MADAM

• Molecular Formula: C16H20N2S
• Molecular Weight: 272.4 g/mol
• InChI: InChI=1S/C16H20N2S/c1-12-8-9-16(14(17)10-12)19-15-7-5-4-6-13(15)11-18(2)3/h4-10H,11,17H2,1-3H3
• InChIKey: UABIXNSHHIMZEP-UHFFFAOYSA-N

MADAM (CAS 411208-45-2), chemically N,N-dimethyl-2-(2-amino-4-methylphenylthio)benzylamine, belongs to the diphenyl sulfide class of serotonin transporter (SERT/5-HTT) ligands. It was developed as a positron emission tomography (PET) radiotracer for in vivo quantification of SERT density in the human brain.

MADAM (N,N-Dimethyl-2-(2-amino-4-methylphenylthio)benzylamine): A High-Selectivity Serotonin Transporter PET Radioligand – Procurement-Relevant Characteristics

MADAM (CAS 411208-45-2), chemically N,N-dimethyl-2-(2-amino-4-methylphenylthio)benzylamine, belongs to the diphenyl sulfide class of serotonin transporter (SERT/5-HTT) ligands. It was developed as a positron emission tomography (PET) radiotracer for in vivo quantification of SERT density in the human brain [1]. In vitro, [3H]MADAM binds to SERT with a dissociation constant (Kd) of 60 pM on rat cortical membranes, representing an affinity at least 2-fold higher than the most commonly used radioactive probes at the time of its characterization [2]. Its selectivity for SERT over the norepinephrine transporter (NET) and dopamine transporter (DAT) exceeds 1,000-fold [2]. The carbon-11 labeled analogue, [11C]MADAM, has been validated in non-human primates and humans, demonstrating high specific binding in SERT-rich regions such as the raphe nuclei, with the cerebellum confirmed as a suitable reference region for non-displaceable binding [1][3].

• [1] Lundberg J, Odano I, Olsson H, Halldin C, Farde L. Quantification of 11C-MADAM binding to the serotonin transporter in the human brain. J Nucl Med. 2005;46(9):1505-1515. View Source
• [2] Chalon S, Tarkiainen J, Garreau L, Hall H, Emond P, Vercouillie J, Farde L, Dasse P, Varnas K, Besnard JC, Halldin C, Guilloteau D. Pharmacological characterization of N,N-dimethyl-2-(2-amino-4-methylphenylthio)benzylamine as a ligand of the serotonin transporter with high affinity and selectivity. J Pharmacol Exp Ther. 2003;304(1):81-87. View Source
• [3] Halldin C, Guilloteau D, Tarkiainen J, et al. [11C]MADAM: a highly suitable radioligand for examination of the serotonin transporter with PET [abstract]. Eur J Nucl Med. 2001;28:973. View Source

Why Generic Substitution of SERT PET Radioligands Fails: MADAM's Structural and Pharmacological Differentiation

The diphenyl sulfide class of SERT radioligands—including [11C]DASB, [11C]AFM, [123I]ADAM, and [11C]MADAM—shares a common N,N-dimethyl-2-(arylthio)benzylamine scaffold, yet subtle aryl substitutions drive clinically relevant divergence in affinity, selectivity, nonspecific binding, and reference region suitability [1]. Direct comparative autoradiography has demonstrated that the Bmax cortex-to-cerebellum ratio—a critical parameter determining the validity of the cerebellum as a reference region for simplified quantitative models—differs substantially among these congeners (e.g., MADAM: 3–6; DASB: 3–4; citalopram: 17) [1]. Moreover, in vitro radioligand binding across 75 receptors, ion channels, and transporters confirmed that MADAM's selectivity profile is not interchangeable with other SERT probes [2]. These quantitative differences preclude simple generic substitution without re-validation of the entire PET quantification pipeline, as even structurally close analogs can produce different binding potential (BPND) estimates in the same brain regions under identical modeling conditions [1].

