Screening: general concerns
No clear consensus exists across medical specialties with respect to the benefits of PSA screening in the general population. The American Urological Association and the American Cancer Society recommend annual PSA and DRE screening for men stardng at 50 years old (40 or 45 in high-risk patients); the American College of Preventive Medicine and the American College of Physicians-American Society of Internal Medicine, on the other hand, do not. Survey data across three US states of practices by generalists and internists suggest that only 67% of family physicians and 40% of internists routinely screen men over 50.

The presence of LUTS does not predict prostate cancer incidence, but it does increase the likelihood of PSA testing, as do a diagnosis of Benign Prostatic Hypertrophy (BPH) and/or prior history of prostate surgery. BPH patients are also, of course, more likely to be seen by a urologist; a 1997 Gallup survey found that 92% of urologists employed routine PSA testing among their BPH patients. The Agency for Health Care Policy and Research and the International Consultation on BPH both recommend PSA measurement in those BPH patients for whom an incidental diagnosis of prostate cancer would spur a change in management (generally referring to patients with at least a 10-year life expectancy).
Prostate cancer detected incidentally during work-up or surgery for BPH may not necessarily be clinically significant. In the series of Tombal and colleagues, 8% of T1a patients opting for surveillance progressed at a mean of 73 months, whereas 29% of T1b patients progressed at a mean of 17 months. Kearse and associates followed 304 TURP patients for a minimum of 8 years, concluding that the risk of progression and death among T1a prostate cancer patients was not significantly elevated over those confirmed to have only BPH on pathological examination. Epstein and coworkers found, conversely, that even low-volume T1a tumors may progress and cause mortality with extended follow-up. Fowler and colleagues also point out that, with the decline in surgical management for BPH, men with T1a/b prostate cancer are increasingly likely to be undiagnosed.

A fundamental caveat with respect to interpretation of studies of diagnostic tests’ predictive ability is raised periodically in reviews of the prostate cancer screening literature and should be reiterated here: most studies report sensitivity (i.e. the proportion of cancers that will be diagnosed) and specificity (i.e. the proportion of unnecessary biopsies that may be avoided) of a test in a given population, and frequently present receiver operating characteristic (ROC) curve areas as a means of comparing competing tests. The calculation of these measures, however, is highly dependent on the prevalence among the study population of the disease in question. Thus, if a test’s threshold value is calculated to yield high sensitivity among a referral population in whom prostate cancer is relatively common, in a general screening population the same test would retain high sensitivity but may have very low specificity.

For this reason, the positive and negative predictive values (PPV and NPV) of a test, reported less frequently, are more relevant to most clinical decision-making. For ROC comparisons among unfractionated PSA and other assays between prostate cancer and BPH patients, Jung and colleagues have suggested matching patients by unfractionated PSA, to compensate for unequal distribution of PSA values between the two groups, but such adjustments are not usually performed.

More information and facts about PSA and DRE screening and prostate cancer, you can found in the site –