Cancer cells, because of their rapid growth, have a voracious appetite for glucose, the main nutrient used to generate energy. And tumors often use glucose differently from healthy cells, an observation first made by a German biochemist in the 1920s.

That observation is already used to detect tumors in the body using PET scans. A radioactive form of glucose is injected into the bloodstream and accumulates in tumors, lighting up the scans.

Now, efforts are turning from diagnosis to treating the disease by disrupting the special metabolism of cancer cells to deprive them of energy.

The main research strategy of the last decade has involved so-called targeted therapies, which interfere with genetic signals that act like accelerators, causing tumors to grow. But there tend to be redundant accelerators, so blocking only one with a drug is usually not enough.

In theory, however, depriving tumors of energy should render all the accelerators ineffective.

“The accelerators still need the fuel source,” said Dr. Chi Dang, a professor of medicine and oncology at Johns Hopkins University. Indeed, he said, recent discoveries show that the genetic growth signals often work by influencing cancer cells’ metabolism.

The efforts to exploit cancer’s sweet tooth are in their infancy, with few drugs in clinical trials. But interest is growing among pharmaceutical companies and academic researchers.

“Nutrient supply and deprivation is becoming potentially the next big wave,” said Dr. David Schenkein, chief executive of Agios Pharmaceuticals, a company formed two years ago to develop drugs that interfere with tumor metabolism. Among its founders was Dr. Craig B. Thompson, the new president of Memorial Sloan-Kettering Cancer Center in New York City.

Other small companies, like Cornerstone Pharmaceuticals and Myrexis, are pursuing the approach, and big drug companies are also jumping in. Earlier this year, AstraZeneca agreed to work with Cancer Research UK, a British charity, on drugs that interfere with cancer metabolism.

One factor spurring interest in cancer metabolism is the intriguing interplay between cancer and diabetes, a metabolic disease marked by high levels of blood glucose. The possible link between the two great scourges has garnered so much attention that the American Cancer Society and the American Diabetes Association jointly published a consensus statement this summer summarizing the evidence.

People with Type 2 diabetes tend to have a higher risk of getting certain cancers. And preliminary evidence suggests that metformin, the most widely used diabetes pill, might be effective in treating or preventing cancer.

It is still not clear if high blood glucose is the reason diabetics have a higher cancer risk. A more likely explanation is that people with Type 2 diabetes have high levels of insulin, a hormone that is known to promote growth of certain tumors, according to the consensus statement.

Similarly, metformin might fight cancer by lowering insulin levels, not blood sugar levels. But there is some evidence that the drug works in part by inhibiting glucose metabolism in cancer cells.

Even if blood sugar levels fuel tumor growth, however, experts say that trying to lower the body’s overall level of blood sugar — like by starving oneself — would probably not be effective. That is because, at least for people without diabetes, the body is very good at maintaining a certain blood glucose level despite fluctuations in diet.

“When a patient with cancer is calorically restricted, the amount of glucose in the blood until they are almost dead is close to normal,” said Dr. Michael Pollak, professor of medicine and oncology at McGill University in Montreal. Also, Dr. Pollak said, tumors are adept at extracting glucose from the blood. So even if glucose is scarce, he said, “the last surviving cell in the body would be the tumor cell.”

So efforts are focusing not on reducing the body’s overall glucose level but on interfering specifically with how tumors use glucose.

This gets to the Warburg effect, named after Otto Warburg, the German biochemist and Nobel Prize winner who first noticed the particular metabolism of tumors in the 1920s.

Most healthy cells primarily burn glucose in the presence of oxygen to generate ATP, a chemical that serves as a cell’s energy source. But when oxygen is low, glucose can be turned into energy by another process, called glycolysis, which produces lactic acid as a byproduct. Muscles undergoing strenuous exercise use glycolysis, with the resultant buildup of lactic acid.

What Dr. Warburg noticed was that tumors tended to use glycolysis even when oxygen was present. This is puzzling because glycolysis is far less efficient at creating ATP.