• [1] Elfving B, Madsen J, Knudsen GM. Neuroimaging of the serotonin reuptake site requires high-affinity ligands. Synapse. 2007;61(11):882-888. View Source
• [2] Larsen AK, Brennum LT, Egebjerg J, Sanchez C, Halldin C, Andersen PH. Selectivity of 3H-MADAM binding to 5-hydroxytryptamine transporters in vitro and in vivo in mice: correlation with behavioural effects. Br J Pharmacol. 2004;141(6):1015-1023. View Source

MADAM vs. Closest SERT Radioligand Analogs: Head-to-Head and Cross-Study Quantitative Evidence for Procurement Decisions

MADAM Demonstrates 2-Fold Higher Affinity for SERT Compared to Standard Radioactive Probes in Rat Cortical Membranes

In a direct in vitro comparison using rat cortical membrane preparations, [3H]MADAM exhibited a dissociation constant (Kd) of 60 pM, which is at least 2-fold better (i.e., lower Kd, higher affinity) than the most commonly used radioactive probes for SERT at the time of the study [1]. In competition assays within the same study, the SSRI paroxetine displayed a Ki of 0.32 nM (320 pM) and citalopram a Ki of 1.57 nM (1,570 pM) against [3H]MADAM binding, confirming that MADAM's intrinsic affinity substantially exceeds that of these reference inhibitors [1]. The Bmax in rat cortex was 543 fmol/mg protein [1].

Evidence Dimension	SERT binding affinity (Kd/Ki) in rat cortical membranes
Target Compound Data	[3H]MADAM Kd = 60 pM; Bmax = 543 fmol/mg protein
Comparator Or Baseline	Paroxetine Ki = 0.32 nM (320 pM); Citalopram Ki = 1.57 nM (1,570 pM); commonly used radioactive probes (Kd not explicitly stated but reported as at least 2-fold worse than MADAM)
Quantified Difference	MADAM Kd is 5.3-fold lower (higher affinity) than paroxetine Ki and 26-fold lower than citalopram Ki; reported as at least 2-fold better than standard probes
Conditions	Rat cortical membrane preparations; [3H]MADAM radioligand binding assay; competition studies with SSRIs; Chalon et al. 2003

Why This Matters

Higher affinity at the target enables lower administered mass and reduced pharmacologic occupancy, which is critical for achieving true tracer-dose conditions in PET imaging and avoiding pharmacological effects during scanning.

• [1] Chalon S, Tarkiainen J, Garreau L, Hall H, Emond P, Vercouillie J, Farde L, Dasse P, Varnas K, Besnard JC, Halldin C, Guilloteau D. Pharmacological characterization of N,N-dimethyl-2-(2-amino-4-methylphenylthio)benzylamine as a ligand of the serotonin transporter with high affinity and selectivity. J Pharmacol Exp Ther. 2003;304(1):81-87. View Source

MADAM Exhibits >1,000-Fold Selectivity for SERT Over NET and DAT, Surpassing the Selectivity Margins of Early-Generation SERT Radioligands

Competition binding studies in rat cortical membranes evaluated the affinity of MADAM at NET and DAT using selective inhibitors. Nisoxetine (NET-selective) exhibited a Ki of 270 nM, while desipramine and GBR 12935 (DAT-selective) each displayed Ki values greater than 1,000 nM (i.e., >1,000,000 pM) against [3H]MADAM binding [1]. This corresponds to a selectivity window of >1,000-fold for SERT over NET and DAT. In a broader profiling study, [3H]MADAM binding was tested across 75 different receptors, ion channels, and transporters, and binding was confirmed to be highly selective for SERT with no significant cross-reactivity [2]. By comparison, the earlier tracer [11C]McN 5652 is known to exhibit measurable affinity for NET and DAT, limiting its specificity in regions with mixed transporter expression [3].

