Tumor markers are often heterogeneous substances that may be present in elevated concentrations in the serum of cancer patients. Typically measured by immunoassay, they contribute to clinical management, particularly in screening, case-finding, prognostic assessment, and post-treatment monitoring. Data both from external quality assessment (EQA) schemes and clinical studies demonstrate significant variation in tumor marker results obtained for the same specimen using different methods. Between-method between-laboratory coefficients of variation (CV) reported by EQA schemes generally reflect the complexity of the measurand, ranging from <5% for the structurally relatively simple α-fetoprotein (AFP) to >25% for the complex mucinous cancer antigen 19-9 (CA19-9). Improving the standardization of tumor marker measurements is particularly important for three reasons. The primary use of tumor markers is in monitoring cancer patients over long periods of time. Clinical interpretation of trends may consequently be affected if results are obtained in different laboratories using different methods or if a laboratory has to change method. Differences in results may have major implications for adoption of area-wide decision cut-offs and make implementation of these difficult. Method-related differences also make it difficult to compare clinical studies. Improving comparability of tumor marker results requires broad international agreement about which molecular forms of the measurand have clinical utility, identifying and adopting pure molecular forms as calibrants, and defining antibody specificities for their optimal detection. These aims have been achieved to varying extents for the most frequently measured serum tumor markers as described in this paper.
Keywords: AFP; CA antigens; CEA; International Standards; PSA; antibody mapping; between-method comparability; cancer biomarkers; hCG.