A positive EFSA opinion is the moment the trade press reports and the moment founders tell investors about. It is not the moment a product can be sold. Between the two sits an administrative stage that, across the nine amendments the European Commission made to the Union list of novel foods in 2026, took anywhere from 151 days to 320 days.
That is a spread of 169 days — five and a half months of variance on a step that most launch plans treat as a formality.
The statutory rule governing that stage is precise, and it is not the rule most applicants think it is. Article 12(1) of Regulation (EU) 2015/2283 opens: “Within seven months from the date of publication of the Authority’s opinion, the Commission shall submit to the committee referred to in Article 30(1) a draft implementing act authorising the placing on the market within the Union of a novel food and updating the Union list …”
Three things in that sentence do work that is easy to miss. The clock starts at publication, not adoption. It ends at submission of a draft to a committee, not at authorisation. And neither of those is the date the product may legally be placed on the market.
Methodology
Every date below comes from the instrument itself. The Commission’s Union list page indexes the implementing regulations that have amended Implementing Regulation (EU) 2017/2470; nine carry 2026 numbers, and each was read in the Official Journal. The recitals of each regulation state the date the Authority adopted its scientific opinion. The signature block states the date the regulation was adopted.
The regulations do not state the date the opinion was published, which is the date Article 12 actually keys to. For that we used the EFSA Journal record for the opinion cited in each regulation’s footnotes, taking the date the article record was registered as the publication date. This is the best public proxy available; it is not a Commission-certified date, and readers who need certainty for a filing should confirm it against the EFSA Journal entry directly.
Two entries in the table are extensions or changes of use rather than first authorisations. They are included because Article 12 governs them identically, and excluding them would shrink an already small sample.
The nine 2026 amendments
| Novel food | Regulation | Opinion adopted | Opinion published | Regulation adopted | Adopted to published (days) | Published to regulation (days) |
|---|---|---|---|---|---|---|
| Lacto-N-tetraose | (EU) 2026/397 | 10 Jul 2025 | 13 Aug 2025 | 23 Feb 2026 | 34 | 194 |
| Inulin-propionate ester | (EU) 2026/1219 | 25 Jun 2025 | 24 Jul 2025 | 9 Jun 2026 | 29 | 320 |
| cRG-I enriched carrot fibre | (EU) 2026/1306 | 25 Jun 2025 | 29 Aug 2025 | 11 Jun 2026 | 65 | 286 |
| Defatted rapeseed powder | (EU) 2026/386 | 27 Aug 2025 | 15 Sep 2025 | 20 Feb 2026 | 19 | 158 |
| Pasteurised Akkermansia muciniphila (extension of use) | (EU) 2026/391 | 27 Aug 2025 | 25 Sep 2025 | 23 Feb 2026 | 29 | 151 |
| Jatropha curcas L. kernels (conditions and specifications) | (EU) 2026/1378 | 29 Sep 2025 | 27 Oct 2025 | 23 Jun 2026 | 28 | 239 |
| Rhizomucor pusillus mycelium | (EU) 2026/1507 | 29 Sep 2025 | 1 Dec 2025 | 3 Jul 2026 | 63 | 214 |
| Egg membrane hydrolysate | (EU) 2026/1427 | 29 Sep 2025 | 3 Dec 2025 | 2 Jul 2026 | 65 | 211 |
| Galacto-oligosaccharide (extension of use) | (EU) 2026/1749 | 19 Nov 2025 | 16 Dec 2025 | 20 Jul 2026 | 27 | 216 |
Median time from publication of the opinion to adoption of the regulation: 214 days. Seven months is roughly 213. The median case, in other words, lands almost exactly on the statutory figure — while individual cases sit 62 days inside it and 107 days outside it.
Practical consequence: if you are modelling a launch date from a positive opinion, seven months is the central estimate and not the ceiling. Two of the nine took more than ten months. Budget for the tail, not the median.
