Approximately 30% of adult AML subjects are refractory to induction therapy. Furthermore, of those who achieve CR, approximately 75% will relapse. FLT3-mutated AML comprise an especially poor prognosis group.

Until now, there was no established standard for relapsed subjects with FLT3 mutations and less than 20% will achieve CR with subsequent treatment.

In phase 3 Study ADMIRAL Trial, gilteritinib has resulted in CRc in over 25% of subjects receiving 120 mg/day before on study HSCT. With this treatment, the median overall survival is at 9.3 months, furthermore, gilteritinib was well tolerated at the proposed doses. This study has been designed for R/R patients for which gilteritinib as single agent has been showed to be superior to high- and low-intensity chemotherapy (Perl, NEJM 2019, Supp Table S4) and patients included in this study will receive this treatment. Beyond high- or low-intensity chemotherapy, other options available are best supportive car or other clinical trials.

The aim of this study is to assess the efficacy and safety of the addition of oral-azacitidine to salvage treatment by gilteritinib in subjects ≥18 years of age with relapsed/refractory FLT3-mutated acute myeloid leukemia

AML is characterized by the clonal expansion of myeloid blasts in the BM, peripheral blood and extramedullary tissues which disrupts normal hematopoiesis. It is a heterogeneous disease, encompassing a large number of distinctly subtypes that may have different clinical presentations and responses to treatment. AML is defined by the WHO as a myeloid neoplasm with 20% or more blasts in the peripheral blood or BM.

Internal tandem duplication (ITD) in the FLT3 gene is one of the most frequent mutations found in AML. FLT3-ITD is associated with poor prognosis and has emerged as a relevant therapeutic target. FLT3-ITD is usually conserved at relapse, suggesting that FLT3-ITD AML-initiating cells are key targets for long-lasting remission. FLT3-TKI, developed as ATP-competitive inhibitors, are currently the focus of new development strategies in FLT3-mutated AML, particularly in combination with intensive or non-intensive chemotherapies in newly diagnosed (ND) FLT3 mutated AML or in R/R situation. Type I inhibitors, including midostaurin, gilteritinib, and crenolanib, have activity against FLT3-ITD and tyrosine kinase domain (TKD) mutations. Type II inhibitors, including sorafenib and quizartinib, do not have activity against FLT3-TKD mutations. Although first-generation FLT3 inhibitors, such as midostaurin 10 and sorafenib 11, have marginal single-agent activity in active disease, they have shown promising activity as maintenance therapies, notably after HSCT for sorafenib 14. Conversely, several FLT3-TKI, such as quizartinib, crenolanib, and gilteritinib, have single-agent activity that lead to complete or near-complete remission, providing a strong rationale to combine these agents with other chemotherapies.

The ADMIRAL phase 3 trial, also designed for R/R FLT3-mutated AML patients, recently demonstrated the superiority of gilteritinib as single agent over the control treatment arm, which was determined by investigators prior to 2:1 randomization between mitoxantrone, etoposide, cytarabine; fludarabine, cytarabine, granulocyte colony-stimulating factor, idarubicin; AZA or LDAC. These regimens are recognized as acceptable salvage strategies in this situation, although other combinations based on intermediate- or high-dose cytarabine or even single-agent cytarabine are also widely used. In the ADMIRAL trial, OS was significantly improved in the gilteritinib arm compared to the control arm with HR at 0.64 (95%CI: 0.49-0.83; p < 0.001). The median OS was 9.3 months in the gilteritinib arm and 5.6 months in the control arm. The CR and CRi rates were 21.1% and 25.5% in the gilteritinib arm vs. 10.5% and 4.8% in the standard arm 2. Gilteritinib was generally well tolerated but was associated with increased incidence of GI side effects, most frequently diarrhea although nausea has been occasionally observed. Increase in bilirubin and transaminase can be seen with gilteritinib but are usually self-resolving and transient. Posterior reversible encephalopathy and pancreatitis are rare (<1-2%) but important side effects to be aware of. These results led to the approval of gilteritinib monotherapy in the US and Europe in patients with R/R FLT3 mutated AML and it is now the gold standard treatment in such situation. However, long lasting remission with gilteritinib as single agent are uncommon since various resistance mechanisms allow AML to escape. Innovative strategies have to be established in order to improve these results. In that setting, associations with ICT, HMA, other targeted therapies are currently developed.

