ICCVAM Biennial Report 2024-2025

Profiling chemical effects on transcription factor activity can help characterize the mechanisms by which chemicals may disrupt biological systems. The Attagene cis-FACTORIAL™ assay uses a reporter system to detect transactivation of 46 transcription factors, providing a quantitative assessment of chemical effects. A later version of this assay, CYP-FACTORIAL™, added nine key CYP450 enzymes to enable the evaluation of chemical effects on transcription factor activity with and without CYP-mediated Phase 1 metabolism. These two complementary assay formats support a better understanding of whether CYP-mediated oxidation results in altered bioactivity profiles between parent and metabolite compounds.

To explore the application of this system to toxicity prediction, NIEHS and Attagene collaborators tested 24 chemicals of regulatory concern using both systems. Profiling across all 46 transcription factors produced chemical toxicity signatures, and applying a biological read-across approach identified chemicals with similar effects. Additionally, comparing the profiles of “toxic” vs. “nontoxic” chemicals yielded insights into the potential biological mechanisms underlying adverse outcomes. Dose–response modeling using the tcplfit2 R package (Sheffield et al. 2022) further revealed how the addition of CYP enzymes altered bioactivity and response sensitivity across chemicals and transcription factors. This study exemplifies how integrating metabolism into a multiplexed in vitro assay system can provide additional mechanistic insights needed to understand chemical-elicited bioactivity, thereby facilitating the development of human-relevant test systems. A manuscript is in preparation and is planned for submission in 2026.

Scientists at NIEHS are using zebrafish (Danio rerio) embryos as a test model to study how chemical exposure might impact development. Zebrafish are useful for developmental studies because their embryos are transparent, develop quickly, and respond similarly to humans. However, different laboratories use different testing methods, and these discrepancies make it difficult to compare results, limiting the wider use of zebrafish embryos for regulatory chemical safety testing. To address the discrepancies in testing methods, the Systematic Evaluation of the Application of Zebrafish in Toxicology (SEAZIT) initiative was created. This project aimed to improve understanding of what aspects of experimental protocols, data analysis, and reporting causes variability among zebrafish testing across laboratories and help identify ways to make methods more consistent.

In a study involving three laboratories (Hamm et al. 2024), 42 chemicals were tested under different conditions; for example, using embryos with the chorion intact versus removed or using static versus static renewal chemical exposure. The study compared how these methods influenced observed developmental toxicity. The results showed good agreement across laboratories, with 33 of the 42 chemicals (~79%) producing the same activity outcomes. However, differences in chemical potency linked to varying laboratory protocols underscored the need for method standardization. The findings of this study provide valuable insights into the application of standardized ontologies for cataloging, comparing, and sharing developmental toxicity data. This work could help improve the reliability and regulatory acceptance of zebrafish embryos as a tool for evaluating developmental toxicity.

An interactive web resource, SEAZIT Data Integration and Visualization Enabling Resource (SEAZIT-DIVER), has been developed to allow the public to explore and visualize SEAZIT data. This platform also provides methodological details, information on test chemicals, tools to evaluate data quality, and options to download data for additional exploration.

Scientists at NCATS, EPA, and NIEHS are studying how chemicals interact with important biological systems to better predict potential risks, such as developmental toxicity. One key system involves all-trans retinoic acid, a molecule produced from vitamin A, which is crucial for regulating development and cell growth. Levels of all-trans retinoic acid are tightly controlled in the body, and the molecule is broken down by specialized enzymes called CYP26 (subtypes A1, B1, and C1) to maintain balance. This research focused on the detection of chemicals that inhibit CYP26A1, one of the enzymes involved in this process, using a new experimental method.

Researchers developed a high-throughput screening (HTS) assay using a system called P450-Glo (Sakamuru et al. 2024). This cell-free system quickly identifies chemicals that interfere with CYP26A1 activity. The assay performed well in terms of consistency, reliability, and ability to distinguish signals, with a signal-to-background ratio of 25.7 and a Z-factor of 0.7 (indicating strong performance). To validate the method, 39 chemicals were tested, including known CYP26A1 inhibitors, compounds related to retinoids, and control chemicals. The assay accurately confirmed known inhibitors and identified new ones, such as chlorothalonil, prochloraz, and SSR126768. This successful validation highlights the reliability of the new assay in the detection of chemicals that may disrupt all-trans retinoic acid balance, which is critical during development. This innovative tool could be pivotal for studying developmental toxicity and ensuring chemical exposure safety, particularly for substances that impact embryonic development.
 

