COVID-19 is an emerging, rapidly evolving situation.

Get the latest public health information from CDC and research information from NIH.

U.S. flag

An official website of the United States government

Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.


ICCVAM Biennial Report 2018-2019

ICCVAM Biennial Report 2018-2019

Data Resources

As momentum grows toward adoption of alternative methods for chemical safety testing, curated data are needed to support method validation and establish scientific confidence in new approaches. ICCVAM agencies address that need by compiling data and making them publicly available.

Integrated Chemical Environment Updates

NICEATM's Integrated Chemical Environment (ICE) provides data and tools to help develop, assess, and interpret chemical safety tests. An update of ICE was launched in May 2019 that provides a home page with links to all ICE tools and detailed user guides. An update of the Search tool simplified access to chemical information, and tools were added for IVIVE and chemical characterization.

Other data updates to ICE during 2018 and 2019:

  • Expanded physicochemical property predictions to include over 720,000 chemicals using an updated set of predictive models (Mansouri et al. 2018).
  • Updated the curated HTS data coming from the Tox21 initiative to include increased curation and the most up-to-date data release.
  • Added predictions of endocrine activity using the CERAPP (Mansouri et al. 2016) and CoMPARA (Mansouri et al. 2020) models.
Unsupervised Data-driven Analysis of Tox21 Assay Data

Scientists at NIEHS are using computational data organization methods based on patterns to identify chemicals that exhibit biological properties similar to those of well-characterized toxicants in the Tox21 10K library. Users will soon be able to perform bioactivity-based correlations to compare query chemicals with well-characterized Tox21 toxicants using the Tox21 Correlation Browser. An upgrade to the Correlation Browser is in progress and an improved web tool will be available in 2020.

Results of these analyses are being used in combination with another tool, Tox21 Enricher, to help prioritize compounds for more extensive toxicological testing. The Tox21 Enricher performs neighborhood enrichment analysis to determine if one or more input chemicals have characteristics that show similar bioactivity patterns to characterized chemicals. Tox21 Enricher has been described in a publication (Hur et al. 2018).

Tox21 High-throughput Screening Assay Target Mapping

NICEATM is mapping assay targets used in Tox21 and the EPA ToxCast programs to known modes of action for developmental toxicity, acute toxicity, and carcinogenicity based on established modes of action from literature. The assay mappings are used in ICE to facilitate targeted access to the Tox21 data in ICE tools. This mapping draws on established ontologies and controlled vocabularies such as the NCI Metathesaurus and UMLS Metathesaurus to promote the interoperability of ICE data with partners.

Compilation of Human Skin Sensitization Data

To support the evaluation of non-animal approaches for skin sensitization assessment, scientists from CPSC, FDA, and NICEATM worked with industry and international collaborators to collect human predictive patch test data from approximately 1800 publications. Results considered to be sufficiently reliable were classified using GHS categories. Decision tree and weight-of-evidence approaches were used to help resolve ambiguity and discordance in individual tests for each substance. This classification approach was applied to a Cosmetics Europe reference list of 128 substances to support the evaluation of defined approaches for skin sensitization proposed for inclusion in a new OECD guideline, for which reliable classifications were obtained for 80 substances. Results of the analysis were described in an abstract (Strickland et al.) accepted for presentation at the 2020 Society of Toxicology Annual Meeting. The entire human skin sensitization patch test database will be made publicly available in the future for additional evaluation of alternative skin sensitization methods and development of new models.

Extraction and Annotation of Legacy Developmental Toxicity Study Data

To support the evaluation of non-animal approaches for developmental toxicity assessment, NICEATM scientists extracted information from about 250 NTP legacy prenatal developmental toxicity animal studies and a subset of about 50 studies submitted to ECHA that were deemed high-quality by NTP subject matter experts. Study details extracted included species, strain, administration route, dosing duration, and treatment related effects. Efforts are underway to standardize the effects extractions by applying controlled vocabularies and ontologies to facilitate computational analyses and integration with other structured databases such as EPA’s ToxRefDB. 

Variability Analysis of In Vivo Data to Set Performance Thresholds for Alternative Methods

To better characterize the reproducibility of the in vivo rabbit skin irritation assay, NICEATM assessed variability within a data set of over 3000 results for over 700 chemicals tested at least twice. The in vivo rabbit skin irritation assay has historically been the benchmark against which NAMs have been compared. However, a limiting factor in identifying a full replacement for the in vivo method could be the variability inherent to the subjective scoring of responses in the rabbit test. The NICEATM analysis found that chemicals classified as moderate irritants at least once were classified as mild irritants or non-irritants over 40% of the time when tested repeatedly. The level of variability was greatest between mild and moderate irritant classifications. This analysis indicates that the level of variability present in the rabbit skin irritation test should be taken into consideration when evaluating the performance of nonanimal alternative methods.

