Xanomeline and Trospium Chloride: Mechanism of Action, Clinical, and Translational Science

ABSTRACT

Xanomeline‐trospium chloride (Cobenfy, formerly KarXT) is a first‐in‐class, oral central M1/M4 muscarinic receptor agonist approved for the treatment of schizophrenia in adults. Cobenfy represents a paradigm shift by avoiding direct dopamine D2 receptor blockade and thus significantly reduces the burden of metabolic, extrapyramidal, and sedative side effects observed in standard first and second‐generation antipsychotics. Trospium chloride is a peripheral muscarinic antagonist with negligible blood–brain barrier penetration. Cobenfy is co‐formulated with trospium to minimize peripheral cholinergic effects caused by xanomeline. Under fasting conditions, xanomeline and trospium reach peak plasma concentrations (T _max) in 2 and 1 h, and maximum concentrations (C _max) of 1955 and 5787 pg/mL, respectively. While food intake does not change xanomeline pharmacokinetics, it reduces trospium C _max by around 70%. Xanomeline has a high volume of distribution (10,800 L) and is metabolized by several cytochrome P450 (CYP) enzymes. Trospium is a P‐glycoprotein (P‐gp) substrate and is 85% eliminated in feces. In phase III clinical trials (EMERGENT‐2, EMERGENT‐3), Cobenfy significantly reduced PANSS total scores (least squares mean difference −9.6 and −8.4, respectively; p < 0.0001) compared to placebo. Treatment‐emergent adverse events were primarily mild to moderate in severity and were mainly transient nausea (19%), constipation (17%), dyspepsia (15%), and vomiting (14%). Discontinuation rates due to adverse events were low (6.2% vs. 5.2% for placebo), and Cobenfy was not associated with weight gain, extrapyramidal symptoms, or cardiometabolic risks seen in conventional antipsychotics.

Clinical and Translational Card for Fruquintinib.

• Mechanism of Action

  • ○
    Muscarinic acetylcholine receptor (mAChR) combined agonist/antagonist. Xanomeline acts as a preferential M1 and M4 receptor agonist in the central nervous system (CNS). M1 receptor activation modulates NMDA signaling and neuronal plasticity, while M4 receptor activation inhibits D1 striatal neurons. There is no direct D2 receptor blockade.
• Indication(s)

  • ○
    Treatment of schizophrenia in adults.
• Dosage and Administration

  • ○
    Oral, fixed dose combination ranging from 50 mg xanomeline/20 mg trospium chloride orally twice daily up to 125 mg xanomeline/30 mg trospium chloride orally twice daily.
• Key pharmacokinetic characteristics

  • ○
    The AUC under fasting conditions is 15,280 pg*h/mL for xanomeline and 49,460 pg*h/mL for trospium. The peak plasma concentration (C _max) is 1955 pg/mL for xanomeline and 5787 pg/mL for trospium (fasted). Peak concentrations (T _max) are reached in about two hours for xanomeline and one hour for trospium. The elimination half‐life (t _1/2) is about five hours for xanomeline and 6 h for trospium. Xanomeline has a high volume of distribution (10,800 L) and high CNS penetration; trospium is a P‐gp substrate with negligible CNS penetration.

1. Introduction

Schizophrenia is a chronic and severe neurodevelopmental disorder affecting approximately 24 million people worldwide [1]. It is associated with markedly reduced life expectancy—individuals may die 15–20 years earlier than the general population, with a 3.7‐fold increased all‐cause mortality [2]. Cardiovascular disease is the leading cause of death, accounting for approximately one‐third of natural deaths, followed by pneumonia and chronic obstructive pulmonary disease; suicide has the highest cause‐specific mortality ratio [3]. The clinical presentation of schizophrenia is characterized by positive symptoms (hallucinations, delusions, and disorganized thought/behavior) and negative symptoms (blunted affect, alogia, avolition, anhedonia, social withdrawal), resulting in progressive functional decline. Over the past 70 years, all FDA‐approved antipsychotics relied on dopamine receptor blockade, dopamine receptor partial agonism, or serotonin 5‐HT2A receptor antagonism. These agents target such regions as the striatum, prefrontal cortex, hippocampus, thalamus, and anterior cingulate cortex [4]. While first and second‐generation antipsychotics demonstrate robust efficacy for treating positive symptoms, their impact on cognitive or negative symptoms remains weak. Moreover, conventional antipsychotics carry significant burden, including severe sedation, metabolic syndrome, hyperprolactinemia, QT prolongation, and extrapyramidal symptoms (EPS) such as tardive dyskinesia [4]. Cobenfy (xanomeline‐trospium, formerly KarXT) highlights an alternative mechanism from traditional dopamine‐based treatments. It is an oral, fixed‐dose combination of xanomeline tartrate, a muscarinic acetylcholine receptor agonist, and trospium chloride, a peripherally restricted muscarinic antagonist. This review highlights the regulatory status, mechanism of action, clinical and pharmacological profiles, and key clinical trials for Cobenfy.

