NGS better discriminates true MRD positivity for the risk stratification of childhood ALL treated on an MRD-based protocol

We compared minimal/measurable residual disease (MRD) levels evaluated by routinely used real-time quantitative polymerase chain reaction (qPCR) patient-specific assays and by next-generation sequencing (NGS) approach in 780 immunoglobulin (IG) and T-cell receptor (TR) markers in 432 children with B-cell precursor acute lymphoblastic leukemia treated on the AIEOP-BFM ALL 2009 protocol. Our aim was to compare the MRD-based risk stratification at the end of induction. The results were concordant in 639 of 780 (81.9%) of these markers; 37 of 780 (4.7%) markers were detected only by NGS. In 104 of 780 (13.3%) markers positive only by qPCR, a large fraction (23/104; 22.1%) was detected also by NGS, however, owing to the presence of identical IG/TR rearrangements in unrelated samples, we classified those as nonspecific/false-positive. Risk group stratification based on the MRD results by qPCR and NGS at the end of induction was concordant in 76% of the patients; 19% of the patients would be assigned to a lower risk group by NGS, largely owing to the elimination of false-positive qPCR results, and 5% of patients would be assigned to a higher risk group by NGS. NGS MRD is highly concordant with qPCR while providing more specific results and can be an alternative in the front line of MRD evaluation in forthcoming MRD-based protocols.