Abstract: Objective To analyze intralaboratory precisions and interlaboratory variations of the measuring system for national tumor marker detection, and to investigate the quality levels among different measuring systems. Methods External quality assessment(EQA) data were collected from the National Center for Clinical Laboratory in the past 3 years. The measuring results were classified into different groups according to the measuring systems, after excluding the data of "mean ± 3s". The coefficients of variation (CV) were calculated respectively. Internal quality control data and corresponding CV from nationwide participanting laboratories in May 2012 were collected by Web-based EQA software system. Acceptable rates were calculated according to Clinical Laboratory Improvement Amendments(CLIA)′ 88 1/3 total allowable error (TEa), 1/4 TEa and the specifications based on biological variation including the minimal, appropriate and optimal imprecisions. Results After classification based on analysis systems, the average CVs of 4 inlet systems were <10%. However, from the groups of radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA) showed large CV. More than 70% of laboratories from most systems met 1/3 TEa requirement. More than 85% of laboratories from most systems met appropriate specification based on biological variation. Conclusions The inlet system 2 shows best interlaboratory precision, and inlet system 5 shows best intralaboratory precision. To ensure the reliability of the measuring results, manufacturers and academic institutions should pay more attention to the traceability of tumor markers, meanwhile clinical laboratories should strengthen their quality controls.

Key words: Tumor marker, External quality assessment, Comparability, Immunoassay