Graphical AbstractIs a two-fold approach — preliminary studies based on small samples followed by a large-sample study to check reproducibility — in the search for biomarkers really prudent?