Impact of hematopoietic cell transplantation and quizartinib in newly diagnosed patients with acute myeloid leukemia and FMS-like tyrosine kinase 3-internal tandem duplications in the QuANTUM-First trial

Abstract

QuANTUM-First (ClinicalTrials.gov identifier: NCT02668653) was a randomized phase III trial in patients with newly diagnosed FLT3-internal tandem duplication (ITD)-positive acute myeloid leukemia (AML) treated with quizartinib or placebo plus standard induction and consolidation chemotherapy and/or allogeneic hematopoietic cell transplantation (allo-HCT), followed by single-agent maintenance therapy. We evaluated the impact of allo-HCT performed in first complete remission (CR1) or composite CR1 (CRc1) on overall survival (OS), considering treatment randomization. Post-hoc extended Cox regression multivariable analyses were conducted in patients who achieved complete remission/composite complete remission by the end of induction, including allo-HCT in CR1/CRc1 as a time-dependent variable to identify prognostic and predictive factors for OS. There were 297 patients with complete remission by the end of induction (quizartinib, N=147; placebo, N=150); of these, 157 (52.9%) underwent allo-HCT in CR1 (quizartinib, N=84; placebo, N=73). There were 368 patients with composite complete remission by the end of induction (quizartinib, N=192; placebo, N=176); of these, 196 (53.3%) underwent allo-HCT in CRc1 (quizartinib, N=110; placebo, N=86). Multivariable analyses revealed quizartinib treatment and allo-HCT in either CR1 (hazard ratio [HR]=0.553, 95% confidence interval [95% CI]: 0.383-0.798, P=0.0015 and HR=0.527, 95% CI: 0.349-0.796, P=0.0023, respectively) or CRc1 (HR=0.645, 95% CI: 0.470‒0.886, P=0.0068 and HR=0.557, 95% CI: 0.391-0.793, P=0.0012, respectively) as significant predictive factors for a longer OS. No new safety signals were identified. Patients who underwent protocol-specified allo-HCT in CR1/CRc1 experienced post-transplant-related complications, mostly grade ≥2 graft-versus-host disease, as expected. This post-hoc analysis further supports the use of quizartinib and allo-HCT in CR1/CRc1 as an efficacious and well-tolerated treatment strategy for newly diagnosed FLT3-ITD-positive AML patients fit for intensive chemotherapy.

Introduction

FMS-like tyrosine kinase 3-internal tandem duplication (FLT3-ITD) is among the most common genetic alterations in newly diagnosed acute myeloid leukemia (AML), with an incidence of ≈25%; it is associated with poor prognosis, high relapse rate, and inferior overall survival (OS).¹,² Standard chemotherapy with FLT3 inhibitors followed by allogeneic hematopoietic cell transplantation (allo-HCT) is the mainstay treatment for patients with newly diagnosed FLT3-ITD-positive AML who are fit for intensive chemotherapy.^3,4

According to the most recent National Comprehensive Cancer Network (NCCN)⁴ and European LeukemiaNet (ELN)³ AML guidelines, all patients with FLT3-ITD AML are categorized as intermediate risk, regardless of FLT3-ITD allelic ratio (FLT3-ITD/FLT3 wild-type) or nucleophosmin 1 (NPM1) mutation status. Allo-HCT in first complete remission (CR1) improves OS and relapse-free survival (RFS) in patients with AML.^5,6 However, even with allo-HCT, the risk of relapse remains high in patients with newly diagnosed FLT3-ITD-positive AML (30%-59%), particularly in comparison to that of patients without FLT3-ITD (16%-19%).^7,8 Therefore, inhibiting the tyrosine kinase signaling pathway has been a successful new therapeutic strategy for these patients.^9-13 Furthermore, in patients with newly diagnosed FLT3-mutated AML, treatment with midostaurin in combination with standard chemotherapy, including the option to undergo allo-HCT in CR1 or complete remission (CR) with incomplete neutrophil or platelet recovery (CRi), improved survival substantially, relative to placebo. Of note, the benefit provided by midostaurin over placebo was more pronounced in patients who underwent allo-HCT than in those who did not.^12,14

