NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Devito M, Farrell P, Hagiwara S, et al. Value of Information Case Study on the Human Health and Economic Trade-offs Associated with the Timeliness, Uncertainty, and Costs of the Draft EPA Transcriptomic Assessment Product (ETAP). Washington (DC): U.S. Environmental Protection Agency; 2024 Jul.
Value of Information Case Study on the Human Health and Economic Trade-offs Associated with the Timeliness, Uncertainty, and Costs of the Draft EPA Transcriptomic Assessment Product (ETAP).
Show details3.1. DESIGN OF THE CASE STUDY
For regulatory agencies, case studies have been an important tool to understand the strengths and weaknesses of new methods and gain familiarity with new methods or approaches before application (Kavlock et al. 2018). The present report uses the VOI framework developed by EPA ORD (Hagiwara et al. 2022) in a case study to evaluate the human health and economic trade-offs associated with the timeliness, uncertainty, and costs of different toxicity testing and assessment approaches. The case study focuses on a VOI comparison between two options: 1) a five-day, repeated dose in vivo transcriptomic study and the EPA Transcriptomic Assessment Product process [hereafter, the combination is referred to as the ETAP]. The ETAP incorporates a standardized and structured data collection and analysis procedure, reporting template, and review process that is intended to facilitate the rapid development and release of the assessment. The ETAP is intended to be applied to substances with no existing or publicly accessible repeated dose toxicity studies or human evidence suitable for use as a POD and reference value derivation; and 2) two-year rodent chronic toxicity test with traditional human health assessment process [hereafter, the combination is referred to as THHA]. The two-year rodent chronic toxicity test was selected as the basis for comparison since the transcriptomic PODs from the five-day, repeated dose in vivo study showed robust concordance with the traditional apical PODs from the chronic studies (EPA 2024b). The concordance between the transcriptomic and apical POD values was approximately equivalent to the observed inter-study variability in the repeated dose toxicity studies (EPA 2024b). For the purposes of the case study, the THHA is assumed to be the gold standard that the ETAP is compared against. The ETAP was chosen as the comparator because it underwent parallel review by the EPA Board of Scientific Counselors (BOSC) (EPA 2024b) and has recently been released as a new EPA human health assessment product.
The case study was constructed to evaluate the VOI under a variety of chemical exposure and decision contexts that could impact the costs of exposure mitigation (i.e., control costs) and overall public health burden (i.e., health costs). To evaluate the relative benefits driving the choice between two processes, the case study inputs included important components from the VOI framework paper as well as others that were unique to the comparison being performed, including (Figure 3-1):
- Toxicity testing and human health assessment: The costs associated with conducting each type of toxicity test, uncertainty around the experimentally determined point of departure, and the time required for toxicity testing and developing the assessment.
- Exposure: Chemical exposure defined by the mean population exposure level and population variability in exposure.
- Affected population size: Size of the exposed population.
- Quality of exposure data: Consideration of more accurately knowing mean population exposure level and population variability in exposure.
- Health effects and chemical control costs: Economic valuation for the adverse health effects resulting from exposure and costs of exposure mitigation actions.
- Toxicological concordance uncertainty: Uncertainty associated with the five-day, repeated dose in vivo transcriptomic study.
- Decision type: Benefit-risk and target-risk decision makers.
- Target risk level: The specified target risk level required for the target-risk decision maker to take action.
Figure 3-1
Overview of the VOI analysis case study comparing the ETAP and THHA. The two toxicity testing and human health assessment approaches were compared across multiple exposure characteristics and decision contexts to capture the range of values for a chemicals (more...)
For components above, a range of values informed by real-world data was evaluated to understand the relative sensitivity of the results to the inputs. For the exposure and decision type, the range of values for each component were combined into a set of baseline scenarios to capture the multiple exposure characteristics and decision contexts for a chemical for which ETAP may be applicable. In addition, a separate set of sensitivity analyses (i.e., sensitivity analysis scenarios) were performed to evaluate additional considerations, including chemical characteristics and other potential sources of uncertainty. The results of the baseline and sensitivity analysis scenarios were summarized across multiple VOI metrics to bound the relative difference in value between the non-traditional and traditional toxicity testing and human health assessment process.
The report is organized to present the following information:
- Description of the toxicity testing and human health assessment methods and processes for ETAP and THHA.
- The analytical framework used for VOI analysis, as well as the decision-making paradigms used to evaluate the VOI from the two toxicity tests and human health assessment processes.
- Parameterization of the VOI analytic framework informed by real-world inputs.
