A proof-of-concept practical approach for achieving equivalent results from non-harmonized measurement methods while...

Area of Science:

• Clinical chemistry
• Laboratory medicine
• Metrology in diagnostics

Background:

• Inconsistent laboratory test results hinder clinical decisions due to a lack of measurement harmonization.
• Higher-order references are often unavailable, leading to comparability issues across different diagnostic assays.

Purpose of the Study:

• To introduce and validate a method for improving the comparability of laboratory test results from different measurement systems.
• To provide a solution for consistent test result interpretation until full metrological harmonization is achieved.

Main Methods:

• Developed a linear transformation formula using 2.5th and 97.5th percentiles from source and target methods.
• Tested the formula with carbohydrate antigen 125 (CA 125) data from Roche and Abbott assays.
• Evaluated effectiveness using difference plots and Passing-Bablok regression, and developed a web application for the process.

Main Results:

• Initial median relative difference between CA 125 assays was 37.7%, with a slope of 1.450.
• After transformation, the median relative difference reduced to 6.0%, within acceptable bias goals.
• Regression slope and intercept improved significantly, indicating enhanced result comparability.

Conclusions:

• The proposed linear transformation method effectively enhances the comparability of results between different diagnostic assays.
• This approach enables more consistent interpretation of patient test results during follow-up care.