Quizartinib is a highly potent, selective, second-generation, type II FLT3 inhibitor for once-daily, oral treatment.^15-18 Based on data from the QuANTUM-First trial (ClinicalTrials. gov identifier: NCT02668653),¹⁹ quizartinib has recently been approved in combination with chemotherapy across induction, consolidation, and as maintenance monotherapy by the United States Food and Drug Administration,²⁰,²¹ the Japanese health agency,²² the European Medicines Agency,^23,24 and by the United Kingdom health agency,^25,26 for the treatment of adult patients with newly diagnosed FLT3-ITD-positive AML (but not after allo-HCT in the United States). In QuANTUM-First, the addition of quizartinib to standard induction and consolidation chemotherapy, including allo-HCT, followed by up to 36 cycles (≈3 years) of quizartinib maintenance monotherapy resulted in a statistically significant 16.8 months OS extension compared with standard therapy (placebo group) in patients with newly diagnosed FLT3-ITD-positive AML, with a significant reduction in the risk of death (hazard ratio [HR]=0.78, P=0.032), and a manageable safety profile.¹⁹ A clinically meaningful reduction in cumulative incidence of relapse (CIR), an increased duration of CR, and a reduction in measurable residual disease contributed to the OS benefit. The objectives of this QuANTUM-First post-hoc analysis were to assess the impact of allo-HCT in CR1 and in composite CR1 (CRc1), with composite CR (CRc) including both CR as well as CRi and the interplay with quizartinib treatment on clinical outcomes in patients with newly diagnosed FLT3-ITD-positive AML.

Methods

A detailed description of the QuANTUM-First trial was previously published.¹⁹ Per protocol, patients were permitted to undergo allo-HCT after CR or CRi had been achieved (protocol-specified allo-HCT).¹⁹ Allo-HCT for consolidation could be performed after induction, any time during consolidation, or within the first 3 months of maintenance.¹⁹ After engraftment, patients who received protocol-specified allo-HCT were eligible to enter maintenance and received 36 cycles (28 days per cycle) of quizartinib 60 mg daily or placebo, according to their initial study randomization. Any allo-HCT performed for other reasons (e.g., molecular relapse or without CR or CRi response) was considered as non-protocol-specified AML therapy.¹⁹ Patients receiving non-protocol-specified AML therapy were discontinued from the allocated treatment and followed up for clinical outcome data. Any allo-HCT performed after treatment discontinuation was non-protocol-specified.¹⁹ The trial was conducted in compliance with the Declaration of Helsinki and Good Clinical Practice guidelines outlined by the International Council for Harmonisation. Institutional review boards or independent ethics committees approved the protocol at each participating site. Patients provided written informed consent before enrollment.

Statistical analyses of efficacy

OS was analyzed in the intent-to-treat analysis set, comprising all randomized patients. The median duration of follow-up was calculated by the reverse Kaplan-Meier estimate.²⁷ The effect of allo-HCT on OS as a time-dependent intervening event was tested by using the Mantel-Byar method²⁸ for univariable and extended Cox regression (Andersen-Gill model)²⁹ for multivariable analyses. The Mantel-Byar univariable OS analysis was performed in patients who achieved CR/CRc by the end of induction comparing OS for patients who underwent allo-HCT with OS for patients who did not undergo allo-HCT in each treatment arm. The multivariable extended Cox regression OS analysis was performed in patients who achieved CR/CRc by the end of induction, stratified by region, age, and white blood cell count, including allo-HCT in CR1/CRc1 as a time-dependent variable and adjusted for FLT3-ITD variant allele frequency (FLT3-ITD/total FLT3) and sex. The method of Simon and Makuch³⁰ was used to assess the time-dependent effect of allo-HCT in CR1/CRc1 on OS according to initial randomization, in patients who achieved CR/CRc by the end of induction. The Simon and Makuch plot considers the timing of allo-HCT occurrence; therefore, once a patient undergoes allo-HCT, the patient switches from the “without allo-HCT” category to the “with allo-HCT” category and is censored at that time in the “without allo-HCT” curve. P values were not adjusted for multiplicity. The analyses presented are post-hoc analyses. RFS and CIR were analyzed in patients who achieved CR/ CRc by the end of induction based on independent review committee assessment. RFS was defined as the time from randomization until documented relapse or death from any cause, whichever occurred first. The medians of OS and RFS were estimated based on the Kaplan-Meier method, the two-sided 95% confidence interval (95% CI) using the method of Brookmeyer and Crowley, and the HR (with 95% CI) using an unstratified Cox proportional hazards model.

Graft- versus-host disease analyses

The incidence and severity (including grading and staging)³¹,³² of acute or chronic graft-versus-host disease (GvHD) were assessed in the safety analysis population (i.e., patients who received at least one dose of quizartinib or placebo and underwent protocol-specified allo-HCT) every 4 weeks after allo-HCT and at the following post-transplant timepoints: day 100, 6 months (±3 months), 12 months (±3 months), 18 months (±3 months), and 24 months (-3/+6 months). Details on patients, randomization, and safety analyses are provided in the Online Supplementary Methods.

