Prostate Cancer Screening
Should We or Shouldn’t We? The Debate Continues…
IN RECENT MONTHS, A VERY PUBLIC DEBATE ABOUT THE EFFECTIVENESS AND APPROPRIATENESS OF PROSTATE CANCER SCREENING HAS BEEN TAKING PLACE. IN OCTOBER 2011, THE U.S. Preventive Services Task Force (USPSTF) issued a draft recommendation against screening. Since then, the medical community has been deeply divided over the issue. The opinions of Levine Cancer Institute physicians, like the medical community at large, reflect this divide. In the spirit of debate, Derek Raghavan, MD, and Chris M. Teigland, MD, have collaborated to produce these differing opinions.
PRO USPSTF RECOMMENDATIONS
By Derek Raghavan, MD, PhD, FACP, FRACP President, Levine Cancer Institute
“SCREENING”SPECIFICALLY RELATES TO the detection of disease in a person with no symptoms or signs; for example, a mammogram in a woman without symptoms, a breast mass or relevant family history. The key point of screening for any disease is to save a life or to prolong a life, not just to find it earlier. For example, if an exercise electrocardiogram for heart disease in an asymptomatic patient were to reveal narrowing of four vessels, leading to coronary artery bypass graft, and if the procedure were to cause an operative death, then the screening process would have accidentally shortened his life, and thus have failed. Because of this issue, when one introduces a new screening test for any disease, it has to be validated by a trial that demonstrates that a group of patients who underwent screening had better overall survival than the group who didn't undergo screening, and it's not sufficient to have improved outcomes relating only to death (in this case) from heart disease. For many years, it seemed logical that a routine chest X-ray would help improve survival for patients with lung cancer, figuring that an earlier diagnosis would improve survival. It turned out, when tested in a clinical trial, that the idea was false. Although lung cancers were diagnosed a little earlier than without the chest X-ray, the test wasn’t accurate enough to find lung cancer early enough to improve survival when treatment was given by surgery, radiotherapy or chemotherapy. As a result, patients were no longer screened by annual chest X-rays. More recently, spiral CT scans, which actually can detect lung cancer at an even earlier time, have been shown to improve survival among habitual smokers in a randomized trial.
So, how does this relate to the debate on prostate cancer and PSA screening? As Dr. Teigland notes, PSA is a protein released into the blood by a number of prostate diseases, including prostate cancer, and has often been very useful in helping to identify the presence of prostate cancer or in following its course after treatment. The problem is that noncancer diseases (benign prostatic hyperplasia, inflammation) cause PSA levels to rise, and some cancers don’t produce PSA at all! Prostate cancer is a highly variable disease, with some cases occurring and growing quickly, but the majority are very slow growing. It has been estimated that 50 to 60 percent of men ages 70 and older have prostate cancer, but only a tiny proportion of those patients actually are harmed by the disease, and many men die of other causes without their prostate cancers ever showing themselves clinically (i.e. they don’t spread and don’t cause symptoms). To add to the problem, we still aren't very good at determining which cancers are dangerous and which are not, and as a result, some patients may actually have been treated for the nondangerous type of prostate cancer. The particularly difficult patient group comprises men who have Gleason scores that range between 6 and 7.
Several randomized clinical trials have been reported with half the patients allocated to a PSA-screening procedure and half continuing their lives normally without having PSA tests ordered. The results have been very controversial. Of importance, some of the patients in the “nonscreening” group actually decided later to have screening tests done, which probably altered the outcomes in favor of the nonscreened group. Despite that, there‘s no doubt that most of the studies have shown a reduction in the number of deaths from prostate cancer in the screened group, a clearly desirable result. However, none of the studies have shown a reduction in deaths overall, meaning that the patients in the screened group have had a higher death rate from other causes. It‘s still not clear why that has occurred, and these studies will have to be followed longer for a final explanation. As a result, the USPSTF issued its recommendation against routine screening in the community.
