Cancer screening markers are defined as tests. When systematically applied to populations, they identify asymptomatic individuals at sufficient risk of cancer to justify use of additional markers or clinical intervention. The goal of applying screening markers is to detect malignant disease or its premalignant precursors at a stage early enough that existing clinical interventions are effective. Since screening assays would be typically applied to subjects without clinical symptoms, the ideal screening marker must be noninvasive, inexpensive and robust enough to permit population-based application. Although maintaining high sensitivity, that is, a high true-positive rate indicating the proportion of subjects with disease that are marker positive, a key performance characteristic of screening markers is very high disease specificity, that is, a low FPR indicating the proportion of subjects without disease who are marker positive, since, in this setting, even a small FPR, typically less than 1%, will still result in a relatively large number of disease-free subjects who will be unnecessarily subjected to additional and/or invasive diagnostic procedures. For nonbinary markers, that is, those quantitated as a continuous measure, a minimally acceptable FPR implies the setting of a threshold for the marker usually ascertained using a receiver operating characteristic (ROC) analysis.1 Finally, the ultimate validation of a screening marker is whether its use results in a reduction of the burden of cancer in the screened population. It should also be emphasized that even if a marker is effective in detecting disease early, there are reasons why a screening program utilizing the marker might not provide an overall benefit to the screened population. These include (1) lack of effective treatment for the marker-detected cases; (2) poor compliance with the screening modality; (3) excessive economic or health-related costs of the screening test or associated subsequent diagnostic tests with false-positive results; and (4) overdiagnosis, such that the screening test detects otherwise undiagnosed malignancies that do not threaten the health or life of the individual. This evaluation is best accomplished in a standard parallel-arm randomized clinical trial wherein one arm is composed of subjects undergoing the screening and the other unscreened individuals. Such trials are massive undertakings, for example, to detect a 20% reduction in a cause-specific mortality with 80% power, typically several hundred overall deaths would need to be observed.2 The prevalence and mortality of individual organ site malignancies will, however, determine the relative size of the trial required, that is, fewer patients are required to demonstrate efficacy in lung cancer versus ovarian cancer.
In contrast to screening markers, cancer early detection markers indicate the presence of an early cancer or that cancer will occur with nearly 100% certainty within a very short time interval. Historically, a number of tests have been investigated for cancer screening and early detection. These include screening mammography for breast cancer; the Papanicolaou cell cytology (PAP) test for cervical cancer; fecal occult blood (FOB), flexible sigmoidoscopy, and serum carcinoembryonic antigen (CEA) for colorectal cancer; chest x-ray and sputum cytology for lung cancer; ultrasound and serum CA125 for ovarian cancer; and digital rectal examination (DRE) and serum prostate-specific antigen (PSA) for prostate cancer. Of these, only breast mammography, the cervical PAP test, and colorectal FOB testing have been demonstrated to reduce site-specific cancer mortality when applied as a screening measure in the general population and, in the case of screening mammography, the results are still somewhat controversial.3, 3a
Given the proven efficacy of the above screening approaches, immunodiagnostic methods for the quantitation of only three specific tumor markers in serum are widely used in routine screening and clinical diagnostic applications. These include CA125 in ovarian cancers, CEA in gastrointestinal malignancies, and PSA in urological cancers. In the organ site-specific tumor marker sections that follow, the biological and technical bases of these immunodiagnostic tests are summarized, the historical development and detailed performance of the tests are discussed, and the strengths and current limitations of these methods for screening, early detection, and patient management are reviewed.