Results

Patients’ disposition

In QuANTUM-First, 539 patients were randomized to receive either quizartinib (N=268) or placebo (N=271) (Figure 1).¹⁹ At the time the trial was conducted, the risk stratification outlined in the NCCN and ELN guidelines³³ of that time indicated that patients with a high FLT3-ITD allelic ratio and NPM1 wild type had an adverse risk, those with a low FLT3-ITD allelic ratio and NPM1 wild type had an intermediate risk, and those with a low FLT3-ITD allelic ratio and NPM1 mutation had a favorable risk. Allo-HCT was not recommended by these guidelines for patients with low FLT3-ITD allelic ratio and NPM1 mutation.

Overall, 272 (50.4%) patients of the 539 in the intent-to-treat population underwent allo-HCT (both protocol-specified and non-protocol-specified; 53.7% [144/268] of patients in the quizartinib arm and 47.2% [128/271] in the placebo arm) (Table 1). There were 83 patients who underwent non-protocol-specified allo-HCT (44 in the quizartinib arm and 39 in the placebo arm), but four of these patients (2 in each treatment arm) also underwent protocol-specified allo-HCT (Table 1). Of the intent-to-treat population, 64.6% of patients in either the quizartinib arm (173/268) or the placebo arm (175/271) entered consolidation (Figure 1). Of these, allo-HCT in CR1/CRc1 was performed in 31.3% (84/268)/41.1% (110/268) of patients in the quizartinib arm versus 26.9% (73/271)/31.7% (86/271) in the placebo arm (Figure 1, Table 1). In addition, there were 115 allo-HCT performed outside CR1 (22.4% [60/268] of patients in the quizartinib arm and 20.3% [55/271] in the placebo arm) (Table 1). Of the 115 allo-HCT conducted outside CR1, 93 (80.9%) occurred in patients who did not achieve CR1 (52/268 [19.4%] in the quizartinib arm and 41/271 [15.1%] in the placebo arm) and 22 (19.1%) occurred after relapse (8/268 [3.0%] in the quizartinib arm and 14/271 [5.2%] in the placebo arm) (Table 1). Of 93 patients who underwent allo-HCT not in CR1, 39 had allo-HCT in CRc1 (26/268 [9.7%] in the quizartinib arm and 13/271 [4.8%] in the placebo arm) (Table 1). Of the intent-to-treat population, 43.3% (116/268) of patients in the quizartinib arm versus 33.9% (92/271) in the placebo arm entered maintenance (Figure 1). Of these, allo-HCT in CR/CRc1 was performed in 22.8% (61/268)/28.4% (76/268) of patients in the quizartinib arm versus 13.3% (36/271)/16.6% (45/271) in the placebo arm (Figure 1). The disposition of patients who achieved CR/CRc is presented in Online Supplementary Figure S1.

Conditioning regimen and graft characteristics in patients undergoing allogeneic transplantation in first remission

The conditioning regimens and graft characteristics were well balanced between treatment arms in the two cohorts (Table 1). Among patients who underwent allo-HCT in CR1/CRc1 with quizartinib (N=84/N=110), those who did so with placebo (N=73/N=86), 50.0%/50.9% in the quizartinib arm received a myeloablative conditioning regimen versus 45.2%/47.7% in the placebo arm. Patients predominantly received grafts from unrelated donors (47.6%/49.1% in the quizartinib arm vs. 52.1%/50.0% in the placebo arm, respectively). The majority of patients received matched grafts (72.6%/70.9% in the quizartinib arm vs. 69.9%/67.4% in the placebo arm, respectively). Peripheral blood was the major source of stem cells for 78.6%/75.5% in the quizartinib arm versus 84.9%/83.7% in the placebo arm, respectively. In the quizartinib arm, allo-HCT was performed in CR1/ CRc1 after a median time of 3.5 months; in the placebo arm, allo-HCT was performed in CR1 after a median of 3.3 months and in CRc1 after a median of 3.2 months (Table 1). For GvHD prophylaxis, patients received mainly calcineurin inhibitors (cyclosporine and tacrolimus), methotrexate, and mycophenolate mofetil.

Baseline patients’ demographics and disease characteristics

Patients’ demographics and disease characteristics were well balanced across the five cohorts, including patients who achieved CR (N=297), patients who underwent allo-HCT in CR1 (N=157), patients who achieved CR and did not undergo allo-HCT in CR1 (N=140), patients who achieved CRc (N=368), and patients who underwent allo-HCT in CRc1 (N=196) (Table 2). The median age of patients achieving CR/ CRc was 56 years/55 years, similar to that of the overall population of QuANTUM-First of 56 years.¹⁹ Patients who underwent allo-HCT in CR1/CRc1 were younger (median age of 51 years) with lower rates of Eastern Cooperative Oncology Group performance status (ECOG PS) of 2 (13.4%/11.2%, respectively) than those who did not undergo allo-HCT in CR1 (median age of 61.5 years; ECOG PS of 2 in 18.6%) (Table 2). In the overall population, 52.3% had a concomitant NPM1 mutation, while patients who achieved CR/CRc and those who underwent allo-HCT in CR1/CRc1 had a slightly higher rate of concomitant NPM1 mutation (65.7%, 63.9%, 64.3%, and 62.8%, respectively). Most patients across the five cohorts (roughly 52.4%-55.6%) had FLT3-ITD variant allele frequency values ranging from >25% to ≤50%, which was similar to that of the overall population of QuANTUM-First (52.1%) (Table 2). As expected, patients who did not undergo allo-HCT in CR1 had the highest rates of concomitant NPM1 mutation (67.1%) as well as the highest rates of FLT3-ITD variant allele frequency values, ranging from >25% to ≤50% (57.9%) (Table 2). The patients’ demographics and disease characteristics in these five cohorts, separated by treatment arms, are provided in Online Supplementary Table S1.

