Prostate Cancer: To Screen or not To Screen? That is the Question.
Prostate Cancer: To Screen or not To Screen? That is the Question.
ABSTRACTS & COMMENTARY
The principles of cancer management that have emerged over the years are straightforward. First, prevent cancer if at all possible. Second, if you can't prevent cancer, detect it early before it causes symptoms because it will certainly be easier to cure when it is small and asymptomatic than when the tumor burden is large and symptomatic. It is virtually a universal in cancer that curability is inversely related to tumor burden. This is kid's stuff; everyone knows this.
We know we can prevent a number of cancers. Don't smoke, stay out of the sun, stay away from asbestos, exercise three times a week, eat a well-balanced diet rich in fruits and vegetables and low in saturated fat; these active choices will significantly reduce the incidence of cancer. For a couple of common cancers (breast and colon), screening of the general population above age 50 has been demonstrated to save lives in a cost-effective fashion.
But what about prostate cancer? What approach to its diagnosis and management is supported by evidence? We know that prostate cancer is the most common cancer in men. We know that it increases in incidence dramatically with age. We know that prostate cancer is associated with increased serum levels of a rather specific marker, prostate-specific antigen (PSA). So it is clear what needs to be done; all men over age 50 should annually undergo a digital rectal examination and have their serum PSA level measured, regardless of whether they are symptomatic. Right?
Well, many of us appear to think so. Fowler and colleagues conducted a survey of 444 primary care physicians and 394 urologists to assess their practices regarding the screening of patients for prostate cancer. About 90% of primary care physicians said they almost always perform digital rectal examinations on their male patients between the ages 50 and 80 years. Eighty percent routinely do digital rectal examinations on men older than age 80. Urologists agreed with this practice in men aged 50-75 years. About 75% supported the idea of doing rectal examinations to assess the prostate in men younger than age 50 and older than age 80.
Most primary care physicians ordered serum PSA levels on all males older than 50 years and ordered the test with the same frequency in men older than 80 years. By contrast, fewer urologists recommended PSA testing in older men; only half thought this test was useful in men 75-79 years, and 25% recommended routine PSA testing in men older than 80 years. Similarly, urologists' enthusiasm for prostate biopsy was greatly influenced by the age of the patient. Only 3% of urologists would recommend a prostate biopsy for an 80 year old patient whose PSA was between 4 and 10 ng/mL; half of the primary care physicians claimed that they almost always refer patients older than 80 years with elevated PSA levels for biopsy.
Surprisingly, many primary care physicians (63%) who believe that treatment (either radical prostatectomy or radiation therapy) is unlikely to benefit someone with less than 10 years of life expectancy nevertheless perform PSA tests at least half the time on their patients 80 years or older. Why do they do a test to find a disease for which they believe that standard therapy is ineffective? The answer is not known. However, the facts are clear. In another study, Meigs and colleagues surveyed 33,028 male health professionals without prostate cancer between the ages of 47 and 85 years for whether they had received prostate cancer screening. Thirty-nine percent of men in their 50s and 53% of men in their 80s had had PSA levels determined in the past year. The likelihood of PSA testing was greatly increased in the presence of symptoms of benign prostatic hyperplasia.
Thus, screening for prostate cancer is widespread. It is common in men beginning at age 50 and it appears to increase in frequency with age, with the frequency being highest in men 80 years and older. Despite skepticism about the efficacy of radical prostatectomy (which is greater in primary care physicians than urologists) and radiation therapy (which is greater in urologists than in primary care physicians), prostate cancer screening continues in men who are believed to be unlikely to benefit from treatment. (Fowler FJ, et al. Am J Med 1998;104:526-532; Meigs JB, et al. Am J Med 1998;104:517-525.)
