Does Diet Pop Cause Cancer? What the Science Says

No reliable evidence shows that drinking diet soda at normal levels causes cancer in humans. Two large systematic reviews covering dozens of studies and multiple sweetener types have found no consistent link between artificial sweetener intake and any form of cancer. That said, the question is more layered than a flat “no” suggests. In 2023, a World Health Organization body classified aspartame as “possibly carcinogenic,” a handful of cohort studies have reported small elevated risks, and animal experiments at high doses have produced tumors. Sorting through these seemingly contradictory findings requires understanding what each piece of evidence actually shows and where its limits lie.

What the Largest Reviews Conclude

The strongest tool scientists have for settling a question like this is to pool results from many studies and look for patterns. A 2025 systematic review examined 90 epidemiological studies of sweeteners including aspartame, acesulfame potassium, saccharin, cyclamate, and sucralose. It looked at 17 specific cancer types and found no consistent association between any individual sweetener or sweeteners as a group and any cancer. It also found no evidence of a dose-response relationship, meaning heavier users did not show higher risk in any reliable pattern.¹

An umbrella meta-analysis published the same year pooled data from earlier meta-analyses and arrived at a summary risk ratio of 0.99 for overall cancer, which is statistically indistinguishable from no effect at all. Subgroup analyses by cancer type were also null, with one exception: gynecological cancers actually showed a slightly reduced risk among sweetener users, though the clinical significance of that finding is unclear.²

When two independent, large-scale reviews using different methodologies reach essentially the same conclusion, that conclusion carries real weight. Individual studies can point in different directions for all kinds of reasons, but the pooled picture is clear: at the population level, drinking diet soda does not appear to raise cancer risk.

The IARC “Possibly Carcinogenic” Label

In July 2023, the International Agency for Research on Cancer classified aspartame as a Group 2B substance, meaning “possibly carcinogenic to humans.” That phrase sounds alarming until you know what Group 2B actually contains. It is a large category that includes things like aloe vera extract, pickled vegetables, and certain textile dyes. Being placed in Group 2B means that the evidence of cancer in humans is limited and the evidence from animal studies is less than sufficient. It does not mean the substance has been shown to cause cancer.

At the same time IARC made its classification, a separate WHO expert committee called JECFA reaffirmed aspartame’s safety and maintained the existing acceptable daily intake. JECFA concluded that the epidemiological evidence for a link between aspartame and cancer was “not convincing” and that there were no consistent associations across the research.³ The two bodies were asking different questions: IARC asks whether something could theoretically be hazardous under some circumstances, while JECFA asks whether it is dangerous at the levels people actually consume it. The answer to the first question was a cautious “maybe.” The answer to the second was “no.”

IARC pointed to three studies on liver cancer that it rated as high quality, but acknowledged that chance, bias, and confounding could not be ruled out as explanations for the positive findings in those studies.³ That is a significant caveat. When you cannot rule out confounding, you cannot draw a causal conclusion.

Studies That Did Find Associations

Not every study comes back clean, and it is worth understanding the ones that did find something. A large French cohort study of over 100,000 adults tracked artificial sweetener consumption through detailed dietary records over roughly eight years. It reported that people who consumed above-median amounts of artificial sweeteners had about a 13% higher risk of cancer overall compared to non-consumers. For aspartame specifically, the increase was about 15%, and there were elevated signals for breast cancer and obesity-related cancers.⁴

A 13 to 15 percent increase in relative risk is quite small in epidemiological terms. For context, smoking raises lung cancer risk by roughly 1,500 to 3,000 percent. When an observed effect is this modest, the likelihood that it is actually caused by confounding factors rather than the sweetener itself becomes very real. Researchers acknowledged that the association may have been shaped by characteristics of the consumers themselves, such as higher rates of obesity or metabolic disorders, rather than the sweeteners directly.

Meanwhile, other major studies found no link. One of the largest ever conducted on aspartame and cancer used data from the NIH-AARP Diet and Health Study and found that even at high intakes, aspartame was not associated with blood cancers or brain tumors in men or women.⁵ The fact that the biggest studies tend to show null results while some smaller or more recent cohorts show small positive signals is part of why the systematic reviews land where they do: the overall pattern does not support a real effect.

