Blood Cancer Journal www.nature.com/bcj ARTICLE OPEN Survival improvement of patients with FLT3 mutated acute myeloid leukemia: results from a prospective 9 years cohort Guadalupe Oñate 1, Marta Pratcorona1, Ana Garrido1, Alicia Artigas-Baleri1, Alex Bataller2, Mar Tormo 3, Montserrat Arnan4, Susana Vives5, Rosa Coll6, Olga Salamero 7, Ferran Vall-Llovera8, Antònia Sampol 9, Antoni Garcia10, Marta Cervera11, ✉ Sara Garcia Avila12, Joan Bargay13, Xavier Ortín14, Josep F. Nomdedéu 1, Jordi Esteve 2,15, Jorge Sierra 1,15 and Spanish Cooperative Group for the Study and Treatment of Acute Leukemias and Myelodysplasias (CETLAM) © The Author(s) 2023 Midostaurin added to intensive chemotherapy is the standard of care for acute myeloid leukemia (AML) with FLT3 mutations (FLT3mut). We analyzed the impact of midostaurin in 227 FLT3mut-AML patients included in the AML-12 prospective trial for fit patients ≤70 years (#NCT04687098). Patients were divided into an early (2012–2015) and late (2016–2020) cohorts. They were uniformly treated except for the addition of midostaurin in 71% of late group patients. No differences were observed in response rates or the number of allotransplants between groups. Outcome was improved in the late period: 2-year relapse incidence decreased from 42% vs 29% in early vs late group (p = 0.024) and 2-year overall survival (OS) improved from 47% vs 61% 1234567890();,: (p = 0.042), respectively. The effect of midostaurin was evident in NPM1mut patients (n = 151), with 2-yr OS of 72% (exposed) vs 50% (naive) patients (p = 0.011) and mitigated FLT3-ITD allelic ratio prognostic value: 2-yr OS with midostaurin was 85% and 58% in low and high ratio patients (p = 0.049) vs 67% and 39% in naive patients (p = 0.005). In the wild-type NPM1 subset (n = 75), we did not observe significant differences between both study periods. In conclusion, this study highlights the improved outcome of FLT3mut AML fit patients with the incorporation of midostaurin. Blood Cancer Journal (2023)13:69 ; https://doi.org/10.1038/s41408-023-00839-1 INTRODUCTION pro-survival and proliferative signals via the RAS/MAPK, JAK/STAT5 The treatment of patients with acute myeloid leukemia (AML) is and PI3K/AKT pathways [9]. rapidly evolving due to the advances in targeted therapy, risk- The prognosis of patients harboring FLT3-ITD depends on adapted protocols and measurable residual disease (MRD) guided several variables, such as the allelic ratio of the mutation, the decisions [1–4]. The implementation of molecular techniques with presence of determinant co-mutations like NPM1, or the insertion high sensitivity in everyday practice is now essential to accurately site [10–14]. Regarding allelic ratio, several studies, including the classify the different AML entities and to identify potential National Comprehensive Cancer Network (NCCN) guidelines and therapeutic targets in the AML cells. Therefore, a comprehensive the European LeukemiaNet 2017 classification (ELN-17) [15] molecular characterization of the disease is mandatory both at emphasized its relevance in risk assessment. However, there is diagnosis and in the relapse and refractory setting for optimal some controversy regarding the specific FLT3-ITD allelic threshold treatment choice [5, 6]. that can accurately divide high- and low-risk patients as well as FMS-like tyrosine kinase 3 (FLT3) is one of the most frequently concerns regarding the reproducibility of the technique. These are mutated genes in AML with an incidence of around 30% and is some of the reasons which led the experts to remove the generally associated with a negative outcome [7]. The most modulating effect of FLT3-ITD allelic ratio in the latest prognostic frequent mutations of FLT3 are the internal tandem duplication classification of the ELN for AML (ELN-22) [16]. Anyhow, FLT3-ITD (FLT3-ITD) near the juxta-membrane domain and point mutations has in most instances a strong adverse impact and consequently, of the TKD2 domain (FLT3-TKD), with an incidence of 22% and 8% allogeneic hematopoietic cell transplantation (alloHCT) remains respectively [8]. FLT3 mutations (FLT3mut) result in the constitutive the recommended post-remission treatment in fit patients. activation of the FLT3 receptor and the continuous transduction of Nonetheless, the indication of alloHCT in AML with FLT3mut 1 Hospital de la Santa Creu i Sant Pau. Institut d’investigació Biomèdica Sant Pau (IIB SANT PAU) Department of Medicine, Universitat Autonoma of Barcelona, Barcelona, Spain. 2 Hospital Clinic. August Pi i Sunyer Biomedical Research Institute (IDIBAPS), University of Barcelona, Barcelona, Spain. 3Hospital Clinico Universitario, Biomedical Research Institute INCLIVA, Valencia, Spain. 4Institut Catala d’Oncologia, Hospital Duran i Reynals, Institut d’Investigacio Biomèdica de Bellvitge (IDIBELL), L’Hospitalet de Llobregat, University of Barcelona, Barcelona, Spain. 5Institut Catala d’Oncologia, Hospital Germans Trias i Pujol. Josep Carreras Leukemia Research Institute, Badalona, Universitat Autonoma of Barcelona, Barcelona, Spain. 6Institut Català d’Oncologia, Hospital Josep Trueta, Girona, Spain. 7Hospital Universitari Vall d’Hebron and Institute of Oncology (VHIO), Universitat Autonoma of Barcelona, Barcelona, Spain. 8Hospital Universitari Mutua de Terrassa, Barcelona, Spain. 9Hospital Universitari Son Espases, Palma de Mallorca, Spain. 10Hospital Arnau de Vilanova, Lleida, Spain. 11Institut Catala d’Oncologia, Hospital Joan XXIII, Tarragona, Spain. 12Hospital del Mar, Barcelona, Spain. 13Hospital Son Llatzer, Palma de Mallorca, Spain. 14Hospital Verge de la Cinta, Tortosa, Spain. 