- Results of the VOI analysis across baseline scenarios that are defined by the mean population exposure level, variability in exposure across the population, and the type of decision maker.
- Results of the sensitivity analysis scenarios defined by quality of exposure information, valuation of adverse health effects, costs of exposure mitigation actions, prevalence of exposure in a population, target risk level, distribution of potential toxicological potencies for an untested chemical, the degree of uncertainty reduction achieved using the ETAP, the time horizon considered, variation in the testing costs incurred, and differences in the time required to complete the testing and associated human health assessments.
- Overall conclusions from the case study and implications for application of the ETAP.
3.2. CHARACTERISTICS OF THE TESTING AND ASSESSMENT PROCESSES CONSIDERED IN THE CASE STUDY
3.2.1. TWO-YEAR RODENT BIOASSAY AND HUMAN HEALTH ASSESSMENT
The two-year rodent bioassay is a standard toxicity testing method that is used to determine adverse health effects that may arise due to prolonged and repeated exposure to a substance. Most bioassays expose rodents to at least three dose levels of the test substance or vehicle control for up to two years, beginning at 5–6 weeks after birth, and ending prior to the natural lifespan of the animal [i.e., approximately three years for rats (Huff et al. 2008), which is the typical species of choice for combined chronic toxicity and carcinogenicity studies] (Bucher 2002). At least 100 rodents (50 male and 50 female) are used for each dose group and the vehicle control group, with additional numbers added for interim sacrifices at earlier time points to investigate temporality, as the design may require. Typical studies involve analysis of clinical observations, hematology, clinical chemistry, and urinalysis on a subset of animals. In addition, changes in body weight over time, organ weights and gross examination, and histopathologic assessment of a wide variety of organs are performed to ensure adequate biological coverage. Histopathology findings are frequently subjected to additional independent review by a pathology working group to gain agreement on the endpoints and responses identified (Bucher 2002).
The process within the EPA to develop a human health assessment for existing substances involves multiple steps that have evolved over time (EPA 2014, 2022c)6. The process usually begins with problem formulation and scoping to identify the regulatory need, specific environmental or exposure conditions, and the specific assessment questions to be answered. Following problem formulation, relevant animal and human studies are compiled and evaluated for quality, consistency, and relevance. In recent years within the EPA ORD, the literature survey and study evaluations may be conducted using systematic review principles (NASEM 2021; Whaley et al. 2020). The hazard evidence is integrated for each health outcome and the studies are selected for dose-response assessment. The critical effect(s) is identified and used to derive reference values using appropriate uncertainty factors (UFs) that capture important experimental, variability, and extrapolation considerations (EPA 2002). In large federal agencies such as EPA, the human health assessments often undergo a multi-step intra- and inter-organization review process, external peer review, and public comment period. The human health assessments are then revised based on the reviews and public comments prior to final publication. The case study performed in this report assumes that a two-year rodent bioassay was used as the basis of the critical effect(s) to derive the reference value.
3.2.2. SHORT-TERM IN VIVO TRANSCRIPTOMIC STUDY AND EPA TRANSCRIPTOMIC ASSESSMENT PRODUCT (ETAP)
The ETAP was developed to provide timely information to support decision making for chemicals lacking existing or publicly accessible repeated dose toxicity studies or human evidence suitable for use as a POD and reference value derivation. The proposed methods are summarized in this document to orient the reader as to the derivation of values pertinent to the VOI case study. Detailed methods and the scientific studies supporting the development of the ETAP are available in separate EPA reports (EPA 2024b, c).
Briefly, the ETAP Scientific Support Document evaluates the relationship between transcriptomic PODs from short-term exposures and their concordance to apical PODs from traditional two-year toxicity studies in rodents. Data from 33 independent studies of over 140 chemicals with diverse physicochemical and toxicological properties demonstrated that transcriptomic benchmark dose (BMD) and benchmark dose lower confidence bound (BMDL) values, when integrated at a gene set level, were concordant with BMD and BMDL values for apical responses in traditional subchronic and chronic rodent toxicity studies. The error associated with the concordance between the transcriptomic BMD values versus apical BMD values was approximately equivalent to the inter-study variability in the repeated dose toxicity study itself. The transcriptomic and apical dose concordance was robust across different exposure durations, exposure routes, species, sex, target tissues, physicochemical properties, toxicokinetic half-lives, and technology platforms.
