Identification of Relapse Risk Using a Multiparameter Flow Cytometry-Based Detection of Minimal Residual Disease in Adult Patients with Acute Myeloid Leukemia at ELN Intermediate-Risk

Background: Minimal Residual Disease (MRD) measurement is a useful tool to refine prognosis in adult Acute Myeloid Leukemia (AML) patients in morphological complete remission (CR). Multiparametric flow cytometry (MFC) may provide independent prognostic information particularly in intermediate (Int)-I and Int-II ELN 2010 risk groups.

Patients and methods: Analysis was performed on bone marrow (BM) aspirates of 120 ELN Int-I/II patients (pts) from 2 independent institutions, separately analyzed in two MFC labs. Treatment consisted of one-two idarubicin-based induction courses followed by High-dose-Ara-C consolidation and by allogeneic transplantation (NILG, n: 18/51; GIMEMA, n: 20/69). The median age was 49yr (18-73yr), F/M ratio was 49/71; de-novo and secondary AML cases were 94.2% and 5.8%, respectively. FLT3 gene was mutated in 41.7% of the patients (50/120), being wild type in 56.6% of the cases (68/120). FLT3-ITD was detected in 35.0% of the patients (42/120) and FLT3-D835 was detected in 7.5% of the patients (9/120). NPM1 gene was mutated in 31.6% of the patients (38/120) being wild type in 65.8% of the cases (79/120). ELN 2010 int risk groups were distributed as follows: 79.2% Int-I (n: 95) and 20.8% Int-II (n: 25). At diagnosis, a panel of 22 monoclonal antibodies, combined in four 8-color tubes, was employed to detect leukemia-associated immunophenotypes (LAIPs). MRD was measured after the first cycle (TP1) and after the second cycle (TP2). MRD was considered positive if residual leukemic cells were equal or higher than 0.1% and 0.035% at TP1, and than 0.035% at TP2, respectively. Relapse-free survival (RFS) and overall survival (OS) were analyzed using Kaplan-Meier curves and the log-rank test.

Results: One-hundred-eighteen and 116 patients were tested at TP1 and TP2, respectively. At TP1, MRD-positivity was detected in 72.9% (86/118) at a threshold of 0.1% and 83.9% pts (99/118) at a threshold of 0.035%. Both thresholds were significant in predicting relapse (P=0.0002 and P=0.0158 for the 0.1% and the 0.035% threshold, respectively). We observed a shorter RFS in MRD-positive versus MRD-negative patients: median RFS was 10.8 months (0.1% threshold) and 11.3 months (0.035% threshold) in MRD-positive pts as compared to 28.5 months (0.1% threshold) and 19.5 months (0.035% threshold) in MRD-negative patients. Furthermore, MRD-positive pts presented with a significantly shorter OS as compared to MRD-negative patients (0.1% threshold, P=0.0185), with a median OS of 18.2 months as compared to 33.7 months for MRD-negative pts. When the 0.035% threshold was used, OS was not different between the two groups of pts. At TP2, MRD-positivity was detected in 75.8% of the pts (88/116) and significantly predicted relapse (P=0.0023, 0.035% threshold): the median RFS was 10.8 months in MRD-positive pts vs 26.3 months in MRD-negative. Moreover, we observed a significant improvement in terms of OS in MRD-negative vs MRD-positive pts (P=0.0411), the median OS being 37.0 and 18.2 months, respectively. Notably, the proportion of MRD-positive and MRD-negative patients did not differ in terms of age, sex, ELN risk classification, FLT3 mutational status, and number of allogeneic transplanted patients.

Conclusions: Overall, our findings demonstrate a correlation between MRD levels by MFC and relapse risk in ELN-intermediate risk patients. The use of an allogeneic transplantation approach rather than other consolidation strategies in ELN-intermediate risk AML patients is often debated. MFC-based MRD potentially allows for a more appropriate and tailored post-remission therapeutic strategy in these AML patients. Furthermore, harmonization of MRD analysis between different labs and different protocols sharing a common technical approach seems to be feasible. Future prospective studies are warranted for prospective evaluation of MRD in ELN-intermediate AML patients.