3Rs (Replacement, Reduction, Refinement)

The foundational framework for ethical and scientifically sound use of animals in research, first articulated by Russell and Burch in 1959: Replacement of animals with non-animal methods, Reduction in the number of animals used to obtain reliable and useful information, and Refinement of animal use to minimize pain, suffering, distress, or lasting harm.²¹ The 3Rs are the ethical motivation for New Approach Methodology (NAM) development and are formally embedded in European Union Directive 2010/63 and in the policies of U.S. agencies including NIH and FDA.<sup>2,3

‍</sup>NAMs most often address Replacement, the first R that should be considered in any project, but can sometimes support Reduction or Refinement by minimizing the number of animals within a still required animal study or reducing their pain and distress.  Therefore, NAMs can advance all three Rs, improving both the quality of the science and animal welfare.

‍Relevant Links:
‍The 3Rs — NC3Rs
‍The 3Rs – 3RsC

Adverse Outcome Pathway (AOP)

A structured, biology-based representation of existing knowledge linking a Molecular Initiating Event (MIE) to a series of measurable Key Events (KEs) across levels of biological organization, culminating in an adverse outcome (AO) of regulatory concern at the organism or population level.²² AOPs are chemical-agnostic: they describe the biology of injury, not any one compound.

AOPs are a conceptual backbone linking NAM measurements to regulatory endpoints, although not every biological area from which NAMs are built yet has an associated AOP. Where one exists, a well-developed AOP identifies the KEs a NAM could measure, the quantitative response–response relationships between them, and the conditions under which in vitro perturbation reliably predicts in vivo adversity. AOPs are curated in the OECD-hosted AOP-Wiki.<sup>22,23

‍</sup>Relevant Links:
‍AOP-Wiki (OECD)
‍OECD AOP Programme

Air-Liquid Interface (ALI)

A culture configuration in which the apical surface of a cell bilayer is exposed to air while the basolateral surface remains immersed in medium, typically on a permeable support (Transwell). ALI cell culture enables mature, physiologically relevant epithelial models of the airway, alveolus, skin, intestines and cornea, where differentiation into ciliated, mucus-producing, or keratinized phenotypes does not occur, or is vastly reduced, under submerged culture.

ALI is commonly combined with primary or iPSC-derived epithelial cells in Transwell inserts or integrated into MPS. ALI enables direct application of aerosolized test articles (inhaled drugs, vapors, particulates) in toxicology, and is a defining feature of OECD-accepted skin irritation and corrosion test methods (e.g., TG 431, TG 439).^9,19

Analytical Validation

Establishing that the performance characteristics of a test, tool, or instrument are acceptable for its intended purpose. Analytical validation addresses accuracy, precision, sensitivity, specificity, reproducibility, robustness, and reference (dynamic) range. For a NAM, analytical validation demonstrates that the assay reliably and reproducibly measures what it is intended to measure, independent of whether those measurements are biologically or clinically meaningful.<sup>1

‍</sup>For in vitro, ex vivo, and microphysiological systems, analytical validation typically addresses inter- and intra-laboratory reproducibility, lot-to-lot variability consumables, transferability across operators, and fitness of readouts (e.g., TEER, imaging, omics) for the intended context of use.<sup>19,20

‍</sup>Relevant Links:
‍BEST Resource — Analytical Validation
‍OECD GIVIMP Guidance Document

Applicability Domain

‍The region of chemical, biological, or physicochemical space within which a model — typically a (Q)SAR or other in silico model, but the concept applies equally to in vitro NAMs — produces predictions of acceptable reliability. Substances outside the applicability domain should not be evaluated with the model, or predictions should be flagged as extrapolations.<sup>27

‍</sup>Applicability-domain definition is one of the OECD Principles for the Validation of (Q)SAR Models. For in vitro NAMs, analogous considerations include concentration range, solubility in culture medium, cytotoxicity interference, and chemical properties (volatility, reactivity, PDMS absorption) that can place substances outside the reliable operating range of the assay.<sup>27,13

‍</sup>Relevant Links:
‍OECD — Guidance Document on the Validation of (Q)SAR Models

Assembloid

A three-dimensional culture formed by fusing two or more region-specific organoids — or an organoid with a defined population of migratory or connective cells — to reconstitute interactions between tissue compartments. Assembloids extend organoid technology to circuits and multi-tissue interfaces that cannot form within a single organoid, such as cortico-striatal projections, neuromuscular junctions, or tumor–stroma interfaces.³¹

Basement Membrane Extract (BME); e.g., Matrigel

A solubilized basement-membrane preparation — most often supplied commercially as Matrigel (Corning), whose brand name is frequently used generically for the material — rich in laminin, collagen IV, entactin/nidogen, and heparan sulfate proteoglycans, together with a variable complement of growth factors. Basement membrane extract (BME) is derived from the Engelbreth-Holm-Swarm (EHS) mouse sarcoma, gels at 37 °C, and is the most widely used extracellular matrix for organoid culture, 3D invasion assays, and many CIVMs.<sup>32

‍</sup>Its murine origin, batch-to-batch variability, and undefined composition are recognized limitations, especially for regulatory-grade NAMs. Chemically defined hydrogels (e.g., PEG-based, recombinant laminin–entactin) and decellularized ECMs are increasingly used as alternatives.^30,32

Biological Validation

The process of establishing that a measurement or assay — including a NAM — is biologically meaningful and reflects the physiological, pathological, or pharmacological process it is intended to represent. In the BEST framework, the closely related concept of clinical validation asks whether a biomarker is reliably associated with the presence or state of a disease or biological condition of interest; for NAMs, biological validation extends this idea to whether the model system recapitulates the species-specific biology relevant to the decision being made.<sup>1

‍</sup>Biological validation is context-of-use dependent. In a NAM context, an organ, organoid or MPS may be biologically validated for one application (e.g., predicting CYP3A4 induction) and not another (e.g., predicting idiosyncratic hepatotoxicity). Evidence typically includes benchmarking against reference compounds, comparison to clinical or ex vivo data, and mechanistic plausibility.^3,20

Biomarker

A defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes, or biological responses to an exposure or intervention, including therapeutic interventions. Molecular, histologic, radiographic, or physiologic characteristics are types of biomarkers. A biomarker is not an assessment of how an individual feels, functions, or survives.<sup>1

‍</sup>BEST recognizes eight biomarker categories: susceptibility/risk, diagnostic, monitoring, prognostic, predictive, pharmacodynamic/response, safety, and surrogate endpoint. In a NAM context, biomarkers are the measurable outputs (e.g., albumin secretion, TEER, transcriptomic signatures, troponin release, signature of neural activity) that translate the biology captured by the model into interpretable data for a regulatory or research decision.<sup>1,9

‍</sup>Relevant Links:
‍BEST Resource — Biomarker chapter

Bioprinting (3D Bioprinting)

‍Additive-manufacturing techniques — extrusion, inkjet, laser-assisted, and stereolithographic — that pattern cells, biomaterials, and bioactive factors layer by layer to produce three-dimensional tissue constructs with controlled geometry and composition.²⁹ Bioprinting enables precise spatial placement of multiple cell types and vascular-like channels that are difficult to achieve with self-assembling organoid or spheroid approaches, while often incorporating matrices with different mechanical properties.

