Standardizing NAMs: Why the Field Needs a Unified Voice Before Regulation Can Catch Up

As non‑animal models (NAMs) move rapidly from specialist innovation into mainstream drug development, the field is facing a growing mismatch between scientific capability and regulatory readiness. The Industry Alliance for Microphysiological Systems (IAMPS), the world’s first dedicated Microphysiological Systems (MPS) industry association, was established to address this gap. IAMPS aims to unify NAM developers to standardize terminology, clarify context of use, and create shared data infrastructures that can support future regulatory qualification and acceptance.

Few voices can speak to this challenge with the clarity and authority of Dr. Madhu Lal‑Nag, the chief scientific officer at InSphero and former director of the Trans NIH RNAi Facility at the National Center for Advancing Translational Sciences (NCATS/NIH) and Program Director at USFDA. Drawing on her dual vantage point, Lal‑Nag explains why NAMs have outpaced the shared language and evidence regulators need, why misconceptions are proliferating, and how new coalitions can lay the foundation for credible guidance. Her insights provide a grounded, practical view on what must happen before policy can truly catch up.

Clearing the fog: Why NAM definitions must be standardized

Across industry and academia, enthusiasm for NAMs has surged—but so has confusion about what each model is designed to do. Lal‑Nag stresses the need to define spheroids, organoids, and microphysiological systems. She explains the recent hype cycle has amplified misconceptions, encouraging businesses and researchers to adopt NAMs without understanding their proper domain.

Figure 1: Non-animal models (NAMs), including spheroids, organoids, and microphysiological systems, require standardization. Credit: AI-generated image created using Microsoft Copilot (2026).

IAMPS directly addresses this fragmentation by aligning MPS developers around harmonized terminology, qualification processes, and data‑sharing standards with regulators, including the European Medicines Agency, the Food and Drug Administration (FDA), the Medicines and Healthcare products Regulatory Agency, and the Product Development and Management Association. The organization aims to help end‑users and agencies better interpret what each technology can—and cannot—deliver.  

“Let’s come together to create this NAMs ecosystem…so that we can actually deliver to the community what you should use, where, to get what result.”— Dr. Madhu Lal Nag

Why clear definitions matter:

• Ambiguous language leads to misuse and unrealistic expectations of NAMs.
• Coalitions like IAMPS unite developers to align terminology and context of use.
• Standardized definitions are the foundation for credible regulatory evaluation.

Regulators want to move faster—but lack the data to do so

Regulators appear receptive to NAMs but lack sufficient shared datasets and defined performance boundaries to issue formal guidance. Lal‑Nag notes that agencies “don’t see this data,” meaning they cannot evaluate how NAMs should be used in submissions. Without established boundaries, performance metrics, or negative results, they cannot confidently issue broad guidance.

Meanwhile, the FDA Modernization Act 3.0, passed in late 2025, directs FDA to update regulations by replacing references to “animal” testing with “nonclinical” terminology—explicitly acknowledging NAMs, MPS, and AI‑driven in silico models, and requires the agency to build a formal, accelerated process for evaluating non‑animal testing methods. Whilst this legislation intends to modernize nonclinical terminology, formal guidance on implementation remains limited, contributing to uncertainty among sponsors.  A recent survey found that 77% of researchers are not using in vitro alternatives to animal testing, often citing uncertainty about regulatory acceptance as a major barrier.

The Critical Path Institute (C‑Path) has developed frameworks to support FDA qualification of complex in vitro models (CIVMs), demonstrating how shared datasets, structured contexts of use, and collaborative consortia can build reviewer confidence.

Why regulators need better evidence:

• Agencies cannot issue reliable guidance without systematic performance datasets.
• FDA Modernization Act 3.0 accelerates regulatory flexibility but depends on high‑quality evidence.
• C‑Path frameworks show how shared data enables structured qualification.

The essential role of negative data in building credibility

One of Lal‑Nag’s most urgent points is the need for negative data. Historically, sponsors avoided submitting failures, slowing regulators’ ability to understand model limits. Without clear evidence of where models break down, guidance cannot reflect real‑world performance.