15These authors contributed equally: Jordi Esteve, Jorge Sierra. * ✉email: jsierra@santpau.cat Received: 10 February 2023 Revised: 5 April 2023 Accepted: 13 April 2023 G. Oñate et al. 2 might be redefined in the future given the potential benefit of the beginning of the protocol. Favorable-risk patients completed two more FLT3 inhibitors. HDAC courses and continued with MRD monitoring, while intermediate or Midostaurin is a first-generation type 2 FLT3 inhibitor, with multi adverse patients were intended for alloHCT in CR1 after one or two HDAC kinase inhibitory effect over various protein kinases such as FLT3, consolidations. MRD was assessed in bone marrow samples after each KIT, or PDGFR. The RATIFY phase III trial demonstrated a significant treatment cycle by either multiparameter flow cytometry or molecular monitoring of NPM1 transcripts (as described by Gorello et al.) and, in a few improvement in overall survival (OS) and event free survival (EFS) cases, of rearrangements RUNX1::RUNX1T1 and CBF::MYH11. with the addition of midostaurin to standard 7 plus 3 chemother- apy in fit FLT3mut AML patients [17]. Based on this finding midostaurin was approved by the US Food and Drug Association Genomic analysis All analysis were performed at diagnosis on bone marrow samples. and the European Medicines Agency in 2017 for all adult patients Molecular testing of NPM1 mutation was studied as previously described with AML with FLT3 mutations. [25]. All patients were tested for FLT3 internal tandem duplication (FLT3- A post-hoc analysis of the RATIFY trial demonstrated the benefit ITD) [26] and classified according to its ITD/wt allelic ratio following of midostaurin in reducing relapse risk in all ELN-17 prognostic previous recommendations (<0.5 low ratio, FLT3low; and ≥0.5 high ratio, categories [18]. However, the study was restricted to patients up FLT3high) [11, 15]. FLT3-TKD mutations were not routinely reported until to 59 years old. To date, only a few prospective phase II trials 2015. Next-generation sequencing with a targeted panel of 42 genes was included patients older than 60 years who were intensively introduced in 2017 as part of the protocol diagnostic work-up allowing the treated and also received midostaurin [19, 20]. These studies detection of additional, less frequent FLT3 mutations (Supplemental Fig. 2). showed the feasibility of this combination in old fit patients as well as an overall outcome improvement compared to historical Statistical analysis cohorts. Additionally, these studies also demonstrated the safety The analysis of the relationship between categorical variables was of midostaurin in the setting of post-alloHCT maintenance. performed using the Chi-square test or the Fisher exact test. Differences In 2016 the Spanish CETLAM group (Grupo cooperativo de between groups for continuous variables were studied by means of the estudio y tratamiento de las leucemias agudas y mielodisplasias) independent-samples t-test or Mann-Whitney U. All tests were two sided and considered significant if p < 0.05. Follow-up duration was calculated incorporated midostaurin to the therapy protocol for fit adults with the inverted Kaplan-Meyer method [27, 28]. OS and EFS were studied with AML and FLT3 mutations. Patients were treated according the with the Kaplan-Meier method, whereas cumulative incidences were CETLAM AML-12 phase II trial (#NCT04687098). This included calculated to estimate relapse risk and non-relapse mortality (NRM) intensive chemotherapy (CT) induction followed by a consolida- considering death in remission and relapse as competitive end-points, tion approach within a risk-adapted post-remission strategy, with respectively. Differences between groups were assessed with the log-rank high-dose cytarabine or alloHCT according to the genetic risk at test (OS, EFS) and the Gray test (CIR, NRM) and were considered significant diagnosis and the persistence of MRD after treatment. This report when p < 0.05. Cox-proportional hazard regression was used for multi- focuses on the outcomes of FLT3mut patients up to 70 years variable analysis. AlloHCT was analyzed as a time-dependent variable. All before and after the introduction of midostaurin. statistical analyses were performed with the SPSS software (Version 26, Two recent studies have analyzed the resulting effect of the IBM, Armonk, NY, USA) and R statistics (Version 4.0.2, R Foundation for Statistical Computing, Vienna, Austria). incorporation of midostaurin [19, 21]. Up to the best of our knowledge, this constitutes the first study to address the issue in a patient population up to 70 treated with an homogenous RESULTS chemotherapy backbone and a subsequent transplant decision Characteristics of patients based on a risk adapted criteria FLT3 mutations were detected in 227 cases (25%) out of 906 patients included in the AML-12 protocol between 2012 and 2020, and constitute the study population. Median follow up is SUBJECTS AND METHODS 42 months (95%CI 35–49). The early cohort included 94 patients Patients and samples whereas the late included 133. Characteristics of these patients, The series included patients with 18–70 years-old diagnosed with de novo considering both time cohorts (2012–2015 and 2016–2020) are AML and eligible for intensive chemotherapy within the CETLAM AML-12 summarized in Table 1. Cytogenetics was available in 93% of phase II trial (#NCT04687098). Patients were treated at 15 academic centers in Spain between January 2012 and December 2020. Since 2016, patients cases. No significant differences were observed between the two with FLT3mut could receive midostaurin added to the therapy as part of study cohorts including the ELN prognostic category, prevalence the AML-12 trial. The trial was approved by the IRB boards at