New Approach Methodologies (NAMs) have demonstrated significant potential to improve how safety and risk are evaluated in preclinical research and toxicology. By delivering results earlier in development and generating more human-relevant data, these approaches have moved beyond the conceptual stage. Yet despite that progress, NAMs are still not used as broadly or consistently as their capabilities suggest they should be.
The reasons are not rooted in scientific uncertainty. Instead, they stem from a set of structural challenges tied to sourcing, standardization and regulatory pathways. These barriers reinforce each other, slowing the transition from technical promise to routine practice.
Understanding where those constraints arise is critical to closing the gap.
Structural Obstacles to NAMs Adoption
NAMs encompass a wide range of approaches, including in vitro models such as organoids, three‑dimensional tissue cultures and organ‑on‑chip systems, as well as computational and in silico tools. While all NAMs face adoption challenges, those obstacles are most pronounced for in vitro models, where biological sourcing, methodological consistency and regulatory readiness are tightly intertwined.
Despite growing momentum, there remains a disconnect between what NAMs can achieve in controlled laboratory settings and how they are applied in high‑stakes decision‑making. Three structural barriers explain much of that disconnect.
Barrier 1: Cell Sourcing
Every cell‑based NAM depends on the quality and reliability of its biological starting material, and obtaining that material at scale presents real trade‑offs.
Primary human cells offer strong physiological relevance, but their supply is limited and inherently variable. Donor‑to‑donor differences can influence results, and identical material cannot be replenished over time. Induced pluripotent stem cell (iPSC)‑derived cells address supply constraints, but they do not always reach the functional maturity required for certain safety endpoints. Immortalized cell lines provide consistency and scalability, yet their genetic modifications may alter how they respond to compounds compared with native human tissue.
Beyond these individual trade‑offs lies a broader challenge. Many NAMs studies still rely on cells from a single donor, representing one genetic background, age and sex. Clinical responses, by contrast, vary widely across populations. For NAMs to deliver meaningful human relevance, they need to reflect that diversity rather than relying solely on the most accessible materials.
Barrier 2: Standardization
Even when biological sourcing issues are addressed, progress is limited by the absence of shared standards. Across the field, there is little agreement on protocols, reagents, study designs or acceptance criteria. As a result, similar experiments conducted in different laboratories may yield results that cannot be directly compared.
Without comparability, data cannot accumulate into a coherent evidence base. That is particularly problematic in regulatory contexts, where confidence depends on reproducibility across studies and institutions.
Traditional in vivo models reached their current level of acceptance through decades of shared use, eventually leading to common expectations around variability and study design. NAMs are still earlier in that process. While guidance such as the OECD’s work on good in vitro method practices provides an important foundation, questions remain about acceptable variability for different applications and the level of evidence needed to support specific decisions.
Barrier 3: Regulatory Acceptance
Regulatory pathways continue to evolve, but gaps remain, especially for complex endpoints. In many cases, there are no established precedents defining what a NAMs‑based submission should include or how it will be evaluated.
This creates a challenging dynamic. Companies may hesitate to submit NAMs data without clearer expectations, while regulators face difficulty building frameworks without sufficient submissions to review. Although regulatory momentum is real, institutional inertia and the practical burden of maintaining parallel data packages remain obstacles, particularly for smaller organizations.
Breaking a Self‑Reinforcing Cycle
These barriers don’t operate in isolation. Inconsistent cell sourcing complicates standardization. Without standardization, reproducibility is difficult to demonstrate. Without reproducible data entering regulatory channels, frameworks develop slowly. And without clearer frameworks, incentives to invest in better sourcing and standardization remain limited.
The encouraging reality is that progress at any point in this cycle can drive progress elsewhere. Improvements in sourcing can support consistency. Better standardization can yield more interpretable data. That data can then help inform regulatory approaches and build confidence over time.
The ecosystem supporting traditional in vivo models took decades to mature. NAMs are unlikely to require that long. Advances in technology, increased investment and the urgency of reducing late‑stage failure all favor faster movement. Still, meaningful adoption will depend on sustained, coordinated effort across industry, academia and regulatory agencies.
Addressing these challenges requires approaches that prioritize rigor and regulatory relevance from the outset. Generating reproducible NAMs data within quality systems designed for regulatory use is essential to building the cumulative evidence base that supports confident decision‑making.
NAMs have the potential to reshape how safety and efficacy are evaluated. Realizing that potential will depend not only on better tools, but on the structures that allow those tools to be used consistently, credibly and at scale.
The editorial staff had no role in this post's creation.