Figure 1.Diagram of the patients flow in the study. ^aIncludes protocol-specified allogeneic hematopoietic cell transplantation (allo-HCT). ^bIncludes protocol-specified allo-HCT and non-protocol-specified allo-HCT. FLT3-ITD: FMS-like tyrosine kinase 3-internal tandem duplication; ITT: intent-to-treat; HiDAC: high-dose cytarabine; CR1: first complete remission; CRc1: first composite complete remission.

Table 1.Allogeneic hematopoietic cell transplantation timing, conditioning regimens, and graft characteristics.

Efficacy analysis

A univariable Mantel-Byar OS analysis found, among patients who achieved CR by the end of induction, a statistically significant OS benefit with allo-HCT in CR1 in the quizartinib arm (P=0.0248) and in the placebo arm (P=0.0046). Similarly, among patients who achieved CR by the end of induction, an OS benefit was found with allo-HCT in CRc1 in the quizartinib arm (P=0.0234) and in the placebo arm (P=0.0008).

According to a multivariable extended Cox regression analysis conducted in patients who achieved CR by the end of induction, both quizartinib treatment (HR=0.553, 95% CI: 0.383-0.798, P=0.0015) (Figure 2A) and allo-HCT in CR1 (HR=0.527, 95% CI: 0.349-0.796, P=0.0023) (Figure 2A) were found to be predictive factors for a better OS. Similarly, in patients who achieved CRc by the end of induction, both quizartinib treatment (HR=0.645, 95% CI: 0.470-0.886, P=0.0068) (Figure 2B) and allo-HCT in CRc1 (HR=0.557, 95% CI: 0.391-0.793, P=0.0012) (Figure 2B) were predictive factors for longer OS. As expected, a FLT3-ITD variant allele frequency >50% was an unfavorable prognostic factor for OS (Figure 2B). Neither ECOG PS nor cytogenic risk category was a predictive factor for longer OS in patients who achieved CR/CRc by the end of induction (Figure 2A, B). Based on this model, quizartinib-treated patients who achieved CR/CRc by the end of induction and proceeded to allo-HCT in CR1/CRc1 had an OS advantage (HR=0.292, 95% CI: 0.168-0.505 for CR1; HR=0.359, 95% CI: 0.225-0.573 for CRc1) at any given time, compared with placebo-treated patients who achieved CR/CRc by the end of induction who had not yet received allo-HCT in CR1/CRc1 by that time. According to a Kaplan-Meier OS analysis in patients who achieved CR/CRc by the end of induction, longer OS was observed in patients treated with quizartinib versus placebo, irrespective of whether they received allo-HCT in CR1/CRc1 or not (Figure 3). Among patients who underwent allo-HCT in CR1, the HR was 0.591 (95% CI: 0.330-1.059) (Figure 3A). Among patients who did not undergo allo-HCT in CR1, the HR was 0.579 (95% CI: 0.353-0.949) (Figure 3B) when OS was censored at the date of starting the conditioning regimen for allo-HCT. Similar findings were obtained in the analysis by allo-HCT in CRc1 (Figure 3C, D). According to a Simon and Makuch OS analysis by allo-HCT, patients who achieved CR by the end of induction in the quizartinib arm had longer OS compared with those in the placebo arm, regardless of whether they received allo-HCT in CR1 or not (Figure 4A). Similar findings were obtained in the analysis by allo-HCT in CRc1 (Figure 4B). The time-dependent Simon and Makuch analysis of the effect of allo-HCT in CR1/CRc1 on OS was consistent with the earlier analysis conducted with the Kaplan-Meier method (Figure 3).

