Looking into a way to predict breast cancer risk
” Breast cancer” is one of the scariest pairs of words a woman can hear. Perhaps that is why so many women opt for the most drastic treatment, a double mastectomy, even when doctors detect a tiny mass that may never cause symptoms. Part of the problem is that doctors have no way to predict which small, early cancers are likely to become invasive (grow and spread) and which ones won’t.
That ambiguity is one reason why the U.S. Preventive Services Task Force, which advises the government on health matters, controversially recommended in November 2009 that most women should not get mammograms until age 50 and thereafter have them only every other year. The Task Force noted that many women are being treated for cancers that would have done them no harm — and those treatments can be frightening, painful and harmful of themselves.
If only doctors knew which early cancers are truly worrisome and which are unlikely to spread. Researchers at UC San Francisco may have made a stride toward that goal in the case of ductal carcinoma in situ, or DCIS, the most common form of noninvasive breast cancer. In DCIS, abnormal cells grow in one of the breast’s milk ducts. These cancerous cells may grow and spread beyond the duct, or they may do nothing at all.
After analyzing the outcomes for nearly 1,200 women who had a DCIS surgically removed, the UCSF researchers were able to categorize low, medium or high risk for future DCIS or invasive, spreading cancer.
In their study, published online April 28 in the Journal of the National Cancer Institute, the researchers found that the worst offenders were DCIS detected by palpation — that is, obvious to the touch — or those that carried three particular chemical marks on their surface. DCIS found by mammogram, or missing those marks, was far less likely to cause trouble.
DCIS is considered to be the earliest form of breast cancer. It occurs when cells outgrow their normal boundaries and form a mass. On its own, DCIS is not deadly, says study first author Dr. Karla Kerlikowske, an associate professor of medicine at UCSF and director of the San Francisco VA Women Veterans Comprehensive Health Center. Left untreated, more than half of DCIS cases will not progress to serious disease, according to the Preventive Services task force report.
Since the introduction of mammograms in the ‘80s, DCIS detections have skyrocketed. Women diagnosed with DCIS are likely to receive aggressive treatment. Most have a lumpectomy, sometimes followed by radiation. A full quarter of women with DCIS opt to have both breasts removed.
Another option is to do nothing, and have frequent mammograms, to see if the mass grows or another appears. Few patients and doctors choose this “wait-and-see” approach.
The uncertainty is the most trying thing, says Cynthia Artiga-Faupusa, a 35-year-old English teacher in San Francisco who was diagnosed with DCIS in October. Having watched her mother, grandmother and family friends suffer from breast cancer, she imagined terrifying scenarios.
Doctors at UCSF could give only vague predictions about her future risk of another DCIS or more invasive breast cancer, and the risk would increase with each passing year. With a lumpectomy, she’d have to return for mammograms — and the accompanying anxiety over her results — every six months.
If anyone had been able to tell her that she was unlikely to face DCIS or a more invasive cancer again, Artiga-Faupusa says, she could have opted for the lumpectomy or “wait-and-see” with less fear. If her DCIS was high-risk, she would have known she had to get more aggressive.
Scared and lacking clear information, Artiga-Faupusa went with the mastectomy. She’ll never know if that decision saved her life or put her through unnecessary surgery.
Though researchers still do not know exactly what distinguishes a DCIS that could lead to invasive cancer from a DCIS that just sits there harmlessly, the new UCSF study gives them clues.
The three chemical marks found on the surface of high-risk masses are proteins called p16, Ki67 and COX-2. All three are associated with a biochemical pathway that leads to cancer, and the presence of any one indicates that cells are growing too much. But it was only the trifecta that strongly presaged invasive cancer in the UCSF study.
The other key risk factor found in the study was how the DCIS was first detected. The ones that tend to be trouble are the ones you can feel, not the lower-risk ones that mammograms alone pick up, according to the study. But that isn’t because palpable lumps are bigger, Kerlikowske says — they’re much the same size. So there must be something else about their biology that makes them a bad sign.
The most dangerous masses are found not during a routine screening, Kerlikowski adds. Instead, a woman might brush against something unusual in the shower, or roll over in bed and notice something doesn’t feel quite right.
It should be possible to develop a diagnostic test based on the three proteins from the study, Kerlikowske says, to determine which DCIS’s are of highest concern. Detection method, meanwhile, is something that doctors can consider right now to offer slightly better advice to women faced with treatment decisions, she adds.
Kerlikowske would like to see more low-risk women choose a lumpectomy and eschew radiation treatment or mastectomy. As many as 44% of DCIS patients might be able to get away with watchful waiting, she estimates.
Even after enduring multiple surgeries in which doctors removed her breasts and constructed new ones from belly fat, Artiga-Faupusa feels she made the best decision she could with the information she had at the time. “I’ve got a great new pair of pretty darn perfect breasts and a flat stomach again, and no cancer,” she says. “And I never have to do a mammogram again in my life.”
Still, Artiga-Faupusa wishes she’d had more information, such as the predictions from the new study, six months ago. “That would have made all the difference in decision-making,” she says.
Yet, as is the case with all scientific studies, this one needs to be repeated and confirmed before doctors change their practice, says Dr. Craig Allred, a pathologist at Washington University in St. Louis.
“I hope it’s true,” he says. “One of the big goals … is to avoid overtreatment.”
