Fructose, Corn Syrup & Cancer
New research found that fructose metabolism enhances tumor growth. What does this mean for dietary fructose consumption?
Not medical advice.
Podcast quotes may be lightly edited for concision & clarity.
Questions we will explore in light of recent scientific findings on cancer and fructose metabolism:
Are there any negative consequences to consuming too much fructose?
How much is too much?
What about fructose from natural whole food sources, versus high-fructose corn syrup (HFCS) and ultra-processed foods (UPFs)?
Some interesting new research may help you decide how much fructose to consume, and in what form. I’ll share how these results inform my own decision-making heuristics, but I won’t be spoon-feeding you “the answer.” (If you don’t develop enough agency to set your own guidelines, you probably won’t like my content anyway.)
The latest science is an important source of information informing my consumption decisions, but it’s not all I rely on. Lots of bad research gets published—if you blindly follow the conclusions of whatever “the experts” publish in a journal, you will often be led astray. Case in point: the entire field of nutrition epidemiology.
One way people get led astray by “the science” is by internalizing the (false) conclusions of pseudo-fields like nutrition epidemiology. Such fields are built upon a shaky foundation of unreliable data inputs, preventing even the mere correlations they catalog from resting on firm empirical ground.
In contrast to nutrition epidemiology, the "gold-standard” for understanding cause-and-effect in human health are randomized controlled trials (RCTs), in which large numbers of real people are subjected to tests and measured against placebo controls. But here too we run into trouble: many RCTs are poorly conducted or underpowered (low sample sizes). A well-done RCT is very valuable indeed. However, even if all RCTs were well-done, they would still leave us with blind spots—for practical or ethical reasons, many informative human RCTs simply cannot be performed. For example, conducting a human RCT comparable to the animal studies we’ll describe below would involve giving different groups of people who already have cancer different amounts of supplemental fructose, with the expectation this will accelerate tumor growth in the high-fructose groups. No ethics board would approve such a study.
Given the above limitations in the clinical literature, we must often turn to animal studies. This type of research enables for rigorous cause-and-effect experiments that uncover underlying biological mechanisms. The drawback? Laboratory conditions are controlled but contrived, not naturalistic. Things often don’t work the same in humans as in mice or zebrafish.
This is where evolutionary thinking can help guide thinking: Are the biological systems at work highly conserved or divergent between animal models and humans? The higher the degree of evolutionary conservation, the more likely the results generalize to other species. Some aspects of biology work in similar or even near-identical fashion in humans, mice, and other creatures. Other things are variable between sub-groups of the same species.
Since we’re reviewing an animal study about fructose metabolism, it helps to know how much fructose metabolism differs between mice and humans. For an intro, I recommend these two articles (especially the first):
Related content:
Article: Dietary Fructose & Metabolic Health: An Evolutionary Perspective
Article: Sugar Has No Impact On Obesity: Just-so stories in metabolic health
Before reviewing this study, let’s prime our intuition using a combination of common sense and real-life human experience. How much fructose intake is too much? A safe bet is that the answer is somewhere between zero and this much.
Believe it or not, The Cooking Goth is providing valuable data. The sample size is low. The dosing, totally uncontrolled. But we’re looking at behavior of a real person in their natural habitat. It may be performative, but that’s now just another part of our hyper-modern human environment. Millions of people consume this and similar fructose-infused products, based on two factors that all animals use these to guide food intake: palatability and availability. How attractive something is to the senses, divided by how easy it is to consume. Appeal over effort. The “food” here may be an artificially produced human concoction, but this is ordinary animal behavior—eat what’s available, according to taste. The real point here is that fructose is added to processed foods to enhance palatability, which promotes feeding.
Most of our focus here will be on how fructose consumption relates to cancer growth. For general background on liver metabolism and metabolic health, read this article first. It describes how fructose metabolism in humans and other apes differs from that of other animals (we’re effectively more metabolically sensitive to fructose than many other species).
In M&M #215, Dr. Gary Patti provided an excellent overview of the basics of cancer metabolism, unpacking recent work linking dietary fructose metabolism to tumor growth, which we will examine here.