Similar to the OS analysis, the RFS analysis in patients who achieved CR/CRc by the end of induction and underwent allo-HCT showed a longer RFS in patients treated with quizartinib versus placebo, irrespective of whether they received allo-HCT in CR1/CRc1 or not (Online Supplementary Figure S2). Among patients who underwent allo-HCT in CR1, the HR was 0.607 (95% CI: 0.351-1.052) (Online Supplementary Figure S2A). Among patients who did not undergo allo-HCT in CR1, the HR was 0.682 (95% CI: 0.458-1.016) (Online Supplementary Figure S2B). Similar findings were obtained in the RFS analysis by allo-HCT in CRc1 (Online Supplementary Figure S2C, D). Consistently, the CIR rates were numerically lower in the quizartinib arm than in the placebo arm, irrespectively of whether the patients received allo-HCT in CR1/CRc1 or not (Online Supplementary Figure S3).

Table 2.Baseline demographics and disease characteristics of specific cohorts of patients.

Safety analysis

Of the 102 patients in the quizartinib arm and 91 in the placebo arm who underwent protocol-specified allo-HCT in CR1/CRc1, 57 (55.9%)/41 (45.1%), respectively, had post-transplant–related complications (Table 3). Most complications were grade ≥2 GvHD, which affected 34 (33.3%) patients in the quizartinib arm and 22 (24.2%) patients in the placebo arm (Table 3). Among patients who underwent protocol-specified allo-HCT, acute GvHD was reported in 46 (45.1%) patients in the quizartinib arm and in 35 (38.5%) patients in the placebo arm, and the percentage of patients with grade 3/4 acute GvHD was higher in the quizartinib arm (17 [16.7%]) than in the placebo arm (6 [6.6%]) (Table 3). Chronic GvHD was reported in 30 (29.4%) patients in the quizartinib arm and 18 (19.8%) in the placebo arm (Table 3). Of the 85 patients with grade 2-4 acute GvHD and chronic GvHD, the percentage of patients who discontinued in the quizartinib arm (31/51 [60.8%]) was lower than the percentage of those who discontinued placebo (24/34 [70.6%]) (Table 3). Online Supplementary Table S2 describes the maximum score for each organ. For patients who underwent allo-HCT, clinically relevant medical conditions that had their onset and resolution during the allo-HCT period, as well as other medical conditions that started during the allo-HCT period and were still ongoing on day 1 of the phase, were collected. The most common medical conditions in patients who underwent protocol-specified allo-HCT were gastrointestinal disorders (stomatitis, diarrhea, nausea, and vomiting), infections (cytomegalovirus infection and pneumonia), and immune system disorders (GvHD) (Online Supplementary Table S3). Similar proportions of patients in the quizartinib versus placebo groups had at least one adverse event (100%, each) and one grade 3 or worse adverse event (96.1% vs. 94.5%), with serious adverse events and adverse events associated with dose modifications being more frequent in the quizartinib group than in the placebo group (Online Supplementary Table S4). Among patients who underwent protocol-specified allo-HCT, the most common grade 3/4 adverse events (occurring in ≥10% of patients) were febrile neutropenia, neutropenia, hypokalemia, pneumonia, and anemia in both groups; neutrophil count decreased and gamma-glutamyl transferase concentration increased more in the quizartinib group; and thrombocytopenia occurred more in the placebo group (Table 4).

Figure 2.Extended Cox regression analysis of overall survival, stratified by region, age, and white blood cell count. (A, B) Multivariable extended Cox regression post-hoc analysis was conducted in patients who achieved complete remission by the end of induction (A) and in patients who achieved composite complete remission by the end of induction (B), including allogeneic hematopoietic cell transplantation in first complete remission (A) and in first composite complete remission (B) as time-dependent variables and adjusted for FLT3-ITD variant allele frequency and sex. OS: overall survival; CR: complete remission; allo-HCT: allogeneic hematopoietic cell transplantation; CR1: first complete remission; HR: hazard ratio; CI: confidence interval; FLT3-ITD: FMS-like tyrosine kinase 3‒internal tandem duplication; VAF: variant allele frequency; ECOG: Eastern Cooperative Oncology Group; CRc: composite complete remission; CRc1: first composite complete remission.

Figure 3.Kaplan-Meier plot of overall survival by treatment arm in patients who achieved complete remission (CR)/composite CR by the end of induction per independent review committee, by allogeneic hematopoietic cell transplantation in first CR/first composite CR. Post-hoc analysis. (A) The group of patients who underwent allogeneic hematopoietic cell transplantation (allo-HCT)^a in first complete remission (CR1). (B) The group who did not undergo allo-HCT^a in CR1.^b (C) The group who underwent allo-HCT in first composite complete remission (CRc1). (D) The group who did not undergo allo-HCT in CRc1.^{b a}Includes protocol-specified and non-protocol-specified allo-HCT. Some patients had both protocol-specified and non-protocol-specified allo-HCT. ^bOverall survival was censored at the starting date of the conditioning regimen for allo-HCT. OS: overall survival; HR: hazard ratio; 95% CI: 95% confidence interval; NR: not reached; NE: not evaluable; CRc1: first composite complete remission.