the of NPM1 co-mutation or the proportion of patients harboring a participating institutions and by the health authorities and was conducted FLT3-ITD with a high allelic ITD/wt ratio, with the exception of the in accordance with the Declaration of Helsinki. Informed consent for both virtual absence of TKD sole mutations in the early cohort. Since bone marrow analyses and treatment was obtained in all cases. the analysis of FLT3-TKDs started in 2015, only 2 patients from the In order to analyze the impact of midostaurin on the outcome, we early cohort were identified compared to 25 from the late period considered two study cohorts: “early cohort” (treated between 2012 and cohort. Only one less frequent FLT3 mutation (F594I) was detected 2015) and “late cohort” (2016–2020). All patients received the same CT in a patient with a complex karyotype. protocol except for the addition of midostaurin since 2016 in most FLT3mut patients. The criteria to indicate alloHCT in first CR (CR1) were unchanged throughout the study period. Impact of midostaurin use Full protocol details are available at www.clinicaltrials.gov and are Seventy-one per cent (94/133) of patients from the late cohort detailed in supplemental material. In brief, all patients enrolled in the AML- received midostaurin at some point during frontline treatment 12 trial received induction chemotherapy with idarubicin and cytarabine. (Fig. 1): 71 patients with an ITD (22 FLT3low and 49 FLT3high), 20 The definition of complete remission (CR), overall survival (OS), event-free with TKD, and 3 with other FLT3mut. The onset of the treatment survival (EFS) and cumulative incidence of relapse (CIR) followed the was during induction-1 in most patients (81%), although recommended ELN criteria [22]. Response assessment was performed after treatment start was deferred in the remaining patients due to one cycle: patients with a partial response received a second induction an administrative gap between initial request and real availability cycle; if first complete remission with or without complete hematological recovery (CR or CRi) [23] was achieved, they proceeded with first of the drug during the initial period, which required compassio- consolidation with high dose cytarabine (HDAC) according to the classical nate request on a case-to-case basis. Fourteen patients of this CALGB scheme [24]. Risk stratification was based on ELN recommendations group received midostaurin as maintenance, mainly (n = 12) due (ELN-2010 [23] until the introduction of ELN-2017 [15]) but patients with to a non-transplant allocation given a favorable genetic risk NPM1mut and FLT3-ITD low allelic ratio were considered favorable since according to ELN-2017. Maintenance was administered in one Blood Cancer Journal (2023)13:69 G. Oñate et al. 3 patient with NPM1mut/FLT3high during the posttransplant period, Table 1. Characteristics of patients treated according CETLAM AML-12 whereas one unfavorable patient with NPM1wt/FLT3high was not protocol and with an identified FLT3 mutation. eligible for alloHCT after chemotherapy, and started maintenance CETLAM AML-12 patients < 70 years with FLT3mut (n = 227) following consolidation-1. Response rate after induction was similar in both time cohorts 2012–2015 2016–2020 P n = 94 n = 133 (CR/CRi rate 78% (n = 73) in the early group and 82% (n = 109) in Female gender n (%) 52 (55) 75 (56) 0.9 the late period group; p = ns). Treatment failure was related to refractory disease (12% in the early vs 11% in the late cohort, Median age (range) 54 (21–70) 55 (20–70) 0.65 p=ns) or death in aplasia (11% vs 8%, respectively p=ns). An early <60 years n (%) 65 (69) 87 (65) relapse in the first 3.5 months (median time from CR to transplant) ≥60 years n (%) 29 (31) 46 (35) occurred in 9 patients between 2012 and 2015 and in 5 patients in ECOG 0–1 n (%) 77 (82) 102 (77) 0.3 the latter period 2016–2020 (p = 0.089). AlloHCT was performed in Median WBC ×109/L 53 (1.6–314) 45 (0.42–395) 0.9 a similar proportion of patients in both time cohorts (in 56% (range) (n = 53) and 62% (n = 82) of patients, respectively [p = ns]). For Median BM blasts % 80% (21–100) 80% (21–100) 0.44 patients achieving CR1, the alloHCT rate was 64% (n = 47) in the (range) early group and 68% (n = 74) in the late group. In the overall Cytogenetics 0.9 cohort, MRD persistence following first consolidation was asso- prognostic category& ciated with higher relapse-risk (2 yr 48 ± 15% vs 28 ± 8% n (%) p = 0.021) and a trend towards worse survival (2 yr 57 ± 8% vs Favorable 4 (4) 4 (3) 70 ± 4% p = 0.055, Supplemental Fig. 3). However, we did not Intermediate 82 (88) 111 (83) observe a significant association between MRD negativity and Adverse 4 (4) 5 (4) time period (73% MRD negative in early and late cohort) or No metaphases 4 (4) 13 (10) midostaurin exposure (74% in both midostaurin naive and FLT3 mutations n (%) exposed patients). Early cohort FLT3mut patients presented with a significantly FLT3-ITD 92 (98) 100 (75) 0.54 higher CIR compared to the late cohort (2-year (2 yr) CIR of Low ratio 33 (36) 27 (31) — 42 ± 11% vs 29 ± 10% for early and late group, respectively; High ratio 59 (64) 73 (69) — p = 0.024), without differences in NRM (2 yr NRM of 12% vs 13% FLT3-TKD# 2 (2) 25 (19) — respectively; p = 0.8; Fig. 2). This translated into an improved FLT3-ITD with TKD# 0 7 (5) outcome of late cohort patients, regarding EFS (2-yr 37 ± 5% vs Other — 1 (1) 50 ± 5% for early and late groups; p = 0.021) and OS (2-yr of 47 ± 5% vs 61 ± 5%, in each group, respectively; p = 0.042, Fig. 2). Concomitant 61 (65) 90 (68) 0.7 NPM1mut n (%) To establish the influence of different variables on survival, we ELN-17 prognostic 0.8 performed