Tumor marker studies questioned
Ioannina – Tumor markers are specific to a certain type of cancer and its developmental stage. They are used to estimate a prognosis for the development and treatment of the cancer.
Researchers led by John Ioannidis from the University of Ioannina School of Medicine undertook an empirical approach to reports from studies of these cancer prognostic factors to revise their informational value (J. natl. cancer inst. 99,(3): 236-43). From 20 meta-analyses of cancer prognostic markers, 331 studies were identified. Applying eight quality measures to the study design and assay methods (i.e. blinding and power calculation), they found that estimates of prognostic effects of tumor markers varied considerably and came to the conclusion that the literature was largely unreliable. According to their assessment, improved design and reporting was deemed necessary, given the clinical importance of prognostic markers.