Invited Talk 9: Importance of Commutable Human Specimens and Test Value Distributions for Assay Calibration

George G. Klee, M.D., Ph.D.
Dept. of Laboratory Medicine
Mayo Clinic
Hilton 730, 200 First St. SW
Rochester MN 55905
Phone: 507-284-8213
FAX: 507-284-1927
e-mail: klee.george@mayo.edu

Certified Reference Materials are important for standardization and traceability of clinical assays; however, the use of these reference materials does not assure harmonization of test values. Two major issues causing this discordance are the presence of cross-reacting substances and matrix differences between the reference material and human specimens. Standard reference methods that perform equally well on reference materials and human specimens can help resolve these differences. These reference methods can be used to assign values to panels of commutable human specimens that subsequently can be used to calibrate clinical assays assuming that the reference methods and clinical assays have good correlation. Some clinical assays may not give results proportional to reference methods with all specimens in these panels because of differences in assay specificity. Such assays could not be "calibrated" globally, but may be standardized for limited applications.

Distributions of test values from well-characterized groups of patients also may serve as calibration reference standards for some assays. The biologic variation of these patient groups often can be minimized by standardizing preanalytic specimen handling and normalizing the test results for known demographic variables. With these provisions, the median or central indices of the distribution of test values in these patient groups is fairly constant across groups of patients. For example the median calcium concentration in ambulatory patients varies less than + 0.1 mg/dL over time. Therefore, median values for population test distributions established by reference methods, could be used to statistically "calibrate' assays without using traceable reference materials. Similarly commutable control material could be used to "calibrate" clinical assays if reference methods standardized with Certified Reference Materials are used to assign specific traceable values to this material.

Return to the Invited Talks page