Clonal evolution of acute myeloid leukemia with FLT3-ITD mutation under treatment with midostaurin

Abstract

In the international randomized phase 3 RATIFY (Randomized AML Trial In FLT3 in patients less than 60 Years old) trial, the multikinase inhibitor midostaurin significantly improved overall and event-free survival in patients 18 to 59 years of age with FLT3-mutated acute myeloid leukemia (AML). However, only 59% of patients in the midostaurin arm achieved protocol-specified complete remission (CR), and almost half of patients achieving CR relapsed. To explore underlying mechanisms of resistance, we studied patterns of clonal evolution in patients with FLT3-internal tandem duplications (ITD)-positive AML who were entered in the RATIFY or German-Austrian Acute Myeloid Leukemia Study Group 16-10 trial and received treatment with midostaurin. To this end, paired samples from 54 patients obtained at time of diagnosis and at time of either relapsed or refractory disease were analyzed using conventional Genescan-based testing for FLT3-ITD and whole exome sequencing. At the time of disease resistance or progression, almost half of the patients (46%) became FLT3-ITD negative but acquired mutations in signaling pathways (eg, MAPK), thereby providing a new proliferative advantage. In cases with FLT3-ITD persistence, the selection of resistant ITD clones was found in 11% as potential drivers of disease. In 32% of cases, no FLT3-ITD mutational change was observed, suggesting either resistance mechanisms bypassing FLT3 inhibition or loss of midostaurin inhibitory activity because of inadequate drug levels. In summary, our study provides novel insights into the clonal evolution and resistance mechanisms of FLT3-ITD-mutated AML under treatment with midostaurin in combination with intensive chemotherapy.

Graphical abstract

Figure 1.

Presence of FLT3-ITD clones at diagnosis and disease progression based on a Genescan-based routine diagnostic assay. Each column represents 1 patient. (A) Presence of FLT3-ITD clones at diagnosis (D) and relapse/refractory disease (R) in patients treated with midostaurin. (B) Presence of FLT3-ITD clones between D and R in 21 intensively treated patients without the addition of midostaurin. Green indicates that the same ITDs were detected at both time points. A change of ITDs between D and R is color-coded in orange (light, loss of ITDs at progression with at least 1 persistent FLT3-ITD between both time points; medium, switch of ITD insertion site or ITD length; dark, gain of ITD at R). Blue indicates that no FLT3-ITDs were detected at R.

Figure 2.

Distribution of somatic variants detected with WES and stability of mutations at D and R. (A) Number of all mutations detected with WES in patients treated with midostaurin (n = 54), which were either present only at D (blue), at R (red), or detectable at both time points (purple intersection). (B) Median number of mutations in patients treated with midostaurin present at the time of diagnosis (blue) or relapse/refractory disease (RD, red). Each dot represents the number of mutations found in a given patient at a given time point. P value was calculated using the Mann-Whitney test. (C) Stability of mutations between D and R in patients treated with midostaurin and the control group. Each bar represents 1 patient. Given is the percentage of mutations present only at D (blue), at R (red), and present at both time points (yellow). Bars are arranged in descending order of stably detectable mutations (yellow). Information on the presence of FLT3-ITD (+ = positive; − = negative) at R is given on the top of each bar.

Figure 3.

Recurrently mutated genes at D and at R. (A) Frequencies of recurrent gene mutations in all patients treated with midostaurin (n = 54) present only at D (blue), only at R (red), and at both time points (yellow). Genes are arranged according to functional groups as described at the top of the graph. (B) Presence of gene mutations in patients treated with midostaurin and the control group at D and R. Genes are arranged with regard to functional groups. Each column represents a single patient. For each patient, the presence of mutations is given at both time points in subsequent rows. Blue, presence of a mutation; gray, wild type. Mutations present only at R are highlighted in red. The number (n) and percentage (%) of mutations at the respective time points are given next to the figures.

Figure 4.

Patterns of clonal evolution in relapse patients. Fish plots visualizing patterns of clonal evolution in relapse patients according to the variant allele frequencies at diagnosis, remission, and relapse in patients with persistent FLT3-ITD (relapse occurred after midostaurin maintenance cycle 11; A), change of FLT3-ITD (relapse after cycle 3; B), and loss of FLT3-ITD (relapse after cycle 3; C). Information on treatment and BM blast count is given below each plot. HDAC, high-dose cytarabine; HSCT, hematopoietic stem cell transplantation; NA, not available.

Figure 5.

Patterns of clonal evolution in patients with refractory disease. Patterns of clonal evolution in refractory (RD) patients with loss of FLT3-ITD at RD (A-B) or persistent FLT3-ITD at RD (C-D). VAFs are given on the left of each graph and the allelic ratios of FLT3-ITDs on the right. Persistent mutations are color-coded in yellow, lost mutations in blue, and gained or upregulated mutations in green. FLT3-ITDs are indicated by dotted lines.