IAMPS’ emphasis on retrospective case studies and shared datasets directly supports this. By pooling both successes and failures, MPS developers and end‑users can help regulators identify context-appropriate applications and avoid premature policy shifts. C‑Path uses similar safe‑harbor approaches within its collaborative consortia.

“If you only submit positive results, you don’t know what it has failed in.”— Dr. Madhu Lal Nag

Why negative data shapes regulatory trust:

• Reveals performance boundaries regulators must understand to write guidance.
• Safe‑harbor data‑sharing platforms accelerate tool qualification.
• Prevents unrealistic expectations about NAMs’ generalizability.

Diving Into Multiomics: From Data to Discovery

This infographic explores the key multiomics fields, their integration across research applications, and real-world case studies spanning cancer, diabetes, agriculture, and environmental science.
View Infographic

Advertisement

Where NAMs are ready today—and where they still need roadmaps

Lal‑Nag highlights liver toxicology as the area closest to regulatory acceptance. Thanks to long‑term investment in biomarkers, donor consistency, and reproducible microtissue systems, she believes liver toxicology NAMs are approaching regulatory qualification for small molecules, with potential acceptance of non-animal data in IND submissions within the next one to two years.

She also emphasizes oncology as an area where NAMs offer a unique advantage: patient-derived organoid and spheroid systems can model clinical heterogeneity in ways homogeneous animal populations cannot. This aligns with broader regulatory shifts toward identifying subpopulation responses rather than relying solely on aggregate outcomes.

However, large molecules and non‑human primate (NHP) alternatives require structured, staged roadmaps, as the scientific foundation is less mature.

Where NAMs stand on the adoption curve:

• Liver toxicology NAMs are closest to full regulatory acceptance.
• Large molecule and NHP replacement pathways require multi‑year roadmaps.

Reproducibility: The non‑negotiable requirement for NAM validation

Reproducibility determines whether NAMs can scale from innovation to regulatory‑grade tools. Lal‑Nag describes InSphero’s stringent quality control (QC) for its 10‑donor pooled liver microtissues, including biological markers, 28‑day metabolic stability, enzyme activity, albumin secretion, and plate‑level performance. Oncology models likewise rely on primary patient tissues and immune microenvironment profiling. These QC strategies create baseline stability, enabling reliable interpretation of responses.

She also stresses that endpoint assays alone miss important biological nuance. High‑content imaging, spatial transcriptomics, and kinetic readouts capture early predictive signals that improve both translational relevance and model qualification. However, these data‑rich assays must be designed for automation and AI from the outset, as retrofitting fragile biology into high-throughput formats leads to data that is not robust or reproducible.

“We need to co-develop the machine learning and the biology hand in hand.”— Dr. Madhu Lal Nag

Why reproducibility underpins NAM adoption:

• Consistent donor‑ and lot‑level QC enables interpretable data.
• Multimodal, time‑series readouts reveal signals endpoint assays miss.
• NAMs must be designed for automation and AI—not adapted later.

Predict Drug-Induced Liver Injury With Greater Confidence

This poster introduces programmed hiPSC-derived hepatocytes that combine the scalability and batch consistency researchers need to address liver functionality and clinically relevant toxicity responses.
View Poster

Advertisement

What a realistic roadmap requires

For NAMs to become widely accepted, Lal‑Nag argues for specificity: indication‑based roadmaps, detailed action items, and transparent datasets. Mature areas such as liver toxicity already have the evidence base; others require clearer milestones and a deliberately phased approach. Coalitions like IAMPS are essential to coordinating standards, definitions, and evidence collection across companies and regulators.

Policy momentum—from FDA Modernization Act 3.0 to EU standardization initiatives—provides tailwinds, but evidence unity remains the true rate‑limiting factor. NAMs are scientifically powerful, but only a standardized, collaborative approach can transform them into regulatory norms

“They need to set up not just a two‑page idea, but an actual roadmap with action items for what you need to test… and the correlation that you have to establish with a human being.”— Dr. Madhu Lal Nag

What a credible roadmap demands:

• Roadmaps must be indication‑specific, with measurable milestones.
• Coalitions unify definitions, standards, and data infrastructure.
• Policy tailwinds enable progress, but unified evidence determines pace.

This content includes text that has been created with the assistance of generative AI and has undergone editorial review before publishing. Technology Networks' AI policy can be found here.