Objectives: The carbohydrate antigen 19-9 assay is a widely used biomarker as a complement to other tests in the detection and follow-up of pancreatic, bile duct and colon cancer in high-risk individuals. The CA 19-9 assays often give different results with different methods. This study intended to characterize the harmonization problem in CA 19-9 assays using results obtained through external quality assessment.

Material and Methods: According to the 2018-External Quality Assessment program by the Randox International Quality Assessment Scheme, pooled serum specimens involving three levels of CA19-9 were analyzed. In addition, the results of four assays which are Abbott Architect, Beckman Access DXI600/800, Roche Cobas 6000/8000, and Siemens Advia Centaur XP/XPT/Classic were contrasted.

Results: Data from External Quality Assessment schemes demonstrate significant variation in CA 19-9 assay results obtained for the same specimen using different assays. The mean CA 19-9 evaluation of the peer groups differed for all four systems, and the interlaboratory CVs also varied. Generally, the highest peer groups mean values were gathered by employing the Abbott Architect system, which was followed by Siemens Centaur XP/XPT/Classic, while the lowest means were acquired utilizing the Roche Cobas system. Serum CA 19-9 assays also show a wide range in CV, which vary from 4.5-10.1%.

Conclusion: According to our findings, the harmonization of the CA 19-9 results obtained from four immunoassays have not been achieved yet. Systematic differences(different antibodies etc.) are still present among the most popular commercial methods for CA 19-9 measurement and may produce some difficulties in interpreting laboratory test results. Consequently, it is suggested that the assay or method used to identify CA 19-9 concentrations should be included in the laboratory result report.