As described in the ETAP Standard Methods Document, the ETAP consists of three primary components with associated processes and decision points within each component. The three primary components consist of: 1) initial database searches and systematic evidence map development; 2) short-term (five-day) in vivo transcriptomic study and POD determination; and 3) assessment development and reporting. The main concepts of the ETAP are that the underlying methods and data analysis procedures are highly standardized and structured, and the decision context is narrowly focused on chemicals with no existing or publicly accessible repeated dose toxicity studies or human evidence suitable for use as a POD and reference value derivation. Candidate chemicals for ETAP are screened for publicly available repeated dose toxicity data using the EPA ToxVal database (ToxValDB). If no suitable studies are identified in the ToxValDB, then systematic evidence map (SEM) methods are used to identify and organize the research available on a specific substance (Thayer et al. 2022a; Thayer et al. 2022b). For the ETAP, a SEM is developed to identify and evaluate the literature base associated with the candidate substance for mammalian in vivo repeated dose toxicity studies or suitable human evidence. Resources searched include databases of published research (e.g., PubMed, Web of Science, ProQuest) as well as repositories of studies that may not have been peer-reviewed, such as those summarized in European Chemicals Agency (ECHA) registration dossiers or EPA’s ChemView database. In addition, searches may be conducted to discern whether studies exist in such regulatory reporting databases but are classified as confidential business information. Based on the SEM, chemicals confirmed to have no publicly available mammalian in vivo repeated dose toxicity studies or suitable human studies may be eligible for development of an ETAP.
The next component of an ETAP is a short-term (five-day) in vivo transcriptomic study and POD identification. Transcriptomics is the characterization of gene expression changes in a cell, tissue, organ, or organism of interest. When analyzed following dose-response treatment with a chemical substance, transcriptional changes provide an understanding of the signaling pathways, biological processes, and molecular functions that are disrupted and the dose at which this occurs (Thomas et al. 2007). In the ETAP, a five-day repeat dose design in both male and female rats is used as the basis for the transcriptomic study. Transcriptomic measurements for ETAP development are performed using targeted ribonucleic acid (RNA) sequencing (RNA-seq) in twelve tissues. Additional endpoints typically assessed for the purposes of the THHA, such as clinical chemistry and histopathology, are not assessed for the ETAP.
Transcriptomic BMD modeling is performed consistent with the expert-reviewed, National Toxicology Program's (NTP) Approach to Genomic Dose-Response Modeling (NTP 2018), but adapted for the targeted RNA-seq gene expression platform used for the ETAP. A comprehensive series of analyses was performed to identify and support the choices and parameters used in each step of the transcriptomic dose-response modeling process to promote detection of transcriptional changes concordant with adverse apical effects, maximize inter-study reproducibility, and minimize detection of false dose response changes (EPA 2024c). A combination of dose-response modeling parameters was identified that resulted in transcriptomic BMD values with a concordance to apical BMD values from a two-year rodent bioassay that was approximately equivalent to the combined inter-study variability from both studies. To select the transcriptomic POD, the BMDL from the gene ontology (GO) biological process class with the lowest median BMD in the most sensitive sex (male or female) and across all the tissues examined is identified. No determination of a specific type of hazard caused by the substance nor mechanistic interpretation of the gene expression changes is performed.
For the development of the assessment product and reporting, the transcriptomic POD obtained from the five-day in vivo oral exposure study is converted to a human equivalent dose (HED) using an oral dosimetric adjustment factor (DAF) based on allometric cross-species scaling (EPA 2011a). The PODHED is used in the derivation of a chronic transcriptomic reference value (TRV) through application of UFs that are consistent with traditional human health assessment guidance and practice. The quantitative values of the individual UFs and the overall composite value are the same across the individual ETAP assessments due to the standardized nature of the studies and data analysis procedures. The TRV is defined as an estimate of a daily oral dose to the human population that is likely to be without appreciable risk of adverse non-cancer health effects over a lifetime. The results from the SEM, five-day transcriptomic study, and TRV derivation are compiled and reported in a standardized ETAP reporting template, which is made available at a target date within six months of initiating the experiments. The analysis detailed in the series of ETAP documents provide scientific support for considering a short-term transcriptomic study as an alternative method for chemicals lacking existing or publicly accessible repeated dose toxicity studies or human evidence suitable for use as a POD and reference value derivation (EPA 2024b, c).
Footnotes
- 6
- VOI CASE STUDY - Value of Information Case Study on the Human Health and Economi...VOI CASE STUDY - Value of Information Case Study on the Human Health and Economic Trade-offs Associated with the Timeliness, Uncertainty, and Costs of the Draft EPA Transcriptomic Assessment Product (ETAP)
Your browsing activity is empty.
Activity recording is turned off.
See more...