Release the Rabbits: What Two Decades of EU Skin Irritation Testing Trends Reveal About Replacing Animal Models
By Vesna Adjei | September 17th, 2026
Innovation alone does not replace animal testing. Regulation does.
Over the past two decades, toxicology has undergone a major shift away from animal testing and toward New Approach Methodologies (NAMs), including in vitro and computational methods. This transition has been driven not only by ethical concerns, but by growing scientific criticism of animal models themselves.
Few policies shaped this transition more than the European Union’s 2006 REACH regulation (Registration, Evaluation, Authorisation and Restriction of Chemicals). REACH transformed chemical safety regulation by requiring companies to use non-animal methods whenever possible and establishing vertebrate animal testing as a “last resort.” Because compliance was necessary for access to one of the world’s largest chemical markets, REACH created strong incentives for industry investment in alternatives to animal testing.
At first glance, one might expect validated non-animal methods to immediately reduce animal use. But trends in skin irritation testing between 2004 and 2024 reveal a much more complicated transition. Rather than a simple replacement story, the EU experienced periods where both animal and non-animal testing increased before long-term declines in animal use finally emerged.
REACH Created Pressure Both For and Against Animal Testing
REACH dramatically increased demand for toxicological data. By the 2008 preregistration deadline, more than 143,000 substances had been preregistered with the European Chemicals Agency (ECHA), far exceeding early estimates. This created a major problem: completing all required testing using traditional animal models was projected to require approximately 54 million animals and cost nearly €10 billion.
Before non-animal approaches became widely accepted, skin irritation testing relied heavily on OECD Test Guideline (TG) 404, a rabbit-based method introduced in 1981. Under TG 404, chemicals were applied to shaved rabbit skin and irritation responses were measured over time.
Beginning in the early 2000s, however, reconstructed human epidermis models began emerging as alternatives. OECD TG 439, formally adopted in 2010, represented a major milestone because it specifically evaluated skin irritation using lab-grown human skin models rather than animals.
However, even as alternatives emerged, validation did not automatically produce adoption.
Early in vitro models struggled to satisfy European classification systems because regulators required nuanced hazard categories, while many early methods could only produce binary “corrosive/non-corrosive” outcomes. As a result, regulators often continued requiring confirmatory animal testing even when in vitro data already existed.
One of the clearest findings from EU testing trends is that animal use did not see steady decline until years after REACH implementation. Between 2006 and 2008, in vivo skin irritation testing increased substantially. Several factors drove this rise:
Companies rushed to generate data before REACH deadlines
TG 404 remained the only universally accepted regulatory standard
Regulators remained cautious about relying exclusively on in vitro methods
Industry feared delays in market approval if submissions were rejected
The result was a paradox: REACH accelerated investment in non-animal alternatives while simultaneously increasing short-term animal testing demand.
The Turning Point
In vivo testing continued to see an unsteady decline between 2009 and 2012, as TG 439 gained regulatory recognition, funding for alternatives expanded, and political pressure surrounding measures such as the EU Cosmetics Regulation intensified. Still, animal tests remained deeply embedded within regulatory systems. Thus, in practice, validated alternatives initially supplemented animal tests rather than replacing them.
However, after 2012, sustained declines in animal-based skin irritation testing finally emerged across the EU. Several forces drove this shift, such as the sheer scale of REACH making continued reliance on animal testing economically unsustainable and criticism of animal models intensifying as concerns about reproducibility and human relevance gained attention. At the same time, regulatory support for NAMs strengthened considerably:
The cosmetics testing ban increased pressure to avoid animal methods
Updates to REACH increasingly encouraged read-across and computational approaches
Directive 2010/63/EU required ethical review of all new animal tests
OECD TG 439 became more widely accepted
In 2017–2018, in vitro testing reached a fever pitch. The sudden peak aligned with the final 2018 deadline for REACH registrations, but also and the positive stage set for regulatory NAM support over the preceding years, as well as the fact that this deadline was specifically for low-tonnage chemicals, which are more amenable to simpler non-animal tests.
Finally, this peak was followed by a decline in both in vivo and in vitro testing which has lasted until today, as most substances have already met the REACH deadline, and toxicology has increasingly shifted toward integrated approaches combining computational modeling and existing datasets.
What the EU Experience Means for the United States
The EU experience demonstrates that replacing animal testing is not simply a scientific challenge, but also a regulatory and institutional one.
Importantly, validated alternatives existed years before widespread adoption occurred. TG 439 was validated in 2007, formally recognized in Europe in 2009, and adopted by the OECD in 2010. Yet, major increases in non-animal testing only followed broader regulatory acceptance and legal pressure from measures such as the cosmetics animal testing ban.
The availability of alternatives did not create the same level of trust or urgency as international regulatory acceptance, legal mandates, and economic incentives.
This has major implications for the United States as NAM-friendly legislation continues expanding through policies such as the FDA Modernization Act 2.0 and state-level cosmetic testing bans; Funding innovative methods is essential, but funding alone will not drive widespread replacement. Policies that actively disincentivize unnecessary animal use while creating clear regulatory pathways and incentives for validated alternatives are what ultimately shift testing behavior at scale.
The EU’s transition away from rabbit skin irritation testing was neither immediate nor linear. But over two decades, it demonstrated that meaningful reduction in animal testing becomes possible when scientific innovation is paired with strong regulatory pressure, institutional reform, and sustained political will.
References
Hartung, Thomas. “Research and Testing without Animals: Where Are We Now and Where Are We Heading?” Animal Experimentation: Working towards a Paradigm Change, 17 Apr. 2019, pp. 673–687, brill.com/view/book/edcoll/9789004391192/BP000035.xml, https://doi.org/10.1163/9789004391192_029.
Knight, Jean, et al. “4.2 Million and Counting… the Animal Toll for REACH Systemic Toxicity Studies.” ALTEX, vol. 40, no. 3, 2023, pp. 389–407, pubmed.ncbi.nlm.nih.gov/37470350/#:~:text=The%20EU, https://doi.org/10.14573/altex.2303201.
OECD. “Test No. 439: In Vitro Skin Irritation: Reconstructed Human Epidermis Test Method.” OECD, 2025, www.oecd.org/en/publications/2021/06/test-no-439-in-vitro-skin-irritation-reconstructed-human-epidermis-test-method_g1g59b2f.html.
Rovida, Costanza, and Thomas Hartung. “Re-Evaluation of Animal Numbers and Costs for in Vivo Tests to Accomplish REACH Legislation Requirements for Chemicals - a Report by the Transatlantic Think Tank for Toxicology (T(4)).” ALTEX, vol. 26, no. 3, 2009, pp. 187–208, pubmed.ncbi.nlm.nih.gov/19907906/.
Taylor, Katy. “Recent Developments in Alternatives to Animal Testing.” WBI Studies Repository, 11 Nov. 2019, https://www.wellbeingintlstudiesrepository.org/cgi/viewcontent.cgi?article=1005&context=appamet
Taylor, Katy. “Ten Years of REACH — an Animal Protection Perspective.” Alternatives to Laboratory Animals, vol. 46, no. 6, 1 Dec. 2018, pp. 347–373, https://doi.org/10.1177/026119291804600610.
Author Bio
Vesna Adjei is an intern with the Johns Hopkins Toxicology Policy Program. Her work has focused on the efficacy of historical New Approach Methodology (NAM) regulations and what that can mean as US agencies work toward the reduction of animal testing.
The views expressed do not necessarily reflect the official policy or position of Johns Hopkins University or Johns Hopkins Bloomberg School of Public Health.