2. Drug Regulatory Approval

Cobenfy is a first‐in‐class, oral, fixed‐dose muscarinic agonist/antagonist combination approved for the treatment of schizophrenia in adults. The FDA granted full regulatory approval in September 2024 based on data from five core clinical trials: three 5‐week, randomized, double‐blind, placebo‐controlled acute efficacy studies (NCT03697252, NCT04659161, NCT04738123) and two 52‐week open‐label extension trials (NCT04659174, NCT04820309) [5]. Trospium chloride is a known muscarinic antagonist previously approved by the FDA for the treatment of overactive bladder (approved NDA021595, trade name Sanctura, Allergan) [6, 7]. The combination of trospium chloride with xanomeline was developed to enable therapeutic dosing of xanomeline by reducing peripheral cholinergic side effects.

3. Mechanism of Action

The muscarinic acetylcholine receptor (mAChR) family belongs to the G‐protein coupled receptor (GPCR) superfamily and is expressed throughout the CNS and peripheral tissues including the heart, smooth muscle, and pancreas. These receptors regulate memory, smooth muscle function, cardiac function, behavior, and learning. The mAChR family comprises five muscarinic receptor subtypes, designated M1 through M5. The M1, M3, and M5 subtypes are stimulatory receptors coupled to Gq proteins, whereas M2 and M4 are inhibitory receptors coupled to G_i proteins (Figure 1). The M1 and M4 subtypes are highly concentrated in the striatum, hippocampus, and fronto‐cingulate cortex, and their expression levels are reduced in individuals with schizophrenia [8]. Neuroimaging studies show reduced cortical and striatal muscarinic levels correlating with cognitive deficits and positive symptoms. M1 activation is proposed to enhance NMDA signaling and synaptic plasticity relevant to learning and memory, while M4 activation is hypothesized to inhibit D1 striatal neurons and modulate dopamine release. One proposed model suggests that muscarinic agonism may suppress positive symptoms without direct D2 antagonism, though this remains under study [9].

FIGURE 1.

Mechanism of action of Cobenfy (KarXT), a fixed‐dose combination of xanomeline tartrate and trospium chloride. Xanomeline is a preferential agonist at central M1 (Gq‐coupled) and M4 (Gi‐coupled) muscarinic acetylcholine receptors. In the brain, M1 activation stimulates the phospholipase C‐IP3/DAG pathway, increasing intracellular calcium and protein kinase C to produce an overall excitatory effect that supports cognition and may improve negative symptoms. M4 activation inhibits adenylyl cyclase, lowers cAMP and protein kinase A, and opens G‐protein‐gated potassium channels, producing an overall inhibitory effect, thereby reducing positive symptoms of schizophrenia. Xanomeline has lower affinity for M2, M3, and M5 receptors and can cause peripheral cholinergic side effects. Trospium chloride, a quaternary amine that does not cross the blood–brain barrier, antagonizes peripheral M2 and M5 receptors in the gastrointestinal tract, heart, bladder, and glands to mitigate these side effects without diminishing central M1/M4 activity. Ach, acetylcholine; cAMP, cyclic adenosine monophosphate; DAG, diacylglycerol; Gi/o, inhibitory G protein; GPCR, G‐protein coupled receptor; IP3, inositol triphosphate; PKC, protein kinase C.

Xanomeline acts as a preferential M1 (Ki = 10 nM) and M4 (Ki = 7 nM) receptor agonist. Xanomeline can also bind to M2, M3, and M5 receptors at lower affinities (Ki = 12 nM, Ki = 17 nM, Ki = 22 nM, respectively) [10]. Despite xanomeline's preferential M1/M4 activity, its initial development for Alzheimer's disease was abandoned due to peripheral cholinergic adverse effects including gastrointestinal hypermobility, nausea, diaphoresis, increased salivation, bradycardia, and urinary urgency. To overcome this, trospium chloride was added. As a charged quaternary amine, trospium cannot cross the blood‐brain barrier and is restricted to peripheral tissues. The combination of xanomeline with trospium resulted in as much as 46% reduced cholinergic adverse events compared to xanomeline monotherapy [6].

