Technique Can Predict Likelihood of Tumors Spreading, Scientist Says
A new technique can predict the likelihood that tumors of the breast and colon will spread throughout the body, a Canadian researcher reported here Tuesday.
If the technique shows that a tumor will not spread, then debilitating radiation and drug therapy may be safely avoided after the tumor has been surgically removed, chemist Ian C. P. Smith of the National Research Council in Ottawa said at the Third Chemical Congress of North America.
In addition, he said, unnecessary radical mastectomies, in which surgeons remove a breast, lymph nodes in the armpit and muscles linking the upper chest and shoulder, might also be avoided.
Doctors currently use such factors as the size of a tumor and its appearance under the microscope to help predict which tumors are likely to spread, but it is a “subjective” science, Smith said.
The technique, known as nuclear magnetic resonance spectroscopy, or NMR, is a variation of magnetic resonance imaging, which is now becoming widely used in place of X-rays for seeing inside the body.
Smith also reported that the same technique could be used on blood samples to monitor the course of therapy for all types of cancer and perhaps even to detect cancer at an early stage.
“There are two big problems in cancer: early detection and, when you’ve found a tumor, what is the prognosis,” Smith said. “What I am reporting about is the use of a chemical technique that will help the medical profession with both of those problems.”
NMR has been used by chemists for more than 30 years to determine the structure of molecules. The chemical to be studied is placed in a powerful magnetic field and bombarded with radio-frequency energy. By measuring which frequencies of the energy are absorbed, scientists are able to deduce the structure of the chemical.
High Concentration
In their studies, Smith and Caroline E. Mountfors of the University of Sydney in Australia ultimately concluded that a sharp NMR spectrum was produced by a high concentration of fat molecules on the surface of the tumor cells.
In applying the technique to rat tumors, Smith noticed something peculiar about the tumors. Some got rid of the absorbed energy extremely rapidly--vibrating it away in less than 35/100 of a second--while others took much longer. Further studies showed that tumors that shed the energy slowly had a very high probability of spreading to other sites in the animals’ bodies (metastasizing), while those that shed it rapidly never metastasized.
In the last three years, Smith has studied breast and colon tumors excised from 167 patients at Ottawa General Hospital and Ottawa Civic Hospital. Twenty-five of the tumors shed the absorbed energy quickly, and none of the patients from whom the tumors were removed have subsequently developed metastases, he said.
But virtually all of the patients from whom the other tumors were removed either had metastases at the time of the surgery or developed them later. Both groups had similar follow-up therapy consisting of either chemotherapy or radiation.
“It should be most useful for breast cancer,” he said. “If the tumor has no metastatic potential, a lumpectomy will suffice, but if it can metastasize, more radical surgery will be necessary. The surgeon will know which is necessary during the operation.”
NMR spectroscopy can also be used to detect tumor-related fats in blood samples drawn from patients--a technique first reported in 1986 by radiologist Eric Fossel of Beth Israel Hospital in Boston. The presence of such fats in blood samples indicates the presence of a tumor in the patient, and their absence after cancer therapy indicates that the therapy was successful.
Smith reported that he had used the technique on the blood of 1,400 patients. He found, unfortunately, that the test gave a 20% incidence of false positives (indicating the presence of a tumor when none was there). He concluded that this rate was too high for routine screening for cancer, but that it was low enough for monitoring therapy.
In a telephone interview, Fossel said, however, that he had studied more than 3,400 patients and had a false-positive rate of only 5%. A Norwegian group had a similar false-positive rate for more than 1,000 patients, he added.
Furthermore, he said, preliminary results indicate that they have developed a way to tell which positives are false. These results make it very likely, he said, that the technique will be used for widespread cancer screening in one or two years.