The punchline of that work: Cancerous tumors grow faster in the presence of supplemental fructose. This effect was seen in two model animal species (fish and mice) and multiple cancer types, pointing to evolutionary conservation in the underlying biology. This suggests that it’s plausible that results will extend to humans. For ethical reasons, the equivalent work cannot be done in people, as this would involve giving those with cancer varying amounts of fructose with the expectation that those receiving more fructose will experience accelerated tumor growth.
Related content:
Podcast: Cancer Metabolism: Sugar, Fructose, Lipids & Fasting | Gary Patti
Article: Dietary Fructose & Metabolic Health: An Evolutionary Perspective
Study Overview: Fructose metabolism boosts tumor growth indirectly via lipids
I won’t be providing a thorough dissection of the technical details. Here are the basic findings and conclusions of the study, followed by some further commentary:
Fructose supplementation enhances tumor growth in animal models of melanoma, breast cancer, and cervical cancer (without causing weight gain or insulin resistance*).
*Note: the mouse experiments used female animals, as there are sex-dependent effects of fructose on metabolic health. This enabled researchers to focus on the effects of fructose on cancer without extra confounding from obesity and insulin resistance (which would likely have been seen in males).
Cancer cells themselves did not readily use fructose as a nutrient, as they did not express key enzymes for fructose metabolism.
Fructose enhanced tumor growth indirectly, via liver metabolism. Hepatocytes (liver cells) metabolize fructose and excrete various lipids, which can be consumed by cancer cells to enhance growth.
Supplementing animal diets with high-fructose-corn syrup increased serum levels of some of those same lipid by severalfold, indicating that HFCS consumption can greatly boost levels of fructose-derived lipids found to increase tumor growth.
This involved both in vitro (cell cultures of various cancer lines) and in vivo experiments in two species (zebrafish and female mice). Fructose supplementation in both animal species boosted tumor growth. Despite enhancing tumor growth in animals, cancer cells couldn’t readily metabolize fructose in vitro. If cancer cells directly use fructose to fuel growth, you would expect that bathing cancer cells in fructose directly would also facilitate growth. But it didn’t, which is puzzling.
Here’s what the data looked like, showing that fructose enhances tumor growth in animals, yet isn’t metabolized efficiently by cancer cells on their own.
The effect here is indirect. Fructose is first metabolized by other, non-cancer cells, which secrete lipids (fats) that cancer cells can then consume. Recall that most of our cells do not metabolize dietary fructose. With fructose consumption, the bulk of the metabolism occurs in the liver and certain gut cells. I have discussed fructose metabolism by the liver in more detail here, in the context of the evolution of ape metabolism.
The picture that emerges from this study is that fructose is metabolized largely by liver cells, just as it normally is. Metabolism by hepatocytes produces lipids such as lysophosphatidylcholines (LPCs). These LPCs can then be consumed by tumors to enhance growth. Giving LPCs to mice directly, instead of fructose, increased tumor growth. Supplementing their diet with high-fructose corn syrup (HFCS) boosted serum levels of multiple LPCs by a lot.
As Dr. Gary Patti explained to me, the LPCs produced from fructose metabolism are more water soluble than other lipids, allowing them to travel through the bloodstream without having to be packaged into complex lipoproteins, as with LDL or HDL cholesterol.
GP: What we found is that, in addition to making those kinds of lipids [LDL, etc.], fructose causes the liver to excrete this other kind of lipid called lysophosphatidylcholines [LPCs]. And LPCs are different than the other lipids that are packaged [such as LDLs, etc]. LPCs are a little more water soluble. When you think about the lipoprotein particles, it's a very complex biochemical mechanism.
NJ: It's a way for the body to move non-water soluble fatty stuff around in the watery environment of our blood.
GP: Exactly. [But these LPCs] get excreted, and they can circulate without the complexities of lipoprotein particles.
NJ: So these other cells [liver cells] that are not tumor cells, they're metabolizing the fructose. They're creating these LPCs, these fairly water soluble lipids. Those can now just go through the blood, and the tumor cells will suck them up like they would any other energy source.
GP: [The tumor cells are] able to utilize [LPCs] much more directly than than you could with the more complex machinery that's involved in lipoprotein particles—which is amazing, you know, it's kind of a magnificent way for a cell, a cancer cell, to get access to lipids faster.
