License to Build: New Theory of Cancer Puts Metabolism at Center

By Matthew Tontonoz,

Tuesday, January 12, 2016

Montage of beer, reproducing cells, and building blocks

Scientists have known for 100 years that cancer cells metabolize nutrients in a unique way, though they haven’t understood why. In a new paper, MSK researchers reconsider the evidence and offer an unorthodox explanation, turning some commonsense wisdom on its head.

  • Normal cells in the body do not take up nutrients without the appropriate growth signals.
  • Cancer cells acquire mutations that allow them to take up nutrients autonomously.
  • This altered metabolism may lie at the heart of cancer.
  • Targeting cancer metabolism opens up new avenues for treatment.

Long before Louis Pasteur became famous for proving that diseases were caused by germs, he worked in a beer factory. His job: finding a way to make beer from sugar, hops, and yeast without having the yeast take over the vat, gunking up the beer.

He failed.

Turns out yeast are very good at converting sugar into more yeast, and nothing Pasteur did could change that — which is why today, most beer is filtered.

This long-familiar fact about beer making is inspiring some unconventional thinking about cancer. In a paper published today in Cell Metabolism, Memorial Sloan Kettering President and CEO Craig Thompson and postdoctoral fellow Natasha Pavlova argue that cancer cells take up and use nutrients much like yeast in a vat of sugar, reproducing with wild abandon. Further, they claim that it’s this altered metabolism of nutrients — rather than any quirk of a disordered cell cycle — that lies at the heart of cancer.  

“All of the information that drives the cell cycle — drives cell growth — comes from cells recognizing that they have adequate nutrients,” says Dr. Thompson.

If he’s right, then much of what we think we know about cancer is wrong.     

The Challenges of Living Together

For Dr. Thompson and his colleagues, the problem of cancer is intimately tied up with another biological question: how living things evolved from single-celled organisms, such as yeast, to multicellular organisms like fish, birds, and biologists.

“The fundamental thing that allows us to live as a collaborative multicellular organism — a society of cells — is that every cell agrees it will not take up and utilize the shared resources available to the body except on its instruction from other cells,” says Dr. Thompson.

In this view, cancer results when cells stop playing by the food rules. Through mutations, they develop the capacity to acquire nutrients autonomously. After that, it’s every cell for itself.

Through mutations, cancer cells develop the capacity to acquire nutrients autonomously.

The easiest way to see this, Dr. Thompson says, is to consider what happens when a person gets a yeast infection in the blood, a common problem on hospital wards where people are often immunocompromised. The yeast find their way from the bloodstream to the liver — a nutrient smorgasbord — and grow out of control.

“Metastatic cancers do the same thing,” he explains. “They act just like yeast, and that’s why the liver is the most common site of metastasis.”

Back to top

Revisiting Cancer’s Hallmarks

Craig Thompson and Natasha Pavlova.

Drs. Thompson and Pavlova named their paper “The Emerging Hallmarks of Cancer Metabolism” in homage to a famous article called “The Hallmarks of Cancer,” published in 2000 by Douglas Hanahan of UCSF and Robert Weinberg of MIT.

In that article — required reading for every cancer biologist — the authors describe six features, or hallmarks, of cancer that they believe characterize the disease, including things like activated growth factor signaling and evasion of cell death. Not included on the list is anything having to do with altered metabolism, the subject of the new paper.

Metabolism did later receive attention from these authors, but even then it was accorded a secondary status.

“The repeated refrain from traditional biochemists is that altered metabolism is merely an indirect phenomenon in cancer, a secondary effect of cancer cells gaining signals to survive and proliferate,” Dr. Thompson says. “The heretical part of our argument is that we’ve gotten all of this wrong.”

In support of their view, Drs. Thompson and Pavlova point out that many cancer-causing genes, or oncogenes, have direct effects on metabolism. For example, the oncogenes AKT and RAS directly increase glucose consumption by cells. Other cancer-associated genes, such as MYC and Rb, boost uptake of amino acids, another important nutrient.

Oncogenes do not simply transmit instructions for a cell to survive, grow, and divide, but in fact directly control nutrient uptake and utilization.
Natasha Pavlova
Natasha Pavlova cancer biologist

“It is becoming increasingly clear that oncogenes do not simply transmit instructions for a cell to survive, grow, and divide, but in fact directly control nutrient uptake and utilization,” says Dr. Pavlova, a postdoctoral fellow in the Thompson lab and the paper’s first author.