Adoption and publication are different dates, by up to 65 days
The first column pair is the one that trips people up. EFSA adopts an opinion at a panel meeting; the EFSA Journal publishes it later. Across these nine, that gap ran from 19 days to 65 days, median 29.
The regulation cites the adoption date. Article 12 runs from the publication date. So the document that tells you when the opinion happened is silent on the date that starts the Commission’s clock.
For Rhizomucor pusillus mycelium — the Dutch company The Protein Brewery’s Fermotein, and the first mycelium ingredient through the EU novel food framework — the two dates are 29 September 2025 and 1 December 2025. Sixty-three days apart. An applicant reading recital (6) of Regulation (EU) 2026/1507 and adding seven months to the date printed there would have expected a draft act at the committee by late April 2026. The clock did not start until December.
Practical consequence: when you are told an opinion has been adopted, the Article 12 clock has not started. Ask for the EFSA Journal publication date, and date the schedule from that.
What the seven months actually binds
Here the honest reading matters more than the rhetorically convenient one.
Five of the nine regulations were adopted after the date seven months from publication of the opinion. That is not evidence that the Commission missed a deadline. Article 12 requires the Commission to submit a draft to the Standing Committee on Plants, Animals, Food and Feed within seven months. What follows — the committee’s examination procedure under Article 30(3), the vote, and formal adoption — sits outside the seven-month window entirely.
In the Fermotein case the committee voted in favour in May 2026, comfortably inside seven months of the December publication, and the regulation followed on 3 July. The Commission appears to have complied. The applicant still waited 214 days after publication, and 277 days after the opinion was adopted, before an instrument existed.
The deadline, in other words, binds a step no applicant can observe, and does not bind the step every applicant is waiting for.
And there is one more date after that. Every one of the nine regulations carries the same final article: “This Regulation shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union.” Adoption is not market access either. Twenty more days, plus the interval between adoption and Official Journal publication.
Practical consequence: a launch plan needs four dates, not one — opinion published, draft submitted to committee, regulation adopted, regulation in force. Only the first and the last are visible to you, and only the last permits a sale.
The stage before this one is worse
Article 11(1) gives the Authority nine months from receipt of a valid application to adopt an opinion. Article 11(4) lets that period be extended whenever EFSA asks the applicant for additional information, and in practice that provision does most of the work.
Where the 2026 regulations state the date the Commission asked EFSA for an opinion, the elapsed time to adoption was:
| Novel food | Commission request to EFSA | Opinion adopted | Days |
|---|---|---|---|
| Galacto-oligosaccharide (extension of use) | 18 Jul 2025 | 19 Nov 2025 | 124 |
| Pasteurised Akkermansia muciniphila (extension of use) | 12 Feb 2024 | 27 Aug 2025 | 562 |
| cRG-I enriched carrot fibre | 8 Nov 2023 | 25 Jun 2025 | 595 |
| Jatropha curcas L. kernels | 19 Jan 2024 | 29 Sep 2025 | 619 |
| Rhizomucor pusillus mycelium | 7 Jan 2021 | 29 Sep 2025 | 1,726 |
| Inulin-propionate ester | 16 Jul 2018 | 25 Jun 2025 | 2,536 |
One of six came in under the nine-month statutory period. Inulin-propionate ester was referred to EFSA in July 2018 and authorised in June 2026 — 2,885 days, just under eight years, from referral to instrument.
The Protein Brewery’s chief executive gave AgFunderNews a specific account of why its own assessment ran to 1,726 days: two rounds of additional studies, requested two and four years after submission, because “assessment questions were being defined as we went.” That is Article 11(4) operating as designed. It is also why any timeline built on the nine-month figure is a fiction.
A correction to our own record
This publication got the same distinction wrong three weeks ago.
Our 13 August analysis of The Protein Brewery’s divergent EU and US outcomes stated that the Commission adopted the implementing regulation on 17 June 2026, and that EFSA adopted its positive opinion in December 2025. Both were sourced from the Commission’s comitology register and contemporaneous trade reporting, and both are wrong against the Official Journal.