Why Observational Studies of Diet Soda Are Especially Tricky

Diet soda research has a well-known methodological headache called reverse causality. People who are already overweight or worried about their health are more likely to switch from regular to diet soda. That means diet soda drinkers, as a group, tend to have higher rates of obesity and metabolic conditions from the start. When researchers then observe that diet soda drinkers have somewhat higher rates of certain diseases, they may be measuring the consequences of the underlying health problems, not the effects of the drink.

This pattern has been demonstrated repeatedly. A review of diet beverage research noted that reverse causality is a “particular problem” in these studies, since individuals at higher risk of weight gain are the ones who tend to choose artificially sweetened beverages in the first place.⁶ Another study found that the apparent link between diet soda and diabetes was largely explained by BMI at the time people first reported their diets.⁷ The same dynamic applies to cancer: obesity is itself a risk factor for multiple cancer types, and if obese individuals disproportionately drink diet soda, any link between diet soda and those cancers could be a statistical shadow of the obesity link.

Research has confirmed that low-calorie sweetener use is closely tied to a person’s prior intent to lose weight, which strongly suggests that the health profile of the average diet soda drinker is not the same as the average non-drinker.⁸ Even the most careful statistical adjustments cannot fully eliminate this kind of confounding.

What Animal Studies Show and Why the Translation Is Uncertain

Much of the original concern about artificial sweeteners and cancer came from studies in rodents. The Ramazzini Institute in Italy published a series of experiments exposing rats to aspartame starting from prenatal life. At high doses (2,000 parts per million in feed), the researchers reported significant increases in lymphomas, leukemias, and mammary tumors.⁹ These findings were dramatic and received a lot of media attention.

However, regulatory agencies in the United States and Europe reviewed the Ramazzini data and raised serious questions about the methods, including high rates of respiratory infections in the animals that can cause the same lymphomas seen in the results. The doses used were also far beyond what any human would consume. For aspartame, the acceptable daily intake set by regulators is equivalent to about 14 to 18 cans of diet soda per day for an average adult, and even the highest consumers in population surveys fall well below that threshold.¹⁰

The broader lesson from animal studies is that many substances, including some vitamins, produce tumors in rodents at extreme doses through mechanisms that do not operate in humans. Saccharin provides the most famous example.

The Saccharin Story

Saccharin was arguably the original artificial-sweetener cancer scare. Starting in the 1970s, studies showed that male rats fed very high doses of saccharin developed bladder tumors at rates up to 30 percent. Canada banned it, and the United States required cancer warning labels on saccharin products starting in 1981.¹¹

It took years of follow-up research to understand what was actually happening. Rats have unusually high urine concentrations compared to humans. At massive saccharin doses, calcium phosphate crystals formed in rat bladders and physically damaged the bladder lining, leading to a cycle of tissue repair and, eventually, tumors. The same mechanism worked with vitamin C at comparable doses. The key finding was that this crystal-formation pathway does not operate in humans because our urine chemistry is different.¹² The National Institute for Environmental Health Sciences eventually removed saccharin from its list of potential carcinogens, and the warning labels came off in 2000.

The saccharin episode is instructive because it illustrates how a mechanistically real finding in animals can be completely irrelevant to human risk when the biology differs. It also shows that regulatory agencies can and do reverse course when the science warrants it.

Other Sweeteners Under the Microscope

Diet sodas contain a variety of sweeteners depending on the brand, and the safety data vary somewhat from one to the next.

Acesulfame potassium (ace-K), commonly paired with aspartame or sucralose in diet beverages, has been tested in genetically modified mice designed to be highly sensitive to carcinogens. A nine-month National Toxicology Program study found no evidence of carcinogenic activity at any dose tested.¹³ A separate systematic evaluation of ace-K’s potential mechanisms of carcinogenicity concluded that it is unlikely to produce a carcinogenic response, based on a lack of activity across all key cancer-related pathways.¹⁴

Sucralose has generated more recent debate. A 2023 lab study reported that sucralose-6-acetate, a compound formed when the body metabolizes sucralose, was genotoxic in screening assays, meaning it appeared capable of damaging DNA.¹⁵ That study made headlines and genuinely rattled people. But a subsequent battery of tests conducted under international regulatory guidelines found that commercial-grade sucralose showed no genotoxic potential in bacterial mutation assays, cell-based tests, or a transgenic rodent assay designed to detect mutations in living animals.¹⁶ The conflict between the two sets of results may come down to the difference between testing a specific metabolite in a lab dish and testing the real-world substance in a whole organism, where metabolism and detoxification systems are operating. The science here is genuinely unsettled, but the more comprehensive follow-up study is reassuring.