Where does that leave us? I agree with Dr. Teigland that the USPSTF has been extreme in its view and may actually cause harm. There are two very important groups that were not well covered by the trials: men with a family history of prostate cancer and African-American men. Both of these groups are known to be at increased risk of the disease, and African-Americans seem to have a higher death rate than Caucasians. Dr. Teigland and I agree that both groups should undergo screening until evidence is available that indicates whether that approach is truly beneficial or perhaps even that it doesn’t help improve their survival (in which case, a new recommendation should be made).
Where we disagree is on the topic of men with no family history, no symptoms and who come from relatively low-risk groups. My view of the trials is that they have not shown overall survival benefit, meaning that the test doesn’t save lives overall. It‘s excellent news that it reduces deaths from prostate cancer, but if the overall death rate isn’t improved, then it is not a beneficial test. PSA screening does not allow men in the screened group to live longer than those in the nonscreened group. We don’t know why that is the case, but it is the current situation. The reality is that, despite the introduction of PSA screening, the absolute number of deaths from prostate cancer in America has not changed dramatically in the past 30 years; what has changed is the number of “reservoir” cases (i.e. those heretofore undetected, and which mostly never caused clinical problems) that have been found, and these have artificially inflated the denominator of total cases.
CON USPSTF RECOMMENDATIONS
By Chris M. Teigland, MD
THE USPSTF‘S RECENT DRAFT RECOMMENDATION AGAINST THE USE OF prostate-specific antigen (PSA) for prostate cancer screening has brought a long-standing controversy to a boil.
The flaws of PSA screening are many and well known by both primary care physicians and prostate cancer specialists:
Yet physicians practicing in America over the past 20 years, since PSA screening has been widely adopted, have seen the absolute number of deaths from prostate cancer fall by more than 35 percent and the rate of metastatic disease at presentation fall from more than 25 percent to less than 2 percent. Indeed, a number of CHS’s Carolinas Physicians Network primary care physicians have shared with me that in their own consistently screened patient population, they have never had a prostate cancer death.
The USPSTF‘s recommendation against widespread PSA screening suggests that the weight of the epidemiologic evidence firmly supports the abandonment of our major tool in the fight against the second-leading cancer killer in men. This is simply not the case.
The USPSTF report is based on a meta-analysis of three randomized, controlled screening trials: PLCO, ERSPC and the Göteborg screening trials. While the PLCO trial showed no benefit to screening, the ERSPC trial showed a 27 percent reduction in prostate cancer mortality after nine years and the Göteborg study produced a 56 percent reduction after 14 years. Furthermore, in ERSPC the relative risk of developing metastatic disease was reduced by 41 percent.
Unfortunately, the USPSTF equally weighted the value of the evidence presented by these three trials even though the PLCO trial is so methodologically flawed as to be of questionable value. Fifty-two percent of the control group was in fact PSA screened (“drop-in participants”) and 40 percent of the screened group found to be PSA positive was never biopsied!
However, even if one discounts the findings of the USPSTF, the knowledgeable and concerned physician must deal with the clear problem of overdiagnosis and overtreatment of clinically indolent cancers caused by excessive PSA screening and transrectal ultrasound prostate biopsies. This is being done today by the use of PSA velocity, PSA density, percent free PSA and the use of prostate cancer risk calculators such as the one derived from the Prostate Cancer Prevention Trial (http://deb.uthscsa.edu/URORiskCalc/Pages/calcs.jsp).
Furthermore, current National Comprehensive Cancer Network guidelines for the management of prostate cancer emphasize the use of active surveillance for patients who have very low-risk and low-risk disease.
PSA screening presents significant clinical challenges if it‘s to be used appropriately. The black or white approach of completely abandoning its use is neither an appropriate response to an accurate assessment of the evidence nor is it fair to our patients.
This issue is far from being resolved. The topic will continue to be the subject of media coverage and medical debate as experts weigh in and the recommendations are finalized, but what both Dr. Raghavan and Dr. Teigland emphasize is that this is a complex and confusing issue. All men, especially men ages 40 to 50 or older, should certainly seek advice from their personal physicians/urological cancer experts so they can make educated, relevant choices for their unique situation.
See More About Prostate Cancer Research
Watch Dr. Raghavan’s video interview about the progress in prostate cancer research and treatment from the American Society of Clinical Oncology’s website.