N-nitroso-atomoxetine

• Molecular Formula: C17H20N2O2
• Molecular Weight: 284.35 g/mol
• InChI: InChI=1S/C17H20N2O2/c1-14-8-6-7-11-16(14)21-17(12-13-19(2)18-20)15-9-4-3-5-10-15/h3-11,17H,12-13H2,1-2H3/t17-/m1/s1
• InChIKey: ZILIWANDBPOVTK-QGZVFWFLSA-N

N-Nitroso-Atomoxetine: Identity & Regulatory Class

N-Nitroso-atomoxetine (CAS 3028194-77-3; molecular formula C₁₇H₂₀N₂O₂; MW 284.35 g/mol) is the N-nitroso derivative of the secondary N-methylamine moiety of atomoxetine, a selective norepinephrine reuptake inhibitor (sNRI) used for ADHD treatment [1]. It belongs to the nitrosamine drug substance-related impurity (NDSRI) class, a sub-category of N-nitrosamine impurities that share structural similarity with the corresponding active pharmaceutical ingredient [2]. N-Nitroso-atomoxetine has been confirmed mutagenic in vitro (Ames assay) and in vivo (transgenic rodent mutation assay, TGR) [2], and is classified under ICH M7(R2) as a class 2 mutagenic impurity with structural alerts for genotoxicity [3]. It is supplied as a highly characterized reference standard by multiple vendors, typically at ≥95% HPLC purity, with full COA documentation (H-NMR, MASS, HPLC) compliant with USP, EMA, JP, and BP regulatory standards [1].

Type: Nitrosamine drug substance-related impurity (NDSRI) reference standard

Regulatory class: ICH M7 Class 2 mutagenic impurity with structural alerts

Documentation: Highly characterized; full COA with H-NMR, MASS, HPLC

• [1] Veeprho. N-Nitroso Atomoxetine | CAS 3028194-77-3 – Product Datasheet. Veeprho.com, 2022. Available at: https://veeprho.com/impurities/n-nitroso-atomoxetine/ View Source
• [2] Jolly RA, Cornwell PD, Noteboom J, Sayyed FB, Thapa B, Buckley LA. Estimation of acceptable daily intake values based on modeling and in vivo mutagenicity of NDSRIs of fluoxetine, duloxetine and atomoxetine. Regulatory Toxicology and Pharmacology, 2024;152:105672. DOI: 10.1016/j.yrtph.2024.105672 View Source
• [3] PMDA Japan. Safety Review Committee Discussion on N-Nitroso-Atomoxetine Risk Assessment (資料3-1), August 2024. Summarized on USP Nitrosamines Exchange. Available at: https://nitrosamines.usp.org/t/current-nitrosamine-landscape-in-japan/1626/123 View Source

N-Nitroso-Atomoxetine: Non-Substitutable Standard

Nitrosamine drug substance-related impurities exhibit compound-specific mutagenic potency, metabolic activation profiles, and regulatory acceptable intake (AI) limits that preclude interchangeable use in analytical method development, validation, or quality control. N-Nitroso-atomoxetine differs from its closest NDSRI analogs—including N-nitroso-fluoxetine, N-nitroso-duloxetine, and N-nitroso-desmethyl-atomoxetine—in its chromatographic retention behavior, mass spectrometric fragmentation pattern (MRM transitions specific to the atomoxetine scaffold), and critically, its regulatory AI assignment [1]. The EMA has set an AI of 100 ng/day for N-nitroso-atomoxetine via read-across, while the FDA assigns an AI of 26.5 ng/day under Category 1 [1]. These limits differ substantially from those of N-nitroso-fluoxetine and N-nitroso-duloxetine, which have distinct AI values based on their own toxicological datasets [2]. Furthermore, N-nitroso-atomoxetine's formation propensity is exquisitely sensitive to environmental NOx concentrations during API processing—a behavior documented quantitatively via grinding studies under controlled NOx atmospheres [3]—meaning that reference standard purity, storage stability, and handling protocols must be validated specifically for this compound. Generic substitution with an incorrect NDSRI standard introduces quantifiable risk of analytical inaccuracy, failed regulatory submission, or missed batch contamination.

Regulatory AI Compound-specific acceptable intake limits diverge between EMA and FDA and differ from other NDSRIs; generic substitution compromises method compliance.

