PHASE 0—The FDA Just Changed What Preclinical Science Looks Like

Animal testing was the default because nothing better existed. Something better now exists.

92% of drugs that pass animal trials fail in humans. That number has been the quiet indictment of preclinical science for decades — cited in papers, acknowledged in conferences, and largely ignored in practice because there was no scalable alternative.

On March 18, 2026, the FDA formally changed that. The agency released its validation framework for New Approach Methodologies (NAMs), a document that signals a genuine paradigm shift in what preclinical data looks like before a drug reaches a human being.

This isn't a minor regulatory update. It's the official beginning of the end for animal testing as the default.

What Are New Approach Methodologies?

NAMs is the umbrella term for a set of technologies that replicate human biology with greater accuracy than animal models — and do so at a fraction of the time and cost. The four primary categories the FDA is now formally validating:

Organ-on-a-chip systems place human tissue constructs inside microfluidic devices that mimic the mechanical and biochemical environment of living organs. A lung-on-a-chip breathes. A liver-on-a-chip metabolizes. A gut-on-a-chip absorbs. These aren't analogies — they're functional human tissue systems that respond to drugs the way human organs do, not the way rodent organs do.

Organoids are lab-grown three-dimensional tissue structures derived from human stem cells. They self-organize into mini-organs that replicate the architecture and function of real tissue at a level of biological fidelity that animal models simply cannot match for human drug response.

AI and in silico modeling uses computational frameworks trained on molecular and clinical data to predict toxicity, pharmacokinetics, and drug-target interactions before a compound is synthesized. The FDA's embrace of this category is significant — it signals that regulatory-grade preclinical evidence can now be partially computational.

Real-world human safety data from international sources will now be formally integrated into preclinical assessments, reducing the need for duplicative animal studies when equivalent human data already exists.

The Regulatory Timeline That Got Us Here

This didn't happen overnight. The FDA has been moving toward this position for over a year.

In April 2025, the agency announced its intention to phase out animal testing as the default requirement for monoclonal antibodies and began formally encouraging NAMs data in all IND applications. In December 2025, draft guidance was released specifically targeting the reduction of nonhuman primate testing for mAb toxicity studies — a major signal that even the most entrenched animal testing requirements were being reconsidered.

The March 18 framework is the culmination of that trajectory. It provides the validation criteria, submission standards, and evidentiary weight that sponsors need to actually use NAMs data in regulatory submissions. It's the document the field has been waiting for.

Why Capital Needs to Pay Attention

The ethical dimension of this shift is real and worth acknowledging. But for the purposes of understanding what this means for the biotech ecosystem, the more important framing is infrastructural.

The companies building organ-on-chip platforms, organoid systems, and in silico toxicology tools are not peripheral research suppliers. They are becoming critical infrastructure for the next generation of drug development. Every IND application that incorporates NAMs data creates a dependency on the platforms that generate it. That dependency compounds.

Watch for increased M&A and licensing activity around NAMs technology platforms over the next 18 months. Large pharma companies that have been quietly piloting these technologies in their preclinical workflows are about to formalize those relationships. The validation framework gives them the regulatory cover to do so.

The Signal

Preclinical data is about to look fundamentally different. The trials designed on the back of NAMs evidence will carry different risk profiles, different timelines, and different cost structures than anything the industry has built before. Capital that hasn't priced this transition in yet will need to.