Unpacking the 'ketogenic diet cures cancer' myth: Part 1
Dr Thomas Seyfried claims that a ketogenic diet can prevent and cure cancer. Extraordinary claims require extraordinary evidence; let's see if he delivers the goods!
"Are you familiar with Diary of a CEO podcast/YouTube channel? The channel has a large following (7m+ on YouTube alone) and I’ve noticed that the last few guests have been ‘experts’ in the area of metabolic health and cancer prevention. I’ve watched them and it’s been disappointing to learn that they are pushing a ‘low carb, high meat and ketogenic’ diet and they insinuate that anything else (including a whole food vegan diet) is damaging to mitochondria and therefore promotes the formation of cancer. I’d be interested to hear your thoughts and research on this in a newsletter article in the future. I imagine that your Substack readers would possibly be across some of these interviews on Diary of a CEO.
This is the latest interview with a professor from Boston College:
Sincerely,
Cxxxxx"
I must confess, I'd never heard of Diary of a CEO; my podcast feed is dominated by COVID dissidents and conspiracy realists! But judging by its subscriber numbers, it's reaching a huge audience. So it's deeply concerning when a platform with this kind of reach brings on well-credentialled and apparently credible guests, who then proceed to make claims that are completely untrue about a topic as serious as cancer, and the host does not have the subject matter expertise to challenge these claims.
This interview with Dr Thomas Seyfried, a professor of biology at Boston College, was incredibly frustrating to watch. Seyfried's contention that cancer is primarily a metabolic disease, rather than a genetically-driven condition, strikes me as having considerable validity, although I doubt that altered metabolism is the only causal factor in malignancy. And I agree 100 per cent with Seyfried's criticisms of the cancer treatment industry, and its focus on blockbuster drugs that cost an absolute bomb, while delivering little to no meaningful benefits to patients.
But so much of what Seyfried comes out with in support of his theory of cancer causation and treatment is complete bunkum, easily disprovable by anyone with internet access and a sceptical mind. Fortunately, I have both. And so, as a service to you, my dear reader, I decided to dissect the Seyfried interview and present to you all the evidence that disproves his claims about ketosis and cancer.
I want to stress at the outset that I'm not doing this to nitpick Seyfried. But I firmly believe that when health professionals, scientists and researchers are communicating with the public, they have an ethical obligation to be incredibly conscientious - to meticulously fact-check themselves. Regular readers will have noticed that my posts are densely referenced. It takes me a long time to write each article, because every time I write a statement that asserts something as factual (for example, obesity increases the risk of cancer), I stop and ask myself 'Is this true? How do I know that? Is it true in all cases, or are there exceptions to the rule?' Then I trawl through the scientific literature to find out what the current state of knowledge is on the topic, taking into account the opposing views. And often, I end up amending what I'd started to write, because it turns out that my understanding was incomplete. Not only do I want to ensure that I have as complete a grasp of the topics that I write about as is humanly possible, I want you to have access to the most complete understanding too.
Unfortunately, Seyfried managed to pack so many false claims into this interview, that this enterprise is going to take several posts! We'll kick off with his claims about the altered metabolism of cancer cells. I've included timestamps so you can easily locate the relevant claims.
7:36 "They all [i.e. all cancer cells] have one thing in common: they depend on fermentation - energy without oxygen - so all cancers are a singular type of disease, it's just that they happen in different tissues. But when you look at the underlying problem, they're all very very similar - they can't live without fermentation, which means energy without oxygen. So that's the common pathophysiological problem in all cancers whether it's a colon, brain, breast, bladder, skin, lung - we've looked at all these cancers and they're all essentially using the same mechanism [i.e. the mechanism of fermentation] to grow out of control."
Here, Thomas Seyfried is discussing the Warburg effect, named after the German biochemist Otto Warburg, who won the 1931 Nobel Prize for his discovery of an iron-containing enzyme central to cellular respiration - the process by which organisms extract usable energy in the form of adenosine triphosphate (ATP) from molecules derived from food. Most of the body's normal, healthy cells use the process of oxidative phosphorylation - burning components derived from food in the presence of oxygen, inside the mitochondria (the cell's 'power plants') - to produce the bulk of their ATP. In 1923, Warburg published his finding that cancer cells use glycolysis - the anaerobic (i.e. without oxygen) breakdown of glucose in the cytosol (the fluid inside the cell) for energy production, even in the presence of adequate oxygen to carry out oxidative phosphorylation.1 (Seyfried uses 'fermentation' as a user-friendly shorthand for the phenomenon that Warburg dubbed 'aerobic glycolysis'.)