Analytical specificity MRM transition (m/z 302→177) and chromatographic retention are scaffold-specific; methods validated for other NDSRIs are not transferable.

Formation sensitivity Nitrosation is highly sensitive to environmental NOx; documented multi-fold increases under manufacturing-relevant conditions require compound-validated handling.

• [1] USP Nitrosamines Exchange. FDA – Recommended Acceptable Intake Limits for NDSRIs Guidance for Industry (comparison of FDA vs EMA AI lists). 2023. Available at: https://nitrosamines.usp.org/t/fda-recommended-acceptable-intake-limits-for-ndsris-guidance-for-industry/7145/17 View Source
• [2] Jolly RA, Cornwell PD, Noteboom J, Sayyed FB, Thapa B, Buckley LA. Estimation of acceptable daily intake values based on modeling and in vivo mutagenicity of NDSRIs of fluoxetine, duloxetine and atomoxetine. Regulatory Toxicology and Pharmacology, 2024;152:105672. DOI: 10.1016/j.yrtph.2024.105672 View Source
• [3] PMDA Japan. Safety Review Committee Discussion (資料3-2: Nihon Iyakuhin Risk Assessment Report). August 2024. Summarized on USP Nitrosamines Exchange. Available at: https://nitrosamines.usp.org/t/current-nitrosamine-landscape-in-japan/1626/123 View Source

N-Nitroso-Atomoxetine: Quantitative Differentiation Evidence

Mutagenic Potency vs. NDMA and NDEA

N-Nitroso-atomoxetine, together with N-nitroso-fluoxetine and N-nitroso-duloxetine, was assessed via a weight-of-evidence approach combining read-across informed by physicochemical property modeling and quantum mechanical activation modeling. All three NDSRIs were 10–100-fold less potent compared with highly potent small-molecule nitrosamines NDMA and NDEA [1]. All three were confirmed mutagenic in the Ames assay (in vitro) and the TGR assay (in vivo), yet the in vivo mutation response potency corresponded to an AI of ≥4400 ng/day—an order of magnitude higher than the published regulatory AI limits for these NDSRIs (100 ng/day for N-nitroso-atomoxetine) [1]. This quantitative discrepancy—a ≥44-fold difference between the science-derived BMDL-based AI and the read-across regulatory AI—is specific to the atomoxetine NDSRI and its analogs, and is not observed for NDMA (AI = 96 ng/day, derived from rodent carcinogenicity data) or NDEA (AI = 26.5 ng/day) [2]. Note: individual compound-level TGR BMDL values for N-nitroso-atomoxetine were not disaggregated in the published study; the ≥4400 ng/day AI value is a class-level inference for the atomoxetine/fluoxetine/duloxetine NDSRI group.

Mutagenic potency vs. NDMA

Class-level

10–100-fold lower potency vs. NDMA/NDEA; ≥44-fold discrepancy between BMDL-based AI (≥4400 ng/day) and regulatory AI (100 ng/day)

Supports science-based AI derivation context vs. conservative read-across

Class-level inference; compound-specific BMDL not disaggregated

Evidence Dimension	In vivo mutagenic potency (TGR BMDL-derived AI) vs. regulatory AI
Target Compound Data	N-Nitroso-atomoxetine: TGR BMDL-derived AI ≥4400 ng/day (class-level); Regulatory AI (EMA) = 100 ng/day
Comparator Or Baseline	NDMA: Rodent carcinogenicity-derived AI = 96 ng/day; NDEA: AI = 26.5 ng/day; N-Nitroso-fluoxetine: 30–50× less mutagenic than NDMA in vivo (FDA SOT 2024 poster data)
Quantified Difference	≥44-fold discrepancy between BMDL-based AI (≥4400 ng/day) and regulatory AI (100 ng/day) for N-nitroso-atomoxetine; 10–100-fold lower potency than NDMA/NDEA
Conditions	Weight-of-evidence: in vitro Ames (Enhanced Ames Test with hamster S9), in vivo TGR (transgenic rodent mutation) assay; quantum mechanical modeling of C–N bond dissociation energy for diazonium metabolite

Why This Matters

Understanding the 10–100-fold potency differential relative to NDMA/NDEA and the ≥44-fold discrepancy between data-derived and regulatory AI values enables scientifically informed selection of N-nitroso-atomoxetine as a reference standard for risk assessment, rather than defaulting to conservative read-across limits that may overstate hazard.

