The number of women diagnosed with ductal carcinoma in situ, abnormal cells that sometimes become breast cancer, has soared since the 1970s. That's mostly because more women have been getting screening mammograms that can detect the tiny lesions.
The vast majority of women diagnosed with DCIS have surgery, even though there's considerable debate whether it's needed, since DCIS sometimes never becomes invasive cancer.
Shifts in treatment since 1999 away from single mastectomy and toward lumpectomy with radiation for DCIS haven't changed breast cancer survival rates, according to a study that looked at data on over 120,000 women.
The highest overall survival rate after 10 years, 89.6 percent, was in women who had lumpectomy with radiation. The survival rate for women who had mastectomies was 86 percent, followed by lumpectomy alone at 80.6 percent.
But most women in the study group who died didn't die of breast cancer; cardiovascular disease was the major killer, with just 9 percent of deaths overall due to breast cancer.
Looking at deaths from breast cancer alone, the 10-year survival rates were pretty much identical: 98.9 percent for lumpectomy plus radiation; 98.5 percent for mastectomy and 98.4 percent for lumpectomy alone.
Between 1991 and 2010, the number of women who chose lumpectomy with radiation almost doubled, the study found, rising from 24 percent to 47 percent. The number of women choosing single mastectomy dropped from 45 percent to 19 percent.
The number of women who chose no treatment, which usually involves screening mammograms, rose from 1 percent to 3 percent.
But more women also started choosing bilateral mastectomy, which usually involves removing a healthy breast as well as a breast with DCIS. Those numbers rose from zero in 1991 to 8.5 percent in 2010. They tended to be younger women.
The results were published online in the Journal of the National Cancer Institute.
"When we treat these precancers women do very, very well; they have a 99 percent chance of not getting breast cancer," says Shelley Hwang, senior author of the study and chief of breast surgery at the Duke Cancer Institute.
She and her colleagues were expecting to see differences in survival based on treatment, "but there really didn't seem to be any difference at all. Which sort of argues for doing the bare minimum versus doing the most you can do."
Problems after surgery can be significant, she notes, and include long-term pain, disfigurement and lymphedema if lymph nodes are removed..
"The troubling trend in my point of view is that more women are getting bilateral mastectomies," Hwang says. "That's because there's a limited understanding on how good the treatment is for DCIS. I'm not saying that women don't die from breast cancer; they do. But the cures have never been better, and your likelihood of surviving is greater than 90 percent."
Far better, Hwang says, would be to have a test that can tell which types of DCIS will become dangerous cancer, and which will never cause any harm at all. Other researchers are trying to create those tests, and Hwang is hoping to run a big clinical trial that will compare surgery to taking hormone-suppressing medication instead.
"Wouldn't it be wonderful if we take a disease that women are feeling compelled to have a bilateral mastectomy for, and it can be eradicated by taking a pill once a day?" Hwang asks.
That's what she's working for.