Acute myeloid leukemia (AML) with t(8;21)(q22;q22.1); RUNX1::RUNX1T1, along with AML with inv(16)(p13.1q22)/t(16;16) (p13.1;q22);CBFB::MYH11 considered as core binding factor (CBF)-AML, is known to confer a favorable prognosis.1-3 However, a considerable proportion of patients with CBF-AML, especially those with t(8;21) AML, still experience relapse, emphasizing the need of novel therapeutic approaches.4,5 Several studies have identified additional molecular alterations, such as KIT or FLT3 mutations, as risk factors for relapse and impaired survival in CBF-AML. Here, in particular the prognosis of patients with t(8;21) AML seems to be negatively impacted by these additional mutations.6-9 Thus, there is a molecular rationale to implement KIT/FLT3 inhibitors into treatment of patients with AML with t(8;21).
Accordingly, we conducted a prospective, single-arm, multi-center, phase II trial MIDOKIT (clinicaltrials gov. Identifier: NCT01830361) to evaluate the molecular guided addition of midostaurin, an oral multi-kinase inhibitor with activity on KIT and FLT3, to standard daunorubicin/cytarabine (DA)-based intensive chemotherapy (IC) in adult patients with newly diagnosed AML with t(8;21) with evidence of KIT and/or FLT3 internal tandem duplication (FLT3-ITD) mutations. Patients aged 18-65 years with previously untreated AML according to the World Health Organization (WHO) classification were eligible for molecular prescreening. Those with evidence of t(8;21)(q22;q22.1); RUNX1::RUNX1T1 with additional KIT and/or FLT3-ITD mutations and response to one course of DA induction chemotherapy (IT) performed outside the trial could be enrolled. Inclusion and exclusion criteria are summarized in the Online Supplementary Table S1.
The trial design and treatment algorithm is shown in Figure 1A. Enrolled patients were scheduled to receive a second course of DA induction chemotherapy (cytarabine 100 mg/m2/day continuously intravenously, days 1-7 and daunorubicin 60 mg/m2/day intravenously, days 3-5) and three courses of high-dose cytarabine consolidation (cytarabine 3000 mg/m2/ intravenously twice daily, days 1, 3, 5) each followed by midostaurin 50 mg orally twice daily on days 8-21, and subsequent midostaurin maintenance for 12 months.
The primary endpoint of the study was 2-year event-free survival (EFS). The primary objective was to improve 2-year EFS from a historical benchmark of 50% to 80%. The historical benchmark was based on data of 103 patients with AML with t(8;21) treated within the AML96 trial (clinicaltrials gov. Identifier: NCT00180115), the AML2003 trial (clinicaltrials gov. Identifier: NCT00180102) and the AML60+ trial (clinicaltrials gov. Identifier: NCT00180167), hereof 32 patients -matched for sex, age and number of courses of induction and consolidation chemotherapy - with evidence of KIT and/or FLT3-ITD mutations (28.1% of patients harboring FLT3-ITD mutations) showing chemo-responsiveness after IT1. Secondary endpoints included composite complete remission (CR) rate, relapse-free survival (RFS), overall survival (OS), incidence of treatment-emergent adverse events (AE) and measurable residual disease (MRD) kinetics.
For prospective molecular screening the presence of AML with t(8;21)(q22;q22.1); RUNX1::RUNX1T1 was assessed via reverse transcription polymerase chain reaction (RT-PCR) (evidence of RUNX1::RUNX1T1) and fluorescence in situ hybridization (FISH) (evidence of t(8;21)(q22;q22.1)), results were validated in one of the SAL reference laboratories. Patients were evaluated for FLT3-ITD and KIT mutations via fragment length analysis (limit of detection [LOD] 0.05 allelic ratio) and Sanger sequencing (LOD 10% variant allele frequency [VAF]), respectively.10 In addition, patients with available samples were retrospectively assessed via amplicon-based next-generation sequencing (NGS; MiSeq, Illumina, Inc; DHS-003Z human myeloid neoplasms panel + custom AML panel, Qiagen N.V.; LOD 1% VAF). Quantitative bone-marrow RUNX1::RUNX1T1 MRD was assessed via RT-PCR according to the Europe Against Cancer (EAC) initiative guidelines.11
The trial was conducted in accordance with the principles of the Declaration of Helsinki; the trial protocol was approved by the Ethics Committee of the TU Dresden. Written informed consent was provided by all patients before screening.
