Saccharin: The Oldest Artificial Sweetener and Its Cancer Scare

Saccharin was the first artificial sweetener anyone ever tasted, and it was found the way a surprising number of chemistry discoveries happen: by accident, and by a researcher who forgot to wash his hands. In 1879 a chemist named Constantin Fahlberg, working in Ira Remsen’s lab at Johns Hopkins, noticed that his dinner roll tasted sweet. He traced the sweetness back to a coal-tar compound he had been synthesizing earlier that day. That compound was saccharin, and more than a century later it is still in the pink packets on diner tables.

It is also the sweetener with the most dramatic regulatory story of any on the market. Saccharin was nearly banned outright, then saved by an act of Congress, then made to wear a cancer warning label for two decades, and then cleared and delisted once the science caught up. That arc is worth telling in full, because it is the clearest example we have of how a real laboratory finding can be both true and irrelevant to humans at the same time.

What Saccharin Is

Saccharin is benzoic sulfimide, a small synthetic molecule with no nutritional relatives in food. It is roughly 300 to 400 times sweeter than table sugar, so the amount needed to sweeten a drink is tiny. Pure saccharin is not very water-soluble, so the form used in food is almost always a salt, usually sodium saccharin or calcium saccharin, which dissolves easily.

The body does not metabolize it. Saccharin is absorbed and then excreted unchanged in the urine, contributing zero calories and zero effect on blood glucose. That metabolic inertness is exactly why it was attractive to people with diabetes and to dieters long before low-calorie food was a mainstream category.

Its weakness is taste. At low doses saccharin reads as clean and intensely sweet, but as the concentration climbs it develops a bitter, metallic, slightly tinny aftertaste that many people can detect. This single flaw shapes almost everything about how saccharin is actually used, which we will come back to.

The Full Regulatory Timeline

The short version of the saccharin scare appears on our artificial sweeteners overview , but the standalone story has more moving parts than most people realize.

1958 to 1977: GRAS, then doubt. When the US food additive law was overhauled in 1958, saccharin was already in wide use and was placed on the Generally Recognized as Safe (GRAS) list. Through the 1960s and early 1970s, a series of animal studies raised questions, and the FDA moved saccharin off the unconditional GRAS list and put it under interim study.

1977: the proposed ban. A Canadian study reported that rats fed high doses of saccharin developed bladder tumors. Under the Delaney Clause, a provision that required the FDA to ban any food additive shown to cause cancer in animals at any dose, the FDA proposed to ban saccharin entirely. At the time it was the only artificial sweetener left on the US market, because cyclamate had been banned in 1969. The public reaction was loud and immediate.

1977: Congress steps in. Rather than lose the only sugar substitute available, Congress passed the Saccharin Study and Labeling Act, which placed a moratorium on the ban and ordered further study. In place of a ban, the law required a warning on every saccharin product and on store shelves. The label read that use of this product may be hazardous to your health because saccharin had been determined to cause cancer in laboratory animals.

1977 to 2000: the warning-label era. For more than twenty years, diet sodas, tabletop packets, and other saccharin products carried that warning while researchers worked out what the rat tumors actually meant. The moratorium was renewed repeatedly. During this stretch the mechanism behind the rat tumors was identified, and it turned out to undercut the entire premise of the ban.

2000: the NTP delisting. The US National Toxicology Program reviewed the accumulated evidence and removed saccharin from its Report on Carcinogens, where it had been listed as reasonably anticipated to be a human carcinogen. The NTP concluded that the rat bladder mechanism did not apply to humans and that human studies showed no clear cancer link (National Toxicology Program, 2000, Report on Carcinogens 9th Edition).

2000: the label goes away. In December 2000, Congress repealed the warning-label requirement through Public Law 106-554, ending the labeling era that the 1977 law had created. Products manufactured after that point no longer carry the warning. The FDA continues to permit saccharin as a sweetener and lists it under 21 CFR 180.37.

The Rat-Bladder Mechanism, and Why It Does Not Apply to Humans

This is the part that justifies the whole reversal, and it is genuinely interesting chemistry.

The tumors appeared in male rats fed saccharin at doses equivalent to a person eating hundreds of grams a day for life. At those doses, sodium saccharin changed the chemistry of the rat’s urine. Male rats produce urine that is unusually high in protein and high in pH, and they carry a specific protein called alpha-2u-globulin. In that particular environment, the high sodium load and the high concentration of saccharin caused a tiny calcium phosphate-containing precipitate, with possible silicate involvement, to form in the urine.

Here is the analogy. Think of those crystals as fine grit in a shoe. A single grain does nothing, but constant grit rubbing against the same patch of skin will eventually inflame it, and the skin responds by growing thicker and faster. In the rat bladder, the crystals physically scraped the lining (the urothelium), which responded with ongoing cell proliferation. Cells that are forced to divide over and over for a lifetime have more chances to pick up the mutations that lead to cancer. The tumor was the end result of mechanical irritation, not of saccharin acting as a direct carcinogen on DNA.

