Does Aspartame in Diet Soda Cause Cancer?
Ordinary diet-soda drinking hasn't been shown to cause cancer. In 2023, IARC called aspartame "possibly carcinogenic." The same year, JECFA, another WHO-linked committee, read much of the same evidence with one difference: it asked what happens at the amounts people actually consume, and left its daily intake limit where it was. IARC's label flags a possible hazard, not a proven risk at ordinary intake. The signals behind it are limited. Some observational studies report modest cancer associations, and the animal and mechanistic evidence has big gaps. What follows is what the evidence supports, where it's shaky, and who has reason to cut back.
Group 2B, where aspartame landed in 2023, is an evidence label, not a danger measure. IARC asks a narrow question: could a substance cause cancer under some conditions? It doesn't estimate the risk from a can of soda. To classify aspartame, it drew on three streams of evidence, and judged every one limited.
What IARC actually weighed
Human studies gave the most weight to liver cancer, where artificially sweetened beverages, not always aspartame itself, were linked to higher risk in some analyses. The animal evidence came mostly from lifetime rodent studies at the Ramazzini Institute that reported more lymphomas and leukemias. The mechanistic evidence centered on oxidative stress, a weak bridge to human cancer unless it's consistent, dose-related, and tied to a plausible pathway.
Each stream has a hole. The liver studies often couldn't separate aspartame from the other sweeteners in the same drinks. A 2026 pooled analysis of 11 prospective cohorts found no independent link between artificially sweetened beverages and liver cancer after adjustment for diabetes and BMI; sugar-sweetened beverages were linked to liver-cancer subtypes. That keeps confounding high on the list. The Ramazzini studies were later rated not reliable by quality reviewers and criticized by regulators, with concerns about chronic infections, high background tumor rates, end-of-life pathology problems, and dosing patterns far from ordinary intake. Standard genotoxicity tests came back negative.
The human studies don't agree with each other
A causal signal at ordinary intake would be more persuasive if large cohorts pointed the same way and showed more cancer with more intake. They don't. France's NutriNet-Santé cohort linked higher aspartame intake to about 15% more overall cancer, plus more breast and obesity-related cancer. But the Nurses' Health Studies, following women for nearly 30 years, found no link between aspartame and breast cancer. A 2026 Cancer Prevention Study-II analysis of about 100,000 U.S. adults found more cancer among people reporting any aspartame use, but no rise with higher intake. Other large cohorts and reviews cut the other way: a big NIH-AARP cohort found no link to blood cancers or brain tumors, an umbrella review of ten meta-analyses found no overall cancer association for artificial sweeteners as a group, and a 2025 review of 90 studies found no consistent tie between any nonsugar sweetener and any cancer.
The 2012 lymphoma and leukemia study shows why the signal is hard to read. It found more non-Hodgkin lymphoma and multiple myeloma among men who drank at least one diet soda a day, but not among women, and it flagged the same kind of signal for regular sugar-sweetened soda. Later cohorts didn't back up a clear lymphoma or myeloma link. Some meta-analyses still flag leukemia with artificially sweetened beverages, but they don't isolate aspartame well and rely on few datasets. When a finding shows up in one study, in one sex, and fades when others look, chance and confounding move up the list. Confounding is the central problem here. Heavy diet-soda drinkers differ from everyone else, often because weight, blood sugar, or a health scare pushed them toward it in the first place. A signal may be tracking the person, not the can.
Four agencies, one pile of evidence, different verdicts
The same evidence went to several expert bodies at nearly the same time, and they split. Not over the data, but over what a hazard label should mean for you.
| Authority | What it concluded | What that means for a consumer |
|---|---|---|
| IARC | Classified aspartame as possibly carcinogenic (Group 2B) on limited evidence | A cancer-hazard flag; it does not estimate risk at ordinary intake |
| JECFA | Found the cancer evidence not convincing and kept the 0 to 40 mg/kg daily limit | Considers intake within that limit safe |
| FDA | Disagreed with IARC's reading and kept its 50 mg/kg daily limit | No safety concern under approved use |
| Health Canada | Kept its 40 mg/kg daily limit and took no further action after the 2023 review | No safety concern at current use levels |
The bodies that weigh actual intake landed together. IARC asks whether something could cause cancer under some conditions; JECFA, the FDA, and Health Canada ask whether approved use is safe at expected doses. That's why the scary category and the unchanged limits aren't a contradiction.
How much would you have to drink?
A lot. WHO's own example: a 70-kg adult would need more than about 9 to 14 cans of diet soft drink a day to cross the JECFA limit, assuming 200 to 300 mg per can and no other source. Many U.S. diet drinks contain less than that per can, but the exact number depends on the product and any other aspartame sources. Real intake usually sits far below the limit. The animal studies that raised concern don't map neatly to ordinary use: they used lifelong rodent exposure, sometimes before birth, and were criticized on pathology and study-quality grounds.
Who should think twice
A few people should. Anyone with phenylketonuria has to avoid aspartame, because they can't safely process the phenylalanine it breaks down into. That's what the label's warning is about, not cancer. Heavy daily drinkers have a reason to cut back too, though not the one they'd expect: observational studies tie heavy intake of artificially sweetened beverages to cardiovascular events, type 2 diabetes, and earlier death more consistently than to cancer. Those links carry the same reverse-causation problem, but they make a better case for moderation than the cancer signal does. WHO's advice against using nonsugar sweeteners for weight control is a nutrition recommendation, not a cancer verdict. Diet soda can replace sugar. It won't make the rest of your diet healthy.
For a healthy adult, none of this makes the IARC label proof that diet soda causes cancer. What would change that verdict is specific: studies that measure aspartame itself rather than a mix of sweetened drinks, show cancer risk rising with intake, report the absolute numbers plainly, and fit a mechanism that works in humans. Until then, the warning is a reason to keep an eye on the evidence, not a reason to put down the can.