By the time TPO became prohibited in EU cosmetics on 1 September 2025, the decisive signal was already several years old.
The final regulation created the hard deadline. The scientific evidence, classification proposal, committee opinion and harmonised classification created the time to prepare for it.
TPO, or diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide, was used as a photoinitiator in some UV-cured nail products. It helped the gel harden when exposed to light. Its route from a useful ingredient to a prohibited substance shows how regulatory change travels from science into product law, and where regulatory intelligence gives a business room to act.
The deadline came at the end of a longer chain
From 1 September 2025, cosmetic products containing TPO could no longer be placed or made available on the EU market. Nail technicians and salons could not continue using affected products on clients, including stock purchased before the deadline. The European Commission confirmed that there was no general sell-through or professional use-up period.
For manufacturers, distributors and professional users, the consequences were immediate and practical:
- Identify affected formulations and stock-keeping units (SKUs).
- Reformulate or qualify alternatives.
- Withdraw remaining stock.
- Brief distributors and salons.
- Assign accountable owners before the cut-off date.
The compliance date was one point in a sequence. The sequence began with changing scientific evidence.
TPO was already under scrutiny
TPO had previously been permitted for professional use in artificial nail systems at concentrations of up to 5%. Earlier animal studies identified the testes as a target organ and indicated possible effects on male fertility. The available evidence supported classification as Reproductive Toxicant Category 2, meaning suspected of damaging fertility.
In 2014, the European Commission's Scientific Committee on Consumer Safety concluded that TPO was safe for use in professional nail-modelling products at concentrations up to 5% under the exposure conditions it assessed. The committee also identified TPO as a moderate skin sensitiser. The full reasoning is set out in the SCCS opinion on TPO (SCCS/1528/14).
That conclusion was a use-specific risk assessment based on the evidence and exposure assumptions available at the time.
Then a new study closed an important evidence gap.
One study changed the regulatory pathway
A 2019 reproductive and developmental toxicity screening study conducted under OECD Test Guideline 421 used oral exposure in rats and included an extended pre-mating period covering a complete spermatogenic cycle.
In the high-dose group, the fertility index was 0%. The study also reported severe effects involving the testes, epididymides and sperm. ECHA's Committee for Risk Assessment (RAC) concluded that the evidence supported classifying TPO as Reproductive Toxicant Category 1B, with the hazard statement H360Fd. The RAC opinion records the evidence and committee conclusion.
The earlier evidence showed reproductive-organ toxicity and raised concern about fertility. The newer study added a functional reproductive outcome. That materially increased the probability of regulatory escalation.
The scientific qualification matters. This was animal hazard evidence involving oral exposure at experimental doses. It provides no direct evidence of fertility effects among nail technicians or consumers under ordinary salon exposure.
Hazard classifications still carry legal consequences. Once the evidence supported a stronger classification, the issue moved beyond toxicology.
Classification activated product law
Sweden submitted a proposal to revise TPO's harmonised classification. RAC adopted its opinion in September 2021, supporting classification as:
- Reproductive Toxicant Category 1B, H360Fd.
- Skin Sensitiser Category 1B, H317.
The European Commission formalised the classification through Commission Delegated Regulation (EU) 2024/197, with the new classification applying from 1 September 2025.
That change activated a second legal mechanism. Article 15 of the EU Cosmetics Regulation generally prohibits substances classified as carcinogenic, mutagenic or toxic for reproduction in cosmetic products unless tightly defined conditions for an exception are met.
No exception request was submitted for TPO following the new classification. Commission Regulation (EU) 2025/877 therefore removed TPO from the restricted-substances list and added it to Annex II, the list of substances prohibited in cosmetic products.
The full pathway was:
New scientific evidence → classification proposal → scientific opinion → harmonised classification → cosmetics prohibition → product withdrawal
The mechanism matters because it makes the downstream consequence legible before the final sector-specific amendment arrives.
Each stage created a different action window
A company watching only final cosmetics legislation might have treated Regulation (EU) 2025/877 as the start of its compliance project. Most of the strategic time had already passed by then.
Each earlier stage supported a proportionate response:
| Signal stage | What became knowable | Proportionate company action |
|---|---|---|
| New scientific study | A stronger reproductive classification had become plausible. | Increase monitoring and assess the likelihood of reclassification. |
| Formal classification proposal | A regulatory process had begun. | Map affected substances, formulations, suppliers and products. |
| RAC opinion supporting Category 1B | Downstream restrictions had become substantially more likely. | Start alternatives assessment, supplier engagement and reformulation planning. |
| Harmonised classification adopted | The legal trigger and application date were clear. | Move the issue into an owned implementation programme. |
| Cosmetics prohibition published | The remaining requirements were settled. | Execute withdrawal, communication and deadline controls. |
Early action could start before the final outcome was known. The response should match the maturity of the signal. Monitoring could intensify while uncertainty was high. Product mapping and supplier questions could begin once the formal process started. Reformulation planning could follow when the committee opinion made escalation materially more likely.
This is the same discipline described in how regulators think about risk: preserve the distinction between hazard and exposure, make uncertainty visible, and still prepare for the legal and commercial pathways that credible evidence can activate.
Monitoring finds the document. Intelligence connects the consequence.
Regulatory monitoring tells you that a document has been published. Regulatory intelligence helps you understand:
- What changed in the evidence?
- How mature is the signal?
- Which legal mechanisms could it activate?
- Which products, substances, suppliers and markets may be affected?
- What action is proportionate at this stage?
The TPO pathway crossed scientific literature, a substance dossier, a committee opinion, a classification amendment, a sector-specific prohibition and an implementation deadline. A regulatory team should be able to follow that chain without reconstructing it by hand across six disconnected sources.
Run the pathway test on your own watch
Choose one substance that matters to your portfolio and work backwards from its latest regulatory event.
Can your current system show:
- The evidence that first changed the signal?
- Every formal step between that evidence and the current legal position?
- The products, suppliers and markets exposed at each stage?
- The person accountable for the next decision?
- The point when monitoring should become active preparation?
Any missing link shortens the time available to make a considered decision.
TPO is a narrow example: one substance, one product category and one hard deadline. That narrowness makes the lesson unusually clear. The final regulation was the point at which earlier scientific and regulatory signals became commercially unavoidable.
The value of regulatory intelligence lies in seeing that pathway before the deadline becomes a stock-withdrawal problem.
