023).
• Neuroplasticity Improves Bipolar Disorder: A Review. PubMed (2020).
• Future Pharmacotherapy for Bipolar Disorders: Emerging Trends and Personalized Approaches. MDPI (2025).
• Russo AJ. Decreased Serum Hepatocyte Growth Factor in Individuals with Bipolar Disorder Normalizes after Zinc and Anti-oxidant Therapy. Nutrition and Metabolic Insights (2010).
• Russo AJ. Decreased Serum Hepatocyte Growth Factor in Depression Correlates with Severity of Disease. Biomarker Insights (2010).
• Hepatocyte growth factor — reference overview.
• Hajek T et al. Smaller hippocampal volumes in patients with bipolar disorder are masked by exposure to lithium: a meta-analysis. Journal of Psychiatry & Neuroscience (2012).
• Kakhki MK et al. The influence of lithium on hippocampal volume in elderly bipolar patients: a study using voxel-based morphometry. Translational Psychiatry (2016).
• Yucel K et al. Bilateral hippocampal volume increases after long-term lithium treatment in patients with bipolar disorder: a longitudinal MRI study. Neuropsychopharmacology (2007).
• Lithium and neuroprotection: translational evidence and implications for the treatment of neuropsychiatric disorders. Neuropsychiatric Disease and Treatment (2013).
• Lithium-induced neuroprotection in bipolar disorder: Translational insights into cytoskeletal and proteomic mechanisms. PMC (2025).
• An analysis of lithium requesting across three hospital trusts in the UK: many people are managed with lithium levels below the current NICE guidance lower limit.
• Serum lithium test requesting across three UK regions: an evaluation of adherence to monitoring guidelines.
• Nottinghamshire APC — Lithium Prescribing Guideline for Mental Health Indications.
• Updated NICE Guideline CG185 — Bipolar disorder: assessment and management (Ashtons summary).
• Cariprazine in the Treatment of Bipolar Disorder: Within and Beyond Clinical Trials. PMC (2021).
• Observational Study of Oral Cariprazine Capsules in Adult Participants With Bipolar I Disorder — ClinicalTrials.gov.
• Cariprazine for Bipolar Disorder — Phase 4 Clinical Trial (Power.com).
• Full-spectrum efficacy of cariprazine across manic and depressive symptoms of bipolar I disorder: Post hoc analysis of pooled randomised controlled trial data. Journal of Affective Disorders (2024).
• Salloum NC et al. Efficacy and Safety of Ketamine/Esketamine in Bipolar Depression in a Clinical Setting. Journal of Clinical Psychiatry (2024).
• Risk of manic switch and suicidal outcomes in bipolar depression treated with esketamine: A one-year retrospective cohort study of 2,126 patients. European Neuropsychopharmacology (2025).
• An Update on the Efficacy of Single and Serial Intravenous Ketamine Infusions and Esketamine for Bipolar Depression: A Systematic Review and Meta-Analysis (2024).
• Ketamine for Bipolar Depression: A Systematic Review. International Journal of Neuropsychopharmacology.
• Hayes JF et al. Rates and associations of relapse over 5 years of 2,649 people with bipolar disorder: a retrospective UK cohort study. BMC Medicine (2023).
• NCISH — Suicide in mental health patients with bipolar disorder (2025 annual report, University of Manchester).
• UCL News — Why are suicide rates so high in bipolar disorder, and what can we do about it? (March 2025).
• NHS England Digital — Chapter 11: Bipolar disorder, Adult Psychiatric Morbidity Survey 2023-24.
• Mental Health Foundation — UK bipolar disorder statistics.
• Bipolar UK — UK’s national charity for people affected by bipolar disorder.
• Rethink Mental Illness.
• Mind.
• Samaritans — 24-hour helpline 116 123.
• NIHR Be Part of Research portal.
• Benoist CC et al. The procognitive and synaptogenic effects of angiotensin IV-derived peptides depend on activation of the HGF/c-Met system. JPET (2014).
• Fosgonimeton (ATH-1017) — ALZFORUM therapeutics database.
• Overview of Psychiatric Medications in the Pipeline in Phase III Trials — Innovations in Clinical Neuroscience (June 2024).