Figure 4.Time-dependent Simon and Makuch plot of overall survival from initial randomization by allogeneic hematopoietic cell transplantation in first complete remission (CR)/first composite CR in patients who achieved CR/composite CR by the end of induction per independent review committee. Post-hoc analysis. (A) The group who achieved CR by the end of induction. (B) The group who achieved composite CR by the end of induction. ^a“W/o allo-HCT in CR1” refers to patients who achieved CR without allogeneic hematopoietic cell transplantation (allo-HCT) in the study or patients who achieved CR with allo-HCT outside first CR. ^b“W/o allo-HCT in CRc1” refers to patients who achieved composite CR without allo-HCT in the study or patients who achieved composite CR with allo-HCT outside of first composite CR. OS: overall survival; CR1: first complete remission; CRc1: first composite complete remission; CRc: composite complete remission.

Among 193 patients who underwent allo-HCT, 119 (quizartinib, N=70; placebo, N=49) received maintenance therapy. The exposure during maintenance therapy in these 119 patients was relatively similar between the two treatment arms, with a median adjusted treatment duration of 1.6 years in the quizartinib arm versus 1.7 years in the placebo arm, and with 68.6% of patients in the quizartinib arm versus 73.5% in the placebo arm receiving ≥12 cycles (Online Supplementary Table S5). Patients who underwent allo-HCT and received quizartinib maintenance had higher rates of grade ≥3 treatment-emergent adverse events and rates of adverse events associated with dose modifications than those who received maintenance with placebo (Online Supplementary Table S6).

Table 3.Graft-versus-host disease during the allogeneic hematopoietic cell transplantation period in patients who underwent protocol-specified allogeneic hematopoietic cell transplantation (safety analysis set).^a

Table 4.Adverse events of grade 3/4 occurring in ≥5% of the patients who underwent protocol-specified allogeneic hematopoietic cell transplantation, in either treatment arm (safety analysis set).^a

As expected, based on quizartinib’s safety profile,¹⁹ rates of blood cytopenias (including neutropenia/neutrophil count decrease, anemia, and thrombocytopenia/platelet count decrease), gastrointestinal disorders (nausea, diarrhea, and vomiting), and infections (pneumonia and herpes zoster) were higher in the quizartinib arm than in the placebo arm (Online Supplementary Table S7).

Discussion

These analyses of the QuANTUM-First trial¹⁹ were focused on assessing the impact of allo-HCT in CR1/CRc1 and the interplay with quizartinib treatment compared with placebo on OS in patients with newly diagnosed FLT3-ITD-positive AML. The demographics and disease characteristics in the cohorts of patients who achieved CR/CRc and those who underwent allo-HCT in CR1/CRc1 were well balanced between treatment arms and generally reflective of the patient population with newly diagnosed FLT3-ITD-positive AML with a high burden of aggressive disease.

The multivariable extended Cox regression analysis, including allo-HCT as a time-dependent covariable, demonstrated that among patients who achieved CR/CRc, quizartinib treatment and allo-HCT in CR1/CRc1 were associated with longer OS, with estimated reductions in the risk of death of 44.7% and 47.3%, respectively, for patients who achieved CR, and 35.5% and 44.3%, respectively, for patients who achieved CRc. One of the limitations of the Kaplan-Meier plot is that it does not consider the timing of allo-HCT occurrence, as patients’ categorization has to be defined since randomization; therefore, patients who have undergone allo-HCT during the course of the study are included in the “with allo-HCT” category since the beginning. A more analytically rigorous approach of the impact of allo-HCT on OS is provided by the Simon and Makuch method.³⁰ Therefore, we have illustrated the survival benefit provided by treatment with quizartinib, including the interplay with allo-HCT as a time-dependent intervention in patients with CR/CRc, in a Simon and Makuch plot, confirming improved OS in patients proceeding to allo-HCT in CR1/CRc1 and in patients receiving standard consolidation chemotherapy in the quizartinib arm over the placebo arm. A pre-planned sensitivity analysis of OS of the QuANTUM-First trial that censored at the starting date of the conditioning regimen for allo-HCT revealed a strong trend favoring quizartinib over placebo (HR=0.75), consistent with the primary analysis of OS (HR=0.78).¹⁹ RFS and CIR curves revealed that quizartinib provided higher RFS rates and lower CIR rates versus placebo, regardless of whether patients underwent allo-HCT or not, which could be attributed to higher rates of not detectable measurable residual disease in quizartinib-treated patients than in placebo-treated patients, as previously reported.¹⁹ In the QuANTUM-First trial, patients who underwent protocol-specified allo-HCT experienced post-transplant -related complications, mostly grade ≥2 GvHD, as expected. Rates of acute and chronic post-transplant GvHD in this study were generally in line with those reported for allo-HCT in patients with FLT3-mutated AML.³⁴ More patients treated with quizartinib were able to undergo allo-HCT, resulting in more reporting of grade 3/4 acute GvHD in the quizartinib arm. GvHD events were reported as resolved or improved in the majority of patients. No new safety signals were identified in patients with newly diagnosed FLT3-ITD-positive AML who underwent allo-HCT in CRc1, and the safety profile was in agreement with QuANTUM-First safety data overall.¹⁹

