The FDA is banning Red No. 3 — while saying the human cancer risk is unlikely

The Food and Drug Administration is removing the synthetic dye FD&C Red No. 3 from the American food supply, but not because the agency has decided it harms people: FDA's own scientists concluded the cancer risk to humans is unlikely [s1]. The dye is being delisted because a 1958 statute, the Delaney Clause, bars any color additive found to cause cancer in humans or animals — and Red No. 3 causes thyroid tumors in male rats [s1].

That gap between the finding and the reason is the whole story here, and it is easy to lose in the headlines.

What the order does, and when

On 16 January 2025, FDA granted a color additive petition submitted by the Center for Science in the Public Interest and others, repealing the regulations that permit FD&C Red No. 3 in foods, including dietary supplements, and in ingested drugs [s1]. The order takes effect on 15 January 2027 for food and 18 January 2028 for ingested drugs, giving manufacturers a window to reformulate [s1].

Those dates were briefly in doubt. Objections filed in February 2025 triggered an automatic administrative stay under the Federal Food, Drug, and Cosmetic Act [s2]. On 5 August 2026, FDA lifted that stay, ruling that the objections raised no issue of material fact justifying a hearing, and confirmed the 2027 and 2028 effective dates [s2]. The delisting is now settled.

Why an additive FDA calls low-risk is being banned anyway

The Delaney Clause — the part of the law that applies to ingested color additives — is deliberately absolute. Courts have read it that way. In Public Citizen v. Young (1987), the D.C. Circuit called it "extraordinarily rigid," rejecting FDA's argument that an additive posing only a trivial cancer risk could be spared [s1]. If a substance induces cancer in humans or animals, it must go, regardless of the probability of harm or the dose involved [s1].

Red No. 3 meets that trigger on a narrow technicality. Some studies show it can cause thyroid tumors in male rats, apparently through a rat-specific hormonal pathway: the dye raises circulating thyroid-stimulating hormone, which drives follicular growth that can become tumors [s1]. The dye is likely not genotoxic — it does not react directly with DNA — and the tumors were not seen in female rats, in mice of either sex, or in gerbils or dogs [s1].

FDA is blunt about how little this says about people. The agency, along with the joint FAO/WHO expert committee (JECFA), the European Food Safety Authority and Food Standards Australia New Zealand, has concluded that the male-rat thyroid tumors "are of limited relevance to humans" [s1]. FDA's own Cancer Assessment Committee judged the risk of such tumors in humans at real-world exposure "unlikely" [s1]. The numbers underline it: the agency estimated the highest human exposure from all approved uses at 0.25 mg per kilogram of bodyweight per day, against a no-observed-adverse-effect level of 35.8 mg/kg/day in rats — a 210-fold margin of exposure [s1].

A 35-year loose end

None of this is new science, which is part of why the decision landed the way it did. FD&C Red No. 3 was permanently listed for food and ingested drugs back in 1969 [s1]. In 1990, on the strength of the same rat thyroid data, FDA declined to keep the dye's provisional listing for cosmetics and externally applied drugs — but left the food and ingested-drug listing untouched [s1]. The 2025 order finally closes that inconsistency, three and a half decades later, after the petition forced the question.

What it means for a reader

For anyone worried that Red No. 3 in candy or maraschino cherries has been quietly giving people cancer, the agency's assessment is the opposite of alarming: at the levels people actually consume, it does not consider that risk established [s1]. What changed is legal, not biological. The counterpoint, argued by the petitioners, is that a dye adds color and nothing else, so even a contested animal finding is reason enough to drop it — the Delaney Clause simply encodes that precaution as law.

The episode is also a clean illustration of how the Delaney Clause works: it does not weigh dose against risk, which is why a substance three regulators consider low-concern for humans is nonetheless barred. That same all-or-nothing logic sits behind long-running debates over other food chemicals, from preservatives studied in large cancer-cohort analyses to the broader scrutiny of additives that comes with a diet heavy in ultra-processed food. It is a different question from whether a food is healthy, which is the subject of FDA's separate move to redefine how processed foods are regulated.

What to watch

Whether the confirmed 2027 and 2028 deadlines hold without further legal challenge, how quickly manufacturers switch to other reds, and whether the Red No. 3 precedent accelerates petitions against other synthetic dyes still on the market.