ABBV-932

ABBV-932 is an AbbVie small molecule with an undisclosed mechanism being tested in two Phase 2 indications: depressive episodes in bipolar I/II disorder and generalized anxiety disorder. A first Phase 2 study in bipolar depression (NCT06605599, n=161) has completed but topline data has not yet been publicly presented [1], while two larger Phase 2 trials are now recruiting [2][3]. The asset sits inside AbbVie's roughly $55.5B 2024 net revenue base [4] and is part of the company's effort to build out a neuroscience pipeline behind Vraylar before that franchise faces composition-of-matter patent expiration around 2029 to 2030 [4].

Novel compound, never approved anywhere. AbbVie has been running the program through Phase 1 and Phase 2 in parallel since at least 2024. The completed Phase 2 study in bipolar I/II depression (NCT06605599, n=161) used the Montgomery-Asberg Depression Rating Scale (MADRS, the standard clinician-rated instrument for depression severity) as its primary endpoint [1]. That study is the first piece of efficacy data investors should care about, and AbbVie has not yet disclosed topline results in a press release or earnings call. A second Phase 2 in bipolar depression (NCT07220460, n=200) is now recruiting with adverse events as the listed primary endpoint, meaning it functions as a safety and tolerability extension rather than a key efficacy trial [2]. A third Phase 2 in generalized anxiety disorder, adding ABBV-932 on top of existing antidepressants (NCT06846320, n=315), is also recruiting [3]. Phase 1 work continues with a radiolabeled mass balance study (NCT06953934) that maps how the drug is metabolized and excreted, standard regulatory homework [5]. An earlier Phase 1 relative bioavailability study (NCT06849791, n=9) completed without disclosed safety concerns [8], so the program has cleared its initial Phase 1 gating. No FDA breakthrough, fast track, orphan, or priority review designations have been granted. A Phase 3 program likely will not begin until 2027 at the earliest.

AbbVie has not publicly disclosed what ABBV-932 binds to or how it changes brain chemistry. The company refers to it as a small-molecule oral therapy in pipeline disclosures and SEC filings, but the molecular target is absent from clinical trial records and investor materials accessible by compound code alone [4][6]. This silence is deliberate. Pharma companies often keep early-stage neuroscience targets confidential until Phase 2 efficacy is in hand, to avoid signaling competitors about which receptor or transporter is in play. What can be inferred from indication choice: bipolar depression and generalized anxiety disorder are conditions where existing drugs (SSRIs, atypical antipsychotics, benzodiazepines) work imperfectly and through monoaminergic pathways (drugs acting on serotonin, dopamine, or norepinephrine, the brain chemicals targeted by most existing antidepressants) along with GABA. AbbVie already owns Vraylar (cariprazine), a dopamine D3-preferring partial agonist approved for bipolar depression and the company's largest neuroscience asset [4]. A new molecule pursued in the same indications would likely either hit a non-monoaminergic target (such as glutamate, neuropeptide systems, or sigma receptors) or refine a monoamine mechanism in a differentiated way. Without disclosed target information, mechanism-based confidence in this program is low. The completed Phase 2 readout will be the first real test of whether the underlying biology is doing useful work.

Three Phase 2 trials are in play. NCT06605599, completed with 161 patients, tested ABBV-932 in adults with bipolar I or II depressive episodes using change from baseline in MADRS total score as the primary endpoint [1]. MADRS is a 10-item scale ranging 0 to 60 where a placebo-adjusted drop of roughly 3 to 5 points typically separates real responders from noise. A sample size of 161 is small for psychiatry, where placebo arms typically improve by 8 to 10 MADRS points on average, but adequate for a signal-seeking Phase 2. NCT07220460 is a separate Phase 2 enrolling 200 bipolar I/II patients with adverse events as the listed primary, indicating a longer-exposure safety study running alongside the efficacy work [2]. NCT06846320 enrolls 315 adults with generalized anxiety disorder and layers ABBV-932 on top of background antidepressant therapy, again with adverse events as the listed primary; the design implies AbbVie is positioning the compound as adjunctive rather than monotherapy in GAD [3]. None of the studies appears to use a biomarker-selected population, which is standard for psychiatry but means any efficacy signal must be unusually clean to stand out from placebo noise. Comparator arms in the recruiting trials are placebo, with adjunctive design in the GAD study.

