In checkup center, tumor markers are measured in subjects most of whom are free of cancer. The accuracy and stability of assay kits of tumor markers can be estimated by studying monthly changes of the averages and the positive rates of tumor markers. We have analyzed tumor marker data from clients who visited our center from Sep. 2000 to Aug. 2002 (
n=8, 316) . PSA was measured in male clients over 50 years old, and CA125 was assayed in female clients. Clients who had apparent cancer were excluded in this study.
The average values (positive rates) of CEA, AFP, CYFRA21-1, PSA, CA19-9 and CA125 were 2.1ng/m
l (4.5%), 4.2 ng/m
l (2.5 %), 0.9ng/m
l (6.5%), 1.2 ng/m
l (3.8%), 5.8 U/m
l (0.76%) and 11.0 U/m
l (2.1%), respectively. All tumor markers showed monthly change in the average. Alterations of positive rates were observed in CEA, CYFRA21-1 and CA19-9. The average and the positive rate of CEA were slightly high in July 2001, which was apparently due to malfunction of assay apparatus. The average and the positive rate of CYFRA21-1 showed large consecutive changes with two peaks and one nadir: the average values (positive rates) were 1.0 ng/m
l (9.1%), 0.5 ng/m
l (0.56%) and 1.2 ng/m
l (11%) in Oct. 2000, May 2001 and Feb. 2002, respectively. These alterations were not accompanied by the change of the internal standard. The averages and the positive rates of CA19-9 were decreased by half after Mar. 2002. This decrease was due to poor quality of the particular lots of assay kit. We consider CYFRA21-1 may not be suitable for the use in checkup center. Its average value is close to cutoff value, and small fluctuation of the assay will affect the judgment. Other tumor markers did not show such large alterations as CYFRA21-1.
View full abstract