In this interview we discuss PSA screening for prostate cancer, the compromised results of the PLCO trial, and more.
Oncology (Williston Park). 30(8):694, 764.
Laurence Klotz, MD
1.Recently, doctors from New York-Presbyterian/Weill Cornell challenged the results of a landmark study (the Prostate, Lung, Colorectal, and Ovarian Cancer [PLCO] Screening Trial) that served as the basis for the US Preventive Services Task Force (USPSTF) 2012 recommendations against routine prostate cancer screening. Do you agree that the study was flawed? What are the major concerns with the outcomes?
DR. KLOTZ: It is widely acknowledged that the PLCO study was significantly compromised by contamination. The rate of prostate-specific antigen (PSA) testing in the control arm was approximately 85%. Indeed, the rate of noncompliance with testing in the study arm was 15%; thus, about the same proportion of patients were tested in both arms! PSA testing was significantly, not slightly, more frequent (per patient) in the study arm. Thus, PLCO was a study of programmatic testing vs ad hoc testing, rather than screening vs no screening, so the lack of a significant difference in prostate cancer mortality was not surprising. The USPSTF lumped the results of PLCO and the European Randomised Study of Screening for Prostate Cancer (ERSPC), leading to a negative view of the mortality benefit. This was an error. The ERSPC study, which had a dramatically lower contamination rate, showed a consistent mortality benefit that has increased as the study has matured. Had the USPSTF looked at the mortality benefit from the perspective of ERSPC alone, they would likely have come to a different conclusion about the value of PSA screening. The USPSTF also did not take into account the dramatic fall in prostate cancer mortality since the introduction of PSA screening. Virtually all researchers who have examined this phenomenon have concluded that it is largely (albeit not solely) due to screening.
2.As a result of this new information, should PSA testing be re-evaluated in terms of its role in detecting prostate cancer?
DR. KLOTZ: Yes. The fact that PSA screening results in a 25% to 30% mortality reduction (at the least; perhaps more with longer follow-up) should be accepted by the healthcare community and incorporated into guidelines regarding screening. Two other major changes have also occurred, which improve the outcome of screening significantly: active surveillance has been widely adopted for low-risk prostate cancer, sparing about 40% of newly diagnosed patients the side effects of therapy; and selective, risk-adapted, infrequent testing has been demonstrated to improve the metrics of screening. For example, an initial PSA < 1.0 in men in their 40s indicates that PSA testing can be done every 5 years without incurring any significant risk of delayed diagnosis. A PSA < 1.0 in a 60-year-old man similarly indicates that testing can be discontinued. Both of these practices will reduce overdiagnosis substantially and improve the outcome of screening.
Criteria and procedures for screening, biopsy, and treatment vary; these variations will lead to different outcomes with respect to mortality reduction and overtreatment. Most of the effects of the PSA test are not a function of the test itself; rather, they are a function of the downstream effects of diagnosis and treatment. Although a diagnosis of prostate cancer is anxiety-provoking, it has less impact on quality of life than the adverse effects of treatment (ie, erectile dysfunction and incontinence).
3. What are the current pros and cons of PSA screening?
DR. KLOTZ: PSA screening results in the diagnosis of localized prostate cancer at a point in the disease process when it is more likely to be curable. The benefit is a significant reduction in the risk of dying of prostate cancer. There is also substantial benefit from the reduced risk of complications related to disease progression in those men who ultimately die of other causes. The risks are related to overdiagnosis and overtreatment of prostate cancer. The number needed to diagnose in the ERSPC trial for each mortality, even avoided, was 27. Thus, many patients will be treated for each death avoided. However, the increasing acceptance of active surveillance, and the emergence of new, less morbid treatments, particularly focal therapy, offer the promise of reducing the number needed to treat to very acceptable levels.
4.What should physicians and patients take away from this controversy, and how do you think it will affect clinical practice for prostate cancer going forward?
DR. KLOTZ: Issues surrounding screening are complex and involve many factors, including public health, economics, and social and cultural considerations. Screening for prostate cancer, viewed through an informed lens, is appealing, and results in a net benefit to the individual. It must be used in conjunction with conservative management of low-risk disease, and should be offered infrequently to most men, and more frequently to those at risk. Another game changer is the increasing role of MRI in identifying patients with more aggressive disease, and increasingly replacing biopsy for many men.
I expect PSA will have a resurgence. It is probable that MRI will replace biopsy as the initial diagnostic test in men with an elevated PSA, and biopsy will be reserved for those with a positive MRI, or very-high-risk men with a negative MRI. Screening used in conjunction with active surveillance for low-risk disease; focal therapy employed for many patients with unilateral, small-volume, intermediate disease; and radical therapy reserved for those with aggressive, large-volume, intermediate-to-high-risk cancer will dramatically alter the metrics of screening and promote a reconsideration by policy-makers (like the USPSTF).
Alternatively, if policy-makers don't change their recommendations and PSA testing continues to be progressively abandoned by primary care physicians, the effect is likely to be an increase in prostate cancer mortality in the United States and Canada. There is already evidence that aggressive prostate cancer is being diagnosed less frequently as a result of less PSA testing, meaning that these cancers are no longer being diagnosed early, and will likely show up later as incurable, fatal disease. This mortality increase will take 5 to 10 years to become apparent. In 15 to 20 years, when the effect is incontrovertible, the task forces will be forced to reconsider. But many lives will have been lost unnecessarily in the interim.
Financial Disclosure:The author has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.