Warburg believed that this aerobic glycolysis was a universal property of cancer, and posited that this switch from oxidative phosphorylation to the far less energy-efficient process of aerobic glycolysis was the main cause of the disease. However, other researchers subsequently discovered that certain tumours do not display this shift in energy metabolism. In fact, as early as 1929, Herbert Crabtree found that in almost two-thirds of transplantable mouse tumours studied, oxidative phosphorylation contributed significantly to cells' energy budgets, in many cases outweighing glycolytic energy production. Crabtree also noted that the share of energy generated from aerobic glycolysis varied enormously between tumours of different strains, and even between different tumours from the same strain. In other words, Crabtree found that the environment that the tumour found itself in apparently shaped its preference for using either glycolysis or oxidative phosphorylation to generate energy, and not that the use of glycolysis as the principal means of energy production was an intrinsic feature of cancer.
According to Craig Thompson, a cancer biologist at the University of Pennsylvania, between 60 and 90 per cent of tumours make the shift from oxidative phosphorylation to glycolysis... which means that up to 40 per cent of tumours do not primarily rely on glycolysis for energy production. Thompson discovered that a pro-survival signaling protein called Akt could drive cancer cells to switch to glycolysis "without affecting the rate of oxidative phosphorylation" (emphasis added). Thompson stresses that “the glycolytic shift is not absolutely required for transformation,” but it does gives cancer cells “a higher metastatic potential and a higher invasive potential… because they're now cell-autonomous for their own metabolism.”
On the other hand, Michael Guppy, a now-retired biochemist formerly at the University of Western Australia, argues that the methods used to estimate oxygen consumption in cancer cells are inaccurate, and that in fact, "there is no evidence that cancer cells are inherently glycolytic". In a paper published in 2004, Guppy and Zu synthesised multiple studies of metabolism in normal and cancerous cells, and found that glycolysis contributed, on average, 20 per cent of the energy budget of normal cells, compared to 17 per cent for cancer cells, although there was wide divergence in the glycolytic contribution in different cell types (both normal and cancerous).
Get more from Robyn Chuter in the Substack app
Available for iOS and Android
Contrary to Seyfried's assertion that all cancers use fermentation/aerobic glycolysis to "grow out of control", research on tumour bioenergetics has demonstrated that multiple cancer cell types are metabolically flexible - that is, they can switch from aerobic glycolysis to oxidative phosphorylation, depending on the availability of glucose:
"A subclass of glioma [a type of brain tumour] cells which utilize glycolysis preferentially (i.e., glycolytic gliomas) can also switch from aerobic glycolysis to OXPHOS [oxidative phosphorylation] under limiting glucose conditions [7], [8], as observed in cervical cancer cells, breast carcinoma cells, hepatoma cells and pancreatic cancer cells [9], [10], [11]. This flexibility shows the interplay between glycolysis and OXPHOS to adapt the mechanisms of energy production to microenvironmental changes as well as differences in tumor energy needs or biosynthetic activity. Herst and Berridge also demonstrated that a variety of human and mouse leukemic and tumor cell lines (HL60, HeLa, 143B, and U937) utilize mitochondrial respiration to support their growth [12]. Recently, the measurement of OXPHOS contribution to the cellular ATP supply revealed that mitochondria generate 79% of the cellular ATP in HeLa cells [a cervical cancer cell line], and that upon hypoxia [low oxygen] this contribution is reduced to 30% [4]. Again, metabolic flexibility is used to survive under hypoxia. All these studies demonstrate that mitochondria are efficient to synthesize ATP in a large variety of cancer cells, as reviewed by Moreno-Sanchez [13]. Despite the observed reduction of the mitochondrial content in tumors [3], [14], [15], [16], [17], [18], [19], cancer cells maintain a significant level of OXPHOS capacity to rapidly switch from glycolysis to OXPHOS during carcinogenesis...
