Next-generation digital polymerase chain reaction assay for BCR::ABL1 in chronic myeloid leukemia: comprehensive validation for standardized measurable residual disease detection

Abstract

The sensitivity of real-time quantitative reverse-transcription polymerase chain reaction (RT-qPCR) for detecting measurable residual disease (MRD) in patients with chronic myeloid leukemia (CML) who have achieved a deep molecular response is limited. We validated a digital PCR (dPCR) assay (LQ Check BCR::ABL1 Major %IS Kit) for MRD quantification. Validation was performed according to CLSI guidelines EP17-A2, EP06, EP05-A3, and EP07 using BCR::ABL1-positive and -negative materials and the World Health Organization (WHO) 1st International Standard. A nonparametric analysis of the limit of blank was used to establish a positivity cutoff of ≥ 2 droplets. The limit of detection (LoD) was verified using a detection criterion of ≥ 95%, and the limit of quantitation (LoQ) was defined as the lowest level meeting ≥ 95% detection and a study-defined %CV of ≤ 76%. Clinical validation included 72 CML specimens for molecular response (MR) category comparisons and 67 specimens for paired quantitative comparisons. The LoD and LoQ were both 0.0025% on the International Scale (IS) with a detection rate of 95.8%. The assay showed linearity over 5 logs (R2 = 0.9916–0.9972), within-run %CV ≤ 1.52%, cross-condition %CV ≤ 2.76%, and WHO-traceable accuracy. In 67 paired specimens, the correlation coefficient was r = 0.9740, with a Passing–Bablok slope of 1.0189, intercept of −0.2552, and Bland–Altman mean bias of 0.2266. Among the 72 specimens, dPCR classified 20 as having a deeper MR than RT-qPCR. Reproducibility was characterized by r = 0.99 and a mean bias of 0.0967. Whole blood was preferably processed within 48 h, with processing at 72 h considered acceptable only when BCR QC criteria were met. The LQ Check BCR::ABL1 Major %IS Kit enabled sensitive, precise, and reproducible BCR::ABL1 quantification at MR4.6. The clinical significance of a deeper MR classification using dPCR, particularly for treatment-free remission, requires longitudinal validation.