One theory is that cancer cells need raw materials to build new cells as much as they need ATP. And glycolysis can help provide those building blocks.

“You can have energy that turns on the lights in your house, but that energy can’t build anything,” said Matthew G. Vander Heiden, assistant professor of biology at the Massachusetts Institute of Technology.

Still, as with everything else about cancer, metabolism is complex. Not all tumor cells use glycolysis, and some normal cells do. So it could be challenging to develop drugs that can hurt tumors but not normal cells.

Two early efforts by a company called Threshold Pharmaceuticals to interfere with glucose metabolism did not work well in clinical trials. One of Threshold’s drugs, called 2DG, is the same form of glucose used in PET imaging, but without the radioactivity. Because of a slight chemical modification, this form of glucose cannot be metabolized by cells, so it accumulates.

But much less 2DG buildup is needed to spot a tumor on a scan than to destroy it by gumming up its works. Large amounts of the drug were needed because 2DG lasted only a short time in the body and because it had to compete with the natural glucose that is abundant in the bloodstream.

Efforts have not ended, however. Waldemar Priebe, a professor of medicinal chemistry at the M.D. Anderson Cancer Center, said he had developed a way to deliver up to 10 times as much 2DG to a tumor. It has been licensed to a startup called Intertech Bio.

The other Threshold drug, glufosfamide, consisted of glucose linked to a standard chemotherapy agent. The idea was that, as with the Trojan horse, the tumors would eagerly ingest the glucose only to then be poisoned.

In a late-stage clinical trial involving more than 300 patients with advanced pancreatic cancer, glufosfamide prolonged lives compared with no treatment, but not by a statistically significant amount.

A new company, Eleison Pharmaceuticals, plans to repeat the trial. Dr. Forrest Anthony, Eleison’s chief medical officer, said the original trial would have succeeded had it excluded 43 diabetics who were taking insulin, which is known to impede PET scanning for tumors. Insulin “sends glucose into skeletal muscle and fat tissue and away from the cancer,” he said.

Many other companies and scientists are trying to develop drugs that inhibit enzymes — for example, pyruvate kinase M2, involved in tumor metabolism.

Yet another approach is not to starve a tumor of energy but to give it more energy, and that is the idea behind a substance called dichloroacetate, or DCA. Dr. Evangelos Michelakis of the University of Alberta, who came up with the idea, says there is a mechanism by which cells that become defective can commit suicide for the greater good of the body.

But cancer cells usually do not kill themselves. Dr. Michelakis says this could be because they lack sufficient energy.

DCA, a simple chemical that is formed in small quantities when drinking water is chlorinated, has long been used to treat certain rare diseases in which lactic acid builds up in the body. DCA inhibits an enzyme called pyruvate dehydrogenase kinase. The effect of that inhibition is to move metabolism away from lactic acid-producing glycolysis and toward more normal oxidation of glucose in the mitochondria, the energy factories of the cell.

In 2007, Dr. Michelakis and colleagues published a paper showing that DCA, when put in drinking water, could slow the growth of human lung tumors implanted into rats. It seemed the DCA did not affect normal cells.

Some patients began clamoring for it. Within days, an amateur chemist had synthesized DCA and began offering it for sale. Some clinics still offer it. Dr. Michelakis cautioned that in high doses DCA can cause nerve damage and that it takes months for enough to build up in the body to have any effect.

This spring, in the journal Science Translational Medicine, Dr. Michelakis reported results of the first human testing of DCA, in five patients with glioblastoma multiforme, a deadly brain cancer. There was no control group, making it hard to judge the drug’s effectiveness, though some patients lived longer than might have been expected. There was evidence that the drug bolstered the activity of mitochondria and promoted cell suicide.

Since DCA is not a novel compound, it cannot be patented, making it unlikely a pharmaceutical company would pay for clinical trials. So Dr. Michelakis has been raising money from foundations and governments to conduct larger clinical trials.

“We have only assumptions and theoretical excitement,” Dr. Michelakis said. Still, he added, “there’s no question that this is a new direction that is logical and very appealing.”