Monlunabant

• Molecular Formula: C26H22ClF3N6O3S
• Molecular Weight: 591.0 g/mol
• InChI: InChI=1S/C26H22ClF3N6O3S/c1-16(37)32-24(31)33-25(35-40(38,39)21-13-9-19(10-14-21)26(28,29)30)36-15-22(17-5-3-2-4-6-17)23(34-36)18-7-11-20(27)12-8-18/h2-14,22H,15H2,1H3,(H3,31,32,33,35,37)/t22-/m1/s1
• InChIKey: GYJPQNPVIJXXTA-JOCHJYFZSA-N

Monlunabant Procurement Guide

Monlunabant (INV-202, (S)-MRI-1891) is a small-molecule, orally active cannabinoid CB1 receptor inverse agonist/antagonist with subnanomolar binding affinity (hCB1 Ki = 0.3 nM) and >2,000-fold selectivity over CB2 [1]. Discovered at the NIH as a β-arrestin-2-biased ligand, it is a structurally modified derivative of ibipinabant designed to minimize blood-brain barrier penetration (brain/plasma ratio 7% vs ibipinabant 22%) [1]. Currently in Phase 2 clinical development by Novo Nordisk for obesity and metabolic syndrome, monlunabant represents the most clinically advanced peripherally selective CB1 antagonist with functional signaling bias [2].

CB1 receptor pathway inhibition studies

β-arrestin-2 biased signaling research tool

Peripheral restriction context supports tissue-selective CB1 research

Clinical-stage reference compound (Phase 2 context)

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source
• [2] Novo Nordisk A/S. Monlunabant phase 2a trial in obesity successfully completed. Company Announcement, 20 September 2024. View Source

Monlunabant Differentiation from CB1 Antagonists

CB1 receptor antagonists share a common molecular target but differ critically in three dimensions that preclude simple interchange: (i) blood-brain barrier permeability (brain/plasma ratio ranging from 100% for rimonabant to 7% for monlunabant) [1]; (ii) functional signaling bias (monlunabant shows 286-fold βArr2/G-protein bias vs unbiased rimonabant) [1]; and (iii) receptor binding affinity and selectivity (Ki: monlunabant 0.3 nM, ibipinabant 7.8 nM, rimonabant 1.8 nM; CB1/CB2 selectivity: >2,000-fold vs ~1,000-fold for ibipinabant) [1]. First-generation drugs like rimonabant were withdrawn globally due to neuropsychiatric toxicity mediated by central G-protein signaling blockade; monlunabant's combined peripheral restriction and βArr2 bias are mechanistically designed to decouple metabolic benefit from CNS adverse effects—a profile not achievable with earlier CB1 antagonists [2].

Brain penetration profile

Peripheral restriction design differs markedly from brain-penetrant CB1 antagonists; may shift CNS vs. peripheral target engagement.

Functional signaling bias

β-arrestin-2-preferential antagonism is not shared by unbiased CB1 blockers; signaling outcomes may not replicate across tool compounds.

Affinity and selectivity context

CB1 binding affinity and isoform selectivity relative to ibipinabant/rimonabant require re-verification; direct substitution may confound assay interpretation.

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source
• [2] Sayburn A. European drugs agency withdraws antiobesity drug. BMJ. 2008;337:a2301. doi:10.1136/bmj.a2301. View Source

Monlunabant Quantitative Evidence

CB1 Binding Affinity and Selectivity

Monlunabant binds human CB1 with a Ki of 0.3 nM, yielding >2,000-fold selectivity over CB2 (hCB2 Ki = 613 nM) [1]. The parent compound ibipinabant has a hCB1 Ki of 7.8 nM and ~1,000-fold selectivity (hCB2 Ki = 7,943 nM) . Rimonabant, the first-generation benchmark, has a hCB1 Ki of 1.8 nM . Thus, monlunabant is 26-fold more potent than ibipinabant at CB1, 6-fold more potent than rimonabant, and achieves 2-fold greater CB1/CB2 selectivity than ibipinabant.

CB1 Affinity & Selectivity

Reported

hCB1 Ki = 0.3 nM CB1/CB2 selectivity >2,000-fold

Supports CB1 pathway inhibition with reduced CB2 off-target context

Radioligand displacement, cross-study comparable

Evidence Dimension	CB1 binding affinity (Ki) and CB1/CB2 selectivity
Target Compound Data	hCB1 Ki = 0.3 nM; hCB2 Ki = 613 nM; selectivity >2,000-fold
Comparator Or Baseline	Ibipinabant: hCB1 Ki = 7.8 nM, CB1/CB2 selectivity ~1,000-fold; Rimonabant: hCB1 Ki = 1.8 nM
Quantified Difference	26-fold more potent than ibipinabant; 6-fold more potent than rimonabant; 2-fold greater selectivity than ibipinabant
Conditions	Radioligand displacement using [3H]CP55,940 on human recombinant CB1 and CB2 receptors (Liu et al., 2021 for monlunabant; independent publications for ibipinabant and rimonabant)

Why This Matters

Higher affinity with subnanomolar potency reduces compound mass per experiment and minimizes target occupancy variability; superior selectivity lowers CB2-mediated off-target liability in tissues co-expressing CB2 (e.g., immune cells), directly improving assay signal-to-noise ratios and in vivo dosing accuracy.

