steak
© Istockphoto
The women's hospital at the University of Würzburg used to be the biggest of its kind in Germany. Its former size is part of the historical burden it carries - countless women were involuntarily sterilized here when it stood in the geographical center of Nazi Germany.

Today, the capacity of the historical building overlooking the college town, where the baroque and mid-20th-century concrete stand in a jarring mix, has been downsized considerably. And the experiments within its walls are of a very different nature.

Since early 2007, Dr. Melanie Schmidt and biologist Ulrike Kämmerer, both at the Würzburg hospital, have been enrolling cancer patients in a Phase I clinical study of a most unexpected medication: fat. Their trial puts patients on a so-called ketogenic diet, which eliminates almost all carbohydrates, including sugar, and provides energy only from high-quality plant oils, such as hempseed and linseed oil, and protein from soy and animal products.

What sounds like yet another version of the Atkins craze is actually based on scientific evidence that dates back more than 80 years. In 1924, the German Nobel laureate Otto Warburg first published his observations of a common feature he saw in fast-growing tumors: unlike healthy cells, which generate energy by metabolizing sugar in their mitochondria, cancer cells appeared to fuel themselves exclusively through glycolysis, a less-efficient means of creating energy through the fermentation of sugar in the cytoplasm. Warburg believed that this metabolic switch was the primary cause of cancer, a theory that he strove, unsuccessfully, to establish until his death in 1970.

To the two researchers in Würzburg, the theoretical debate about what is now known as the Warburg effect - whether it is the primary cause of cancer or a mere metabolic side effect - is irrelevant. What they believe is that it can be therapeutically exploited. The theory is simple: If most aggressive cancers rely on the fermentation of sugar for growing and dividing, then take away the sugar and they should stop spreading. Meanwhile, normal body and brain cells should be able to handle the sugar starvation; they can switch to generating energy from fatty molecules called ketone bodies - the body's main source of energy on a fat-rich diet - an ability that some or most fast-growing and invasive cancers seem to lack.

The Würzburg trial, funded by the Otzberg, Germany - based diet food company Tavartis, which supplies the researchers with food packages, is still in its early, difficult stages. "One big problem we have," says Schmidt, sitting uncomfortably on a small, wooden chair in the crammed tea kitchen of Kämmerer's lab, "is that we are only allowed to enroll patients who have completely run out of all other therapeutic options." That means that most people in the study are faring very badly to begin with. All have exhausted traditional treatments, such as surgery, radiation and chemo, and even some alternative ones like hyperthermia and autohemotherapy. Patients in the study have pancreatic tumors and aggressive brain tumors called glioblastomas, among other cancers; participants are recruited primarily because their tumors show high glucose metabolism in PET scans.

Four of the patients were so ill, they died within the first week of the study. Others, says Schmidt, dropped out because they found it hard to stick to the no-sweets diet: "We didn't expect this to be such a big problem, but a considerable number of patients left the study because they were unable or unwilling to renounce soft drinks, chocolate and so on."

The good news is that for five patients who were able to endure three months of carb-free eating, the results were positive: the patients stayed alive, their physical condition stabilized or improved and their tumors slowed or stopped growing, or shrunk. These early findings have elicited "very positive reactions and an increased interest from colleagues," Kämmerer says, while cautioning that the results are preliminary and that the study was not designed to test efficacy, but to identify side effects and determine the safety of the diet-based approach. So far, it's impossible to predict whether it will really work. It is already evident that it doesn't always: two patients recently left the study because their tumors kept growing, even though they stuck to the diet.

Past studies, however, offer some hope. The first human experiments with the ketogenic diet were conducted in two children with brain cancer by Case Western Reserve oncologist Linda Nebeling, now with the National Cancer Institute. Both children responded well to the high-fat diet. When Nebeling last got in contact with the patients' parents in 2005, a decade after her study, one of the subjects was still alive and still on a high-fat diet. It would be scientifically unsound to draw general conclusions from her study, says Nebeling, but some experts, such as Boston College's Thomas Seyfried, say it's still a remarkable achievement. Seyfried has long called for clinical trials of low-carb, high-fat diets against cancer, and has been trying to push research in the field with animal studies: His results suggest that mice survive cancers, including brain cancer, much longer when put on high-fat diets, even longer when the diets are also calorie-restricted. "Clinical studies are highly warranted," he says, attributing the lack of human studies to the medical establishment, which he feels is single-minded in its approach to treatment, and opposition from the pharmaceutical industry, which doesn't stand to profit much from a dietetic treatment for cancer.

The tide appears to be shifting. A study similar to the trial in Würzburg is now under way in Amsterdam, and another, slated to begin in mid-October, is currently awaiting final approval by the ethics committee at the University Hospital in Tübingen, Germany. There, in the renowned old research institution in the German southwest, neuro-oncologist Dr. Johannes Rieger wants to enroll patients with glioblastoma and astrocytoma, aggressive brain cancers for which there are hardly any sustainable therapies. Cell culture and animal experiments suggest that these tumors should respond particularly well to low-carb, high-fat diets. And, usually, these patients are physically sound, since the cancer affects only the brain. "We hope, and we have reason to believe, that it will work," says Rieger.

Still, none of the researchers currently studying ketogenic diets, including Rieger, expects it to deliver anything close to a universal treatment for cancer. And none of them wants to create exaggerated hopes for a miracle cure in seriously ill patients, who may never benefit from the approach. But the recent findings are difficult to ignore. Robert Weinberg, a biology professor at MIT's Whitehead Institute who discovered the first human oncogene, has long been critical of therapeutic approaches based on the Warburg effect, and has certainly dismissed it as a primary cause of cancer. Nevertheless, he conceded, in an email, for tumors that have been affected by the ketogenic diet in animal models, "there might be some reason to go ahead with a Phase I clinical trial, especially for patients who have no other realistic therapeutic options."

Richard Friebe is executive editor of the German science magazine SZ Wissen.