Transferring a Quantitative Molecular Diagnostic Test to Multiple Real-Time Quantitative PCR Platforms

J Mol Diagn. 2018 Jul;20(4):398-414. doi: 10.1016/j.jmoldx.2018.02.004. Epub 2018 Apr 3.


Quantitative gene expression assays are increasingly used for diagnosis and research, but are often restricted to specific instrumentation. We propose a robust technical and statistical framework that enables transferring of established real-time quantitative PCR assays across real-time quantitative PCR platforms without compromising analytical and clinical validity. The feasibility of our approach was tested on MammaTyper, an in vitro diagnostic assay that quantifies breast cancer biomarkers and dichotomizes results according to cutoff points. CFX96, Applied Biosystems 7500 Fast, and Mx3000P were chosen as the candidate platforms, whereas the LightCycler 480 II was used as a reference. Two instruments were used per platform, and they were tested initially for equivalence via Bland-Altman and Deming regression analyses. A method comparison approach was adapted to adjust cutoffs for the new systems and the cross-platform agreement was evaluated. Finally, precision was estimated for each platform. The performance on the candidate devices was highly comparable to the reference platform, with a 7 log quantification range and amplification efficiencies of 97% to 103%. The equivalence tests successfully prequalified instruments, preventing constant and proportional errors and enabling reliable adjustments of cutoffs, which resulted in cross-platform marker and subtype agreements of 91% to 100% and κ values between 0.78 and 1.00. Provided that platform-specific adjustments are implemented, the described process can help expand the operability of quantitative diagnostic tests while maintaining assay performance characteristics.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Biomarkers, Tumor / metabolism
  • Humans
  • Limit of Detection
  • Molecular Diagnostic Techniques / methods*
  • Real-Time Polymerase Chain Reaction / methods*
  • Reference Standards
  • Regression Analysis
  • Transcription, Genetic


  • Biomarkers, Tumor