Breast density becoming an important issue for accurate screening of cancer
January 31, 2019
Are you dense? It sounds offensive, but it’s a question Canadians need to be asking, say advocates of a growing awareness campaign about breast density and its relation to increased cancer risk.
“The denser you are, the more likely you are to get breast cancer and the more likely it is to be missed,” said Dr. Paula Gordon, a Clinical Professor in the Department of Radiology at the University of British Columbia. She is also Medical Advisor to Dense Breasts Canada, a non-profit organization dedicated to educating people about the risks associated with dense breasts.
As early as the mid-1970s, medical experts showed a correlation between extremely dense breast tissue and increased cancer risk.
More recent studies indicate the risk of missing cancer is higher in people who have dense breasts when 2D digital mammography is used as a screening tool because of the masking that occurs.
Extremely dense breast tissue and cancerous tumours both show up white on a mammogram image, so that spotting abnormalities is like looking for a snowman in a snowstorm, said Dr. Gordon. Often, tumours in dense tissue aren’t discovered until they are at an advanced stage and noticed as a lump, even in those who routinely go for mammograms.
Despite the push for awareness, the latest guidelines for breast cancer screening, released by The Canadian Task Force on Preventive Health Care in December 2018, do not include recommendations specific to breast density.
According to the guidelines, “classification of breast density status has poor reliability.” The report goes on to caution that “supplemental screening may increase the number of women experiencing over-diagnosis with unnecessary treatment and negative effects from false-positive results.”
It also states that the task force “did not specifically review evidence on supplemental screening for women with dense breast tissue.”
Leading Canadian breast screening experts believe the recommendations could and should do better.
When reading mammograms, radiologists apply the Breast Imaging-Reporting and Data System (BI-RADS) to standardize their description and assessment of image findings. BI-RADS is also used to assign a density score.
Category A signifies breasts that are almost entirely fatty; B indicates scattered areas of fibrous and glandular tissue; C indicates breasts that are heterogeneously dense, indicating large areas of dense tissue which may hide tumours; and, D indicates extremely dense tissue which will lower the accuracy of mammography.
The argument is that women in the highest two categories should be put into a high risk group and offered supplemental screening, similar to the way other risk factors are considered.
“The task force says density is no big deal, we’re not going to treat women with dense breasts any differently than average-risk women,” said Gordon. “My goal is to get all of Canada using BI-RADS and for doctors and patients to be told.”
British Columbia is the first province to mandate that women and their doctors receive information about breast density after mammogram screening. In making the announcement, B.C. Health Minister Adrian Dix stated the decision was based on evidence.
In other provinces, women can ask their family doctors about their breast density measure and if they are in the extremely dense category D, meaning their breast tissue is more than 75 per cent dense, they may be recommended for annual screening and/or supplemental testing.
The lack of federal recommendations means breast density is treated differently according to where people live. In Alberta, for example, radiologists report density to primary-care providers who then decide the next course of action for women in the higher categories.
Supplemental screening is covered, but the technology available varies between clinics. In Ontario, patients in the extremely dense category are recommended for annual mammograms.
One reason for the push for increased awareness and education about breast density is the advent of new technology, including tools that bring consistency to how it is measured and advanced imaging techniques that do a better job than 2D mammography at detecting early stage cancers in dense breast tissue.
For example, densitasdensity from Halifax-based Densitas Inc. and VolparaDensity from New Zealand’s Volpara Solutions Ltd., are providing software to automatically generate standardized breast density measurements, removing the subjectivity associated with visual assessments and giving radiologists a good indicator to confirm what they see with their eyes.
Volpara uses volumetric breast density, a method that divides the volume of fibroglandular (dense) tissue by the total volume of breast tissue and has been validated by magnetic resonance imaging. Densitas applies artificial intelligence (AI), using an algorithm to ‘teach’ its software to assess the same images a radiologist sees.
“We want to provide a measure of density that is standardized, reliable and reproducible so that we have equitable service delivery, and so that planning and resource planning can be applied uniformly across the healthcare system,” said Densitas founder and CEO Mo Abdolell.
The densitasdensity software appliance installs on a network alongside the mammography machine and picture archiving and communication system (PACS).
Every mammogram acquired is routed through the software, which automatically generates a density report and sends it to the PACS or electronic medical record for inclusion in the radiologist’s report. The technology is deployed across the province of Nova Scotia and is being considered by breast screening programs in other provinces.
According to Abdolell, the company is leaving the decision about what to do with the automated breast density score up to healthcare system policy makers.
“A lot of people are focused now on whether or not women are informed of their density. It’s a complex issue and it has to be a healthcare system decision on how you inform women and what do you inform them,” he said. “That’s an important question that has to be dealt with at the system level. However, we know that by providing a standardized measure of density to the radiologist at the point of care, they can already make a really important decision when it comes to how they manage that patient.”
Dr. Martin Yaffe, Co-director of the Imaging Research Program at the Ontario Institute for Cancer Research and a Volpara founder, has had a long-standing interest in breast density, helping to develop an early tool called Cumulus for measuring breast density in a research lab environment.
