Assessment of acute myeloid leukemia molecular measurable residual disease testing in an interlaboratory study

Abstract

The European LeukaemiaNet (ELN) measurable residual disease (MRD) working group have published consensus guidelines to standardise molecular genetic MRD testing of the t(8;21)(q22;q22.1) RUNX1::RUNX1T1, inv(16)(p13.1q22) CBFB::MYH11, t(15;17)(q24.1;q21.2) PML::RARA and NPM1 type A markers. A study featuring 29 international laboratories was performed to assess interlaboratory variation of testing, and the subsequent interpretation of results, both crucial to patient safety. The majority of participants in this study were able to detect, accurately quantify and interpret MRD testing results correctly, with a level of proficiency expected from a clinical trial or standard of care setting. However, a number of testing and interpretive errors were identified that in a patient setting would have led to misclassification of patient outcomes and inappropriate treatment pathways being followed. Of note, a high proportion of participants reported false positive results in the NPM1 marker negative sample. False positive results may have consequences clinically, committing patients to unneeded additional chemotherapy and/or transplant with the attendant risk of morbidity and mortality, and therefore highlights the need for ongoing external quality assessment (EQA)/proficiency testing (PT) in this area. Most errors identified in the study were related to the interpretation of results. It was noted that the ELN guidance lacks clarity for certain clinical scenarios and highlights the requirement for urgent revision of the guidelines to elucidate these issues, and related educational efforts around the revisions to ensure effective dissemination.

Measurable Residual Disease Detection in Acute Myeloid Leukemia: Current Challenges and Future Directions

“…However, the use of different transcripts makes comparisons between different assays difficult. Furthermore, the same qPCR tests run in different laboratories, e.g., for testing mutated NPM1 , may lead to discrepant results [ 23 ]. Although the sensitivity of MRD detection using PCR was high, Scott et al were able to demonstrate an issue with NPM1 analysis in an interlaboratory study.…”

“…During the process of reverse transcription before analysis, errors in NPM1 exon 12 can be introduced artificially. As NPM1 is regarded as a leukemia-specific mutation, falsely positive MRD results could lead to unnecessary treatment [ 23 ]. Therefore, current efforts undertaken by groups, such as the ELN–DAVID MRD Working Group, focus on standardization and quality control of qPCR-based MRD assays to generate comparable MRD results [ 24 ].…”

Recommendations from the AML molecular MRD expert advisory board

“…Further to this, external quality assessment (EQA) programs have been established to assess the quality of testing [ 4 ], For molecular MRD testing, EQA data has shown that interlaboratory variation for molecular AML MRD testing was substantially greater than that seen for BCR :: ABL1 testing in chronic myeloid leukemia (CML), despite both tests being reverse transcription quantitative PCR-based (Fig. 1 ) [ 5 , 6 ]. However, BCR :: ABL1 MRD testing has been subject to many years of standardization, including the development of the BCR :: ABL1 international scale (IS) [ 7 ].…”

“…

Fig. 1 Comparison of interlaboratory variation for standardized BCR :: ABL1 IS RT-dPCR (gray triangle markers) and RT-qPCR (black triangle markers) [ 5 ] testing with unstandardized RUNX1 :: RUNX1T1 (blue circle markers), CBFB :: MYH11 (red circle markers), PML :: RARA (green circle markers) and NPM1 (orange circle markers) RT-qPCR testing [ 6 ].

…”