There is little doubt that mammography saves lives. Since the advent of mass screening programmes, death rates from breast cancer have dropped around the globe, in part due to better diagnosis. And while precise recommendations vary by country, most women aged over 40 will be invited for regular mammograms for as long as they remain in good health.

However, mammography is far from an infallible tool, with the associated issues well documented. False positives are fairly common, requiring healthy women to undergo additional tests and a potentially agonising wait for results. They may even face unnecessary treatment since some of the cancers identified through mammography would never have become symptomatic or life-threatening.

On the flip side, not all tumours are caught in time for the woman to stand a chance of recovery. And a further group of women are led into a false sense of security, with as many as 20% of breast cancers missed altogether.

False negatives are particularly prevalent in women with dense breasts, which are constituted less from fat, and more from fibrous and connective tissue. Because their scan will tend to show large areas of white – conveying either tumours or normal areas of density – abnormalities can prove difficult to spot.

"Cancers are harder to find in people with dense breasts because the tumour can be hidden in normal tissue," explains Dr Laurie Margolies, director of breast imaging and associate professor of radiology at Mount Sinai Hospital.

"In the very dense group, up to 40% of breast cancers will not have one of the features that make it possible for it to be identified on a routine mammogram."

"If a woman knows her breasts are dense, even the best mammogram possible could have missed the cancer. So there are other things one could do in the attempt to find a cancer early."

In an unfortunate twist, dense breasts are themselves a known risk factor for cancer. Since the 1990s, the American College of Radiology has used a four-tier model of breast density: almost entirely fat, scattered fibroglandular densities, heterogeneously dense and extremely dense. Compared with women in the least dense category, those with ‘extremely dense’ breasts are four to six times as likely to succumb to the disease.

Laws of disclosure

Given this double blow – the heightened risk of cancer and of having it missed – it is not surprising that the issue is gathering momentum. While historically very few women were notified about their breast density, there is lobbying underway to make that information more accessible. Pressure is mounting in Japan and the EU, and legislation has been proposed in Canada.

The US is some way ahead of the curve. Starting with Connecticut in 2009, 21 states so far have implemented disclosure laws. While the legislation varies in the specifics, it all requires the physician to notify a woman if she has dense breasts. There is also a federal bill under consideration that would mean all women are kept informed, irrespective of how dense their breasts are.

As Margolies explains: "The idea is that if a woman knows her breasts are dense, even the best mammogram possible could have missed the cancer. So there are other things one could do in the attempt to find a cancer early. The most common thing would be a breast ultrasound, which is readily available and inexpensive, and then, for a higher-risk patient, breast MRI."

The laws do not typically specify that insurance will pay for supplemental screening, only that the woman should discuss the options with her provider, but it goes without saying that many women will want to look into extra tests. Since the laws were passed, a number of people have undergone an additional breast ultrasound, albeit with some geographic variability.

"It’s more common on the East Coast and will tend to happen there if the radiologist thinks it’s a good idea," says Dr Karla Kerlikowske of the University of California, San Francisco. "But California passed a law two years ago this April, and there’s not been much screening ultrasound here."

Along with fellow experts, including the Dartmouth Institute’s Dr Anna Tosteson, Kerlikowske has studied the purported benefits of ultrasound after a negative mammogram. Their recent study, published in the Annals of Internal Medicine in December, used information from breast cancer surveillance databases to look at the pros, cons and cost-efficiency of the practice. It came to a stark conclusion.

"We found that supplemental ultrasound screening for all US women with dense breasts would substantially increase healthcare costs with little improvement in overall health," explains Tosteson. "For every 10,000 women with dense breasts receiving ultrasound screening exams after a normal mammogram between the ages of 50 and 74, about four breast cancer deaths would be prevented, but an extra 3,500 biopsies would occur in women who did not have breast cancer."

In other words, the rate of false positives is arguably insufficient to justify funding the practice. While an ultrasound would indeed find the majority of cancers, it would also find many more abnormalities requiring biopsy, most of which would be benign.

The situation seems particularly untenable given the number of women with dense breasts. More than 40% of US women aged 40-74 are estimated to fall into this category, and that figure is inversely correlated with age. Around half of women in their 40s would be classed in the heterogeneously or extremely dense groups.

Surveying the options

Kerlikowske does not feel that ultrasound should be written off altogether, however. She points out that while the study explored handheld ultrasound, it did not look into automated whole-breast ultrasound, and the jury is out on whether this newer technique might benefit women with dense breasts.

Overall, though, it seems clear we would want a greater degree of nuance from the ideal diagnostic tool. It would need high sensitivity – the ability to detect cancer when it is present – and high specificity, to avoid false alarms.

One contender is digital breast tomosynthesis (DBT), otherwise known as 3D mammography, which has just received a major reimbursement boost in the US: from 1 January, Medicare has covered payment for DBT in conjunction with a normal screening mammogram.

While DBT may show greater sensitivity – it enables the radiologist to view the breast in thin slices, rather than just as a single 2D image – there have been relatively few studies performed to date.

"Tomosynthesis is being widely adopted without much formal evaluation," explains Tosteson, who is working with the Breast Cancer Surveillance Consortium (BCSC) to further appraise its benefits.

"There is some evidence that while callbacks may be less frequent with tomosynthesis than digital mammography, callbacks following tomosynthesis may result in a biopsy more frequently. It is important that this be investigated."

"We’re trying to figure out who’s at high risk of a missed cancer and then ask, ‘Do they need whole-breast ultrasound, do they need tomosynthesis, do they need MRI?’."

"It makes it easier to see some of those cancers by eliminating the overlapping tissue, so some of those cancers will stand out," adds Margolies. "But if a cancer doesn’t have calcification, if it doesn’t distort the parenchyma or there is a border with fat, it can still be missed on the 3D mammography."

Look for the negatives

Kerlikowske feels that in the absence of a single perfect imaging modality, our best bet is to home in on the highest-risk women and offer them whichever tests they need. She points out that ordinary digital mammograms do, in fact, find the majority of cancers, even among women with extremely dense breasts. However, the critical question, is: how do we identify the subgroup that is most likely to present with false negatives?

"At this point, 45% of the people being screened have dense breasts, so we can’t do supplementary imaging on all those women, and it’s unclear whether 3D tomosynthesis will work," Kerlikowske says. "So we’re trying to figure out who’s at high risk of a missed cancer and then ask, ‘Do they need whole-breast ultrasound, do they need tomosynthesis, do they need MRI?’. We’re trying to see which one of those might work in that group."

Her team helped create the BCSC Risk Calculator, an interactive tool estimating a woman’s five-year risk for breast cancer.

This is the first predictive tool to look at breast density alongside factors like age and family history. She is now looking to ascertain which of its users are prone to having their cancer missed.

Margolies contends that knowing your risk can only tell you so much and that finding out you have low-density breasts is no grounds for complacency.

"There are many people who develop breast cancer who don’t have risk factors, and a lack of a family history of breast cancer or other risk factors does not mean that you’re safe. It makes sense that we screen all women each year, beginning at age 40," she says.

It seems clear, however, that information is power when it comes to deciding the next steps. As more women start being notified about their breast density, we can expect surging demand for options beyond mammography. Further research is needed to establish what these options should be.

"We initially tried to discourage disclosure laws at a national level because we didn’t feel we had enough information to tell people what to do. We’re closer maybe, and we’re actively trying to produce results, but we’re still not quite there," says Kerlikowske.

For the time being, dense-breasted women are advised to talk to their healthcare provider in order to go through the pros and cons of additional screening. As research mounts, we can only hope that the evidence will throw up some clearer answers.