IARC: Can't Be Aspar-Tamed
Sorry, Miley
I missed the announcement of the Inaugural Global Olympics for Botching Science Communication but the competition has been heating up in recent years. The newest round of games began last week with the Reuter’s Exclusive informing us of a leak stating that the World Health Organization (WHO)’s International Agency for Research on Cancer (IARC) will classify aspartame, the non-nutritive sweetener used in many sugar-free and sugar-reduced products (sodas, gums, etc), as ‘possibly carcinogenic’. There’s been dozens of media articles and social media reacts to this news, some on point (IMO) and some missing the mark or using this information to spread unnecessary fear (though my favorite takes were from creators letting DietCoke defend herself). As much as I’d like to be critical of the more fearmongering takes, the contrived approach to #scicomm coming from IARC concerning hazard vs risk is mostly to blame here and the confusion is quite predictable (more on that below).
Some of the responses to this aspartame news fell along the lines of “duh, we’ve known this”. Artificial/Non-nutritive sweeteners have a long history of both scientific & public concern, especially when it comes to cancer. In 1969, the FDA removed the generally recognized as safe (GRAS) status of the NNS cyclamate for possible bladder cancer risk (especially when used in combination with saccharin, another NNS). Saccharin similarly has had a tumultuous history, being added to the US National Toxicology Porgram’s Report on Carcinogens as an anticipated human carcinogen (also linked to bladder cancer in rats), but then later removed in 2000, largely because of negative mutagenicity and genotoxocity testing as well as the bladder cancer phenotype requiring a set of physiological conditions unique to the (male) rat (male rats excrete high levels of protein, contributing to a high urinary pH and sodium concentration as well as crystal precipitates needed to induce a proliferative response and neoplasms).
Aspartame, a methylated derivative of 2 amino acids (phenylalanine and aspartate), was approved for use as an additive in the USA in 1974 with clear guidance on the doses able to be used in foods and beverages as well as requiring a warning that it contains phenylalanine, a concern for individuals with the inborn error of metabolism, phenylketonuria (PKU). This approval came after its discovery in 1965 when James Schlatter was working to develop antiulcer drugs and licked his fingers, noticing the sweet taste (such bad laboratory practices also led to cyclamate’s discovery). Aspartame has quite a storied history since its approval, with early claims about brain tumor risk as well as consumers claiming seizures following its use (likely due to popularized claims that aspartame is neurotoxic informed by extremely high dosing studies performed in animal models). More drama would hit aspartame following rat carcinogenicity studies published by the Ramazzini Institute, the results of which would be heavily criticized by several governmental panels including the European Food Safety Authority (EFSA) for issues with study conduct and tumor diagnosis at the institute. Such concerns are not unique to Ramazzini’s aspartame investigations; indeed, the US EPA and National Toxicology Program (NTP) have limited their use of data on other exposures (e.g. methanol) from the institute for similar reasons, making data from the institute hard to trust (though many consumer advocate and industry watchdog groups still side with Ramazzini in opposition to artificial sweeteners). Despite all of this controversy, aspartame is still considered safe for use by many authoritative scientific bodies and there are Acceptable Daily Intakes set by the FDA (50mg aspartame/kg body weight), Europe (40) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA) (0-40mg/kg) - if you’re interested in JECFAs overview of the chemical toxicology data (which it is currently updating concurrently), you can find it over at INCHEM.
The historical, seemingly shotgun approach to claims about aspartame and broad cancer risk might have you questioning whether something about aspartame as a chemical compound makes it likely to cause cancer? It’s not particularly genotoxic or mutagenic either in vitro or in vivo across a range of dose exposures. It’s also pretty rapidly metabolized, being digested in the intestine to it’s constituent amino acids and methanol. The methanol product is often used in claims surrounding the ‘risks’ of aspartame because methanol is first oxidized in the liver to formaldehyde and then to formic acid. Because methanol and formaldehyde are more colloquially understood as toxins, these have long presented a low-hanging fruit rationale for why folks should avoid the substance, despite the amounts of methanol generated from aspartame-sweetened products being less than a the equivalent formed in the digestion of apple juice. As with any substance, the dose makes the poison.
Central to understanding the IARC ruling on Aspartame is understanding Hazard vs Risk. You can consider whether something is hazardous (can a substance in any circumstance cause cancer) vs risk (hazard X dose i.e. what is the likelihood of increased cancer at a specific intake level). The classic lay example of HvR is that swimming in the ocean is hazardous, because drowning is a possibility and sharks exist, but the risk of harm from swimming is extremely low. It is quite easy to scare someone out of swimming in the ocean if you are constantly yelling shark (and weaving conspiracy about how BigSnorkle is hiding the truth about swimming’s risk of shark attacks) - but this doesn’t change the fact that its extremely unlikely that you’ll be attacked by a shark when swimming in the ocean. Hazard vs risk has long plagued areas of nutrition, such as in the well-publicized case of monosodium glutamate/MSG - old studies where MSG was given at extremely high doses, sometimes injected rather than fed, led to concerns (along with quite a bit of xenophobia) about consuming very low culinary doses of MSG (literally just sodium and an amino acid that bind tightly to glutamate receptors in the taste bud to induce the umami flavor/sensation).