At the intracellular level, xanomeline binds to M1 receptors and activates the Gq/11 pathway, activating phospholipase C and the phosphatidylinositol triphosphate cascade. This generates the second messengers diacylglycerol (DAG) and inositol triphosphate (IP3), increasing intracellular Ca²⁺ release and protein kinase C expression for an overall excitatory signaling cascade proposed to contribute to enhanced cognition. For the M4 receptor, coupling to Gi/o proteins inactivates adenylyl cyclase activity, lowers cyclic adenosine monophosphate (cAMP) levels, and protein kinase A expression. The inhibitory cascade is hypothesized to downregulate hyperdopaminergic signaling, thus contributing to treatment of positive symptoms [11].

4. Pharmacokinetic/Pharmacodynamic (PK/PD) Characteristics

4.1. Absorption, Distribution, Metabolism, Elimination (ADME)

Cobenfy pharmacokinetics have been characterized across one phase 2 clinical trial (NCT03697252), two phase 3 clinical trials (NCT04659161, NCT04738123), and two 52‐week open‐label extension trials (NCT04659174, NCT04820309), which encompass the approved dosage range of 100 mg/20 mg to 125 mg/30 mg administered orally twice daily (BID). Additional studies remain ongoing, including evaluations of the safety and efficacy of xanomeline/trospium in first episode/early phase schizophrenia and treatment in adolescents (NCT06923891, NCT07288567) [6, 12, 13]. Pharmacokinetic parameters are shown in Table 1.

TABLE 1.

Pharmacokinetic summary for Xanomeline and Trospium.

ADME characteristic	Xanomeline	Trospium
Absorption (fasted)	T max ~2 h C max 1955 pg/mL AUC 15280 pg*h/mL	T max ~1 h C max 5787 pg/mL AUC 49460 pg*h/mL
Food intake	Minimal impact on pharmacokinetics	Significant: T max prolonged to 3 h, marked decrease in C max and AUC
Low‐fat meal	C max 1749 pg/mL AUC 16640 pg*h/mL	C max 753 pg/mL AUC 14460 pg*h/mL
High‐fat meal	C max 2130 pg/mL AUC 19830 pg*h/mL	C max 672 pg/mL AUC 13440 pg*h/mL
Distribution (V d )	10,800 L	Peripherally restricted (quaternary ammonium)
Protein binding	95%	None specified (low CNS penetration)
CNS penetration	High: Passive diffusion, not a substrate for P‐gp/BCRP efflux	None: Active P‐gp substrate; limited BBB crossing
Metabolism	CYP2D6, 2B6, 1A2, 2C9, 2C19, FMO1, FMO3	Ester hydrolysis and glucuronic acid conjugation
Elimination	78% urine/12% feces (inactive metabolites)	85% feces/6% urine
Half‐life (t 1/2)	~5 h	~6 h
Clearance (CL)	1950 L/h	29 L/h

Note: Unless otherwise noted, AUC, C _max, and t _1/2 values are reported as geometric mean, and T _max as median, consistent with standard pharmacokinetic reporting conventions.

Abbreviations: AUC, area under the curve; C _max, maximum concentration; CNS, central nervous system; CYP, cytochrome P450 enzyme; T _max, time to maximum concentration.

Following oral administration, Xanomeline and trospium are absorbed with time to peak plasma concentrations (T _max) of approximately two hours and one hour under fasting conditions, respectively. Trospium has undetectable CNS penetration at steady‐state peak plasma concentrations (NDA 021595). While food intake had minimal effect on xanomeline pharmacokinetics, it significantly altered trospium exposure. With both low and high‐fat meals, trospium T _max was prolonged to three hours, while C _max and AUC decreased by 70%–75% and 85%–90%, respectively [6].

