Computational toxicology of <i>N</i> -nitrosamine impurities: from molecular structure to regulatory concern

Since 2018, drug recalls and other regulatory actions related to impurities, particularly <i>N-</i>nitrosamines, have raised significant toxicological and public health concerns. These potent carcinogens, characterized by the R₂N-N = O moiety, may form through nitrosation of secondary amines, contaminated raw materials, or excipients. Regulatory bodies, such as the United States Food and Drug Administration (USFDA), World Health Organization (WHO), and Central Drugs Standard Control Organization (CDSCO), have issued guidelines to mitigate these risks. Conventional analytical methods, including Liquid Chromatography-Mass Spectrometry (LC-MS), Gas Chromatography-Mass Spectrometry (GC-MS), and High-Resolution Mass Spectrometry (HRMS), remain standard for detection and quantification, but are constrained by cost, methodological complexity, and sensitivity. Recent advances in computational approaches, including quantum mechanics (QM), quantitative structure-activity relationship (QSAR) modeling, and artificial intelligence (AI), do not directly quantify impurities but estimate carcinogenic potency, identify mutagenic structural drivers, and derive acceptable intake limits, etc. This review presents a mechanism-driven and regulatory-oriented framework integrating nitrosamine formation pathways, metabolic bioactivation, analytical detection strategies, and predictive computational techniques, providing a coherent perspective on their capabilities, limitations, and complementary role alongside analytical techniques in nitrosamine safety assessment.