It was recently leaked that the World Health Organization (WHO) is about to change their classification of the artificial sweetener, aspartame, to category 2b – possibly carcinogenic to humans. This has caused some confusion, and has fed existing conspiracy theories about the sweetener. For some reason, even before the web, there was a viral campaign to spread fear and misinformation about aspartame. This took on a life of its own, and could not be quelled by something as pedestrian as evidence.
There must be something about the narrative that resonates – aspartame seems too good to be true. It is 200 times sweeter than sugar, and so can give the foods we love a sweet taste without the calories. Something in our “Simpsons” programmed sensibilities and baseline cynical nature says – it must be unhealthy. Also, the appeal-to-nature crowd that demonizes anything they arbitrarily deem not natural found a convenient target to sow fear and mistrust in mainstream institutions.
Meanwhile, there have literally been thousands of studies, and over 60 scientific and regulatory bodies around the world have reviewed the evidence and found aspartame to be safe for human consumption. There are also many reviews of the evidence in the published literature, generally finding no clear evidence of any link between aspartame consumption and any disease. It always needs to be noted that the one exception is people with phenylketonuria, who should not consume aspartame.
But researching risk is complex and the evidence is often misunderstood and misrepresented. One type of such research is pre-clinical, basic toxicology research. This focuses on whether the chemical in question does stuff to cells or biochemistry. How is it metabolized, what are the breakdown products, does it increase or decrease inflammation or oxidative stress, is it a potential mutagen, and many other assays. This kind of research just tells us if a substance is a potential hazard, but not whether it will confer a health risk.
The difference between hazard and risk is important to understand in terms of this research. A good analogy I often go to is – a shark in a tank is a hazard, meaning that it can potentially cause harm in the right circumstance. But as long as you don’t swim in the tank with the shark, the risk is zero. Something happening chemically may be a hazard, but we need to know how the substance is metabolized, will it get to the target tissue and in what dose, and what compensatory mechanisms are there? A potential hazard can be of zero risk depending on exposure.
Then we get to animal research, which more directly assesses risk. But in animal research the goal is often to “Mythbuster” the question – stress the system until it breaks, meaning increasing the exposure to a potential hazard to see where it starts to become a risk. Sometimes this is explicitly done in what is called the LD50 test – what is the dose that kills half of test animals? Toxicity is all about dose, and so this research is used to know where the threshold dose of toxicity is. Typically, regulations then build in 1-2 orders of magnitude of a safety buffer.
Finally there is clinical and ecological research – is there any actual increase in a negative outcome (such as disease incidence) from controlled or real-world use? This is the definitive evidence about risk. It doesn’t matter if a substance changes a chemical in a petri dish, or kills rats when given in massive doses. What matters is if there is any net health risk when used by humans in the usual dose.
With aspartame, as with anything that gets studied enough, there is a lot of noise in terms of things that happen at a petri dish level, or in animal research, but there is no apparent risk of any disease when consumed by humans in normal amounts. Actual risk may come into play with levels of consumption that are inhuman – on the order of 60 liters of aspartame-sweetened drinks per day every day.
With this as context, how do we make sense of the new WHO classification? We need a bit of further context. The WHO classification system is based on hazard, not risk. They are not determining that something actually causes cancer, only whether or not it is a carcinogen. Further, the classification system is based upon the strength of the evidence, not the magnitude of the effect. Classification 1 is for substances known to be carcinogenic to humans. This means the evidence is clear, but it says nothing about the effect size – we could be really confident in a tiny effect size. Classification 2A is probable, and 2B (the apparent new classification for aspartame) is possible. Category 3 is unclassifiable as a carcinogen, and class 4 is probably not carcinogenic.
A lot of things that we consume every day are in category 2B. Drinking hot beverages is in category 2A, above that of aspartame. Here is a full list of 2B.
But even the categorization of 2B is likely to be controversial. Here is the latest review of the question in the published literature, from 2022, which concludes:
Taken together, available evidence supports that aspartame consumption is not carcinogenic in humans and that the inconsistent findings of the RI studies may be explained by flaws in study design and conduct (despite additional analyses to address study limitations), as acknowledged by authoritative bodies.
Even if we can quibble about the evidence and whether or not there is the possibility of a carcinogenic effect from aspartame, it is not controversial that exposure levels would have to be much higher than what most people consume to confer any risk. At the end of the day the evidence still supports the conclusion that aspartame is safe at human levels of consumption.