Xanomeline has a high volume of distribution (10,800 L) and 95% plasma protein binding, indicating extensive tissue distribution despite a relatively small unbound plasma fraction. It is not a substrate of major uptake or efflux transporters including P‐glycoprotein (p‐gp), breast cancer resistance protein (BCRP), organic anion transporting polypeptide (OATP1B1 and OATP1B3), organic anion transporters (OAT1 and OAT3), organic cation transporters (OCT1 and OCT2), and multidrug and toxin extrusion protein (MATE1 and MATE2K). The absence of P‐gp and BCRP efflux liability, combined with its high tissue distribution, supports efficient blood–brain barrier penetration and sustained central nervous system exposure, as the unbound fraction diffuses into the brain without active transport‐mediated removal. In contrast, trospium is a substrate of P‐gp and is a quaternary ammonium compound, promoting active efflux into the periphery and limiting passive diffusion across the blood–brain barrier [6]. Thus, xanomeline maintains central exposure despite limited systemic availability, while peripherally restricted trospium minimizes peripheral cholinergic adverse effects.

Regarding elimination, xanomeline is eliminated via urine (78%) and feces (12%) as mostly inactive metabolites, with 1950 L/h clearance and a five‐hour half‐life. Xanomeline is metabolized by CYP2D6, CYP2B6, CYP1A2, CYP2C9, and CYP2C19, as well as by flavin monooxygenases FMO1 and FMO3. Trospium is eliminated via feces (85%) and urine (6%), with a 29 L/h clearance and a six‐hour half‐life. Trospium is hypothesized to be metabolized by ester hydrolysis with the formation of benzylic acid and azoniaspironotropanol followed by glucuronic acid conjugation, though this has not been fully characterized [6]. Despite co‐formulation, the two agents remain pharmacokinetically independent.

Mild renal impairment increases exposure to both agents, but no adjustment is needed since safety remained comparable to normal renal function. Hepatic impairment, however, disproportionately affects xanomeline, with exposure increasing 2.6‐fold at mild impairment and more at moderate impairment, while trospium is minimally affected. Thus, Cobenfy is not recommended in mild hepatic impairment and is contraindicated in moderate/severe hepatic impairment.

4.2. Special Populations

Cobenfy exhibits dose‐proportionate pharmacokinetics, yielding higher predictable xanomeline and trospium levels, but high interpatient variability causes substantial AUC 0–12 overlap between the 100 mg/20 mg and 125 mg/30 mg doses. Xanomeline and trospium have no weight, race, or age‐based dose adjustments. Cobenfy is not recommended in patients with moderate (eGFR: 30 to < 60 mL/min) to severe (eGFR: ≤ 30 mL/min) renal impairment and mild (Child‐Pugh Class A) hepatic impairment. It is contraindicated in patients with moderate to severe hepatic impairment (Child‐Pugh Class B or C), urinary retention, and untreated narrow‐angle glaucoma. The maximum approved dose evaluated for geriatric patients is 100 mg/20 mg BID. The safety and efficacy among pediatric patients are unknown [6]. The evidence base is limited by the populations studied. The EMERGENT trials enrolled adults aged 18–65 without significant comorbidities, so no data exist in pediatric, geriatric, or pregnant populations. This novel mechanism raises safety considerations that cannot be extrapolated from existing antipsychotic data. While avoiding direct dopamine antagonism may offer metabolic benefits, muscarinic agonism and anticholinergic effects raise concerns for gastrointestinal effects, tachycardia, and urinary retention.

4.3. Drug–Drug Interactions

At the systemic level, xanomeline and trospium do not inhibit or induce major CYP enzymes or transporters. Since xanomeline is metabolized by several CYP enzymes, clinically significant drug–drug interactions are less likely. At the gut level, xanomeline inhibits CYP3A4 and P‐gp. At the systemic and gut level, trospium is a noninhibitor and non‐inducer of all major CYP enzymes and major drug transporters. Consequently, the core clinical recommendation for Cobenfy is proactive monitoring for adverse effects and potential drug–drug interactions (DDIs). Close monitoring is recommended when Cobenfy is co‐administered with strong CYP2D6 inhibitors, drugs eliminated by active tubular secretion, or other cholinergic or muscarinic agents [6].

