SAN ANTONIO -- Removing just one to three key lymph nodes instead of the usual dozen or more can spare women lifelong arm problems and reliably indicate whether breast cancer has spread and needs aggressive treatment, the first big study to test this approach has found.
Many doctors have been doing this without proof that it is as good as the standard operation, and they still don't know whether it will hurt women's survival odds.
But the large, federally funded study has answered at least the accuracy question, finding that the less severe surgery is 97 percent accurate at revealing whether cancer has spread beyond the breast.
"There is a high degree of accuracy here," said Dr. Thomas Julian, a breast cancer surgeon at Drexel University College of Medicine and Allegheny General Hospital in Pittsburgh. "This offers an option for the majority of women."
He presented the research Thursday at a breast cancer conference in Texas.
Dr. Stephen Edge, a surgeon at Roswell Park Cancer Institute in Buffalo, N.Y., called it "a landmark study that many of us, including me, thought was undoable," because so many women already demand the less-severe surgery.
When a woman has breast cancer surgery, doctors typically remove a third of the lymph nodes in her arm -- about 10 to 20 -- to see if the cancer has spread. The answer determines whether she needs further treatment with chemotherapy and radiation.
But the surgery leaves many women with motion problems and less feeling in their shoulder and arm, and up to 20 percent develop lymphedema, painful and severe arm swelling that can recur throughout their lives. They are also at greater risk of infection because they have lost so many lymph nodes that drain fluid from the arm.
"Patients often get caught by friendly fire," suffering ill effects from a procedure intended to help save their lives, said Dr. Mark Kissin of the Royal Surrey County Hospital in Gilford, England.
The alternative is called sentinel node biopsy. At the time of a woman's breast cancer surgery, doctors inject a dye that travels and collects in the lymph nodes most involved in draining the area nearest the tumor. The theory is that these "sentinel nodes" would be most likely to contain malignant cells if the cancer had spread beyond the breast.
Julian's study involved 5,260 women in one of the longest-running cancer studies ever, the National Surgical Adjuvant Breast and Bowel Project. It is known for such watershed findings as proving that removing just a cancerous lump was as good as removing the entire breast for most women, and establishing the benefits of chemotherapy and tamoxifen.
All women in the new study had an average of three sentinel nodes taken out. Half then went on to have the usual 10 to 20 nodes removed. The others had more nodes removed only if the sentinel ones had cancer.
In 96 percent of cases, when the sentinel nodes did not contain cancer, no other nodes in the armpit did. The false negative rate -- when the sentinel nodes were thought to be cancer-free but the disease was present in other lymph nodes -- was nearly 10 percent.
That is higher than some smaller studies have found but comparable to that of the tissue examination tests used now to declare cancer present or absent. And it is probably close to what women can expect in the real world, since this study involved more than 230 surgeons throughout North America, including many who do not specialize in breast cancer.
"It shows that a broad population of surgeons can carry this out," Julian said.
Kissin presented a separate study of more than 1,000 women by British doctors that documented how devastating the standard lymph node operation can be: 37 percent of women had some loss of arm sensation at six months, compared with only 14 percent who had the sentinel node approach. They also had higher rates of lymphedema.
"There shouldn't really be a choice any more. Sentinel node, for the patient, should be the standard of care," he said.
But Edge and others said such a recommendation should wait until after the federally funded trial can provide information on long-term survival. Some worry that sentinel node biopsies will miss too many cancers and lead to deaths that could have been prevented.