Lyle D. Burgoon

A framework for the next generation of risk science.

Objectives: In 2011, the U.S. Environmental Protection Agency initiated the NexGen project to develop a new paradigm for the next generation of risk science. Methods: The NexGen framework was built on three cornerstones: the availability of new data on toxicity pathways made possible by fundamental advances in basic biology and toxicological science, the incorporation of a population health perspective that recognizes that most adverse health outcomes involve multiple determinants, and a renewed focus on new risk assessment methodologies designed to better inform risk management decision making. Results: The NexGen framework has three phases. Phase I (objectives) focuses on problem formulation and scoping, taking into account the risk context and the range of available risk management decision-making options. Phase II (risk assessment) seeks to identify critical toxicity pathway perturbations using new toxicity testing tools and technologies, and to better characterize risks and uncertainties using advanced risk assessment methodologies. Phase III (risk management) involves the development of evidence-based population health risk management strategies of a regulatory, economic, advisory, community-based, or technological nature, using sound principles of risk management decision making. Conclusions: Analysis of a series of case study prototypes indicated that many aspects of the NexGen framework are already beginning to be adopted in practice. Citation: Krewski D, Westphal M, Andersen ME, Paoli GM, Chiu WA, Al-Zoughool M, Croteau MC, Burgoon LD, Cote I. 2014. A framework for the next generation of risk science. Environ Health Perspect 122:796–805; http://dx.doi.org/10.1289/ehp.1307260

The Next Generation of Risk Assessment Multi-Year Study—Highlights of Findings, Applications to Risk Assessment, and Future Directions

Background: The Next Generation (NexGen) of Risk Assessment effort is a multi-year collaboration among several organizations evaluating new, potentially more efficient molecular, computational, and systems biology approaches to risk assessment. This article summarizes our findings, suggests applications to risk assessment, and identifies strategic research directions. Objective: Our specific objectives were to test whether advanced biological data and methods could better inform our understanding of public health risks posed by environmental exposures. Methods: New data and methods were applied and evaluated for use in hazard identification and dose–response assessment. Biomarkers of exposure and effect, and risk characterization were also examined. Consideration was given to various decision contexts with increasing regulatory and public health impacts. Data types included transcriptomics, genomics, and proteomics. Methods included molecular epidemiology and clinical studies, bioinformatic knowledge mining, pathway and network analyses, short-duration in vivo and in vitro bioassays, and quantitative structure activity relationship modeling. Discussion: NexGen has advanced our ability to apply new science by more rapidly identifying chemicals and exposures of potential concern, helping characterize mechanisms of action that influence conclusions about causality, exposure–response relationships, susceptibility and cumulative risk, and by elucidating new biomarkers of exposure and effects. Additionally, NexGen has fostered extensive discussion among risk scientists and managers and improved confidence in interpreting and applying new data streams. Conclusions: While considerable uncertainties remain, thoughtful application of new knowledge to risk assessment appears reasonable for augmenting major scope assessments, forming the basis for or augmenting limited scope assessments, and for prioritization and screening of very data limited chemicals. Citation: Cote I, Andersen ME, Ankley GT, Barone S, Birnbaum LS, Boekelheide K, Bois FY, Burgoon LD, Chiu WA, Crawford-Brown D, Crofton KM, DeVito M, Devlin RB, Edwards SW, Guyton KZ, Hattis D, Judson RS, Knight D, Krewski D, Lambert J, Maull EA, Mendrick D, Paoli GM, Patel CJ, Perkins EJ, Poje G, Portier CJ, Rusyn I, Schulte PA, Simeonov A, Smith MT, Thayer KA, Thomas RS, Thomas R, Tice RR, Vandenberg JJ, Villeneuve DL, Wesselkamper S, Whelan M, Whittaker C, White R, Xia M, Yauk C, Zeise L, Zhao J, DeWoskin RS. 2016. The Next Generation of Risk Assessment multiyear study—highlights of findings, applications to risk assessment, and future directions. Environ Health Perspect 124:1671–1682; http://dx.doi.org/10.1289/EHP233

Using In Vitro High-Throughput Screening Data for Predicting Benzo[k]Fluoranthene Human Health Hazards

Today there are more than 80,000 chemicals in commerce and the environment. The potential human health risks are unknown for the vast majority of these chemicals as they lack human health risk assessments, toxicity reference values, and risk screening values. We aim to use computational toxicology and quantitative high-throughput screening (qHTS) technologies to fill these data gaps, and begin to prioritize these chemicals for additional assessment. In this pilot, we demonstrate how we were able to identify that benzo[k]fluoranthene may induce DNA damage and steatosis using qHTS data and two separate adverse outcome pathways (AOPs). We also demonstrate how bootstrap natural spline-based meta-regression can be used to integrate data across multiple assay replicates to generate a concentration-response curve. We used this analysis to calculate an in vitro point of departure of 0.751 μM and risk-specific in vitro concentrations of 0.29 μM and 0.28 μM for 1:1,000 and 1:10,000 risk, respectively, for DNA damage. Based on the available evidence, and considering that only a single HSD17B4 assay is available, we have low overall confidence in the steatosis hazard identification. This case study suggests that coupling qHTS assays with AOPs and ontologies will facilitate hazard identification. Combining this with quantitative evidence integration methods, such as bootstrap meta-regression, may allow risk assessors to identify points of departure and risk-specific internal/in vitro concentrations. These results are sufficient to prioritize the chemicals; however, in the longer term we will need to estimate external doses for risk screening purposes, such as through margin of exposure methods.

