Efficacy and safety of FLAG-IDA as front-line therapy in de novo paediatric acute myeloid leukaemia population

Acute myeloid leukaemia (AML) requires intensive chemotherapy for cure. Regimens from the Children's Oncology Group (COG) represent the most common treatment approach to paediatric AML in North America. These regimens utilize up to five cycles of intensive chemotherapy, with a high anthracycline burden (~492–684 mg/m2 doxorubicin equivalents),1 contributing to acute morbidity and long-term cardiotoxicity.2 The United Kingdom Medical Research Council (MRC) AML 15 randomized trial tested a new double induction regimen of fludarabine, high-dose cytarabine (HDAc), G-colony stimulating factor (GCSF) and idarubicin (FLAG-IDA) in adults and a small subset of paediatric patients.3 MRC 15 included four cycles of chemotherapy with lower intensity and less anthracycline exposure (cumulatively, ~240 mg/m2 doxorubicin equivalents), and demonstrated excellent survival.3 As a result, we adopted MRC-15-based therapy as routine care. Between 2014 and 2021, we treated 30 patients with de novo AML (Table 1) and describe here disease response, morbidity and survival. Each case was characterized by cytogenetics, chromosomal microarray and a proprietary molecular testing panel (OncoKids®).4 Treatment was delivered per MRC 15,5 adapted to reflect institutional standard of care. Briefly, patients received double induction chemotherapy (fludarabine 30 mg/m2 and cytarabine 2 g/m2 IV days 2–6, idarubicin 8 mg/m2 IV days 4–6, GCSF 5 mcg/kg days 1–7 [optionally omitted for patients with leukocytosis]) followed by double consolidation (cytarabine 3 g/m2 q12 hours days 1,3,5). Following regulatory approval in 2017, patients with CD33+ received gemtuzumab ozogamicin (GO, 3 mg/m2 IV day 7, maximum dose 4.5 mg) as per treating physician discretion. Tyrosine kinase inhibitors (TKI) were optionally included for patients with FLT3 mutations. Disease response was assessed by flow-cytometric minimal residual disease (MRD) following Induction I (and II, if positive) using a ‘difference from normal’ approach.6 MRD+ was defined using COG threshold at the time of treatment (≥0.1%).7 Final risk category was assigned from presenting molecular analyses and post-Induction I MRD using COG classification at the time of diagnosis (per AAML1031 or AAML1831) (Table S1). Stem cell transplantation (SCT) in first complete remission (CR1) was reserved for high-risk patients. Supportive care included hospitalization during neutropenia and antimicrobial prophylaxis. During consolidation cycles, patients afebrile without infection were discharged with close follow-up and readmission for monitoring once neutropenic (ANC <500/μL). This retrospective study was approved by the Institutional Review Board. Patients' demographics, presenting features, treatment, disease response, cardiac outcomes, length of hospitalizations and survival were collected. To analyse disease response and survival, patients were classified using the contemporary AAML1831 risk stratification, including MRD threshold (≥0.05%).8 Cardiac, liver and infectious toxicity during treatment cycles were captured. Invasive fungal infections were classified following central review by an infectious diseases expert using international consensus definitions.9 Cardiac dysfunction was defined as left ventricular shortening fraction (LVSF) <28%. Event-free survival (EFS) was defined as start of therapy to first of refractory/relapsed disease, second malignancy or death. Overall survival (OS) was defined as start of therapy to death or latest follow-up. EFS and OS were estimated at 3 years and stratified by risk category. All patients completed double induction with FLAG-IDA, and 27/30 (90%) patients completed all planned therapy (persistent thrombocytopenia, early relapse post-consolidation I, physician discretion). Following Induction I, 28/30 patients (93%, 95% confidence interval [CI] 78%–99%) achieved MRD-negative CR1. Both MRD+ patients (one with high-risk genetics) achieved MRD negativity following Induction II. All patients were MRD negative post-Induction II (100%, 95% CI 88%–100%). Of patients classified as high risk at the time of treatment, 9/9 (100%) underwent HSCT in CR1. For the overall cohort, median follow-up time for patients alive at last follow-up was 2.3 years. Three-year EFS and OS rates for the cohort were 73 ± 9% and 80 ± 8% (Figure 1A,B). Following risk