Our model gives this drug an 18% chance of eventually being approved. It starts from the historical rate for Phase 2 drugs in this area, about 24%, then adjusts based on ten facts about the trial and its sponsor. The estimate is pushed higher by the trial's non-randomized design and open-label blinding, and pulled lower by its few secondary endpoints and weak earlier-phase results. The remaining facts fall close to average and have little net effect on the final number.

Efficacy risk dominates. Bipolar depression is one of the harder psychiatric indications. Placebo groups typically improve by 8 to 10 points on the MADRS by themselves, and only a small handful of drugs (Vraylar, lurasidone, lumateperone, quetiapine, the olanzapine-fluoxetine combination) have ever beaten placebo cleanly in registration trials. A 161-patient Phase 2 has limited statistical power to detect modest effects, so a borderline result is plausible even if the drug works. For generalized anxiety disorder, the field is more crowded and more forgiving: SSRIs, SNRIs, and buspirone are generic and cheap, so any new entrant needs a real differentiation story to get prescribed, let alone reimbursed. Safety risk depends entirely on the undisclosed mechanism. CNS drugs frequently develop unexpected liabilities at higher exposures: QT prolongation (a delay in the heart's electrical recovery visible on an EKG that can trigger dangerous arrhythmias at high doses), seizure threshold lowering, hepatic enzyme elevations, suicidal ideation signals. AbbVie's parallel mass balance and bioavailability studies indicate the pharmacokinetic profile is still being characterized, which is normal for this stage but means surprises remain possible [5][8]. Execution risk is low. AbbVie has the capital, the trial infrastructure, and the commercial neuroscience footprint from Vraylar to run a full program and launch a drug [4]. The bigger commercial risk is positioning. If ABBV-932 reads out only marginally better than placebo, payers will require step therapy (insurers mandating that cheaper generic drugs be tried and fail first before covering a new branded option) through cheap generic atypicals first. Even a clean win has to compete with Caplyta (lumateperone, Bristol-Myers Squibb after the Intra-Cellular acquisition), which already secured a bipolar depression label and is generating real prescription volume.

Worth watching, with a single clear signal to wait for. NCT06605599 is the asset's first real efficacy readout in bipolar depression, and its results are not yet publicly disclosed [1]. AbbVie's neuroscience strategy needs a successor to Vraylar, which generated more than $3 billion in 2024 sales and faces composition-of-matter patent expiration (the core patent protecting the molecule itself, after which generic copies can enter) around 2029 to 2030 [4]. That pressure intensified after the late-2024 Phase 2 failure of emraclidine (ABBV-552), the schizophrenia muscarinic agonist AbbVie acquired in the Cerevel deal, which removed what was supposed to be the company's next big neuroscience asset. ABBV-932 is one of the remaining shots on goal in that program, and a positive Phase 2 would justify both a Phase 3 commitment and a higher line-item in AbbVie's pipeline disclosures [6]. The addressable U.S. market across bipolar depression (roughly 3 to 5 million diagnosed adults) and GAD (approximately 6.8 million diagnosed adults) totals around 10 million patients, with peak branded sales in these categories typically running $500M to $2B annually for a differentiated entrant. Specific checkpoints: ABBV-932 was not called out in AbbVie's Q1 2026 earnings commentary [9], which is consistent with either a pending analysis or a negative result being managed quietly. The next scheduled catalyst is the Q2 2026 earnings call expected in late July 2026, with Q3 2026 earnings (late October) a more likely venue for presentation-ready data. The American College of Neuropsychopharmacology annual meeting in December and the American Society of Clinical Psychopharmacology meetings are natural disclosure venues. Until then, the program is in informational equipoise, and the model's 22% probability of approval is a reasonable anchor. If the bipolar depression readout misses or the GAD trial reads out flat, the program is likely shelved quietly. If it hits, expect target disclosure, a Phase 3 announcement, and a re-rating of AbbVie's neuroscience pipeline narrative.