Humans do not produce urine with that protein profile or that chemistry. We lack the high alpha-2u-globulin levels, our urine pH and protein content are different, and the crystals simply do not form. The dose-and-chemistry combination that drove the rat tumors has no human equivalent. Researchers who worked out this mechanism argued for years that the rat data should not be read across to people (Ellwein and Cohen, 1990, Critical Reviews in Toxicology), and the broader body of work on rat-specific bladder responses to high-dose sodium salts supported that conclusion (Cohen et al., 1998, Cancer Research).

Large human studies backed it up. Bladder-cancer studies in people, including heavy long-term saccharin users such as people with diabetes, did not show the consistent excess that the rat data would have predicted if the mechanism translated. This is why the International Agency for Research on Cancer moved saccharin into Group 3, not classifiable as to its carcinogenicity to humans, the same neutral category for substances where the evidence does not support a human cancer hazard.

The honest framing today is straightforward. Saccharin caused tumors in one sex of one species through a mechanism that depends on that animal’s unique urinary chemistry, and that mechanism is not present in humans. The current consensus is that saccharin is not a human carcinogen at the levels people consume.

E954 and Global Approvals

Saccharin is approved across the major regulatory systems, not just in the United States.

In the European Union, saccharin and its sodium, potassium, and calcium salts carry the additive number E954 and are permitted in a defined list of food categories under EU additive rules. The Joint FAO/WHO Expert Committee on Food Additives (JECFA) has long held saccharin to be acceptable within a group ADI of 0 to 5 mg/kg per day. EFSA re-evaluated the same toxicology in November 2024 and concluded saccharin does not damage DNA and is unlikely to be linked to cancer in humans, raising its own ADI to 9 mg/kg per day (EFSA FAF Panel, 2024, EFSA Journal). It is also cleared in Canada, where it was once restricted to pharmacies and is now permitted as a tabletop sweetener, and in dozens of other countries through the JECFA framework.

So the substance that the United States nearly banned in 1977 is now one of the most widely cleared sweeteners in the world, sold under names like Sweet’N Low, Sweet Twin, and Necta Sweet.

How the ADI Was Set

An acceptable daily intake (ADI) is the amount a person can consume every day for life with no expected harm, and it is built with a large safety cushion. The method is the same one used across the other sweeteners on this site.

Toxicologists start from the highest dose in animal studies that produced no adverse effect, the no-observed-adverse-effect level. They then divide by a safety factor, usually 100, to account for differences between animals and humans and differences among people. JECFA sets the saccharin ADI at 0 to 5 mg per kg of body weight per day, expressed as saccharin. The FDA has historically cited a higher figure of 15 mg/kg/day, which is the number you will see in the comparison table on our acesulfame potassium page.

Put the JECFA number in real terms. A 70 kg (154 lb) adult has a daily limit of about 350 mg of saccharin. A single packet of tabletop sweetener contains on the order of 36 mg of saccharin, with the rest of the packet being a bulking agent like dextrose. That adult would need around 9 to 10 packets every single day to reach the JECFA ADI, and the ADI itself sits well below any level where effects were seen. Normal use does not come close.

Taste-Masking Blends

Remember saccharin’s one real flaw, the metallic bitter aftertaste. The food industry solved it not by fixing saccharin but by mixing it with partners that cover its weak spots.

The classic example is the saccharin-and-cyclamate blend, which dominated diet sodas for years in countries where cyclamate is still legal. Saccharin and cyclamate each have an off-taste, but the two off-tastes partially cancel each other, and the blend tastes cleaner and more sugar-like than either one alone. A roughly 10-to-1 cyclamate-to-saccharin ratio became a standard formula.

More commonly today, saccharin is blended with aspartame or with acesulfame potassium . Two useful things happen in these mixes. First, the sweeteners are synergistic, meaning the combined sweetness is greater than the sum of the parts, so manufacturers can use less of each. Second, blending dilutes any single sweetener’s characteristic aftertaste below the threshold where most people notice it. The result tastes rounder and closer to sugar. This is why the saccharin in a finished diet drink is often less detectable than the saccharin in a packet you stir into black coffee.

What the Body Does With It

Saccharin is one of the few additives that genuinely passes through unchanged. It is absorbed in the gut, circulates briefly, and is filtered out by the kidneys into the urine without being broken down. It is not stored and it does not accumulate.

The newer and still-open question is not cancer but the gut. A controlled human trial published in 2022 reported that saccharin, among other non-nutritive sweeteners, was associated with changes in the gut microbiome and in glucose response in some people, though the effects were small and the doses were at the high end of typical intake (Suez et al., 2022). We cover that research in depth on the artificial sweeteners overview and on our gut microbiome basics page. A changed microbiome is not the same as a damaged one, and the clinical meaning at real-world doses is not settled. That uncertainty is about metabolism, not the cancer fear that defined saccharin’s history.

The Bottom Line

Saccharin earns a safe verdict at food levels. The bladder-tumor finding that nearly ended it was real, but it ran through a rat-specific urinary chemistry that humans do not share, and the human evidence never confirmed the risk. The NTP delisted it in 2000, Congress repealed the warning label in December 2000, and major agencies worldwide permit it within a generous ADI. The most practical thing to know is mundane: at high doses it tastes metallic, which is why you usually meet it blended with another sweetener rather than on its own.