Cancer Miracles

A cancer patient, given just months to live, stages a miraculous recovery. Doctors dismiss it as a fluke. Yet the mystery may offer crucial clues to fighting cancer.

Charles Burrows was given two months to live in 2005. Then, with no treatment, his liver tumor vanished. 'I won a lottery,' he says.

The immune system work is part of a new twist on the war on cancer. For decades cancer researchers have focused mostly on killing cancer cells with drugs and radiation, or removing them with surgery. But this is often impossible to accomplish. So scientists are studying the environment around tumors in order to invent drugs that will halt their spread. Such drugs, like Genentech (nyse: DNA - news - people )'s Avastin, would be the medical equivalent of cutting terrorist-cell supply lines or putting up security checkpoints to stop them from getting into vital areas.

One of the first to try to trigger the immune system to attack cancer was New York surgeon William Coley. He was inspired by a patient with sarcoma who recovered after suffering an acute bacterial infection. In the 1890s Coley started vaccinating other patients with killed bacteria. He claimed that his toxins spurred the immune system to destroy tumors in a minority of cases.

In the 1980s the natural immune protein interleukin-2 was touted as a breakthrough. But it turned out to help only a small minority of cancer patients and to sport an array of nasty side effects. Over the years numerous trials of anticancer vaccines designed to train the immune system to recognize cancer have shown mostly lackluster results. None of these new therapeutic vaccines is approved in the U.S.

But intriguing data suggest that the immune system can combat cancer sometimes. "To the body, a tumor looks like the biggest bacteria it has ever seen," says Robert Schreiber, an immunologist at Washington University School of Medicine in St. Louis. He has found that mice lacking key components of their immune system are far more likely to develop cancers. In one experiment 60% of mice missing something called the gamma interferon receptor on their cells got tumors after being exposed to a carcinogen, versus only 15% of normal mice.

Schreiber theorizes that many early cancers arising in the body are killed off by the immune system. Over time, however, some develop mutations that allow them to thwart the immune system, and a long stalemate ensues. Eventually some tumors escape the control of the immune system entirely.

Moreover, a 2006 study by the University of Paris Descartes' Wolf H. Fridman found that the number of certain kinds of white blood cells inside colon tumors is a stronger predictor of a relapse than criteria pathologists traditionally use, such as whether cancer cells have spread to the lymph nodes. He analyzed tumor samples from 415 people who had been operated on for early-stage colon cancer over the last two decades. Those with the highest number of these cells in their tumor rarely relapsed; those with few immune cells almost always did.

No broad-based anticancer antidotes have emerged from the immune system work yet. In trials to date, immune-boosting drugs have generally helped only a minority of patients, particularly those with melanoma. But when they do work, it can be spectacular.