classification per AAML1831, 3-year EFS for low- versus high-risk patients was 87 ± 9% and 44 ± 17% (LRT p = 0.009) and 3-year OS 94 ± 6% and 51 + 18%, respectively (LRT p = 0.010) (Figure S1). All patients with low-risk molecular findings at diagnosis were alive at latest follow-up (3-year OS 100%). Survival outcomes following reclassification using only presenting genetic features (per AAML1831) are described in supplemental data (Figure S2A,B). The regimen was well tolerated, with acceptable toxicity, and preserved cardiac function in all patients (Table S2). As described here, our MRC-15-adapted regimen to treat children and young adults with de novo AML was well tolerated, decreased burden from hospitalization (Table S3), limited cardiac morbidity, excellent disease response and maintained survival comparable to other paediatric regimens.5, 10 Reduction in cardiotoxicity from AML therapy is critical; recent studies have reported that late cardiomyopathy following AML therapy is common, affecting 15%–27% of survivors.11 The MRC-style FLAG-IDA regimen, with markedly reduced anthracycline exposure, addresses this need. While a small cohort, it is noteworthy that no patients developed clinical cardiotoxicity. Moreover, this was achieved while preserving excellent disease response, with 100% of patients achieving MRD negativity after double induction, and favourable survival rates, particularly for low-risk patients. The regimen was well tolerated, with manageable infectious toxicity, and no treatment mortality. We would anticipate that consistent integration of contemporary targeted therapies (e.g. FLT3 TKI, GO), and supportive care (e.g. antimicrobial prophylaxis for all patients), would further improve survival and reduce treatment morbidity. While direct comparisons to larger trials are challenging, MRD and survival data in our cohort compare favourably to front-line COG regimens. Of the 1096 patients enrolled in the COG trial AAML1031, MRD was available in 95% of patients and was negative (<0.05%) post-Induction I in only 75% of patients. Similarly, the 3-year EFS for the superior treatment arm on AAML1031 and AAML0531 was 45.9% and 53.1%, respectively, with a corresponding 3-year OS of 65.4% and 69.4% for each trial.5, 10 Reported survival for FLAG-IDA remained comparable when stratified by risk group. The use of FLAG-IDA in our patients preserving disease response while also reducing the number of treatment cycles and duration of inpatient hospitalization to decrease the burden on patients and families. We must acknowledge that the data are from a single institution study describing routine chemotherapy, delivered outside of a clinical trial and with short-term follow-up. Nonetheless, these findings support continued use of the regimen and support the need for further study and replication in a multicenter, and ideally, randomized trial setting. Holly Huang and Dr Hadar contributed to the original study conception and design, performed data acquisition, assisted in data analysis and interpretation, performed the literature search, and revised subsequent versions of the manuscript. Dr Doan and Dr Orgel designed the study, aided with data acquisition, supervised data interpretation, drafted the initial manuscript and revised subsequent versions. Jemily Malvar analysed and interpreted the data and revised subsequent versions of the manuscript. Dr Rushing and Dr Dao assisted in the identification of potential subjects for the study and revised subsequent versions of the manuscript. Dr Bhojwani, Dr Freyer, Dr Gaynon, Dr Stokke and Dr Parekh aided with revision of data interpretation, and revised subsequent versions of the manuscript. Dr Raca and Dr Kovach aided with data acquisition and verification of molecular classification of subjects and revised subsequent versions of the manuscript. Dr Posch aided with data acquisition and verification of infectious classification of subjects and revised subsequent versions of the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. Dr Parekh is a consultant for Appia Bio and has equity in PLUTO and Amgen. His spouse is also an Amgen employee. Dr Gaynon chairs a DSMC for Takeda (Ponatinib in Ph + ALL). All conflicts are unrelated to the information presented in the manuscript. The data that support the findings of this study are available from the corresponding author upon reasonable request. Appendix S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.