Other studies demonstrated the value of allo-HCT as a valuable consolidation treatment that improves survival outcomes for patients with FLT3-ITD-positive AML, particularly when performed in CR1.^6,12,35-38 Among the patients who underwent allo-HCT in CR1 in the RATIFY trial, a remarkable improvement in OS was seen in those receiving induction and consolidation therapy with midostaurin and chemotherapy compared with those receiving placebo and chemotherapy.¹² When the OS analysis was performed starting from the time of allo-HCT, patients in the midostaurin arm had a 24% risk reduction of death (compared with 22% risk reduction, based on the primary OS analysis), although allo-HCT was not mandatory and patients were not randomized to undergo allo-HCT.¹²

The risk of relapse is still higher in patients with newly diagnosed FLT3-ITD-positive AML than in patients without FLT3-ITD, even after undergoing allo-HCT.⁷ Therefore, there is still an unmet need to decrease relapse rates after consolidation. The FLT3 inhibitor gilteritinib was investigated in patients with FLT3-ITD AML as maintenance therapy after allo-HCT in the phase III MORPHO trial (NCT02997202).³⁹ The MORPHO trial did not meet its primary RFS endpoint nor its key secondary OS endpoint; hence, it did not provide a definitive answer on the role of maintenance therapy after HCT with a FLT3 inhibitor.³⁹ Post-hoc analyses from QuANTUM-First suggest that quizartinib would allow more patients to receive maintenance therapy after HCT than would placebo, but due to statistical limitations with small sample sizes and small numbers of OS events, more research is warranted.⁴⁰ Two phase III trials of FLT3 inhibitors in patients with newly diagnosed FLT3-mutated AML are ongoing, but their results are not yet available (ARO-021 [NCT03258931]⁴¹ and HOVON 156 AML [NCT04027309]).⁴²

These unplanned post-hoc analyses have several limitations. The fact that allo-HCT was performed per investigator’s decision adds some variability to the study. In addition, only slightly more than 50% of patients who achieved CR/CRc received allo-HCT. As a result, some analyses included a small number of patients, which could limit the generalization of the results. The P values were not adjusted for multiplicity; therefore, results should be interpreted appropriately.

These post-hoc analyses of the QuANTUM-First trial showed that quizartinib treatment and allo-HCT in CR1/CRc1 were favorable factors for longer OS for patients in the study. Patients who achieved remission on quizartinib had longer OS than patients on placebo, irrespective of whether they underwent allo-HCT in first remission. Rates of post-transplant GvHD seemed to be in line with those reported for allo-HCT in previous studies. No new safety signals were identified in patients who underwent allo-HCT. Therefore, taken together, these data further support quizartinib as an efficacious and well-tolerated treatment option for patients aged 18-75 years with newly diagnosed FLT3-ITD-positive AML.

