A solution for distinguishing Le(a-b-) sera in CA19-9 assays using SphereLight 180 and Architect i2000 assays

Clin Chim Acta. 2012 Jan 18;413(1-2):278-81. doi: 10.1016/j.cca.2011.09.009. Epub 2011 Sep 14.

Abstract

Background: Measurement of carbohydrate antigen (CA) 19-9 is not applicable in patients with Lewis (Le) blood type, Le(a-b-). It is important to distinguish cases with Le(a-b-) before CA19-9 measurement. Therefore, we prepared a cut-off solution that gives a clear index to distinguish Le(a-b-) sera.

Method: The frequencies of Le blood types and the distribution of the CA19-9 values in each Le blood type were examined in 188 healthy subjects. The CA19-9 values for all Le(a-b-) sera and for a portion of Le(a-b+) sera exist below the limit of quantitation as measured by the SphereLight 180 kit. The cut-off solution, which gives a clear cut-off index (COI), was prepared to differentiate Le(a-b-) from Le(a-b+), and was evaluated using the SphereLight 180, Architect i2000, UniCel DxI 800, Elecsys 2010, and Lumipuls ƒ kits.

Results: The COI was calculated as the mean +3 SD of the CA19-9 values of a cut-off solution that is adjusted to the limit of detection. Both the sensitivities and specificities of the COIs were 100% using the SphereLight 180 kit and 100% and 91.7%, respectively, using the Architect i2000 kit, but these values were not satisfactory using the other CA19-9 assay kits.

Conclusion: The COIs, determined by the cut-off solution, correctly identified all Le(a-b-) sera as Le(a-b-) and differentiated Le(a-b-) sera from other types of sera in CA19-9 assays using only the SphereLight 180 and Architect i2000 kits.

MeSH terms

  • CA-19-9 Antigen / blood*
  • Humans
  • Lewis Blood Group Antigens*
  • Limit of Detection

Substances

  • CA-19-9 Antigen
  • Lewis Blood Group Antigens