4.4. Exposure‐Response

Exposure‐response analyses examined the relationship between the proportion of subjects with AEs and xanomeline and trospium exposure. Safety was assessed for endpoints experienced by at least 5% of subjects. Only constipation (16%) and dry mouth (7.8%) had frequencies > 5%. Although higher xanomeline exposure was associated with greater improvement in PANSS total score, no meaningful difference in clinical efficacy was observed between dose changes. There was no clinically relevant effect on the QTc interval at doses up to 125 mg/30 mg Cobenfy twice daily. In phase two and three studies, results from the univariate exposure‐response analyses revealed a statistically significant exposure‐response relationship between the probability of constipation and C _avg up to the event for xanomeline. These findings suggest that subjects with higher xanomeline or trospium C _avg up to an event appear to have a decreased incidence of constipation or dry mouth. In addition, there was no apparent difference in the occurrence of constipation or dry mouth between 125/30 and 100/20 mg dose levels. The exposure‐tolerability analysis results do not suggest a positive exposure‐response relationship between trospium exposure and proportion of subjects with constipation or dry mouth. An additional consideration is that these analyses spanned only the relatively narrow exposure range achieved with the clinical dosing and titration regimen. The absence of a clear exposure‐response relationship should not be interpreted as evidence that tolerability is independent of exposure more broadly, but rather that exposure differences within the studied range may have been insufficient to reveal such a relationship. Mechanistically, while the observed trospium exposure‐response relationship may initially appear counterintuitive, it is likely that as the concentrations of both components increase, xanomeline's ability to mitigate trospium‐mediated anticholinergic adverse effects predominates over the trospium‐mediated effects [6]. This apparent dissociation warrants interpretation. Across the EMERGENT trials, gastrointestinal adverse events were most common in weeks 1–2 and declined despite continued dose escalation and steady‐state exposure. Several complementary mechanisms may explain this. First, trospium, a peripherally restricted antagonist with minimal CNS penetration, is co‐escalated with xanomeline, maintaining peripheral blockade while preserving central M1/M4 activation. Second, agonist‐induced receptor desensitization and downregulation are well‐established in muscarinic receptor biology and may allow peripheral gastrointestinal receptors to adapt during the initial weeks, attenuating gastrointestinal adverse effects despite increasing drug exposure [14]. Finally, the titration schedule may exploit the adaptation period by delaying maintenance doses until desensitization begins. Together, these mechanisms might explain why GI events peak early and diminish despite continued exposure.

5. Key Clinical Trials

Key clinical studies evaluating Cobenfy's safety and efficacy include Phase‐I Healthy Volunteer Study (NCT02831231), Phase‐II EMERGENT‐1 (NCT03697252), Phase‐III EMERGENT‐2 (NCT04659161), and EMERGENT‐3 (NCT04738123), and 52‐week open‐label extensions EMERGENT‐4 (NCT04659174) and EMERGENT‐5 (NCT04820309) (Table 2) [6, 15]. In EMERGENT‐2 and EMERGENT‐3, the least squares (LS) mean differences in PANSS total score versus placebo were −9.6 (95% CI: −13.9, −5.2, p < 0.0001) and −8.4 (95% CI: −12.4, −4.3, p < 0.0001), respectively.

TABLE 2.

Overview of key clinical trials evaluating Cobenfy for schizophrenia.