While glutamine, glycine, alanine, glutamate, and proline [amino acids, derived from protein] are typically oxidized in normal and tumor mitochondria, alternative substrate oxidations may also contribute to ATP supply by OXPHOS. Those include for instance the oxidation of fatty-acids, ketone bodies, short-chain carboxylic acids, propionate, acetate and butyrate"
Or, in plain English, many types of cancer cells - including several of the specific types that Seyfried insisted were entirely reliant on aerobic glycolysis for survival - can and do adapt to utilise whatever fuel is available to them - glucose from carbohydrate, amino acids from protein, fatty acids and ketone bodies derived from fat, or short chain fatty acids derived from microbial metabolism.
10:51 "The organelle inside the cell that generates energy is not efficient. It's inefficient and the cells are using ancient fermentation [instead of oxidative phosphorylation]... The cancer cell in our body is doing nothing [other] than falling back on these ancient fermentation pathways that become accelerated, upregulated in the tumour cell because the efficiency of the energy coming from the mitochondria is now depleted. It's defective in many different ways. So this is very clear and this happens in lung cancer, colon cancer, we've looked at all the major cancers and we found out these common defects are seen in all the cancers, so they're all very similar in their metabolism."
The organelles that Seyfried is referring to here are the mitochondria, often dubbed the powerhouses of the cell. The prevailing theory for the origin story of our mitochondria is that they were originally prokaryotes (primitive single-celled organisms, of which bacteria are an example), which became internalised into the cells of complex multicellular organisms. Warburg believed that cancer cells switched to aerobic glycolysis because their mitochondria were damaged or defective, a position that Seyfried endorses. But contrary to this assertion, the mitochondria of cancer cells remain both active and functional and, as detailed above, contribute significantly to those cells' energy needs. Even in cancer cells that have switched over to glycolysis as their principal means of energy production, oxidative phosphorylation (which occurs only in mitochondria) continues unabated.
It's also a little misleading to describe glycolysis as "ancient fermentation" as if it's a means of generating energy that complex multicellular organisms have long since abandoned, since in fact glycolysis is the first step of cellular respiration, which is carried out in every cell, every day, and - as reported by Guppy and Zu - contributes around 20 per cent of the average energy budget of normal, healthy cells. In fact, Otto Warburg himself documented "examples of normal tissues possessing aerobic glycolysis" in 1929.
16:57 "So all we have to know with cancer is that they're - how are they growing so rapidly; why are they going out of control; how come it's so hard to kill them? - because as long as you have those fermentable fuels that drive this ancient fermentation pathway they will continue to grow. They're very hard to kill. And the fermenting fuels are glucose and glutamine... So a solution to the cancer problem, to manage cancer without toxicity, is to simultaneously restrict the two fuels that are needed to drive this dysregulated growth while transitioning the whole body off to a fuel that the tumour cells can't use, which is fatty acids and ketone bodies."
Once again, Seyfried's assertion that only glucose and the amino acid glutamine fuel cancer growth, is contradicted by the evidence cited above that many tumour types have been demonstrated to exhibit a high degree of metabolic flexibility, adeptly switching between multiple amino acids (not just glutamine), glucose, fatty acids, ketone bodies and short chain fatty acids, depending on availability and circumstances. I discussed the ability of multiple cancer types to efficiently utilise ketone bodies in Ketogenic diets: Part 5 – Cancer:
"In a study of 33 human cancer cell lines, all were found to express key ketolytic enzymes, which enable them to take up ketone bodies and use them as a fuel. The level of expression varied significantly between different cell lines, meaning that some cancer cell types have a stronger preference than others for using ketone bodies.
Interestingly, the researchers found "no correlation between glycometabolism and ketone body metabolism", meaning that cancer cells with high glycometabolism (use of glucose as a fuel) didn’t necessarily have a lower level of key ketolytic enzymes. Some cancer cells, it seems, are able to use glucose and ketone bodies quite interchangeably, depending on what is available to them.
Some cancer cells show a strong preference for using ketone bodies, and they cause the body to make more ketones in order to fuel their growth. An oncogenic (cancer-causing) mutation in the BRAF V600E gene activates MEK1, which stimulates the growth of cancer. The ketone body acetoacetate selectively enhances BRAF V600E mutant-dependent MEK1 activation in human cancers. Which types of cancer does this mutation occur in?