Regulation (EU) 2026/1507 is dated 3 July 2026. EFSA adopted the opinion on 29 September 2025 and published it in December. The 17 June date corresponds to the comitology stage, which the trade press reported at the time as approval; the instrument came sixteen days later.
The article’s central finding — that the EU route finished while the US GRAS notice closed with no decision — is unaffected, and its “six years” figure holds: application 16 May 2020, authorisation 3 July 2026. But the specific dates were wrong, and we are correcting them here and on the original piece.
We are flagging our own error rather than quietly fixing it because it is the best available evidence for the argument. If a publication reading the primary instruments conflates comitology with adoption and adoption with publication, an applicant working from a press release will do the same — and will build a supply agreement around it.
The counter-argument
The strongest case against reading much into this: nine observations is a small sample, drawn from one year, and most of these ingredients are not fermentation-derived proteins. Two are extensions of use, which may move faster or slower than first authorisations for reasons that have nothing to do with the Commission’s workload. The 320-day and 286-day outliers both involve opinions adopted on 25 June 2025, which may reflect a single scheduling backlog rather than a structural pattern.
That case is fair. The narrower claim survives it: the interval is not fixed, it is not short, and the statutory number that appears to govern it does not govern the step applicants care about. Those three things are true of all nine.
A second objection is that the Commission is doing what the regulation requires, so there is nothing to report. That is also true, and it is the point. The problem is not non-compliance — it is that compliance with Article 12 is compatible with an applicant waiting most of a year, and nothing in the public record tells that applicant which of the two outcomes they are in.
What we could not establish
- Submission dates to the Standing Committee. The step Article 12 actually times is not published in a form we could retrieve for each of the nine. We can measure publication to adoption; we cannot measure publication to submission, which is the only interval the seven-month deadline binds. No claim of a missed deadline is made anywhere above.
- Official Journal publication dates. We recorded each regulation’s adoption date from its signature block and its entry-into-force rule from its final article, but did not independently establish the OJ publication date for each, so we cannot state the exact date each entered into force.
- EFSA publication dates are proxied. We used the EFSA Journal article registration date. For the Rhizomucor pusillus opinion this gives 1 December 2025, consistent with AgFunderNews’s contemporaneous report of a December publication. We did not obtain a Commission or EFSA statement confirming the date used to start the Article 12 clock in any individual case.
- Completeness of the nine. The list is taken from the Commission’s own Union list index as it stood when we read it. If an amendment adopted in 2026 had not yet been added to that index, it is not in our table.
- What happened on 17 June 2026. The comitology register records an adopted attachment on that date and the trade press reported approval. We could not establish the formal status of that step well enough to name it precisely, beyond that it preceded the instrument by sixteen days.
What to watch
Three testable things, each of which will settle a claim above within about a year.
Whether the 2027 cohort shows the same spread. If publication-to-adoption again ranges across roughly five to eleven months, the variance is structural. If it tightens toward the seven-month figure, 2026 was a backlog year.
Whether the EU Biotech Act changes the pre-submission stage. The Protein Brewery’s specific ask — a mandate for EFSA to discuss study designs with applicants before submission — targets the Article 11(4) extensions that turned nine months into 1,726 days. Our earlier reading of that instrument found that a fermentation process can enter the sandbox while the food it makes cannot. Whether scientific pre-advice arrives is the test of whether the Act touches the binding constraint.
Whether the next mycelium applicant moves faster. The Protein Brewery’s argument is that the six years were the cost of being first, and that the precedent is now set. There are enough biomass-fermentation dossiers behind it — and our survey of what the 1997 date decides identified which of them need authorisation at all — that the claim will be measurable. If the second Rhizomucor-class authorisation still takes four years from referral to opinion, “being first” was not the explanation.