Mammograms and PSA Tests: To Do or Not To Do

28 Oct Mammograms and PSA Tests: To Do or Not To Do

 You’ve probably noticed the ground shifting on the question of screening for the three major gender-specific cancers—prostate cancer in men, breast and cervical cancer in women. It all comes from a series of earth-shaking reports in the past two years by the federal government’s Preventive Services Task Force, an influential panel of independent experts that evaluates all kinds of widely accepted disease prevention practices.

In October 2011, the task force released a report saying that PSA testing to detect prostate cancer doesn’t save lives and actually leads to many unnecessary and even harmful procedures. The panel also said there’s no reason for most women to have annual Pap tests for cervical cancer—every three years is enough. This followed the task force’s hugely controversial assessment two years ago questioning the benefit of annual mammograms and advising women to consider having them less frequently—and not at all before age 50.

What to do? The questions are far from settled—in fact, the arguments are just beginning. “You don’t turn decades of thought around immediately,” a task force member told Gina Kolata of The New York Times.

Ms. Kolata framed the dilemma well and offered some much-needed context in her article headlined: Considering When It Might Be Best Not to Know About Cancer. But like other coverage of the recent turnabout in thinking, it didn’t do much to help a given person decide whether, and how often, to make cancer screening part of their regular checkups. That’s because there isn’t one answer that’s right for everyone. And that’s why discussing cancer screening with patients has become a daily part of my practice. Routine screening isn’t routine anymore. At least it shouldn’t be.

I welcome the new thinking. As Ms. Kolata points out, many of my peers do not—partly because they and the public at large are “stuck in a sort of cancer time warp.” It reflects a view of cancer at odds with the vast knowledge and understanding we’ve gained from decades of advances in cancer biology.

In 1845, Rudolph Virchow, a German physician, looked at tumors taken from autopsies and famously defined cancer as a disease that spreads until it finally kills. But it was a textbook case of “selection bias”: All the cancers he saw, the only ones, had killed. But our modern ability to peer into the living body relatively non-invasively has opened up a much larger universe of abnormal cells than Virchow could have imagined. Under a pathologist’s microscope, they look very similar to the lethal cancer Virchow saw, but many of them will never go on to metastasize or grow uncontrollably and kill. Some will regress under the surveillance of the immune system or do nothing at all. But it’s still cancer, and most people would probably regard leaving it alone, keeping an eye on it, as a life sentence of anxiety at least, and a form of Russian roulette.

My father was a case in point. He was diagnosed with prostate cancer when he was 74 and decided to have the recommended surgery. Though a very rational college professor who prided himself on “not letting emotions get in the way of making good decisions,” he couldn’t stand the idea of a cancer sitting in his prostate. The surgery significantly reduced his quality of life and was almost surely unnecessary at his age, even if the cancer had grown moderately fast for the eight more years he lived. (He died of Alzheimer’s disease.) It has been estimated from autopsies that more than half the men who die in their 80s of other causes have “cancer cells” sitting in their prostates.  What would Virchow have done with this information?

Neither mammograms nor PSA tests can distinguish between cancers that will kill prematurely and those that will spontaneously regress or just sit there indolently until one dies of some other cause.  This is the sine qua non of a screening test: find a disease early enough so that its course can be altered and be sure that you are not unnecessarily treating a disproportionate number of people.  In short, alter the mortality rate  with an acceptable false positive rate and morbidity from the treatment.  The federal task force found that the screening tools being used have not. But, of course, that doesn’t mean that they don’t save some lives.

The stakes can be very high.  In his Pulitzer Prize-winning book, The Emperor of All Maladies, Siddhartha Mukherjee recounts the devastatingly tragic effects of this “heal with steel” approach to cancer.  Between 1891 and 1981 more than  500,000 radical mastectomies were performed, resulting in an incredible amount of disfiguring and suffering.  When the definitive study to assess the effectiveness of this treatment was finally done, it was found that the vast majority were unnecessary and didn’t save any more lives than breast-conserving lumpectomies and radiation.  But even these less disfiguring procedures are not without considerable pain and suffering. What if the majority of them are unnecessary?

The new analyses demonstrate that 87 to 97 percent of the 138,000 mammograms that “find cancer” find it in women who either have cancers that would never cause them any problems or would kill them regardless of the treatment they subsequently receive.

Given this accumulating body of evidence documenting the shortcomings of mammography and PSA testing, it is heartening that Dr. Otis Brawley, chief medical officer of the American Cancer Society, is now saying, “We need to be a little more rigorous about what we accept about screening.” And we need a better understanding of prostate and breast cancer so that better screening tests can be developed.

Where does this leave us?  Should you dispense with these tests altogether?  As with so many other decisions, there’s no one-size-fits-all answer. For prostate cancer screening, there is certainly good evidence that if your PSA is less than 1 at 60 years old or older, then the chances of your dying from prostate cancer in the next 25 years are less than two-tenths of one percent. (2). If you’re in this lucky group—and there are many—and you’re not prone to inflammation and cancer, I see no reason to continue with PSA testing.  Should a man who is under 50 begin PSA testing?  In the absence of a strong family history, it is perfectly rational not to, though that’s a personal choice that I leave to my patients.

For women, the new recommendations–not starting mammograms until after age 50 and doing them every other year (to avoid excessive radiation exposure and pain)—is also rational, but choosing not to get one at all is not irrational either. Part of the equation is the incidence of false positives and the morbidity attached to tracking them down versus not knowing.  This is a personal decision that can be influenced by family history.

I believe that the next paradigm shift in cancer screening will be much more revolutionary than just saying no to mammograms and PSAs, and potentially far more effective.  Aging adults are making millions of new “abnormal” cells every day, the result of the normal processes of metabolism and inflammation that  constantly bombard our DNA with free radicals that cause mutations.  The vast majority of these damaged cells are repaired, kept latent, or destroyed by the repair and maintenance mechanisms of our immune systems. There is something about the general biological milieu of one person that keeps these abnormal cells from killing them that differs from that of another person who ultimately succumbs to cancer. Age is one factor.

Judith Campisi of the Buck Institute for Aging Research has done elegant experiments in which cancer cells from tumors in an old mouse were transplanted into a young mouse. The cancer cells did not grow into tumors (3).  But when the reverse experiment was done—cancer cells transplanted from the young mouse to the old mouse—they grew uncontrollably. There is something suppressive to tumor growth in the younger mouse that turns permissive in the older mouse’s milieu. The factors include the proportion of senescent to naive T cells, short to long telomeres, inflammatory to anti-inflammatory cytokines, and free radical generation to antioxidant capacity.

In the coming years, rather than focusing on abnormal cells medicine will target the poorly aging immune system and supporting tissues by monitoring biomarkers. This will decrease the chance that cancer will develop or better enable us to target our screening toward higher risk individuals. In future posts, I’ll be writing more about advances in these exciting areas of age management medicine and cancer prevention.

1.       Haas GP, Delongchamps N, Brawley OW, Wang CY, de la Roza G. 2008. The worldwide epidemiology of prostate cancer: perspectives from autopsy studies. Can J Urol 15: 3866-71

2.       Vickers AJ, Cronin AM, Bjork T, Manjer J, Nilsson PM, Dahlin A, Bjartell A, Scardino PT, Ulmert D, Lilja H. 2010. Prostate specific antigen concentration at age 60 and death or metastasis from prostate cancer: case-control study. BMJ 341: c4521

3.       Krtolica A, Campisi J. 2003. Integrating epithelial cancer, aging stroma and cellular senescence. Adv Gerontol 11: 109-16