On March 10, 2023, the FDA issued a rule making it a requirement to send letters to patients undergoing screening mammography to inform those who have dense breasts. Where do I begin? After all, my professional study of mammograms that miss cancers dates back 30 years with my first academic paper: “Establishing a Histologic Basis for False-negative Mammograms,” published in the American Journal of Surgery….and the only publication of mine that was ever honored later as “One of the Top Contributions to Oncology in the Past 50 Years” by the same journal in a later anniversary issue.
Funny thing, though. At that time, I was focused on the impact of histology of various types of breast cancer, that is, what tumors look like under the microscope. And it was clear that some tumor growth patterns escape detection by mammography because they grow diffusely, without forming a “mass.” That publication was only the second time that “invasive lobular” cancer had been implicated in false-negative mammograms (common knowledge today). And, we made brief reference to the additional theory that tumors might be missed on mammography due to breast density. Only a few authors had raised this question….in 1993.
Long story short, in writing that first article, I wanted to start off with what was known about mammographic sensitivity, that is, the percentage of cancers detected. To my dismay, I could not find the reference that supported the oft-quoted, official, widely accepted… 90% sensitivity. I’ve chronicled this search in my books, and more recently, as the most comprehensive coverage of the story, written in a review article published in the American Journal of Surgery (same journal as above, circling back from my first publication). It’s a fascinating story of scientific paralysis, where the medical community refused to accept the true miss-rate of screening mammograms, a problem that continues to this day.
The question of missed cancers was hard to answer back then: How do you know you missed a cancer if it’s not on the mammogram? Quick Answer: you don’t. Instead, they used arbitrary follow-up periods, usually one year — any cancer appearing quicker than one year after a negative mammogram was a “miss.” It would take many pages to explain why this was an absolutely terrible way of judging sensitivity. But when breast ultrasound was introduced, the cracks in 90% started to form, and by the time breast MRI was introduced to the clinic in 2002, the belief in 90% was shattered (for many, but not all). If MRI is compared to mammography head-to-head, then we discover the chilling truth that mammograms miss 60% of cancers. That’s right, the number is remarkably similar from study to study. The sensitivity of mammography is only 40% when compared to MRI….and that includes all levels of density. If we restrict this to Level D density (extreme), then mammographic sensitivity for these women is around 20%. Stated alternatively, if mammograms tried to get FDA approval today, they would be rejected for Level D women and perhaps Level C as well.
When breast MRI first came on the scene, the “father of breast MRI,” Dr. Steven Harms, asked me to join his team of radiologists who educated the rest of the country’s radiologists on the use and interpretation of breast MRI. We traveled to so many cities I can’t even recall where. Nevertheless, one of my jobs as the token surgeon was to “crack the 90%”…”Do that 90% thing you do about mammography” wherein I would share the bizarre story of how we ended up with a grossly exaggerated number that doctors and patients grabbed onto, then refused to let go.
Back to the FDA. What are they doing sticking their collective noses into how we report screening mammography results? Well, little known fact — breast imaging became the first radiologic test governed by the FDA in the early 1980s. Sad to say, when mammograms were introduced in the 1970s, “mammography factories” opened up, and did a terrible job of quality assurance. Horror stories were often exposed on the evening news. The FDA took over, and still to this day, every breast center has to be FDA approved, then undergo a yearly inspection.
Here’s where it gets weird. For years, through state legislation (Oklahoma passed its law about 5 years ago), nearly every state in the Union already had this legislation in place (hard to believe it required legislation of any type, isn’t it?) So, the FDA has made it 100% (all states) and has standardized the reporting. So what? The final few states now have to comply, but that’s about it. There are no guidelines for these women who are left floundering, and are not counseled accordingly. It’s not an easy subject, and I won’t dwell on the mountains of misinformation that are out there for women with dense breasts. Oh, yes, the American College of Radiology issued guidelines, recommending the addition of breast ultrasound to mammography for women with Level C and D mammography, and this is all well and good. Problem is, we live in an upside-down world where the expert providers of medical service are not allowed to establish their own guidelines because of “known bias.” So, neutral organizations have to take the leap and confirm the need for additional imaging in women with dense breasts because third-party payors won’t pay attention to the expert clinicians.
