The emerging science of mammography would soon beat breast cancer.

That’s what Dr. Donald R. Lannin, then a young breast surgeon, and others in the field believed in the 1980s. The mammogram, an X-ray picture of the breast, had recently emerged as a mainstream procedure. In October 1985, the American Cancer Society and other organizations designated October as Breast Cancer Awareness Month, and the importance of early detection through regular screenings for women beginning in their 40s was taught to the public.

In the decades since mammogram screenings became more mainstream, Lannin says although the number of small breast cancers diagnosed has gone up dramatically, decreases in deaths from breast cancer have not kept pace.

“I’ve realized that early detection is really only a minor benefit; it’s probably some benefit, but it’s pretty small,” says Lannin, now a professor of surgery at the Yale School of Medicine. “If you look at the figures for the reduction in mortality due to mammography, it’s about 19 percent.” He adds that given the number of incredibly tiny cancers found on screenings, “that really ought to reduce the mortality from cancer at least 50 to 75 percent. So, why is it only reduced 19 percent?”

It’s a question Lannin believes he has answered in a study published in June in the prestigious New England Journal of Medicine. The study, coauthored by Dr. Shiyi Wang, assistant professor of epidemiology at the Yale School of Public Health, builds on previous studies showing 22 percent of breast cancers are overdiagnosed. This does not mean that radiologists made a mistake in diagnosing these tumors, but rather that some cancers are so slow-growing and biologically favorable, that they would never bother the woman if they were not found.

“Many small cancers are small not because they were caught early, but because they are biologically so [inactive] that they will never get large,” Lannin says. “In contrast, many large cancers are large not because they were ignored or not found, but because they grow so fast that they get large before they could ever be detected by mammography.”

New findings

Previously, researchers believed the lead time (the time between when a tumor can be detected by mammography and when it becomes apparent in other ways) for all cancers was essentially the same. Lannin’s study shows large, aggressive cancers have a lead time of two years or less, while many smaller cancers have a lead time of 15-20 years, which is beyond the life expectancy of some older patients.

“It’s deceptive because you see a whole lot more small cancers and they all do very well, and you think, ‘it’s because you caught it when they were small,’” Lannin says. “For years, I was taking out these small cancers thinking, ‘Thank goodness we caught it then and took it out when it was small before it got large.’ Now I realize that many of those small cancers would never have gotten large or bothered the patient.”

For the study, Lannin and Wang analyzed invasive breast cancers diagnosed between 2001 and 2013 as part of a publicly available database that contains cancer data on about 28 percent of the U.S. population. They divided the cancers into favorable, unfavorable and intermediate categories based on tumor grade and hormone receptor expression. Dr. Wang then developed a sophisticated set of models to explain which patients and tumor types most likely account for the 22 percent expected rate of overdiagnosis. “We compared lead time and life expectancy,” Wang says. “If the lead time is longer than life expectancy, this means the cancer will not become clinically apparent before the patient dies from other causes and therefore this patient is overdiagnosed.”

Their models showed that women were increasingly likely to be overdiagnosed as they aged, because the odds increased that they would die from other causes before their slow-growing cancer could become fatal. For example, one of the models showed that for a woman in her 70s diagnosed with a biologically favorable tumor, the chances are 65 percent that this represents overdiagnosis. If a woman in her 50s were diagnosed with that same cancer, the chances of overdiagnosis would only be 40 percent. In contrast, if the woman in her 70s were diagnosed with a biologically unfavorable tumor, the chances of overdiagnosis would only be 5 percent.

Practical implications

The difficult part for patients and researchers is that some small cancers do become dangerous. Lannin says that “out of 100 cancers found on a mammogram, 50 percent of them are slow-growing cancers that would eventually become clinically apparent but would still be curable at that time. About 25 percent of the cancers are slow-growing cancers that would never become apparent during the patient’s lifetime. These are the overdiagnosed cancers.” He adds, “There is another 20 percent that are rapidly growing cancers that are already incurable even when found on a mammogram. Then, there’s about 5 percent that are curable when discoverable on mammography but that would become incurable by the time they are large enough to be found without mammography.”