This “metabolic reprogramming” has long-term consequences for the cell, she says, transforming it from one that obeys the instructions of its neighbors to one that seeks only to feed and reproduce itself. 

Back to top

Rediscovering Metabolism

Kidney cancer cells growing in culture.

The new theory helps to make sense of a long-standing cancer conundrum. Scientists have known since the 1920s that cancer cells take up much more glucose than normal cells (which, incidentally, is the basis for PET scans). However, cancer cells don’t process this sugar in the same way normal cells do.

Typically, cells burn sugar in their mitochondria to make ATP, which they use to power chemical reactions. (Recall from your high school biology class that mitochondria are the cell’s “power plants.”) Cancer cells, on the other hand, use the less efficient process of glycolysis, which produces fewer molecules of ATP and leaves much of the glucose energy untapped. (Yeast do this too; it’s called fermentation.)

This fact has perplexed biologists over the years, since cancer cells presumably have a great need for energy. “For a hundred years, we’ve thought that the whole problem of life was gaining a battery supply,” Dr. Thompson says. “We weren’t thinking about the problem of how you actually make two of something.”

For a hundred years, we've thought that the whole problem of life was gaining a battery supply.
Craig B. Thompson
Craig B. Thompson cancer biologist

To make two (and four and eight) of themselves, cancer cells need more than just energy; they need building materials — sugar to construct DNA and lipids to make membranes, for example. Shunting nutrients from the power plant to the assembly line, therefore, makes perfect sense.

The effects of cancer metabolism do not end there. Metabolic products also affect gene expression, which shapes a cell’s identity. Cancer cells even pump out metabolic products that trick neighboring normal cells into helping them proliferate and spread.

Back to top

New Treatment Options

If correct, Drs. Thompson and Pavlova’s new theory suggests that targeting altered metabolism could be an effective approach to cancer treatment. Drugs not previously thought of in the context of cancer — such as the diabetes drug metformin — might become useful in the fight against the disease. We might even find that many existing cancer drugs, including certain targeted therapies, achieve their effects by affecting a cell’s access to nutrients, in contrast to current theories of how they work.

The study of cancer metabolism is more than a century old, but these recent discoveries are revitalizing interest in the subject. Asked if he thinks the time will come when altered metabolism will be seen as a full-fledged cancer hallmark, Dr. Thompson is bullish: “I actually don’t think it’s a hallmark of cancer,” he says. “I think it’s the root cause.”


Back to top

Craig Thompson is supported by grants from the NIH and NCI, and is a founder of Agios Pharmaceuticals. Natasha Pavlova is supported by the Terri Brodeur Breast Cancer Foundation.


Can one extrapolate then if one has cancer they should be on a low glycemic index and low fat diet ? Or might fasting be beneficial ? Or is it too soon to tel ? Thank you .

Dear Edward, thanks for your comment. The short answer is "no," ­ switching to a low-glycemic index or low-fat diet or fasting is not something that can be recommended based on this information. Here's a more detailed response from Dr. Pavlova:

"Our bodies vigilantly monitor and correct the level of glucose in the blood, maintaining it within a very narrow range, more or less regardless of diet (except for pre-diabetics/diabetic patients)." [This means that altering your diet to reduce sugar, for example, will not significantly change the level of glucose in your blood.] "Excess sugar in diet may increase the risk of cancer in the long run by contributing to the development of obesity and metabolic syndrome, but the exact causative links are still being investigated and may include chronic systemic inflammation associated with obesity, changes in intestinal microbial composition, increased levels of insulin and other growth factors ­in addition to inefficient glucose control. But high sugar consumption per se would not cause our cells to divide improperly and grow into a tumor."

That said, eating a well-balanced diet that is low in added sugar and fat is recommended for health regardless of cancer status. You can read more about nutrition guidelines here:…

I am presently a patient at MSK. A two year breast cancer survivior , I was diagnosed with diabetes last April. I am currently on Metformin and have a high ESR which I was told is inflammation in my body. I suffer from lower back and hip pain. I have had every scan possible. Thank God no mets. With this new discovery I'm concerned that between diabetes and inflammation in my body I am more likely to get mets. Is there anything I can do or a specialist at MSK who could help ? Should I get I touch with my oncologist for additional recommendations due to this theory?