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source

β-Arrestin-2 Biased Signaling Profile

Monlunabant (MRI-1891) inhibits CB1R-agonist-induced β-arrestin-2 (βArr2) recruitment with an IC50 of 21 pM, while its inhibition of G-protein activation (GTPγS binding) requires an IC50 of 6 nM—a 286-fold functional bias toward βArr2 blockade [1]. By contrast, rimonabant tested under identical conditions is nonbiased, showing no preferential inhibition of either pathway [1]. This biased antagonism is attributed to monlunabant's unique interaction with N-terminal loop residue F108 and transmembrane helix-1 residue S123 on CB1R, which rimonabant does not engage [1].

β-Arrestin-2 Bias

Head-to-head

Monlunabant: βArr2 IC50 21 pM, G-protein IC50 6 nM → 286-fold bias Rimonabant: nonbiased (ratio ~1)

Enables selective βArr2 pathway interrogation; G-protein-mediated CNS signaling context reduced

HEK293-hCB1R; CP-55,940 EC80, F108/S123 engagement

Evidence Dimension	Functional signaling bias (βArr2 recruitment vs G-protein activation)
Target Compound Data	βArr2 IC50 = 21 pM; G-protein (GTPγS) IC50 = 6 nM; bias ratio = 286
Comparator Or Baseline	Rimonabant: nonbiased (no preferential pathway inhibition)
Quantified Difference	286-fold bias for monlunabant; rimonabant shows no bias (ratio ~1)
Conditions	CB1R-agonist CP-55,940 at EC80 concentration; βArr2 recruitment and GTPγS binding assays in HEK293 cells expressing hCB1R (Liu et al., 2021)

Why This Matters

The βArr2 bias enables selective investigation of arrestin-dependent CB1R pathways (e.g., muscle insulin resistance) without engaging G-protein-mediated CNS effects—impossible with unbiased antagonists. This functional selectivity is the molecular basis for monlunabant's reduced anxiety liability and directly informs tissue-specific experimental design.

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source

Brain Penetration and Peripheral Restriction

Following acute oral administration at 3 mg/kg in male wild-type mice, monlunabant (MRI-1891) achieved a brain/plasma concentration ratio of 7%, compared to 22% for its structural parent ibipinabant and approximately 100% for the brain-penetrant comparator rimonabant [1]. The 3.1-fold reduction in brain partitioning relative to ibipinabant was achieved through increased total polar surface area and hydrogen bonding capacity introduced during structural optimization [1]. Brain CB1R occupancy measured by PET was not significant at 1 or 10 mg/kg monlunabant, whereas 3 mg/kg rimonabant occupied brain CB1R substantially [1].

Brain Penetration

Head-to-head

Monlunabant: 7% brain/plasma Ibipinabant: 22% | Rimonabant: ~100%

Supports peripheral CB1R research; CNS exposure minimized

3 mg/kg p.o. mice; PET occupancy at 1–10 mg/kg

Evidence Dimension	Brain/plasma concentration ratio
Target Compound Data	7% brain/plasma ratio
Comparator Or Baseline	Ibipinabant: 22%; Rimonabant: ~100%
Quantified Difference	3.1-fold lower than ibipinabant; >14-fold lower than rimonabant
Conditions	Acute oral dose of 3 mg/kg in male wild-type C57BL/6J mice; brain and plasma sampled post-dose; brain CB1R occupancy by [18F]FMPEP-d2 PET (Liu et al., 2021)

Why This Matters

Reduced brain exposure directly correlates with lower brain CB1R occupancy and diminished anxiogenic liability, enabling researchers to isolate peripheral CB1R pharmacology in adipose, liver, kidney, and muscle without the confounding central effects that led to rimonabant's market withdrawal.