Between Mar 13, 2013 and Dec 15, 2017 a total of 53 patients were diagnosed with t(8;21) AML at 11 participating centers, of those, 23 patients were screened and 18 patients with AML with t(8;21) and evidence of KIT and/or FLT3-ITD mutations were enrolled. All 18 patients had received DA for IT1 outside the trial (most patients received daunorubicin at a dose of 60 mg/m2) and had shown chemo-responsiveness. Table 1 and Figure 1B, C show the baseline characteristics, molecular features, and clinical course of the enrolled patients. Patient disposition is provided in the Online Supplementary Figure S1A. The median age was 50 years (range, 27-65 years), ten patients (55.6%) were male. Sixteen patients were diagnosed with de novo AML. A total of 24 KIT mutations were found in 16 patients (88.9%), some patients harbored several (subclonal) variants, median KIT VAF was 11.94% (range, 1.13-44.47%), and most patients had exon 17 mutations (Figure 1B, D). Four patients had evidence of an FLT3-ITD mutation, the median FLT3-ITD allelic ratio was 0.16 (range, 0.05-0.36); three patients harbored FLT3-TKD mutations. (Figure 1B, E). Two patients were positive for both, KIT and FLT3-ITD. The mutational landscape was in line with previous reports, none of the patients harbored CEPBA or NPM1 mutations. Sixteen patients (88.8%) completed IT2, 12 patients (66.6%) completed consolidation therapy and started midostaurin maintenance and four patients (22.2%) completed midostaurin maintenance (Figure 1C; Online Supplementary Figure S1A). Allogeneic hematopoietic cell transplantation (HCT) was performed in seven patients (38.9%), of whom three patients were in first CR and four patients received salvage HCT after relapse. Of note, one patient was diagnosed with colorectal cancer (CRC) AJCC stage IVC after IT2, he was withdrawn from the trial and died 5 months later (UPN 003-06). Another patient was withdrawn from the trial due to non-compliance (UPN 010-02) and one patient was lost to follow-up during consolidation (UPN 014-18) (Figure 1C).
Figure 1.Trial design, treatment algorithm, patient characteristics, molecular features, outcome and RUNX1::RUNX1T1 measurable residual disease kinetics. (A) MIDOKIT trial design and treatment algorithm. (B) Patient clinical characteristics and molecular features, each row represents 1 unique patient, columns represent clinical features or molecular alterations (only recurring variants/variants of special interest [ASXL2] considered pathogenic/likely pathogenic are shown). (C) Swimmer plot showing individual patient outcomes. (D) KIT variant alle frequency (VAF) according to KIT variants. (E) FLT3-ITD AR and FLT3-TKD VAF. (F-H) Event-free survival, relapse-free survival and overall survival in MIDOKIT patients as compared to historical controls (patients with acute myeloid leukemia [AML] with t(8;21) with KIT and/or FLT3-ITD mutations treated with intensive chemotherapy (IC) without midostaurin within the AML96 trial (clinicaltrails gov. Identifier: NCT00180115), the AML2003 trial (clinicaltrails gov. Identifier: NCT00180102) and the AML60+ trial (clinicaltrails gov. Identifier: NCT00180167). (I and J) Quantitative bone-marrow RUNX1::RUNX1T1 measurable residual disease (MRD), 11 patients achieved persistent hematologic complete remission (CR) (left panel), 7 patients experienced subsequent molecular and/or hematologic relapse (right panel). AE: adverse event; AR: allelic ratio; CI: confidence interval; DNR: daunorubicin; EOT: end of treatment; F: female; HCT: hematopoietic cell transplantation; HC: historical control; IT1: induction therapy 1; M: male; MRD: minimal residual disease; N: no/mutation absent; NA: not assessed ; RT-PCR: reverse transcription polymerase chain reaction; Y: yes/mutation present.
After IT2 CR rate was 100%, 16 patients (88.9%) achieved CR and two patients achieved CR with incomplete neutrophil recovery (11.1%). After a minimum follow-up of 2 years, seven patients suffered from relapse (of which 2 were molecular and 5 were hematologic relapses) and two patients died after preceding relapse. Here, patients with low KIT VAF seem to more frequently experience relapse. Detailed information on salvage therapies for patients experiencing relapse is provided in the Online Supplementary Table S2. One patient was lost to follow-up (Figure 1C). In the intention-to-treat analysis, this results in eight events and a 2-year EFS of 55.6% (95% confidence interval [CI]: 30.8-78.5), accordingly the alternate hypothesis of a 2-year EFS ≥80% was not met (Figure 1F). The median EFS, RFS and OS were not reached. At 2 years, rates of RFS and OS were 58.8% (95% CI: 39.5-87.6) and 88.2% (95% CI: 74.2-100), respectively (Figure 1G, H). No differences in EFS and RFS between patients treated within MIDOKIT and historic controls were seen. However, we observed a trend for an improved OS in patients treated within the MIDOKIT trial as compared to the historical controls (2-year OS 88.2%, 95% CI: 74.2-100 vs. 65.6%, 95% CI: 51.1–84.3; P=0.05), which was stable during longterm follow-up (5-year OS 73.5%%, 95% CI: 43.4-89.3 vs. 49.3%, 95% CI: 31.0-65.2; P=0.08) (Figure 1H; Online Supplementary Figure S1B).
