When New York City medical oncologist Vicky Makker meets a patient with endometrial cancer that has spread or recurred, she knows the outlook isn’t good. Even after radiation and drug treatments, most women with advanced disease die within 5 years.
But this spring, Makker is helping launch two clinical trials she hopes will change the picture. The drug patients will receive, called a phosphatidylinositol 3-kinase (PI3K) inhibitor, has already failed in multiple cancer trials. But the new studies are taking an unconventional tack to resurrect the drug: putting patients on a ketogenic diet, a low-carbohydrate regimen that typically involves loads of meat, cheese, eggs, and vegetables. The researchers hope the diet will render tumors more vulnerable to the drug, which blocks a growth-promoting pathway in cells. “It’s very outside of the mainstream thinking,” says Makker, a researcher at Memorial Sloan Kettering Cancer Center.
The trials are the brainchild of cell metabolism researcher Lewis Cantley of Weill Cornell Medicine (WCM). Decades ago, he discovered the PI3K signaling pathway, which the drugs aim to target. More recently, his lab showed in mice that a ketogenic diet can counter tumors’ resistance to those drugs.
Cantley isn’t the first to suggest that a particular diet, such as fasting or selectively reducing certain nutrients, can make cancer treatments work better. For at least a century, doctors and self-styled nutrition experts have touted the idea in bestselling books and, more recently, on popular websites. “There’s a big industry there, but it’s not based on a real understanding of what’s going on in a tumor cell,” says cancer biologist Karen Vousden of the Francis Crick Institute in London. Still, some early clinical trials showed hints of an effect. Now, studies from high-profile labs are spawning a new wave of trials with more rigorous underpinnings.
Scientists including Vousden, who cofounded a company with Cantley to test diet-drug combinations in cancer trials, are unraveling the molecular pathways by which slashing calories or removing a dietary component can bolster the effects of drugs. In mice with cancer, “the effects are oftentimes on the same order of magnitude as those from the drugs that we give patients. That’s a powerful thing to think about,” says physician-scientist Matthew Vander Heiden of the Koch Institute for Integrative Cancer Research at the Massachusetts Institute of Technology (MIT) and the Dana-Farber Cancer Institute. And the idea appeals to patients, he adds. “Diet is something that people feel like they can control.”
STILL, COMPELLING RESULTS in patients will be needed to overcome some oncologists’ view of special diets as fringy alternative medicine. The doubts often focus on a pioneer in the field, biochemist Valter Longo of the University of Southern California and the Italian Foundation for Cancer Research’s Institute of Molecular Oncology, who has built a huge popular following with his fasting research. Critics worry the media attention encourages cancer patients to diet without adequate evidence. Longo agrees patients should not improvise and says fasting needs more clinical testing.
His labs in Los Angeles and Milan are full of hungry mice. Longo began his career studying caloric restriction, which can extend the life spans of diverse species and has been shown to reduce the incidence of cancer in rodents and monkeys. Because few people can stay on low-calorie diets in the long term, Longo shifted his focus to fasting, a treatment offered for various ailments as far back as ancient Greece. In two key papers in 2008 and 2012, his team reported that reducing nutrients in the medium used to grow cells in a dish protected normal cells from the toxic effects of chemotherapy drugs such as cyclophosphamide and doxorubicin, yet made cancer cells more likely to die. In mice with cancer, fasting—drinking only water for 2 or 3 days—helped the drugs curb tumor growth and boosted the animals’ survival.
Longo’s explanation is that fasting, which lowers levels of glucose in the blood, causes healthy cells to hunker down in a protective mode. But cancer cells need to keep growing, which puts them at risk of starvation. Fasting also reduces the body’s production of hormones, such as insulin, that can drive tumor growth. Both effects may make the cancer cells more susceptible to chemotherapy.
Hoping to make fasting easier on cancer patients, Longo’s team showed that merely limiting calories for a few days has similar effects on blood hormones and other biomarkers. A company Longo started in 2009, L-Nutra, supplies that “fasting-mimicking” diet for clinical trials: packets of crackers, soups, teas, and nut bars. The company also sells the meal kits online to the public, touting them as a way to combat aging. After facing criticism for profiting off a product that hasn’t been fully validated, Longo began to donate profits from his shares in the company to charity in 2017.
