Zhu, Dennis, Qu, Yang and Ye, Ning (2024) Automated Carcinogenic Potency Categorization Approach for Nitrosamine Drug Substance-related Impurities. Current Green Chemistry. ISSN 1463-92621463-9270

Abstract

Nitrosoamines are known carcinogens that can form during chemical syntheses. Recently released health authority guidelines have defined acceptable intake limits for various nitrosoamines, requiring improved risk assessment approaches. We have developed a web-based application that autonomously predicts the nitrosoamine potency category of compounds from their SMILES notation as high, moderate, or low risk per health authority limits. An automated prediction algorithm categorized nitrosoamine potency based on the encoded molecular structure. The application was trained and validated on datasets of compounds with known nitrosoamine formation potential. It provides instant potency screening to identify high risk nitrosoamine formations. Making these predictions widely accessible enables chemists to proactively mitigate nitrosoamine hazards during process development. This represents a step toward greener, safer chemicals and processes in alignment with public health priorities. The web-based interface and focus on regulatory compliance differentiates this tool from prior in silico models. User testing demonstrates its utility for accelerating low-risk chemical design. This predictive toxicology approach could be extended to other compound classes. Overall, the application integrates computational and regulatory science to advance environmental and human health priorities.

Item Type:	Article
Keywords:	N-Nitrosamine, Digitization, Autonomous Tool, CPCA, FDA, EMA
Date Deposited:	16 Apr 2024 00:46
Last Modified:	16 Apr 2024 00:46
URI:	https://oak.novartis.com/id/eprint/51631