Trial name/phase (clinical trial registration)	Population	N	Design	Dose	Primary endpoint	Key findings	Commonly reported AE (> 2.0%)	Severity of AE (> 1.0%)
Healthy volunteer/phase‐I (NCT02831231)	Healthy volunteers	68 (33 xanomeline alone, 35 KarXT)	Single site, 9‐day, double‐blind comparison of xanomeline alone versus KarXT	Xanomeline alone: 75 mg three times daily; xanomeline 75 mg TID + trospium 20 mg twice daily	Rates of five predetermined cholinergic adverse effects (nausea, vomiting, diarrhea, excess sweating, salivary hypersecretion)	KarXT reduced composite incidences of cholinergic AEs by 46% compared to xanomeline alone. Each AE reduced by ≥ 29%; rate of postural dizziness 11.4% (KarXT) versus 27.2% (xanomeline alone); 0 syncope episodes versus 2 syncope episodes with xanomeline alone	Nausea, vomiting, excessive sweating, postural dizziness, diarrhea, salivary hypersecretion	Mild—moderate; no severe AE reported
KAR‐004/EMERGENT‐1 Phase 2 (NCT03697252)	Adults 18–65 years with DSM5 diagnosis schizophrenia, acute psychosis requiring hospitalization, PANSS 80–120, CGI‐S ≥ 4	182 (90 Xanomeline‐trospium, 92 placebo)	5‐week, randomized, double‐blind, placebo‐controlled, inpatient, across 12 US sites	Flexible; 50/20 mg BID up to 125/30 mg BID max	Change in PANSS score from baseline to week 5	Least squares mean difference: −11.6 points (95% CI –16.1 to −7.1; p < 0.001); Cohen's d = 0.75; significant improvement in PANSS positive, negative subscales	Constipation (17%), dry mouth (9%), dyspepsia (9%), vomiting (9%)	Mild or moderate; no discontinuations due to GI AEs, serious AE similar between groups
KAR‐007 EMERGENT‐2 Phase 3 (NCT04659161)	Adults 18–65 years, with DSM5 diagnosis of schizophrenia, acute psychosis requiring hospitalization, PANSS ≥ 80, CGI‐S ≥ 4	252 (126 KarXT, 126 placebo)	5‐week, randomized, double‐blind, placebo‐controlled, flexible‐dose inpatient, across 22 US sites	Flexible; 50/20 mg BID up to 125/30 mg BID max	Change in PANSS score from baseline to week 5	Least Squares mean difference: −9.6 (95% CI –13.9 to −5.2; p < 0.0001), Cohen's d = 0.61; all secondary endpoints met	Constipation (21%), dyspepsia (19%), nausea (19%), vomiting (14%), headache (14%), hypertension (10%), dizziness (9%), GERD (6%), diarrhea (6%)	Majority mild/moderate, transient, resolved in 2–3 weeks; serious AE: 2 suicidal ideation (KarXT); discontinuation due to AEs: 7% versus 6%
KAR‐009 EMERGENT‐3 (Phase 3) NCT04738123	Adults 18–65 years, with DSM5 diagnosis of schizophrenia, acute psychosis requiring hospitalization, PANSS ≥ 80, CGI‐S ≥ 4	256 (125 KarXT, 131 placebo)	5‐week, randomized, double‐blind, placebo‐controlled, flexible‐dose inpatient, across 30 sites (18 US, 12 Ukraine)	Flexible; 50/20 mg BID up to 125/30 mg BID max	Change in PANSS score from baseline to week 5	Least Squares mean difference: −8.4 (95% CI –12.4 to −4.3; p < 0.001); Cohen's d = 0.60; all secondary endpoints met	Nausea (19.2%), dyspepsia (16%), vomiting (16%), constipation (12.8%), hypertension (6.4%)	Mild/moderate, AE‐related discontinuation: 6.4% versus 5.5%
EMERGENT‐4 KAR‐008 (52‐week OLE) (NCT04659174)	Participants completed EMERGENT‐2 or 3	156 enrolled, 152 treated, 34 completed, however results non‐finalized	52‐week, open label extension	Flexible; 50/20 mg BID up to 125/30 mg BID max	Proportion with ≥ 1 TEAE	Results non‐finalized at time of review. 53.3% experienced ≥ 1 TEAE; PANSS total improved −33.8 points from acute baseline to week 52; no new safety signals; no clinically meaningful motor symptoms; weight gain, hyperprolactinemia, or metabolic effects	Nausea, vomiting, dyspepsia, dry mouth	Mild or moderate; resolved with continued treatment
EMERGENT‐5 KAR‐011 (52‐week OLE) NCT04820309	Psychiatrically stable adults with schizophrenia	566 enrolled, 277 (48.9%) completed	52‐week, open label, outpatient, across 54 US sites	Flexible; 50/20 mg BID up to 125/30 mg BID max	Long term safety, efficacy, tolerability	Results non‐finalized at time of review. 82.3% experienced ≥ 1 TEAE; severe TEAEs in 5.8%; continued improvement in PANSS total, positive/negative subscales, CGI‐S over 52 weeks	Nausea (23.1%), vomiting (20.3%), constipation (18.0%), hypertension (10.4%), dry mouth (9.4%), diarrhea (9.4%), dizziness (8.8%), headache (8.1%), dyspepsia (7.2%), somnolence (6.2%), weight loss (5.7%), hyperhidrosis (5.1%), urinary retention (1%)	Mostly mild/moderate; severe TEAEs: 5.8%; no new safety issues emerged

Abbreviations: AE, adverse event; BID, twice daily; KarXT, capsules containing xanomeline and trospium chloride mixed in various ratios; N, number of patients in study; NCT, national clinical trial; PANSS, positive and negative syndrome scale; TEAE, treatment‐emergent adverse event; TID, three times daily; US, United States.

