McDermott Will & Schulte, a global law firm

On 1 June 2026, the European Commission published a communication setting out a roadmap to phase out animal testing for chemical safety assessments, supported by a detailed Staff Working Document (SWD(2026) 144). It promises 22 concrete actions and more than 30 technical recommendations across 15 legislative domains, organized around three pillars and a new governance structure.

The communication is a statement of policy intent rather than binding law: its bite will come later, through amendments to sectoral legislation such as Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH). For business and legal teams, the document is best read now as an early-warning map of where regulatory requirements, data expectations, and commercial opportunities are heading over the next several years.

What was published and its legal status

The roadmap takes the form of a Commission Communication. In EU law, a communication is a nonbinding (“soft-law”) instrument: It expresses the commission’s policy direction and commits the Commission to act, but it does not, by itself, create or change obligations for companies. Concrete duties will arise only when the Commission adopts or makes legislative proposals to implement the actions – a step it commits to taking, for the short-term actions, by the end of 2029.

The roadmap rests on an existing legal foundation rather than creating a new one. Article 13 of the Treaty on the Functioning of the EU recognizes animals as sentient beings, and Directive 2010/63/EU already sets the objective of replacing animal procedures as soon as scientifically possible. Politically, the trigger was the 2023 European citizens’ initiative “Save Cruelty-Free Cosmetics,” backed by more than 1.2 million signatures, to which the Commission pledged this response.

The scale of the issue is significant: Between 2015 and 2023, more than 15 million animals were used for regulatory testing in the EU, with almost 40% of that use attributable to chemical safety assessments.

Scope: Broad, but with carve-outs

The roadmap reaches across 15 legislative domains, including industrial chemicals (REACH), chemical pharmaceuticals, plant protection products, biocides, food and feed additives, consumer products, and medical-device biocompatibility.

Crucially for legal and regulatory teams, the scope is defined as much by what it excludes as by what it covers. Explicitly excluded domains include:

• Biologicals, vaccines, gene therapies, advanced therapy medicinal products, and novel veterinary therapies (only chemical pharmaceuticals are covered by the scope);
• Final-stage safety and efficacy testing of veterinary medicines in target animal species, reflecting specific requirements of EU law; and
• Replacement of nonprotected life stages and invertebrates, which is treated as a longer-term goal, consistent with the 3Rs prioritization in Directive 2010/63/EU.

The governing principle throughout (and the one most likely to be litigated and negotiated in practice) is that nonanimal approaches must deliver a level of protection equivalent to that of the methods they replace. This equivalent protection test is the legal and scientific pivot of the entire exercise.

The three pillars at a glance

A recurring theme across the pillars is the shift to Next-Generation Risk Assessment (NGRA), assessing safety from a mechanistic, molecular understanding of toxicity rather than from observed effects in animals. The Commission frames this transition openly as a paradigm shift that will, in time, require revisions to legislation and guidance and dovetails with the recently adopted “One Substance, One Assessment” package and the future Common Data Platform on Chemicals.

Key milestones to watch

Regulatory and business implications

For companies placing products on the market

• Anticipate a gradual reweighting of data requirements toward in vitro, in silico, and computational evidence. Several short-term actions are about omitting redundant or duplicative animal studies, which is a potential cost and time saving for dossiers if firms position early.
• Engage with the new “safe spaces” and “regulatory exploration spaces.” These allow confidential, nonbinding dialogue with regulators on whether a given alternative method would be accepted – a low-risk way to de-risk a future submission. EMA is furthest ahead; ECHA and EFSA are building comparable mechanisms.
• REACH registrants: Note the Commission’s intention to push ECHA and its Member State Committee to accept animal testing only as a genuine last resort, and to monitor testing proposal numbers as a key indicator. Testing proposal strategy and read-across justifications will come under sharper scrutiny.
• Plan for the Common Data Platform on Chemicals and “One Substance, One Assessment.” Greater data findability and reuse across sectors cuts duplication but raises questions about data ownership, confidentiality, and competitive use of shared information.

For pharmaceutical companies

Drug developers sit in a particular position under the roadmap and should read it closely, both for what it covers and for what it leaves untouched.