COMMENTARY
What do we know for sure? No prospective, randomized trial has yet demonstrated that screening for prostate cancer saves lives. Given that the disease is common and that a serum screening test is available, that sounds ridiculous, but it is true. Several features of prostate cancer make it difficult to know whether to screen for it. First, serum PSA elevations are not specific for prostate cancer; they also occur in a much more common malady, benign prostatic hyperplasia. However, beyond the PSA elevation and the fact that both entities can cause symptoms of urinary outflow obstruction, prostate cancer and prostate hyperplasia have little in common, and, in particular, prostate hyperplasia is not a premalignant state. Prostate hyperplasia tends to occur centrally in the gland near the urethra; this is why quite modest hyperplasia can produce serious symptoms. By contrast, prostate cancer is often multifocal and begins more peripherally in the gland, often not producing symptoms until the disease has become quite large. While prostate cancer can occur in a gland that is affected by benign hyperplasia, it also occurs in nonhyperplastic glands. Indeed, the prevalence of prostate cancer is not higher in men with benign prostate hyperplasia than in age- and ethnic group-matched asymptomatic men.1
Second, the nature of prostate cancer is unusual. While the cancer is highly prevalent, it appears that most of the cancers are clinically unimportant. It has been estimated that for a 50-year-old man with a 25-year life expectancy, the lifetime risk of microscopic prostate cancer is 42%; the lifetime risk of prostate cancer that is clinically symptomatic is only 10%, and the risk of the prostate cancer being fatal is only 3%.2
Third, the approach to the cure of prostate cancer confined to the gland is no picnic. Radical prostatectomy is a major surgical procedure with acute and chronic complications. Permanent sexual dysfunction is reported in about one-half of the men who undergo this procedure, and 25% have some degree of incontinence. Radiation therapy may have fewer long-term complications (10% incidence of bowel dysfunction), but it is also less effective according to some series. Even the costs and anxieties associated with working-up an elevated PSA or abnormal digital rectal examination are significant. About one in four asymptomatic men will have an abnormal PSA or rectal examination, and two-thirds of these will be false-positives. Among 1000 men with symptomatic prostate hyperplasia, 400 will have elevated PSA levels or abnormal rectal exams and only 30 of these 400 will have prostate cancer.3 Thus, a large number of men will undergo unnecessary procedures at some expense and a small percentage of them may experience either bleeding or infection from the biopsy.
What should be done about this? Perhaps the most important issue is to learn how to distinguish prostate cancers that will kill you from prostate cancers that won't. In the staging system for prostate cancer, stage T1 is used for cancers discovered incidentally that are not palpable; thus, T1a is nonpalpable disease found incidentally in a transurethral resected specimen obtained for relief of prostate hyperplasia symptoms in which less than 5% of the resected tissue is involved. T1b is the involvement of more than 5% of the resected tissue and T1c is disease detected by PSA screening. The natural history of untreated stage T1a disease is just beginning to be reported. In one series, 8% of men with T1a disease developed metastatic disease within up to 15 years of follow-up.4 In another series of 102 men, 55 died of intercurrent disease, 28 were lost to follow-up, 14 died of unknown causes, 14 were documented to have developed clinical progression of prostate cancer, and four died from it. Of the 54 men followed for more than eight years, eight (15%) developed clinical progression.5
The group of cancers detected by elevated PSA levels are more heterogeneous. When radical prostatectomy specimens are classified by size and Gleason grade, 29% were insignificant or minimal, 52% were moderate (0.5 cc3vol, Gleason grade < 7), and 19% were advanced (capsular penetration).6 Certain clinical characteristics can be used to determine whether the tumor is significant: Gleason score of 7 or greater, three or more needle cores involved with cancer, 50% or more involvement of any single needle core, and PSA density (i.e., PSA level divided by volume of the prostate on CT or MRI scan) 0.15 ng/mL/gm of prostate. When these criteria were used to select patients for operation, only 16% of operated patients had insignificant tumors.
In addition to these clinical criteria, other efforts have been made to refine the use of PSA levels to discern life-threatening from non-life-threatening prostate cancer. For example, the ratio of free to total PSA in the serum is an early marker of aggressive cancer. The amount of free PSA decreases in more aggressive cancers. Thus, when less than 13% of the total PSA is free, the patient is very likely to have an aggressive cancer; this index detects clinically significant cancers 10 years before symptoms develop based on a retrospective analysis of banked serum on men who developed aggressive cancers.7 Thus, an elevated total PSA with a low fraction of free PSA may identify patients most likely to benefit from radical surgery.
Ongoing clinical trials are addressing the predictive value of PSA refinements and clinical predictors of disease aggression. However, until such data are available, random PSA screening is difficult to justify. It should not be done in men whose life expectancy is less than 10 years. It is useful in someone who has already been diagnosed with prostate cancer. However, the early detection of clinically insignificant disease is not an advance, and subjecting such patients to curative therapy may do more harm than good.
References
1. American College of Physicians. Ann Intern Med 1997; 126:480-484.
2. Whitmore WF. Lancet 1994;343:1263-1267.
3. Wilt TJ. Am J Med 1998;104:602-604.
4. Johansson JE, et al. JAMA 1997;277:467-471.
5. Cheng L, et al. J Natl Cancer Inst 1998;90:1105-1107.
6. Carter HB, et al. J Urol 1997;157:2206-2209.
7. Carter HB, et al. Urology 1997;49:379-384.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.