Endoxifen

Endoxifen, the active metabolite of tamoxifen, is in a Phase 3 trial (NCT06608641) for acute mania in bipolar I disorder, sponsored by Jina Pharmaceuticals [1]. The rationale is mechanism-based repurposing. Tamoxifen has produced rapid anti-manic effects in small placebo-controlled trials, attributed to direct inhibition of protein kinase C (PKC), an intracellular signaling enzyme that runs hot during manic episodes [5][6]. Endoxifen is roughly 30 to 100 times more potent than tamoxifen at the estrogen receptor and substantially more potent at PKC, with a reported IC50 of ~350 nM against PKCβ1 versus ≥5 μM for tamoxifen, on the order of a 14-fold improvement at the in vitro target [10]. The bet is that an enteric-coated oral formulation can deliver therapeutic CNS exposure without the gram-level tamoxifen doses prior bipolar studies required [2]. Because endoxifen is a metabolite of an FDA-approved drug (tamoxifen), Jina is most likely pursuing the 505(b)(2) regulatory pathway, which can reference existing tamoxifen safety data but typically yields shorter market exclusivity than a novel chemical entity. If positive, this becomes the first PKC-targeted mood stabilizer in a market dominated by lithium, valproate, and atypical antipsychotics. Commercially, this is a small specialty bet by a sponsor with no prior CNS approval. The same molecule has a separately developed breast cancer program at Atossa Therapeutics that is further along in public disclosure [4][9].

Endoxifen is not approved anywhere for any indication. The molecule is well-characterized as the metabolite responsible for most of tamoxifen's anti-estrogen activity in breast cancer treatment and prevention, and oral endoxifen formulations have been studied in Phase 1 and 2 breast cancer and mammographic-density trials by Atossa Therapeutics and the NCI [2][3][4]. The bipolar I program run by Jina Pharmaceuticals uses an enteric-coated oral formulation and is listed at Phase 3 under NCT06608641 [1]. No FDA Breakthrough Therapy, Fast Track, Orphan Drug, RMAT, or Priority Review designation has been confirmed in a primary source for the bipolar indication. Current enrollment status (Recruiting / Active Not Recruiting / Suspended) and site count for NCT06608641 are not transparently disclosed in the sources reviewed and should be reconfirmed at ClinicalTrials.gov before any modeling. Jina Pharmaceuticals is a small Pittsburgh-based specialty pharma known for nanoparticle and reformulation work. It has not previously moved a CNS asset through Phase 3, which leaves sponsor execution as an open question, and the company's financing runway through readout is not disclosed in primary sources reviewed. A readout date has not been disclosed in any verifiable filing reviewed. The much louder corporate signal on the molecule comes from Atossa Therapeutics (ATOS), which has filed 8-K and 10-Q updates on its breast oncology endoxifen program through 2026, but those filings address a different indication, formulation, and clinical pathway than the Jina bipolar trial [9].

In bipolar disorder, manic episodes are tied to overactive intracellular signaling in neurons. PKC (protein kinase C, a family of enzymes inside neurons that amplify signals coming from neurotransmitter receptors) is one of the central nodes stuck in the 'on' position during mania [7]. Lithium and valproate, the two oldest mood stabilizers, both reduce PKC activity indirectly after weeks of treatment, and that lag is part of why patients suffer breakthrough episodes during titration [7]. Tamoxifen, originally developed as an anti-estrogen for breast cancer, also blocks PKC directly at high doses and produced rapid anti-manic effects in small placebo-controlled trials, with separation in days rather than weeks [5][6]. Endoxifen is tamoxifen's main active metabolite, generated in the liver primarily by CYP2D6 (a cytochrome P450 enzyme that varies widely in activity between individuals; roughly 7 to 10 percent of Caucasian patients are poor metabolizers and convert very little tamoxifen to endoxifen). Endoxifen binds the estrogen receptor roughly 30 to 100-fold more tightly than tamoxifen and inhibits PKCβ1 with an IC50 of approximately 350 nM versus ≥5 μM for tamoxifen, on the order of a 14-fold gain in PKC potency in cell-free assays [2][8][10]. That pharmacology is the basis for the bipolar program: deliver the active species directly, skip the CYP2D6 conversion step that varies widely across patients, and reach CNS PKC at doses that do not require the gram-level tamoxifen exposures used in prior investigator-initiated mania studies. The case is biologically credible and has the strongest human proof-of-concept of any non-monoamine target in mood disorders, but PKC has at least ten isoforms with different cellular roles, the specific isoform driving mania is not pinned down, and no PKC inhibitor has been approved in psychiatry.

