Days ago, the U.S. Preventive Services Task Force issued a new draft for its recommendations on routine PSA measurements in asymptomatic men. The panel's report is published in the Annals of Internal Medicine. The main findings are two: first, the absence of evidence that routine PSA testing prolongs men's lives, and second, that PSA evaluation may, on balance, cause more harm than good.
Not surprisingly, there's been considerable coverage of this by the media, and some controversy. For decades, many men have had their PSA checked, knowingly or not, by their physicians. The PSA test measures the level of Prostate Specific Antigen, a protein produced and sometimes secreted by prostate cells, normal, inflamed or malignant, into the bloodstream.
As an oncologist, I don't find the panel's recommendations surprising. There's never been strong data to support the hypothesis that routine PSA testing reduces mortality for men in any age group. Prostate cancer is often indolent, a slow-growing kind of tumor for which a "watch and wait" approach may be best, especially when it occurs in elderly men who are most likely, even in the absence of treatment, to die of another cause.
The complication rate of prostate surgery is fairly high, although this "cost" of screening likely varies, depending on the skill of the surgeon. Still, and understandably, there are men who swear by this measurement, whose lives have been, in some cases, saved by early detection of a high-grade tumor upon screening.
For today, I'd like to consider some key differences between breast and prostate cancers, and the potential value of screening:
Breast cancer tends to affect younger patients than prostate cancer.
Based on SEER data, the median age of a breast cancer diagnosis in the U.S. is 61 years. The median age of death from breast cancer is 68 years. For prostate cancer, the SEER data show a median age of 67 years at diagnosis, and for death from prostate cancer, 80 years.
So the potential number of life-years saved by early detection and intervention is, on average, greater for breast cancer than for prostate cancer.
Screening for breast cancer has improved over the past 25 years.
Because the blood test for PSA hasn't changed much in decades, it's reasonable to consider studies and long-term survival curves based on data going back to the 1980s.
Mammography, by contrast, is much safer and better than it was 25 years ago, for various reasons: increased regulation of mammography facilities (more care with the procedure, better training and credentialing of technicians) according to the FDA's Mammography Quality Standard Acts Program; development of ultrasound methods to supplement mammograms in case of suspicious lesions (lessens the false positive rate overall); the advent of digital technology (lessens the false positive rate in younger women and others with dense breasts); more breast radiology specialists (expertise).
The data reviewed by the USPSTF in issuing their 2009 recommendations for breast cancer screening were decades old, and, as I've considered previously, irrelevant to modern medical practices. A recent article in the New England Journal of Medicine points to the problem of the panel's reliance on the Age trial for women in their 40s. That trial involved the obsolete method of single-view mammography.
Mammography involves a woman's consent (in the absence of dementia, a separate ethical issue).
A woman knows if she's getting a mammogram. She may not ask sufficient questions of her doctor, or her doctor may not answer them well, but in the end she does or doesn't enter into a radiology room, volitionally. She decides to get screened, or not. She can choose to have a mammogram every year, or every other year, or not at all.
There's no ethical problem, as reported for some men, of patients learning they have an abnormal PSA, after blood was drawn indiscriminately, without their knowing the test was being performed.
This perspective might, and should, later extend to consider additional differences between these two kinds of malignancies (each of which is really a group of cancer subtypes), a fuller discussion of the impact of treatment on survival for each type, and the relative risks of screening due to differential complication rates of biopsies and other procedures.
To be clear, there's no perfect screening test for either cancer type. Far from it. But the merits and risks of each procedure should be weighed separately, and with care.
This post originally appeared at Medical Lessons, written by Elaine Schattner, ACP Member, a nonpracticing hematologist and oncologist who teaches at Weill Cornell Medical College, where she is a Clinical Associate Professor of Medicine. She shares her ideas on education, ethics in medicine, health care news and culture. Her views on medicine are informed by her past experiences in caring for patients, as a researcher in cancer immunology and as a patient who's had breast cancer.