Characteristics of FLT3-ITD and Efficacy of Sorafenib Treatment in Pediatric AML: A Report from AML-BFM Study Group

Introduction

Internal tandem duplication (ITD) of FMS-related tyrosine kinase 3 (FLT3) gene is one of the most common somatic mutations in acute myeloid leukemia (AML). We and others have previously shown that the prognostic significance of FLT3-ITD in children and young adults relies on its characteristics e.g., co-occurrence with WT1 mutations and NUP98::NSD1 fusions (Niktoreh et al., 2019) or higher allelic ratio (AR) (Tarlock et al., 2018) lead to worse survival outcomes. Also, recent studies suggested that treatment with FLT3 inhibitors, including sorafenib, might be effective in children with FLT3-ITD positive AML. In this report, we summarize the characteristics of FLT3-ITD and outcomes of treatment with sorafenib in patients recruited in AML-BFM consecutive trials.

Methods

Patients aged 0-21 years old, diagnosed with FLT3-ITD positive de novo AML between Jan. 2004 and Dec. 2021 in Germany and recruited in AML-BFM 2004 and 2012 studies or the 2012 and 2017 registries were retrospectively analyzed. All patients with FLT3-ITD with high- or intermediate risk stratification (HR or IR) were recommended to receive 100 mg/m² sorafenib taken orally twice daily on days 7 to 21 of their induction and consolidation chemotherapy blocks, the final decision however was made by the treating physicians. Patient samples were examined in AML-BFM reference laboratory in Essen, Germany using standard methods. Patients with Down syndrome and AML-M3 were excluded from the current analysis and survival rates were calculated using Kaplan-Meier method and Log-Rank test (significance threshold p<0.05).

Results

In total, 137 patients (51% male) with median age of 7.78 (1.3-19, IQR=7.27) years at diagnosis were included. Most of the patients were initially stratified in IR group (70%, N=95) followed by HR (26%, N=35). FABs M2, M4 and M1 consisted 74% of all morphologic subtypes with 26%, 25% and 23% of patients, respectively and with a median of 64,000 (500 -700,400, IQR=53,178) x10⁹/L, the majority of patients showed a leukocytosis at diagnosis. The median ITD length in the 107 patients with available data was 51 (9-111, IQR=39) bps and 54% of the 91 patients with available data had an AR above 0.5. WT1 mutations and NUP98::NSD1 were detected in 25% (N=34) and 7% (N=10) of patients, respectively and 4% (N=6), harbored both mutations simultaneously. The three-year probability of overall and event-free survival (3y-OS and 3y-EFS, respectively) in all FLT3-ITD patients were 68.8% (±4.3) and 47.6% (±4.5), respectively and although harboring ITDs longer than the median length had a tendency to worsen the 3y-OS (54.4% in ITD>51 bps vs. 73.3% in ITDs<51 bps, p=0.071), it did not influence the 3-y EFS (41.7% vs. 54.2%, p=0.204). Similarly, an AR higher than 0.5 did not affect the survival outcomes (3y-OS: 67.4% vs. 60.4%, p=0.744, 3y-EFS: 46.8% vs. 47%, p=0.618). In total, 36 patients (27%) were treated with sorafenib of which, 23 (63%) received a consecutive HSCT. We observed no survival advantage between sorafenib-treated and sorafenib-naive groups of patients (3y-OS: 73.7% vs. 67.3%, p=0.408, 3y-EFS: 48.3% vs. 47.5%, p=0.929). Patients with longer ITDs did not benefit from sorafenib treatment (3y-OS: 68% vs. 54.3%, p=0.279, 3y-EFS: 48.9% vs. 38.4 %, p=0.649) but higher AR, showed a tendency of improved survival outcomes when sorafenib treatment was implicated (3y-OS: 80.8 % vs. 61.9 %, p=0.159, 3y-EFS: 67.3 % vs. 39.9 %, p=0.131). Although the number of patients were limited, we observed a possible survival advantage in double or triple FLT3-ITD, NUP98::NSD1 and WT1 mutated patients who received Sorafenib: from 28 patients with WT1 mutations only, four patients received sorafenib three of which, are currently alive and in CR but contrarily, from the 24 untreated patients, 14 (58%) are deceased (78% due to relapse). Additionally, from 10 patients with NUP98::NSD1 (w/o WT1 mutations), two who were treated with sorafenib are alive and in CR, whereas from the remaining 8 patients, 6 (75%) experienced relapse and unfortunately did not survive.

Conclusions

Our preliminary analysis implicates that although addition of sorafenib to the intensive chemotherapy of children and adolescence with FLT3-ITD positive AML, did not improve the survival outcomes in all patients; specific genetic subgroups might benefit from this treatment. These results must be validated in bigger cohorts of patients with longer follow-ups.