In phase three clinical trials (EMERGENT‐2 and EMERGENT‐3), short term efficacy studies and long‐term safety studies were used to evaluate Cobenfy safety. In five‐week trials, anticholinergic treatment‐emergent adverse events (TEAEs) occurred in 35.0% of Cobenfy‐treated subjects (vs. 14.9% placebo), primarily constipation (17.1%), dyspepsia (15.3%), dry mouth (5.0%), dizziness (4.7%), blurred vision (2.4%), tachycardia (2.4%), and dysuria/urinary retention (0.3% each). Pro‐cholinergic TEAEs affected 32.4% (vs. 9.9% placebo), primarily nausea (18.5%), vomiting (13.5%), GERD (4.1%), somnolence (3.8%), salivary hypersecretion (2.1%), and hyperhidrosis (1.8%). Long term studies were consistent and reported constipation (15.4%), dry mouth (9.3%), dyspepsia (8.3%), dizziness (7.1%), tachycardia (2.2%), blurred vision (2.1%), urinary retention (1.9%), and dysuria (0.4%), nausea (18.4%), vomiting (16.8%), diarrhea (8.0%), somnolence (4.7%), hyperhidrosis (3.3%), GERD (2.8%), and salivary hypersecretion (2.7%). Of note, Cobenfy was not associated with standard antipsychotic cardiovascular or cerebrovascular risks (e.g., weight gain, hyperglycemia, dyslipidemia, hypertension, and insulin resistance) [6].

5.1. Pooled Safety Analysis

A pooled analysis (Table 3) combining three five‐week, phase II trials (EMERGENT‐1, 2, and 3) examined Cobenfy safety and tolerability in 683 adults with schizophrenia using twice‐daily flexible dosing (maximum xanomeline 125 mg/trospium 30 mg) [16]. Cobenfy significantly reduced PANSS total scores compared to placebo. While Cobenfy demonstrated higher TEAE rates (51.4% vs. 29.7% placebo), discontinuation rates due to AEs were low and comparable (6.2% cobenfy, 5.2% placebo). Common AEs corresponded to Cobenfy's muscarinic and cholinergic profile, primarily mild‐to‐moderate, transient gastrointestinal symptoms (nausea 19%, constipation 17%, dyspepsia 15%, vomiting 14%). Extrapyramidal symptoms were negligible (0.6%, 0% placebo). Cobenfy‐treated patients had less significant weight gain (> 7% body weight; 5.2% vs. 11.2% placebo) and no hyperglycemia, elevated hemoglobin A1c (HbA1c), or hyperlipidemia. To date, there are no placebo‐controlled maintenance or long‐term studies evaluating long‐term efficacy [6].

TABLE 3.

Pooled analysis data.

Pooled analysis (NCT03697252, NCT04659161, NCT04738123) a

Adults with schizophrenia with acute psychosis

683 (340 KarXT, 343 placebo)

5‐week, randomized, double‐blind, placebo‐controlled, flexible‐dose inpatient, across multiple sites

Flexible; 50/20 mg BID up to 125/30 mg BID max

Change in PANSS score from baseline to week 5

TEAE's: 66.8% (KarXT) versus 49.0% (placebo); treatment‐related AEs: 54.1% versus 29.7%; EPS: 0.6% versus 0%; akathisia: 3.2% versus 0.6%; weight gain: 5.2% versus 11.2%

Nausea (19%), constipation (17%), vomiting (14%), dyspepsia (15%), hypertension (6.2%)

Mild/moderate, transient; discontinuation rates similar (27.6% vs. 22.7%); AE‐related discontinuation: 6.2% versus 5.2%

^a The pooled analyses (EMERGENT‐1, NCT04659161, NCT04738123) are an integrated safety and efficacy analysis and do not contain an NCT number.

6. Summary of Clinical Efficacy and Safety

Overall, Cobenfy demonstrated significant superiority over placebo, with greater mean improvement in positive and negative syndrome scale (PANSS) total scores from baseline to week five across phase II and two phase III (NCT04659161 and NCT04738123) trials. Cobenfy was generally well tolerated. The most common AEs (> 10% incidence)—nausea, vomiting, dyspepsia, constipation, and headache—aligned with the cholinergic/anticholinergic effects from xanomeline and trospium. Post hoc analysis revealed these AEs were not dose‐dependent. Most were mild‐to‐moderate, reported within the first seven days, and resolved by trial completion.

