We performed serial measurable residual disease (MRD) monitoring in bone marrow (BM) and peripheral blood (PB) samples of 155 intensively treated patients with RUNX1-RUNX1T1+ AML, using a qRT-PC-based assay with a sensitivity of up to 10-6. We assessed both reduction of RUNX1-RUNX1T1 transcript levels (TLs) and achievement of MRD negativity (MRD-) for impact on prognosis. Achievement of MR2.5 (>2.5 log reduction) after treatment cycle 1 and achievement of MR3.0 after treatment cycle 2 were significantly associated with a reduced risk of relapse (P = .034 and P = .028, respectively). After completion of therapy, achievement of MRD- in both BM and PB was an independent, favorable prognostic factor in cumulative incidence of relapse (4-year cumulative incidence relapse: BM, 17% vs 36%, P = .021; PB, 23% vs 55%, P = .001) and overall survival (4-year overall survival rate BM, 93% vs 70%, P = .007; PB, 87% vs 47%, P < .0001). Finally, during follow-up, serial qRT-PCR analyses allowed prediction of relapse in 77% of patients exceeding a cutoff value of 150 RUNX1-RUNX1T1 TLs in BM, and in 84% of patients exceeding a value of 50 RUNX1-RUNX1T1 TLs in PB. The KIT mutation was a significant factor predicting a lower CR rate and inferior outcome, but its prognostic impact was outweighed by RUNX1-RUNX1T1 TLs during treatment. Virtually all relapses occurred within 1 year after the end of treatment, with a very short latency from molecular to morphologic relapse, necessitating MRD assessment at short intervals during this time period. Based on our data, we propose a refined practical guideline for MRD assessment in RUNX1-RUNX1T1+ AML.