In the NAM context, bioprinting is used to produce reproducible tissue constructs such as those from liver, skin, kidney, or tumor tissues for drug efficacy and safety testing, and to fabricate scaffolds or chambers that complement MPS. Key considerations include bioink rheology, cell viability during printing, post-print maturation, and scalability for screening.^29,36

Body-on-a-Chip (Multi-Organ Chip)

‍An extension of the MPS concept in which two or more organ-specific tissue compartments are interconnected by a shared circulating medium, enabling study of inter-organ communication, first-pass metabolism, systemic distribution, and organ-to-organ toxicity. Typical configurations link liver with intestine, heart, kidney, or tumor compartments.<sup>7,36

‍</sup>Multi-organ chips are central to mechanistic pharmacokinetic / pharmacodynamic (PK/PD) modeling with NAMs, because systemic drug exposure in a single-organ chip is difficult to replicate physiologically. Scaling allometry — matching tissue mass, flow rate, and residence time to physiology — is an active area of methodological research.³⁶

Cell Painting

Cell Painting is a widely used, openly published high-content imaging (HCI) assay that stains six cellular components — the nucleus, endoplasmic reticulum, nucleoli and cytoplasmic RNA, actin/Golgi/plasma membrane, mitochondria, and cytoplasm — with multiplexed fluorescent dyes and extracts roughly 1,500 morphological features per cell using automated image-analysis software such as CellProfiler. Because the assay is unbiased and does not require a predefined mechanistic hypothesis, Cell Painting can classify compounds by mechanism of action, flag off-target and cytotoxic effects, and support read-across between structurally related chemicals; the JUMP Cell Painting Consortium has generated one of the largest public image-based profiling datasets built on the method. HCI is a primary readout for many NAMs because it is non-perturbative to the system, sensitive to subtle phenotypes, and scales to thousands of conditions.³³

CIVM (Complex In Vitro Model)

‍An umbrella term used by the U.S. FDA and others for in vitro systems that go beyond traditional 2D monoculture to more faithfully represent in vivo tissue structure and function. CIVMs include 3D cultures, co-cultures, organoids, spheroids, and microphysiological systems. The term is deliberately broad and is used in regulatory dialogue to describe NAMs that capture key in vivo features such as tissue architecture, cell–cell interactions, and mechanical or flow cues.<sup>9,7

‍</sup>CIVM is closely related to but broader than MPS: every MPS is a CIVM, but not every CIVM (e.g., a static organoid culture) is an MPS. Regulatory submissions increasingly use CIVM to encompass the full range of advanced non-animal models being qualified for specific contexts of use.<sup>9

‍</sup>Relevant Links:
‍FDA — Advancing Alternative Methods

Co-culture

Any cell-culture system containing two or more distinct cell types cultured together — either in direct contact (same compartment) or separated by a permeable membrane or hydrogel. Co-cultures are used to recapitulate paracrine signaling, cell–cell contact, and tissue microenvironments (e.g., hepatocytes with non-parenchymal liver cells, neurons with astrocytes and microglia, epithelial cells with immune cells). Co-cultures are often multiple cell types of the same tissue but can also involve culturing multiple cell types from different tissues (e.g. lung epithelial cells with peripheral immune cells).

Co-culture is a defining feature of most CIVMs, including organoids and MPS, and is a prerequisite for modeling biology driven by stromal, vascular, or immune components. Quality control considerations include ratio stability, differential proliferation rates, media compatibility, and selective identification of each population in downstream readouts.^7,19

Context of Use (COU)

A statement that fully and clearly describes the manner and purpose of use of a drug development tool (DDT). The COU specifies the setting in which the tool will be used (disease, population, decision to be informed), the specific role it plays, and the interpretive framework for its measurements. Qualification decisions — including those for NAMs — are always tied to a specific COU, never to the tool in the abstract.<sup>16

‍</sup>A narrow, well-justified COU (for example, “use of MPS X to rank-order compounds for hepatic clearance in early candidate selection”) is typically easier to qualify than a broad one. Expansion of a COU after initial qualification requires supplementary evidence proportionate to the new use.^16,17

Decellularized Extracellular Matrix (dECM)

The native, tissue-specific extracellular matrix remaining after cells have been removed from an organ or tissue by physical, chemical, or enzymatic decellularization. dECM preserves structural proteins (collagens, elastin, laminins), glycosaminoglycans, and tissue-specific bound factors, and is used as a scaffold or as a solubilized hydrogel (“dECM ink”).<sup>30

‍</sup>In NAMs, dECMs can improve the biological fidelity of in vitro models compared with generic matrices (e.g., Matrigel, collagen I coating of plastic), particularly for liver, lung, heart, and tumor systems where matrix composition strongly influences cell phenotype.^29,30

Defined Approach (DA)

A regulatory-ready, transparent combination of one or more information sources (typically NAMs such as in vitro assays and in silico predictions) with a fixed data-interpretation procedure (DIP) that yields a categorical or quantitative result for a specific regulatory endpoint. DAs are designed to be fully reproducible: given the same inputs, different users must arrive at the same conclusion.<sup>23

‍</sup>DAs are the practical output of an IATA for a defined use. OECD has adopted DAs for skin sensitization (TG 497), and others are under development for eye irritation, genotoxicity, and inhalation toxicity. DAs illustrate how NAMs can replace, rather than merely supplement, in vivo tests when combined in validated decision logics.²³

Disease Model

An experimental system — in silico, in vitro, ex vivo, animal or combined — designed to reproduce selected pathophysiological features of a disease to study mechanisms of disease, identify biomarkers, or evaluate candidate interventions. A disease model is defined by the set of disease features it recapitulates (genetic lesion, cellular phenotype, tissue architecture, clinical endpoint correlate), not by any one technology.<sup>5,42

‍</sup>NAM-based disease models include patient-derived organoids (tumor, cystic fibrosis airway, inflammatory bowel disease intestine, cerebral organoid …), iPSC-derived cellular models of  disease such as neurodegeneration and cardiomyopathy as well as genetically modified iPSC cells reflecting genetic risk factors of disease, MPS configured with disease-relevant cells and mechanical cues (fibrotic liver-on-chip, preeclamptic placenta-on-chip, brain-on-chip, blood-brain-barrier), and in silico models of disease development, such as tumor growth or immune dynamics. Well-qualified disease NAMs are a principal driver of ISTAND-type regulatory qualifications focused on efficacy.^7,8,17,42

Efficacy

‍The ability of a medicinal product to produce the intended therapeutic, pharmacological or disease-modifying effect under specified, typically well-controlled conditions. In drug and device development, efficacy is conventionally established through adequate and well-controlled clinical trials and is distinguished from effectiveness, which refers to the same ability under real-world conditions. <sup>38