univariate analysis for all FLT3mut patients, and found categories that leucocyte count (WBC) at diagnosis, treatment period, ELN-17 Favorable 30 (32%) 45 (34%) FLT3 categories, and the presence of NPM1mut significantly impacted on OS (supplemental Table 1). The HR for OS of RUNX1::RUNX1T1 3 3 midostaurin administration was 0.62; 95%CI 0.41–0.94; p = 0.024. CBFB::MYH11 1 1 Thus, midostaurin led to improved survival in the whole cohort by NPM1mut/FLT3low 25 25 decreasing relapse incidence (Fig. 3A) with a 2 yr RR of 40 ± 9% vs NPM1mut/FLT3-TKD 1 16 28 ± 10% (p = 0.034) for naive and exposed patients, and a 2 yr OS Intermediate 38 (40%) 57 (43%) of 49 ± 4% vs 65 ± 5% (p = 0.023) respectively. It is noteworthy NPM1mut/FLT3high 34 47 that patients in the late cohort who did not receive midostaurin NPM1wt/FLT3low 3 5 had very similar survival to those from the early cohort (Fig. 3B). In a multivariate model including age, midostaurin use, WBC at NPM1wt/FLT3-TKD 1 5 diagnosis and ELN-17 prognostic categories, midostaurin main- Adverse 26 (28%) 31 (23%) tained its independent prognostic value both for OS and EFS with NPM1wt/FLT3high 23 24 a HR for OS of 0.55 (95%CI 0.36–0.85; p = 0.007), and for EFS of Other* 3 7 0.51 (0.34–0.76; p = 0.001, Table 2), for patients exposed to the CR rate n (%) 73 (78) 109 (82) 0.7 drug. AlloHCT n (%) 53 (56) 82 (62) 0.6 Impact of midostaurin in AML with NPM1 mutation CR1 39 (41) 64 (48) A total of 151 patients harbored NPM1mut and FLT3mut CR2 6 (6) 9 (7) simultaneously. There was an improved outcome of AML- Active disease 5 (5) 7 (5) NPM1mut patients who received midostaurin (n = 65), with a Unknown status 3 (5) 2 (2) 2-yr EFS of 42 ± 5% vs 60 ± 7% for naive vs exposed patients AlloHCT allogeneic hematopoietic cell transplantation, BM bone marrow, CR (p = 0.002) and 2-yr OS of 50 ± 5% vs 72 ± 6% respectively complete remission, ELN European LeukemiaNet, CR1 first complete (p = 0.011) (Fig. 4A). Moreover, the HR for OS of midostaurin remission, CR2 second complete remission, NA not applicable, OS overall exposure in the univariate analysis was 0.50 (95%CI 0.29–0.86, survival, WBC White blood count &Cytogenetic risk defined according to p = 0.013), a finding that was maintained in the multivariate MRC/NCRI recommendations (Grimwade et al. Blood 2010) #TKD mutations analysis both for OS (HR 0.40 (0.22–0.72; p = 0.002) and EFS (HR were not routinely analyzed until 2015 0.34 (0.20–0.59; p < 0.001, Table 2). *Two cases in the early period presented with t(6;9) while the other had a complex karyotype (CK), in the late period 2 patients had CK, 1 a t(6;9), 2 The allelic ratio of FLT3-ITD retained its prognostic value in both patients harbored mutated TP53, and 2 cases a mutated RUNX1 with an time cohorts (Fig. 4B). Thus, the 2-yr OS of patients treated during absence of NPM1mut 2012 and 2015 was 65 ± 9% for FLT3low (n = 26) vs 32 ± 8% for Blood Cancer Journal (2023)13:69 G. Oñate et al. 4 Fig. 1 CONSORT diagram of patients in both time periods. AlloHCT: allogeneic stem cell transplant, AutoSCT: autologous stem cell transplant, C-1: first consolidation, CR1: first complete remission, CR2: second complete remission, HiDAC: high-dose cytarabine, MRD + : positive measurable residual disease. FLT3high (n = 34; p = 0.005) whereas in the late period the FLT3high patients (n = 47) (p = 0.033). Similar findings were seen adverse prognosis of FLT3-ITD was highly mitigated with a 2 yr OS in terms of EFS (Supplemental Fig. 4) and when midostaurin of 81 ± 9% for FLT3low patients (n = 26) and 57 ± 8% for the exposure was directly contrasted (2-yr OS with midostaurin was Blood Cancer Journal (2023)13:69 G. Oñate et al. 5 Fig. 2 Outcome of CETLAM AML-12 FLT3mut patients (n = 227), according to treatment period. CIR: cumulative incidence of relapse, EFS: event-free survival, NRM: non-relapse mortality, OS: overall survival. Fig. 3 Survival outcome of midostaurin use in AML-12 FLT3mut patients. A OS, CIR and NRM of the overall cohort according to midostaurin exposure. B OS, CIR, and NRM in both time periods among patients who did not receive midostaurin OS: overall survival, CIR: cumulative incidence or relapse, NRM: non-relapse mortality. 85% in low and 58% in high ratio patients (p = 0.049) vs 67% and midostaurin was administered to 83% of patients from the late 39% in naive patients; p = 0.005). On the other hand, NPM1mut/ cohort that presented with non-high-risk disease (NPM1mut with FLT3-TKD patients (n = 18) presented with a 2-yr OS of 72 ± 10%.It FLT3low or TKD) who were never transplanted, and their 2-yr CIR is worth mentioning that NPM1mut/FLT3low patients from the late was 5% in the late group (n = 29) vs 29% in the early group cohort had a strikingly favorable outcome with a 2 yr OS of 81%, (n = 14, p = 0.022) with similar NRM (7%) in both cohorts and more than half of those patients did not receive an alloHCT (Supplemental Fig. 5). (64%, 16 out of 25 patients alive after induction). Among the 9 Seventy-eight percent of NPM1mut/FLT3high patients (63 out of patients that did undergo alloHCT, 3 were performed due to 81) received an alloHCT in the overall cohort, with 84% of them positive MRD, 4 in CR2 status, 1 had active disease, and the disease transplanted in CR1 (Fig. 5). We observed a significantly lower CIR status of the remaining patient was unknown (Fig. 5). Moreover, of NPM1mut/FLT3high patients who received an alloHCT in the Blood Cancer Journal (2023)13:69 G. Oñate et al. 6 Table 2. Multivariate Cox-proportional hazard regression in each FLT3mut subset. OS EFS FLT3mut HR 95% CI p HR 95% CI p WBC 1.002 1–1.005 0.025 1.003 1.001–1.005 0.003 Age <60 years 0.61 0.41–0.91 0.016 0.61 0.42–0.88 0.008 ELN-17 subcategories