Animal studies supporting the benefits of fasting-mimicking diets in cancer are now plentiful. Last year, Longo’s team reported that restricting calories enhanced the effects of hormone therapies in mice with breast cancer. Another team reported a synergistic effect with immunotherapies, again in mice. “There are probably 100 papers on this, almost all positive,” Longo says.
In small, preliminary clinical trials, Longo’s team and others showed that the fasting-mimicking diet may protect against some side effects of chemotherapy. Carolina Sandoval of Pasadena, California, age 40, participated in one of Longo’s intermittent fasting trials for 2 months last fall during four cycles of chemotherapy to fight breast cancer. “It was really hard,” she says, especially when the chemo affected her taste buds. “I couldn’t bear the taste of some of the food,” and she lost weight, she says.
But Sandoval also says she thinks the diet spared her from some of the nausea and fatigue of chemotherapy and allowed her to avoid taking days off from her job teaching high school online. She hopes the fasting “put my good cells to sleep, and the chemotherapy was able to attack more of my cancer cells,” she says. “I would do it again.”
However, the first trial aiming to rigorously test whether a fasting-mimicking diet can make chemotherapy work better faltered, partly because participants found the diet unappetizing. The study, launched in 2014 and led by oncologist Judith Kroep of Leiden University Medical Center, monitored 131 Dutch women with early stage breast cancer who were slated to receive chemotherapy before surgery. They were randomly assigned to follow either Longo’s fasting-mimicking diet or a regular diet for 4 days leading up to each round of chemo. But many women disliked the taste and lack of choices in the diet, and just 20% completed all eight cycles. In part because of the dropout rate, Kroep’s team couldn’t go on to compare biomarkers predicting overall survival in the two groups.
Yet the trial did yield hints that chemotherapy was more potent and less toxic to healthy cells in women who completed at least two cycles of the fasting-mimicking diet. Scans showed their tumors were more likely to shrink, and immune cells in their blood had less DNA damage from chemotherapy, Kroep, Longo, and colleagues reported last year in Nature Communications. Longo calls that “remarkable evidence” that the diet worked. But without definitive evidence that dieters were more likely to survive longer, other researchers found the trial inconclusive.
And when Kroep and a patient from the trial were featured on a Dutch TV program in late 2019, an uproar ensued from physicians and dietitians worried women with cancer would fast on their own. Medical groups and Kroep’s own institute released cautionary statements. “I agree that confirmation is needed before we can advise patients to fast, also because it is not always easy,” says Kroep, who is planning a new trial with changes to make the diet more appealing.
Longo hopes for a bigger test. His team has applied to the U.S. National Cancer Institute for a $12 million grant to run a 460-patient clinical trial at 11 hospitals of a fasting-mimicking diet and chemotherapy for breast cancer. The agency includes whether fasting works on a list of “provocative questions” in cancer. “If it happens, it’s going to be very exciting,” Longo says.
PATIENTS WHO BALK at cutting calories may have an easier time with the popular low-carb ketogenic diet that, like fasting, lowers glucose and hormone levels in blood. “Generally speaking, the ketogenic diet and fasting are two roads to a similar metabolic state,” says Princeton University biochemist Joshua Rabinowitz. A ketogenic diet also forces the liver to turn excess fat into molecules called ketone bodies that glucose-craving cancer cells struggle to burn for energy, some scientists suggest. The approach has been used to treat epilepsy since the 1920s, when researchers discovered the seizure-reducing effects of the ketogenic diet on brain metabolism.
Animal studies going back a decade suggest a ketogenic diet can enhance the effects of chemotherapy and radiation. Case reports and some small clinical trials hint that the diet may extend the lives of cancer patients—particularly those with the brain cancer glioblastoma, which tends to use large amounts of glucose.