• Confirm scope first: Only chemical pharmaceuticals are caught; biologicals, vaccines, gene therapies and advanced therapy medicinal products are expressly excluded, as is final-stage safety and efficacy testing of veterinary medicines in target species. Map which parts of a portfolio actually fall within the roadmap before drawing conclusions.
• Use EMA’s head start: EMA is the agency furthest aligned with the “safe-space” concept, through its Innovation Task Force, voluntary “safe-harbor” data submission, scientific advice and qualification of novel methodologies. Sponsors can test regulator appetite for a nonanimal method through these channels now, before committing to a development strategy.
• Note existing flags: Near-term reductions are already flagged for human medicines, signaling fewer repeated-dose toxicity (RDT) studies for advanced-cancer or severely debilitating and life-threatening indications; complex in vitro models to predict drug-induced liver injury, cardiotoxicity, immunotoxicity, and pharmacokinetics; and “virtual control groups” to cut the number of control animals in RDT studies.
• Mind the harmonization layer: For medicines, acceptance runs through ICH (human) and VICH (veterinary) rather than REACH-style routes. A method’s cross-market usability turns on uptake in those fora, so global development plans should track ICH/VICH developments, not EU acceptance alone.
• Watch the governance signals: EMA sits on the EMA-ECHA-EFSA inter-agency working group and brings its 3Rs Working Party; coordination there will indicate where qualification expectations are converging. Separately, an ongoing Horizon Europe call on generative AI in biomedical research is expected to produce usable tools for the chemical safety assessment of medicines.

For method developers and the life sciences sector

• The Commission frames the phase-out as an industrial competitiveness opportunity, citing a projected in vitro toxicology market of up to €30 billion by 2032 and a cell-based technologies market of €26.5 billion by 2028, with the EU targeting a share of approximately 30%.
• Funding and matchmaking instruments (a €49 million Horizon Europe call, the European Innovation Council’s Advanced Innovation Challenge, and a startup/investor interface) are explicitly aimed at converting EU science into scalable businesses, many of them SMEs.

For legal and regulatory advisers

• Track the legislative follow-through. The real obligations will appear in sectoral amendments through to 2029. The Communication itself changes nothing overnight but signals drafting direction.
• Master the acceptance pathways. Whether a nonanimal method is usable in a dossier turns on validation, qualification, and standardization routes: OECD test guidelines (and mutual acceptance of data), ICH/VICH for medicines, and ISO/CEN standards, notably for medical devices.
• Watch the international layer. The Commission will press for recognition of methods at OECD, ICH/VICH, and within the GHS; cross-border acceptance will shape multinational testing strategies.

A roadmap without a destination date

Commentators have welcomed the ambition of the roadmap, while questioning its delivery. As one scientific editorial observed, even calling the proposals a roadmap might be a bit of a stretch: the document is rich in milestones but offers no estimate of how long the journey will actually take, and no firm end date for full replacement.

Three open questions stand out for anyone relying on the roadmap:

• Funding. Over two decades, the EU reports close to €1.5 billion invested in relevant research, yet critics note that spending specifically on nonanimal alternatives has averaged only around €50 million a year across all Member States, a sum widely seen as far below the level required to meet the stated ambitions.
• Scientific limits. For complex endpoints such as repeated-dose or reproductive toxicity, validated full replacements do not yet exist, and in some cases may require entirely new scientific frameworks built “from first principles.” The timeline for these long-term actions is, by definition, open-ended.
• Accountability. Without timelines attached to each milestone, it will be hard to hold institutions to account. This principle underpins advocates’ emphasis on the promised public dashboard and indicator framework.

The international backdrop adds urgency, with other jurisdictions moving in parallel (e.g., the US Environmental Protection Agency, which recently added 13 nonanimal methods to those it will accept). As a result, EU alignment (or divergence) will have direct consequences for companies operating across markets.

Practical action points

• Map your product portfolio against the 15 in-scope domains and flag where animal data currently drives your dossiers.
• Identify endpoints where short-term actions (e.g., omitting redundant studies, computational acute-toxicity models) could be leveraged soon.
• Consider early, confidential engagement through safe spaces or regulatory exploration spaces, before committing to a testing strategy.
• Assign someone to monitor the Roadmap Steering Team outputs, the 2027 regulatory-needs report, and the public dashboard from end-2026.
• For multinationals, watch OECD/ICH/VICH and GHS developments to keep testing strategies internationally coherent.