Dear Lisa, we are sorry to hear about your health issues. You may be interested in learning more about how diabetes may affect your cancer care here: If you would like to make an appointment with one of our specialists to address your questions, please call our Physician Referral Service at 800-225-2225. Thank you for reaching out to us.

A relative is in hospice with terminal colon cancer. Any suggestions on last resort treatment?

Dear Carol, we are sorry to hear about your relative. Hospice care provides comfort and support at the end of life. It also helps to manage a person's pain and other symptoms. When these symptoms are controlled, the patient can live each day as fully as possible. Your loved one's doctor may have suggested hospice care because his or her cancer stopped responding to treatment. If he or she has questions about whether there are any additional therapies that may help, please have your relative follow up with his or her oncologist. It is important to understand that hospice care does not speed up or delay death. It provides comfort and support during this time. Thank you for reaching out to us.

Though I'm not a doctor, I am a patient at MSK. I'm under the impression that there are as many different types of cancer cells as types of cancer. Does this effect the metabolic mutation of the cancer cells? Meaning are all cancer cells thriving on sugar in the same fashion? (I realize this paper is new and there's more investigation ahead.) thank you so much and thank you for the never ending dedication to cancer research!

Dear Rose,
Thank you for your comment. We forwarded your question to Dr. Pavlova and this is her response:

“It is indeed true that different types of tumors display different metabolic alterations. While many tumor types display extreme avidity for glucose - which is evident by the broad utility of radiolabeled glucose as an imaging agent in PET scans - some types of tumors (for instance, prostate tumors) are generally not amenable for imaging with radiolabled glucose, because they do not consume much of it at all. Furthermore, while some tumors, such as Myc-oncogene driven liver tumors consume large quantities of glutamine, lung tumors driven by the same oncogene actually accumulate glutamine. Finally, pancreatic tumors were recently shown by our group and others to engulf and digest entire proteins from around them, and derive amino acids this way - something that normal tissues, and other tumor types, lack the capacity of doing. This is a very good question, and is something we are just beginning to understand in depth.”

So this article is statung that too much glucose and sugar can be a cancer causing agent in all of us? So diabetics are at higher risk of developing cancer?

Any thoughts on a ketogenic diet (high in good fat, moderate protein, and low carbs) used with conventional treatment to starve cancer cells?

Dear Marcia,
Thanks for your comment. We forwarded your question to our experts here at MSK and this was their response:
“There is currently not enough evidence to recommend ketogenic diets (or any other specific diet) as a mechanism for cancer prevention or treatment. However, there is evidence to support the health benefits (including cancer prevention and cancer outcome) of avoiding metabolic syndrome, diabetes and/or obesity via healthy lifestyle choices.

Patients with advanced cancer may face a challenging situation in terms of adequate caloric intake and, thus, restrictive diets are often not advisable. At this point, it is reasonable for cancer patients to follow a healthy diet with a variety in nutrient composition, and for those not in a state of caloric deficit to avoid foods with excess added sugars or refined foods with low nutrient content (“junk foods”).

Each patient’s nutritional needs are different, depending on his/her health conditions. There is no universal diet that is good for everyone. Patients should discuss their nutritional needs with their health care professionals.”

There are some doctors that have started that they believe Cancer cells are Candida Yeast cells run amok. Had this theory ever been tested ? Could Cancer be a mutation of Yeast rather than our own cells ?

Dear Charles,
Cancer cells are not yeast cells. Cancer cells are altered versions of our own cells, while yeast cells are a different organism. Cancer cells and yeast cells can be clearly distinguished based on genetics, morphology, and behavior. We hope this information answers your question.

Given the similarities to how yeast grows, is it reasonable to assume that systemic yeast overgrowth is somehow involved ? What about bio films?

Dear Cheryl,
Thank you for your comment. No, there is no reason to think that yeast is a cause of cancer. Sometimes people being treated for cancer can get infections because their immune systems are weakened, but yeast itself is not a cause of cancer.

Does the pH of the cancer cell have anything to do with the metabolism?

As a thyroid cancer patient, I've been told COUNTLESS times by alarmed friends that eating sugar will feed my cancer and cause it to grow more aggressively. I eat very well these days, but still consume a little sugar every day to avoid diet boredom. I have very low body fat and exercise every day. Do I no longer need to worry about a modest sugar intake?