‍</sup>NAMs contribute to efficacy evaluation at the nonclinical stage by providing species-relevant evidence of target engagement, pathway modulation, and disease relevant biologic effects using in vitro and computational models, such as patient-derived organoids or MPS of the target tissue. Efficacy-oriented NAMs are a growing focus of ISTAND qualification and of precompetitive consortia working in oncology, fibrosis, and rare disease.<sup>17

‍</sup>Relevant Links:
‍FDA — Providing Clinical Evidence of Effectiveness

Endpoint

A precisely defined variable intended to reflect an outcome of interest that is statistically analyzed to address a particular research question. A precise definition typically specifies the type of assessment performed, the timing of the assessment(s), the assessment tool(s) used, and other details, such as unit and numerical value that characterizes the result.<sup>1

‍</sup>In NAM-based studies, endpoints may be functional (barrier permeability, contractility, neural activity), biochemical (enzyme activity, cytokine secretion), morphological (high-content imaging features), or molecular (transcriptomic, proteomic, or metabolomic signatures). Clear endpoint definition — including acceptance criteria — is essential for both analytical and biological validation.<sup>1,19

‍</sup>Relevant Links:
‍BEST Resource — Endpoints chapter

Ex Vivo Normothermic Machine Perfusion (NMP)

A technology that maintains an intact, donated organ outside the body under near-physiological conditions by continuously perfusing it with oxygenated blood or blood-substitute perfusate at normal body temperature (typically 35–37°C). Normothermic machine perfusion preserves tissue viability, cellular metabolism, and organ-level physiological function, enabling real-time assessment of organ function, pharmacology, and response to experimental interventions.<sup>44

‍</sup>Unlike static tissue culture or MPS, ex vivo normothermic machine perfusion preserves the native three-dimensional architecture of an intact human organ, including its vascular network, extracellular matrix, resident immune cells, and multicellular interactions. Within the NAM landscape, normothermic machine perfusion provides a human-relevant platform for studying pharmacokinetics, pharmacodynamics, efficacy, toxicity, device performance, biomarker discovery, and AI-ready biological data generation while avoiding interspecies extrapolation.^44,45

Fit-for-Purpose

A conclusion that a tool, method, or dataset has sufficient performance characteristics to support a specific decision within a defined context of use. Fit-for-purpose is explicitly relative: a NAM can be fit for early-stage, exploratory hazard ranking yet unfit for quantitative regulatory risk assessment, and vice versa.<sup>16

‍</sup>The fit-for-purpose principle shifts NAM evaluation away from a binary “validated / not validated” framing toward an evidence-proportionate question: is the method good enough for this decision? It is the unifying principle — the standard against which fitness for a given decision is judged — behind FDA qualification, ICH and OECD validation schemes, and tiered NAM frameworks.^16,20

High-Content Imaging (HCI); also High-Content Screening (HCS) or Phenotypic Screening

Automated, microscope-based screening that measures biological activity in single cells or whole organisms by extracting many (often hundreds to thousands) of morphological and subcellular features per cell from fluorescently labeled or label-free images. HCS combines high-content imaging, performed using specialized hardware (spinning-disk confocal, widefield with deconvolution) with image-analysis pipelines to produce quantitative, multi-dimensional phenotypic profiles suitable for mechanism-of-action analysis and toxicity screening.<sup>33

‍</sup>The EU Reference Laboratory for alternatives to animal testing operates high-content imaging platforms for regulatory science applications.

‍Relevant Links:
‍NIH Assay Guidance Manual — High-Content Screening
‍EURL ECVAM (EU Reference Laboratory for Alternatives to Animal Testing)

High-Throughput Screening (HTS) or High-Throughput Testing (HTT)

Technology involving the process of screening and assaying large chemical or biological libraries, typically thousands to hundreds of thousands of conditions, in miniaturized, parallel assay formats (384-, 1536-, or 3456-well plates). HTS combines liquid-handling robotics, plate readers, and informatics to measure a defined endpoint across a broad compound space.<sup>25,37

‍</sup>HTS in vitro assays are a rapid and cost-effective source of NAM data for chemical prioritization, read-across support, and AOP development, primarily at the exploratory, pre-regulatory screening tier. ^25,37,26

Relevant Links:
‍EPA ToxCast
‍FDA — ToxCast/Tox21 High-Throughput Screening Program

IATA (Integrated Approach to Testing and Assessment)

‍A structured, flexible approach to hazard or risk assessment that integrates existing and new information sources — including NAMs, read-across, QSAR, and targeted in vivo data — through a weight-of-evidence evaluation. An IATA describes the problem formulation, the assessment elements used, and the decision logic; it may or may not include a formally fixed Defined Approach within it.<sup>23

‍</sup>IATAs are the dominant OECD construct for organizing the use of NAMs in regulatory decision-making, because they accommodate both fully validated DAs for well-understood endpoints and expert-judgment WoE for endpoints still under development.<sup>23

‍</sup>Relevant Links:
‍OECD IATA Programme

In Chemico

Refers to the abiotic measurement of chemical activity. In the NAMs framework in chemico methods are approaches that use chemical systems to measure direct chemical interactions or reactions associated with biological processes, without the involvement of biological material such as animals, cells, or tissues. Examples include the Direct Peptide Reactivity Assay (DPRA, OECD TG 442C) for skin sensitization and the FRAME assay for protein-binding hazard. In chemico methods sit alongside in silico and in vitro approaches as distinct non-animal method categories.<sup>3

‍</sup>Relevant Links:
‍OECD — Guidance on Grouping of Chemicals

IND Application (Investigational New Drug Application)

‍A submission by a sponsor to the U.S. FDA containing the nonclinical, chemistry-manufacturing-and-controls (CMC), and details of the proposed clinical study required to allow an unapproved drug to be administered to humans in a clinical investigation. A pre-IND submission is a briefing package sponsors use to obtain FDA feedback before filing the full IND application.<sup>39,40

‍</sup>Section 3209 of the FDA Modernization Act 2.0 and FDA’s subsequent implementation expressly allow sponsors to use NAMs — including cell-based assays, organ chips, MPS, bioprinted tissue, and computer models — to satisfy portions of the nonclinical evidence base for an IND, where scientifically justified for the specific context of use. Early engagement with the relevant FDA review division (pre-IND meeting) is strongly recommended when NAM data substitute for traditional animal studies.^18,9 The FDA Modernization Act 3.0, which would direct FDA to fully update its regulations to reflect this statutory shift away from mandatory animal testing, had passed both the House and Senate but had not yet been signed into law as of August 2026.<sup>43

‍</sup>To date, several drug developers, including Sanofi, Dianthus, and Cantex, have successfully used NAM data in IND and pre-IND applications to support drug development and safety and efficacy evaluation. 

