Real Life Analysis of the Rete Ematologica Lombarda on ELN Favorable Acute Myeloid Leukemia: The Molecular Remission Is a Necessary Goal for a Good Outcome in All Different Cytogenetic/Molecular Subgroups

Introduction: The European Leukemia Net (ELN) stratification classifies as favorable prognosis four different subgroups of acute myeloid leukemia (AML) with specific cytogenetic/molecular abnormalities. These forms are potentially curable with standard chemotherapy, even though about 30% of patients (pts) still relapse. Overall survival (OS) of 60% at 5 years is reported but it is not clear if all different subgroups show an homogeneous outcome. The purpose of this study was to assess, in a real life setting, the response rate and the outcome of newly diagnosed AML patients with favorable prognosis based on ELN classification. Data were collected from six hematological centers of the regional network REL (Rete Ematologica Lombarda).

Methods: Between 2007 and 2014 we analyzed adult AML patients with t(8;21)(q22;q22)/RUNX1-RUNX1T1(group1), inv(16)(p13q22) or t(16;16)(p13q22)/CBFbeta-MIH11(group 2), normal karyotype and mutated NPM1 and negative FLT3-ITD (group 3) or double mutated CEBPA (CEBPA dm) (group 4). All patients received induction, mainly standard 3+7 chemotherapy (or reduced 1+5 in older pts) in 66% of cases, followed by high dose cytarabine based consolidation in 77% of pts. Clinical and molecular data were analyzed at diagnosis and at different time points during treatment and follow up with qualitative and quantitative PCR. The Kaplan-Meier product-limit method was used to estimate survival curves, and the log-rank test was adopted to evaluate differences between groups of pts.

Results: We studied 177 AML pts (96 males and 81 females) with a median age of 55 years (range 20-80): 27 of group 1, 35 of group 2, 98 of group 3 and 17 of group 4. Complete hematological response (CHR), assessed in 176 pts, was achieved in 161 pts (92%): 23(85%), 32(91%), 90(93%), 16(94%) respectively in group 1, 2, 3, 4 without significant difference (p=0.26). Molecular complete remission (mCR), evaluated in 152 pts out of 161 in CHR, was documented in 102 pts. Forty pts entered mCR after induction and 62 pts at the end of consolidation or later (1-40 months); therefore, mCR rate was 67%. Considering the type of the method, 48 pts resulted in mCR with qualitative PCR, the remaining 54 (53%) with qualitatitive and quantitative PCR. Median time to mCR was 4 months and was different among groups (p=0.03); in particular, t(8;21) pts obtained mCR later than NPM1 pts (8 versus 2.6 months) (p=0.01). The relapse rate was 41% and the median time to relapse was 10.2 months, without any difference among groups (respectively p=0.5 and p=0.8). Five years overall survival (OS) and disease free survival (DFS) were 65% and 47% with no significant difference among groups (respectively p=0.3 and p=0.9). Age (<60 years) was the only parameter that affected OS (80% versus 41%, p<0.001) and DFS (56% versus 33%, p=0.038). In a time-dependent analysis there is a significant association between mCR and DFS, therefore, if mCR was reached the risk of relapse decreased of 60% (p<0.001). Moreover, if qualitative mCR was confirmed by quantitative analysis, 5 years-DFS was much higher (78%).

Conclusion: This study, conducted in a non selected population of favorable risk AML patients, confirms high CR rate and the prognostic value of age, favoring patients younger than 60 years. Our data confirm that the ELN classification identifies a clinically homogeneous group of good risk AML patients sharing comparable outcome even outside clinical trials setting. Molecular complete remission is associated with better outcome, in particular when assessed by quantitative PCR.