MRD monitoring in AML: results of AMLSG and AML17 studies

On 16 June 2019, at the 24^th European Hematology Association Congress in Amsterdam, Frank G. Rücker and Richard Dillon presented evidence for minimal residual disease (MRD) monitoring in AML from the AML study group (AMLSG)¹ and the UK National Cancer Research Institute (NCRI) AML17 study.²

The AML17 study

The study sought to correlate pre-transplant molecular MRD status with outcome in patients with NPM1 mutant AML to identify clinically relevant MRD thresholds, assess interaction with standard prognostic factors and observe the effect of transplant type. They analysed pre-transplant blood and bone marrow samples using reverse-transcription polymerase chain reaction (RT-qPCR) and compared MRD to overall survival (OS). They also examined the effect of FMS-like tyrosine kinase 3 (FLT3) internal tandem duplication (ITD), donor source, conditioning regimen and T-cell depletion on outcome.

Patient characteristics (N=107)

• Younger adults (aged 17–69 years) eligible for intensive chemotherapy

Results

• MRD negativity (MRD^neg) pre-transplant was associated with good outcome regardless of FLT3-ITD, remission status or transplant type
• FLT3-ITD at diagnosis was associated with a poor outcome in MRD^low patients (HR=7.14; p=0.04)
• In the MRD^neg group, neither donor source, conditioning regimen or T-cell depletion, correlated with outcome
• In the MRD positive (MRD^pos) group, T-cell depletion was significantly associated with inferior survival (34% vs. 100% for patients who did not undergo T-cell depleted transplant; p=0.002)
• No effect of donor source or conditioning protocol was detected
• Peripheral blood (PB) MRD status after second induction was also associated with poor outcome in the MRD^low group (5-year OS 82% vs. 45%; p=0.033)
• Combined FLT3-ITD and PB MRD statuses stratified patients by outcome with 2y OS of 10% and 81% (adjusted HR=7.70; p<0.0001)

The AMLSG study

The AMLSG study also aimed to correlate molecular MRD status with outcome. They studied patients with AML featuring recurrent genetic abnormalities (RUNX1-RUNX1T1 gene fusion), quantifying RUNX1-RUNX1T1 transcription levels (TL) in bone marrow (BM) and PB at various treatment stages.

Patient characteristics (N=155)

• Adults with t(8;21)-positive AML eligible for intensive chemotherapy

Results

• Reduction of RUNX1-RUNX1T1 TL and MRD^neg at defined time points was a significant prognostic factor
• MR⁵ (>2.5 log reduction) after treatment cycle 1 and achievement of MR^3.0 after cycle 2 were associated with a reduced risk of relapse (p=0.034 and p=0.028)
• MRD^neg in both BM and PB after completion of therapy, was an independent favourable prognostic factor for cumulative incidence of relapse (4-year cumulative incidence of relapse (CIR) BM: 17% vs. 36%, p=0.021; PB: 23% vs. 55%; p=0.001) and OS (4-year OS rate BM: 93% vs. 70%, p=0.007; PB: 87% vs. 47%; p<0.0001)
• Follow-up RT-qPCR analyses predicted relapse in 77% of pts exceeding a cut-off of 150 RUNX1-RUNX1T1 TL in BM, and in 84% of pts with >50 RUNX1-RUNX1T1 TL in PB
• Virtually all relapses occurred within one year after end of treatment with a very short latency from molecular to morphologic relapse, indicating the need for MRD assessment at regular intervals during this period

Conclusion

The results of these studies suggest that a refined practical guideline for MRD assessment in AML is needed. The AM17 study demonstrated that in patients with NPM1 mutant AML, MRD negativity prior to transplant predicted outcome, with above 200 copies/10⁵ of Abelson murine leukemia viral oncogene homolog gene (ABL) in the PB or 1000 copies in the BM indicative of an adverse outcome. Below this level FLT3-ITD and post-induction PB status were strongly associated with poor outcome. T-cell depletion was also associated with adverse outcome in patients who were MRD positive.

In RUNX1-RUNX1T1-positive AML MRD monitoring allows for the discrimination of pts at high and low risk of relapse. Therefore, MRD in BM and PB should be analyzed after each treatment cycle and in follow-up according to MRD status as below:

• MRD^neg: PB monthly
• MRD^pos (TL <150 in BM and <50 in PB): BM 3-monthly and PB monthly
• MRD^pos (TL >150 in BM and/or >50 in PB) or increase of MRD >1-log or conversion from MRD^neg to MRD^pos: BM and PB monthly

This study also noted very short latency from molecular to morphologic relapse, indicating that MRD assessment at shorter intervals during the first year post-treatment is indispensable.