Dear Dave,
We are unable to make specific dietary recommendations or give medical advice because each person's case is unique. You can find general dietary recommendations for cancer patients here:…

Because other readers have had similar questions about whether the information in this blog post warrants a change in diet (the general answer being "no"), we asked Dr. Pavlova to provide more details:

"What we argue in our paper is that we need to rethink how oncogenes work - specifically, that one of the primary effects of oncogenes (and the loss of tumor suppressor genes) is that they perpetuate signals that allow cancer cells to break "social conventions" of their respective tissues of origin, and begin consuming glucose and amino acids from blood and tissue fluid in an unregulated manner, fueling tumor cell proliferation. In conditions of a severely limited carbohydrate intake, the body will continue to maintain blood glucose at a steady level of 3.5-5 mmol/L via a combined effects of reduced insulin levels, physiological insulin resistance and upregulation of liver gluconeogenesis, so that enough glucose is available for the consumption by the glucose-dependent tissues (brain to a certain extent, erythrocytes etc) as well as for the anabolic needs of tissues. Coincidentally, even when dietary carbohydrates are reduced or eliminated, glucose will continue to be available for a tumor, which is already primed for its consumption by the combined action of oncogenes.

At the same time, dietary approaches that normalize fasting blood glucose levels and prevent or reduce obesity/metabolic syndrome/diabetes remains to be an unequivocally good advice for prevention of cancer, since these conditions are linked to the increased incidence of numerous types of cancer. It is important to keep in mind that the mechanisms behind these reduced risks are complex and most likely include reduction in insulin levels, reduction in inflammation, changes to the intestinal microbiota, among the many other factors."

We hope this additional information helps to clarify why changing one's diet to reduce or omit carbohydrates or sugar will not affect the ability of cancer cells to obtain nutrients, but why eating a balanced diet that prevents overweight and obesity is generally good advice for everyone.

Would a ketogenic diet make a more substantial impact on halting cancer? What of claims of reversing cancer with a raw foods diet etc? A candida diet? These all shift metabolic pathways, don't they?

Dear Susan,
Thanks for your comment. We forwarded your question to our experts here at MSK and this was their response:

“There is currently not enough evidence to recommend ketogenic diets (or any other specific diet) as a mechanism for cancer prevention or treatment. However, there is evidence to support the health benefits (including cancer prevention and cancer outcome) of avoiding metabolic syndrome, diabetes and/or obesity via healthy lifestyle choices.

Patients with advanced cancer may face a challenging situation in terms of adequate caloric intake and, thus, restrictive diets are often not advisable. At this point, it is reasonable for cancer patients to follow a healthy diet with a variety in nutrient composition, and for those not in a state of caloric deficit to avoid foods with excess added sugars or refined foods with low nutrient content (“junk foods”).

Each patient’s nutritional needs are different, depending on his/her health conditions. There is no universal diet that is good for everyone. Patients should discuss their nutritional needs with their health care professionals.”

Your Lab also found that mTor inhibitors might actually increase cell growth in some cancers. Based on these two results, is there a clue already which gene mutations will be "candidates" for metformin treatment and which will speed up cancer growth?

Dear Jane,
We forwarded your question to Dr. Wilhelm Palm, who had this to say: "The underlying hypothesis of this question is whether AMPK activity, by suppressing mTORC1, could promote cell growth that utilizes proteins as nutrients. This question is of high interest and we are currently initiating experiments to address this. Because the role of AMPK in extracellular protein-dependent growth remains to be established, it is too early to speculate whether this would cause some cancers to respond to metformin treatment with enhanced growth." We hope this answer is of some help to you!

A family member has had 2 surgeries for a stage IV glioblastoma removal from his brain in March and is completing a clinical trial study in NC this Feb. Was wondering what to do next besides watch and wait for regrowth in light of this new theory of cancer cell behavior. Any insights would be appreciated. Thank you for sharing your research!

Dear Maria, we are sorry to hear that one of your family members has been diagnosed with Stage IV glioblastoma and hope that he is responding to his treatment. We would encourage him to follow up with his oncologist regarding specific questions he may have about the next steps in his care plan. His physician is the most appropriate person to make those kinds of recommendations specific to his circumstances. Thank you for reaching out to us.

Are there any new trials or treatments for trout cances

Dear Debbie, assuming you mean tongue cancer, you can get more information by going to our Head and Neck Cancer page here:

You can also browse through our clinical trials evaluating new treatments for people with head and neck cancer here:

We hope this information is helpful. Thank you for reaching out to us.