NICEATM also assessed the reproducibility of the rat acute oral lethality test using a comprehensively curated dataset of over 2000 chemicals. Quantitative and categorical analyses were conducted and a global confidence interval was established to characterize the inherent variability of the test method. These analyses revealed that independent studies can yield LD50 values orders of magnitude apart, which can have significant implications from a regulatory perspective due to the resulting impact on hazard classification and labeling. Similar to the analysis of the rabbit skin irritation test, no physicochemical properties or differences in test method protocol were identified that could readily explain differences between study results. The resulting publicly available dataset can be used for modeling, and the calculated confidence interval can be applied when assessing performance of NAMs. A publication describing this work is planned for submission in 2020.

Establishment of Ontologies for Zebrafish Developmental Toxicity Screening Studies

NICEATM is coordinating collaborative projects to establish ontologies for zebrafish screening. Three laboratories tested the same 90-chemical set using similar study designs and morphology assessments. Data from these studies are being curated and assessed to refine testing approaches and support development of informatics resources. A poster describing this project (Ceger et al.) was presented at the 2019 annual meeting of the Society for Birth Defects Research and Prevention (formerly the Teratology Society).

To further define ontologies for zebrafish screening, NICEATM coordinated an online evaluation of heterogeneity in terms used to describe zebrafish phenotypes following chemical exposure. Zebrafish images were posted online for researchers to evaluate the phenotypes using their in-lab terminology. Results were collected, compiled, and terms mapped to the Zebrafish Phenotype Ontology. A second online evaluation is scheduled for early 2020 and will employ controlled vocabulary; results of this evaluation will be presented in a NICEATM webinar and published.

Retrospective Analysis of Acute Fish Toxicity Data

To assess potential hazards to wild fish species, EPA currently requires testing in each of three different fish types: warmwater, coldwater, and marine/estuarine. NICEATM and the EPA Office of Pollution Prevention and Toxics are extracting and evaluating acute fish toxicity data from pesticide safety studies submitted to the EPA office. These data will be analyzed to determine whether one or more of the three fish types can be eliminated from testing requirements. Extracted data will be submitted to EPA’s Toxicological Reference Database.

Retrospective Analysis of Triple-pack Studies for Dermal Absorption

To provide an estimate of dermal absorption, the EPA Office of Pesticide Programs combines data from in vivo rat, in vitro rat, and in vitro human dermal absorption studies, commonly referred to as the “triple pack”, to calculate a chemical-specific dermal absorption factor. The dermal absorption factor estimates the percentage of the external dose that will cross the skin barrier and end up in the systemic circulation. To assess the feasibility of using data from in vitro studies alone to estimate this value, EPA and NICEATM, in conjunction with the Health Canada Pest Management Regulatory Agency and industry stakeholders, are conducting a retrospective analysis of triple pack reports for agrochemicals completed between 2005 and 2015. Data were extracted from study reports provided to NICEATM by EPA and agrochemical products companies ADAMA, BASF, Bayer, Corteva, FMC, Syngenta, and Valent. The aim of this analysis is to determine whether the rate of in vitro absorption is equal to or greater than the rate of in vivo absorption. Absorption values from in vitro rat studies are also being compared to in vitro human studies to determine the differences in permeability between rat skin and human skin. A stakeholder meeting was convened in May 2019 at EPA to discuss the preliminary findings from this review and to identify and discuss additional information needs for both U.S. and Canadian regulators to finalize the assessment. A paper detailing this work is planned for submission in 2020.

Human BioMolecular Atlas Program

In October 2018, NIH issued its first set of research funding awards for the Human BioMolecular Atlas Program. This is an open, global framework to support the research community’s efforts to map the adult human body at the level of individual cells. The project will award $54 million between 2018 and 2021 to support:

  • Generating, standardizing, and validating data sets on cell organization and variability.
  • Developing new tools and techniques to construct high-resolution tissue maps.
  • Coordinating program activities, managing Human BioMolecular Atlas Program data, and building an atlas of tissue maps.
Changes to NLM TOXNET Resources

NLM created the Toxicology Data Network (TOXNET) in 1985 to improve access to NLM’s online chemical and toxicity databases. As part of the reorganization associated with its current strategic plan, NLM retired TOXNET in December 2019. Most of TOXNET’s databases have been incorporated into other existing NLM resources. Information about how to access these databases is available on the NLM website, and more detailed information is available in an NLM Technical Bulletin issued in November 2019. New locations for other NLM resources such as ToxTutor and ALTBIB are under discussion. NLM will conduct webinars and other presentations in 2020 to inform former TOXNET users how they can continue to access these resources.