Prevention is the best medicine, doctors say, and nowhere is that truer than in cancer. Picking up early signs of disease is the best way to prevent cancer from taking root, so doctors have urged people get screened for all types of cancer on a regular basis. The assumption is that screening will save them from developing advanced cancer, which is less treatable and deadlier.
That was the reason behind a worldwide push to have every woman get regular mammogram screenings. The idea was to lower the rates of advanced cancers and ultimately deaths from the disease. But in a new study published in BMJ, researchers show that mammography did little to reduce either deaths or advanced breast cancer over a period of 23 years in the Netherlands. Instead, they found that the X-ray based test designed to pick up tumors led to overdiagnoses 60% of the time.
The study involved all Dutch women who were screened with mammograms every other year between 1989 and 2012—about 8 million women in all. The researchers, led by Dr. Philippe Autier from the University of Strathclyde Institute of Global Public Health, wanted to see if the screening affected the number of advanced breast cancers recorded over that time, as well as deaths from the disease. A previous study using some of the same data had found decreases in the incidence of some advanced breast cancers from 1989 to 1997, hinting that widespread mammography was effective.
However, over the longer time period, the scientists failed to find the same reduction. In fact, from 1989-2012, there was no significant decrease in the incidence of stage 2 to stage 4 breast cancers. On the other hand, incidence of early stage in situ cancers, in which the tumor hasn’t spread to surrounding tissues, increased by 4% each year. And Autier calculated that 59% of lesions detected by mammograms were overdiagnosed, meaning they didn’t necessarily need treatment.
Mammography doesn’t actually pick up tumors but rather the changed breast tissue surrounding tumors; it’s possible that many advanced cancers are growing without significantly changing breast tissue, and therefore not getting picked up by mammograms in time.
The findings support a growing number of other population-based studies from the U.S., Australia, and Norway that have also shown that since mammograms were recommended in the 1970s, rates of advanced breast cancer have not dropped dramatically. While deaths from breast cancer have been declining in the U.S., experts say that better treatments, rather than a reduction in advanced cases picked up by mammograms, may be responsible.
Based on such accumulating data, in 2009, the U.S. Preventive Services Task Force recommended that most women begin regular screening at age 50 instead of age 40, and receive the test once every two years instead of every year.
“I don’t think the accumulating data shows that continuing mammography screening is a good solution,” says Autier, “essentially because the price to pay by women in terms of overdiagnosis is enormous.” Previous studies have documented that overdiagnosis of breast cancer can lead to additional biopsies and even treatments that expose women to side effects, without necessarily protecting them from cancer.
Autier says that he was a champion of mammogram screening in previous decades, believing that the benefits for women in detecting cancer outweighed the harms of developing advanced cancer. However, he says the evidence no longer supports that position. “We, and I include myself in the we, promoted mammography screening for plenty of good reasons then,” he says. “I am not glad at all that we found [in this study] that mammography is probably not the ideal solution for protecting women from breast cancer. There is no good evidence that mammography makes a difference in mortality and if it does, it’s probably marginal. But the overdiagnosis toll is enormous.”
Not everyone agrees with his conclusion. Robert Smith, vice president of cancer screening at the American Cancer Society, notes that not all women eligible and invited to have mammograms received them, and of those that did, not all continued to get them regularly. That means that while 80% of women “attended” screening, as Autier says, this does not mean that women were getting screening consistently every other year. Attendance rates also tend to decline over time, says Smith. In addition, Smith says the study could not account for an overall increasing incidence of breast cancer (due to changing environmental exposures, for example), and different rates of the disease in different age groups. If breast cancer rates overall were rising in the study period, then little change in the diagnosis of advanced cancer could still mean that rates of advanced disease were declining, since more cancers were being reported.
Autier says he is not against screening for cancer, but says that based on his findings, mammography is likely not the ideal way to detect breast cancer. In contrast, he says, screening for cervical cancer and colon cancer are extremely effective; those methods have been linked to a 70% to 80% drop in deaths from those cancers.
He argues that his findings support the need for a more effective way to detect breast cancer, which other researchers are currently developing. While mammography isn’t an ideal tool, it is the best available technology for now. All women, and especially those at high risk of developing breast cancer because of a family history of the disease, should discuss with their doctors the benefits and risks of screening and decide if they should continue screening, and if so, how frequently and for how long.
Doctors need to be more upfront in discussing the pros and cons of mammography with their patients, he says, and informing them that data suggests the test may not be as effective as doctors once thought. Based on their own personal risk for breast cancer, he says, women should then make their own decisions about whether they should get regular screening.