CEO, Taconic Biosciences
FDA guidance encouraging reduced reliance on animal models in preclinical research reflects a well-intentioned push toward more ethical, human-relevant science. Central to this evolution is the growing adoption of new approach methodologies (NAMs), including organoids, microphysiological systems, and computational models. These technologies are already transforming early discovery, yet they are not designed to function in isolation.
Many critical questions, particularly those involving systemic exposure, immune interactions, and long-term safety, still require fully integrated biological systems that capture the complexities of drug efficacy or safety in patients.
Across the life sciences, researchers are deeply committed to responsible and compassionate research through the 3Rs of animal research: replacement, reduction, and refinement. One way to adhere to these principles is to optimize how animal models are selected and applied. This has also led to introspection on the concept of NAMs, which is continually being refined. While often associated with non-animal systems, NAMs are increasingly defined by function rather than format, emphasizing approaches that improve human relevance with refined models, reduce overall animal use, or replace less predictive legacy methods.
In this context, FDA guidance on NAMs emphasizes approaches that replace, reduce, or refine animal use, creating space for highly refined and fit-for-purpose in vivo models to align with NAM objectives when they demonstrably reduce animal numbers or replace more resource-intensive or less informative in vivo studies.
Rather than framing progress as a binary choice between animal and non-animal technologies, the field is moving toward a more integrated paradigm. Advanced genetically engineered models (GEMs) and humanized systems can be deployed alongside in vitro and in silico NAMs, each addressing distinct translational gaps. When used strategically, these complementary tools can reduce total animal use, improve reproducibility, and generate more predictive datasets that support confident regulatory decision-making.
The path forward lies in integration, not elimination. By combining next-generation non-animal technologies with advanced, fit-for-purpose in vivo models, scientists can accelerate development timelines and improve translatability while honoring the spirit and practical application of the 3Rs. This balanced approach currently offers the most credible route to safer, more successful therapies for patients.
For regulators, this integrated model selection paradigm supports a central objective of modern guidance: enabling flexible, science-based evidence packages that are proportionate to risk, fit for purpose, and grounded in biological relevance rather than adherence to any single methodological category.
The engine behind the breakthrough
The FDA approved a milestone HIV prevention therapy: Lenacapavir, a twice-a-year injectable that represents one of the most significant advances in HIV care in more than a decade. For those of us in drug development, it was a landmark clinical success as well as a reminder that thoughtful model selection can simultaneously advance innovation and reduce animal use.
Behind the headlines, the approval was enabled by developments in advanced in vivo modeling. A highly specialized genetically engineered rasH2 mouse model, designed specifically to accelerate carcinogenicity assessment, allowed researchers to shorten preclinical development timelines by approximately 1.5 years. By replacing lengthy, more resource-intensive two-year carcinogenicity studies with a targeted approach, the six-month rasH2 model reduced total animal use while delivering faster, decision-ready safety data.
Wins like this will become more common as the life sciences industry embraces integration over elimination. The most effective preclinical strategies increasingly pair non-animal NAMs, such as in silico models and organoids, with specialized translational animal models that address questions those systems cannot yet resolve independently. Together, they enable preclinical studies that are more reproducible, more efficient, and more predictive of human outcomes.
The most meaningful reductions in animal use come not from abandoning models wholesale, but from replacing legacy approaches with smarter, more precise models—whether in vitro, in silico or in vivo—that generate the data regulators need with fewer animals and greater confidence.
This approach aligns closely with the 3Rs framework by prioritizing replacement where possible, reduction through efficiency, and refinement through improved model design. Better models reduce attrition, conserve resources, and help deliver medicines to patients more efficiently.
Expanding translational reach
The rise of humanized immune system (HIS) mice offers another example of how refined animal models can complement NAM-driven pipelines. By engrafting human immune cells into immunodeficient mouse strains, these models enable the study of human immune responses, disease mechanisms, and therapeutic interventions within an integrated biological context that cannot yet be fully replicated ex vivo.
Advanced HIS mice have become indispensable in areas such as immuno-oncology, autoimmunity, and infectious disease. When paired with patient-derived xenografts, this combination allows researchers to interrogate human tumor-immune interactions directly and evaluate emerging modalities, including checkpoint inhibitors and cell-based therapies, with greater translational relevance.
By generating richer, more human-relevant data per study, these models can reduce the number of animals required while helping to mitigate the high attrition rates that continue to challenge clinical development.
This evolution reflects a broader truth: the future of drug discovery will not be defined by choosing between animal and non-animal technologies, but by integrating both into a smarter, evidence-driven preclinical ecosystem. Human-relevant animal models, such as refined GEMs and humanized systems, will continue to play a critical role alongside organoids, microphysiological systems, and AI-enabled modeling. Each contributes distinct strengths, and together they provide a more complete and trustworthy picture of human biology and disease, which gives researchers and regulators the confidence to move forward into clinical development.
The question is no longer whether animal models still have a place in modern drug development, but how they can be used more judiciously and responsibly. When integration guides decision-making, the result is a more efficient therapeutic pipeline, greater regulatory confidence, and meaningful progress toward the goals of the 3Rs. A win for science; a win for patients; and, ultimately, a win for animals.
Mike Garrett is the CEO of Taconic Biosciences.
The post Integration Over Elimination appeared first on GEN – Genetic Engineering and Biotechnology News.
