To Eat or Not to Eat: With Cancer Therapies, That Is the Question ▲ NCI Cancer Bulletin for July 10, 2012 - National Cancer Institute

NCI Cancer Bulletin for July 10, 2012 - National Cancer Institute

To Eat or Not to Eat: With Cancer Therapies, That Is the Question

When we are healthy, we tend to eat what we want, when we want, and without much thought about how our bodies process food and anything else we ingest. But what we eat and when we eat it can affect the way our bodies absorb and react to medications, sometimes to the extent of altering treatment outcomes.

Food intake, therefore, is an important variable when determining the optimal treatment for many diseases. And cancer researchers are now exploring whether manipulating food intake could help reduce the side effects of some treatments or make them more effective, as well as more cost-effective.
A Double Challenge to Cancer Cells

What we eat and when we eat it can affect the way the body absorbs and reacts to cancer treatments.

In 2008, the laboratory of Dr. Valter Longo, a professor of gerontology and biological science at the University of Southern California (USC), showed that fasting for 2 to 3 days protected normal cells in culture and mice with xenograft tumors from chemotherapy drugs without protecting cancer cells—an effect they called differential stress resistance.

Dr. Longo and oncologists from USC later published a study of 10 elderly cancer patients who voluntarily underwent short-term fasting before and/or after cytotoxic chemotherapy infusion. Patients reported fewer side effects, including fatigue, weakness, and gastrointestinal problems, when they fasted. However, some doctors still worried that fasting could also protect cancer cells, explained Dr. Longo, which would negate its use in cancer patients.

A recent study by the USC research team, published March 7 in Science Translational Medicine, addressed this concern by showing that, contrary to such fears, fasting renders cancer cells more sensitive to chemotherapy.

The researchers found that fasting conditions in cell culture and in mice caused normal and cancer cells to radically change their gene expression patterns—but in very different ways. Normal cells reduced the expression of genes associated with cell growth and division and diverted their energy to cellular maintenance pathways that protect normal cells from stressful conditions and repair stress-induced damage. In contrast, cancer cells reduced the expression of many protective genes, which made them more likely to die, explained Dr. Longo.

Fasting results in "more investment in a variety of systems that protect the [normal] cell," Dr. Longo said. This shift to maintenance (instead of growth) has an added benefit for normal cells: Nondividing cells that enter a maintenance mode are less likely to be damaged by chemotherapy drugs that target the process of cell division.

In contrast, cancer cells contain mutations that may hinder their ability to respond to starvation conditions by shifting their resources away from growth, as normal cells do. Fasting also deprives cancer cells of the glucose and other molecules they need to fuel their endless cell division. Therefore, fasting adds a second stressor on top of chemotherapy, forcing cancer cells to deal "with two extreme environments at once," explained Dr. Longo.

Fasting adds a second stressor on top of chemotherapy, forcing cancer cells to deal with two extreme environments at once.
—Dr. Valter Longo

 

 

This combination of stressors led to promising results in animal studies. In mice with implanted breast cancer cells, short-term fasting alone delayed tumor growth to the same extent as treatment with the drug cyclophosphamide. Fasting before administering the drug had a stronger effect: the tumors of fasting mice given cyclophosphamide grew to less than half the size of those in nonfasting mice. The researchers saw similar results in mice implanted with melanoma or glioma cells.

In mouse models of metastatic melanoma, breast cancer, and neuroblastoma, fasting combined with high-dose chemotherapy extended survival compared with high-dose chemotherapy without fasting. The combination also reduced the overall number of metastatic tumors. Moreover, 20 to 40 percent of fasting mice with neuroblastoma had a long-term remission, which was not observed in mice that received chemotherapy without fasting.

The USC team is now studying how fasting can reduce side effects in people receiving chemotherapy. Dr. Longo has helped design three ongoing early-phase clinical trials examining this question (at USC, the Mayo Clinic, and Leiden University in the Netherlands).

And a consortium of 12 hospitals in the United States and Europe is planning two trials, each with more than 800 patients, Dr. Longo noted. One trial will look at whether fasting can reduce chemotherapy side effects, and the other will look at whether fasting can influence both side effects and drug efficacy (as observed in mice).

According to a survey by the USC team, more than 70 percent of eligible patients would refuse a water-only fast, so the international trials will use a substitution diet called Chemolieve that the research team developed and commercially marketed under an NCI Small Business Innovation Research (SBIR) contract. The researchers designed the diet to provide a minimum amount of nutrients to cancer cells while providing nourishment to the patients, sparing them the discomfort of fasting.

