Molecular Residual Disease Monitoring Provides Insufficient Lead-Time to Prevent Morphologic Relapse in the Majority of Patients with Core-Binding Factor AML

Introduction

Core-binding factor (CBF) acute myeloid leukemias (AML) are characterized by the chromosomal abnormalities t(8;21) or inv(16)/t(16;16) and their fusion proteins RUNX1/RUNX1T1 and CBFB-MYH11, respectively. Despite their relatively favorable prognosis, it has been reported that up to 45% of patients with CBF-AML relapse. Current guidelines recommend monitoring the peripheral blood (PB) or bone marrow (BM) for molecular evidence of residual disease (RD) every 3 months for 2 years following remission. By monitoring patients for rising molecular transcripts, those at risk of impending relapse can be identified and treated prior to the emergence of overt disease. However, the relapse kinetics of CBF-AML are poorly-understood, and it is unknown if serial RD monitoring can detect molecular relapses with sufficient lead-time to intervene and prevent morphologic relapse.

Methods

After local REB approval, we identified patients with CBF-AML treated at the Princess Margaret Cancer Centre in Toronto, Canada between 2000 and 2017. Patients underwent induction and consolidation chemotherapy according to standard protocols followed by RD monitoring with polymerase chain reaction (qPCR) of RUNX1/RUNX1T1 or CBFB-MYH11 transcripts in a CAP/CLIA certified lab every 3 months for a median of 1.2 years (range 0-5.3). RD assessment was performed on BM aspirates, but PB could be tested in patients who could not tolerate repeat BM aspirates. Morphologic relapse was defined as emergence of >5% blasts in the PB or BM, and molecular relapse as a positive transcript PCR (if previous PCR measurements were undetectable) or an increase by 1 log (if previous PCR measurements were detectable) on 2 successive samples without morphologic relapse. Rapid relapse was defined as <100 days from molecular to morphologic relapse (i.e. timeframe insufficient to administer pre-emptive treatments such as allotransplant to prevent morphologic relapse). We extracted relevant clinical and cytogenetic data and performed unpaired t-tests and Fisher's exact tests as appropriate using GraphPad statistical software and version 9.4 of the SAS System for Windows (SAS Institute, Cary, NC).

Results

We included 114 patients with CBF-AML. Median age was 46.5 years (range 18-79). t(8;21) was present in 59% and inv(16)/t(16;16) in 41% of patients. All patients achieved remission with 7+3 induction chemotherapy. Over a median follow-up time of 3.7 years (range 0.2-14.3), RD measurements were performed a mean of 5 times per patient with mean sampling interval of 103±54 days. Remission was maintained in 71 (62%) patients but 43 (38%) developed morphological (n=34) or isolated molecular relapse (n=9), with median time to relapse of 4.4 months (range 1.4-31.4). Patients with relapsed disease were significantly less likely to have achieved ≥3 log reduction in RUNX1/RUNX1T1 or CBFB-MYH11 BM transcripts at the end of consolidation chemotherapy compared to patients who remained in remission (75% vs 90%, p=0.046). Of the 43 patients who relapsed, the majority (74.4%, n=32) had rapid relapse kinetics. Of these 43 patients, 25 received reinduction chemotherapy with achievement of CR2, 6 had refractory disease or death, 1 was lost to follow-up, and 17 went on to receive allotransplantation. RD monitoring enabled timely detection of impending relapse and permitted intervention prior to morphologic relapse in only 11 patients (25.6%). Among these 11 patients with slower relapse kinetics, 6 received reinduction chemotherapy with achievement of CR2, 2 received reinduction chemotherapy with refractory disease, and 8 went on to receive allotransplantation. Median overall survival was 20 months (range 3-128) for patients with rapid relapse kinetics and 28 months (range 14-166) for patients with slow relapse kinetics (p=0.02). Clinical features, BM vs PB RD measurements, additional molecular mutations, and cytogenetic abnormalities did not distinguish patients with rapid vs slow relapse kinetics.

Conclusions

Current guidelines recommend molecular RD monitoring every 3 months for CBF-AML. However, in the majority of patients who relapsed at our institution, RD monitoring every 3 months provided insufficient lead-time to identify molecular relapses prior to morphologic relapse. Further research is warranted to identify the patients with CBF at the highest risk of relapse and the best strategies to monitor these patients over time.