How to use surrogate data and read-across in the qualification of non-mutagenic pharmaceutical impurities.
Read-across uses data from one or more scientifically justified analogues to address a data gap for a target impurity. The quality of the conclusion depends more on the biological relevance of the analogue than on a single structural-similarity number.
Analogue selection
The surrogate should be relevant to the expected toxicological behavior and exposure context.
Similarity dimensions
Structure, physicochemical properties, metabolism, pharmacokinetics, target-organ profile and mechanism may all be relevant.
Differences matter
Important structural or metabolic differences should be identified and explained rather than hidden by a global similarity score.
Uncertainty
The conclusion should state what is known, what remains uncertain and why the evidence is adequate for the proposed specification.
Frequently asked questions
Is high structural similarity sufficient?
Not by itself.
Can more than one surrogate be used?
Yes, when multiple analogues strengthen the weight of evidence.
Should metabolism be considered?
Yes when it can change the toxicological relevance of the analogue.
Does read-across automatically establish a limit?
No. Exposure and the strength of the evidence must be integrated.