Over 50% of melanomas;
10% of colorectal cancers;
100% of hairy cell leukemias; and
5% of multiple myelomas.
For people whose cancers express this BRAF V600E mutation, adopting a ketogenic diet would pour fuel on the fire; in mice who were implanted with human melanomas expressing the BRAF V600E mutation, a high fat diet “increased growth rates, masses and sizes of tumors”.
Researchers have also identified a 'reverse Warburg effect' in which fibroblasts (connective tissue cells that produce the extracellular matrix, or stroma, in which all cells, including cancer cells, live) 'feed' ketone bodies to adjacent cancer cells, stimulating tumour growth and metastasis. The human MDA-MB-231 breast cancer cell line is one that has been found to overexpress key enzymes involved in ketone metabolism.
Here is how the researchers described the two-way relationship between cancer cells and surrounding fibroblasts that increases the amount of ketones available to fuel tumour growth:
"Ketogenic fibroblasts promote the growth of adjacent breast cancer cells, by driving increased mitochondrial biogenesis. Thus, the tumor stroma may serve as a reservoir for ketone body production, while cancer cells upregulate the enzymes required for ketone body re-utilization, driving oxidative mitochondrial metabolism (OXPHOS) in epithelial cancer cells.”
“Our data provide the necessary genetic evidence that ketone body production and re-utilization drive tumor progression and metastasis… In summary, ketone bodies behave as onco-metabolites, and we directly show that the [ketolytic] enzymes HMGCS2, ACAT1/2 and OXCT1/2 are bona fide metabolic oncogenes [genes with the potential to cause cancer]”.
An enzyme involved in the generation of ketone bodies was found to be upregulated in high grade prostate cancer, and the concentration of the ketone body β-hydroxybutyrate was higher in these cells, indicating that aggressive prostate cancer cells use ketone bodies to “gain a survival advantage allowing them to become increasingly aggressive and gain androgen-independent properties.”
This is a critical point, because conventional prostate cancer therapy includes androgen blockade, but androgen deprivation decreases the ability of prostate cancer cells to take up glucose from the bloodstream and use it as a fuel. This “decrease in the activity of the glycolytic pathway places prostate cancer cells under stress to generate energy in a quick manner in order to carry out necessary cellular functions. One avenue through which such an effect can be achieved is to increase energy production through the breakdown of fatty acids via the β-oxidation pathway” – that is, through utilising ketone bodies.
18:18 "We as a species evolved to be in nutritional ketosis for the majority of our existence as a species, like one and a half million years, for centuries and centuries, thousands and thousands of years, our species, you and me, our ancestors were always in a state of nutritional ketosis because there was very few carbohydrates in the environment for them to be consuming, right?"
No, not "right". In fact, every single claim in this sentence is dead wrong. For starters, the fossil record indicates that our species - Homo sapiens - has walked this Earth for only around 300,000 years. One and a half million years ago, the precursor species to our lineage, including Homo antecessor and Homo ergaster, were eatingtubers (the carbohydrate-rich underground storage organs of plants) along with meat. Neither our ancestors nor the ancestors of any other extant species "were always in a state of nutritional ketosis". As I wrote in Ketogenic diets: Part 2 – Is ketosis ‘natural’?:
"No animal on earth lives permanently in ketosis. Omnivorous animals such as bears and dogs, and obligate carnivores such as cats – the ultimate low-carbers – use gluconeogenesis to transform amino acids from protein into glucose, in order to maintain an optimal blood glucose level to supply their bodies’ needs for this vital nutrient. Only in prolonged starvation or a diabetic state will these animals enter ketosis.
Even hibernating bears do not go into ketosis. And predatory animals who undergo extended periods of food deprivation, such as elephant seals, are metabolically resistant to going into ketosis; instead, they have upregulated gluconeogenesis pathways which allow them to maintain steady production of glucose.
This makes perfect sense, since predators’ survival depends on their capacity to catch prey animals, which usually requires intense bursts of activity. Sprinting capacity is dependent on glucose, as humans who adopt a ketogenic diet quickly discover.
“A high fat, low carbohydrate ketogenic diet may impair exercise performance via reducing the capacity to utilize carbohydrate, which forms a key fuel source for skeletal muscle during intense endurance-type exercise.”