‍Relevant Links:
‍FDA IND Application Overview
‍21 CFR Part 312 (eCFR)
‍FDA SBIA — Pre-IND Meeting FAQ
‍Sanofi/Dianthus NAM Case Study (ScienceDirect)
‍Hesperos — Rare Neuromuscular Disorder Efficacy Study

In Silico

Computational modeling and simulation used to predict biological activity, toxicity, or pharmacokinetics. Common in silico NAM categories include (Q)SAR models, physiologically based pharmacokinetic / toxicokinetic (PBPK/PBTK) models, read-across and category approaches, molecular docking, and machine-learning models trained on HTS or omics data.<sup>27,37

‍</sup>The OECD Principles for the Validation of (Q)SAR Models (defined endpoint, unambiguous algorithm, defined applicability domain, appropriate performance measures, and mechanistic interpretation) are the benchmark for in silico NAMs intended for regulatory use.²⁷

In Vitro

Experimental studies performed outside a living organism — literally, “in glass” — typically in controlled culture conditions using isolated cells, tissues, subcellular fractions, or biochemical molecules. In vitro NAMs span simple 2D monolayer assays, 3D cultures, spheroids, organoids, and MPS, together with biochemical and cell-free receptor and enzyme assays.

Together with in silico, in chemico, and ex vivo studies, in vitro work constitutes the non-animal evidence base within NAMs frameworks. In vitro methods are the most developed and most frequently cited category of NAMs in regulatory guidance and are the primary basis for the OECD GIVIMP good-practice standard.^19,3

iPSC (induced Pluripotent Stem Cell)

‍A somatic cell reprogrammed to a pluripotent, embryonic-like state through forced expression of a defined set of transcription factors — classically OCT4, SOX2, KLF4, and c-MYC (the “Yamanaka factors”).^14,15 Like embryonic stem cells, iPSCs self-renew indefinitely and can differentiate into cell types of all three germ layers.

iPSCs are a cornerstone of NAMs because they allow derivation of patient- or population-specific cell types (hepatocytes, cardiomyocytes, neurons, intestinal epithelium, etc.) that would otherwise be inaccessible. Quality considerations include reprogramming method, genomic stability, donor diversity, differentiation protocol reproducibility, and functional maturity of derived cells.^8,19

ISTAND (Innovative Science and Technology Approaches for New Drugs Pilot Program)

A U.S. FDA Center for Drug Evaluation and Research (CDER) program, launched in 2020, that provides a regulatory pathway for qualifying Drug Development Tools (DDTs) that do not fit within FDA’s existing biomarker, clinical outcome assessment, or animal model qualification programs. ISTAND explicitly welcomes novel approaches including digital health technologies, artificial intelligence / machine learning tools, and NAMs such as MPS and CIVMs.<sup>17

‍</sup>Submissions proceed through Letter of Intent (LOI), Qualification Plan (QP), and Full Qualification Package (FQP) stages. A successful ISTAND qualification yields an FDA determination that the tool can be relied upon for its specified context of use across multiple drug development programs, reducing the need for tool-specific justification in each IND/NDA.<sup>16,17

‍</sup>To date, no NAMs have been fully qualified as DDTs in the ISTAND Program, but several have received positive determination decisions on their LOI and QP submissions. 

‍Relevant Links:
‍FDA ISTAND Pilot Program

IVIVE (In Vitro to In Vivo Extrapolation)

The quantitative translation of in vitro concentration–response data (e.g., an AC₅₀ in a cell-based assay) to an equivalent in vivo exposure (e.g., an external dose or plasma concentration) that would be expected to produce the same effect in humans or animals. IVIVE is the bridge between NAM bioactivity data and real-world risk assessment or pharmacokinetic decisions.<sup>26

‍</sup>IVIVE typically combines in vitro assay data with a PBPK/PBTK model and toxicokinetic parameters (plasma protein binding, hepatic intrinsic clearance) derived from additional in vitro assays. EPA’s httk R package and related tools have made IVIVE routine for HTS-scale chemical evaluation.^26,37

Key Event (KE)

A measurable biological change within an AOP that is both necessary for progression of the pathway and essential for the adverse outcome. KEs span levels of organization — molecular, cellular, tissue, organ, individual — and each is associated with a Key Event Relationship (KER) describing its qualitative and, ideally, quantitative link to upstream and downstream events.<sup>22

‍</sup>KEs define what a NAM should measure to be mechanistically relevant. A NAM that reliably measures a KE close to the adverse outcome, within a well-characterized AOP, provides stronger predictive evidence than one measuring an unrelated or distant event.^22,23

MEA (Multi-Electrode Array; also Microelectrode Array)

A non-invasive electrophysiology platform in which multiple electrodes embedded in a culture substrate record extracellular field potentials from excitable cells (neurons, cardiomyocytes) cultured on the array. MEAs capture spontaneous activity, evoked responses, and network-level features such as burst rate, synchrony, and conduction velocity.<sup>34

‍</sup>MEA is a primary functional readout for iPSC-derived neuronal and cardiac NAMs, underpinning the Comprehensive In Vitro Proarrhythmia Assay (CiPA) and related cardiac safety frameworks, and is a leading technology for developmental neurotoxicity (DNT) NAMs.^34,9

Microfluidics

Systems used to precisely manipulate small volumes of fluids at sub-millimeter length scales, typically in channels with cross-sections of 10 µm–500 µm. At these scales, flow is typically laminar, diffusion dominates mixing, surface area‑to‑volume ratios are high, and fluid behavior can be tightly programmed — all features exploited by MPS, organ-on-chip, and lab-on-chip NAMs.<sup>24,7

‍</sup>Microfluidics enables physiologically relevant flow, shear stress, and concentration gradients in cell culture, along with precise delivery and sampling of test articles. Microfluidics is also used for droplet formation, including for cell incorporation in microparticles and microgel fabrication, including cell-free microgels used as porogens and for drug or protein delivery in 3D tissues. Soft lithography in PDMS remains the dominant prototyping approach; thermoplastic injection-molded devices are increasingly used for regulatory-grade applications.^12,24

Molecular Initiating Event (MIE)

‍The initial chemical-biological interaction within an AOP: the point at which a stressor perturbs a specific molecular target (receptor binding, enzyme inhibition, covalent adduction, DNA damage) in a way that can initiate the downstream cascade of Key Events leading to an adverse outcome.<sup>22

‍</sup>The MIE is the natural target for the earliest NAMs in a hazard-screening tier: binding assays, reporter-gene assays, and targeted HTS readouts often measure MIEs directly. However, perturbation of an MIE does not guarantee downstream adversity — interpretation depends on quantitative response–response relationships across subsequent KEs.^22,25

MPS (Microphysiological System)

An in vitro platform — commonly “organ-on-chip” or, for multi-tissue formats, “body-on-chip” — that integrates living cells with engineered microenvironments to recapitulate key physiological functions of a tissue, organ, or organ system. MPS typically combine 3D tissue architecture, relevant cell types (often including iPSC-derived cells), controlled fluid flow, and mechanical or biochemical cues such as shear stress, stretch, or oxygen gradients.<sup>7,8,12

‍</sup>MPS are developed and evaluated for ADME, efficacy screening, disease modeling, and organ-specific toxicity (hepatic, cardiac, renal, intestinal, pulmonary, neural). Readouts commonly include TEER, barrier permeability, contractility, secretome analysis, and high-content imaging.<sup>7,10