To reduce and replace the use of animals in immunotoxicity assessments, the NIEHS Division of Translational Toxicology collaborated with NICEATM to develop a comprehensive human whole-blood NAM battery for in vitro assessment of immune function. The system uses human whole blood diluted in culture media, to which an immune stimulant such as anti-CD3/CD28 antibodies or viral peptides may be added. The model immunotoxicants dexamethasone and benzo[a]pyrene were used to evaluate the utility of the screening assay. Following in vitro culture, immune cells were identified and counted by flow cytometric immunophenotyping and assessed for natural killer cell activity and T cell activation. Proinflammatory cytokine levels were measured in culture supernatants. In vitro treatment with dexamethasone resulted in concentration-dependent suppression of cytokine production, natural killer cell activity, and T cell activation. Treatment with benzo[a]pyrene caused suppression of cytokine production, but immunosuppression only occurred following metabolic activation by addition of S9 fraction enzymes. Advantages of this NAM battery include the ability to assess adaptive immunity and relate immunotoxicity to resistance to infectious diseases. These studies demonstrate that in vitro immunotoxicity assays can measure effects that reflect known outcomes from in vivo studies. This multiendpoint human whole-blood NAM battery may be useful for screening compounds for immunotoxicity hazard identification in a manner that reduces reliance on animal systems and increases translatability to humans.

Exposure to environmental chemicals, such as pesticides and heavy metals, may contribute to neurodegenerative disorders by inducing inflammation in nervous system tissues. EPA, NCATS, and NIEHS collaborated to develop in vitro models using human induced pluripotent stem cell-derived microglia to identify potential neurotoxicants (Yang et al. 2025). The assays measured cytokine secretion, which is a marker of inflammatory response. The cytokine response patterns in these cells closely resembled those under in vivo conditions, with significant increases in interleukin 6 and tumor necrosis factor-alpha levels. These initial results were used as the basis for development of a homogeneous, time-resolved fluorescence HTS assay platform in a 1,536-well plate format for screening of environmental chemicals. This assay was used to test nine environmental chemicals having evidence of cytokine-inducing potential in other in vitro assays. While all nine tested agents stimulated interleukin 6 and tumor necrosis factor-alpha production, three chemicals showed significant stimulation of both cytokines. This study establishes a reliable high-throughput platform for detecting inflammatory effects of environmental toxicants in a microglial cell assay, data from which may provide valuable insights into their neuroinflammatory potential and possible contributions to development of neurodegenerative disorders.

The early detection of neurological disorders is crucial for effective intervention and treatment, but current methods of studying brain tissue directly, such as biopsies, are risky and invasive. Scientists at NIEHS are investigating ways to measure biomarkers, molecules found in bodily fluids such as blood or cerebrospinal fluid, to monitor brain health. One promising biomarker is cell-free DNA, fragments of genetic material released by cells, which could provide insights into brain function and disease. However, understanding how cell-free DNA from brain tissue changes in response to damage or disease has been challenging because many tissues contribute to cell-free DNA levels in the body.

To address this challenge, NIEHS researchers utilized cerebral organoids, which are miniature, lab-grown models of the human brain. These organoids offer a controlled system to track cell-free DNA changes specifically related to brain tissue. The study demonstrated that cerebral organoids release enough cell-free DNA to be analyzed using advanced techniques including droplet digital PCR and whole-genome sequencing. Importantly, when researchers examined the genetic information in the cell-free DNA, they found that many of the fragments were associated with biological processes linked to neurodevelopment and neurological disorders.

This work highlights the potential of cerebral organoids to be powerful tools in the study of cell-free DNA biomarkers. These biomarkers could one day help identify early signs of neurodevelopmental and neurological conditions, paving the way for earlier diagnoses and better-targeted treatments. The findings provide a promising step forward in understanding how cell-free DNA reflects brain health and disease.

The U.S. Army Chemical Biological Center (CBC) has established cooperative research and development agreements with the tissue chip manufacturers Emulate, Inc., and TissUse GmbH to apply their products to investigations of aerosol toxicity.