Our human ancestors were not eating high fat, low carbohydrate diets and therefore could not have been in diet-induced ketosis. African wildlife such as wildebeest, warthog and impala all have low body fat – well under 10%, and as low as 0.3% in the dry season – so even the most successful early hunters could not possibly have consumed enough fat to enter diet-induced ketosis.
And in fact, humans develop a condition dubbed “rabbit starvation” when they eat a diet that is low in fat and carbohydrate, and high in protein – so our early ancestors would have to have been eating ample carbohydrate-rich plant foods to prevent protein toxicity induced by eating such low-fat meat.
“But what about the Inuit?”
The Inuit (Eskimo) peoples inhabiting the Arctic regions of Greenland, Canada and Alaska are frequently cited by ketogenic diet advocates as an example of a human population adapted to eating a high fat, low carbohydrate and relatively low protein diet. Here’s a typical example:
Yet as far back as 1928, researchers conducted experiments on Inuit people who were still eating their traditional diet (comprising on average 280 g of protein, 135 g of fat, and 54 g of carbohydrate per day, the latter derived primarily from muscle glycogen from raw meat) which established two important facts:
Even in the fasting state, Inuit people showed resistance to entering ketosis: “On fasting he develops a ketosis, but only of mild degree compared to that observed with other human subjects.”"
Inuit people were not in ketosis on their regular diet; instead, their high protein intake resulted in gluconeogenesis – just like carnivores and omnivores.
18:45 "If we stopped eating and we took a low carbohydrate diet and just did water only fasting, we would get into nutritional ketosis where the normal cells - our brain, our kidneys, our heart - can be burning these ketone bodies because they have a good mitochondria and they can burn these fuels effectively. The tumour cells have a bad mitochondria; they can't burn those fuels - they're dependent on glucose and glutamine. We can replace glucose and glutamine with ketone bodies in the normal cells of our... so we selectively marginalise these tumour cells slowly over time, they slowly start to die; the blood vessels disappear and the body comes in and dissolves them."
Here, Seyfried conflates two distinctly different physiological states:
The state of ketosis that naturally occurs when energy (calorie/kilojoule) intake either ceases entirely, or drops below a critical threshold (as occurs during, for example, Buchinger fasting which provides up to 250 calories per day from fruit juice, vegetable soup, and honey); and
Nutritional ketosis, which is induced by eating a diet that contains up to 90 per cent fat (by calories) and limiting carbohydrate intake to 20-50 g per day.
But fasting-induced ketosis and nutritional ketosis are not the same. The ketone bodies that are utilised for energy production during fasting come from one's own body fat, whilst in nutritional ketosis, the bulk of ketone bodies are synthesised from the fat one eats - unless one is in a calorie deficit. Fasting induces higher blood levels of ketone bodies than the ketogenic diet, and is more effective at promoting autophagy (the process of removing and recycling damaged components of cells, which promotes cellular rejuvenation).
Seyfried specifically mentions the kidneys as organs that can switch to burning ketones as a fuel. But while the cells that comprise the cortex (outer layer) of the kidney can utilise ketone bodies, the renal medulla (inner part of the kidney) cannot, and is heavily reliant on glucose, which is primarily metabolised via glycolysis because of the low oxygen availability in this part of the kidney.
And, as previously discussed, Seyfried's contentions that the mitochondria of cancer cells are "bad", that cancer cells are "dependent on glucose and glutamine", and that replacing glucose and glutamine with ketone bodies will cause tumour cells to die, are quite simply, flat-out wrong. To recapitulate the key points discussed at length above, cancer cells do have functional mitochondria, and many if not most of them can make use of ketone bodies as a fuel.
19:52 "Our brains are addicted to glucose; it's like cocaine and nicotine and whatever."
This is, quite frankly, such an utterly ludicrous statement that it’s hard to imagine that it could have come out of the mouth of a biologist - that is, a scientist who specialises in the study of living things, including their metabolism. Comparing glucose - the preferred fuel of brain cells, without a steady supply of which, we could never have developed our large and energy-expensive brains - to drugs of addiction betrays a breathtaking ignorance of human evolutionary history and biochemistry.
21:01: "You can get more energy bang for buck burning a ketone body than you can burning a pyruvate coming from glucose or even a fatty acid."