Evidence Dimension	Selectivity ratio: SERT affinity vs. NET and DAT affinity
Target Compound Data	MADAM: nisoxetine Ki = 270 nM; desipramine Ki >1,000 nM; GBR 12935 Ki >1,000 nM; selectivity ratio >1,000-fold for SERT
Comparator Or Baseline	[11C]McN 5652 (earlier SERT radioligand with reported NET/DAT cross-reactivity; no quantitative selectivity ratio available from same study but acknowledged as less selective in the literature [3])
Quantified Difference	MADAM SERT-to-NET selectivity >1,000-fold; McN 5652 selectivity is substantially lower (class-level inference based on in vitro profiles)
Conditions	Rat cortical membrane competition assays; Chalon et al. 2003; selectivity profiling across 75 targets by Larsen et al. 2004

Why This Matters

High selectivity minimizes confounding signal from off-target transporter binding in brain regions co-expressing NET or DAT, directly improving the accuracy of SERT quantification in PET studies, particularly in cortical and limbic areas where transporter distributions overlap.

• [1] Chalon S, Tarkiainen J, Garreau L, Hall H, Emond P, Vercouillie J, Farde L, Dasse P, Varnas K, Besnard JC, Halldin C, Guilloteau D. Pharmacological characterization of N,N-dimethyl-2-(2-amino-4-methylphenylthio)benzylamine as a ligand of the serotonin transporter with high affinity and selectivity. J Pharmacol Exp Ther. 2003;304(1):81-87. View Source
• [2] Larsen AK, Brennum LT, Egebjerg J, Sanchez C, Halldin C, Andersen PH. Selectivity of 3H-MADAM binding to 5-hydroxytryptamine transporters in vitro and in vivo in mice: correlation with behavioural effects. Br J Pharmacol. 2004;141(6):1015-1023. View Source
• [3] Huang Y, Hwang DR, Narendran R, Sudo Y, Chatterjee R, Bae SA, Mawlawi O, Kegeles LS, Wilson AA, Kung HF, Laruelle M. Comparative evaluation in nonhuman primates of five PET radiotracers for imaging the serotonin transporters: [11C]McN 5652, [11C]ADAM, [11C]DASB, [11C]DAPA, and [11C]AFM. J Cereb Blood Flow Metab. 2002;22(11):1377-1398. View Source

MADAM Provides a More Favorable Cerebellum Reference Region Profile Than [3H]-(S)-Citalopram, With Quantitatively Defined Bmax Cortex-to-Cerebellum Ratios

In a systematic comparison of seven selective serotonin reuptake inhibitors as SERT radioligands, Elfving et al. (2007) determined the Bmax cortex-to-cerebellum ratio—a key parameter for reference tissue model validity—for [3H]MADAM, [11C]DASB, and [3H]-(S)-citalopram in both rat and monkey brain membranes [1]. In monkey cortex, [3H]-(S)-citalopram demonstrated a Bmax cortex:cerebellum ratio of 17, indicating substantial specific binding in the reference region that would compromise its use as a true reference. In contrast, [3H]MADAM exhibited ratios of 3 (monkey) and 6 (rat), and [11C]DASB showed ratios of 4 (monkey) and 3 (rat) [1]. The lower ratio for MADAM and DASB indicates lower specific SERT binding in the cerebellum, supporting its use as a reference region. However, the differential ratio between MADAM and DASB (3 vs. 4 in monkey; 6 vs. 3 in rat) suggests species-dependent differences in nonspecific binding that may influence the choice of tracer for cross-species translational studies.

Evidence Dimension	Bmax cortex-to-cerebellum ratio (indicator of reference region suitability)
Target Compound Data	[3H]MADAM: monkey ratio = 3, rat ratio = 6
Comparator Or Baseline	[3H]-(S)-Citalopram: monkey ratio = 17; [11C]DASB: monkey ratio = 4, rat ratio = 3
Quantified Difference	MADAM vs. citalopram: 5.7-fold lower ratio in monkey; MADAM vs. DASB: 1.3-fold lower ratio in monkey, 2-fold higher in rat
Conditions	In vitro membrane binding assays using rat cerebrum/cerebellum and monkey cortex/cerebellum; Elfving et al. 2007

Why This Matters

A lower Bmax cortex:cerebellum ratio directly translates to a more valid reference region for simplified quantitative models, reducing bias in BPND estimates and enabling less invasive PET protocols without arterial blood sampling. This is a decisive factor for clinical research centers selecting a SERT radioligand.

