Evaluation of a tiered in vitro testing strategy for assessing the ocular and dermal irritation/corrosion potential of pharmaceutical compounds for worker safety.

INTRODUCTION: Irritation reactions are a frequently reported occupational illness. The potential adverse effects of pharmaceutical compounds (PCs) on eye and skin can now be assessed using validated in vitro methods.

OBJECTIVES: Our overall aim is to reduce animal testing by replacing the historically utilized in vivo test methods with validated in vitro test methods which accurately determine the ocular and dermal irritation/corrosion potential of PCs to inform worker safety within the pharmaceutical space. Bristol-Myers Squibb (BMS) and the Institute for In Vitro Sciences (IIVS) have therefore conceptualized and internally qualified a tiered in vitro testing strategy to inform occupational hazards regarding eye and skin irritation and corrosivity of PCs. For the small scale pre-qualification phase, we paired historical in vivo and newly generated in vitro data for 15 PCs to determine the predictive capacity of in vitro assays already validated for the eye and skin irritation/corrosion endpoints and accepted for certain regulatory submissions. During the post-qualification phase, a group of 24 PCs were subjected exclusively to the developed tiered testing strategy, which is based on three Organisation for Economic Co-operation and Development (OECD) in vitro methods.

MATERIALS AND METHODS: The qualified in vitro testing strategy utilizes the Corrositex® assay for the corrosivity (OECD TG 435), the Bovine Corneal Opacity and Permeability (BCOP) assay for ocular irritation (OECD TG 437), and the EpiDerm™ tissue model-based Skin Irritation Test (SIT) for dermal irritation (OECD TG 439). In the first step, the pH of each PC was determined. For compounds with pH extremes ≥11 or ≤2, the Corrositex® assay was generally conducted first. For compound(s) that were incompatible with or were negative in the Corrositex® assay or had pH values between 2 and 11, the BCOP assay and SIT were performed first.

RESULTS: The results of the tiered testing strategy's qualification phase demonstrated that the BCOP assay is sensitive enough to identify a wide range of eye irritation/corrosion potentials and its over-prediction rate was considered acceptable to inform occupational hazards and ensure the proper handling practices of PCs. The SIT correctly predicted the skin irritation potential of 14 out of the 15 PCs included in the qualification phase, only over-predicting one PC. In the post-qualification phase, four PCs out of four tested were predicted corrosive by the Corrositex® assay and thus no further testing was needed or conducted. The rest of the PCs were evaluated in the BCOP assay (both neat and as a 20% dilution), with the higher response being used for hazard classification. Four PCs were determined to be severe eye irritants, 1 a moderate irritant, 8 were mild irritants, and 8 were non-irritants. The same set of PCs was evaluated using the SIT and were classified as non-irritants to skin. These results are consistent with the BMS historical in vivo results showing a very low number of PCs as skin irritants.

CONCLUSIONS: This tiered in vitro testing strategy, which replaces the use of animal studies, was found to be reasonably accurate in its predictive capacity when compared to historical in vivo results and represents a conservative and reliable platform that can be utilized for the prediction of ocular and dermal irritation/corrosion potential of PCs and for subsequent GHS classification and worker safety hazard communications.