Potential Predictors of Induction Failure and Complete Remission Duration in FLT3-ITD Mutated Acute Myeloid Leukemia

Background

Internal tandem duplication (ITD) mutations inFLT3 (fms-like tyrosine kinase III) are detected in approximately 25% of patients (pts) with acute myeloid leukemia (AML) and are associated with an increased risk of relapse and adverse outcome. In the present study we evaluate the potential predictors of induction failure and CR1 duration in pts with FLT3-ITD mutated AML.

Methods

We retrospectively evaluated adult pts with FLT3-ITDmutated AML treated at our institution from July 2012 to January 2018. Pts with acute promyelocytic leukemia or core-binding factor AML were excluded. Demographic, clinical, laboratory and pathologic data were obtained from clinical records. FLT3-ITD mutational analysis was performed using PCR-based DNA sequencing. FLT3-ITD was quantified by the quantity of mutant allele burden as a percentage of total allele burden, high level was defined as ≥50%. Response to treatment was evaluated according to the 2003 revised International Working Group AML criteria. Patients with available cytogenetic data were classified according to the European Leukemia Net (ELN) guidelines. Statistical analysis included Mann Whitney test for continuous, Fisher's exact test for categorical variables and Kaplan-Meier survival curves for overall survival (OS) and relapse free survival (RFS).

Results

We identified 235 pts.The median age was 64 years (19-92) with 45% being male. Cytogenetic data at diagnosis was available in 166 pts and risk stratification by the ELN criteria included, 33 (20%) favorable, 117 (70%) intermediate and 16 (10%) adverse risk, respectively. Forty-two (18%) pts had a history of antecedent hematological disease (AHD) and 26 (11%) had therapy related AML (t-AML). Pts were treated with several different therapies as induction treatment for AML, including: high dose cytarabine based (HiDAC) in 84 (36%), conventional dose cytarabine based in 59 (26%), hypomethylating agents (HMA) in 70 (29%) and other in 22 (9%). Approximately half of pts (n=120, 51%) received a FLT3 inhibitor during induction. A total of 166 (71%) pts achieved a complete remission (CR)/CR with incomplete hematologic recovery (CRi) and 69 (29%) failed induction. Baseline characteristic of the pts are shown in Table 1. Compared with responders, pts with primary induction failure were more frequently treated with HMA based induction (p=0.001), were older (p=0.002), male (p=0.01), had an increased incidence of AHD (p=0.04), lower levels of FLT3-ITD allele ratio (p=0.04), and adverse cytogenetic risk (p=0.001). There was no significant difference in analyzed molecular mutations (Table 1). Among the 166 complete responders; 92 (61%) remain in CR1 over a median follow-up of 16 months (range, 2-69). Of the 74 (39%) pts that relapsed, 38 (23%) relapsed within 6 months of CR1, 27 (16%) between 6 to 12 months and 9 (5%) relapsed after 12 months of CR1. Median duration of CR1 was 8 months (range, 0.2-68). We classified pts into 2 categories based on their duration of CR1 including: CR1 duration <6 months (n=38 (23%)) and ≥ 6months (n=105, 45%) (Table 2). Among baseline characteristics, older age (p=0.009), t-AML diagnosis (p=0.02) and EZH2mutation (p=0.004) were associated with a CR1 duration less than 6 months. Among treatment related characteristics, achieving CR instead of CRi (p=0.007) and minimal residual disease negativity (MRD^neg) by flow cytometry were associated with a CR1 duration ≥6 months. There was a trend in favor of cytarabine based induction regimens (p=0.08). On univariate analysis, age >65 years (p=0.001), presence of EZH2(p>0.001) or RUNX1(p=0.01) mutations at baseline were associated with early relapse; achievement of a MRD^neg by flow cytometry (p=0.0015), CR vs CRi (p<0.001), receiving a HiDAC based induction (p<0.001) or allogeneic cell transplant (AlloHCT) at CR1 (p<0.001) were associated with longer RFS. Treatment with FLT3 inhibitor, high FLT3-ITD allelic burden as well as NPM1mutation at baseline had no significant effect (p=0.07 and p=0.3, respectively). Once censored for AlloHCT, achievement of a negative MRD by flow cytometry (p=0.01), CR instead of CRi (p<0.003) and FLT3 inhibitor treatment (p=0.001) demonstrated a significant increase in RFS.

Conclusions

Achieving a CR and/or MRD^neg identified by flow cytometry as well as the intensity of the conditioning regimens demonstrated a significant increase in RFS. The role of EZH2andRUNX1warrants further investigation.

View largeDownload slide

View largeDownload slide

Close modal