intermediate risk (vs fav) 2.36 1.38–4.04 0.002 2.21 1.38–3.55 0.001 adverse risk (vs. fav) 3.47 1.99–6.04 <0.001 3.19 1.95–5.23 <0.001 Midostaurin 0.55 0.36–0.85 0.007 0.51 0.34–0.76 0.001 NPM1mut FLT3mut WBC 1.002 0.999–1.005 0.25 1.003 1.000–1.006 0.027 Age<60 years 0.50 0.29–0.85 0.011 0.53 0.33–0.87 0.012 ELN-17 interm vs fav risk 2.39 1.34–4.24 0.003 2.13 1.28–3.53 0.003 Midostaurin 0.40 0.22–0.72 0.002 0.34 0.20–0.59 <0.001 NPM1wt FLT3mut WBC 1.005 1.002–1.009 0.006 1.005 1.002–1.009 0.002 Age < 60 years 0.73 0.38–1.43 0.4 0.64 0.35–1.17 0.15 ELN-17 subcategories intermediate risk (vs fav) 1.93 0.38–9.91 0.43 2.02 0.52–7.89 0.3 adverse risk (vs. fav) 3.65 0.86–15.5 0.08 3.12 0.94–10.4 0.063 Midostaurin 1.07 0.55–2.08 0.9 1.09 0.59–2.03 0.8 CI confidence interval, EFS event free survival, HR Hazard ratio, OS overall survival, WBC leucocyte count Fig. 4 Outcome of AML patients with FLT3 mutations and co-mutated NPM1. A OS and EFS of patients according to midostaurin exposure. B OS of each NPM1mut/FLT3mut subset in the early (left) and the late groups. FLT3-TKD mutations were not available in the early period. Blood Cancer Journal (2023)13:69 G. Oñate et al. 7 Fig. 5 Description of the disease status of NPM1mut/FLT3mut patients in each time period and FLT3 subset. In allogeneic stem cell transplant (AlloHCT) recipients, pre-transplant evaluation is shown, in non-allotransplanted patients the last available evaluation is informed. CR complete remission (first: CR1, second: CR2), MRD + positive measurable residual disease, R/R relapse or refractory disease. late group (71% of which had received midostaurin) in contrast to Supplemental Fig. 10). In the absence of NPM1mut, the small the early group (2-yr 55% vs 30%; p = 0.044) with no differences in subset of non-ITD patients (n = 10: 9 TKDs and 1 other mutation) NRM (p = 0.2; Supplemental Fig. 6). maintained their favorable outcome with a 2-yr OS of 73 ± 16%. Impact of midostaurin in AML with wild-type NPM1 Seventy-five patients presented with FLT3mut without mutation of DISCUSSION NPM1, 33 and 42 from the early and late periods, respectively. The This study confirms the benefit of midostaurin in patients with main clinical characteristics were balanced between cohorts AML and mutations of FLT3 eligible for intensive chemotherapy, a (supplemental Table 2). There were no significant differences in subgroup that includes 30% of adults with this disease and is karyotype distribution which was mostly of intermediate risk associated with adverse outcomes [30–32]. This beneficial effect (76%) according to the Medical Research Council definition [29]. was observed in patients with an age up to 70 years, treated Nine percent of patients presented with CBF rearrangements, and following the same therapeutic protocol with a post-remission FLT3-ITD allelic ratio was high in 78% of ITD patients in both strategy adapted to genetic risk. Moreover, the effect of cohorts. Also, 8 out of the 9 patients who presented FLT3-TKD midostaurin was mostly attributed to a decrease in relapse risk mutations were from the late cohort and none of them were and was confirmed in a multivariate analysis. associated to a favorable karyotype. In our experience, the prognostic improvement was predomi- Fifty-four patients (72%) achieved CR1 following induction and nantly observed in patients with NPM1 co-mutation. Of note, the alloHCT was performed in 58% (19/33) and 74% (31/42) patients outcome of patients both with NPM1mut/FLT3-ITD low ratio with from the early and late cohort respectively (p = 0.3; Supplemental an OS of 81% at two years, and NPM1mut with FLT3-TKD Fig. 7). Twenty-nine out of 42 late cohort patients received mutations (2-year OS 71%) was remarkably improved in the most midostaurin (69%). Although the cohort size was limited, in the recent period 2016–2020, two subgroups without an initial absence of NPM1mut we found no significant survival improve- alloHCT intention in CR1. Overall, our experience is consistent ment with the use of midostaurin with a median EFS of 10 months with the results of the RATIFY trial that provided the most solid for naive and exposed patients (p = 0.9) and a 2-yr OS of 48 ± 7% data on the benefit of adding midostaurin to intensive front-line vs 51 ± 10% in the same groups respectively (p = 0.9, Supple- chemotherapy in FLT3mut AML. This worldwide trial randomized mental Fig. 8), with a HR for OS (midostaurin) of 0.95 (95%CI 717 FLT3mut patients to receive midostaurin or placebo with 0.51–1.79; p = 0.9). A trend towards higher relapse rate was seen induction and consolidation CT and showed an improvement in in the early group (2-yr CIR 54 ± 19% vs 33 ± 17% in the early and survival in the group with FLT3 inhibitor in the whole series as well late groups; p = 0.14, Supplemental Fig. 9), but it did not translate as in all the molecular subgroups. A few studies have validated into a significant survival improvement possibly due to the limited this finding; Larson et al. reported a post-hoc analysis of the cohort size and a higher NRM of late cohort patients in this RATIFY trial emphasizing the benefit of midostaurin in decreasing subgroup (NRM at 2 yr 8% vs 28% in the early and late groups, the cumulative incidence of relapse (HR 0.71 (95% CI, 0.54–0.93); respectively p = 0.10). The global outcome of this molecular p = 0.01). The trial and the analyses included patients up to 59 subgroup of patients remained poor throughout the protocol with years old [18]. The AMLSG 16–10 phase 2 study, showed the a median EFS around 10 months in both groups, and a 2-yr OS of positive impact of midostaurin in 198 younger and 86 older (>60 49 ± 9% vs 48 ± 8% in the early and the late groups (p = 0.9, years) patients treated between 2012 and 2016 and compared the Blood Cancer Journal (2023)13:69 G. Oñate et al. 8 outcomes with historical controls from 5 AMLSG trials between patients had an outstanding OS since the introduction of 1993 and 2008 [19, 20]. Recently, the MD Anderson group midostaurin with a relapse incidence of less than 10% at 2 years. published a series showing the improvement in prognosis of On the other hand, the results observed in the NPM1mut/FLT3-ITD patients with FLT3mut AML, mostly due to the implementation of high ratio who were candidates for alloHCT in CR1 also were targeted therapy with FLT3 inhibitors. In contrast to our study, significantly improved after midostaurin was introduced. In this they included patients mainly treated with sorafenib or quizartinib regard, in this molecular category OS at 2 years was higher than which are not commonly used as front-line therapy in this 50% in the midostaurin era. Further analyses are needed to identify situation, and the report analyzed the addition of the FLT3 agents whether in a subgroup of these patients, allograft could be avoided both in intensive and non-intensive combinations [21]. Differently or delayed to a later moment based on MRD absence and pattern from this, our study included only intensively treated patients and, of co-mutations. On the other hand, additional progress is needed as the AMLSG study, we analyzed patients up to the age of 70 in the NPM1wild/FLT3-ITD group, since midostaurin in our hands did years who were homogeneously treated with intensive CT and not improve their outcome. This particular finding should be transplantation, the latter indicated based on risk assessment. It is weighed carefully due to the limited size number. We analyzed a important to emphasize that in our report 33% of the patients had possible impact of midostaurin exposure in the pretransplant an age between 60 and 70 years. setting in NPM1 mutated vs wild type and found that the median In the AML-12 trial the indication of alloHCT was defined number of days in the first group was 25 vs 18 in the second. This according to risk at diagnosis and MRD evolution; in all difference, however, was not statistically significant. Finally, TKD intermediate and adverse patients the intention was to proceed mutations were only available in the late cohort, and therefore their to alloHCT unless major complications aroused during chemother- outcome could not be comprehensively contrasted. Additional apy and in the favorable genetic category only patients in division of ITD/TKD subsets as well as the impact of other co- remission with persistent or reappearing MRD or those with mutations resulted in too few patient in each subgroup to provide cytological relapse were considered for alloHCT. All these statistical significance in the current analysis and should be interventional strategies did not allow, in our view, a fair explored in larger studies. assessment of the impact of alloHCT in the multivariable analysis. We consider that studies as the one reported here are of value. In fact, when we made an exploratory assessment of the impact of Randomized clinical trials set to register novel agents in AML transplantation in the favorable ELN-17 subgroup (NPM1mut with patients usually present strict inclusion criteria regarding perfor- FLT3low or TKD) and performed a Cox regression with transplan- mance status, organ functions and coexistence of infections at tation as time-dependent covariate, a worse survival in transplant diagnosis, among others. Therefore, it is useful to confirm their recipients was observed (HR for OS (Allo) 4.71 95% CI 1.31–16.95 results in less restrictive trials from academic groups such as ours p = 0.018) that we consider was attributable to the worse that make much fewer limitations to inclusion concerning real-life characteristics of transplanted cases (MRD positive or in relapse). high ECOG scores or active infections and accept adjustments of One of the limitations of our study is that we retrospectively drugs for renal or liver abnormalities that would make the patients compared two different study periods, and the favorable outcome ineligible for trials promoted by corporate sponsors. The of the late cohort might be influenced by other factors such as complementarity of registration trials, academic experiences and changes in alloHCT platforms or advances in support measures. real data gives a more solid picture of the impact of new drugs to However, during the period analyzed, neither chemotherapy improve the prognosis of a rare and difficult disease such as AML. strategies nor transplantation techniques substantially changed in In summary, this study highlights the positive change in our cooperative group and the only major advance was the prognosis of FLT3mut patients due to the association of CT and introduction of midostaurin. Also, the outcomes of patients with targeted therapy with midostaurin, in patients included in a AML that we treated with an identical protocol without FLT3 prospective trial during 9 years comprising patients with an age mutations did not change between 2012 and 2020 (data not up to 70 years and with a risk-adapted pre-defined post-remission shown), and the OS improvement of FLT3mut patients was related alloHCT policy. Further research is needed in patients with AML to a decrease in their relapse risk. These facts reinforce that the and FLT3mut in the absence of NPM1 mutation. It will also be of improved outcome of FLT3mut-AML patients in recent years was interest to know if the results reported here can be further due to the addition of this agent to the chemotherapy courses. improved with other FLT3 