Diet Soda Versus Regular Soda and Cancer

An interesting comparison arises when you look at cancer risk from sugar-sweetened beverages alongside artificially sweetened ones. A large Scandinavian cohort study found that frequent consumption of sugar-sweetened soft drinks was associated with a rising trend in obesity-related cancers, while artificially sweetened soft drinks showed no association at all. People who drank more than one sugar-sweetened soda per day had about an 18% higher risk of obesity-related cancers compared to those who rarely drank them, whereas daily artificially sweetened soda drinkers had no elevated risk.¹⁷

This does not make diet soda a health food, but it provides a useful reality check. If your concern is cancer, the sugar in regular soda is a better-established risk factor than any artificial sweetener. Excess sugar intake drives obesity, and obesity is convincingly linked to at least 13 types of cancer. By that logic, replacing regular soda with diet soda is likely a net positive in terms of cancer risk, not a new source of danger.

The Gut Microbiome Question

One area that has attracted a lot of research attention is whether artificial sweeteners alter the gut microbiome in ways that could eventually promote disease, including cancer. Animal studies have reported decreases in beneficial bacteria and increases in harmful strains, along with disruptions to short-chain fatty acid production and gut hormone signaling.¹⁸ These findings sound concerning, but the same research consistently notes that human studies show much milder or no significant effects.

A review of randomized controlled trials in humans found that while some trials reported shifts in gut bacteria composition with sweetener consumption, many others reported no significant changes at all.¹⁹ The difference between the animal and human findings may be explained by the much higher doses used in animal research, differences in gut anatomy across species, and the greater diversity and resilience of the human microbiome. The gut microbiome remains a plausible pathway through which sweeteners could theoretically exert effects, but the chain from “sweetener alters bacteria in a rat” to “sweetener causes cancer in a person” has too many missing links to be meaningful right now.

Something in the Can That Is Not the Sweetener

There is a cancer-related ingredient in some sodas, both regular and diet, that gets less attention than artificial sweeteners: 4-methylimidazole, or 4-MEI. This compound forms during the production of certain types of caramel coloring, the ingredient that gives colas and other dark sodas their brown color. The National Toxicology Program identified 4-MEI as a carcinogen in animal studies in 2007, and California added it to its Proposition 65 list in 2011.²⁰

A risk assessment measuring 4-MEI in commercially available beverages found wide variation, with some brands containing concentrations high enough that routine consumption would exceed California’s warning threshold of 29 micrograms per day. The highest levels were found in certain malt beverages and some colas, while other colas had far lower concentrations.²¹ Manufacturers have since reformulated many products to reduce 4-MEI levels, but the irony is notable: people worrying about aspartame in their diet cola may be overlooking a compound in the caramel coloring for which the carcinogenicity evidence is, at least from an animal standpoint, more straightforward.

A Scattered Signal for Specific Cancer Types

While the overall picture is null, a few site-specific findings pop up in meta-analyses and deserve mention. A dose-response meta-analysis of prospective studies found that high intake of artificially sweetened beverages was associated with a modestly elevated risk of non-lymphoid leukemia, though the analysis included only three studies for that endpoint. The same meta-analysis found a decreased risk of colorectal cancer with higher intake.²² When a single analysis finds both an increase for one cancer and a decrease for another, and when the number of studies contributing to each sub-finding is very small, the most honest interpretation is statistical noise rather than a real biological signal. The broad systematic reviews, which incorporate these same studies alongside many others, consistently report no overall association.

Children with type 1 diabetes are one group that sometimes consumes large amounts of artificially sweetened products because they are managing their blood sugar closely. Research has found that even in this higher-exposure population, only the very highest consumers (roughly the top 5 percent) occasionally exceeded the acceptable daily intake for certain sweeteners, and that outcome was uncommon enough that researchers described it as presenting “little chance” of exceeding safe limits.²³ For the general adult population, average aspartame intake in the United States sits well below the acceptable daily intake and has actually declined over recent decades.¹⁰