Project 1: ddPCR Control round

Results from a EUTOS Ring trial testing duplex digital PCR assays for monitoring BCR-ABL1 using ABL1, BCR and GUSB as reference genes

Franke G.-N.1, Pallisgard N.2, Maier J.1, Machova Polakova K.3, Lange T.4,5, Salmon M.6, Cross N.6, Gottardi E.7, Saglio G.8, Soverini S.9, Hochhaus A.10, Niederwieser D.1, Ernst T.10

Institute:

1Universitätsklinikum Leipzig, Abt. Hämatologie und internistische Onkologie, Leipzig, Germany, 2Zealand University Hospital, Roskilde, Denmark, 3Institute of Hematology and Blood Transfusion, Department of Molecular Genetics, Prague, Czech Republic, 4Universität Leipzig, Medizinische Fakultät, Leipzig, Germany, 5Asklepios Klinikum Weißenfels, Klinik für Hämatologie, internistische Onkologie und Gastroenterologie, Weißenfels, Germany, 6Wessex Regional Genetics Laboratory, Salisbury, United Kingdom, 7Torino University, Department of Clinical and Biological Sciences, Torino, Italy, 8Torino University, Division of Internal Medicine and Hematology, Torino, Italy, 9Bologna University, Department of Experimental, Diagnostic and Specialty Medicine, Bologna, Italy, 10Universitätsklinikum Jena, Klinik für Innere Medizin II, Jena, Germany

Reverse transcriptase quantitative polymerase chain reaction (RT-qPRC) for BCR-ABL1 mRNA is routinely used for molecular monitoring in chronic myeloid leukemia (CML). The results are aligned to International Scale (IS) applying a laboratory-specific conversion factor (CF). However, RT-qPCR is a relative measurement and relies on standard curves and is prone to errors from batch variation and varying PCR efficiency. In contrast, droplet digital PCR (ddPCR) generates an absolute quantification of BCR-ABL1 mRNA. This could potentially reduce the requirements for standardization and produce more robust and precise results.

Methods: The European Treatment and Outcome Study (EUTOS) has established a network of CML reference laboratories across Europe. Samples and multiplex digital PCR assays for BCR-ABL1 and ABL1 (provided by Bio-Rad laboratories), BCR and GUSB (non-commercial assays) were sent out to seven EUTOS laboratories. We aimed to determine inter-laboratory variation and validate the ability of the method and assays to detect deep molecular responses (MR4.5) by distributing cDNA reagents (Bio-Rad), centrally prepared primer/probes, cDNA, cell lysates and a lyophilized secondary reference (LYO) panel (Cross et al. Leukemia. 2016;30(9):1844-52).

Results: The initial analysis showed that cDNA samples have a very high concordance between labs for both copy numbers (CN) and the BCR-ABL1/reference gene (RG) ratio (%Coefficient of variation (%CV) for CN 2.5 – 10.5 and 2.3 to 11.9, respectively), with the highest CV in the samples with low BCR-ABL1 CN. As expected, variation was higher for cell lysates, which also included RNA extraction and cDNA synthesis. For the LYO panel, %CV for ABL1 and BCR as RG, ranged from 11.3 for MR1 to 78.9 for MR4.5 samples. The ratio %BCR-ABL1/RG was 15.3 and 17.1 at MR1, 1.37 and 1.24 at MR2, 0.012 and 0.0095 at MR4 and 0.008 and 0.0046 at MR4.5 for ABL1 and BCR, respectively. At present GUSB results are incomplete for technical reasons. Conversion factors to IS for each lab and assay have not yet been determined. More detailed results will be shown at presentation.

Conclusions: Overall these results indicate that droplet digital PCR has a low inter-laboratory variation, improving the precision of BCR-ABL1 monitoring and the ability to detect deep molecular responses.