Setting occupational exposure limits for unstudied pharmaceutical intermediates using an in vitro parallelogram approach

Abstract

Occupational exposure limits for unstudied pharmaceutical synthetic intermediates are often established under the assumption that penultimate and near-ultimate intermediates have the same structure-activity and dose-response as the ultimate active pharmaceutical ingredient (API). This is seldom the case because moieties that render biological activity to the API are often protected or modified for synthetic purposes. Incorrectly assuming that intermediates have biological activity similar to the API may lead to excessive exposure controls that in turn impose unnecessary ergonomic hazards on workers and greatly reduces the scale and efficiency of production. Instead of assuming intermediates have the same toxicity profile as the API, it is feasible to use a parallelogram approach to establish exposure limits for synthetic intermediates using low-cost in vitro data. By comparing in vitro responses of intermediates to structurally similar data-rich molecules such as the API, occupational exposure categories can be established for unstudied intermediates. In this contribution (1) methods for setting occupational exposure limits for data-poor compounds are reviewed; (2) applications and limitations of in vitro assays are discussed; (3) two exposure categorization examples are presented that rely on an in vitro parallelogram approach; and (4) inherent safeguards for uncertainties in pharmaceutical risk assessment are identified. In vitro hazard and dose-response information for unstudied intermediates that are structurally similar to well-studied APIs can greatly enhance the basis for setting occupational exposure limits for unstudied synthetic intermediates.

Keywords::

• API
• containment
• control
• ergonomics
• exposure
• hazard
• occupational
• OEL
• OEB
• parallelogram
• pharmaceutical
• potent compound
• risk assessment

Acknowledgements

The author wishes to thank Sean Hays and David Gilkey for insightful suggestions and William Hanneman for critical review of this paper. The technical editing assistance of Mike Bauer is gratefully acknowledged.

Declaration of Interest

The author reports no conflicts of interest. The author alone is responsible for the content and writing of this paper.