To reiterate, the basic story here looks like this: animals consume fructose; liver cells metabolise fructose, producing LPCs and other relatively water soluble lipids; those lipids then circulate in the bloodstream; tumor cells growing anywhere else in the body, even away from the liver, can now uptake those lipids and grow faster.
Two obvious questions that arise here, in terms of human relevance:
Are the supplemental levels of fructose they gave to animals comparable to levels of dietary fructose consumption in humans?
Is the same basic biology likely in play in humans as well?
The short answer to both questions: Yes.
As in other animals, the bulk of human fructose metabolism occurs in the liver. In fact, humans and certain other primates seem to be more metabolically sensitive to fructose than other animals, due to mutations in a gene for the enzyme uricase (more detail here). But how much fructose were these mice being given. Well, in many of the expeirments, they consumed it ad libitum (as much as they wished).
GP: The animals could consume the [food] as they wished. So we put the fructose in the water. We didn't infuse them with them. We did do some experiments where we infused them, but for the vast majority of what we did, we just gave them access to it. And just like if you—if you gave me access to cake and didn't tell me it was bad, I'd eat a lot actually, because it was yummy. And so I think the that's basically what happened. So, it’s not super-physiological in the sense that we didn't force feed them [to eat] amounts more than they wanted to consume.
NJ: So the animals are choosing to eat as much as they want, which is, at a coarse grain level, not unlike the natural situation for a lot of humans. We're in an environment where we can eat as much fructose as we want, and a lot of people do.
One last point to make is that this study showed that fructose enhanced the growth of tumors that were already present. It did not show that fructose was carcinogenic, causing cancer to arise at higher rates in the first place. That’s a distinction worth remembering.
With these results in mind, let’s take a quick peek at how fructose and high-fructose corn syrup (HFCS) consumption in humans has trended over time.
Fructose: Consumption Trends, Formulations & Palatability
Here is a graph from this excellent paper, which I’ve discussed in detail with the senior author. It’s showing you the per capita supply of different caloric sweetners in the US, from the early 1900s until about 2020. The red line shows the supply of corn sweetners, dominated by high-fructose corn syrup (HFCS).
Notice that we hit peak availability around year 2000, which declined after. Despite the 21st century decline, corn sweetner supply has remained much higher than it was for most of recorded history—per capita supply is severalfold higher than it was before the mid-1900s, even after the decline since we hit peak corn sweetener in 2000.
HFCS is a mixture of free glucose and fructose molecules (not bound together). There are multiple HFCS formulations, each with a different fructose:glucose ratio. The most common formulation is HFCS-55, which is 55% fructose. It’s widely used in soft drinks. Unlike HFCS, sucrose (table sugar) is a 50:50 mix, with glucose and fructose molecules bound together one-for-one.
Keep in mind that charts like this are usually not depicting intake per se. They’re showing supply, which is a proxy for consumption. The two variables probably correlate well, but I’ve never seen any cold-hard data showing exactly how well.
The bottom line here is simply that humans have been consuming more sugar overall, and more fructose in particular, for several decades. This remains true even after some declines in sugar supply since the early 2000s. As explained more here, many people today consume as much or more fructose than even species we think of as eating “a lot” of fruit, like chimpanzees. And for the average American, fructose consumption is coming mainly from added fructose found in ultraprocessed foods and sugar-sweetned beverages, devoid of fiber and other nutrients that would otherwise slow down fructose absorption.
HFCS seems to be especially prominent sweetener in the United States:
Why would the most common HFCS formulation be 55:45 fructose:glucose? Food companies have the ability to use any ratio they want, so why that specific one? As I discussed with Dr. Patti, there are at least two factors here. One is economic: fructose is apparently cheaper. So then why not use pure fructose? Why 55%?
Well, the other factor here is palatability. The perceived sweetness of foods varies with the fructose:glucose ratio, as does how reinforcing the formulation is. Over time, food companies presumably experimented with varies HFCS formulations in different foods. By measuring sales (which they naturally seek to maximize), they could select over time whichever formulations were associated with the best sales. (If you have any good sources on the history of HFCS formulations, please share in the comments.)
GP: [Fructose consumption has] gone up quite a bit, and it's been used in many ways as a replacement of glucose. One of the reasons for that is because fructose, is cheaper to make. And per molecule of fructose, you get more sweetness out of it.
