Researchers at the University of Sydney report they have developed a special program that can be used along with magnetic resonance spectroscopy (MRS), a type of scan, to identify breast cancer that is more likely to spread to the lymph nodes.

They hope the technique might one day provide a diagnosis and prognosis before a woman goes into surgery. Patients with early breast cancer might avoid having numerous armpit lymph nodes removed for testing and women with truly benign lesions may be able to avoid surgery to rule out cancer, the Australian researchers said.

The program is a type of mathematical pattern recognition or statistical classification strategy (SCS) that organizes thousands of pieces of information to gauge whether or not breast tissue is malignant, and if so whether it is likely to spread to the lymph nodes.

Dr. Cynthia Lean, scientific director of the Institute for Magnetic Resonance Research, presented results of her research Friday at the Department of Defense Breast Cancer Research meeting here. In an interview with Reuters Health, lead investigator Lean said the ultimate goal of her team is to develop "a total program of breast cancer screening and diagnosis that is completely noninvasive."

Currently, she and her colleagues have collected cell samples from 400 breast cancer patients who participated in three separate breast cancer trials. The samples were from fine needle biopsies, in which cells are collected with a needle to test for cancer after a suspicious mammogram. Initially they used MRS to look at the specimens, and determine if malignant samples had "a different array on the spectrum" than benign specimens. Spectroscopy uses the spectrum of light to identify the chemical components of a substance, each of which absorbs different wavelengths of light.

"And the specimens did appear very different," Lean explained. The next step involved identifying approximately 4,000 pieces of chemical formation in each sample. "With those 4,000 pieces of information, we were able to develop the statistical classification strategy program." She said that she and her colleagues are developing similar recognition programs for "10 or 12 other cancers, but we have achieved the greatest success with breast cancer."

MRS analysis using the statistical classification strategy identified 94% of the cancerous samples, with 2% false positives. It is also useful in termining which tumors were likely to spread to lymph nodes. Lean said MRS may be most useful as an adjunctive test for women "who have inconclusive mammography and fine needle biopsy. As an additional piece of information, MRS could be used to confirm malignancy without the need for an open biopsy." Moreover, she added, it could be used to confirm that the cancer had spread, or metastasized, to the lymph nodes. While Lean and colleagues are currently using MRS to analyze biopsy samples, they hope to use MRS for diagnosing and determining prognosis in patients.

Dr. Kenneth A. Bertram, colonel in the US Army Medical Corps and director of the Congressionally Directed Medical Research Programs, said that MRS is appealing because it "has the potential to reduce the need for open biopsy."

Asked if it was likely that MRS could someday be used as a completely non-invasive technique for diagnosis, he said "the physics folks tell us this is possible. And although it is likely to be expensive, remember CT (CAT scanning) was very expensive in the beginning."