Comparison between protocols and publications for prognostic and predictive cancer biomarker studies

Background:Method prespecification in study protocols is important for controlling bias in reports. The primary goal of this study was to assess potential for discordance between study protocols an...