The paper’s authors and other leading researchers said the data indicate treatment has not made much difference, if any, for the tens of thousands women a year who are told they have ductal carcinoma in situ, or DCIS.

Share story

 More than 30 years after the widespread use of mammograms set off a surge in the detection of tiny lesions in milk ducts, there is debate about how — or even whether — to treat them.

In an era there has been so much study of how to treat more advanced cancer, it might seem odd that there is so much uncertainty about these minute sprinklings of abnormal cells, often called Stage 0 cancer, which some say are not cancers at all.

The latest round of controversy was set off by a paper published Thursday in JAMA Oncology that analyzed 20 years of data on 100,000 women who had the condition, also known as ductal carcinoma in situ, or DCIS. The majority had lumpectomies (with or without radiation) and most of the others had mastectomies.

The death rate from breast cancer of these patients, regardless of their choice of treatment, during the next 20 years was about the same as the lifetime risk in the general population of women, 3.3 percent.

The paper’s authors and other leading researchers in the field said the data indicate that treatment has not made much difference, if any, for the tens of thousands of women a year who are told they have this condition. (Last year about 60,000 in the United States got a DCIS diagnosis.)

One key piece of evidence is that though tens of thousands of cases of DCIS were being diagnosed and aggressively treated each year, there seemed to be no substantial impact on the incidence of invasive breast cancers found annually in the general population.

About 240,000 were diagnosed with it last year. If treating DCIS was supposed to fend off invasive breast cancer, the incidence of invasive breast cancer should have plummeted once DCIS was being found and treated, the experts said.

That has intensified questions about what DCIS really is: cancer, precancer or risk factor for cancer?

New dilemma

Before mammography, only a few hundred women a year were diagnosed with DCIS. It was a condition almost always noticed only at autopsy. Once radiologists began finding DCIS, though, doctors were faced with a dilemma.

At first glance, it seemed wonderful to find the lesions. The cells looked like cancer although they had not broken through the wall of the milk duct into surrounding breast tissue. Their discovery, it seemed, offered a chance to cut out cancers long before they could be felt as lumps.

But there was a problem, noted Dr. Barnett Kramer, director of the division of cancer prevention at the National Cancer Institute. With DCIS, the abnormal cells were not just a well-defined clump. Instead, abnormal cells often were sprinkled along the length of the milk duct. What to do?

The obvious answer seemed to be to cut off the entire breast. Doctors had the example of abnormalities detected by the Pap test: cutting out early lesions on the cervix prevented deaths from invasive cervical cancer. It made sense to do something similar to prevent breast-cancer deaths, said Dr. Barron Lerner, a historian of medicine at New York University Langone Medical Center.

Yet this was happening at a time, in the 1980s and early 1990s, when women with invasive cancers were told they could have a lumpectomy instead of a mastectomy. “It’s very weird, thinking back now. We treated the more aggressive disease less aggressively than we treated the less aggressive disease,” said Dr. Otis Brawley, chief medical officer for the American Cancer Society.

Unlike the situation with invasive breast cancer, no one did clinical trials in DCIS comparing, say, lumpectomies plus radiation to mastectomies, or watchful waiting, and assessing whether there was a difference in the risk of cancer reappearing in the breasts. But as doctors became comfortable treating invasive cancers with lumpectomies, they began treating DCIS that way, too.

Over the years, investigators have come to conclude that the old model of cancer — a few aberrant cells will grow, spread and inevitably become a deadly cancer if not destroyed — is wrong.

Small clumps of abnormal cells may just stop growing, scientists now know. Even invasive cancers do not always grow. Some regress or disappear. That is especially true in prostate cancer, where up to half of all cancers found with screening will not progress if they are left alone. But it seems true in breast cancer, too, researchers say.

Doctors also used to think that cancers that recur at the original site in the breast after treatment are likely to spread outside the breast and kill. That, too, has turned out not to be true. Some cancers metastasize and others remain in the breast and never leave it, Kramer said. The two types of cancers have different properties. That is why radiation after a lumpectomy does not reduce the death rate from breast cancer although it reduces the recurrence of invasive cancer in the breast.

“Treatment that reduces recurrence in the breast is not a good surrogate for reducing the risk of death from breast cancer,” Kramer said. “This gets to the issue of how counterintuitive cancer is.”

Clinical trial unlikely

With DCIS, women have a worse prognosis if they are black, younger than 40 or have tumors with molecular markers such as those found in more aggressive invasive cancers. They may benefit from treatment, but, so far, no one has done a large clinical trial asking if these treatments prevent breast-cancer deaths for the rest of the women with DCIS. Are the surgical treatments better than watchful waiting and perhaps drugs to reduce overall breast-cancer risk?

Medical experts say it is unlikely such a clinical trial will be done. Because the risk of dying of breast cancer is so low for women treated for DCIS, the study would take 10 to 20 years and involve tens of thousands of women, assuming it would even be possible to recruit so many who would agree to have their treatment decided at random. And assuming somehow there was money to pay for it.

And when it was done, doctors might well argue that the study used old-style medicine, that things had progressed since then and so the results were no longer valid.

“Welcome to the world of dealing with low-probability events,” said Dr. H. Gilbert Welch, a professor of medicine at Dartmouth and author of “Less Medicine, More Health.”

“I think it is a classic example of what is and will only increasingly become a recurrent problem in medicine,” he added. “The questions about what to do — if anything — are fundamentally difficult.”