He believes there is a clear path towards action to reduce the likelihood of cancers being missed in women with dense breast tissue, beginning with quantitative, standard measure. He has been pushing the Ontario screening program to start reporting density.
Dr. Yaffe’s lab is currently applying feature and texture analysis to analyze the composition and structure of a mammogram, applying AI to refine their algorithm so it can more accurately identify women with dense breast tissue who should be recommended for supplemental screening.
“If a woman knows she has dense breasts, even if her screening program didn’t offer additional screening … at least until it is offered, she would have that information so she could take her health into her own hands,” said Dr. Yaffe. Once women know they have dense breasts, he said, they may be more likely to have a discussion with their family doctor about recommended next steps.
“Supplemental exams are additional money spent in the health system; the pay-off is you’re going to find more cancers,” he said. “And you’re going to find them at an earlier stage when more can be done and the treatment may not have to be as aggressive.”
Advances in imaging tools used for supplemental breast screening include automated breast ultrasound (ABUS), 3D digital tomosynthesis mammography and contrast enhanced mammography, all of which make it easier to spot cancers in dense breast tissue.
Breast MRI is frequently used for supplemental screening for women considered high risk due to family history and other factors, but it is an expensive test and requires a contrast dye – typically gadolinium – which introduces additional health risks. In the absence of guidelines to recommend supplemental screening for breast dense tissue, however, there is no consistency in how the technologies are applied across Canada.
Dr. Jean Seely, Head of Breast Imaging at The Ottawa Hospital Breast Health Centre, believes there needs to be a better way of communicating breast density to patients and treating physicians, and more awareness that dense breasts will reduce the accuracy of a routine, biannual 2D mammogram.
“If you have 100 women with dense breasts and they all have breast cancer, the sensitivity is only going to be about 60 percent. We’re going to miss 40 percent of those cancers,” said Dr. Seely, referring to the sensitivity of 2D mammogram. “That, to me, is the biggest harm of screening mammography. There are a lot of people discussing over-diagnosis and false alarms. Women aren’t so worried about that. They’re more worried about the cancer being missed,” she said.
The Ottawa Hospital is one of three Canadian clinical trial sites to participate in the first North American randomized, multi-centre study to assess the effectiveness of tomosynthesis called The Tomosynthesis Mammographic Imaging Screening Trial (TMIST). Also under way at Sunnybrook Health Sciences Centre and in Vancouver under the Screening Mammography Program of B.C., the study is comparing conventional mammography to the newer 3D method at more than 90 centres to determine whether screening with tomosynthesis will reduce the number of fatal cancers. The trial will enrol 165,000 women by the end of 2020.
“We know from other studies that tomosynthesis is going to help pick up more breast cancers than 2D mammogram, even for women with the most dense breasts,” said Dr. Seely, explaining that the 3D image provided by tomosynthesis makes it possible to see a mass that is obscured by breast tissue because it provides views of a breast in multiple layers.
The CIBC Breast Assessment Centre at the Juravinski Hospital and Cancer Centre in Hamilton, Ontario, opened in 2014 to provide a comprehensive, one-stop site for breast cancer screening, assessment and surgical consult. The centre’s Medical Director, Dr. Kavita Dhamanaskar, said she suspects tomosynthesis will be recommended for all patients in the C and D BI-RAD categories within the next decade.
At the Juravinski centre, one of the goals is to make the breast a sub-specialty of radiology, similar to the way abdominal or neurological specialties are managed, so that patients receive consistent, high-quality care in a fast and accurate manner. Dr. Dhamanaskar said she believes that breast density is an important factor to consider and that patients in the highest categories of density should be carefully managed.
The centre was scheduled to receive its first tomosynthesis machine in January, and is also considering handheld ultrasound and ABUS for supplemental screening. The challenge, she said, is that supplemental ultrasound is time-consuming and tends to pick up more benign lumps that then require follow up per radiology guidelines.
“It’s extra resource utilization for something that’s not cancerous, so it creates a large amount of flow issues for centres. I’m pro-ultrasound, but I see the hesitation in dedicated breast centres to implement it as a screening tool,” she said, adding that she would recommend ABUS over using a handheld probe for screening by ultrasound because of its ability to scan the entire volume of a breast in a shorter time. “When a province is spending money on a screening tool, it has to be timely, quick, comfortable for the patient and cost effective.”
Each mammogram report produced by the Juravinski centre includes a mandatory reporting of breast density so that patients are informed of their BI-RAD score. “My bottom line is that women should know,” said Dr. Dhamanaskar.
In advance of the CTF recommendations, Dr. Gordon engaged with every provincial health minister and screening program. In a statement released by Dense Breasts Canada, she expressed her disappointment in the new guidelines and maintained that the CTF overstated the risk of over-diagnosis. “False alarms are the main reason not to screen women, but if your general practitioner explains it to you, that there’s a 10 percent chance you’ll get recalled (for supplemental screening or a needle biopsy) even though there’s a one percent chance you have cancer, I think most women understand that,” she said.