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source

Clinical Weight Loss and Safety Profile

In a 16-week Phase 2a randomized controlled trial (N=243; baseline body weight 110.1 kg), monlunabant 10 mg once daily produced a placebo-adjusted weight loss of 6.4 kg (LSM difference, 95% CI -8.0 to -4.9), equivalent to ~5.8% placebo-adjusted reduction [1][2]. This weight loss magnitude aligns with the ~6% placebo-adjusted benchmark historically reported for rimonabant 20 mg at similar time points in the Phase 3 RIO program [3]. Higher monlunabant doses (20 mg: -6.9 kg; 50 mg: -8.0 kg LSM difference) provided limited incremental benefit, suggesting effective peripheral CB1R saturation at 10 mg [1]. Mild-to-moderate neuropsychiatric adverse events (anxiety, irritability, sleep disturbances) occurred with dose-dependent frequency but no serious neuropsychiatric SAEs were reported [2]. In contrast, rimonabant was withdrawn globally by the EMA in 2008 due to serious psychiatric adverse events including suicidal ideation, with a doubling of psychiatric disorder risk vs placebo in post-marketing surveillance [4].

Weight Loss Endpoint

Cross-study

10 mg: −6.4 kg placebo-adjusted (95% CI −8.0, −4.9) at 16 wk Rimonabant benchmark: ~6% weight loss (RIO program)

Reported placebo-adjusted weight loss endpoint context; class benchmark alignment

Phase 2a RCT N=243; no serious neuropsychiatric SAEs

Evidence Dimension	Placebo-adjusted body weight change at 16 weeks
Target Compound Data	10 mg: -6.4 kg (LSM, 95% CI -8.0 to -4.9); 20 mg: -6.9 kg; 50 mg: -8.0 kg
Comparator Or Baseline	Rimonabant 20 mg: ~6% placebo-adjusted weight loss at ~16 weeks (cross-study, RIO program); Placebo: -0.7 kg
Quantified Difference	Equivalent weight loss efficacy at 10 mg vs rimonabant benchmark; no serious neuropsychiatric SAEs vs rimonabant withdrawn for suicidality
Conditions	16-week Phase 2a RCT (NCT05891834); adults with obesity and metabolic syndrome; 25 centers in Canada; Knop et al., Lancet Diabetes Endocrinol 2025

Why This Matters

Monlunabant delivers efficacy comparable to the only clinically validated CB1 antagonist benchmark while demonstrating an improved safety profile—a decisive factor for procurement in obesity research programs where neuropsychiatric risk has historically precluded CB1-targeted approaches.

• [1] Knop FK, et al. Efficacy and safety of monlunabant in adults with obesity and metabolic syndrome: a double-blind, randomised, placebo-controlled, phase 2a trial. Lancet Diabetes Endocrinol. 2025;13(11):911-923. doi:10.1016/S2213-8587(25)00216-5. View Source
• [2] Novo Nordisk A/S. Monlunabant phase 2a trial in obesity successfully completed. Company Announcement, 20 September 2024. View Source
• [3] Skye Bioscience. Skye Bioscience presents PK, PD model on CB1 inhibitors at ECO 2025. Cited via TipRanks, 13 May 2025. Monlunabant 6% placebo-adjusted weight loss aligns with rimonabant historical benchmark. View Source
• [4] Sayburn A. European drugs agency withdraws antiobesity drug. BMJ. 2008;337:a2301. doi:10.1136/bmj.a2301. View Source

Anxiogenic Liability Assessment

In obese wild-type C57BL/6J mice, monlunabant (MRI-1891) at doses up to 30 mg/kg did not elicit anxiogenic behavior in the elevated plus maze, even at 30 mg/kg—a dose producing partial brain CB1R occupancy [1]. By contrast, rimonabant at only 3 mg/kg produced significant brain CB1R occupancy and was highly anxiogenic [1]. In βArr2-knockout mice, rimonabant remained anxiogenic, confirming that anxiety-like behaviors from brain CB1R blockade are mediated entirely via G-protein signaling, a pathway monlunabant spares due to its 286-fold βArr2 bias [1]. The no-observed-anxiogenic-effect dose of monlunabant (30 mg/kg) is 10-fold higher than the anxiogenic dose of rimonabant (3 mg/kg), representing a substantial preclinical safety margin [1].

Anxiogenic Liability

Head-to-head

Monlunabant: no anxiety up to 30 mg/kg Rimonabant: anxiogenic at 3 mg/kg

Reported anxiety endpoint context; ≥10-fold margin in preclinical model

Elevated plus maze, wild-type & βArr2-KO mice, PET occupancy

Evidence Dimension	Anxiogenic behavior induction (elevated plus maze)
Target Compound Data	No anxiogenic effect at up to 30 mg/kg (p.o.), despite partial brain CB1R occupancy at highest dose
Comparator Or Baseline	Rimonabant: significant anxiogenic effect at 3 mg/kg (p.o.), with significant brain CB1R occupancy
Quantified Difference	≥10-fold higher tolerated dose without anxiety; rimonabant anxiogenic at 3 mg/kg; behavioral effect attributable to G-protein pathway engagement absent in monlunabant
Conditions	Obese wild-type and βArr2-KO C57BL/6J mice; elevated plus maze; acute oral dosing; brain CB1R occupancy by [18F]FMPEP-d2 PET (Liu et al., 2021)

Why This Matters

This direct preclinical comparison demonstrates that monlunabant's biased signaling translates into a quantifiable and substantial reduction in anxiety liability—the primary toxicity that caused rimonabant's global withdrawal. For researchers conducting chronic metabolic studies or neuropsychiatric safety screening, this provides a mechanistically validated margin of CNS safety.

