Desirable performance characteristics for BCR-ABL measurement on an international reporting scale to allow consistent interpretation of individual patient response and comparison of response rates between clinical trials

Abstract

AbstractAn international basis for comparison of BCR-ABL mRNA levels is required for the common interpretation of data derived from individual laboratories. This will aid clinical decisions for individual patients with chronic myeloid leukemia (CML) and assist interpretation of results from clinical studies. We aligned BCR-ABL values generated by 38 laboratories to an international scale (IS) where a major molecular response (MMR) is 0.1% or less. Alignment was achieved by application of laboratory-specific conversion factors calculated by comparisons performed with patient samples against a reference method. A validation procedure was completed for 19 methods. We determined performance characteristics (bias and precision) for consistent interpretation of MMR after IS conversion. When methods achieved an average BCR-ABL difference of plus or minus 1.2-fold from the reference method and 95% limits of agreement within plus or minus 5-fold, the MMR concordance was 91%. These criteria were met by 58% of methods. When not met, the MMR concordance was 74% or less. However, irrespective of precision, when the bias was plus or minus 1.2-fold as achieved by 89% of methods, there was good agreement between the overall MMR rates. This indicates that the IS can deliver accurate comparison of molecular response rates between clinical trials when measured by different laboratories.