AZA was first investigated as a cytotoxic agent in the 1960s and 1970s in R/R AML and a dose- and time-dependent effect of AZA was evident, with higher remission rates and reduced toxicity reported in patients who received lower doses or continuous infusion schedule of AZA

QUAZAR trial used an oral formulation of AZA (oral-AZA) that is not bioequivalent to injectable AZA, as maintenance therapy in patients with AML who were in CR/CRi after ICT 22. Median OS from randomization was significantly longer with oral-AZA than with placebo (24.7 months vs. 14.8 months, respectively; p<0.001). Median RFS was also significantly longer with oral-AZA than with placebo (10.2 months vs. 4.8 months, respectively; p<0.001). The most common AE in both groups were grade 1 or 2 GI events. Common grade 3 or 4 AE were neutropenia (in 41% of patients in the oral-AZA group and 24% of patients in the placebo group) and thrombocytopenia (in 22% and 21%, respectively). Overall health-related quality of life was preserved during oral-AZA treatment. Among those who were MRD+ at baseline, patients in the oral-AZA arm had a higher rate of conversion to MRD- status vs. placebo: 37% versus 19%, respectively. These data suggest an effective anti-leukemic therapy and not only an ability to maintain a stable situation, among these MRD responders, 9 of 38 (24%) patients in the oral-AZA arm converted to MRD-negativity after > 6 months on treatment, vs. only 1of 22 (5%) patients in the placebo arm. Moreover, during the 63rd ASH annual meeting, Döhner et al. confirmed that Oral-AZA, NPM1 and FLT3 mutational status, cytogenetic risk, as well as MRD status at baseline were independent predictors of survival.

In front line situation, LACEWING study was a phase 3, randomized study comparing gilteritinib+AZA vs. AZA for ND FLT3-mutated AML in patients unfit for ICT. In this trial in patients with ND FLT3 mutated AML unfit for ICT, gilteritinib+AZA led to significantly higher CRc rates but similar OS vs. AZA alone. Based on these results, an independent data monitoring committee recommended the study be terminated for futility, citing that the results are unlikely to demonstrate a statistically significant increase in OS. However subsequent therapy after AZA failure, notably with other regimen including FLT3-TKI, could have precluded transition of better CRc results to better OS results

Inclusion Criteria:

1. Confirmed diagnosis of acute myeloid leukemia (AML) according to world health organization (WHO) 2016 classification
2. Presence of FLT3-mutation(s) at inclusion: in case of FLT3-ITD, the ITD/wt ratio must be > 0.05 ; in case of FLT3-TKD, the mutation must be at D835 or I836 position with a VAF > 5% by NGS.
3. Subjects must be primary refractory or relapsed (R/R) to 1st line intensive chemotherapy (ICT) for AML. Hydroxyurea is allowed for the control of peripheral leukemic blasts in patients with leukocytosis.

3a. Primary refractory is defined as no CR or CRi after at least one course of ICT (including "7+3", gemtuzumab ozogamycin (GO)-based and CPX-351, including or not midostaurine) or two courses (maximum 4) of AZA and venetoclax 3b. Relapse after 1st line ICT for AML is defined as the first hematologic relapse with bone marrow blasts >5% after one line of treatment for AML that includes at least one course of ICT (one line of treatment for AML can include induction, re-induction, consolidation, allogeneic HSCT and maintenance) 3c. Relapse after 1st line non intensive chemotherapy for AML is defined as the first hematologic relapse with bone marrow blasts >5% after or during treatment by AZA venetoclax regardless of number of cycles 4. 1st line intensive treatment may or may not include previous treatment by tyrosine kinase inhibitor (TKI) except gilteritinib.