Aerosolization of chemical agents is a health concern that presents unique technical challenges. When aerosols are generated in specific environments, the agent can be deposited on surfaces and then reaerosolized into the air due to environmental factors, foot traffic, or passing vehicles. This is particularly likely with substances that are easily disseminated, persist in the environment, and are hydrophobic and thus remain in the surface soil environment. To investigate this concept, CBC scientists developed a novel system to simulate reaerosolization of chemical agents in soil with a high sand content. A vibration system with air inlet and outlet ports was developed to allow sampling of the reaerosolized soil-agent to determine a worst-case scenario concentration. Reaerosolized agent concentrations were identified and tested for toxicity in Emulate organ-chip models.

Toxicity of substances of concern will also be assessed in collaboration with TissUse using an open-top multiorgan chip system that can be exposed to an aerosolized agent or compound of interest. The system functions with an established TissUse microfluidic delivery system and a novel aerosol chip exposure apparatus called the Independent Holistic Air–Liquid Exposure System (InHALES). This system accurately replicates the entire human respiratory tract in vitro. The InHALES system will be implemented at CBC in early 2026, and feasibility studies will begin in summer 2026 to better understand aerosol maturation and dissemination.

Additionally, CBC leveraged success with bespoke aerosol organ-chip delivery systems to establish a cooperative agreement with Emulate. The focus of the collaboration is to provide user input on Emulate’s A1 open-top lung chip and to adapt the chip to CBC’s existing aerosol generation infrastructure.

CBC recently began work that aims to characterize exposure routes in vitro. Human lung chips will be exposed to toxic compounds in liquid directly on the tissue or through aerosols generated by the VITROCELL® Cloud in vitro exposure system (VITROCELL GmbH).

Conventional in vitro models of the gastrointestinal tract, such as Caco-2 monolayers, lack the cellular complexity, mechanical environment, and tissue architecture needed to accurately capture absorption, permeability, and barrier function in vivo. Studies by the FDA National Center for Toxicological Research (NCTR) explored how these systems could be improved to support drug development and safety testing.

• Ongoing studies are comparing animal and nonanimal models, including human intestinal tissue-based MPS, to evaluate the use of NAMs for assessing antiretroviral drug toxicity. This work will be presented at the 2026 MPS World Summit (Yanamadala et al. Abstract 797). Planned additional studies in collaboration with NIEHS will use multiorgan chip platforms to assess the extraintestinal impacts of the microbiome on the liver and heart.
• NCTR researchers have demonstrated the importance of continuous-flow systems for maintaining physiological relevance. Using resected human intestinal tissue, continuous-flow gut-on-a-chip models showed improved preservation of tissue function, greater sensitivity to dose-dependent responses, and distinct transcriptional, inflammatory, and drug-transport profiles compared to static systems. This project was presented at the 2025 Gordon Research Conference for Cellular and Molecular Mechanisms of Toxicity.
• To complement host-focused models, NCTR scientists developed a rotary culture-based NAM ex vivo system to study interactions between the microbiome and xenobiotics. This dynamic system enables controlled, mechanistic evaluation of microbial responses and has been directly compared to animal models to assess translational relevance. A presentation at the 2024 MPS World Summit (Khare et al. Abstract 481) described a bioreactor model that keeps microbiota and chemicals in motion, which mimics gastrointestinal conditions better than static systems. The study noted time- and dose-dependent changes in bacterial communities and metabolite production after exposure. The model may improve gastrointestinal toxicology risk assessment, support regulatory decision-making, and advance FDA goals to modernize toxicology testing.

Technical characterization is a key component of the development of NAMs for widespread use. Technical characterization includes assessing sources of variability in a NAM, designing an assay to include relevant control measurements, evaluating the range of test substances for which the assay can be used, and developing a suitable statistical data analysis approach. A key aspect in demonstrating the scientific validity of a NAM is that the assay is sufficiently well-characterized from a technical perspective to ensure that it is robust, reliable, and reproducible. Researchers at NIST applied these principles to develop two assays: a nitric oxide nanobio assay and an oral mucosal tissue irritation assay. 