inhibitors. In accordance with previous CETLAM publications, we also performed molecular subgroup analyses. We, as the AMLSG, stratified patients according to the FLT3-ITD allelic ratio of 0.5. A DATA AVAILABILITY prior study from our group revealed that this threshold had The datasets generated during and/or analyzed during the current study are available prognostic impact [11] and in a recent analysis we confirmed that from the corresponding author on reasonable request. this allelic ratio remained relevant regardless of the presence of DNMT3A co-mutation [4]. Of note, both studies analyzed FLT3mut patients treated before FLT3 inhibitors were available. Furthermore, REFERENCES a publication by Dohner et al. validating the ELN-17 classification in 1. Fenwarth L, Thomas X, de Botton S, Duployez N, Bourhis JH, Lesieur A, et al. A the RATIFY cohort showed that 0.5 was also the best discriminant personalized approach to guide allogeneic stem cell transplantation in younger value to define patients with different prognoses based on allelic adults with acute myeloid leukemia. Blood. 2021;137:524–32. ratio [12]. This differs from the original RATIFY trial that considered 2. Dillon R, Potter N, Freeman S, Russell N. How we use molecular minimal residual the 0.7 ratio cut-off. However, the limitations in the reproducibility disease (MRD) testing in acute myeloid leukaemia (AML). Br J Haematol. of the technique to establish the ratio and the results from other 2021;193:231–44. investigators have led to the withdrawal of this aspect in the most 3. Bataller A, Onate G, Diaz-Beya M, Guijarro F, Garrido A, Vives S, et al. Acute recent ELN 2022 genetic classification of AML [16]. These ELN-2022 myeloid leukemia with NPM1 mutation and favorable European LeukemiaNet category: outcome after preemptive intervention based on measurable residual guidelines consider all FLT3-ITD patients as intermediate risk disease. Br J Haematol. 2020;191:52–61. regardless of NPM1 co-mutation or ITD allelic ratio and recommend 4. Onate G, Bataller A, Garrido A, Hoyos M, Arnan M, Vives S, et al. Prognostic impact alloHCT in CR1. The results from our study in patients with of DNMT3A mutation in acute myeloid leukemia with mutated NPM1. Blood Adv. NPM1mut/FLT3-ITD low ratio challenges this recommendation; in 2022;6:882–90. our view, the results observed with consolidation CT only support 5. Arber DA, Orazi A, Hasserjian RP, Borowitz MJ, Calvo KR, Kvasnicka HM, et al. continuing our current practice of delaying transplantation after a International consensus classification of myeloid neoplasms and acute leukemias: relapse, provided that there is no persistence of MRD. These integrating morphologic, clinical, and genomic data. Blood. 2022;140:1200–28. Blood Cancer Journal (2023)13:69 G. Oñate et al. 9 6. Khoury JD, Solary E, Abla O, Akkari Y, Alaggio R, Apperley JF, et al. The 5th edition 29. Grimwade D, Hills RK, Moorman AV, Walker H, Chatters S, Goldstone AH, et al. of the World Health organization classification of haematolymphoid tumours: Refinement of cytogenetic classification in acute myeloid leukemia: determina- myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36:1703–19. tion of prognostic significance of rare recurring chromosomal abnormalities 7. Schlenk RF, Dohner K, Krauter J, Frohling S, Corbacioglu A, Bullinger L, et al. among 5876 younger adult patients treated in the United Kingdom Medical Mutations and treatment outcome in cytogenetically normal acute myeloid Research Council trials. Blood. 2010;116:354–65. leukemia. N Engl J Med. 2008;358:1909–18. 30. Kiyoi H, Naoe T, Nakano Y, Yokota S, Minami S, Miyawaki S, et al. Prognostic 8. Papaemmanuil E, Gerstung M, Bullinger L, Gaidzik VI, Paschka P, Roberts ND, et al. implication of FLT3 and N-RAS gene mutations in acute myeloid leukemia. Blood. Genomic classification and prognosis in acute myeloid leukemia. N Engl J Med. 1999;93:3074–80. 2016;374:2209–21. 31. Frohling S, Schlenk RF, Breitruck J, Benner A, Kreitmeier S, Tobis K, et al. Prog- 9. Rosnet O, Buhring HJ, deLapeyriere O, Beslu N, Lavagna C, Marchetto S, et al. nostic significance of activating FLT3 mutations in younger adults (16 to 60 years) Expression and signal transduction of the FLT3 tyrosine kinase receptor. Acta with acute myeloid leukemia and normal cytogenetics: a study of the AML Study Haematol. 1996;95:218–23. Group Ulm. Blood. 2002;100:4372–80. 10. Gale RE, Green C, Allen C, Mead AJ, Burnett AK, Hills RK, et al. The impact of FLT3 32. Port M, Bottcher M, Thol F, Ganser A, Schlenk R, Wasem J, et al. Prognostic internal tandem duplication mutant level, number, size, and interaction with significance of FLT3 internal tandem duplication, nucleophosmin 1, and CEBPA NPM1 mutations in a large cohort of young adult patients with acute myeloid gene mutations for acute myeloid leukemia patients with normal karyotype and leukemia. Blood. 2008;111:2776–84. younger than 60 years: a systematic review and meta-analysis. Ann Hematol. 11. Pratcorona M, Brunet S, Nomdedeu J, Ribera JM, Tormo M, Duarte R, et al. 2014;93:1279–86. Favorable outcome of patients with acute myeloid leukemia harboring a low- allelic burden FLT3-ITD mutation and concomitant NPM1 mutation: relevance to post-remission therapy. Blood. 2013;121:2734–8. ACKNOWLEDGEMENTS 12. Döhner K, Thiede C, Jahn N, Panina E, Gambietz A, Larson RA, et al. Impact of This work was supported in part by the Institut d’Investigació Biomèdica Sant Pau (IIB NPM1/FLT3-ITD genotypes defined by the 2017 European LeukemiaNet in SANT PAU), the CERCA Programme/Generalitat de Catalunya (SGR-Cat 2021) and the patients with acute myeloid leukemia. Blood. 