Use and limitations of clinical data in the identification and classification of low molecular weight chemicals (LMWCs) as respiratory sensitizers: recommendations for improvement.

While progress has been made in recent years, there are still no suitable and accepted in silico, in vitro, or in vivo models that can be used to accurately predict whether a chemical substance has the intrinsic property to cause immune-mediated chemical respiratory allergy, typically manifested as allergic asthma or allergic rhinitis which represents a severe health hazard. Regulatory authorities have relied primarily on clinical evidence (case reports, clinical databases, worker exposure studies) to classify substances as respiratory sensitizers, but this evidence can lack a proven immunological mechanism which is necessary to identify substances which can cause life-long sensitization and clinically relevant allergic symptoms in the respiratory tract in an exposed population (such respiratory allergens may be considered as "true" sensitizers, in analogy to the definition of skin sensitization, and in contrast to respiratory irritants). In light of this, the European Center for Ecotoxicology and Toxicology of Chemicals convened a Task Force to evaluate the types of clinical methods and data sources and the implications of relying on such data for regulatory decision making from a scientific perspective. Recognizing that there are benefits and important insights from using such data, significant shortcomings were identified. With clinical work being focused on treatment and diagnosis of individual patients, the approaches and methods used for clinical guidance, diagnostics and reporting have serious limitations in proving the respiratory sensitization potential of a specific chemical, definitely restricting their suitability in deriving legally binding hazard classifications for human health protection. Even within the current broader regulatory definition of respiratory sensitization, a robust assessment and sound evidence of causation by a specific chemical seems mandatory in order to avoid misclassifications. Application of a systematic weight-of-evidence approach is considered suitable to determine the level of confidence, including a thorough assessment of the specificity or non-specificity of observed bronchial hyperreactivity. Recommendations proposed in this publication may not only aid industry and regulators in their decision making but also facilitate a further exchange between stakeholders to improve the data used to (a) more precisely identify true respiratory sensitizers to effectively protect human health, (b) aid evaluation of potential predictive models, and (c) encourage regulators to clarify guidance and to consider a re-evaluation of the current regulatory definition of respiratory sensitizers.