NJ: I've known that the fructose glucose ratio has a big effect on palatability and and how sweet we perceive things. So I always assumed that the reason that all of these big companies have converged on these very particular fructose:glucose ratios of like 55:45 [because] they're optimizing for the sweetness that comes from having a little bit higher fructose than you would get with sucrose, and that's driving increased palatability and therefore increased consumption, which is going to support their business. What you're saying is [that] there's actually a dual thing here, which is, not only is it modulating palatability by boosting fructose, it actually is more economic for the companies because the fructose is cheaper.
GP: Yeah, that’s right.
All of these things aside, Dr. Patti’s results indicate that dietary fructose enhances tumor growth, at least if a cancerous tumor is already growing. On average, modern people are consuming more fructose than we have for most of history. (And yes, some peoples, like the Hadza of Africa, eat lots of honey, seasonally. But the form factor and lifestyle of these people is nothing like the typical modern sedentary UPF junkie. More detail here.)
The obvious implication here is that the modern high-fructose food environment may be fueling faster tumor growth in people with cancer.
So, how much fructose is too much, exactly?
Is it enough to be mindful of how many HCFS-infused beverages I drink, or should I limit natural sources, such as fruit? If we want to minimize cancer risk, should we aim to cut out fructose entirely? Or is it enough to focus on cutting out sugars added artificially added to processed foods to enhance palability?
Dietary Fructose for Humans: Upside benefit vs. downside risk
To the extent that the results from Dr. Patti’s lab generalize to human cancers, higher fructose exposure drives enhanced tumor growth. This implies that minimizing fructose intake would minimize tumor growth rate. But humans have been consuming fructose from natural whole food sources for ages. It’s not difficult to imagine that someone sipping on liquid HFCS in the form of soda and eating lots of ultraprocessed, sugar-rich sweets is exposing themselves to “excess” fructose, but what about someone who likes to eat oranges and bananas? What counts as “excess”?
I have written about the evolution of fructose metabolism in primates, including how humans and other apes may be more likely than other animals to experience negative metabolic effects from consuming high levels of fructose. But how do we estimate how much “high” is?
One way to think about this is to first ask whether fructose consumption is necessary, and whether there are benefits to consuming some amount of it. On the first point, there is a clear answer: fructose consumption is not necessary. The body can produce fructose endogenously, so there’s no strict need to obtain it from diet. In contrast, many nutrients cannot be endogenously produced by humans, so must be consumed in order to maintain good health and avoid illness. We call those, “essential nutrients.” Fructose is a non-essential nutrient.
If humans can produce fructose endogenously, there is no need to consume it at all. That does necessarily mean, however, that are there are no benefits to consuming some fructose. The question is: are there any benefits that come specifically from consuming exogenous fructose? If so, then we need to balance those benefits with the potential for enhanced cancer growth in deciding what “too much” fructose means. If not, then minimize exogenous fructose as much as possible may a viable strategy.
Here’s what I discussed with Dr. Patti on this subject:
NJ: So if you had to compress the takeaways from the study… Is this basically stating that if you're eating “too much” high-fructose corn syrup, whatever exactly that means, that it has the potential to facilitate tumor growth?
GP: Absolutely, I think that is a take home message. Obviously, I can't tell you that we validated this and tested it million ways in humans—that's all the next step. Everything that's in the paper you've mentioned is in animal models… but I think there's compelling evidence to believe that this will translate to humans. And I can say that it's not going to hurt you if you consume less fructose. Let's say it that way: there is a big possibility that [consuming less] will benefit you. If you have cancer, you consume fructose, there's a possibility it could hurt you, but I don't see any possibility that it would be bad [to consume less].
NJ: So there's certainly downside potential to eating too much fructose, whatever exactly “too much” means. [But] there's probably no downside risk to eating less fructose?
GP: Exactly. And again, I do want to put the caveat that [we are not talking about] avoiding fruit. That's not what we mean.
NJ: Is it fair to say that… we're not saying avoid all fructose or avoid fruit? Part of what that means, to me at least, is natural whole foods are naturally limited in certain ways, right? They have a certain amount of fructose inside of them. They contain fiber and other things that will limit the rate of absorption. Our bodies have evolved to extract nutrition from whole foods that are part of the natural environment and have been for 1000s and millions of years. When we start adding things on top of that is where we can start to get into trouble.
