Technological progress within the last 15-20 years has significantly increased our knowledge about the molecular basis of cancer development, tumor progression, and treatment response. As a consequence, a vast number of biomarkers have been proposed, but only a small fraction of them have found their way into clinical use. The aim of this paper is to describe the specific demands a clinically relevant biomarker should meet and how biomarkers can be tested stepwise. We name this procedure the "triple-R principle": robustness, reproducibility, and relevance. The usefulness of this principle is illustrated with the marker TP53. Since it is mutated in a broad spectrum of cancer entities, TP53 can be considered a very promising marker. Thus, TP53 has been studied in detail but there is still no explicit consensus about its clinical value. By considering our own experience and reviewing the literature, we demonstrate that a major problem of current biomarker research is disregard of whether the biomarker is prognostic or predictive. As an example, it is demonstrated that TP53 is not a prognostic marker, but rather a purely predictive marker, and that disregard of this fact has made this otherwise strong biomarker appear as not being clinically useful so far.
Keywords: cancer; p53; predictive biomarker; prognostic biomarker.
© 2018 New York Academy of Sciences.