• Engineered nanomaterials may induce oxidative stress in cells, potentially leading to cytotoxicity and DNA damage. This can be assessed using assays that measure cellular nitrites as a surrogate for nitric oxide production via macrophage activation in vitro. NIST collaborated with FDA to refine an assay used for this purpose to support its broader application (Fahmi et al. 2026). The assay has been recognized as a standard test method by the American Society for Testing and Materials.
• To support the broader application of an oral mucosal tissue irritation assay for the testing of biological effects from dental materials, NIST researchers utilized 3D human oral epithelium models to characterize irritation mechanisms and improve measurement robustness (Gutierrez et al. 2024). The team identified variability sources using statistical modeling and control charts to track assay performance over time. Robustness was tested under multiple conditions, highlighting the importance of pipetting procedures, exposure durations, and complementary measurements, such as transepithelial electrical resistance. Recommendations for protocol refinement and standardization have been made to ensure that this assay aligns with quality assessment procedures.

Error-corrected sequencing technologies are emerging as highly sensitive alternatives to traditional transgenic rodent gene mutation assays because they can detect extremely rare genetic changes throughout the entire genome. These technologies can also reveal detailed patterns in these changes, helping scientists better understand how and why mutations occur. In separate efforts, FDA NCTR and the NIEHS Division of Translational Toxicology have been exploring applications of these technologies in studies using human cell lines and in vivo rodent models. These projects address key methodological considerations for assay design.

• Molnupiravir, an antiviral prodrug used to treat COVID-19, can induce mutations in bacterial and mammalian cell cultures, raising concerns about patient safety. McKinzie et al. (2026) applied a novel error-corrected sequencing pipeline to peripheral blood cells collected retrospectively from molnupiravir-treated and untreated COVID-19 patients. No differences in mutation frequency, type, or mutational signatures were detected between groups, suggesting, within the limits of the study, that the authorized therapeutic dose carries no detectable mutational risk in treated patients.
• Commercially available TK6 and L5178Y cells have elevated background mutation frequencies compared to freshly derived single-cell clones, likely from accumulated replication errors during long-term culture. These elevated backgrounds reduce the relative signal upon mutagen exposure, potentially confounding error-corrected sequencing-based mutagenicity assessments. Miranda et al. (2025) used error-corrected next-generation sequencing to quantify the background mutation frequencies of freshly derived clonal TK6 cell populations and commercially available cell lines. The research team derived an alternative data analysis approach based on absolute mutation frequency changes within each cell population. This approach may provide more comparable results for commercial and clonal populations.

Standard in vitro genotoxicity assays have limitations that can prevent broader use in replacing in vivo assays. Assays using cell lines that lack metabolic capability must be supplemented with exogenous rat liver S9 fractions, which do not accurately reflect human biotransformation. Additionally, conventional in vitro genotoxicity protocols have typically been of limited utility in predicting effects of longer exposures. NCTR researchers have been applying advances in human-relevant genotoxicity testing that address these gaps through three complementary approaches: engineering metabolically competent TK6 cell lines to better recapitulate liver-like metabolism, using 2D and 3D HepaRG liver cell models to evaluate nitrosamine drug impurities, and applying extended-exposure HepaRG protocols to improve detection of low-level mutagens. Together, these approaches support a shift toward human-based systems for regulatory genotoxicity assessment.

• Li et al. (2025) developed TK6-4CYP, a human lymphoblastoid cell line coexpressing four liver CYP enzymes (CYP2A6, 2E1, 2C19, and 3A4) with metabolic capability, enabling bioactivation of indirect genotoxicants without reliance on a rodent S9 fraction. The cell line correctly identified known promutagens in standard genotoxicity assays. Promutagens such as N-nitrosodimethylamine and benzo[a]pyrene are not mutagenic in their original form but can be metabolized into mutagens. TK6-4CYP expands the existing panel of single-CYP TK6 lines and offers a human-relevant, metabolically competent platform for regulatory genotoxicity testing.
• Seo et al. (2025) assessed 10 nitrosamine drug substance-related impurities for genotoxicity in 2D and 3D HepaRG cultures. They used the CometChip assay, which can detect DNA damage at the level of individual cells, and flow cytometry-based micronucleus assays, which assess the presence of small extracellular nuclei containing chromosomal fragments. Five compounds that previously showed positive results in the enhanced Ames test induced DNA damage in both models, while five impurities with negative results in the Ames test showed no activity. The 3D spheroid format consistently outperformed 2D cultures in sensitivity, supporting its use as a more physiologically relevant system for drug impurity hazard evaluation.
• Building on prior work in which acute cytotoxicity limited mutation detection in HepaRG cells, Guo et al. (2025) used 7- and 14-day low-concentration N-nitrosodimethylamine exposures in both 2D and 3D formats to allow mutation accumulation. Extended treatment produced concentration-dependent increases in DNA damage, micronucleus formation, and mutation frequency, with 3D spheroids showing greater sensitivity. The results support longer-duration protocols in metabolically competent human liver cells as a practical approach for the detection of mutagenicity at environmentally or pharmaceutically relevant concentrations.

