Global regulators have spent the last two years making their position on New Approach Methodologies (NAMs) impossible to miss. The FDA has published a roadmap to phase out animal testing for monoclonal antibodies. The UK has released an aggressive strategy to replace animals in science. The EU will imminently publish its own roadmap toward phasing out animal testing. The CDC is eliminating nonhuman primate research in agency laboratories.
The signal could not be clearer. Replace animals where possible. Reduce everywhere else. Improve human-relevant prediction. Accelerate the science.
And yet, actual NAMs uptake in regulatory submissions remains relatively low.
Despite the hype, the headlines, the conferences, and the policy documents, sponsors are not racing to submit programs built around NAMs. Why?
The Surface Reasons
There is no shortage of explanations on offer.
Large pharma is structurally slow to change. Internal SOPs, validated workflows, and entrenched preclinical packages take years to revise. Quality and regulatory teams have built their careers on what works today, not on what might work tomorrow.
Scientific uncertainty remains real. NAMs cover a substantial proportion of safety risks, but important gaps remain including in metabolism, immune complexity, and systemic responses.
Qualification pathways are still maturing. The FDA’s ISTAND program is now permanent, but only a small number of technologies have been disclosed in the DDT database so far.
Each of these explanations is real. But none of them get to the core problem.
The core problem is simpler. It is a risk calculation.
What Sponsors Actually See
Set aside the policy documents and put yourself in the seat of a Chief Scientific Officer or VP of Regulatory Affairs preparing an IND submission. Your job is to maximize the probability of approval, on schedule, with the data package you have.
Now look at what regulators are actually offering you if you adopt NAMs.
In most jurisdictions, animal testing is no longer strictly “required” by statute. FDA Modernization Act 2.0 directly indicates this in the United States. The same general direction is reflected in EU and UK policy.
But “not required” and “not expected” are very different things.
In practice, the default reviewer expectation across nearly every major program remains a conventional package built on rodent and non-rodent species. A sponsor proposing to depart from that default carries the entire burden of justifying it. The reward for doing so is, at best, a slightly faster or the same approval timeline they would have had with conventional studies. The downside is a longer list:
- Outright rejection of the package as insufficient
- Information requests that delay the program by months or quarters
- Conflicting data between NAM results and any retained animal studies
- Reviewer scepticism that propagates through later cycles
- Internal second-guessing by the sponsor’s own quality and regulatory teams
This is not paranoia. It is a rational reading of the current environment.
These risks could sink a pipeline or, in some cases, an entire company.
The Two-Path Trap
When it comes to NAMs, the sponsor effectively has three options. All of them carry risk, but one is a clear winner.
Path 1: Skip animal studies entirely and submit a NAMs-only package.
This is the option regulators publicly say they are working towards. It is also the option with the largest reputational and timeline risk. If the reviewer is unconvinced, the program stalls. There are very few precedents for sponsors to point to. A small biotech with 18 months of cash runway cannot afford to be the test case.
Path 2: Run animal studies and add NAMs as a complementary package.
Safer in theory. But this introduces a new risk that is rarely discussed openly. What happens if the NAMs data and the animal data disagree?
A negative animal finding paired with a clean NAM result (or vice versa) is not a win. It is a discussion. Reviewers will ask which dataset is correct, which dataset should drive labeling, and whether additional studies are needed to resolve the conflict. Even when the NAM is right, the conflict itself can add a review cycle. Sponsors have learned that submitting more data is not always better. It is sometimes just slower.
Path 3: Run animal studies as you always have and limit NAMs to “other”.
This is the path of least resistance. It is the path most sponsors are actually taking. It produces no innovation credit and no regulatory bonus, but it also produces no novel risk.
Look at this from the perspective of a portfolio manager. Path 3 has a known cost and a known timeline. Paths 1 and 2 have wider distributions on both. For any sponsor whose existence depends on the next program reaching the clinic, the rational choice is Path 3 every time.
