NIH Just Committed $150 Million to Replace Animal Testing. The Biotech Industry Should Be Paying Attention.

The mouse has been biomedical research's default model organism for a century. It has also been its most persistent bottleneck. Nine out of ten drugs that work in mice fail in human clinical trials — a ratio that has barely budged despite billions spent refining rodent genetics, housing conditions, and experimental protocols. The pharmaceutical industry has long accepted this failure rate as the cost of doing science. The National Institutes of Health just signaled that it no longer does.

On March 18, NIH announced the first awards under Complement Animal Research in Experimentation — Complement-ARIE — a $150 million program to develop, standardize, and validate human-based research methods that can complement and, in some cases, replace traditional animal studies. The program is the largest single federal investment in what the regulatory world calls New Approach Methodologies, or NAMs: organ-on-chip systems, human organoids, computational models, and cell-based assays that aim to simulate human biology more accurately than a mouse ever could.

This is not an animal rights initiative wearing a lab coat. It is a scientific infrastructure program with regulatory teeth, and it arrives at the exact moment the industry needs it.

The Failure Rate Problem

The 90 percent attrition rate between animal studies and human clinical trials is not a statistic that pharmaceutical companies enjoy discussing, but it is the central fact driving Complement-ARIE's existence. Drug development timelines average twelve to fifteen years. The cost of bringing a single drug to market now exceeds $2 billion. And the primary reason both numbers keep rising is that the preclinical models used to predict human drug response — overwhelmingly mouse and rat studies — are unreliable predictors of what happens when the same compound enters a human body.

The reasons are well understood. Mice and humans diverged evolutionarily roughly 75 million years ago. Their immune systems, metabolic pathways, and disease progressions differ in ways that matter enormously for drug development. A cancer drug that shrinks tumors in immunocompromised mice may do nothing in a patient with a fully functional immune system. An Alzheimer's therapy that clears amyloid plaques in transgenic mice may fail to prevent cognitive decline in humans because the mouse model does not capture the full complexity of human neurodegeneration.

Congress recognized this problem in the FDA Modernization Act 2.0, signed in December 2022, which eliminated the federal requirement that drugs be tested on animals before entering human trials. That law removed a legal barrier. Complement-ARIE is building the scientific infrastructure to walk through the door.

What Complement-ARIE Actually Funds

The program has three interlocking components, each addressing a different bottleneck in the NAMs pipeline.

Technology Development Centers

Seven new technology centers received the bulk of the $150 million. Each focuses on a disease area where animal models are particularly poor predictors of human outcomes — and where better models could accelerate drug development most dramatically.

The focus areas include gynecological disorders, cardiac disease, neurological disorders, and rare diseases. Johns Hopkins University received $15 million over five years to build a platform for studying neurological diseases and screening environmental chemicals using human brain organoids and microphysiological systems. The team will develop models of Parkinson's disease, autism, and chemical neurotoxicity — conditions where mouse models have been notoriously misleading.

Each center is designed to produce not just research papers but validated, reproducible tools that other labs can adopt. This is a critical distinction. The NAMs field has suffered from a credibility problem: too many one-off demonstrations in single labs that cannot be reproduced elsewhere. Complement-ARIE requires each center to develop standardized protocols and share data through a centralized repository.

The NAMs Data Hub

New York University received $25 million over five years to build the NAMs Data Hub and Coordinating Center — a centralized infrastructure for data sharing, protocol standardization, and cross-center coordination. The hub will create integrated data structures and a searchable NAMs repository, allowing researchers across the country to access validated models and compare results using common metrics.

This is the kind of infrastructure investment that rarely generates headlines but determines whether a field matures from academic curiosity into standard practice. Without common data formats and shared repositories, every lab reinvents the wheel. With them, a validated organoid protocol developed at Johns Hopkins can be deployed at a pharmaceutical company in Basel the same month.

The Validation and Qualification Network

NIH plans to contribute approximately $20 million to the Validation and Qualification Network, or VQN, managed by the Foundation for the National Institutes of Health in partnership with industry. The VQN's job is the hardest and most consequential: taking human-based models that work in a research setting and preparing them for regulatory acceptance.

A NAM is scientifically interesting. A NAM that the FDA accepts as evidence in a drug approval application is transformative. The VQN has selected four pilot projects — preterm birth, developmental neurotoxicity, inhalation toxicity, and acute oral toxicity — with plans to expand. Each project will run a human-based model through the gauntlet of reproducibility testing, inter-laboratory validation, and regulatory documentation that the FDA requires before it will consider a new methodology as supporting evidence.