• [1] Elfving B, Madsen J, Knudsen GM. Neuroimaging of the serotonin reuptake site requires high-affinity ligands. Synapse. 2007;61(11):882-888. View Source

Test-Retest Reproducibility of [11C]MADAM SERT Binding Potential Ranges from Good to Excellent Across Cortical and Subcortical Regions, Supporting Longitudinal Study Designs

Lundberg et al. (2006) assessed the test-retest reproducibility of [11C]MADAM PET measurements of 5-HTT binding potential (BPND) in healthy human subjects (n=9, repeated scans). Using the simplified reference tissue model with cerebellum as reference, the intraclass correlation coefficients (ICC) ranged from 0.75 to 0.93 across cortical, subcortical, and brain stem regions, and the absolute test-retest variability ranged from 8.5% to 14.2% [1]. For comparison, [11C]DASB test-retest studies in humans have reported ICC values generally in the range of 0.65–0.89 and variability of 10–16% across similar regions [2]. While these values are from independent studies and not a direct within-study comparison, the cross-study evidence indicates that [11C]MADAM provides equivalent or marginally superior reproducibility, which is critical for detecting longitudinal changes in SERT density in clinical trials of antidepressants or monitoring disease progression.

Evidence Dimension	Test-retest reproducibility of SERT BPND (ICC and % variability)
Target Compound Data	[11C]MADAM: ICC = 0.75–0.93; variability = 8.5–14.2%
Comparator Or Baseline	[11C]DASB: ICC ≈ 0.65–0.89; variability ≈ 10–16% (literature values from independent studies)
Quantified Difference	MADAM ICC marginally higher and variability marginally lower than DASB, though overlapping ranges; statistical significance of difference not established due to between-study design
Conditions	Human PET imaging; [11C]MADAM test-retest with 9 subjects; simplified reference tissue model; Lundberg et al. 2006; DASB data from published literature

Why This Matters

A radioligand with good-to-excellent test-retest reproducibility reduces the sample size required to detect a given effect size in longitudinal or treatment studies, thereby lowering study costs and increasing statistical power for procurement-justified research.

• [1] Lundberg J, Halldin C, Farde L. Measurement of serotonin transporter binding with PET and [11C]MADAM: a test-retest reproducibility study. Synapse. 2006;60(4):256-263. View Source
• [2] Frankle WG, Huang Y, Hwang DR, et al. Comparison of serotonin transporter availability in healthy controls and patients with MDD using [11C]DASB PET. Biol Psychiatry. 2004;55(8):817-823. (Provides DASB reproducibility context as comparator.) View Source

[11C]MADAM Metabolism Profile Produces Predominantly Hydrophilic Radiometabolites That Do Not Cross the Blood-Brain Barrier, Simplifying Input Function Modeling

In human PET studies, the fraction of unchanged [11C]MADAM in plasma at 50 minutes post-injection was approximately 40%, with the remaining radioactivity attributable to labeled metabolites that were more polar than the parent compound as determined by reverse-phase HPLC [1]. This metabolic profile is favorable because the polar radiometabolites are not expected to cross the blood-brain barrier, eliminating the need to correct for brain-penetrant metabolites in kinetic modeling. A dedicated radiometabolite study by Arakawa et al. (2014) confirmed that [11C]MADAM is rapidly metabolized in both human and non-human primate plasma to hydrophilic species [2]. In contrast, the early-generation tracer [11C]McN 5652 produces a lipophilic radiometabolite that crosses the blood-brain barrier, requiring more complex metabolite correction algorithms and introducing additional variability in outcome measures [3].

