Empirical evaluation demonstrated importance of validating biomarkers for early detection of cancer in screening settings to limit the number of false-positive findings

Abstract

ObjectivesSearch for biomarkers for early detection of cancer is a very active area of research, but most studies are done in clinical rather than screening settings. We aimed to empirically evaluate the role of study setting for early detection marker identification and validation. Study Design and SettingA panel of 92 candidate cancer protein markers was measured in 35 clinically identified colorectal cancer patients and 35 colorectal cancer patients identified at screening colonoscopy. For each case group, we selected 38 controls without colorectal neoplasms at screening colonoscopy. Single-, two- and three-marker combinations discriminating cases and controls were identified in each setting and subsequently validated in the alternative setting. ResultsIn all scenarios, a higher number of predictive biomarkers were initially detected in the clinical setting, but a substantially lower proportion of identified biomarkers could subsequently be confirmed in the screening setting. Confirmation rates were 50.0%, 84.5%, and 74.2% for one-, two-, and three-marker algorithms identified in the screening setting and were 42.9%, 18.6%, and 25.7% for algorithms identified in the clinical setting. ConclusionValidation of early detection markers of cancer in a true screening setting is important to limit the number of false-positive findings.