Droplet digital PCR outperforms conventional methods for early relapse prediction in AML/MDS patients post-transplant

Abstract Measurable residual disease (MRD) and donor chimerism (DC) monitoring are critical for early relapse prediction after allogeneic hematopoietic stem cell transplantation (allo-HSCT). Droplet digital PCR (ddPCR), a highly sensitive quantitative technique, offers advantages in mutation detection and MRD tracking. To explore the advantages of ddPCR in predicting relapse post-transplant, ddPCR for MRD monitoring was compared with conventional methods including WT1-qPCR, multiparameter flow cytometry (MFC), short tandem repeat (STR) and XY-FISH chimerism for 196 patients with myeloid malignancies. ddPCR demonstrated superior sensitivity (81.08%) and specificity (91.19%) compared to WT1-qPCR (75.86%/78.23%) and MFC (56.76%/94.97%). Total-DC (16.13%/94.70%), T-DC (9.68%/93.94%) and XY-FISH (46.67%/79.73%) were significantly less sensitive. In all, from robust to inconclusive, the efficacy in predicting relapse (Youden index) were as follows: 0.72 for ddPCR, 0.54 for WT1-qPCR, 0.51 for MFC, 0.26 for XY-FISH, 0.11 for total-DC and 0.04 for T-DC. Notably, ddPCR provided the earliest relapse prediction, with a median lead time of 60.5 (range: 10–461) days, outperforming MFC (43 days), WT1-qPCR (55 days), STR chimerism (total-DC: 43 days; T-DC: 37 days), and XY-FISH (54 days). This study demonstrated ddPCR as the most effective and timely tool for impending relapse post-transplant in AML/MDS.