Evidence Dimension	Fraction of unchanged parent at 50 min post-injection; metabolite lipophilicity and BBB permeability
Target Compound Data	[11C]MADAM: ~40% unchanged at 50 min; metabolites are more polar (hydrophilic) than parent; no evidence of brain-penetrant metabolites
Comparator Or Baseline	[11C]McN 5652: produces a lipophilic radiometabolite that crosses the BBB (class-level evidence from literature [3]); [11C]DASB: similar favorable metabolite profile (hydrophilic metabolites) reported in literature but not directly compared in same study
Quantified Difference	MADAM avoids lipophilic brain-penetrant metabolites, in contrast to McN 5652; insufficient data for precise quantification of the impact on kinetic modeling accuracy
Conditions	Human PET with HPLC metabolite analysis; Lundberg et al. 2005; Arakawa et al. 2014

Why This Matters

The absence of brain-penetrant radiometabolites simplifies the arterial input function correction, reduces modeling complexity, and increases the reliability of the derived binding parameters. This is a practical procurement consideration for PET centers seeking to minimize technical failure rates in clinical studies.

• [1] Lundberg J, Odano I, Olsson H, Halldin C, Farde L. Quantification of 11C-MADAM binding to the serotonin transporter in the human brain. J Nucl Med. 2005;46(9):1505-1515. View Source
• [2] Arakawa R, Stenkrona P, Takano A, et al. A radiometabolite study of the serotonin transporter PET radioligand [11C]MADAM. Nucl Med Biol. 2014;41(4):362-367. View Source
• [3] Szabo Z, Kao PF, Scheffel U, et al. Positron emission tomography imaging of serotonin transporters in the human brain using [11C](+)McN5652. Synapse. 1995;20(1):37-43. (Reference for McN 5652 metabolite properties.) View Source

[11C]MADAM and [11C]DASB Yield Concordant SERT BPND Estimates in Separate Parkinson's Disease Cohorts, Demonstrating Consistent Performance Across Tracers in Clinical Populations

A 2025 study by Oh et al. assessed SERT availability in non-depressed patients with early-stage Parkinson's disease (PD) using two independent cohorts: one imaged with [11C]MADAM (n=17 PD, 20 controls) at Karolinska Institutet, and the other with [11C]DASB (n=15 PD, 8 controls) at the University of British Columbia [1]. Both tracers yielded consistent results: there were no significant differences between controls and early-stage PD patients regardless of the tracer used. The Cohen d effect sizes for group differences were comparable between the two tracers, ranging from 0.16–0.40 (brain stem), 0.19–0.55 (subcortical), and 0.32–0.61 (cortical) for [11C]MADAM, versus 0.34–0.86, 0.09–0.61, and 0.28–0.70 for [11C]DASB, respectively [1]. Logan BPND and simplified reference tissue model BPND were highly correlated for both tracers (P < 0.001) [1]. This cross-tracer concordance supports the interchangeability of [11C]MADAM and [11C]DASB for multicenter trials when identical tracer supply is not feasible, provided that each site validates its own reference region and kinetic model.

Evidence Dimension	Cohen d effect sizes for PD vs. control group differences; BPND correlation between Logan and SRTM
Target Compound Data	[11C]MADAM: Cohen d = 0.16–0.40 (brain stem), 0.19–0.55 (subcortical), 0.32–0.61 (cortical); Logan-SRTM correlation P < 0.001
Comparator Or Baseline	[11C]DASB: Cohen d = 0.34–0.86 (brain stem), 0.09–0.61 (subcortical), 0.28–0.70 (cortical); Logan-SRTM correlation P < 0.001
Quantified Difference	Effect sizes similar between tracers; no statistically significant tracer × group interaction; both tracers led to identical conclusion of preserved SERT in non-depressed PD
Conditions	Human PET in two independent PD cohorts; identical scanning protocol (high-resolution research tomograph); cerebellum reference; Oh et al. 2025

Why This Matters

For multicenter clinical trials or procurement scenarios where a single tracer cannot be supplied to all sites, the demonstrated concordance between [11C]MADAM and [11C]DASB justifies selection of either tracer with confidence in result harmonization, facilitating flexible procurement strategies.

