Cancer screening is looking for trouble, in the most literal sense. By definition, medical screening takes place before there are any signs or symptoms of disease, when someone is feeling fine. If looking for cancer in someone with no evidence of any health problem at all isn’t looking for trouble, I don’t know what would be.
But despite the precautionary implications of that expression, looking for trouble in medical contexts can be a very good thing. The intent, of course, is to find a problem before it finds us, when it is far more manageable, far less ominous and potentially downright curable.
That is the principle that underlies cancer screening, but the practice of it has proven far more complicated. In essence, we have gone looking for trouble, and all too often, trouble is what we’ve found.
For instance, we know that a very high percentage of men who die past the age of 80 die with, but not from, prostate cancer. In other words, if we men live long enough, we will eventually develop some cancer in our prostate, but there is a good chance it will never cause us any grief. What, though, if that very same cancer, destined to remain quiescent in some nook or cranny of the prostate, is found years earlier with a screening test, such as the PSA assay? Then we may feel obligated to treat what we’ve found, with a considerable risk that the “cure” will be far worse than the disease ever would have been. On the other hand, some prostate cancers are destined to metastasize, and will potentially be lethal if not caught early.
Our principal limitation at present is telling these two varieties apart. Until we can do that reliably, prostate cancer screening is a questionable enterprise; it’s not clear that it produces a net benefit rather than net harm. The United States Preventive Services Task Force, widely considered the gold standard in this area, recommends against it. I have decided against it for myself thus far.
At the other end of the cancer screening spectrum are the situations where screening reliably finds abnormal cells even before cancer is established, and creates a decisive opportunity to banish the attendant peril. Both cervical cancer screening and colon cancer screening fall in this category, with recourse to Pap smears and colonoscopies, respectively.
For quite some time, we had the impression that breast cancer screening, with a combination of self-exam, professional exam and in particular mammography, was reliably beneficial as well. But when looking for trouble, unintended consequences may be more the norm than the exception, and this has proven to be the case for breast cancer. Although I know women who have found breast cancers through self-exam, the formal analysis of that approach has suggested it to be nearly useless at the population level. Mammography is a far more muddled issue than we would wish, finding cancer early at times as intended, but also generating a large number of false positives, and perhaps even at times suggesting treatment is needed when it is not. Even breast biopsies, thought to be the final arbiter of reliable diagnosis, have been shown to be fallible.
But as we all know from epidemiology — and all too many of us know from personal experience involving women we love — breast cancer is a clear and highly prevalent danger. Consequently, most of us opt for screening despite the limitations; and although the recommendation is somewhat qualified, the US Preventive Services Task Force advocates for screening as well.
From my perspective as a preventive medicine specialist, and as someone who has dived pretty deeply into the statistical underpinnings of cancer screening in several editions of an epidemiology textbook, the general considerations pertaining to breast cancer screening establish the context for decisions about dense breast screening protocols. On that basis, I challenge a commentary just published in the Journal of the American Medical Association suggesting that breast density notification laws, and associated screening protocols, are ill-advised. I support them.
As noted in JAMA, revised screening protocols for dense breast tissue are largely due to the tireless efforts of Dr. Nancy Cappello, herself a survivor of cancer in dense breasts, missed by mammography and found at a late stage. Turning that adversity into opportunity as only the true heroes among us ever do, Nancy founded the charity Are You Dense to protect other women with dense breasts from the false sense of security mammography was apt to provide them. Standard mammography is ill-suited to detect the early signs of cancer in dense breast tissue, which is present in as many as 40 percent of all women.
Nancy’s efforts, along with those of many others she has engaged along the way — some famous and some not — have resulted in legislation in 24 states to mandate notification to women about dense breast tissue, and coverage for a modified screening protocol involving ultrasound. National legislation is in the works.
The legislation does not require that women get screened for breast cancer, of course. The pros and cons of cancer screening are, as they have always been, subject to the customized decisions that play out in doctor/patient dialogue. Most of my female patients have opted for breast cancer screening even after discussion of the limitations; a minority have made the opposite decision. Legislation related to dense breast screening does nothing to change this dynamic, it merely ensures that if screening is to be conducted, it be the right screening for the job, and that insurers cover its costs accordingly.
The writers in JAMA challenge this proposition. They note, correctly, that population-level data regarding the net effect of modified screening protocols for dense breasts are lacking. They point out that the US Preventive Services Task Force has thus far concluded that data are insufficient to conclude anything about the specific merits of dense breast screening. On this basis, the writers seem to be suggesting that the legislation is overwrought, and heavy-handed.
With all due respect to my colleagues, they are wrong. They are making a common mistake: conflating absence of evidence for evidence of absence. They are also failing to allow for the population of gaps left by science with the obvious ingredient: sense.
The whole point of breast cancer screening, in anyone, is to find breast cancer early. If mammography fails to do this when breast tissue is dense, it is the wrong approach. If ultrasound (or other imaging) can find early lesions in dense breast tissue just as mammography can in breasts of lesser density, then it is merely leveling the playing field, subjecting women with dense breasts to the same risk/benefit trade-offs screening imposes on everyone. While we certainly should be gathering data to demonstrate the net effects of dense breast screening protocols, it makes no sense to insist on only futile imaging in the interim. If we are going to look for trouble, we need a test that is not blind to it.
As for the US Preventive Services Task Force, they are strictly bound by the highest standards of evidence. The result is that they are often obligated to conclude that they can’t reach a conclusion, because the evidence is not yet fully ripe. But while inconclusiveness can work in public health policy, it cannot work in patient care: any given woman will either be screened for breast cancer, or not. If her time is now, it does her no good to know that the authorities will reach a conclusion … some day. For an anonymous “public,” someday can be some day. For any given woman, someday is today.
The urgencies of today always require we do the best we can with the knowledge we have. That knowledge is never perfect, and all too often, far from it. Debate about standard mammography is far from settled. We have no basis to expect better regarding modified protocols. But science works best when conjoined to sense, and the combination tells us that absence of evidence is not the same as evidence of absent effects. The combination tells us that if we are to screen for breast cancer, we should use imaging modalities that can find it, and that those are the methods our insurance should cover.
The JAMA authors tell us it is premature to adopt, and mandate, modified dense breast screening protocols; in essence, they suggest that we may be dense if we don’t, but we are dense if we do. I disagree, and stand emphatically with Are You Dense. While waiting for the scientific data we lack, we must apply sense to applications of the science we have. Anything other than that is, indeed, rather dense.
More from U.S. News
What Not to Say to a Breast Cancer Patient
A Tour of Mammographic Screenings During Your Life
16 Health Screenings All Women Need
Dense Breast Screening: Dense if We Do, Dense if We Don’t? originally appeared on usnews.com