‍</sup>Relevant Links:
‍FDA — Advancing Alternative Methods
‍NIH Tissue Chip for Drug Screening Program

NAM (New Approach Methodology)

A broad term encompassing any technology, methodology, approach, or combination thereof that can provide mechanistically relevant, species-specific information to support chemical hazard characterization, risk assessment, pharmacology profiling, and drug development while avoiding or reducing reliance on traditional animal testing. NAMs include in vitro assays (2D and 3D), organoids, spheroids, microphysiological systems, in silico (computational/AI) models, and in chemico approaches, together with select whole-organism models, e.g., life stages or organisms not covered as protected species under current frameworks (such as zebrafish embryos prior to independent feeding, C. elegans, and Drosophila). Regulatory status is use-specific: a given platform — a CIVM, an MPS, an in silico model, or any other model type above — functions as a NAM when it is applied to answer a defined regulatory question by replacing or supplementing an existing methodology (often an animal test) for that stated Context of Use. The term originated in regulatory toxicology and has been adopted by EPA, ICCVAM, ECHA, OECD, and FDA. 

In the United States, EPA’s NAM Work Plan and ICCVAM’s Strategic Roadmap coordinate cross-agency NAM development and acceptance. NAMs may be used at different levels of maturity and acceptance, from exploratory to Fit-for-purpose and Qualified. The FDA Modernization Act 2.0 (2022) updated the Federal Food, Drug and Cosmetic Act by substituting “preclinical tests (including tests on animals)” with the broader term “nonclinical tests” — which formally encompasses NAMs and includes cell-based assays, organ chips, MPS, and computer models. In September 2026, FDA followed with a direct final rule formally defining “nonclinical test” and “nonclinical study” across its regulations (21 CFR Parts 312, 314, 315, 361, and 601) to include in vitro, in silico, and in chemico tests, organ chips and MPS, computer modeling, and other nonhuman or human biology-based methods, alongside animal studies.^2,3,18,47

Relevant Links:
‍EPA NAM Work Plan
‍ICCVAM Strategic Roadmap
‍FDA Modernization Act 2.0
‍FDA — Nonclinical Testing Terminology (Direct Final Rule, 2026)

OECD Test Guideline (TG)

‍Internationally recognized standardized testing methods used to assess the potential effects of chemicals on the environment and human health that have been formally adopted by the OECD Working Group of the National Coordinators of the Test Guidelines Programme. Adopted TGs are subject to the OECD Mutual Acceptance of Data (MAD) system: data generated in any OECD member country using an adopted TG must be accepted for regulatory assessment by other member countries, avoiding duplicate testing.<sup>19

‍</sup>A growing number of OECD TGs formalise the use of in vitro or computational NAMs — notably for skin corrosion/irritation (TG 431, 439), skin sensitization (TG 442C/D/E, 497), eye irritation (TG 492, 492B), genotoxicity (TG 487), and endocrine activity (TG 455, 458). Adoption as a TG is widely regarded as the highest validation status a NAM can achieve.<sup>19,23

‍</sup>Relevant Links:
‍OECD Test Guidelines

Omics

A collective term for technologies that measure large, unbiased sets of biological molecules of a given class: transcriptomics (RNA), proteomics (proteins), metabolomics (small molecules), lipidomics (lipids), epigenomics (chromatin/DNA modifications), and others. Omics readouts produce high-dimensional signatures that can classify compounds by mechanism, identify responsive pathways, and link NAM exposures to AOP key events.

Targeted subsets — notably Bioseq / S1500+ for toxicogenomics and shotgun metabolomics panels — make omics practical at screening scale. Reproducibility, batch-effect control, and transparent analysis pipelines are central to fit-for-purpose use of omics in regulatory NAMs.^19,37

Organ Function

The integrated physiological performance of an intact organ, measured through quantitative functional outputs that reflect its ability to carry out its native biological role.<sup>7

‍</sup>Examples include glomerular filtration rate, urine production, renal vascular resistance, bile production, oxygen consumption, pulmonary gas exchange, cardiac contractility, vascular perfusion, electrophysiology, and metabolic activity. Functional endpoints complement molecular and cellular biomarkers by providing direct measures of tissue performance and are particularly valuable during validation of ex vivo whole-organ NAMs.^7,8,46

Organoid

A three-dimensional, self-organizing cell culture derived from pluripotent stem cells (embryonic or induced), adult tissue stem cells, or progenitor cells, which recapitulates key structural and functional features of a native organ or tissue. Organoids form through the intrinsic capacity of stem cells to differentiate and spatially organize, typically within an extracellular matrix such as Matrigel, and can be propagated long-term (multiple weeks or months).^4,5 Additionally mature organoids can be disassociated and seeding to other culture formats (e.g. MPS) to be a source of cells for more complex culture systems.

Established organoid systems exist for intestine, liver, kidney, lung, brain, pancreas, stomach, and tumor tissues, among others. Distinguishing features versus spheroids include self-organization, presence of multiple differentiated cell types, and reproducible tissue-like architecture.^4,5,6

PBPK / PBTK (Physiologically Based Pharmaco- / Toxico-Kinetic Model)

A mechanistic, compartmental mathematical model in which each compartment represents a real anatomical tissue or organ, parameterized with species-specific physiology (blood flows, tissue volumes, partition coefficients) and compound-specific properties (plasma protein binding, intrinsic clearance, permeability). PBPK/PBTK models predict the time course of a compound’s concentration in plasma and tissues following any dose and route of administration.<sup>26

‍</sup>PBPK/PBTK models are the computational counterpart to MPS and the engine of IVIVE: they convert in vitro NAM concentration–response data into predicted whole body exposures and doses. They are routinely used in regulatory submissions to support first-in-human dose selection, special-population adjustments, and chemical risk assessment.^26,37

PDMS (Polydimethylsiloxane)

A biocompatible silicone elastomer that is the most widely used structural material for microfluidic and organ-on-chip devices. PDMS is transparent (enabling optical microscopy), gas-permeable (allowing oxygenation of cultured cells), flexible (permitting mechanical actuation such as stretch), inexpensive, and readily patterned via soft-lithography replica molding.<sup>12,24

‍</sup>A well-known limitation is that PDMS absorbs and adsorbs small hydrophobic molecules, which can confound pharmacokinetic and pharmacological measurements in MPS. Surface treatments, alternative thermoplastics (e.g., cyclic olefin copolymer), and mass-balance accounting are common mitigations.¹³

Permeable Membrane Cell-Culture Insert (also Transwell)

A porous polymeric membrane (polycarbonate, polyester, PTFE) suspended within a well of a multi-well plate, separating an apical (upper) and basolateral (lower) fluid compartment and allowing barrier tissue models to be cultured with distinct apical and basolateral environments. Transwell (Corning) is the best-known commercial brand and, like Matrigel, is often used generically for the product category; other manufacturers sell functionally equivalent inserts.

