15th Milan Breast Cancer Conference
Breast cancer screening with MRI versus mammography
Dr Christiane Kuhl - University of Aachen, Germany
My main message actually was that breast MRI so far appears to be the most promising tool, not only because it offers the highest sensitivity of all imaging methods that we have right now at hand but specifically because it offers a view into a tumour pathophysiology and thereby offers the possibility to diagnose breast cancer that is particularly biologically aggressive. In other words, MRI has a tendency to diagnose breast cancer that has active angiogenesis and has it, so to speak, on the genomic tool box to produce proteases, all aspects that improve cancer’s ability to thrive and to metastasise. So you could say that the better a cancer feels, the better the cancer is able to improve its increased demands for nutrients and oxygen, the better this cancer will be visible with MRI as opposed to the way we diagnose breast cancer with mammography. The pathophysiological processes that cause the mammographic findings that we pick up to diagnose breast cancer are all associated with slow growth, with apoptosis, necrosis, hypoxia such as architectural distortions which are usually caused by fibrotic processes which in turn are caused by hypoxia and, of course, calcifications which are usually caused by necrosis. So mammography has a built-in, so to speak, technology inherent bias to detect slowly growing tumours which means that we have a bias towards diagnosing slowed growth, so to speak, with mammography and which ultimately results in what we know as the length-time bias. In other words, a breast cancer has a better prognosis when it has been diagnosed through mammography as opposed to a breast cancer which was invisible on mammography. The so-called over-diagnosis is nothing else but a length-time bias put to an extreme, in other words it appears that mammography has a tendency to diagnose cancer which may ultimately not even need treatment.
What is the standard diagnostic procedure?
The standard for screening is actually only mammography. We do know that we underserve women with pure mammographic screening because not all women are alike and not all breasts are alike. If you have an intermediately dense or dense breast then it is well established that the mammographic sensitivity will decrease down to 30% or so, way lower than what would be accepted as a screening tool.
Ultrasound has been suggested to be used in women with mammographically dense breasts, so far however it has been very difficult to establish this on a broader scale because ultrasound is a technology that takes time. You could even argue that if physician time was adequately paid then ultrasound would probably be the most expensive of all imaging methods. Wendy Burke’s trial has shown that an average screening ultrasound takes 21 minutes of a specialised physician’s time and that is certainly prohibitively expensive if you try to perform this on a broader scale.
Do you believe that MRI could be a better form of detection?
There is actually… True, yes, you could use MRI for screening. It has been established as a screening tool in women at increased risk of breast cancer. I would argue that using MRI for screening has two major advantages: one is that it, as I said, is very sensitive so we find a lot of cancers but due to the debate around over-diagnosis and over-treatment we know very well that we don’t have to find all cancers, we have to find the cancers that kill women. Based on what we know today, because of its ability to preferably diagnose biologically active tumours, MRI screening will certainly help improve our ability to diagnose these aggressive cancers which are overlooked on mammographic screening and only become clinically apparent as … cancers. It will probably also help avoid the over-diagnosis problem which is an integral part, so to speak, of mammographic screening programmes, just by way of detection of cancer.
Is it true that MRI is more effective in younger women?
MRI, or the gradient between the sensitivity that an MRI offers and the sensitivity that a mammogram offers is independent of patient age or risk. It depends just on the breast density, so to speak, in other words the likelihood with which the mammogram is able to depict the cancer. If a woman has an involuted breast then usually the mammogram can offer very high sensitivity. It is true that on average older women tend to have less dense breast tissue but it is by far no direct one to one correlation. There are enough women, even post-menopausal women, who have dense breast tissue. As we know today, dense breast tissue doesn’t only mean that this specific breast is difficult to be diagnosed with mammography but also that the woman carries an increased risk of breast cancer. So, I’ll put it this way, actually the sensitivity gradient between MRI and mammography depends on the breast composition rather than other factors, other factors like age are actually surrogate for breast composition. Then it’s important to note that the ability to diagnose DCIS, specifically high grade DCIS, is independent of breast density because even in women with involuted breasts, as long as a DCIS is not induced calcifications you will not be able to diagnose it through mammography, not to speak of ultrasound. So for diagnosing high grade DCIS, and we’ve just heard that these are two completely different biological or molecular pathways, high grade DCIS will progress to high grade invasive cancers. So to diagnose high grade DCIS MRI is certainly the most accurate imaging tool that we have right now.
Why is the MRI used more?
Mainly I think that’s a matter of cost and availability. But it’s interesting, actually, if you listen to the people who have witnessed the evolution of mammographic screening they can tell you that all the reservations that people mention to argue against the use of MRI for screening for everybody are quite the same that were also mentioned back in the 1970s and ‘80s against mammographic screening – it’s too expensive, there are not enough trained radiologists, way too many false positives for mammography, the women won’t tolerate… All these arguments were the same so I think it will be a matter of time until the medical community understands that investing into screening, investing into meaningful screening, in other words investing into screening tools, a tool that helps us identify the breast cancers that kill, is worthwhile.