Reporting in studies of protein biomarkers of prognosis in colorectal cancer in relation to the REMARK guidelines

Purpose: The REMARK guidelines give authors comprehensive and specific advice on the complete and transparent reporting of studies of prognostic tumor markers. The aim of this study was to use the REMARK guidelines to evaluate the quality of reporting in a sample of studies assessing tissue-based protein markers for survival after resection of colorectal cancer. Experimental design: Eighty pertinent articles were scored according to their conformity to 26 items derived from the REMARK criteria.

Results: Overall, on a scale of adequacy of reporting that potentially ranged from 26 to 78, the median for these studies was 60 (interquartile range 54-64) and several criteria were adequately covered in a large proportion of studies. However, others were either not dealt with or inadequately covered, including description of the study design (35%), definition of survival endpoints (48%), adjuvant therapy (54%), follow-up procedures and time (59%), neoadjuvant therapy (63%), inclusion/exclusion criteria (73%), multivariable modeling methods and results (74%), and discussion of study limitations (85%). 

Conclusions and clinical relevance: Inadequacies in presentation militate against comparability among protein marker studies and undermine the generalizability of their findings. The quality of reporting could be improved if journal editors were to require authors to ensure that their work satisfied the REMARK criteria.