Transwells are the standard format for TEER, transport, and permeability assays; they enable air-liquid interface culture, apical-to-basolateral transcellular drug transport measurements, and co-culture across the membrane (e.g., epithelium on top, endothelium or immune cells underneath). Many MPS platforms derive conceptually from Transwell geometry with added perfusion.^10,7

Personalized Medicine (also Precision Medicine)

A medical approach that tailors prevention, diagnosis, and treatment decisions to individual characteristics — including genetic, molecular, cellular, environmental, and lifestyle factors — typically through biomarker-guided patient stratification. BEST notes that predictive biomarkers are a central tool of personalized medicine because they identify patients most likely to benefit from, or be harmed by, a particular intervention.<sup>1,41

‍</sup>NAMs are a natural substrate for personalized medicine because many — iPSC-derived cardiomyocytes or neurons, patient-derived tumor organoids, kidney and intestinal organoids from biopsies — can be established from a specific individual. Such patient-specific NAMs enable ex ante prediction of drug response, selection of effective therapy in rare or treatment-resistant disease, and mechanistic investigation of interindividual variability. Biobanked organoid and iPSC collections with linked clinical data extend this approach to population-scale precision medicine.^41,42,5

Relevant Links:
‍NIH All of Us Research Program

Primary Cell

A cell isolated directly from a tissue donor (human or animal) and cultured without prior immortalization or extensive passaging. Primary cells retain in vivo phenotype and donor-specific biology more faithfully than most cell lines, but have finite replicative lifespan, donor-to-donor variability, and limited availability.

Primary cells are often the gold standard for NAM development where phenotypic fidelity is critical (e.g., primary human hepatocytes for metabolism studies, primary airway epithelial cells for ALI cultures). They are increasingly complemented — or replaced — by iPSC-derived cells, which offer comparable fidelity with effectively unlimited supply.^19,8

QSAR (Quantitative Structure-Activity Relationship)

A computational model that predicts a biological activity, physicochemical property, or toxicological endpoint for a chemical based on its molecular structure, using training data from measured activities of related compounds. “(Q)SAR” is sometimes used to encompass both quantitative and qualitative (categorical) models.<sup>27

‍</sup>The OECD Principles for the Validation of (Q)SAR Models — defined endpoint, unambiguous algorithm, defined applicability domain, appropriate performance measures, and, where possible, mechanistic interpretation — are the reference standard for regulatory acceptance of (Q)SAR NAMs.<sup>27

‍</sup>Relevant Links:
‍OECD QSAR Toolbox

Qualification/Regulatory Acceptance

A regulatory determination — by FDA and equivalent authorities — that within a specifically stated context of use, a drug development tool (DDT) or medical device development tool (MDDT) can be relied upon to have a specific interpretation and application in drug or device development and regulatory review. Once qualified, the (M)DDT can be used in any IND, NDA, or BLA for the qualified COU without requiring CDER to reconfirm its suitability.<sup>16

‍</sup>For NAMs, qualification requires evidence of both analytical and biological validation for the stated COU, and typically proceeds through FDA’s DDT qualification programs — Biomarker Qualification, Clinical Outcome Assessment Qualification, Animal Model Qualification, or ISTAND for tools that fit none of the first three. ISTAND exists precisely because these advanced platforms do not fit the legacy Biomarker, Clinical Outcome Assessment, or Animal Model tracks, which were designed around older classes of tools.<sup>16,17

‍</sup>Relevant Links:
‍FDA DDT Qualification Programs
‍FDA MDDT Qualification Program

Read-Across

A data-gap-filling technique in which a toxicological property of an untested target chemical is predicted from measured data on one or more structurally or mechanistically similar source chemicals. Read-across may be carried out one-to-one (analogue approach) or from a chemical category, and typically incorporates in silico, in chemico, and in vitro NAM data as similarity and reactivity justifications.<sup>28

‍</sup>A defensible read-across argument explicitly justifies the similarity hypothesis (structural, metabolic, mechanistic), demonstrates agreement of NAM and (where available) in vivo responses across the category, and states a clear applicability domain. ECHA and EPA have issued read-across assessment frameworks; read-across underpins many REACH registrations.<sup>28,20

‍</sup>Relevant Links:
‍ECHA — Read-Across Assessment Framework (RAAF)

Safety

Freedom from unacceptable risk of harm when an intervention is used as intended. Safety is characterized through a combination of nonclinical studies (general toxicology, safety pharmacology, genotoxicity, reproductive and developmental toxicity, carcinogenicity) and clinical studies that describe adverse events, dose limits, and risks to specific populations. In the BEST framework, a safety biomarker is one that indicates the likelihood, presence, or extent of toxicity as an adverse effect.<sup>1,38

‍</sup>NAMs now contribute to safety evaluation across nearly every traditional nonclinical discipline: organ-on-chip hepatotoxicity, MEA-based cardiac proarrhythmia within the Comprehensive In Vitro Proarrhythmia Assay (CiPA), iPSC-derived neurons for developmental neurotoxicity, high-content imaging for general toxicity, AOP-anchored in vitro assays for endocrine disruption, and computational IVIVE for exposure–response. The FDA Modernization Act 2.0 formally enables NAM data to support safety conclusions in INDs and subsequent marketing applications.^9,18,22

Risk Assessment; also NGRA

A systematic process of organizing information to support a risk decision to be made within a risk management process. It consists of the identification of hazards and the analysis and evaluation of risks associated with exposure to those hazards.

While traditional risk assessment evaluates hazards using animal models, Next Generation Risk Assessment (NGRA) modernizes risk assessment and drug development through an exposure-led, hypothesis-driven approach. It integrates human-relevant New Approach Methodologies (NAMs) to predict safety and inform risk decisions directly aligned with human health outcomes, without relying on animal extrapolation.

‍Relevant Links:
‍FDA — Q9(R1) Quality Risk Management
‍Frontiers in Toxicology — Next Generation Risk Assessment

Shear Stress

The tangential mechanical force per unit area (typically dyn/cm² or Pa) exerted when two surfaces move, or tend to move, relative to one another. Shear stress includes both fluid shear stress — exerted by a flowing liquid or gas on a surface — and frictional shear stress between solids, such as between tissue and an implant surface; in biology and in most NAMs, the term most often refers to fluid shear stress acting on the apical surface of endothelial or epithelial cells. Shear stress is a key physiological cue that regulates gene expression, cytoskeletal organization, barrier function, differentiation, and polarity.<sup>11

‍</sup>Reproducing physiological shear is a principal reason for introducing perfusion in MPS: venous endothelium experiences roughly 1–4 dyn/cm², arterial endothelium 10–70 dyn/cm², and renal proximal tubule epithelium ~0.2–1 dyn/cm². Omitting flow in a vascular or renal model risks a dedifferentiated phenotype with reduced predictive value.^7,11

Spheroid

A three-dimensional, roughly spherical aggregate formed when cells — primary cells, cell lines, or stem-cell-derived cells — are allowed to self-assemble in low-attachment conditions, hanging drops, or non-adherent scaffolds. Unlike organoids, spheroids do not require an intrinsic stem-cell program and generally do not develop the multi-lineage, spatially patterned architecture of a native organ.<sup>6