This 5 percent, or five people out of 100, account, roughly, for the 19 percent relative risk reduction in mortality associated with mammography screenings Lannin cited earlier.

But the role that mammography screenings play in increased longevity is controversial. A study published in March in the Annals of Internal Medicine looked at women in Denmark where screenings were implemented in different regions at different times, so they could compare groups of women who had regular screenings versus those who did not. The study concluded “breast cancer screening was not associated with a reduction in the incidence of advanced cancer,” or decreased mortality.

In 2015, the American Cancer Society issued new recommendations for screenings, advising women to start annual screenings at 45 (it had previously advised starting at 40) and to continue exams until 55. After turning 55, the organization suggests women transition to screenings every other year until they turn 75 (previously, it had suggested continuing annual screenings).

The Canadian Cancer Society advises less screening; its guidelines recommend women start screening every other year at 50 and to continue only until age 70.

“Some people think, ‘Well, we shouldn’t do so much mammography.’ I’m not particularly advocating that,” Lannin says. “I think that mammography overall does have some benefits but the benefit is fairly small, 19 percent instead of 50 or 70 percent, but still that 19 percent is important. So I’m not saying that you shouldn’t do mammography, but you might stop doing it a little younger. … The risks and benefits of mammography are not as favorable in older women. They’re much more likely to have an overdiagnosed cancer and much less likely to have their lives saved from a cancer they found. One of the things we might consider is stopping mammography by age 70. We could then avoid a lot of overdiagnosis in women who get mammograms in their 70s and 80s.”

He’d also like to see doctors who find small cancers do molecular testing to verify whether they are biologically favorable, and therefore likely to grow slower, and to start treating slow-growing cancers, “much less aggressively.”

He says, “The minimum treatment that will probably be used most of the time is a lumpectomy to take the cancer out.” In addition, today lymph nodes might be taken out at the same time, but if it could be determined a patient’s cancer was slow growing, then “there’s probably no reason to take out lymph nodes. Typically we use radiation or drug therapy and we probably wouldn’t have to do either of those if we think it’s an overdiagnosed cancer.”

Lannin does caution that more research is necessary and he hopes his study will lead to trials in the future. “It’s hard to make too radical a change in your treatment based on just our study.”

Less awareness, more understanding

Beyond preventing unnecessary procedures, Lannin hopes his study will help patients avoid the psychological hardships a breast cancer diagnosis causes. “It’s sad. I see a lot of patients with a very tiny, slow-growing cancer that probably would do fine if we didn’t treat it at all, and they’re just terrified. They think, ‘I’ve got a life-threatening cancer.’ And they want both breasts removed and chemotherapy and radiation and all kinds of treatments that they don’t need and it’s because they’re so anxious and fearful of it.”

Thirty-five years after he entered the field, Lannin would like to see less awareness and more understanding of breast cancer.

“It’s always in the news and the press about breast cancer,” he says. “When I started my career 35 years ago, breast cancer wasn’t really very well known, people didn’t talk about it much. Back then there were quite a few women who had palpable, big, obvious breast cancers that didn’t see a doctor just because it wasn’t hurting. We started breast cancer awareness, and October is Breast Cancer Awareness Month, because we thought, ‘Well, it really will help to make women more aware of breast cancer.’ Today, we don’t need any more breast cancer awareness. Every woman is aware of it. In fact, she’s too aware and very petrified by it in many cases. What we need today is breast cancer education and understanding and the realization that breast cancers aren’t all the same, that some are very good and some very bad. That many times it’s a very good cancer and it’s not worth being all anxious and upset and panicking about because it’s quite likely something that would never bother you, if you didn’t find it.”