Bio: Dr. Kan Shao is an Associate Professor of Environmental and Occupational Health at Indiana University (IU) School of Public Health – Bloomington. Dr. Shao is a certified toxicologist by the American Board of Toxicology (DABT) and a fellow of the Academy of Toxicological Sciences (ATS) and was recently appointed to the Committee on Toxicology of the National Academies of Sciences, Engineering, and Medicine (NASEM).

Dr. Shao’s research focuses on advancing computational and statistical modeling methods to support chemical risk assessment. His major contributions to this field include the development of the benchmark dose methodology, Bayesian approaches to quantify various sources of uncertainties in dose-response assessment, and a modeling framework to quantitatively integrate mechanistic information. He has received more than three million dollars in external grants from NIH to support his research projects in computational toxicology. Shao is now an Associate Editor of the journal Drug and Chemical Toxicology and served as a reviewer for a few high-profile risk assessment reports. 

Abstract: A primary objective of dose-response assessment is to estimate a reference dose (RfD) to support chemical risk assessment. Conventional approaches often apply a two-step procedure—point of departure (POD) derivation and low-dose extrapolation—without fully integrating mode of action (MOA) information, particularly for chemicals with carcinogenic potential. This study presents a MOA-based probabilistic dose-response modeling framework that quantitatively synthesizes data from multiple sources, including toxicological, epidemiological, and genomic studies, to estimate RfDs. The framework consists of four key steps: (1) identifying key quantifiable events (KQEs) along the MOA pathway, (2) calculating essential doses from relevant experimental data, (3) deriving MOA-based PODs, and (4) estimating probabilistic RfDs using Bayesian methods. This approach builds dose-response relationships across the entire dose continuum, including the low-dose region, while addressing both uncertainty and variability in a transparent, probabilistic manner. The framework is flexible and generalizable, offering a promising tool to improve current chemical risk assessment for various types of chemicals and better inform regulatory decision-making.