The Replacement Trap: NAMs Are Built for Integration not One-for-One Animal Replacement
By Zachary Liebowitz| August 8th, 2026
New Approach Methodologies are often discussed in terms of their potential for replacement within the regulatory field. Many question whether these methods are ready to “take the place” of animal testing in regulatory decision making as one-for-one replacements. This framing, which has popped up in both scientific and policy discussions, misinterprets a more complex reality. Almost all NAMs are not intended to be a direct replacements for traditional animal studies. At the same time, setting the bar at simple replacement as the primary benchmark obscures the roles these methods can play when integrated and considered together.
The right way to understand NAMs is to view them across the continuum of scientific information needed for decision-making. This continuum reflects how well a method is characterized, how consistently it performs, and how clearly its outputs can be interpreted. Some methods are still being explored and refined. Others are in common use and more reproducible and reliable. A smaller subset has reached a level of standardization that allows for consistent use across a set of contexts. Recognizing this variation is important because it shapes a realistic perspective on how these methods can perform.
At earlier points along this continuum, methods often have less well-defined performance characteristics. Reproducibility may vary across laboratories; sources of variability may not be fully understood, and the relationship between measured outputs and real-world biological outcomes may still be under investigation. However, these limitations do not make the methods uninformative and they still could have value for decision-making. In many cases, they provide valuable insight into how chemicals interact with biological systems, even if that information cannot yet be translated directly into regulatory endpoints.
As methods become more developed and are applied in different contexts, their performance and the value of the information they produce becomes clearer. This growing body of evidence allows these methods to be used with greater confidence, particularly when they are combined with other sources of information, integrated with data produced by other NAMs. In these settings, NAMs can help strengthen conclusions, reduce scientific uncertainty, and guide further testing.
As NAMs become more mature, some reach a level of standardization that allows them to be used in a more structured and reliable way. Protocols are more consistent, outputs are better understood, and interpretation becomes more predictable across different contexts. Even at this stage, however, use is often specific to particular questions rather than broadly applicable across all regulatory decisions.
This progression reflects the evolution of the methods themselves. It is separate from the question of how they are used in policy. Regulatory systems apply their own criteria for evidence, and those criteria often emphasize consistency, comparability, and defensibility. Methods that do not meet these expectations may still be scientifically informative, but they are less likely to be used as primary decision tools.
This helps explain why many NAMs remain in supporting roles. They contribute to weight of evidence assessments, inform prioritization, or provide mechanistic context, but they do not replace traditional tests outright. This is not necessarily a failure of the methods. It reflects how regulatory systems are structured and what they require from evidence.
Focusing on one-to-one replacement can create a misleading picture of progress. It suggests a binary transition from animal testing to alternative methods, when in reality the shift is more nuanced. Methods develop over time, and their roles expand as their performance becomes better understood. During this process, there is a long period in which methods are useful but not sufficient on their own.
Recognizing this spectrum of development allows for a more practical approach. Instead of asking whether a method can replace an animal study, it becomes possible to ask how it can contribute at its current level of development. Early methods can help identify potential hazards and guide research priorities. More developed methods can support structured analyses alongside other data. Methods with higher levels of standardization can inform specific regulatory questions where their performance is well established.
Recognizing the need to phase in NAMs, and the likelihood of integration of data across NAMS. does not lower standards for evidence. It aligns expectations with what methods are capable of delivering at a given point in time. It also allows for more efficient use of available information. Rather than waiting for methods to reach a single threshold of readiness, regulatory systems can begin to incorporate them in ways that reflect their strengths and fill knowledge gaps.
There are already frameworks that support this type of integration. Weight of evidence approaches and integrated testing strategies are designed to combine multiple sources of data in a structured way. These approaches recognize that no single method is likely to provide all the information needed for complex decisions. They also provide a pathway for incorporating NAMs as they develop, rather than only after they reach full standardization.
Even so, the use of NAMs within these frameworks remains uneven. In some cases, methods are included but given limited weight. Greater clarity around how to evaluate methods at different stages of development could help address these challenges. It could also encourage more consistent use of NAMs across agencies and contexts.
The future of regulatory toxicology depends on the successful incorporation of new knowledge from NAMs into decision-making. It is clear that NAMs are already contributing to scientific understanding and can improve decision making when used appropriately. The challenge is to move beyond a binary view of readiness and toward a more nuanced understanding of how these methods develop, how they are used, and where they fit.
Most NAMs are developing as valuable scientific tools and their future use will increase to the benefit of public health protection and disease treatments and cures. Right now, they're not yet ready to replace animal testing. That reality should not be seen as a limitation of the field, but as a reflection of where it is in its development and shows a need for more support. Recognizing the spectrum of development across NAMs provides a more realistic foundation for thinking about their role, both now and in the future.
The views expressed do not necessarily reflect the official policy or position of Johns Hopkins University or Johns Hopkins Bloomberg School of Public Health.