Why cancer cells don't settle and grow inside the blood vessels where they have easy access to glucose and nutrients?

Thanks for being there.

Dear Joan, thanks for your comment. Some cancers do form in the blood — these are called leukemias — but most cancer cells do not remain in blood vessels because the force of blood flow pushes them along and they end up in other tissues. Keep in mind that it takes more than ready nutrients for cancers to form; it takes genetic mutations that alter the way the cells take up and use these nutrients (as well as other aspects of cell behavior). Abundant nutrients alone are not a cause of cancer. We hope this information helps.

Very interesting article

To follow up on Dr. Pavlova answer regarding low-carb diet, wouldn't reduction in insulin and IGF-1 (assuming it's a consequence of such a diet) slow cancer growth?

Dear Maria,
Thank you for your comment. The relationship between insulin and insulin-like growth factor (IGF-1) signaling and cancer development is an active and ongoing area of research. We know that obesity and diabetes (which can be associated with higher insulin levels) are risk factors for numerous cancers, but the exact reasons for this association are not yet fully known. The best we can say at the moment is “stay tuned.”

This is very interesting information. I actually was watching a video recently where Craig Thompson was talking about how carbs, not fat, increase the risk of cancer. The video was from 2010 but it doesn't seem like this information has filtered into the nutritional advice from Sloan Kettering. It's still about low fat, avoiding saturated fats. How come? The science says one thing and you agree with it but when it comes to executing it you don't seem to be able to do this. Your recommended diet is still basically high carb, low fat, which certainly doesn't fit into your carbohydrate increasing cancer risk theory.

Dear Elliot,
Thank you for your comment. The relationship between diet and cancer is an ongoing area of research. MSK's nutritional guidelines (such as those listed here:…) reflect the current weight of evidence. They stress the health benefits of eating a balanced diet that limits the amount of added sugars and saturated fats, and avoids obesity. Obesity is a known risk factor for cancer, and so eating a diet that helps you maintain a healthy weight is a reasonable goal.

Although carbohydrates are discussed in this article, it's important to note that the main point is about how cancer cells acquire mutations that enable them to use nutrients (like sugar and amino acids), not about the value or risk of eating carbohydrates per se. Carbohydrates--especially complex carbs that include fiber--are a normal part of a healthful diet.

Not to be critical of your observations, but isn't this very similar to Otto Warburg's theory of anaerobic respiration? Consequently reiterated by Thomas Seyfried and also by Travis Christofferson.

Dear Nicholas,
Thank you for your comment. You are right that Otto Warburg was one of the first to note that cancer cells take up sugar and metabolize it anaerobically (by glycolysis). However, we now know that Warburg was wrong in his interpretation about why they do this. Warburg thought that cancer cells had a defect in their mitochondria that prevented them from metabolizing glucose aerobically even when oxygen is present, and so needed another way to obtain energy. This is incorrect. Cancer cells can and do metabolize glucose aerobically using their mitochondria. However, shifting to glycolysis allows the cancer cells to obtain necessary building blocks they need to grow. So, this recent work by Dr. Thompson and others provides a new interpretation of Warburg's observations.

Are you actually doing treatment with immune therapy or metabolic therapy currently for breast cancer patients. My daughter was diagnosed yesterday with metastasized breast cancer in both lungs. She is 27yrs old and otherwise in incredible health. She went through hell with prior chemo. She has had radiation, hormonal therapy and surgery. She was HER2 positive, ER/pr positive. Unclear if mutated in new mass, biopsy set for Wed. Very interested in this research. New tumor is very large so cannot take risk of clinical trial - need actual treatment. Is it available someplace?

Hi Pamela, we're sorry to hear about your daughter's diagnosis. Immunotherapy is not yet part of the standard of care for breast cancer, but does MSK have several trials looking at whether it is an effective treatment. Participating in a clinical trial does not mean forgoing the current standard treatment, but that additional therapies may be used in addition to the standard of care.

If your daughter would like to make an appointment to learn more she can call 800-525-2225 or go to for more information on making an appointment.

Add new comment

We welcome your questions and comments. While we share many of them with our world-class doctors and researchers, we regret that in order to protect your privacy, we are not able to make personal medical recommendations on this forum, nor do we publish comments that contain your personal information. If you would like to consult with an MSK doctor, we encourage you to make an appointment at 800-525-2225 or request an appointment online.

Your email address is kept private and will not be shown publicly.