Table 1.Patient characteristics.
Table 2.Non-hematologic adverse events of any grade reported in >10% of patients and the corresponding frequencies of grade ≥3 events, regardless of attribution, according to CTCAE 4.03.
Sixteen patients (88.9%) achieved complete molecular remission (CRMRD-) after IT2 (≥3-log reduction of RUNX1::RUNX1T1). Of these, five patients experienced subsequent molecular and/or hematologic relapse, 11 patients showed persistent CRMRD- (Figure 1I, J). Overall, 315 non-hematologic AE - hereof 55 AE ≥ CTCAE grade 3 - and 13 serious AE (SAE) were observed (Table 2). SAE observed included sepsis, pneumonia, sinusitis and pneumonitis; no cardiac SAE were observed. Overall, four patients had an AE entailing premature end of treatment: pneumonitis and biliary tract infection in UPN 036-05, left ventricular systolic dysfunction in UPN 072-17, tremor in UPN 010-11, and diagnosis of CRC in UPN 003-06. Two patients died; deaths were not considered to be related to midostaurin. Interruptions of mid-ostaurin administration and dose reductions were observed in five (27.7%) and eight (44.4%) patients, respectively. AE commonly associated with dose modifications were nausea, vomiting, diarrhea, and electrocardiogram QTc interval prolongation. The median number of days with midostaurin administration was 148.5 (range, 14-416 days). The median ratio of administered/scheduled cumulative dose of midostaurin was 0.34 (range, 0.04-1.06), 13 patients (72.2%) received <80% of the scheduled cumulative dose of midostaurin.
The MIDOKIT trial was the first prospective clinical trial evaluating the molecular-guided implementation of the multi-kinase inhibitor midostaurin in combination with IC in patients with newly diagnosed AML with t(8;21)(q22;q22.1); RUNX1::RUNX1T1 with evidence of KIT and/or FLT3-ITD mutations. The addition of midostaurin to IC and single-agent maintenance therapy was tolerable and feasible, and no unexpected excess in toxicity was observed. Unfortunately, the MIDOKIT trial failed to reach the assumed primary endpoint of 80% 2-year EFS, due to - among others - early relapses in 44% of patients. However, with a composite CR rate of 100% and an improved 2-year OS of 88.2% as compared to patients with AML with t(8;21) and KIT and/or FLT3-ITD mutations treated within historic AML trials without midostaurin, the results of our trial are still promising. Our findings are comparable with two similar trials conducted by the German-Austrian AML Study Group (AMLSG) and the Cancer and Leukemia Group B (CALGB) evaluating the multi-kinase inhibitor dasatinib in a non-biomarker guided fashion in combination with IC in patients with CBF-AML (4-year OS 75%, 3-year OS 77%, and 5-year OS 73% in the AMLSG, the CALGB, and the MIDOKIT trial, respectively), arguing for further evaluation of tyrosine kinase inhibitors in combination with IC in patients with CBF-AML.12,13 Moreover, a subgroup-analysis of the RATIFY trial found CBF rearrangements to be an independent predictor for favorable OS and EFS, again supporting the evaluation of midostaurin in a larger cohort of patients with CBF-AML.14,15
Of note, the MIDOKIT trial has some limitations in order to draw conclusions applicable to a larger cohort of patients, e.g., the small number of patients and the singlearm trial design. Further, non-adherence, mostly due to intolerance as well as individual protocol deviations such as allogeneic HCT in first CR, limited the time and dose of midostaurin exposure and might have contributed to the non-achievement of the primary endpoint (only 27.8% of patients received >80% of the scheduled dose of midostaurin). Moreover, the implementation of midostaurin not before IT2 might have reduced its antileukemic potential since a post hoc analysis of the RATIFY trial indicates the early use of midostaurin from IT1 on to be essential for its antileukemic effect.14 Thus, our results suggest re-evaluation of molecular informed implementation of midostaurin (or novel FLT3/KIT inhibitors) in patients with CBF-AML through a larger, redesigned, randomized, controlled trial.