Cantley turned to a ketogenic diet because of a major disappointment: Drugs based on his discovery of the PI3K pathway that drives growth in many tumors largely flopped in trials in the 2010s. Except for blood and breast cancer studies that led to drug approvals, the trials were a disaster, Cantley says. “Billions of dollars went into efforts that failed.”
He thinks he knows why. The drugs cause a side effect—a rise in blood sugar—that doctors often treat with insulin. But insulin stimulates the PI3K pathway in tumors and cancels out the cancer drug’s effects. In a study published in 2018, Cantley’s team fed a ketogenic diet designed to lower the body’s natural insulin production to cancer-afflicted mice receiving a PI3K drug. The researchers found that the diet allowed the drug to keep working and curbed tumor growth.
The two trials Makker is co-leading will soon test whether that hypothesis holds up in people with endometrial and some other cancers bearing a mutation in one of two genes that rev up the PI3K pathway. In one trial, participants will shop for and prepare meals according to instructions. In the other, the company Faeth Therapeutics that Cantley co-founded will ship meals to patients to help them stay on track.
Memorial Sloan Kettering and WCM are already running a feasibility test, supplying 4 weeks of packaged meals to about 30 women with endometrial cancer awaiting surgery. “They like the food, which is really great to see,” Makker says. And blood tests showed a drop in insulin and other changes that “mirrored the mouse model,” says WCM endocrinologist Marcus Goncalves, a co-investigator for the study and the two new trials.
If those trials show the ketogenic diet helps curb tumor growth for a year or two longer than the PI3K inhibitor otherwise would, the diet “could become the standard of care,” Cantley says. “That will be what physicians will tell patients to do.”
A ketogenic diet may enhance other cancer treatments, too. Immunologist Laurence Zitvogel of the Gustave Roussy Institute in France recently studied mice with skin, kidney, or lung cancers receiving a drug known as a checkpoint inhibitor that helps the immune system’s T cells kill tumors. In animals on a ketogenic diet, the ketone bodies they produced boosted the T cells’ power, her team reported in January. Rabinowitz and collaborators have begun enrollment for a 40-person trial to see whether the diet can enhance the impact on pancreatic cancer of a chemotherapy cocktail.
But researchers warn that a ketogenic diet could backfire and fuel the growth of fat-loving cancers such as those of the breast and prostate and others with certain mutations. Cantley found the diet spurred tumor growth in mice with leukemia. In a recent study, researchers found that, contrary to prevailing thought, glioblastoma tumors can get around the dearth of glucose by feeding on ketone bodies. To safely harness a ketogenic diet as treatment, “you need to really understand how and where it works,” Vander Heiden says.
OTHER RESEARCHERS ARE exploring an even more precise dietary limitation: cutting out specific amino acids, best known as the building blocks of proteins but also key to many other metabolic processes. Vousden unexpectedly veered into that line of research while studying a cancer-preventing gene called p53. The protein it encodes can trigger cells that have DNA damage to self-destruct, stopping them from turning cancerous. The gene is mutated in many tumors, allowing unrestrained growth.
But in 2005, a U.S. lab reported a surprising finding: The intact p53 protein helps healthy cells survive when glucose is scarce, suggesting p53-mutated cancer cells are especially vulnerable to glucose limitation. Vousden wondered whether the protein also helps cells survive a shortage of other, less explored nutrients, such as amino acids—and whether the mutated p53 in cancer would make the cells less resilient.
To find out, her postdoc Oliver Maddocks methodically removed various amino acids from cancer cells’ culture medium. Many types of cancer cells grew more slowly when deprived of two related amino acids, serine and glycine, and deleting p53 ramped up that effect. The scientists then tested the effects of a serine- and glycine-free diet in mice. To some colleagues, doing so seemed “kind of a pointless experiment,” Maddocks says, because the molecules are nonessential amino acids, meaning the body can make them even if they are absent from the diet.