‍</sup>Spheroids recreate cell–cell contact, diffusion gradients of oxygen, nutrients, and drugs, and a necrotic core in larger aggregates. They are widely used in oncology (tumor spheroids), hepatotoxicity screening (hepatocyte spheroids), and as building blocks for more complex CIVMs and MPS.^6,8

Standardization/Harmonization

The use of harmonized methods, protocols, controls, data formats, and interpretation criteria to ensure that data generated by a NAM is reliable, reproducible, and sufficiently consistent for regulatory review and decision-making.<sup>19,23

‍</sup>OECD Test Guidelines, GIVIMP good practice, and IATA reporting formats are the primary harmonization instruments for NAMs: they let data generated in one laboratory or jurisdiction be interpreted, compared, and mutually accepted by regulators and collaborators elsewhere, reducing duplicate testing and easing cross-border regulatory acceptance.^19,23

TEER (Transepithelial / Transendothelial Electrical Resistance)

A non-invasive, quantitative measurement of the electrical resistance across a cellular monolayer, expressed in ohms·cm². TEER is the most widely used functional readout of barrier integrity and tight-junction formation in epithelial and endothelial cultures, including gut, blood-brain barrier, lung, and kidney models.<sup>10

‍</sup>TEER is measured with chopstick electrodes, cup-style systems, or integrated electrodes embedded in MPS chips. Interpretation requires attention to temperature, electrode geometry, cell-free blank subtraction, and frequency of the applied signal; impedance spectroscopy extends TEER by resolving resistive and capacitive contributions.¹⁰

Tox21 / ToxCast

Complementary U.S. federal high-throughput toxicology programs. Tox21 is a collaboration among NIH/NCATS, EPA, FDA, and NIEHS that has screened roughly 10,000 chemicals across ~70 cell-free and cell-based assays targeting nuclear receptors, stress response pathways, and other molecular initiating events. ToxCast is EPA’s program that expands on this with hundreds of additional assays, including higher-content and targeted endpoints.<sup>25,37

‍</sup>The resulting data (publicly available through EPA’s CompTox Chemicals Dashboard and ICE at NIEHS) are a principal evidence base for NAM-driven hazard prioritization, read-across support, and AOP development across the U.S. regulatory landscape.<sup>25,37

‍</sup>Relevant Links:
‍EPA CompTox Chemicals Dashboard
‍NICEATM ICE (Integrated Chemical Environment)

Weight of Evidence (WoE)

A structured, transparent process for integrating multiple lines of evidence — each with distinct relevance, reliability, and uncertainty — to reach a defensible scientific conclusion where no single study is decisive. A WoE evaluation explicitly identifies the information sources used, weights their quality and relevance, and reasons across agreement and disagreement.<sup>35

‍</sup>WoE is the dominant integration framework within IATAs, read-across justifications, and NAM-based regulatory submissions, because NAMs typically generate multiple complementary but imperfect data streams (in silico, in vitro, in chemico, legacy in vivo). EFSA and ECHA have published formal WoE guidance; good WoE is transparent enough that an independent reviewer can trace each conclusion back to its supporting evidence.^35,23

References

1. FDA-NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools) Resource. Silver Spring (MD) and Bethesda (MD): FDA and NIH; 2016. Updated continuously.  https://www.ncbi.nlm.nih.gov/books/NBK326791/

‍2. U.S. Environmental Protection Agency. New Approach Methods Work Plan: Reducing Use of Animals in Chemical Testing. EPA 600/X-21/209; 2021.  https://www.epa.gov/research/new-approach-methods-work-plan

‍3. ICCVAM. A Strategic Roadmap for Establishing New Approaches to Evaluate the Safety of Chemicals and Medical Products in the United States. NIEHS/NICEATM; 2018.  https://ntp.niehs.nih.gov/whatwestudy/niceatm/natl-strategy

‍4. Lancaster MA, Knoblich JA. Organogenesis in a dish: Modeling development and disease using organoid technologies. Science. 2014;345(6194):1247125.  https://doi.org/10.1126/science.1247125

‍5. Clevers H. Modeling Development and Disease with Organoids. Cell. 2016;165(7):1586-1597.  https://doi.org/10.1016/j.cell.2016.05.082

‍6. Fennema E, Rivron N, Rouwkema J, van Blitterswijk C, de Boer J. Spheroid culture as a tool for creating 3D complex tissues. Trends Biotechnol. 2013;31(2):108-115.  https://doi.org/10.1016/j.tibtech.2012.12.003

‍7. Low LA, Mummery C, Berridge BR, Austin CP, Tagle DA. Organs-on-chips: into the next decade. Nat Rev Drug Discov. 2021;20(5):345-361.  https://doi.org/10.1038/s41573-020-0079-3

‍8. Marx U, et al. Biology-inspired microphysiological systems to advance patient benefit and animal welfare in drug development. ALTEX. 2020;37(3):365-394.  https://doi.org/10.14573/altex.2001241

‍9. U.S. Food and Drug Administration. Advancing Alternative Methods at FDA. FDA/ORA; updated 2024.  https://www.fda.gov/science-research/about-science-research-fda/advancing-alternative-methods-fda

‍10. Srinivasan B, Kolli AR, Esch MB, Abaci HE, Shuler ML, Hickman JJ. TEER measurement techniques for in vitro barrier model systems. J Lab Autom. 2015;20(2):107-126.  https://doi.org/10.1177/2211068214561025

‍11. Davies PF. Flow-mediated endothelial mechanotransduction. Physiol Rev. 1995;75(3):519-560.  https://doi.org/10.1152/physrev.1995.75.3.519

‍12. Huh D, Matthews BD, Mammoto A, Montoya-Zavala M, Hsin HY, Ingber DE. Reconstituting organ-level lung functions on a chip. Science. 2010;328(5986):1662-1668.  https://doi.org/10.1126/science.1188302

‍13. Toepke MW, Beebe DJ. PDMS absorption of small molecules and consequences in microfluidic applications. Lab Chip. 2006;6(12):1484-1486.  https://doi.org/10.1039/b612140c

‍14. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126(4):663-676.  https://doi.org/10.1016/j.cell.2006.07.024

‍15. Takahashi K, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131(5):861-872.  https://doi.org/10.1016/j.cell.2007.11.019

‍16. U.S. Food and Drug Administration. Qualification Process for Drug Development Tools: Guidance for Industry and FDA Staff. CDER/CBER; 2020.  https://www.fda.gov/regulatory-information/search-fda-guidance-documents/qualification-process-drug-development-tools-guidance-industry-and-fda-staff

‍17. U.S. Food and Drug Administration. Innovative Science and Technology Approaches for New Drugs (ISTAND) Pilot Program. CDER; 2020, updated 2023.  https://www.fda.gov/drugs/drug-development-tool-ddt-qualification-programs/innovative-science-and-technology-approaches-new-drugs-istand-pilot-program