This is the conundrum. The science is improving. The policy is supportive. The technology is here. The economics still point the wrong way.
Who Owns the Risk
Regulators have been clear about wanting NAMs to succeed. What they have been less clear about is who absorbs the risk when something goes wrong.
Right now, that answer is the sponsor. The sponsor pays for the NAMs studies. The sponsor pays for any conventional study they retain alongside it. The sponsor absorbs the timeline cost of any reviewer pushback. The sponsor carries the liability if a clinical signal emerges that an animal study “might have caught”.
Large pharma can technically afford this risk but has no incentive to take it on a single asset. The portfolio is large, the standard pathway works, and the marginal benefit of a NAMs-led program is unclear. Biotech and small sponsors cannot afford the risk at all. A single delay can be terminal.
The result is a market where the people best positioned to use NAMs in real submissions have the strongest incentives not to.
What Regulators Could Actually Do
If regulators want adoption to match their rhetoric, the risk-reward balance has to shift. A few concrete suggestions follow.
1. Publish clear guidance on what happens when NAMs and animal data conflict.
Sponsors need to know the decision rule before they submit, not after. Which dataset takes precedence under what circumstances? When does a conflict trigger an information request versus a labelling decision versus a request for additional studies? Reducing this ambiguity alone would meaningfully change the risk calculations.
2. Offer concrete review-timeline incentives for qualified NAMs.
The FDA roadmap explicitly references streamlined reviews and fast-track meeting requests for sponsors who use NAMs. That language is welcome, but it is not operational. There is no published clock. There is no defined reduction in review days.
The most useful place to start is the information request cycle. IRs are where most NAMs-inclusive programs will actually lose time, because reviewers will reasonably have more questions about novel data than about a conventional package. If programs that include a qualified Drug Development Tool within its qualified Context of Use received accelerated IR review, prioritized response handling, or a defined cap on IR cycles, the timeline penalty for adopting NAMs would shrink dramatically. Combine that with a published review-clock reduction and fee adjustments, and sponsors finally have something a CFO can put in a slide.
3. Provide novel resolution pathways for early NAMs adopters.
If an IND or CTA is rejected or delayed specifically due to the use or inclusion of NAMs, new mechanisms must be made to accelerate the resolution of these issues. These may include options for secondary review by a NAMs-trained review panel or the creation of a binding reviewer meeting to resolve concerns and agree on next steps (e.g., “post”-IND meeting).
4. Define which specific scientific questions NAMs can and need to answer.
The March 2026 FDA draft NAMs guidance is explicit that it “is not intended to address specific NAMs”, and the December 2025 monoclonal antibody draft says a Weight of Evidence assessment “may include” NAMs without specifying how NAMs would be included. These are reasonable for regulatory guidance but lack the clarity sponsors need. Endpoints like hepatotoxicity, immunotoxicity, and developmental toxicity are offered as illustrations of how to think about Context of Use, not as endpoints with defined acceptance criteria. The real list of endpoints will need to be granular (DILI alone covers hepatocellular injury, cholestasis, mitochondrial dysfunction, and immune-mediated injury). Without defining exactly which endpoints NAMs are permitted to answer and what combination can replace a specific test, “early consultation with the review division” becomes the only path forward, which is exactly the kind of bilateral, unscalable process that keeps adoption slow.
The Path Forward
NAMs will not fail because of bad science. The science is good and getting better. They will fail to translate, or translate slowly, if the regulatory framework continues to ask sponsors to volunteer for risk that the framework itself created.
The hype is real. The policy is real. The technology is real. The translation gap is also real, and it is not a scientific gap. It is an incentive gap.
Regulators have built the runway. Now they need to give sponsors a reason to take off from it.
At InnovApproach Consulting, we help sponsors and NAM developers build qualification strategies, regulatory packages, and ISTAND submissions that turn promising science into accepted practice. If you are weighing the trade-offs of including NAMs in your next program, get in touch.
No responses yet