But the experiment wasn’t pointless. Several tumor types, it turned out, could not efficiently make serine on their own. Maddocks and Vousden reported in 2013 and 2017 that the special diet slowed cancer growth and extended the lives of mice implanted with colon cancer cells lacking p53 as well as in mice engineered to develop lymphoma or colon tumors. Cells need serine or glycine to make a compound that sops up DNA-damaging free radicals, and the deprivation made tumor cells more sensitive to that oxidative stress. Radiation and some chemotherapies kill cells by generating free radicals, so the results suggested the diet could prime tumors for those treatments.
[In mice], effects are oftentimes on the same order of magnitude as those from the drugs that we give patients. That’s a powerful thing to think about.Matthew Vander Heiden, Koch Institute for Integrative Cancer Research
Similar findings have emerged for other amino acids. Limiting the essential amino acid methionine appears to amplify the effects of radiation and chemotherapy in mice with colon cancer and sarcomas. And removing asparagine, an amino acid abundant in asparagus, from mouse diets curbed the spread of metastatic breast cancer, suggesting the diet could enhance drug treatments.
As with serine, depriving mice of one of those amino acids apparently disrupts metabolic cycles by which cancer cells respond to oxidative stress, synthesize DNA, and turn genes off and on. The idea isn’t totally new. Cancer drugs known as antifolates developed in the 1940s also “broadly affect pathways that are sensitive to the levels of nutrients in our diet,” says cancer biologist David Sabatini of MIT’s Whitehead Institute.
Removing a specific amino acid from a person’s diet won’t be easy. Patients will have to eliminate all protein-rich foods, including grains, meat, and beans, and drink a specially formulated shake lacking the relevant amino acid. But that approach is not unprecedented: People born with the metabolic disorder phenylketonuria, for whom the amino acid phenylalanine is a neurotoxin, stay healthy by getting amino acids from a phenylalanine-free drink.
The company Vousden and Cantley founded, Faeth (Welsh for nutrition), is gearing up to test amino acid–depleted diets in two clinical trials this year. Faeth, also cofounded by Maddocks, will combine chemotherapy with a shake lacking specific amino acids, delivered to participants’ homes alongside other meal components, such as salads. The researchers got support from private investors after failing to win research grants for their idea, says Maddocks, now at the University of Glasgow. “It’s quite out of the box.”
LOOMING OVER ALLthose trials is the question of whether patients will be able to stick to the diets. Many people struggle with the high fat content of a strict ketogenic diet, for example. “It’s like eating butter” all the time, Vander Heiden says. Some very sick patients may be too underweight to follow any diet that restricts their caloric intake. And people getting treatments such as PI3K inhibitors or hormone therapy for months or years may have to keep up a special diet for just as long—a daunting prospect.
If a diet turns out to work long term, the cost of delivering meals to patients could add up. But Maddocks notes that meals should still be cheaper than many cancer drugs.
A simpler alternative to diet changes might be drugs that have some of the same effects—for instance, an enzyme that blocks cells from synthesizing a specific amino acid. In Zitvogel’s mouse study, adding ketone bodies to the animals’ chow worked as well as a ketogenic diet at boosting immunotherapies; she now plans to compare the two in a clinical trial.
Another concern is that tumors may harbor some cells that resist the effects of special diets. Longo argues that’s unlikely with a fasting-mimicking diet because “it takes away so many things from the cancer that some of them are bound to be important.” But work led by Vousden and Maddocks showed that some tumors with a mutation in a gene called KRAS can get around a serine-free diet by ramping up their own serine synthesis. And a serine-free diet would have little effect on breast cancer cells that spread to the pancreas, which is naturally awash in that nutrient, Vander Heiden’s group reported. “Every tissue has its own metabolic flavor,” says bioengineer Christian Metallo of the University of California, San Diego.
Maddocks expects the cancer-diet field will take years to move from “piecemeal forays” to a clear understanding of each diet’s pros and cons. Establishing that a specific diet works well enough to become part of routine clinical care also will take time. But Zitvogel says fighting cancer with diet is no longer a fringe idea. The field is at the start of “a new era where people will really take diet seriously into account,” she says. “The time is ripe.”
Read the full announcement: