More than 200,000 US women are diagnosed with invasive breast cancer every year. Doctors generally recommend post-operative radiation for any woman who has surgery that removes the tumor but leaves most of the breast, often called lumpectomy. They also recommend radiation for many patients who have mastectomy, or removal of the entire breast. But those recommendations are not always followed.

According to the American College of Surgeons National Cancer Data Base, at least 25 percent of lumpectomy patients skip radiation. And the more advanced their disease, the more likely they are to do without.

Some may live far away from a hospital and find a six-week course of radiation too burdensome. Others may be too ill, or may fear the side effects of radiation -- which can include cancer and heart disease. But until now, many doctors believed the only thing those women were risking was a local recurrence -- a relapse in the same breast, which could be treated by further surgery.

Significant Long-Term Survival Advantage

They based that belief on a landmark study, supported by the National Cancer Institute and first reported in 1985, that compared mastectomy, simple lumpectomy and lumpectomy followed by radiation for women with early breast cancer. That study found that the women who got lumpectomy without radiation had a much higher chance of a local recurrence, but they were no more likely to die of their disease.

Thursday's issue of the prestigious British medical journal The Lancet turns that on its head. An overview of 40,000 breast cancer patients by researchers at Oxford University found that radiotherapy confers a significant long-term survival advantage.

Fifteen years after breast-sparing surgery, 30.5 percent of women who got radiation had died of their disease, compared with 35.9 percent of those who did not -- five extra breast cancer deaths for every 100 women. (The numbers were slightly higher when deaths from all causes were counted, but the difference was the same -- five extra deaths per 100 women.)

Radiation offered the same benefit to women who got mastectomies if their cancer had already spread to the lymph nodes under their arms: After 15 years, 54.7 percent of those who had radiotherapy had died of breast cancer, compared with 60.1 percent of those who did not.

There was no advantage for women who had their entire breast removed if the cancer had not spread beyond the breast. The risk of a recurrence for those patients is so low that any benefit of radiation is likely to be outweighed by its side effects.

Cancer experts said this study would have a major impact on breast cancer treatment.

Reduction in Collateral Damage

Dr. William Gradishar of Northwestern Memorial Hospital said it's now very clear that "reducing the risk of local recurrence -- whether in the preserved breast or in the chest wall -- translates into a significant improvement in overall survival."

Gradishar and Dr. Kathy Albain of Loyola University Health System noted that the Lancet overview found a significant increase in second breast cancers, lung cancers and heart attacks in the patients who received radiation.

But even taking that extra risk into account, the two experts agreed there's a net benefit to radiotherapy for the vast majority of patients. And exposure to other parts of the body is likely to be smaller today, with advances in radiation that reduce collateral damage.

The 1985 lumpectomy study was conceived by Dr. Bernard Fisher of the University of Pittsburgh, a visionary in his day who believed breast cancer was a systemic disease and that there was little point, therefore, in local treatment, such as removal of the entire breast. His theory -- which has been largely accepted by the oncology community -- was that, if a breast tumor was likely to spread and become fatal, it would already have done so by the time it was detected.

Local Control Makes a Difference

Dr. Samuel Hellman, a professor of radiation oncology at the University of Chicago, was among those critical of Fisher's theory all along. On Thursday he said, "It's very hard to explain" the new data "by saying the disease is systemic at its outset and, therefore, local control makes no difference. It does make a difference."

Hellman said the survival benefit to radiation therapy was even greater in women who also got systemic treatment with anti-cancer drugs.

Sarah Darby, a professor of medical statistics at Oxford and a coordinator of the overview, said the new data "very strongly disagree with [Fisher's] point of view" and show the relationship between local control and survival.

"For every four local recurrences avoided," she said, "you avoid one breast cancer death." But the deaths don't show up until many years after diagnosis, whereas nearly all of the local recurrences occur within the first five years.

In the new overview, the five-year risk of a local recurrence in women who had breast-sparing surgery was 7 percent with radiotherapy and 26 percent without it. For women who had their breasts removed, the risk of a local recurrence within five years was 6 percent with radiotherapy, compared with 23 percent without it.