2020;135:371–80. Instituto de Salud Carlos III, Ministerio de Economía y Competitividad, Spain (PI17/ 13. Rucker FG, Du L, Luck TJ, Benner A, Krzykalla J, Gathmann I, et al. Molecular 01246, PI20/01621 and CM20/00061). landscape and prognostic impact of FLT3-ITD insertion site in acute myeloid leukemia: RATIFY study results. Leukemia. 2022;36:90–99. 14. Schlenk RF, Kayser S, Bullinger L, Kobbe G, Casper J, Ringhoffer M, et al. Differ- AUTHOR CONTRIBUTIONS ential impact of allelic ratio and insertion site in FLT3-ITD-positive AML with GO and MP designed research, GO analyzed data and wrote the paper. JS and JE respect to allogeneic transplantation. Blood. 2014;124:3441–9. supervised research and wrote the paper; JFN and MP coordinated genomic analysis, 15. Dohner H, Estey E, Grimwade D, Amadori S, Appelbaum FR, Buchner T, et al. and AA created the molecular supplemental tables and graphs, AG and AB performed Diagnosis and management of AML in adults: 2017 ELN recommendations from data extraction from the cooperative data base, MA, SV, RC, MT, AS, MC, AG, FVLL, OS, an international expert panel. Blood. 2017;129:424–47. SGA, XO, and JB provided the clinical data and all authors reviewed the final 16. Dohner H, Wei AH, Appelbaum FR, Craddock C, DiNardo CD, Dombret H, et al. manuscript. Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood. 2022;140:1345–77. 17. Stone RM, Mandrekar SJ, Sanford BL, Laumann K, Geyer S, Bloomfield CD, et al. CONFLICT OF INTEREST Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 muta- JS reports personal fees from AbbVie, Vyxeos, Gilead, CSL Behring, Astellas, and tion. N Engl J Med. 2017;377:454–64. Gilead; grants and personal fees from Novartis and Daiichi-Sankyo; and grants from 18. Larson RA, Mandrekar SJ, Huebner LJ, Sanford BL, Laumann K, Geyer S, et al. Amgen and Jazz Pharmaceuticals. JE reports consultancy-honoraria from Abbvie, Midostaurin reduces relapse in FLT3-mutant acute myeloid leukemia: the alliance Novartis, Astellas, Jazz Pharmaceuticals, BMS-Celgene, Pfizer, Daichii-Sankyo and CALGB 10603/RATIFY trial. Leukemia. 2021;35:2539–51. research grants from Novartis and Jazz Pharmaceuticals. The remaining authors 19. Dohner H, Weber D, Krzykalla J, Fiedler W, Wulf G, Salih H, et al. Midostaurin plus declare no competing financial interests. intensive chemotherapy for younger and older patients with AML and FLT3 internal tandem duplications. Blood Adv. 2022;6:5345–55. 20. Schlenk RF, Weber D, Fiedler W, Salih HR, Wulf G, Salwender H, et al. Midostaurin ADDITIONAL INFORMATION added to chemotherapy and continued single-agent maintenance therapy in Supplementary information The online version contains supplementary material acute myeloid leukemia with FLT3-ITD. Blood. 2019;133:840–51. available at https://doi.org/10.1038/s41408-023-00839-1. 21. Reville PK, Sasaki K, Kantarjian HM, Daver NG, Yilmaz M, Dinardo CD, et al. Improved outcomes among newly diagnosed patients with FMS-like tyrosine Correspondence and requests for materials should be addressed to Jorge Sierra. kinase 3 internal tandem duplication mutated acute myeloid leukemia treated with contemporary therapy: revisiting the European LeukemiaNet adverse risk Reprints and permission information is available at http://www.nature.com/ classification. Am J Hematol. 2022;97:329–37. reprints 22. Dohner H, Estey EH, Amadori S, Appelbaum FR, Buchner T, Burnett AK, et al. Diagnosis and management of acute myeloid leukemia in adults: recommen- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims dations from an international expert panel, on behalf of the European Leuke- in published maps and institutional affiliations. miaNet. Blood. 2010;115:453–74. 23. Cheson BD, Bennett JM, Kopecky KJ, Buchner T, Willman CL, Estey EH, et al. Revised recommendations of the International Working Group for Diagnosis, Standardi- zation of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol. 2003;21:4642–9. Open Access This article is licensed under a Creative Commons 24. Mayer RJ, Davis RB, Schiffer CA, Berg DT, Powell BL, Schulman P, et al. Intensive Attribution 4.0 International License, which permits use, sharing, postremission chemotherapy in adults with acute myeloid leukemia. Cancer and adaptation, distribution and reproduction in any medium or format, as long as you give Leukemia Group B. N. Engl J Med. 1994;331:896–903. appropriate credit to the original author(s) and the source, provide a link to the Creative 25. Boissel N, Renneville A, Biggio V, Philippe N, Thomas X, Cayuela JM, et al. Pre- Commons license, and indicate if changes were made. The images or other third party valence, clinical profile, and prognosis of NPM mutations in AML with normal material in this article are included in the article’s Creative Commons license, unless karyotype. Blood. 2005;106:3618–20. indicated otherwise in a credit line to the material. If material is not included in the 26. Thiede C, Steudel C, Mohr B, Schaich M, Schakel U, Platzbecker U, et al. Analysis of article’s Creative Commons license and your intended use is not permitted by statutory FLT3-activating mutations in 979 patients with acute myelogenous leukemia: regulation or exceeds the permitted use, you will need to obtain permission directly association with FAB subtypes and identification of subgroups with poor prog- from the copyright holder. To view a copy of this license, visit http:// nosis. Blood. 2002;99:4326–35. creativecommons.org/licenses/by/4.0/. 27. Shuster JJ. Median follow-up in clinical trials. J Clin Oncol. 1991;9:191–2. 28. Xue X, Agalliu I, Kim MY, Wang T, Lin J, Ghavamian R, et al. New methods for estimating follow-up rates in cohort studies. BMC Med Res Methodol. 2017;17:155. © The Author(s) 2023 Blood Cancer Journal (2023)13:69