Core Binding Factor Acute Myeloid Leukemia In Pediatric Patients Enrolled In The AIEOP AML 2002/01 Trial: The Impact Of Minimal Residual Disease On Patient Outcome

Acute Myeloid Leukemia (AML) accounts for 15% of pediatric leukemia. Improvements of prognosis achieved in the last twenty years are also due to a better stratification of patients into risk groups that allowed to deliver tailored treatment. The AIEOP LAM 2002/01 protocol classified patients with t(8;21)(q22,q22)AML1-ETO and inv(16)(p13;q22)CBFB-MYH11, who responded to induction therapy as belonging to the Standard-Risk (SR) group. These patients were considered at good prognosis; unfortunately, although they all reached morphological complete remission (CR) after the first induction course, they showed a high incidence of relapse (24%) (Pession A. et al., Blood 2013). At present, little is known about the kinetics of relapse or the development of resistance mechanisms; moreover, in many studies post-remission therapy in AML does not take into account the level of residual leukemia. Here, we evaluate the prognostic impact of molecular minimal residual disease (MRD) levels in terms of onset of relapse.

We studied bone marrow of 49 and 30 patients carrying either t(8;21) or inv(16) abnormalities, respectively, at time of diagnosis, and at the end of first and second course of induction therapy (ICE: Idarubicine, Citarabine and Etoposide). MRD was evaluated as number of fusion transcripts by quantitative RT-PCR using absolute quantification, and calculated as logarithmic (Log) disease reduction after the first and second ICE course as compared to diagnosis. Fusion transcripts were normalized to 10(5) ABL copies. The prognostic impact was assessed by the calculation of Overall Survival (OS) and cumulative incidence of relapse (CIR) according to the different MRD levels.

After I ICE, t(8;21)-rearranged patients distributed equally in the four classes of MRD reduction. In particular, 10 patients reduced MRD levels of less than 1 Log, 11 patients reduced 1-2 Log, 17 patients reduced 2-3 Log and 11 patients reduced MRD levels of more than 3 Log as compared to diagnosis. After II ICE, the majority of patients increased the reduction of MRD levels (> 3 Log). We focused on the 10 patients who did not respond to the I ICE course and we found that 7 out of 10 reduced MRD levels of more than 1 Log after the II ICE, confirming a slow clearance of blasts for this subgroup of patients. We found that a reduction of MRD lower than 1 Log conferred a worst outcome to t(8;21)-rearranged patients (OS being 53.3% vs 85.2%, p=0.073 after I ICE; 33.3% vs 82.4 p= 0.062 after II ICE), and higher CIR after I and II ICE (52% vs 12.3%, p=0.0049, and 66.7 %, vs 17.2, p=0.024, at 6 years respectively). The group of inv(16)-rearranged achieved more than 1 Log reduction of MRD since I ICE course. After II ICE course, the majority of of patients reduced MRD of more than 3 Log, and 20% of them were completely negative. Inv(16) rearranged patients were all alive at last follow-up, and CIR never showed statistically significant differences for MRD levels in inv(16)-rearranged patients after the induction therapy. The impact of copy numbers (i.e. transcript levels) at diagnosis and after induction courses on relapse risk never showed significant results in both subgroups of CBF-rearranged patients.

Our data indicate that MRD evaluation using quantitative RT-PCR is an important diagnostic tool that permits the assessment of response to CT and the identification of patients at greater risk of relapse after induction therapy for AML1-ETO rearranged patients. We propose that at the end of induction therapy the cut-off of MRD< 1 Log be used to guide therapeutic decisions for this subgroup of SR patients. On the contrary, MRD levels of CBFB-MYH11 rearranged patients do not have prognostic value on CIR after induction therapy, this suggesting that novel molecular features should be investigated for this subgroup of AML.

No relevant conflicts of interest to declare.