An estimated 12.66 million new cancer cases and 7.56 million cancer-related deaths occurred in 2008, and this figure is increasing. Recent years have seen much progress in the development of various diagnostic and prognostic techniques, and of particular interest are tumor markers. These can be used as a measurable tool, within specific laboratory tests, to diagnose cancer and predict a patient’s response to treatment. Since these can inform clinical management of the patient, it is imperative that these tumor marker tests are appropriately validated and reported.
In this era of personalized medicine, where tumor marker ‘signatures’ can potentially predict which patients will respond to certain therapies, erroneous and inconsistent reporting of potential tumor biomarkers can have far reaching consequences. Therefore, in 2005, the Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) checklist for researchers was produced by a renowned team of biostatisticians from the National Cancer Institute and the European Organisation for Research and Treatment of Cancer. The guidelines have been cited several hundred times, highlighting the usefulness of the checklist to clinicians and researchers.
Now, seven years later, in a multi-year collaboration among the same authors of the original guidelines, Doug Altman and colleagues have published an elaborated and eagerly anticipated version of the REMARK checklist in BMC Medicine. In this updated version, the original 20 checklist items are explained and elaborated one by one. Aimed at clinical researchers and clinical test developers, they provide a further discussion of the importance of stringent reporting by supplying specific recommendations for each checklist item along with real life examples of good practice and reporting.
Importantly, this also extends to the statistical development and analysis when translating the basic research findings to the creation of clinically informative tumor marker tests. In a related commentary for BMC Medicine, Lisa McShane, one of the co-authors of the elaborated REMARK guidelines, discusses these challenges, providing an overview of common statistical errors and underlining the importance of rigorous statistical evaluation in tumor marker test development.
The publication of the original REMARK guidelines was motivated by understanding that variations in designs and analyses can lead to different study results and interpretations. The elaborated guidelines aim to further augment their use and effectiveness through better understanding of the objectives behind each checklist point. It is anticipated that these elaborated guidelines will replace the original REMARK guidelines in becoming the standard against which studies of this nature are reported. Moreover, it is hoped that it will positively influence both the conduct and reporting of research in molecular diagnostics. This should, therefore, result in more accurate tumor biomarker tests so that cancer patients can be treated with more appropriate, reliable and ‘personalized’ therapies.