Danger, But Also Opportunity
On the flip side of the coin, researchers at the University of Chicago are exploring whether the bioavailability of some oral cancer drugs—the amount of drug absorbed and used by the body—can be increased by taking the drugs with food.

For many oral drugs, whether a patient takes them with food is irrelevant. But some oral drugs have a clinically significant food effect, which means that taking them at the prescribed dose with food causes a substantial change in their bioavailability. If a food effect leads to a marked decrease in bioavailability, too little drug will reach the bloodstream. If a food effect leads to a large increase in bioavailability, patients taking the drug with food risk overdosing.

This second scenario is a concern for several oral cancer drugs, including nilotinib (Tasigna) for chronic myelogenous leukemia and lapatinib for advanced breast cancer. The risk of sudden cardiac death from taking nilotinib at its prescribed dose with food is so high that the manufacturer has included a boxed warning about the dangers and developed a corresponding risk evaluation and mitigation strategy.

Dr. Mark Ratain, professor of medicine at the University of Chicago, sees opportunity instead of danger in the food effect, as well as a major flaw in what has become the default strategy for oral cancer drug development.

Oral Cancer Drugs That Are More Potent When Taken with Food

Cancer Drug	Approximate Increase in AUC* When Taken with Food	Estimated Monthly Cost at the Prescribed Dose (2011)
Lapatinib	150%	$3,400
Nilotinib	100%	$8,800
Erlotinib	50%	$4,800
Pazopanib	100%	$6,000
Abiraterone	300%	$5,000

*AUC = area under the curve; a measurement used to estimate the bioavailability of drugs. (Data courtesy of Dr. Mark Ratain, University of Chicago)

For many noncancer drugs that have a greater bioavailability with food, that food effect has been exploited, explained Dr. Ratain. For example, drugs such as darunavir for HIV or telaprevir for hepatitis C are prescribed at lower doses to be taken with a meal.

In oncology, the opposite has happened. Discovery of a food effect has led to the development of a high prescribed dose to be given without food. "That's not convenient for patients" who may take these drugs for years and are otherwise healthy, said Dr. Ratain, such as patients with chronic myelogenous leukemia who achieve complete remission on nilotinib but who must continue taking the drug every day.

Testing oral oncology drugs that have a food effect at lower doses with food might substantially reduce side effects and costs, suggests Dr. Ratain. His research group is testing this concept in a phase II clinical trial of abiraterone acetate (Zytiga), approved for metastatic prostate cancer. Dr. Ratain and his colleagues are testing whether men can safely reduce their dose of the drug by 75 percent by taking it with food. And, in turn, reducing the dose might cut drug costs, he added.

Some oral drugs have what scientists call a clinically significant food effect—taking them at the prescribed dose with food causes either a major increase or decrease in bioavailability.

 

The researchers are randomly assigning participants to one of two treatment groups: the approved dose of 1,000 mg without food or 250 mg taken with a low-fat breakfast. The reduction in prostate-specific antigen (PSA), variability in pharmacokinetics, and effects on the hormonal targets of the drug will be compared between the two groups.

"When one is starting with a drug and wants to study the pharmacokinetics in healthy volunteers, the cleanest thing to do is to study [the drug during] fasting. But just because fasting potentially provides less variability in dose between patients, that doesn't mean it's the best way to administer any given drug," said Dr. Ratain. "We're asking questions that I think the FDA should require companies [to answer]—what is the variability [in dose] over time, with food and with fasting?"

This is beginning to happen. The FDA's Center for Drug Evaluation and Research (CDER) now recommends to all pharmaceutical sponsors that "the impact of food intake on oral oncology drugs should be assessed early in drug development—during the pre-Investigational New Drug (IND) and phase I development periods," said Dr. Atiqur Rahman of CDER's Office of Clinical Pharmacology.
"Information obtained from these evaluations should be incorporated in the phase II and phase III development trials to guide dosing recommendations with regard to food intake," he continued. CDER also informs the sponsors that studying food effect in the late phase of drug development may be necessary if the formulation or dosage is significantly altered from the one tested during early clinical development.

But "whether a particular oral oncology drug can be allowed to be developed with food will depend on many factors, such as the magnitude and variability of the food effect, the therapeutic window of the drug, as well as the characteristics of the disease and the patient population," Dr. Rahman concluded.
—Sharon Reynolds