• [1] Jolly RA, Cornwell PD, Noteboom J, Sayyed FB, Thapa B, Buckley LA. Estimation of acceptable daily intake values based on modeling and in vivo mutagenicity of NDSRIs of fluoxetine, duloxetine and atomoxetine. Regulatory Toxicology and Pharmacology, 2024;152:105672. DOI: 10.1016/j.yrtph.2024.105672 View Source
• [2] Li X, Guo X, et al. Mutagenicity and genotoxicity evaluation of 15 nitrosamine drug substance-related impurities in human TK6 cells. Regulatory Toxicology and Pharmacology, 2024;154:105730. DOI: 10.1016/j.yrtph.2024.105730 (NDMA/NDEA AI reference values cited within) View Source

Regulatory AI Divergence: EMA vs. FDA

A direct comparison of published regulatory AI lists reveals a striking divergence for N-nitroso-atomoxetine: the EMA assigns an AI of 100 ng/day based on read-across methodology, while the FDA assigns an AI of 26.5 ng/day under Category 1 [1]. This 3.8-fold inter-agency discrepancy is not observed uniformly across NDSRI analogs. For context, the PMDA Japan review ultimately adopted the 100 ng/day limit, rejecting Eli Lilly's proposed 4400 ng/day (BMDL-derived) and 1500 ng/day (non-Cohort of Concern) values on the grounds that current ICH M7 guidance does not permit setting an AI >1500 ng/day for Ames- and TGR-positive nitrosamines lacking carcinogenicity data [2]. The resulting regulatory landscape means N-nitroso-atomoxetine is subject to a more stringent and less harmonized AI framework than many other NDSRIs, directly impacting analytical sensitivity requirements (LOQ must be ≤10% of AI) and procurement specifications for reference standards.

Regulatory AI divergence

Head-to-head

EMA AI = 100 ng/day; FDA AI = 26.5 ng/day; 3.8-fold inter-agency difference

Requires method validation to the most stringent applicable AI limit

PMDA Japan adopted 100 ng/day; FDA Category 1

Evidence Dimension	Regulatory acceptable intake (AI) limit
Target Compound Data	N-Nitroso-atomoxetine: EMA AI = 100 ng/day; FDA AI = 26.5 ng/day (Category 1); PMDA Japan adopted = 100 ng/day
Comparator Or Baseline	N-Nitroso-fluoxetine: EMA AI listed separately (distinct SAR-based value); N-Nitroso-duloxetine: separate AI listing; NDMA: FDA AI = 96 ng/day; NDEA: FDA AI = 26.5 ng/day
Quantified Difference	3.8-fold divergence between EMA (100 ng/day) and FDA (26.5 ng/day) for the same compound; 26.5–100× lower than the BMDL-supported AI of ≥4400 ng/day
Conditions	Regulatory AI assignment methodologies: EMA via read-across (SAR approach); FDA via Category 1 classification; PMDA Japan via safety review committee deliberation (August 2024)

Why This Matters

This inter-agency AI divergence means procurement of a single N-nitroso-atomoxetine reference standard must support analytical method validation at LOQ levels compliant with the most stringent applicable limit (FDA 26.5 ng/day), imposing a sensitivity requirement that may exceed what is needed for other NDSRIs with higher or harmonized AI values.

• [1] USP Nitrosamines Exchange. FDA – Recommended Acceptable Intake Limits for NDSRIs Guidance for Industry. 2023 (AI list comparison uploaded by community member). Available at: https://nitrosamines.usp.org/t/fda-recommended-acceptable-intake-limits-for-ndsris-guidance-for-industry/7145/17 View Source
• [2] PMDA Japan. Safety Review Committee Discussion on N-Nitroso-Atomoxetine Risk Assessment (資料3-1), August 2024. Summarized on USP Nitrosamines Exchange. Available at: https://nitrosamines.usp.org/t/current-nitrosamine-landscape-in-japan/1626/123 View Source

Formation Sensitivity to Atmospheric NOx During API Grinding

N-Nitroso-atomoxetine displays extreme sensitivity to environmental nitrogen oxides (NOx) during API processing—a property documented quantitatively in the PMDA Japan risk assessment review. Under a laboratory atmosphere with NOx gas at 2 ppm, grinding atomoxetine API in an agate mortar increased N-nitroso-atomoxetine content from 54 ppb to 5247 ppb—a 97-fold increase [1]. In commercial manufacturing, 174 API lots showed N-nitroso-atomoxetine levels ranging from 15 ppb to 5749 ppb (against a limit of 729 ppb, equivalent to the 100 ng/day AI) [1]. Relocation of the grinding step from a site with measured NOx at 8.4 ppm (proximity to forklift exhaust) to a site with NOx <0.03 ppm reduced maximum contamination from 5749 ppb to a maximum of 222 ppb [1]. Independently, a three-pronged NOx-mitigation approach (NOx-free air for API, excipients, and film-coating processes) successfully suppressed N-nitroso-atomoxetine in finished tablets to 0.097 ppm—well below the regulatory NMT of 0.83 ppm [2]. This NOx-dependent formation behavior is mechanistically linked to the secondary N-methylamine moiety of atomoxetine and is not directly extrapolable to NDSRIs with different amine scaffolds, tertiary amine precursors, or differing nitrosation kinetics.

NOx formation sensitivity

Cross-study comparable

97-fold increase (54→5247 ppb) under 2 ppm NOx during API grinding

Supports process marker utility; informs standard handling protocols

Documented in PMDA Japan risk assessment

Evidence Dimension	N-Nitroso-atomoxetine formation as a function of atmospheric NOx concentration and processing step
Target Compound Data	N-Nitroso-atomoxetine: 54 ppb → 5247 ppb (agate mortar grinding, 2 ppm NOx); API process trajectory: intermediate 106 ppb → after purification 762 ppb → after grinding 3444 ppb; 174-lot API range: 15–5749 ppb; mitigated tablet value: 0.097 ppm
Comparator Or Baseline	No published NOx-sensitivity data for N-nitroso-fluoxetine or N-nitroso-duloxetine API grinding under controlled NOx; NDMA formation from dimethylamine in metformin has reported NOx sensitivity but via a different amine precursor and API matrix
Quantified Difference	97-fold increase (54 → 5247 ppb) under 2 ppm NOx during grinding; 26-fold reduction (5749 → 222 ppb max) by relocating grinding to <0.03 ppm NOx environment; 0.097 ppm final tablet level (8.6-fold below the 0.83 ppm NMT limit)
Conditions	Laboratory: agate mortar grinding, NOx 2 ppm; Manufacturing: API grinding and milling steps, NOx measured at old site 8.4 ppm vs. new site <0.03 ppm; Finished tablets: NOx-free air, three-pronged approach (Nomura et al., 2025)

Why This Matters

This quantifiable NOx sensitivity makes N-nitroso-atomoxetine a uniquely informative process marker for atomoxetine manufacturing quality; reference standard procurement must account for potential formation during standard handling if not protected from NOx exposure.

• [1] PMDA Japan. Safety Review Committee Discussion (資料3-2: Nihon Iyakuhin Risk Assessment Report). August 2024. Summarized on USP Nitrosamines Exchange. Available at: https://nitrosamines.usp.org/t/current-nitrosamine-landscape-in-japan/1626/123 View Source
• [2] Nomura R, Taniguchi T, Kitada K, Otsugi M, Tamai T, Suzuoki K, Yokoi R, Kondo K, Uchikawa O. Three-Pronged Approach (Active Pharmaceutical Ingredients, Excipients, and NOx-Free Air) Successfully Suppressed the Formation of Nitroso-atomoxetine to 0.097 ppm. Organic Process Research & Development, 2025 (ASAP). DOI: 10.1021/acs.oprd.5c00184 View Source

Analytical Method Sensitivity: LC-MS/MS for Oral Solution Matrix

A fully validated LC-MS/MS method for N-nitroso-atomoxetine in atomoxetine oral solution achieved a limit of detection (LOD) of 0.1 ppb and a limit of quantification (LOQ) of 0.2 ppb, with excellent linearity (R² = 0.9995) and precision (RSD < 2.6%) [1]. Recovery ranged from 104.3% to 109.2% [1]. In a separate UFLC-MS/MS method validated for atomoxetine hydrochloride API, a similar sensitivity profile was reported [2]. A Chinese patent application further described a UPLC-MS method achieving an LOQ of 0.100 ppm (100 ppb) and LOD of 0.033 ppm (33 ppb) for N-nitroso-atomoxetine in atomoxetine hydrochloride API [3]. These method sensitivities are sufficient to quantify N-nitroso-atomoxetine at levels required to meet the most stringent regulatory AI (FDA: 26.5 ng/day, corresponding to ~0.265 ppm in a 100 mg daily dose), with the 0.2 ppb LOQ method [1] providing >1000-fold margin below the regulatory threshold. Compared with published methods for N-nitroso-fluoxetine or N-nitroso-duloxetine, the chromatographic conditions (reversed-phase C18, acetonitrile/water/formic acid mobile phase, positive ESI-MRM at m/z 302→177) are specific to the atomoxetine NDSRI scaffold and are not transferable without revalidation.

Analytical method sensitivity

Cross-study comparable

LC-MS/MS LOQ 0.2 ppb, R²=0.9995, recovery 104.3–109.2%

Validated for quantification below all current AI-based regulatory limits

C18, positive ESI-MRM m/z 302→177

Evidence Dimension	Analytical method sensitivity (LOD, LOQ, linearity) for N-nitroso-atomoxetine quantification
Target Compound Data	LC-MS/MS: LOD = 0.1 ppb, LOQ = 0.2 ppb, R² = 0.9995, RSD < 2.6%, Recovery 104.3–109.2% (Adabala et al., 2025); UFLC-MS/MS: comparable LOQ (Patel et al., 2025); UPLC-MS: LOQ = 0.100 ppm, LOD = 0.033 ppm (Chinese Patent CN202410XXXXX)
Comparator Or Baseline	Other NDSRI LC-MS/MS methods typically report LOQ in the 0.1–10 ppb range depending on compound ionization efficiency and matrix; specific N-nitroso-fluoxetine and N-nitroso-duloxetine validated LOQ values not collated in a single comparative study
Quantified Difference	The 0.2 ppb LOQ provides >1000-fold analytical margin below the 100 ng/day AI-based limit (~0.83 ppm for 120 mg max daily dose) and >1300-fold margin below the FDA 26.5 ng/day AI-based limit (~0.265 ppm for 100 mg daily dose)
Conditions	LC-MS/MS: reversed-phase C18 column, mobile phase acetonitrile:HPLC water:methanol (75:20:5 v/v/v) with 0.1% formic acid, positive ESI, MRM mode; UFLC-MS/MS: C18 column, acetonitrile/0.1% formic acid gradient; UPLC-MS: C18 column, formic acid aqueous/ acetonitrile gradient, positive ESI-MRM

Why This Matters

The validated LOQ of 0.2 ppb confirms that commercially available N-nitroso-atomoxetine reference standards of ≥95% purity are analytically fit-for-purpose for quantifying this impurity at levels far below all current regulatory AI thresholds, supporting both ANDA submission and routine QC release testing.

• [1] Adabala JS, Hemalatha K, Angirekula N, Mukthinuthalapati MA. Advanced LC-MS/MS method for the determination of N-Nitroso Atomoxetine in Atomoxetine oral solution: Ensuring drug safety. Research Journal of Pharmacy and Technology, 2025;18(10):5011–5016. DOI: 10.52711/0974-360X.2025.00724 View Source
• [2] Vadariya S, Patel J, Patel H, Hirpara H, Patel T. Development And Validation Of N-Nitroso Atomoxetine Impurity (NDSRI) Determination Method In Atomoxetine Hydrochloride With UFLC-MS/MS (ESI). SEEJPH, 2025; pp. 2238–2248. View Source
• [3] Chinese Patent Application. Method for detecting trace nitrosamine impurity N-nitroso-atomoxetine in atomoxetine hydrochloride (一种盐酸托莫西汀中痕量亚硝胺杂质N-亚硝基-托莫西汀的检测方法). Published December 2024. Available at: https://www.jigao616.com/zhuanlijieshao_43551747.aspx View Source

Process Capability: NOx Mitigation in Finished Drug Product

The Towa Pharmaceutical team demonstrated that implementing a three-pronged NOx-free approach (NOx removal filters on air conditioning, compressed air, and film-coating supply air) reduced N-nitroso-atomoxetine content in finished atomoxetine tablets to 0.097 ppm—an 8.6-fold margin below the regulatory NMT limit of 0.83 ppm [1]. Without NOx filtration, tablet levels reached 0.267 ppm (3.1-fold below NMT but 2.75-fold higher than the filtered process) [1]. At intermediate processing stages, the NOx-filtered API showed 0.014 ppm, granulated powder 0.050 ppm, uncoated tablets 0.049 ppm, and film-coated tablets 0.097 ppm, versus unfiltered values of 0.087–0.267 ppm [1]. This quantitative process capability demonstration provides a benchmark for atomoxetine manufacturers that is specific to N-nitroso-atomoxetine: the NOx-mediated formation pathway is mechanistically linked to the secondary amine in the atomoxetine scaffold and the data cannot be directly extrapolated to NDSRIs derived from tertiary amines (which require dealkylation prior to nitrosation) or to NDSRIs with different nitrosation kinetics.

