By Rick Weiss
Washington Post Staff Writer
Thursday, March 4, 1999; Page A2

Sophisticated X-ray studies conducted on a single hair may reveal whether a woman has breast cancer, and could ultimately help doctors diagnose other cancers, new research suggests.

The surprising link between breast cancer and microscopic changes in hair structure remains unexplained, and is "almost unbelievable," said Veronica James, who led the research at the University of New South Wales in Australia. Others cautioned that, although the test appears highly accurate, it has been tested on too few women to know how useful it will be.

If the test's value is confirmed, however, it could be done for a "few dollars" on hairs received by mail, and could be offered to women without access to mammograms, said Tom Irving of the Illinois Institute of Technology in Chicago. "Even if it is not reliable enough to work as a diagnostic, it may tell us something about how the disease progresses and how to treat it."

James, Irving and three colleagues found that hairs from breast cancer patients generate atypical X-ray images when they are exposed to high-intensity beams from a "third generation" synchrotron, a radiation-generating machine bigger than a football field. The cancer patients' hairs produced a ring-shaped shadow indicating a particular kind of breakdown in the highly ordered protein structure of hair.

The technique sometimes erred if women had permed or colored their hair within the past three months. But when James switched to pubic hair, the results were 100 percent accurate.

Combined results of scalp and pubic hair tests, published today in the journal Nature, were positive for all 23 women with breast cancer diagnosed. Only some of the women had undergone treatment, so the test was not responding to that. And none had been treated with chemotherapy, which is known to affect hair.

Inexplicably, the test was also positive for all five women who did not have cancer but who had both a family history of breast cancer and a mutation in the BRCA1 gene, which together confer very high odds of getting breast cancer.

It was also positive in one of eight healthy women with a family history of breast cancer but no BRCA1 mutation, and in three of 20 healthy women with no special risk factors. For those women, it is not known whether the results were "false positives," indications of high risk or evidence of early, undiscovered tumors.

"It's an interesting finding and the implications are very important," said Stuart Yuspa, a senior investigator and hair expert at the National Cancer Institute. Cancer patients often have high levels of cell-growth factors in their blood, which may alter hair structure, Yuspa said. And "it's conceivable," he said, that hormonal changes associated with cancer could affect hairs. "But at this point it seems preliminary and people should not jump to conclusions" about the test's value, he said.

The discovery was serendipitous, said James, who for years had studied skin changes in breast cancer patients. Before going to Japan, where one of the world's three large synchrotrons is located, James stopped at a hospital in England to pick up skin specimens that had been saved for her, only to learn that a custodian had thrown them out by mistake.

"I was halfway to Japan without a sample in my box, so I said, 'Could we take some hair?'‚" Using the same "X-ray diffraction" technique she had used on skin, James compared the molecular structure of cancer patients' hairs to those of hairs from healthy people. "Amazingly enough there were changes," she said.

Preliminary studies show other, unique abnormalities in hairs from people with other kinds of cancer, James said. And while more studies are needed, she said, she feels certain the technique will lead to an accurate, affordable cancer test.

"My reason for believing this is purely woman," James said. "Women know about hair that doesn't behave."