5. Patients who never received oral azacitidine 6. Age ≥ 18 years 7. Adequate baseline organ function defined by the criteria below:

• adequate renal function as demonstrated by a creatinine clearance ≥ 50 ml/min; calculated by the Cockcroft gault formula or measured by 24-hours urine collection

• aspartate aminotransferase (AST) ≤ 2.5 × ULN

• alanine aminotransferase (ALT) ≤ 2.5× ULN

• bilirubin ≤ 1.5 × ULN

• adequate cardiac function with LVEF ≥45% 8. ECOG < 3 (appendix 1) 9. Absence of any psychological, familial, sociological, or geographical conditions potentially hampering compliance with the study protocol and follow-up schedule 10. Patient is suitable for oral administration of study drug. 11. A female subject is eligible to participate if she is not pregnant and at least one of the following conditions applies: 9a. Not a woman of childbearing potential (WOCBP) as defined in post-menopausal (defined as at least 1 year without any menses) prior to Screening, or documented as surgically sterile (at least 1 month prior to Screening) 9b. WOCBP agrees to follow the contraceptive treatment starting at screening and continue throughout the study period, and for at least 180 days after the final study drug administration 12. Patient must be affiliated to the french social security (health insurance) 13. Signed written informed consent for the study 14. Female subject must agree not to breastfeed starting at screening and throughout the study period, and for 60 days after the final study drug administration.

  15. Female subject must not donate ova starting at screening and throughout the study period, and for 180 days after the final study drug administration.

  16. A male subject with female partner(s) of childbearing potential must agree to use contraception starting at screening and continue throughout the study period, for at least 120 days after the final study drug administration.

  17. Male subject must not donate sperm starting at screening and throughout the study period and for 120 days after the final study drug administration.

Exclusion criteria

1. Subjects with any of the following current or previous diagnoses:

1a. AML secondary to prior myeloproliferative syndrome (MPN)

1b. Acute promyelocytic leukemia (APL) and core binding factor (CBF) AML

1c. DNA fragility or bone marrow (BM) failure syndromes

1d. Blastic plasmacytoid dendritic cell neoplasm

1e. Acute lymphoblastic leukemia including ambiguous lineage 2. Patients ≥ 3rd line of treatment, HSCT being not considered as a line of treatment 3. Patients previously treated by AZA as single agent for AML are not allowed 4. Subjects that have previously been treated by gilteritinib 5. Subjects that have previously been treated by oral azacitidine 6. Clinically active central nervous system (CNS) leukemia 7. Subjects who have received more than 1 prior allogeneic HSCT 8. Subjects who have relapsed within 100 days after allogeneic HSCT 9. Presence of Grade 2 or above graft-versus-host disease (GVHD), including acute, chronic, or overlap; or escalation of therapy for GVHD within 14 days prior to randomization 10. Subject requires treatment with concomitant drugs that are strong inducers of cytochrome P450 (CYP)3A (Annexe 7) 11. Subject requires treatment with concomitant drugs that are strong inhibitors or inducers of P-gp with the exception of drugs that are considered absolutely essential for the care of the subject (Annexe 7) 12. Severe liver disease (e.g. cirrhosis, non-alcoholic steatohepatitis, sclerosing cholangitis or hyperbilirubinemia) 13. Participant has clinically significant abnormality of coagulation profile, such as disseminated intravascular coagulation (DIC).

14. Subject exhibiting evidence of other clinically significant uncontrolled systemic infection requiring therapy (viral, bacterial, or fungal).

15. Isolated extramedullary leukemia relapse 16. History of another malignancy within the past 3 years except basal cell carcinoma of the skin or cervix in situ carcinoma 17. Any other serious medical condition, laboratory abnormalities or psychiatric illness that would place the participant at an unacceptable risk or prevent them from giving informed consent 18. Severe medical or mental condition precluding the administration of protocol treatments 19. persons deprived of their liberty by judicial or administrative decision, persons subject to a legal protection measure (guardianship, curatorship, legal protection), persons under psychiatric care 20. Other comorbidity that the physician judges to be incompatible with conventional intensive chemotherapy which must be reviewed and approved by the study medical monitor before study enrolment 21. Subject with Positive HIV test (due to potential drug-drug interactions). HIV testing will be performed at screening, if required per local guidelines or institutional standards. Subject known to be positive for hepatitis B virus (HBV), or hepatitis C virus (HCV) infection. Inactive hepatitis carrier status with undetectable PCR viral load on antivirals (non-exclusionary medications) are not excluded 22. Known hypersensitivity to the study medication 23. Subject has congestive heart failure classified as New York Heart Association Class III and IV unless a screening echocardiogram performed within 3 months prior to study entry results in a left ventricular ejection fraction that is ≥ 45% 24. Subject with mean Fridericia-corrected QT interval (QTcF) > 450 ms at screening based on central reading 25. Subject with a history of Long QT Syndrome at screening