Minimal Residual Disease (MRD) Monitoring in CBFB-MYH11 Acute Myeloid Leukemia (AML) Is of Prognostic Relevance for Relapse-Free Survival.

Detection of minimal residual disease (MRD) in acute myeloid leukemia (AML) associated with specific gene fusions is an important tool for the assessment of response to treatment and the individual risk of relapse. The real-time quantitative RT-PCR (RQ-PCR) method allows the quantification of fusion transcript levels at distinct time points during treatment. While in acute promyelocytic leukemia (APL) MRD monitoring has been clearly shown to be predictive for clinical outcome, the prognostic value of MRD in CBFB-MYH11 AML could not consistently been demonstrated yet. Small patient populations and the availability of bone marrow (BM)/peripheral blood (PB) samples at defined time points mainly hamper most studies. We evaluated the prognostic impact of MRD in a large cohort of CBFB-MYH11 AML by RQ-PCR. A total of 44 patients (16–60 years) were treated within one of the AMLSG treatment trials (AMLHD93 n=4, AMLHD98A n=27, AMLSG07-04 n=13). Patient samples (BM and/or PB) were collected at study entry (n=75), during treatment (n=199), and during follow up (n=140). Following high-dose cytarabine (HiDAC) consolidation therapy, patients received a second course of HiDAC (n=25); autologous stem cell transplantation (SCT) (n=13) or allogeneic SCT from a matched related family donor (n=6) depending on the treatment protocol. Median follow up was 22.5 months. Quantitative CBFB-MYH11 fusion transcript expression was measured by RQ-PCR using TaqMan technology. Primers and probes were chosen according to Europe Against Cancer (EAC) standard protocols. Sensitivities ranged from 10⁻³ to 10⁻⁴.Transcript levels at diagnosis ranged from 6208 to 312987 (median 34293.5). There was no prognostic impact of pretreatment transcript levels on relapse free survival (RFS). The ratio of transcript levels after 2 induction cycles and pretreatment levels ranged from 0 to 0.0049; again, this ratio had no impact on RFS. In contrast, during consolidation therapy 63% of the patients became RQ-PCR negative and RFS was significantly superior (RFS after 2 years 75%) compared to RQ-PCR positive patients (RFS after 2 years 32%) (p=0.03). After consolidation, seven of the RQ-PCR negative patients became positive at least in one BM-sample during follow up. Four patients developed transcript levels above 10 and all relapsed, whereas the three patients with transcript levels remaining below 10 are in continuous remission (p=0.0001). In our study, transcript levels during and after consolidation therapy are significantly associated with clinical outcome in CBFB-MYH11 AML. Risk-adapted therapy may be considered for those patients remaining positive during consolidation therapy. The identification of transcript levels above 10 after consolidation therapy might allow early treatment decisions.