IARC does not hide, rather it states quite openly in its mission and on its Q&A that it is concerned with whether an agent is capable of causing cancer (i.e. hazard) but does not measure the likelihood that cancer will occur at a particular exposure level (i.e. risk). IARC is essentially asking whether swimming in the ocean can result in shark attacks, not how likely this is. Many individuals do not care if their aspartame-sweetened diet coke could cause cancer, if that carcinogenicity only occurs when consuming 10,000 cans per day (though I know folks up for that challenge..). For any genuinely concerned folks from these headlines, IARC’s ruling is not something to dismiss but to interpret as “okay so aspartame is ‘possibly carcinogenic’ - but, assuming it can cause cancer, how likely is cancer when I’m consuming X amount?”.
IARC’s forthcoming Monograph labels Aspartame as “possibly carcinogenic (its Group 2b classification). It’s worth digging into what “possibly carcinogenic” means. This is defined as having “some evidence that it can cause cancer in humans but at present it is far from conclusive”. To make a 2B categorization, there needs to be 1 of 3 factors present for an IARC Working Group:
limited evidence of carcinogenicity in humans
sufficient evidence of carcinogenicity in experimental animals
strong evidence that the agent exhibits key characteristics of carcinogens
Ideally, we’d want data from all 3 of these groups to align to be really confident something is carcinogenic, but only 1 is required for the 2B designation, consistent with its higher degree of uncertainty and label as only ‘possibly’ causing cancer. Most of the Group 2B carcinogens are industrial/occupational chemicals as well as viruses, but the list includes a number of nutrition relevant determinations, including Kava extract, Ginkgo biloba extract, Aloe Vera leaf extract, and traditional Asian pickled vegetables. Pickled vegetables is worth diving into since its designation details the uncertainty in this category - it received this grouping because of the low levels of carcinogenic N-nitrosamine compounds in pickled vegetables as well as very mixed epidemiological data, with individual studies (varying in design, the dietary assessment methods, dose differentials, methods of preparation, etc) showing correlations between intake and stomach & esophageal cancer (deemed as ‘limited evidence’ by the IARC working group). The IARC determination on its own doesn’t tell you much about how to act, as it’s not clear what doses of what kinds of pickled vegetables are associated with increased risk (e.g. is a single vinegar pickle slice on a sandwich a real risk? does the degree of salt added or fermentation matter?). IARC is not necessarily wrong for pointing out this possible hazard, it’s just pretty useless information for policy makers and consumers on its own. Other organizations, in this case, provide much more informative guidelines; the WCRF/AICR guidelines include a dose-response meta-analysis of salt-preserved vegetables (predominantly from Asian cohort studies) indicating a 9% relative risk increase for every 0.5 serving consumed per day. WCRF/AICR also look broadly at other salt preserved foods (e.g. meats in Norway) and observes similar increased risk, classifying foods preserved with salt broadly as probable cancer risks, noting the consistent epidemiology as well as experimental laboratory evidence pointing to plausible mechanisms (while also noting some uncertainty in the epidemiology given that pickled vegetable consumption may be associated with lower socioeconomic status and higher incidence of H. pylori infection, a causal risk factor for stomach cancer). Much like the case of pickled vegetables, we shouldn’t expect much actual dietary guidance coming from IARC - other organizations are charged with going much beyond the “can it cause cancer?” question to provide specific risk-reducing guidance.
The situation we are in with IARC determining that aspartame is a Group 2B carcinogen can’t be fully evaluated until the entire Monograph comes out but the drama its causing and unnecessary fear over a common food additive it is fostering is quite predictable. Actually, so predictable that the Mara Burr, JD, Office of Multilateral Relations in the Department of Health and Human Services’ Office of Global Affairs sent a letter to the WHO with some concerns about the IARC ruling in August 2022. Burr’s concerns stem from the fact that 2 WHO agencies, IARC and JECFA, are both performing reviews of aspartame. Burr raises concerns that IARC’s assessment will be incomplete relative to JECFAs, as JECFA’s assessment includes all safety endpoints, including carcinogenicity, and has access to non-public proprietary data, whereas IARC only examines carcinogenicity and is limited to public data. Burr’s letter doesn’t emphasize the hazard vs risk characterizations as much as I would’ve liked - JECFA here is charged with possibly setting an updated ADI (safe amount you can consume per day), essentially making a risk characterization based on all relevant health endpoints, whereas IARC is simply looking at hazard with respect to cancer. Nevertheless, Burr’s conclusions are very relevant, noting that IARC’s report may come to different conclusions than JECFA’s and cause public confusion, despite JECFA’s review being of most use to regulators, consumers and ultimately, public health. I eagerly await JECFA’s update as it’s been several decades since it's last review.
I don’t think there’s anything particularly nefarious about IARC’s Monographs - understanding whether something can be carcinogenic is the basis for going on to do risk characterizations and ultimately improve public health. Unfortunately, IARC’s not done a great job of interfacing with the media to clarify what the difference between hazard and risk assessments are - in the current media clickbait ecosystem, it is more essential than ever for authoritative bodies to get out ahead of the narrative and be aggressive about combatting spin that ultimately fosters misinformation. There are innumerable environmental factors that may be carcinogenic but consumers need clear, simple communication about what modifiable risk factors with the biggest bang for their buck that they should pursue.
If you made it to the end and are disappointed I didn’t answer your question of whether you should be worried about cancer risk from consuming aspartame, here’s a link to the National Cancer Institute’s page on artificial sweeteners and cancer risk as well as the Cancer Research UK’s page.