Revisiting Cohen et al. 2015, Cohen et al. 2014 and Waalkes et al. 2014: a bayesian re-analysis of tumor incidences

by Cohen et al. Specifically, we tested the null hypothesis: The control animal tumor incidences reported by Tokar et al. in 2011 are not different from the study reported by the same group in the Waalkes et al. 2014 publication. Our full analysis can be found in Burgoon and Druwe (2015); however, we briefly discuss the approach and results here. To test the null hypothesis, we used a Bayesian approach. Initially, we examined whether or not a difference existed without using any prior knowledge of what the tumor incidence in Waalkes’ laboratory was; therefore, we used a flat prior distribution. We modeled the control tumor incidences from the Tokar 2011 and Waalkes studies as Bernoulli distributions. The posterior distributions were calculated and are shown in Burgoon and Druwe (2015). There is a difference of about 13 % between the means of these two distributions. However, that fact alone does not mean that the incidences are from different distributions. In fact, what we could be observing is a case where the incidences for each study were taken from different sides of the same distribution. Thus, to test the hypothesis, we took samples from the posterior distributions and calculated the difference to obtain a distribution of the differences. If the incidences in both studies were from the same distribution, we would expect a difference of 0, or close to 0, to be a credible value. In order to accomplish this, we used an approach that sets a region of practical equivalence (ROPE) around the zero difference, and the 95 % highest density interval (HDI) of the difference distribution. The ROPE demarcates a region around zero difference that is functionally equivalent to no difference. In general, if any part of the 95 % HDI is within the ROPE, then we accept the null hypothesis that the control tumor incidences from the studies are the same. Else, we reject the null hypothesis. A complete explanation of our decision rules can be found in Burgoon and Druwe (2015). Over the past year, this journal has published a lively discussion in the form of letters to the editor between Waalkes et al. and Cohen et al. regarding an article published in this journal titled “Lung Tumors in Mice induced by ‘whole Life’ inorganic arsenic exposure at human relevant doses” by Waalkes et al. in 2014. Cohen et al. raised a series of thoughtful questions with respect to the reproducibility of the control animal tumor incidences in the arsenic exposure studies published by Tokar et al. (2011) and Waalkes et al. (2014). In addition, Cohen et al. brought into question whether the development of lung tumors in the mice was related to the genetic background of the mice used in the study rather than arsenic exposure. Many of the questions raised by Cohen and colleagues centered around what they deemed to be uncertainty in the tumor incidences in the control animals used in both studies, and by extension, the quality of the studies performed by Tokar and Waalkes. If the assertions made by Cohen et al. proved true, risk assessors would be unable to use Waalkes et al.’s (2014) study in hazard and dose–response assessments of inorganic arsenic. Thus, we performed the analysis requested

Building a developmental toxicity ontology.

BACKGROUND As more information is generated about modes of action for developmental toxicity and more data are generated using high-throughput and high-content technologies, it is becoming necessary to organize that information. This report discussed the need for a systematic representation of knowledge about developmental toxicity (i.e., an ontology) and proposes a method to build one based on knowledge of developmental biology and mode of action/ adverse outcome pathways in developmental toxicity. METHODS This report is the result of a consensus working group developing a plan to create an ontology for developmental toxicity that spans multiple levels of biological organization. RESULTS This report provide a description of some of the challenges in building a developmental toxicity ontology and outlines a proposed methodology to meet those challenges. As the ontology is built on currently available web-based resources, a review of these resources is provided. Case studies on one of the most well-understood morphogens and developmental toxicants, retinoic acid, are presented as examples of how such an ontology might be developed. DISCUSSION This report outlines an approach to construct a developmental toxicity ontology. Such an ontology will facilitate computer-based prediction of substances likely to induce human developmental toxicity.

AOP: An R Package For Sufficient Causal Analysis in Pathway-based Screening of Drugs and Chemicals for Adversity

Summary How can I quickly find the key events in a pathway that I need to monitor to predict that a/an beneficial/adverse event/outcome will occur? This is a key question when using signaling pathways for drug/chemical screening in pharmacology, toxicology and risk assessment. By identifying these sufficient causal key events, we have fewer events to monitor for a pathway, thereby decreasing assay costs and time, while maximizing the value of the information. I have developed the “aop” package which uses back-door analysis of causal networks to identify these minimal sets of key events that are sufficient for making causal predictions. Availability and Implementation The source for the aop package is available online at Github at https://github.com/DataSciBurgoon/aop and can be installed using the R devtools package. The aop package runs within the R statistical environment. The package has functions that can take pathways (as directed graphs) formatted as a Cytoscape JSON file as input, or pathways can be represented as directed graphs using the R/Bioconductor “graph” package. The “aop” package has functions that can perform backdoor analysis to identify the minimal set of key events for making causal predictions. Contact lyle.d.burgoon@usace.army.mil