‍18. U.S. Food and Drug Administration. FDA Modernization Act 2.0 (Public Law 117-328, Section 3209); 2022.  https://www.congress.gov/bill/117th-congress/senate-bill/5002

‍19. OECD. Guidance Document on Good In Vitro Method Practices (GIVIMP). Series on Testing and Assessment No. 286. Paris: OECD Publishing; 2018.  https://doi.org/10.1787/9789264304796-en

‍20. Parish ST, et al. An evaluation framework for new approach methodologies (NAMs) for human health safety assessment. Regul Toxicol Pharmacol. 2020;112:104592.  https://doi.org/10.1016/j.yrtph.2020.104592

‍21. Russell WMS, Burch RL. The Principles of Humane Experimental Technique. London: Methuen; 1959 (reissued Universities Federation for Animal Welfare, 1992).  https://caat.jhsph.edu/principles/the-principles-of-humane-experimental-technique

‍22. Ankley GT, Bennett RS, Erickson RJ, et al. Adverse outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment. Environ Toxicol Chem. 2010;29(3):730-741.  https://doi.org/10.1002/etc.34

‍23. OECD. Guidance Document on the Reporting of Defined Approaches and Individual Information Sources to be Used within Integrated Approaches to Testing and Assessment (IATA). Series on Testing and Assessment No. 255/256. Paris: OECD Publishing; 2017.  https://www.oecd.org/chemicalsafety/testing/adverse-outcome-pathways-molecular-screening-and-toxicogenomics.htm

‍24. Whitesides GM. The origins and the future of microfluidics. Nature. 2006;442(7101):368-373.  https://doi.org/10.1038/nature05058

‍25. Tice RR, Austin CP, Kavlock RJ, Bucher JR. Improving the human hazard characterization of chemicals: a Tox21 update. Environ Health Perspect. 2013;121(7):756-765.  https://doi.org/10.1289/ehp.1205784

‍26. Wetmore BA. Quantitative in vitro-to-in vivo extrapolation in a high-throughput environment. Toxicology. 2015;332:94-101.  https://doi.org/10.1016/j.tox.2014.05.012

‍27. OECD. Guidance Document on the Validation of (Quantitative) Structure-Activity Relationship [(Q)SAR] Models. Series on Testing and Assessment No. 69. Paris: OECD Publishing; 2007.  https://www.oecd.org/env/ehs/risk-assessment/validationofqsarmodels.htm

‍28. Patlewicz G, Ball N, Becker RA, et al. Read-across approaches — misconceptions, promises and challenges ahead. ALTEX. 2014;31(4):387-396.  https://doi.org/10.14573/altex.1410071

‍29. Murphy SV, Atala A. 3D bioprinting of tissues and organs. Nat Biotechnol. 2014;32(8):773-785.  https://doi.org/10.1038/nbt.2958

‍30. Crapo PM, Gilbert TW, Badylak SF. An overview of tissue and whole organ decellularization processes. Biomaterials. 2011;32(12):3233-3243.  https://doi.org/10.1016/j.biomaterials.2011.01.057

‍31. Pașca SP. Assembling human brain organoids. Science. 2019;363(6423):126-127.  https://doi.org/10.1126/science.aau5729

‍32. Hughes CS, Postovit LM, Lajoie GA. Matrigel: a complex protein mixture required for optimal growth of cell culture. Proteomics. 2010;10(9):1886-1890.  https://doi.org/10.1002/pmic.200900758

‍33. Bray MA, Singh S, Han H, et al. Cell Painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes. Nat Protoc. 2016;11(9):1757-1774.  https://doi.org/10.1038/nprot.2016.105

‍34. Johnstone AF, Gross GW, Weiss DG, Schroeder OH, Gramowski A, Shafer TJ. Microelectrode arrays: a physiologically based neurotoxicity testing platform for the 21st century. Neurotoxicology. 2010;31(4):331-350.  https://doi.org/10.1016/j.neuro.2010.04.001

‍35. EFSA Scientific Committee, Hardy A, et al. Guidance on the use of the weight of evidence approach in scientific assessments. EFSA Journal. 2017;15(8):e04971.  https://doi.org/10.2903/j.efsa.2017.4971

‍36. Ronaldson-Bouchard K, Vunjak-Novakovic G. Organs-on-a-chip: A fast track for engineered human tissues in drug development. Cell Stem Cell. 2018;22(3):310-324.  https://doi.org/10.1016/j.stem.2018.02.011

‍37. Thomas RS, Bahadori T, Buckley TJ, et al. The next generation blueprint of computational toxicology at the U.S. EPA. Toxicol Sci. 2019;169(2):317-332.  https://doi.org/10.1093/toxsci/kfz058

‍38. International Council for Harmonisation (ICH). General Considerations for Clinical Studies: E8(R1). ICH Harmonised Guideline; 2021.  https://www.ich.org/page/efficacy-guidelines

‍39. U.S. Food and Drug Administration. 21 CFR Part 312: Investigational New Drug Application. Code of Federal Regulations; updated 2024.  https://www.ecfr.gov/current/title-21/chapter-I/subchapter-D/part-312

‍40. U.S. Food and Drug Administration. IND Applications for Clinical Investigations — Content and Format (Guidance for Industry). CDER/CBER; 2013, updated 2022.  https://www.fda.gov/drugs/investigational-new-drug-ind-application

‍41. Collins FS, Varmus H. A new initiative on precision medicine. N Engl J Med. 2015;372(9):793-795.  https://doi.org/10.1056/NEJMp1500523

‍42. Drost J, Clevers H. Organoids in cancer research. Nat Rev Cancer. 2018;18(7):407-418.  https://doi.org/10.1038/s41568-018-0007-6

‍43. U.S. Congress. FDA Modernization Act 3.0 (H.R. 2821, 119th Congress); passed by the House and Senate 2026, pending enactment.  https://www.congress.gov/bill/119th-congress/house-bill/2821

‍44. Clarke G, Mao J, Hann A, et al. A reproducible extended ex-vivo normothermic machine liver perfusion protocol utilising improved nutrition and targeted vascular flows. Commun Med. 2024;4:216.  https://doi.org/10.1038/s43856-024-00636-2

‍45. Czogalla J, et al. A high-throughput drug discovery pipeline to optimize kidney normothermic machine perfusion. Front Physiol. 2022;13:974615.  https://doi.org/10.3389/fphys.2022.974615

‍46. van der Hoek JL, Krommendijk ME, Manohar S, Arens J, Groot Jebbink E. Ex-vivo human-sized organ machine perfusion: a systematic review on the added value of medical imaging for organ condition assessment. Transpl Int. 2024;37:12827.  https://doi.org/10.3389/ti.2024.12827

‍47. U.S. Food and Drug Administration. Nonclinical Testing Terminology; direct final rule. 91 Fed. Reg. 59988 (Sept. 22, 2026) (to be codified at 21 CFR pts. 312, 314, 315, 361, 601).  https://www.federalregister.gov/documents/2026/09/22/2026-19350/nonclinical-testing-terminology