Molecular analyses of leukemia-specific markers has led to an improvement of the prognosis evaluation in patients (pts) with acute myeloid leukemia (AML). The European Leukemia Net (ELN) has published a classification which separates different subgroups by cytogenetic and molecular genetic analyses. Nevertheless, there are still pts suffering from disease recurrence within the ELN favorable risk group. To identify these pts at high risk for relapse the monitoring of minimal residual disease (MRD) of leukemia-specific markers could become an important diagnostic tool. In this study the potential of MRD monitoring by quantitative real-time PCR (RT-PCR) of NPM1 A mutation (NPM1 A) at different checkpoints within the ELN favorable risk group of pts with NPM1 A and without FLT3-ITD was investigated.
Pts participating in the AMLCG99, AMLCG2004, and AMLCG2008 trial were prospectively or retrospectively screened for NPM1 mutation and FLT3-ITD by melting curve analyses. 334 pts were screened positive for NPM1 mutation and 262 pts showed a NPM1 A, 78.4 % of all NPM1 mutations. For MRD monitoring a relative RT-PCR was performed in 538 samples of 178 NPM1 A positive pts with a sensitivity of 10-6. MRD was monitored at diagnosis, in aplasia, after induction therapy, after consolidation therapy, and during the follow-up. MRD levels were normalized to the housekeeping gene ABL1 and expressed as a ratio to an internal control of known concentration.
In the analysis of the NPM1 A positive and FLT3-ITD negative pts (ELN favorable risk group) 82.5% (n=85) achieved complete remission (CR) after induction therapy. With a median follow-up of 26 (range 1–118) months, 36 (42.9%) pts relapsed within this subgroup. In aplasia, and after induction therapy, pts with a long-lasting remission showed significantly lower NPM1 A ratios in contrast to pts who relapsed during the follow-up. Via Receiver-Operating Curves (ROC) we analyzed the diagnostic power to identify pts at high risk for relapse and determined clinical useable cut-offs at the different checkpoints. ROC were significantly associated with disease recurrence at the checkpoints in aplasia and after induction therapy, but not after consolidation therapy. After induction therapy, a cut-off with a ratio of 0.01 was determined. This cut-off separates the patient cohort into two prognostic groups. NPM1 A MRD levels above the cut-offs result in an increased risk of relapse compared to pts with MRD level below this cut-off. This is reflected in a significantly lower 2-year relapse free survival (RFS) of 18% versus 72% (Figure 1). In 25 pts of this favorable risk group follow-up samples in CR were available for analysis of an upcoming relapse within 100 days of sampling. Only 2 of these pts developed relapse within of the next 100 days, but both pts showed increasing MRD levels prior to relapse. 18 relapse samples were available in this subgroup and interestingly, one patient (5.5%) was NPM1 A negative at relapse. When we further enrolled the FLT3-ITD positive pts into our analyses, not surprisingly we found a negative impact on the RFS of MRD positive and MRD negative pts.
Our results confirm the observations of other studies that showed the prognostic impact of NPM1 MRD monitoring by RT-PCR. With the MRD monitoring we could identify pts at high risk for relapse within the ELN favorable risk group. Particularly high MRD levels after the induction therapy were strongly associated with a worse RFS. This and previously published data of others demonstrate that in addition to pre-therapeutic factors, the individual MRD course should be used as prognostic factor for the guidance of treatment and pts with high or increasing levels of MRD should undergo allogeneic stem cell transplantation, if eligible.
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
Asterisk with author names denotes non-ASH members.