Serious adverse events (SAEs) were infrequent. In short‐term trials, suicidal ideation occurred in two Cobenfy‐treated individuals and one placebo‐treated individual. In long‐term studies, four individuals reported suicidal ideation, and there was one suicide attempt. In long‐term safety studies, only 5% of individuals reported SAEs, and of these SAEs, only 3% were psychiatric disorders, around 1% were infections, and the remainder of SAEs varied widely across system organ class and were likely unrelated to Cobenfy.

7. Limitations

Several important evidence gaps remain. First, the durability of treatment response remains incompletely characterized. All placebo‐controlled efficacy data are derived from 5‐week trials, and no randomized withdrawal or relapse‐prevention studies have been conducted. The 52‐week open‐label EMERGENT‐4 and EMERGENT‐5 studies suggest sustained improvement but lack a comparator and had substantial attrition (21.8% and 48.9% completion, respectively), leaving durable relapse prevention an ongoing question.

Second, the comparative effectiveness of Xanomeline/trospium versus existing antipsychotic treatments is unestablished. In a network meta‐analysis including 593 RCTs xanomeline/trospium ranked in the top third for efficacy among all antipsychotics for acute schizophrenia, while avoiding dopamine‐blocking adverse events (weight gain, EPS, prolactin elevation). The top‐ranked agents—clozapine, amisulpride, olanzapine, and risperidone—each outperformed at least three other antipsychotics with small‐to‐medium effect sizes. However, no head‐to‐head trials comparing Cobenfy to any established antipsychotic have been completed; thus, direct studies are needed to confirm whether this tolerability profile translates into long‐term effectiveness [17].

Third, real‐world adherence is an important consideration. Oral antipsychotic adherence is generally suboptimal, and Cobenfy's twice‐daily, fasting‐dependent dosing may add barriers compared with once‐daily agents. Early gastrointestinal effects are common and may drive early discontinuation. Fourth, the cost‐effectiveness remains uncertain as no cost‐per‐QALY analyses have been published. The absence of a long‐acting injectable formulation may limit its applicability for patients in whom adherence is a primary treatment challenge.

Together, these gaps do not diminish the significance of the EMERGENT program but underscore the need to better define Cobenfy's optimal role within the broader treatment landscape for schizophrenia.

8. Future Prospects

Cobenfy represents a mechanistic shift from traditional dopamine‐antagonist antipsychotics for the treatment of schizophrenia. Ongoing clinical trials are evaluating its potential use across broader populations, indications, and formulations. The phase 3 EMERGENT TEEN trial (NCT07288567) assesses the safety and efficacy of KarXT (Cobenfy) in adolescents (age 13–17) with schizophrenia [12]. For first‐episode and early‐phase schizophrenia, a phase 4 clinical trial (NCT06923891) evaluates 24‐week treatment satisfaction, efficacy, and tolerability [13]. In addition, a phase 1, open‐label study (NCT07061288) is evaluating safety, pharmacokinetic factors, and tolerability of a long‐acting injectable formulation of KarXT among participants with a primary diagnosis of schizophrenia [18]. These studies signal efforts for expanding population access and formulation options for Cobenfy. One medication with a similar mechanistic target under evaluation is emraclidine, a highly selective M4 muscarinic receptor positive allosteric modulator (PAM). Emraclidine has shown a favorable safety profile in early‐stage studies, but larger populations are needed for further evaluation [19]. Despite this tolerability, emraclidine failed to demonstrate efficacy in phase 2 EMPOWER trials, raising important questions whether selective M4 modulation alone is sufficient for antipsychotic effect, or whether xanomeline's combined M1/M4 agonism and dual orthosteric/allosteric interactions at M4 is needed for therapeutic activity [20].

Additionally, the identification of a potential muscarinic receptor‐deficit subgroup (25%–44% of patients) suggests biomarker‐guided targeting may eventually be possible, although this remains speculative [21]. Future research directions include controlled maintenance and head‐to‐head trials, real‐world effectiveness and adherence studies, clarifying M1 versus M4 contributions to symptom domains, and developing next‐generation agents with improved selectivity, efficacy, and tolerability.

Beyond schizophrenia, Cobenfy is being evaluated for the treatment of Alzheimer's disease‐associated agitation in two phase 3 trials, ADAGIO‐1 (NCT07011732) and ADAGIO‐2 (NCT07011745), and for acute manic episodes in bipolar I disorder in the phase 3 BALSAM‐2 trial (NCT06951711) [22, 23, 24]. Collectively, these pipelines position muscarinic agents as versatile therapeutics across psychiatric and neurocognitive disorders.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.