Process capability demonstration

Supporting evidence

0.097 ppm in finished tablets with NOx filtration vs. 0.267 ppm without; 8.6-fold margin below NMT

Provides validated benchmark for manufacturing compliance

Three-pronged NOx-free approach

Evidence Dimension	Finished product N-nitroso-atomoxetine level with vs. without NOx mitigation
Target Compound Data	N-Nitroso-atomoxetine: 0.097 ppm (NOx filter) vs. 0.267 ppm (no filter) in film-coated tablets; API: 0.014 ppm (filtered)
Comparator Or Baseline	Regulatory NMT limit = 0.83 ppm (derived from 100 ng/day AI at maximum daily dose); unfiltered process = 0.267 ppm; PMDA Japan reported API lot range 15–5749 ppb with older grinding site NOx at 8.4 ppm
Quantified Difference	8.6-fold margin below NMT for filtered process (0.097 ppm vs. 0.83 ppm); 2.75-fold reduction via NOx filtration (0.267→0.097 ppm); API filter reduction: 0.014 ppm vs. 54–5749 ppb range in unfiltered/older process
Conditions	Commercial-scale tablet manufacturing, NOx removal filters at 3 locations, average post-filter NOx <5 ppb in production atmosphere, LC-MS quantification (ESI, MRM m/z 302→177, biphenyl column)

Why This Matters

This demonstrated process capability provides atomoxetine manufacturers with a validated quantitative benchmark for N-nitroso-atomoxetine control, de-risking regulatory submission by showing that compliance with even the most stringent AI limits is achievable with appropriate NOx management.

• [1] Nomura R, Taniguchi T, Kitada K, Otsugi M, Tamai T, Suzuoki K, Yokoi R, Kondo K, Uchikawa O. Three-Pronged Approach (Active Pharmaceutical Ingredients, Excipients, and NOx-Free Air) Successfully Suppressed the Formation of Nitroso-atomoxetine to 0.097 ppm. Organic Process Research & Development, 2025 (ASAP). DOI: 10.1021/acs.oprd.5c00184 View Source

N-Nitroso-Atomoxetine Reference Standard Applications

ANDA/NDA Submission: Validated Quantification

Regulatory submissions for atomoxetine-containing drug products (oral capsules, oral solutions) now require validated analytical methods for N-nitroso-atomoxetine as part of nitrosamine risk assessment. The compound-specific MRM transition (m/z 302→177) and chromatographic retention on C18 columns under acidic mobile phase conditions [1] mean that the N-nitroso-atomoxetine reference standard is an indispensable calibrant for method development, validation (AMV), and routine QC release testing. The demonstrated method sensitivity (LOQ 0.2 ppb by LC-MS/MS [1]; LOQ 0.100 ppm by UPLC-MS [2]) meets or exceeds the sensitivity required for compliance with both EMA (100 ng/day) and FDA (26.5 ng/day) AI limits [3]. Procurement of a characterized reference standard (≥95% HPLC purity with full COA including H-NMR, MASS, HPLC [4]) is a prerequisite for ANDA/NDA acceptance.

Manufacturing Process Optimization: NOx Exposure Monitoring

The quantitative relationship between environmental NOx concentration and N-nitroso-atomoxetine formation—a 97-fold increase (54→5247 ppb) under 2 ppm NOx during grinding [1]—positions this impurity as a uniquely sensitive sentinel marker for NOx exposure during atomoxetine API and drug product manufacturing. By monitoring N-nitroso-atomoxetine levels at each processing step (intermediate, post-purification, post-grinding, post-formulation), manufacturers can identify NOx ingress points and validate the effectiveness of NOx mitigation strategies (filtration, facility relocation). The Towa Pharmaceutical study demonstrated that a three-pronged NOx-free approach reduces finished product levels to 0.097 ppm [2], providing a quantitative benchmark for process capability assessment.