It's true that acetoacetate and beta-hydroxybutyrate, the two ketone bodies used by the body for energy, yield more energy than pyruvate: 22 ATP molecules for ketone bodies, vs 15 for pyruvate. But a single palmitic acid - a common 16-carbon fatty acid - produces 106 ATP molecules when completely beta-oxidised within a mitochondrion. So ketone bodies do not give you "more energy bang for buck" than fatty acids. Again, I'm not trying to be picky here, but it took me roughly three minutes of searching to disprove this statement, and I'm not a biology professor.
21:31 "It's hard to determine [the prevalence of cancer in our ancestors] from skeletal records."
Actually, it's not that hard, as I discussed in last week's post, and as I pointed out,
"An exhaustive paleopathological study of human remains from ancient Egypt and fifteenth to nineteenth century AD Germany concluded that, once adjusted for age and sex distribution, rates of metastatic cancer (which leaves distinctive marks in bones) in these populations were no different to those found in a well-defined English study population from AD 1901 to 1905. In other words, if humans live long enough, a good proportion of them will develop cancer, and presumably, eventually die from it."
21:42 "I think we can look at modern man who live according to their traditional ways, you know Albert Schweitzer, the great humanitarian physician went to Africa and looked at Africans that were living according to the traditional ways, and he said one of the weirdest things, they don't have cancer it was like, 'What, what?' Cancer was extremely rare in Africans in... [undecipherable]... living in the areas... British when they came you know, in looking at the health conditions of folks that lived in the Arctic Circle, cancer was not there. They had other things, but they didn't have cancer."
It is certainly true that cancer was rare in the Africans whom Schweitzer encountered - just 196 histologically-proven malignancies for 64,297 persons seen between 1950 and 1965 at the hospital of Doctor Albert Schweitzer, Lambaréné, Gabon. But another physician from Europe, Dr Denis Burkitt, was working in Uganda, in eastern Africa, at around the same time, and is remembered chiefly for identifying the cause of a deadly endemic lymphoma associated with Epstein-Barr virus infection, named Burkitt’s lymphoma in his honour. In contrast with the high prevalence of Burkitt's lymphoma, Burkitt noted the low incidence of colorectal cancer in Africans eating their traditional diet. This traditional diet was extremely high in dietary fibre due to heavy consumption of plant foods rich in unrefined carbohydrate (i.e. the polar opposite of the low carbohydrate, high fat ketogenic diet that Seyfried keeps insisting - against all evidence - was the ancestral diet of human beings).
Seyfried reiterates the same claims that cancer is driven by loss of mitochondrial function and the metabolic switch to "fermentation" (aerobic glycolysis), and that humans evolved on a ketogenic diet, multiple times throughout the interview, so I've skipped over these repetitions in my analysis of this interview. Frankly, I find it astonishing that a professor of biology could come out with such statements, given the lack of evidential support for them and the amount of evidence against them. I expect academics to deal with facts, not just-so stories that one would expect to come out of the mouth of a completely unqualified 'health influencer'. I also skipped over his critique of standard medical, surgical and radiation therapies for cancer, with which I largely concur. I'll pick up in Part 2 of this series with Seyfried's dietary recommendations for cancer prevention and treatment. Until then, I'll leave you with my best advice: question everything you see, hear and read!
And finally, this post has taken me approximately 20 hours to research and write. It takes much more time to debunk statements that aren’t true, than it does to spout them! I make all my posts freely available to all readers, because I believe we all deserve access to information that helps us to take greater control over our health. But I rely on my small core of paid subscribers to provide this service to those who genuinely can’t afford it.
Empowered! is a reader-supported publication. Without the generous support of my paid subscribers, I could not continue to bring you this important information.
For information on my private practice, please visit Empower Total Health. I am a Certified Lifestyle Medicine Practitioner, with an ND, GDCouns, BHSc(Hons) and Fellowship of the Australasian Society of Lifestyle Medicine.
There's an excellent summary of the four-step process of cellular respiration, of which glycolysis is the first step and oxidative phosphorylation the final, here.
Robyn, Cannot thank you enough for providing such careful research and healthful information.
Very interesting, thanks again Robyn. Your Substack again proving it’s easily worth a full paid subscription!
What evidence is there, if any, of the benefits of the fast-induced autophagy on cancer?