The roadmap is a comprehensive, multi-stage plan that sets out specific measures, organizational structures, and success criteria for phasing out animal testing in chemical safety assessments in the EU. It relies on innovation, collaboration, and continuous adaptation to the latest scientific knowledge. The roadmap consists of a communication document and a 253-page staff working document.

The roadmap outlines clear, concrete steps for the transition to innovative methods that do not involve animal testing. Animal testing for chemical safety assessment is to be phased out in 15 areas, including industrial and consumer chemicals, pesticides and biocides, pharmaceuticals, and food and feed additives.

The communication document ⁽¹⁾ outlines three columns: The first column aims to advance the phased phase-out of animal testing and accelerate the development and adoption of non-animal methods.  The second column aims to keep Europe at the forefront of research and innovation by supporting a diverse ecosystem of research and entrepreneurial innovation to develop animal-free approaches, including AI-based approaches. The third column focuses on cooperation within Europe and beyond. In its communication document, the EU notes that progress toward the phased-out of animal testing has been too slow on the whole.

The figures show that in 2023, more than 18.3% of all animal experiments in the EU (including Norway)—totaling 1,477,093 animals—were conducted in regulatory tests. Of these, 569,589 animals were used in quality control tests (batch safety and potency tests, as well as pyrogenicity tests), while 457,142 animals were suffering in toxicity tests, primarily in acute and subacute toxicity tests, skin sensitization tests, long-term repeated-dose tests, reproductive and developmental toxicity tests, pharmacokinetic, pharmacodynamic, as well as in ecotoxicity tests. ⁽²⁾ 

In its roadmap, the European Commission acknowledges that “non-animal approaches have the potential to be more cost-effective and faster, thereby boosting competitiveness and shortening the time to market. Further development of non-animal approaches is also necessary to strengthen the innovative capacity of industry.” To this end, the European Commission has allocated nearly €1.5 billion to research over the past two decades and created one of the world’s most advanced industries for alternatives to animal testing.

The roadmap comprises 22 measures ranging from funding research and innovation to supporting the validation, standardization, and qualification of non-animal testing approaches, as well as cooperation both within the EU and with non-EU countries.

2029 appears to be the deadline for short-term methods
For all toxicological study objectives (endpoints, e.g., acute toxicity, genotoxicity, carcinogenicity, repeated-dose toxicity, developmental and reproductive toxicity), specific short-, medium-, and long-term measures are proposed according to the three-basket principle. These range from alternatives that can be implemented immediately (e.g., in vitro and in silico methods) to the development of a new scientific assessment framework (Next Generation Risk Assessment, NGRA). The definitions of short-, medium-, and long-term are provided in the comprehensive Staff Working Document.

Short-term measures include, in particular, the adaptation of legislation and guidelines by 2029 to incorporate available animal-free approaches into regulatory practice and eliminate “unnecessary” animal testing.

Short-term measures: replacement, waiving and virtual control groups
The alternatives that can be implemented immediately primarily include in silico and in vitro methods that have already been validated being regulatory-approved for specific endpoints. In addition, animal testing can be reduced or avoided entirely through optimized study designs and the use of existing data.

The European Commission’s roadmap calls for the immediate replacement of animal testing with non-animal methods wherever possible. These alternatives are already sufficiently developed and can be implemented in regulatory practice in the short term. This includes the use of computer models such as QSAR models (Quantitative Structure-Activity Relationships) to predict the acute oral toxicity of chemicals, or pharmacokinetic modeling to predict the decline of residues in target organisms. Cell culture assays can already be used for predictions in the areas of developmental neurotoxicity, malformations, embryonic lethality, pyrogenicity, or sensitization. In environmental toxicity, the fish cell line test can serve as a substitute for acute fish toxicity tests, provided the chemical falls within the scope of application. Furthermore, animal testing can be avoided if sufficient historical data or information on similar substances are available, e.g., by waiving the second test group in long-term studies of pesticides. A reduction in the number of animals is possible through the use of virtual control groups (e.g., in repeated-dose toxicity studies in human medicine). By integrating additional study objectives into existing studies, further animal testing should be avoidable. A battery of in vitro tests can be used to assess developmental neurotoxicity. If sufficient data is available, additional animal testing can be avoided.

The burden of animals in tests should also be reduced by changing the study objective from “lethality” to “evident toxicity”: A computer model (QSAR) can replace acute toxicity tests, when the test chemical falls within the model’s scope of application. In the area of skin sensitization, validated in vitro tests should be used.