The FDA Connection

Complement-ARIE does not exist in a regulatory vacuum. The same week NIH announced the program's first awards, the FDA released new draft guidance on using NAMs in drug development — the most concrete regulatory signal yet that the agency is prepared to accept human-based models as complementary or, in some cases, alternative evidence to animal studies.

NIH and FDA formalized their collaboration through a memorandum of understanding that establishes a joint framework for developing, qualifying, and implementing NAMs across the research-to-regulatory pipeline. The EPA is also a partner, reflecting the fact that chemical safety testing — not just drug development — relies heavily on animal models that NAMs could improve.

This three-agency alignment matters because the historical barrier to NAMs adoption has been regulatory, not scientific. Researchers have been building organ-on-chip systems and human organoids for over a decade. What they have not had is a clear pathway for those models to influence actual regulatory decisions. Complement-ARIE and the parallel FDA guidance are designed to build that pathway simultaneously from both ends — the scientific infrastructure side and the regulatory acceptance side.

The $7 Million Challenge

In addition to the technology centers and infrastructure investments, NIH launched the NAMs Reduction to Practice Challenge — a $7 million competition in partnership with the FDA and EPA. Research teams must submit human-based models and demonstrate their viability within a three-year period. Successful submissions will be delivered to the VQN for validation and eventual deployment.

The challenge structure is deliberate. Unlike grants, which fund research proposals, the challenge pays for results. Teams that cannot demonstrate a working, reproducible model within three years receive nothing. This mirrors the milestone-based approach that ARPA-H has brought to healthspan research through programs like PROSPR and signals a broader federal shift toward outcome-driven science funding.

Who Should Be Paying Attention

Biotech and Pharmaceutical Companies

If your preclinical pipeline still relies exclusively on animal models, Complement-ARIE is a signal that the regulatory landscape is shifting beneath you. The FDA's draft guidance on NAMs acceptance means that companies investing in human-based models now will have a competitive advantage when the guidance is finalized. Companies that wait will find themselves scrambling to adopt validated NAMs that their competitors helped develop.

The VQN's pilot projects — preterm birth, neurotoxicity, inhalation toxicity, and oral toxicity — are not random selections. These are areas where the FDA is actively seeking alternatives to current testing requirements. Companies with drug candidates in these therapeutic areas should be engaging with the VQN now, not waiting for final regulatory guidance.

Academic Researchers

Complement-ARIE's technology centers are funded, but the program's infrastructure — the data hub, the validation network, the standardized protocols — is designed to be used by the entire research community. Academic labs developing NAMs should treat the NYU data hub as a resource and a publication pathway. Models submitted to the hub and validated through the VQN carry regulatory credibility that a standalone research publication does not.

NIH has also begun requiring researchers to address how NAMs could complement animal models in grant proposals. This is not yet a mandate, but the direction is clear. As former NIH official Danilo Tagle put it: "NIH decided that it is time to move this field forward in a big way." Researchers who integrate NAMs into their proposals now will be better positioned as the requirement tightens.

SBIR and Small Business Innovators

The intersection of Complement-ARIE and the recently reauthorized SBIR/STTR programs creates a specific opportunity for small companies developing NAMs technologies. Companies building organ-on-chip platforms, computational toxicology models, or high-throughput organoid screening systems should examine whether their technology fits the Complement-ARIE roadmap. The VQN's expansion beyond its four pilot areas will create demand for validated tools from commercial suppliers.

The Honest Limitations

Complement-ARIE is not the end of animal research, and NIH has been careful to frame it as complementary rather than replacement. "The proposal is not to end animal experimentation but explore the limits where we can reduce it," as Columbia's Gustavo Stolovitzky has noted.

The field also faces genuine scientific limitations. Arnold Kriegstein at UCSF has warned of "a tendency to prematurely use imperfect models as though they were already perfected" — a particular concern with brain organoids that lack the vascular networks and immune cells present in actual human brains. Complement-ARIE's validation infrastructure is designed to address exactly this problem, but the process of distinguishing validated models from impressive demonstrations will take years.

Still, the trajectory is unmistakable. Europe launched complementary initiatives including the EU Biotech Act and the VISI-ON-BRAIN program for training researchers in computational alternatives. The global scientific community is converging on the same conclusion: human-based models are not just ethically preferable but scientifically necessary.

For researchers and companies navigating this transition, the question is no longer whether NAMs will become standard practice but how quickly — and whether you will be building the tools or buying them. Platforms like Granted can help you identify the federal funding opportunities that are accelerating this shift and position your next proposal at the leading edge of where the science is heading.