The Gap Between NAMs Promise and NAMs Practice

New Approach Methodologies (NAMs) have demonstrated tremendous promise for preclinical research and toxicology, delivering earlier results and more human-relevant data. What’s standing in the way of widespread NAMs adoption? The remaining barriers are more structural than scientific: sourcing infrastructure, standards and regulatory pathways. Understanding those gaps, and how they tend to reinforce each other, is the first step toward closing them.

The Barriers to NAMs Adoption

New Approach Methodologies (NAMs) are a growing family of tools designed to evaluate safety, efficacy and biological risk in ways that are more human-relevant than conventional preclinical approaches. The category spans complex in vitro models (including organoids, three-dimensional tissue cultures and organ-on-chip platforms built from human cells) as well as computational and in silico approaches. Here, we're focusing on the in vitro side: the biological models where the sourcing, standardization and regulatory challenges are most acute.

Despite real momentum, the gap between what NAMs can do in the lab and how they're used in practice remains wider than it should be. Three structural barriers explain most of that gap. They're more difficult to solve than they might appear, in part because they don't operate independently.

Barrier 1: Cell Sourcing

Every cell-based NAM depends on reliable biological starting material, and that's harder to come by than it sounds. Primary human cells (cells taken directly from human tissue) are physiologically relevant but finite; they can't be replenished with identical material, and results can shift from donor to donor. Induced pluripotent stem cell (iPSC)-derived cells (grown from reprogrammed human stem cells) solve the supply problem but don't always achieve the functional maturity that meaningful safety data requires, and iPSC differentiation protocols may differ between labs. Immortalized cell lines (lab-grown cells that have been modified to replicate indefinitely) offer consistency and scalability, but have been genetically altered in ways that may affect how they respond to drugs compared to primary human tissue.

Cutting across all three cell types is a subtler problem: most studies still rely on cells from a single donor, which means one genetic background, one age, one sex. Clinical trials enroll patients across a far broader range of characteristics, and drug responses vary accordingly. If NAMs are going to offer true human relevance, they need to reflect the breadth of real patient populations, not just the cells that are easiest to source.

Barrier 2: Standardization

Even with reliable biological materials, the field faces a deeper problem: the absence of agreed-upon protocols, study designs, reagents and acceptance criteria. Different laboratories running the same experiment with different methods and quality controls produce results that can't be meaningfully compared, and data that can't be compared can't build a collective evidence base.

In vivo models arrived at their current level of shared convention through decades of accumulated use: agreed-upon species and strains, established study designs, and understood variability. NAMs don't yet have that common ground. Progress is being made (the OECD's guidance on good in vitro method practices is a meaningful foundation), but significant questions remain about what acceptable variability looks like for a given application and what level of evidence is sufficient to support a specific regulatory decision.

Barrier 3: Regulatory Acceptance

For most complex endpoints, there are no established guidelines specifying what a NAMs-based submission should look like. Companies may hesitate to submit NAMs data without precedent, and regulators find it difficult to build frameworks without submissions to review. The regulatory environment is genuinely moving in the right direction, but that momentum doesn't dissolve the institutional inertia on both sides or the practical burden of running dual data packages that smaller organizations may find difficult to sustain.

Breaking the Cycle

Each of these barriers is real on its own. But what makes them particularly stubborn is that they feed each other. Unreliable cell sources make standardization harder. Without standardization, reproducibility can't be demonstrated. Without reproducible data flowing into regulatory channels, frameworks can't be built. And without clearer frameworks, the incentive to invest in better sourcing and standardization stays weak.

It's a self-reinforcing cycle. Breaking it requires pressure at multiple points simultaneously. While that is harder than solving any single piece in isolation, it also means that progress anywhere creates momentum everywhere.

The ecosystem for traditional models took decades to reach its current level of maturity and institutional trust. NAMs won't need 50 years. The pace of technology development, the scale of investment flowing into the field, and the urgency of the underlying problems are all working in their favor. But it will take sustained, disciplined effort across industry, academia and regulatory agencies to get there.

This is exactly the kind of complex, multidisciplinary problem that Battelle was built for. We are focused on generating rigorous, reproducible NAMs data within a quality framework designed for regulatory use from the start. That’s the kind of cumulative evidence base that earns confidence over time, both with industry and regulators.

Stay tuned for our next NAMs blog, where we’ll explore what it will take to move NAMs to the mainstream.