E150d is why some sodas carry a cancer warning in California

E150d is sulfite ammonia caramel, Class IV caramel color, made by heating carbohydrates with ammonium and sulfite compounds. It is the deep brown color in most colas, in many soy sauces, and in balsamic vinegar substitutes. The cancer warning you may have seen on a soda can in California is not really about caramel color at all. It is about a specific byproduct of making it, 4-methylimidazole, and the entire case against it comes from one rodent study. No human trial has tested it.^[1] That distinction matters more than the warning label lets on, and the numbers behind it vary by brand far more than most shoppers would guess.

The mouse study behind the warning

The cancer evidence for 4-methylimidazole (4-MEI) comes from a single source: a two year oral bioassay run by the US National Toxicology Program. NTP fed the chemical to mice and rats and found increased lung tumors in mice, with no such finding in rats.^[2] On the strength of that finding, the International Agency for Research on Cancer classified 4-MEI as Group 2B, possibly carcinogenic to humans, in 2011.^[2]

Because NTP’s bioassay assigned mice to controlled doses rather than just observing them, it can support a causal read on high dose 4-MEI exposure and lung tumor formation, but specifically in mice.^[3] That is a real strength of the study design. It is also the limit of what the study can tell you, because the finding has not been replicated in a species closer to humans, let alone in a human trial.

The source of the warning: a rodent bioassay that found lung tumors in mice at high controlled doses is the entire evidentiary basis for the Proposition 65 listing. No equivalent human study exists.

What the human evidence is, and is not

This is the part most coverage skips. A 2018 review of the full caramel color safety record lists everything that has been tested: toxicokinetics, genotoxicity, subchronic toxicity, carcinogenicity, and reproductive and developmental toxicity studies, all conducted in animals or in cell culture, plus separate data on use levels used to estimate how much people eat.^[4] No human clinical trial appears anywhere in that evidence base.^[4]

What exists instead is exposure measurement, which is a different kind of evidence entirely:

• Product testing. Labs measure 4-MEI concentrations directly in food and drink samples.
• Dietary surveys. Researchers combine those concentrations with national consumption data to estimate how much a population takes in.
• Biomonitoring and intake modeling. These tell you what people are exposed to. They do not tell you what that exposure does to them.

The FDA’s own dietary exposure assessment is a good example. It analyzed more than 700 US food and beverage samples collected in 2013 and 2015 using LC-MS/MS, then combined those measurements with two years of national consumption survey data to estimate population intake of 4-MEI from foods with added caramel color, separately from foods where 4-MEI forms naturally during cooking.^[5] That is a careful, well-designed study. It is also, by design, observational: it measures what is already in the food supply and what people already eat, rather than randomizing anyone to different exposure levels and watching for an outcome.^[4]

So the honest answer to “does E150d cause cancer in people” is that nobody has run the study that would answer it. What exists is a mouse finding and a set of careful measurements of how much of the chemical people are eating, which is a meaningfully smaller claim.

How the ammonia process creates 4-MEI

Sulfite ammonia caramel forms through a Maillard-type reaction between a sugar and ammonia under heat, the same basic chemistry that browns toast and sears a steak, just steered toward a stable brown pigment instead of a crust.^[6] 4-MEI is a documented byproduct of that specific, ammonia-driven reaction.^[6]

Not every caramel color goes through that step. 4-MEI formation is tied specifically to caramel colors made with ammonium compounds, which is Class III, ammonia caramel and Class IV, sulfite ammonia caramel.^[5] Class I, plain caramel, and Class II, caustic sulfite caramel, are made without ammonium compounds and sit outside that formation pathway entirely.^[5]

The numbers behind the ADI and the NSRL

Three different numbers get thrown around in coverage of this additive, and they are not measuring the same thing.

Three regulatory numbers, three different questions

Each figure answers a different question about safety margin.

JECFA’s acceptable daily intake for sulfite ammonia caramel is 0 to 200 milligrams per kilogram of body weight per day.^[7] EFSA separately evaluated 4-MEI itself and concluded it is not genotoxic, setting a no-observed-adverse-effect level of 80 milligrams per kilogram of body weight per day and expressing no safety concern about its presence in caramel color.^[7]

California’s Office of Environmental Health Hazard Assessment, under CalEPA, set its No Significant Risk Level of 29 micrograms a day using a non-threshold approach built directly on the NTP rodent tumor data, and it is that figure, separate from the JECFA or EFSA numbers, that triggers the warning label.^[8]

29 micrograms of 4-MEI a day that triggers California's Prop 65 warning label PMID 27822321

What testing has found on store shelves

Independent testing of 12 sodas bought in California and New York found 4-MEI concentrations that varied enormously, from about 10 micrograms per liter in Coca-Cola to more than 900 micrograms per liter in Malta Goya, varying by both brand and the state where it was purchased.^[1]

4-MEI in two tested sodas

From independent testing of beverages purchased in the US.

The gap between those two numbers is the real story here. Routine consumption of lower 4-MEI sodas like Coca-Cola stays well under the Proposition 65 threshold. Routine consumption of higher 4-MEI brands like Malta Goya can clear that threshold within ordinary daily servings, even without heavy consumption.^[1]

Where E150d fits against the other caramel colors

Industry reformulation after the Proposition 65 listing has moved the picture. The FDA’s updated exposure assessment shows that reformulated caramel color suppliers cut the agency’s estimated general population 4-MEI exposure from added caramel sources from about 0.043 milligrams a day in its 2018 assessment to about 0.016 milligrams a day in the most recent update.^[9]

For a shopper who wants to minimize 4-MEI exposure without giving up color, plain caramel or caustic sulfite caramel is the more conservative choice when an ingredient list specifies the type, since neither goes through the ammonia step that produces the byproduct in the first place. Most US labels just say “caramel color” without the class letter, which is itself a real limit on how precisely anyone can shop around this.

That gap between what the label says and what it would take to avoid an additive in practice is the same problem that runs through the two food additives with genuine trial evidence behind the concern, and it is worth reading alongside this one for contrast: most additive scares, including this one, rest on preclinical signals rather than trial data, in the same way maltodextrin’s gut-damage case rests on mice, not people.

What to watch

Regulatory action on 4-MEI has so far concentrated on California’s Proposition 65 listing. The same products sold in markets without that law, such as New York, carry no equivalent warning, and researchers who tested the gap have themselves called for federal regulation that does not yet exist.^[1]

Two things will keep moving this story forward:

• Continued reformulation. The FDA’s own exposure numbers have already dropped by more than half since 2018 as suppliers adjust. That trend is likely to continue.
• Periodic independent testing. Most of what the public knows about brand-level 4-MEI content comes from researchers buying products off shelves and testing them directly. Routine label disclosure does not provide this.

If you want to check which caramel color class a specific product uses rather than guessing from “caramel color” on the label, scanning the ingredient panel directly with the EatMatrix app is a faster way to see what you are working with than trying to track down the manufacturer’s sourcing.

Bottom line: the Proposition 65 warning on your soda traces back to one mouse-only cancer finding. No human trial backs it. The variable worth caring about is which brand you are drinking, since measured 4-MEI levels differ by close to a hundredfold between the lowest and highest tested sodas.

This article summarises publicly available regulatory assessments and peer-reviewed research about food ingredients. It is general information and is not intended to diagnose, treat, cure, or prevent any disease, and it is not personalised dietary or medical advice. Regulatory positions change; the EatMatrix app reflects the current assessments in our database. If you have an allergy, a medical condition, or are pregnant, talk to a qualified clinician or dietitian before changing what you eat.