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source

Monlunabant Research and Procurement Applications

Peripheral CB1R Pharmacology and Metabolic Tissue Studies

Monlunabant's 7% brain/plasma ratio—3.1-fold lower than ibipinabant's 22%—enables investigation of CB1R function in peripheral metabolic tissues (adipose, liver, kidney, skeletal muscle, gastrointestinal tract) without confounds introduced by central CB1R blockade [1]. For researchers studying tissue-specific CB1R signaling in obesity, insulin resistance, or diabetic nephropathy, monlunabant provides a >14-fold window over rimonabant for peripheral selectivity, making it the tool compound of choice for dissecting metabolic benefits from CNS-mediated effects [1].

β-Arrestin-2 Biased Signaling and Insulin Resistance

With a 286-fold βArr2 preferential inhibition (IC50 21 pM) over G-protein signaling (IC50 6 nM), monlunabant uniquely enables selective interrogation of β-arrestin-dependent CB1R pathways [1]. This biased profile directly addresses obesity-induced muscle insulin resistance, which proceeds through βArr2 signaling. Monlunabant, but not the unbiased comparator rimonabant, improves insulin-induced Akt-2 phosphorylation in C2C12 myoblasts and reverses muscle insulin resistance in wild-type mice, effects lost in βArr2-knockout animals—confirming pathway specificity [1]. Researchers studying biased GPCR signaling will find monlunabant a validated probe for βArr2-dependent CB1R pharmacology.

Obesity Pharmacotherapy Development and Reference Benchmarking

Monlunabant's Phase 2a clinical data—6.4 kg placebo-adjusted weight loss at 16 weeks (10 mg)—provide a validated efficacy benchmark for preclinical obesity models and next-generation CB1 antagonist development [2][3]. The concentration-effect relationship established by the 10-50 mg dose range, combined with documented neuropsychiatric safety data (no serious SAEs vs rimonabant's withdrawal for suicidality), positions monlunabant as the optimal reference compound for obesity programs seeking to demonstrate differentiation from first-generation CB1 antagonists [4]. Its oral bioavailability and once-daily dosing further support translational studies.

Neuropsychiatric Safety Screening and Toxicology Reference

Monlunabant provides a uniquely characterized reference standard for CB1 antagonist neuropsychiatric safety profiling. In direct head-to-head comparison, monlunabant at 30 mg/kg produced no anxiogenic effect even with partial brain CB1R occupancy, while rimonabant at 3 mg/kg was highly anxiogenic—a ≥10-fold safety margin underpinned by mechanistic βArr2 bias [1]. For toxicology programs screening novel CB1-targeting candidates, monlunabant offers a quantitative benchmark for acceptable anxiety liability, with matched pharmacokinetic (brain/plasma ratio 7%), pharmacodynamic (PET-measured CB1R occupancy), and behavioral data all from a single integrated study [1].

Application

Selection Property

Validation Focus

Peripheral CB1R metabolic tissue studies

Peripheral selectivity profile

Tissue-specific CB1R occupancy and metabolic endpoints

β-Arrestin-2 biased signaling research

Biased signaling assay context

βArr2 vs. G-protein pathway discrimination

Obesity pharmacotherapy model benchmarking

Reported weight-loss endpoint context

Placebo-adjusted body weight change and tolerability endpoints

Neuropsychiatric safety profiling reference

Anxiogenic margin context

Elevated plus maze behavior and CB1R occupancy

• [1] Liu Z, Iyer MR, Godlewski G, et al. Functional Selectivity of a Biased Cannabinoid-1 Receptor (CB1R) Antagonist. ACS Pharmacol Transl Sci. 2021;4(3):1175-1187. doi:10.1021/acsptsci.1c00048. View Source
• [2] Knop FK, et al. Efficacy and safety of monlunabant in adults with obesity and metabolic syndrome: a double-blind, randomised, placebo-controlled, phase 2a trial. Lancet Diabetes Endocrinol. 2025;13(11):911-923. doi:10.1016/S2213-8587(25)00216-5. View Source
• [3] Novo Nordisk A/S. Monlunabant phase 2a trial in obesity successfully completed. Company Announcement, 20 September 2024. View Source
• [4] Skye Bioscience. Skye Bioscience presents PK, PD model on CB1 inhibitors at ECO 2025. Cited via TipRanks, 13 May 2025. View Source