Anchorage-independent growth, or the ability of cancer cells to grow without being attached to a surface, is a hallmark of tumor development. While 3D spheroid cultures are valuable for studying this phenomenon, their use in research has been limited due to challenges in process scaling. NCI researchers have addressed this issue by developing new, cost-effective methods to grow tumor spheroids and organoids. One key focus of this research is the KRAS gene, mutations of which are linked to the development of several cancers, including lung adenocarcinoma. Scientists have successfully created scalable tumor spheroid cultures for studying KRAS-mutant epithelial cancers. These spheroids provide a more realistic model than traditional 2D cell cultures, allowing researchers to examine unique molecular and genetic features specific to 3D cancer growth.

NCI researchers also developed mouse organoid cultures using alveolar Type 2 cells, which are natural lung stem cells and likely the starting point of lung adenocarcinoma. The organoid cultures include both healthy and KRAS-mutant alveolar type 2 cells and have been used to study how the mutated KRAS gene drives cancer, including its effects on growth, signaling, gene activity, and stem cell behavior. Efforts are underway to establish similar organoids for human alveolar type 2 cells.

These advanced models help researchers better understand the role of the mutated KRAS gene in cancer development and its behavior in realistic 3D environments. This research helps support improved treatments, particularly targeted therapies for KRAS-related cancers, and enables tumor behavior to be further studied in a controlled yet realistic setting.

Cancer development research requires better methods to model the complex gene mutations and expression changes found in human tumors. NCI researchers have made significant progress in using advanced tools such as RNA interference and CRISPR/Cas9 genome editing to modify multiple genes at once, allowing more accurate simulations of cancer cell behavior in the lab. Multiplexed RNA interference is an approach that enables scientists to silence up to seven genes simultaneously in human cancer cells. This approach helps uncover interactions between genes and identify redundant functions, providing insights into how genes work together to drive cancer. NCI researchers have also developed a multiplexed CRISPR/Cas9 system that can knock out three genes at the same time in human cancer cells. They are now adapting this CRISPR method for use in organoid cultures, which are 3D models that mimic tissue structure. This allows analysis of the effects of various genetic changes in organoids derived from both mice and humans, which better represent real-world tumor behavior. Additionally, the NCI team has created specialized CRISPR libraries for targeted genetic screens. These libraries are organized based on the functions of specific genes, making them highly flexible tools for studying cancer-related pathways. Finally, efforts are underway to develop methods to precisely edit genes in mouse organoids. This innovation allows researchers to rapidly create organoids with specific cancer-causing mutations, enabling detailed studies of tumor behavior and drug responses in lab-based models. These tools and techniques provide powerful methods to study the complex genetic environments of cancers and assess potential treatments, promoting the development of more effective cancer therapies.

The Patient-Derived Models Repository, hosted by NCI, provides resources to support cancer research and drug discovery. In addition to patient-derived xenografts, it offers NAMs such as patient-derived in vitro tumor cell cultures and patient-derived organoids. These specialized NAMs retain characteristics of the original tumors, making them valuable for understanding cancer biology and testing potential treatments. 

The repository includes a diverse range of models developed from various types of cancer, reflecting both common and rare tumor types. This diversity helps researchers study different genetic and molecular aspects of cancer and investigate their influence on disease progression and treatment response. Additionally, the repository emphasizes quality control in the production and characterization of these models to ensure researchers using them to obtain high-quality, reproducible results. The repository also offers detailed information about the models themselves. This includes their sources, tumor types, and other important characteristics, such as genomic data. Researchers can query the repository to find models specific to their study needs.

By using these well-characterized models, scientists can gain insights into cancer biology, identify new therapeutic targets, and evaluate the effectiveness of experimental treatments in settings that closely resemble human cancers. These tools ultimately facilitate advancements in cancer treatment and personalized medicine.