Immunoassay of CEA, CA 19-9, CA 125, and CA 15-3 on the automated systems ES 300 and ES 600: methodological evaluation from a multicentre collaborative study.

The analytical performance of the automated Enzymun Test System ES 300 and ES 600 (developed by Boehringer Mannheim) for the assay of the tumour markers CEA, CA 19-9, CA 125, and CA 15-3, was assessed from data collected in a multicentre collaborative study in which eleven laboratories were involved. Results of the 1990 cycle of the external quality assessment (EQA) scheme for tumour markers, supported by the Italian National Research Council (CNR), were also used in this evaluation. The within-assay and between-assay precision was found to be 2.0 and 4.3 CV% for CEA, 2.9 and 6.8 CV% for CA 19-9, 3.6 and 9.4 CV% for CA 125, 2.9 and 6.0 CV% for CA 15-3. The between-lab variability of the four tumour markers on ES 300 and ES 600 systems was 9.4, 10.6, 11.9, 9.2 CV% for CEA, CA 19-9, CA 125 and CA 15-3 respectively. These values were comparable to or better than those obtained with the most precise manual kits used by laboratories participating in the 1990 EQA cycle. The agreement between the results from the Enzymun Test and those obtained using other method/kits was evaluated by assaying control samples previously circulated either in the CNR EQA or in the German EQA. The regression analysis indicates that for CEA, CA 125 and CA 15-3 assays the results produced by ES 300 and ES 600 are in good agreement with the consensus means of the EQAs; CA 19-9 results exhibit a worse correlation and are generally lower than the consensus mean. The linearity of the assays for the four tumour markers was checked by dilution tests performed by participants in the collaborative study; in all cases the dilution of the sample did not affect the values obtained.