NCT06608641 is listed as a Phase 3 trial of enteric-coated endoxifen in adults with acute manic or mixed episodes of bipolar I disorder, sponsored by Jina Pharmaceuticals [1]. Standard acute mania registration design uses the Young Mania Rating Scale (YMRS) total score change from baseline at three weeks as the primary endpoint, with placebo comparator and stratification by site and severity. The exact enrollment target, dose, blinding scheme, and number of sites for this specific protocol could not be independently confirmed in a primary source as of this writing, so any specifics beyond phase, indication, sponsor, and oral enteric-coated formulation should be treated as preliminary pending the published protocol or a sponsor filing. The biggest design question is placebo response. Modern acute mania trials routinely show 30 to 40 percent YMRS improvement on placebo, which has sunk several otherwise pharmacologically defensible compounds. Tamoxifen's prior bipolar trials separated cleanly from placebo but enrolled fewer than 70 patients each and used doses well above the breast cancer label, raising both effect-size and safety extrapolation questions when moving to endoxifen at the lower exposures a chronic mood-stabilizer label would require [5][6]. A material regulatory concern is that there is no disclosed company-sponsored Phase 2 of endoxifen specifically in bipolar I, only the tamoxifen surrogate data. FDA typically expects at least two adequate and well-controlled trials for a first-in-class psychiatric NDA, and proceeding directly to Phase 3 without a sponsor-run Phase 2 in the target indication is atypical. Even on a positive readout, the agency may require a second confirmatory Phase 3, or supplementary Phase 2 evidence, before accepting the package. The 505(b)(2) pathway available because tamoxifen is approved can ease some preclinical and safety bridging requirements, but it does not lower the efficacy bar for a first-in-class psychiatric indication.

About 11% of drugs reaching this stage are eventually approved by regulators. Our model starts from the historical approval rate for Phase 3 drugs in this area (around 51%), then adjusts based on ten specific facts about this trial and its sponsor. The estimate is pulled down mainly by heavier-than-usual blinding requirements, the sponsor's limited approval track record, weaker earlier-phase results, and the trial's randomized design. The remaining factors fall close to average for this stage, so they don't move the number much in either direction.

Efficacy risk is the largest single concern. The mechanistic case rests on tamoxifen trials with small samples (n<70 each) and the translation of a parent-drug signal to a metabolite at chronically tolerable doses is not automatic [5][6]. Acute mania trials have killed several Phase 2-positive compounds because placebo response climbed and discontinuation rates wrecked statistical power. Safety risk is mechanism-based and material. Endoxifen is a potent SERM (selective estrogen receptor modulator, a class of drugs that block estrogen signaling in some tissues while activating it in others), and chronic anti-estrogen exposure carries fertility, vasomotor (hot flashes), thromboembolism, and endometrial signals that are acceptable in oncology but a hard sell in psychiatry, where the bipolar I population skews younger, includes women of reproductive age and pregnancy-considering patients, and has many alternatives [2][8]. Even a small uterine or venous thromboembolism signal in a 3-to-6 week trial would magnify across years of maintenance dosing. The SERM safety profile is also sex-stratified in important ways. Endometrial hyperplasia and endometrial cancer risk is female-only by anatomy, while venous thromboembolism risk applies to both sexes but is amplified by estrogen-containing contraceptives in women. Male bipolar I patients (roughly half the population) face a distinct adverse-event profile that can include reduced libido, hot flashes, and altered bone turnover from systemic anti-estrogen activity, which complicates label language and may shape prescriber uptake by sex. Execution risk is concrete: Jina Pharmaceuticals is small, enrollment status and pace for NCT06608641 are not transparently disclosed in primary sources reviewed, sponsor financing through readout has not been confirmed, and a single Phase 3 is unlikely to support approval without a second confirmatory study. Commercial risk is real even on a clean win. Generic lithium and valproate plus several atypical antipsychotics (olanzapine, quetiapine, risperidone, aripiprazole, cariprazine, asenapine) cover acute mania at low cost, and payers will demand differentiated efficacy or tolerability data, not just a novel mechanism, to support a branded price. The 505(b)(2) pathway also carries weaker IP durability than a novel chemical entity, exposing any branded launch to faster generic or paragraph IV challenges than a comparable NDA.

Worth watching, low priority until a Phase 2 readout or sponsor disclosure tightens the picture. The biology is the most interesting non-monoamine story in mood disorders and would be category-defining if it works, but the public dataset on the specific Jina Pharmaceuticals Phase 3 is too thin to underwrite anything more than tracking. Market context is modest. US bipolar disorder drug-and-treatment spend is estimated at roughly $1.7 to $2.1 billion annually, with acute mania inpatient and outpatient care accounting for a substantial share of that, so even a successful branded product faces real generic-comparator pricing pressure rather than a wide-open commercial runway. The single data point that would convert this from speculative to signal is a clean, adequately-powered Phase 2 result for endoxifen in acute mania showing greater than 5-point YMRS placebo separation with tolerable SERM-related adverse events at the chosen dose. Until that is in a primary source (peer-reviewed paper, ClinicalTrials.gov results posting, or sponsor SEC filing), the program sits in the plausible-but-unproven bucket. Check back when NCT06608641 enrollment status updates publicly, when Jina Pharmaceuticals issues a development milestone press release, or when any investigator-initiated SERM-in-bipolar paper publishes a new dataset. Endoxifen's much more developed breast cancer program at Atossa Therapeutics is the more commercially substantive story on the molecule today [4][9]. The bipolar indication is a separate, higher-risk bet on the same active species running on different pharmacology (PKC vs ER) at a much smaller sponsor.