Abstract
Recent research has raised hopes for impressively accurate screening for cancer with molecular biomarkers. These molecular markers will probably be more sensitive and specific than older screening modalities, as well as easier to use. In this Essay, I argue that these sensitive screening tests might be clinically valuable - but that they will present unique issues in implementation and interpretation. These issues are likely to affect the way clinicians conduct screening and the way that they make diagnoses in individuals who screen positive for cancer.