Variability of non-clinical behavioral CNS safety assessment: An intercompany comparison.

Introduction: Irwin/FOB testing is routinely conducted to investigate the neurofunctional integrity of laboratory animals during preclinical development of new drugs, however, the study design frequently varies to meet specific needs. Representatives of several European-based pharmaceutical companies performed a "state-of-the-art" assessment of how they conduct their CNS safety evaluation using Irwin/FOB tests.

Methods: This assessment consisted of (1) a survey of current/historical practice, (2) an evaluation of historical studies with reference compounds (amphetamine, chlorpromazine) to determine intercompany reproducibility of results, and (3) an interlaboratory test using reference compounds (MK-801, chlorpromazine) to determine whether partially standardized conditions (animals, sex, doses, vehicles, administration route, observation time points, systemic exposure) might reduce variability of results.

Results: Our survey revealed several similarities, e.g., main endpoints of home cage and openfield observations, species, and positive control substances, but also a high level of heterogeneity between different companies with regard to behavioral endpoints during handling and reflex testing, scoring, group size, and timing of studies. Analysis of heterogeneously designed historical studies with amphetamine and chlorpromazine showed the anticipated behavioral changes, albeit with quantitative variability, and identified more robust (e.g., activity, posture, muscle tone, startle reflex, body temperature) and less robust (piloerection, stereotypical behavior, palpebral closure, respiration) Irwin/FOB parameters. A partially standardized interlaboratory test with MK-801 and chlorpromazine showed the expected behavioral changes and principally confirmed the historically-based more/less robust Irwin/FOB parameters, however, it also showed exposure variability and did not show a markedly reduced quantitative variability of behavioral results.

Discussion: Our survey and intercompany test results demonstrate certain heterogeneity in design and conduct of Irwin/FOB tests by pharmaceutical companies. Although the general behavioral profiles for the reference compounds were consistently found, quantitative variability of results remained even under partially standardized conditions. This suggests the importance of a high level of standardization with regard to the Irwin/FOB test modification used, scoring system, and observer training, in order to achieve an improved intercompany comparability of Irwin/FOB results.