• [1] Oh M, Brumberg J, Sossi V, Varrone A. Preserved serotonin transporter availability in Parkinson disease measured with either [11C]MADAM or [11C]DASB: a study including 2 separate cohorts of nondepressed patients. J Nucl Med. 2025;66(2):309-314. View Source

Optimal Procurement and Application Scenarios for MADAM Based on Quantitative Differentiation Evidence

Longitudinal Clinical Trials of Antidepressant SERT Occupancy Requiring High Test-Retest Reproducibility

When designing randomized controlled trials that measure SERT occupancy by SSRIs or novel antidepressants over multiple time points, [11C]MADAM offers test-retest ICC values of 0.75–0.93 and absolute variability of 8.5–14.2% [1]. This level of reproducibility enables the detection of within-subject occupancy changes with smaller sample sizes, making MADAM a cost-effective choice for pharmaceutical companies and academic centers conducting longitudinal pharmacodynamic studies. The simplified reference tissue model using the cerebellum as reference region further reduces the technical burden and risk of arterial line complications.

Multicenter Neuroimaging Studies Requiring Cross-Site Tracer Harmonization

For consortia that cannot standardize on a single SERT radioligand across all sites, the empirically demonstrated concordance between [11C]MADAM and [11C]DASB in clinical populations (comparable Cohen d effect sizes, highly correlated BPND values, identical clinical conclusions) [2] provides a scientifically justified basis for pooling data from sites using either tracer. This flexibility is particularly relevant for large-scale studies in neuropsychiatric disorders where local radiochemistry infrastructure may favor one tracer over the other.

Preclinical Translational Studies Requiring Consistent SERT Quantification Across Species

The availability of Bmax cortex-to-cerebellum ratios for [3H]MADAM in both rat (ratio = 6) and monkey (ratio = 3) [3] supports cross-species translational experimental designs. Researchers can plan rodent studies with [3H]MADAM autoradiography and non-human primate or human PET studies with [11C]MADAM using consistent reference region assumptions, reducing methodological variability when translating preclinical findings to clinical applications.

Studies in Brain Regions With Low SERT Density Where Selectivity Against NET/DAT Is Critical

In cortical and limbic regions where SERT density is moderate and NET/DAT are co-expressed, the >1,000-fold selectivity of MADAM for SERT over NET and DAT [4] minimizes the contribution of off-target binding to the measured PET signal. This makes MADAM a preferred choice for investigations of SERT in the prefrontal cortex, anterior cingulate, and amygdala—regions frequently implicated in depression and anxiety disorders—where a less selective tracer would overestimate SERT availability due to NET/DAT cross-binding.

• [1] Lundberg J, Halldin C, Farde L. Measurement of serotonin transporter binding with PET and [11C]MADAM: a test-retest reproducibility study. Synapse. 2006;60(4):256-263. View Source
• [2] Oh M, Brumberg J, Sossi V, Varrone A. Preserved serotonin transporter availability in Parkinson disease measured with either [11C]MADAM or [11C]DASB: a study including 2 separate cohorts of nondepressed patients. J Nucl Med. 2025;66(2):309-314. View Source
• [3] Elfving B, Madsen J, Knudsen GM. Neuroimaging of the serotonin reuptake site requires high-affinity ligands. Synapse. 2007;61(11):882-888. View Source
• [4] Chalon S, Tarkiainen J, Garreau L, Hall H, Emond P, Vercouillie J, Farde L, Dasse P, Varnas K, Besnard JC, Halldin C, Guilloteau D. Pharmacological characterization of N,N-dimethyl-2-(2-amino-4-methylphenylthio)benzylamine as a ligand of the serotonin transporter with high affinity and selectivity. J Pharmacol Exp Ther. 2003;304(1):81-87. View Source