In conclusion, the MIDOKIT trial was the first prospective clinical trial assessing midostaurin in patients with AML with t(8;21) in a molecular guided fashion. Although MIDOKIT failed to reach the assumed primary endpoint of 80% 2-year EFS, the promising OS supports further biomarker-driven evaluation of TKI in combination with IC in CBF-AML.
Footnotes
- Received July 4, 2022
- Accepted January 30, 2023
Correspondence
Disclosures
Novartis provided the investigational medicinal product and financial support to conduct the trial. The authors have no conflicts of interest to disclose.
Contributions
CR, GE and CT were responsible for study conception and design. CT, SH, DA and LR carried out/interpreted molecular studies. MK and LR performed statistical analyses. AH provided administrative support. LR, CR, TS, CHB, BS, KSE, SWK, MH, AR, SS, AN, JHM, JS, FS, JF, AK, RS, UP, HS, CDB, CMT, MB, GE and CT were involved in care of patients, sample procurement and/or contributed to in the data collection and interpretation. LR drafted the manuscript; and all authors agreed on the final version.
Data-sharing statement
Original data and protocols are available upon reasonable request.
References
- Döhner H, Estey E, Grimwade D. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood. 2017; 129(4):424-447. Google Scholar
- Grimwade D, Ivey A, Huntly BJP. Molecular landscape of acute myeloid leukemia in younger adults and its clinical relevance. Blood. 2016; 127(1):29-41. Google Scholar
- Arber DA, Orazi A, Hasserjian R. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016; 127(20):2391-2405. Google Scholar
- Schlenk RF, Benner A, Krauter J. Individual patient databased meta-analysis of patients aged 16 to 60 years with core binding factor acute myeloid leukemia: a survey of the German Acute Myeloid Leukemia Intergroup. J Clin Oncol. 2004; 22(18):3741-3750. Google Scholar
- Marcucci G, Mrózek K, Ruppert AS. Prognostic factors and outcome of core binding factor acute myeloid leukemia patients with t(8;21) differ from those of patients with inv(16): a Cancer and Leukemia Group B study. J Clin Oncol. 2005; 23(24):5705-5717. Google Scholar
- Paschka P, Marcucci G, Ruppert AS. Adverse prognostic significance of KIT mutations in adult acute myeloid leukemia with inv(16) and t(8;21): a Cancer and Leukemia Group B Study. J Clin Oncol. 2006; 24(24):3904-3911. Google Scholar
- Duployez N, Marceau-Renaut A, Boissel N. Comprehensive mutational profiling of core binding factor acute myeloid leukemia. Blood. 2016; 127(20):2451-2459. Google Scholar
- Jahn N, Terzer T, Sträng E. Genomic heterogeneity in core-binding factor acute myeloid leukemia and its clinical implication. Blood Adv. 2020; 4(24):6342-6352. Google Scholar
- Cairoli R, Beghini A, Grillo G. Prognostic impact of c-KIT mutations in core binding factor leukemias: an Italian retrospective study. Blood. 2006; 107(9):3463-3468. Google Scholar
- Thiede C, Steudel C, Mohr B. Analysis of FLT3-activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood. 2002; 99(12):4326-4335. Google Scholar
- Schuurhuis GJ, Heuser M, Freeman S. Minimal/measurable residual disease in AML: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2018; 131(12):1275-1291. Google Scholar
- Marcucci G, Geyer S, Laumann K. Combination of dasatinib with chemotherapy in previously untreated core binding factor acute myeloid leukemia: CALGB 10801. Blood Adv. 2020; 4(4):696-705. Google Scholar
- Paschka P, Schlenk RF, Weber D. Adding dasatinib to intensive treatment in core-binding factor acute myeloid leukemia-results of the AMLSG 11-08 trial. Leukemia. 2018; 32(7):1621-1630. Google Scholar
- Stone RM, Mandrekar SJ, Sanford BL. Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 Mutation. N Engl J Med. 2017; 377:454-464. Google Scholar
- Voso MT, Larson RA, Jones D. Midostaurin in patients with acute myeloid leukemia and FLT3-TKD mutations: a subanalysis from the RATIFY trial. Blood Adv. 2020; 4(19):4945-4954. Google Scholar
Data Supplements
Figures & Tables
Article Information
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.