Toxicological Risk Reassessment: Science-Based AI Derivation

The substantial discrepancy between the BMDL-derived AI of ≥4400 ng/day (from TGR in vivo data) and the currently adopted regulatory AI of 100 ng/day (read-across) [1] creates an active research need for additional compound-specific toxicological data. Laboratories engaged in mutagenicity/carcinogenicity research can use the N-nitroso-atomoxetine reference standard to conduct independent Ames fluctuation assays, in vitro micronucleus tests in human TK6 cells, or in vivo TGR studies to generate data that may support a revised, science-based AI. The compound's classification as ICH M7 Class 2 (positive in vitro and in vivo mutagenicity, but lacking carcinogenicity data) [2] makes it a relevant model compound for studying the mutagenicity–carcinogenicity correlation in NDSRIs.

Stability Studies: Formation During Shelf-Life Under ICH Conditions

The PMDA Japan review noted that accelerated stability testing (60°C, 2-month storage) was used to confirm that storage-induced N-nitroso-atomoxetine increase remains within acceptable limits [1]. Procurement of the reference standard enables pharmaceutical manufacturers to conduct ICH-compliant forced degradation and stability studies to quantify N-nitroso-atomoxetine formation rates under long-term (25°C/60% RH), intermediate (30°C/65% RH), and accelerated (40°C/75% RH) storage conditions. Given the compound's sensitivity to heat, moisture, and acidity during formation [2], stability-indicating analytical methods must be validated with authentic N-nitroso-atomoxetine reference material to ensure accurate quantification and establish shelf-life specifications.

Application

Selection Property

Validation Focus

ANDA/NDA submission: validated quantification

Compound-specific MRM and comprehensive COA

Method sensitivity below most stringent AI-based limits

Manufacturing process optimization: NOx monitoring

Quantitative NOx sensitivity profile

Process-step NOx ingress identification

Toxicological risk reassessment: science-based AI

Class 2 mutagenicity with in vivo TGR data

Independent mutagenicity/carcinogenicity studies

Stability studies: ICH shelf-life evaluation

Stability-indicating analytical method

ICH storage condition validation

• [1] Adabala JS, Hemalatha K, Angirekula N, Mukthinuthalapati MA. Advanced LC-MS/MS method for the determination of N-Nitroso Atomoxetine in Atomoxetine oral solution: Ensuring drug safety. Research Journal of Pharmacy and Technology, 2025;18(10):5011–5016. DOI: 10.52711/0974-360X.2025.00724 View Source
• [2] Chinese Patent Application. Method for detecting trace nitrosamine impurity N-nitroso-atomoxetine in atomoxetine hydrochloride. December 2024. Available at: https://www.jigao616.com/zhuanlijieshao_43551747.aspx View Source
• [3] USP Nitrosamines Exchange. EMA Q&A Appendix 1 updated Sept 4, 2024 (FDA/EMA AI comparison). Available at: https://nitrosamines.usp.org/t/ema-q-a-appendix-1-is-updated-sept-4-2024/11202/10 View Source
• [4] Veeprho. N-Nitroso Atomoxetine | CAS 3028194-77-3 – Product Datasheet. 2022. Available at: https://veeprho.com/impurities/n-nitroso-atomoxetine/ View Source
• [5] PMDA Japan. Safety Review Committee Discussion on N-Nitroso-Atomoxetine Risk Assessment (資料3-1, 資料3-2), August 2024. Summarized on USP Nitrosamines Exchange. Available at: https://nitrosamines.usp.org/t/current-nitrosamine-landscape-in-japan/1626/123 View Source
• [6] Jolly RA, Cornwell PD, Noteboom J, Sayyed FB, Thapa B, Buckley LA. Estimation of acceptable daily intake values based on modeling and in vivo mutagenicity of NDSRIs of fluoxetine, duloxetine and atomoxetine. Regulatory Toxicology and Pharmacology, 2024;152:105672. DOI: 10.1016/j.yrtph.2024.105672 View Source
• [7] Nomura R, Taniguchi T, Kitada K, et al. Three-Pronged Approach Successfully Suppressed the Formation of Nitroso-atomoxetine to 0.097 ppm. Organic Process Research & Development, 2025. DOI: 10.1021/acs.oprd.5c00184 View Source