GP: Absolutely. I think that's perfectly well said.
NJ: What would happen if someone was consuming zero fructose? Anything break? Would anything go wrong that we know about?
GP: No. In fact, you can make fructose endogenously.
Have good resources suggesting benefits of dietary fructose? Please share them with me in the comments.
My general approach to these things is to take an evolutionary perspective. What did fructose consumption look like before the industrial era, extending from the Neolithic period (after farming began) all the back back through Paleolithic times? As discussed here, quite a bit a metabolic adaptation has occurred (in some human populations), since the dawn of agriculture at the beginning of the Neolithic period, ~10-12,000 years ago.
Specific fruits are native to specific parts of the world. As you get closer to the tropics, fruits are more likely to be available year-round. As you move away, fruits become more sparsely available and more seasonal. I am not aware of any traditional human population that survived exclusively or mostly on fruit, and thrived for any extended period.
There are, however, examples of traditional humans who survived with little no fruit consumption. The Inuit would be one example people may be familiar with (they ate small amounts, seasonally). Ice Age humans, such as the Clovis people (ancestors of Native Americans), were recently discovered to eat a “ketovore” diet, consisting mostly of animal megafauna.
Some traditional people ate quite a bit of fruit, but never exclusively fruit. Some traditional people ate little to no fruit, ever. Lots of traditional people ate some fruits, sometimes. Remember: many kids today are consuming more fructose than wild chimps. I can’t tell you precisely how much fructose is too much, but I’m pretty sure that’s too much…
Related content on the evolution of human diet & metabolism:
Article: What Did Humans Evolve to Eat?
Podcast: Ancient Diets, Human Carnivory, Mammoth Hunting, Clovis Culture & Origin of Native Americans
Again, humans can produce fructose endogenously. Even if you’re descended from recent ancestors who lived in the tropics, where fruit is most abundant, it’s doubtful your body needs exogenous fructose. That does not mean certain peoples could thrive without fruit, however. For many historical populations, this would have been the major source of essential nutrients like vitamin C.
For no less than tens of thousands of years, my ancestors would have had seasonal access to certain fruits. By constrast, I have perpetual year-round access to virtually any fruit, with the bonus option to purchase Fruity Pebbles Syrup, like our friend The Cooking Goth.
Here’s the basic common sense approach I take for myself, based the information I shared above: avoided added fructose whenever reasonably possible, and eat whole fruits, sometimes. Fresh is better than rotten. Seasonal is better than year-round. Locally sourced is better than imported. All of those things are good because I prefer to eat foods from organisms that were well-nourished themselves. Life happens in cycles. Most foods mature and become ripe, and then decay. The nutritional content of those foods changes across their life cycle, often synchronized with the seasons. That’s why animals have evolved sensory abilities to detect and consume specific foods, at specific times.
For most of human evolutionary history, we could eat like animals, using our evolved sensory systems to select appropriate foods from the environment based on taste, from a food supply limited by the natural environment. Today, virtually all foods are always available to us, most of which were consciously engineered by humans who have shrewdly hacked our biology to confuse our senses and thwart the body’s satiety mechanisms.
The modern food environment makes it harder than ever to make conscious choices, without succumbing to carnal desire. We’re embedded in a surplus of hyperpalatable super-stimuli. Not giving into temptation requires tapping into something stronger. Inspiration may be needed.
Plato said, "Self-control is the chief element in self-respect, and self-respect is the chief element in courage." The Bible says, “The spirit is willing, but the flesh is weak." Paramahansa Yogananda even used a sugar analogy: “Temptation is a sugar-coated poison; it tastes sweet at first, but ultimately destroys.”
There are many possible sources of inspiration. Even The Cooking Goth.
Learn more about fructose metabolism:
Study: Dietary fructose enhances tumour growth indirectly via interorgan lipid transfer
Podcast: Cancer Metabolism: Sugar, Fructose, Lipids & Fasting | Gary Patti
Podcast: Sugar & Sweeteners: Are They Really Addictive or Unhealthy? | Nicole Avena
Article: Dietary Fructose & Metabolic Health: An Evolutionary Perspective
Article: Sugar Has No Impact On Obesity: Just-so stories in metabolic health