Medium-term measures: Expansion and validation of in vitro methods
Medium-term alternatives include the further development, validation, and regulatory integration of more complex animal-free methods, the combination of various approaches in integrated testing strategies, as well as the optimization of existing animal tests. This includes, for example, the development and standardization of more complex cell culture models (e.g., organoids, organs-on-a-chip) to assess genotoxicity, carcinogenicity, as well as developmental and reproductive toxicity (DART). This also includes the expansion of the application areas of QSAR models to additional chemical classes, as well as the development of guidelines and regulatory frameworks, along with case studies for integrated testing strategies (IATA) and weight-of-evidence approaches, in order to systematically combine various non-animal methods (in vitro, in silico, read-across) for complex endpoints such as carcinogenicity or endocrine disruption. 
In environmental toxicity testing, this includes the development and validation of alternatives for bioaccumulation tests (e.g., in vitro biotransformation assays, IVIVE models for air and aquatic organisms). Additional methods are listed in the document. ⁽³⁾

Long-term alternatives: a new assessment framework is in development
Long-term alternatives aim for a fundamental paradigm shift: the development and implementation of a new scientific assessment framework (Next Generation Risk Assessment, NGRA) that relies entirely on animal-free methods. This requires new concepts, methods, and regulatory adjustments to address complex toxicological questions without animal testing. The use of omics technologies (e.g., genomics, proteomics), high-throughput methods, and computer models is planned. The focus should be on describing toxicity through molecular changes rather than through observed harm to animals. In silico methods should be expanded and standardized, and big data, AI, and machine learning should be used to integrate and analyze large amounts of data from various sources. 

The testing strategies should be mechanism-based: so-called Adverse Outcome Pathways (AOPs) and quantitative AOPs (qAOPs) should be developed and used to link molecular and cellular changes to regulatory endpoints. Test batteries to study endpoints of developmental neurotoxicity, reproductive toxicity, and carcinogenicity are being developed. Complex study objectives should be achievable using validated organoids, organ-on-a-chip systems, and advanced in vitro models for complex issues such as carcinogenicity, developmental toxicity, and reproductive toxicity. In long term, for example, the in vitro test battery to assess developmental neurotoxicity is to be further developed and validated in order to establish it as a standard for the assessment of developmental neurotoxicity. To assess carcinogenic potential, weight-of-evidence approaches are being developed that combine in vitro, in silico, and, where appropriate, short-term in vivo tests.

In environmental assessment, the scope of application of alternatives such as the fish cell line test (OECD TG 249) and the fish embryo test (OECD TG 236) will be expanded and validated. For example, an animal-free framework for environmental assessment (e.g., for fish, birds, mammals) is being developed, and the assessment of biotransformation should be possible without animal testing. A uniform assessment framework based on the “One Health” principle is to be developed, integrating human and environmental aspects.

However: Important questions regarding funding remain unresolved
It should be noted that no separate funding programs have been established to date for the implementation of the roadmap. There is no single source of funding for the roadmap, according to Georg Streck, Policy Officer at the European Commission. However, the steering committee is tasked with identification and prioritization of funding opportunities for the implementation of the measures contained in the roadmap. The Commission has proposed to provide information by 2027 on options for funding activities that support these measures. 

Unfortunately: Not all animals will have a chance
Another disappointment is that not all animals used in safety testing will benefit from the phase-out plan outlined in the roadmap. Animals used to test biologics, vaccines, gene therapies, advanced therapy medicinal products, and advanced therapy veterinary medicinal products that are not classified as chemicals do not fall within the scope of the roadmap. Also excluded are the final stages of safety testing of veterinary medicinal products on target animal species.

Sources:
⁽¹⁾ European Commission (2026). Roadmap towards phasing out animal testing for chemical safety assessments. Online. https://single-market-economy.ec.europa.eu/publications/roadmap-towards-phasing-out-animal-testing-chemical-safety-assessments_en

⁽²⁾ European Commission (2026). ALURES – Animal-use reporting - EU System. Statistics Database on the use of Animals for Scientific Purposes under Directive 2010/63/EU. Section 2 – Details of all uses of animals for research, testing, routine production and education and training purposes in the EU. Online:  https://webgate.ec.europa.eu/envdataportal/content/alures/section2_number-of-uses.html

⁽³⁾ European Commission (2026).  Commission Staff Working Document. Accompanying the document Communication from the Commission Roadmap towards phasing out animal testing for chemical safety assessments. SWD (2026) 144 final. Brüssel 01.06.2026. https://ec.europa.eu/transparency/documents-register/detail?ref=SWD(2026)144&lang=en