Now, let me take it to a different plane. A few states (e.g., Connecticut) have lobbied for insurance coverage for screening US added to mammography for women with dense breasts and have WON. Great. But compliance is a huge disappointment (they might be up to 25% by now). Everyone drank the Kool-Aid of 90% sensitivity for mammos years ago, and now, even when covered by insurance, women settle for mammograms alone, and ignore those who are shouting: “Mammograms are not enough!”
This is why, btw, that I continue blood test research (currently with Syantra, Inc) so that women who have settled for mammograms alone will be inspired to proceed with additional imaging if the blood test is positive.
With my research career still active, we are ramping up for a definitive clinical trial of SyantraDX in high-risk and/or high-density women with normal mammograms who would then undergo additional imaging if blood-test-positive.
My coverage of the true complexity of the breast density problem and the fascinating history of the 90% can be found elsewhere on this web site, or in either of my two breast cancer books. It’s a fascinating story, especially the breakthrough in awareness that began with a grassroots movement prompted by one woman whose mammograms failed her miserably. Read about it here:
Mammography and Early Breast Cancer Detction: how screening saves lives
The Best Breast Blogatorials (essays taken from this web site)
I continue to fight the fight. MRI breast cancer screening guidelines were flawed from the beginning in 2007, and I’ve tried my best to draw attention to major issues that need to be corrected. I even served on the Screening Guideline Committee for the American Society of Breast Surgeons, and still couldn’t conjure any major changes. My major point is that breast density barely fits into screening algorithms, in spite of all the attention it gets today. But if you’re proposing a two-method imaging strategy, shouldn’t the first question be: “What is the probability that mammograms will fail?” That probability is based on density…yet the question is never asked using current guidelines. Instead, density is relegated to a modest risk factor (which is unrelated to whether or not a cancer will be found on MRI).
Evidence that clinicians are coming out of hibernation on this issue has finally arrived — The DENSE trial in The Netherlands is now reporting results greatly favoring MRI over mammography based on a single criterion — breast density. If you have Level D mammograms, you qualify. Very simple. And from that trial, the European Society of Breast Imaging, in 2023, changed MRI screening guidelines to include women with no other risk factors than Level D mammograms, to be done every 2-4 years in addition to mammography.
This idea of extending the intervals for MRI for women at lower risk levels (although density is a modest risk in and of itself) was first proposed in 2008 where a point system gave equal weight to density and to risk. In this system, the points are added, and the total number dictates the interval for MRI — every 1, 2, 3 years or not at all. This point system was published in Seminars in Breast Disease (now defunct) by my radiologist colleague, Dr. Rebecca Stough and me in 2008, a full 15 years before these new MRI screening guidelines with longer intervals from Europe.
At the time, we had our initial experience with MRI to go on, but after that, it was common sense, often dismissed as “weak” rational thought in this era of evidence-based medicine. That said, our system that focused on density yielded DOUBLE the number of MRI-detected cancers as the American Cancer Society’s guidelines. Those 2007 guidelines from the ACS remain the same. Oher organizations in the U.S. have made modifications to the original, but have not corrected most of the issues we have brought to light in a series of critical editorials.
A few weeks ago, we had an Editorial published in the online journal CUREUS, calling for a complete overhaul of the MRI screening guidelines in order to allow density to play a greater role.
Here’s the link to the new article about breast density and MRI guidelines: bit.ly/3Mh22uz
ABH
PS: I wonder if there has ever been simultaneous publication of a medical article and a medical novel at the same time? If you’re more interested in my novels, esp the new trilogy by “John Albedo, here’s the link to the book trailer: https://bit.ly/3lzaRot
Early Breast Cancer 