I attended the Society of Toxicology (SOT) annual meeting in Orlando (#SOT2025) where I presented our poster, “Best Practices of Read-Across for the Qualification of Non-Genotoxic Impurities.” This addresses a growing need in pharmaceutical safety: how to evaluate the risks of non-genotoxic impurities (NGIs) that are data-poor.
NGIs are covered under regulatory guidelines ICH Q3A and Q3B, which define qualification thresholds, i.e., levels below which impurities are considered unlikely to pose additional toxicological risks. But when these thresholds are exceeded, a deeper safety assessment is needed. Often, however, NGIs are novel structures with limited or no toxicological data available. This is where read-across, supported by in silico methods, plays a key role in enabling a fit-for-purpose safety assessment.
Our poster explores how read-across, guided by best practices, can support the safety qualification of both API-like and chemical-like NGIs. We focus on building scientifically robust justifications across four similarity domains: structural, physicochemical, metabolic, and toxicological.
Our study aligns with ongoing regulatory discussions, including the EMA’s draft reflection paper on non-mutagenic impurities. The message is clear: read-across, when applied rigorously and transparently, is not only scientifically valid but essential in impurity qualification strategies.
This work underscores the value of integrating in silico approaches into pharmaceutical development. I’m grateful to my co-authors and collaborators for their contributions to this important work.
Have questions or thoughts? Feel free to get in touch. We’re always open to continuing the conversation!