A Framework for Systematic Evaluation and Quantitative Integration of Mechanistic Data in Assessments of Potential Human Carcinogens

• Published in: Toxicological sciences Vol. 167; no. 2; pp. 322 - 335
• Main Authors: Wikoff, Daniele S, Rager, Julia E, Chappell, Grace A, Fitch, Seneca, Haws, Laurie, Borghoff, Susan J
• Format: Journal Article
• Language: English
• Published: United States Oxford University Press 01.02.2019
• Subjects: Animals; Biomedical Research - methods; Biomedical Research - trends; Carcinogens - toxicity; Endpoint Determination; High-Throughput Screening Assays; Humans; Neoplasms - chemically induced; Neoplasms - diagnosis; Risk Assessment; cancer; systematic review; data integration; mechanisms; high-throughput screening
• ISSN: 1096-6080; 1096-0929
• DOI: 10.1093/toxsci/kfy279

Abstract Recently, the key characteristics of carcinogens (KCC) have been proposed as an organizational approach for the evaluation of mechanistic data related to carcinogenicity. Our objective was to develop a framework to systematically and quantitatively integrate KCC data using elements that are important to risk assessment. Methods for developing the framework included: defining objectives, identifying and accommodating key considerations for components, input, and output of the framework, and operational development via iterative testing by a multidisciplinary team. The proposed framework involves 3 steps: (1) a structured, yet flexible, appraisal of individual studies and endpoints, (2) a structured and transparent evaluation of the body of evidence for each key characteristic, and (3) an evaluation of all of the KCC-relevant data relative to tumors and/or cancer types. In step 1, data are assessed and scored for reliability, strength, and activity. In step 2, a mathematical algorithm is used to integrate (and weight) the quality, relevance, and activity for each of the KCCs. These scores facilitate subsequent evaluations related to the overall body of evidence in step 3 in which KCCs can be linked, assessing potential adverse outcome pathways or networks, and finally, considered in the context of observed carcinogenic responses in animals and/or humans. The output is an overall conclusion regarding KCC activity as it relates to carcinogenic responses. The proposed framework provides a flexible solution to quantitatively integrate KCC data in a systematic and transparent manner that provides weighting of data most well-suited for the assessment of potential human carcinogenicity.