Footnotes

• Received September 10, 2024
• Accepted March 4, 2025

Correspondence

°Current affiliation: King’s College London, London, UK

Richard F. Schlenk richard.schlenk@nct-heidelberg.de

Disclosures

RFS reports consulting fees from Daiichi Sankyo, AbbVie, Jazz, and Pfizer; payment for lectures/speakers’ bureaus from Daiichi Sankyo, Novartis, and Pfizer; participation on an advisory board/DSMC for BerGenBio and Novartis; and having received equipment from AbbVie, AstraZeneca, Boehringer Ingelheim, Daiichi Sankyo, PharmaMar, Pfizer, Recordati Rare Diseases, and Roche. PM reports research grants from AbbVie, Bristol Myers Squibb, Jazz, Menarini, Stemline, Novartis, Pfizer, and Takeda; consulting fees from AbbVie, Astellas, BeiGene, Bristol Myers Squibb, Gilead, Incyte, Jazz, Kura Oncology, Menarini, Stemline, Nerviano, Novartis, Otsuka, Pfizer, Ryvu, and Takeda; and payment for speakers’ bureaus from AbbVie, Astellas, Bristol Myers Squibb, Gilead, Jazz, and Pfizer. H-JK reports research grants from BL&H; consulting fees from AbbVie, AIMS BioScience, Amgen, AMLHub, Astellas, Aston BioSciences, Bristol Myers Squibb, Celgene, Boryung, Daiichi Sankyo, GC Biopharma, Handok, Ingenium, Janssen, LG Chem, Meiji Seika, Novartis, Pfizer, Sanofi Genzyme, SL VaxiGen, Takeda, and VigenCell; payment for lectures/speakers’ bureaus from AbbVie, AMLHub, Astellas, Bristol Myers Squibb, Handok, Jazz, Novartis, and Takeda; travel support from AbbVie, the APLC, Astellas, Jazz, and Takeda; being on a data safety monitoring board/advisory board for AbbVie, AMLHub, the APBMT, the APLC, Astellas, Bristol Myers Squibb, Celgene, Daiichi Sankyo, Handok, Janssen, Novartis, Pfizer, and Sanofi Genzyme; and being a leader in other boards, societies, committees, or advocacy groups for AMLHub, the APLC, the APBMT, Bristol Myers Squibb, the KSABMT, and Novartis. RV reports consulting fees from AbbVie, Astellas, and Pfizer; and payment for lectures/speakers’ bureaus from AbbVie, Astellas, Merck Sharp & Dohme, Novartis, Pfizer, PharmaS, and Servier. EP reports consulting fees from KCR US; payment for lectures/speakers’ bureaus from Amgen, Angelini, Astellas, Novartis, and Servier; and travel support from Angelini, Astellas, Bristol Myers Squibb, Jazz, Novartis, Pfizer, and Servier. JC reports research grants from AbbVie, Daiichi Sankyo, Novartis, Sun, and Pfizer; consulting fees from AbbVie, Bio-Path, Daiichi Sankyo, Gilead, Forma, Novartis, Pfizer, and Takeda; payment for lectures/ speakers’ bureaus from Novartis, Pfizer, and Takeda; and having stock options with Bio-Path. MAS reports payment for participation on an advisory board/DSMC from Bristol Myers Squibb, Kurome, and Schrödinger; and having stock options with Kurome. HD reports research grants from Amgen, Astellas, Bristol Myers Squibb, Celgene, Incyte, Jazz, Pfizer, and Servier; and personal fees from Daiichi Sankyo, Incyte, Jazz, and Servier. JW reports payment for participation on an advisory board from AbbVie and for participation on a DSMC from AstraZeneca. AEP reports research grants from AbbVie, Astellas, Daiichi Sankyo, FujiFilm, Bayer, and Syndax; consulting fees from AbbVie, Astellas, Daiichi Sankyo, Forma, and Foghorn; payment for lectures from Astellas and Daiichi Sankyo; travel support from Daiichi Sankyo; payment for participation on DSMC from the Leukemia & Lymphoma Society and Foghorn; and honoraria for participation on advisory boards from AbbVie, Actinium, Aptose, Astellas, BerGenBio, Bristol Myers Squibb, Curis, Daiichi Sankyo, Genentech, Immunogen, Rigel, and Schrödinger. MJL reports research grants from Astellas and FujiFilm; consulting fees from AbbVie, Amgen, Bristol Myers Squibb, Daiichi Sankyo, Jazz, Menarini, Pfizer, and Takeda; and payment for lectures/speakers’ bureaus from Astellas. HPE reports research grants from AbbVie, Agios, ALX Oncology, Amgen, Aptose, Ascentage, Celgene, Daiichi Sankyo, Forma, Forty Seven, Gilead, GlycoMimetics, ImmunoGen, Jazz, Kura Oncology, MacroGenics, Novartis, Oryzon Genomics, PTC, Servier, Sumitomo Dainippon, and Taiho Oncology; consulting fees from AbbVie, Agios, Astellas, Bristol Myers Squibb, Celgene, Daiichi Sankyo, Genentech, GlycoMimetics, ImmunoGen, Incyte, Jazz, Kura Oncology, MacroGenics, Novartis, Pfizer, Schrödinger, Servier, Stemline, Sumitomo Dainippon, Syros, Takeda, and Trillium; payment for speakers’ bureaus from AbbVie, Agios, Bristol Myers Squibb, Celgene, Incyte, Jazz, Novartis, and Servier; and having served as a steering committee member for GlycoMimetics, as chair of the Myeloid Neoplasms Repository study for Bristol Myers Squibb and Celgene, and as chair of the independent review committee of the VIALE A and VIALE C studies for AbbVie. LL and KI are employees of Daiichi Sankyo. YMK is a former employee of Daiichi Sankyo; reports grants/contracts from Bexon Clinical Consulting and Stemline; and has received consulting fees from Stemline. AL and JH are former employees of Daiichi Sankyo. The other authors have no conflicts of interest to disclose.

Contributions

Daiichi Sankyo and the study steering committee (including RFS, PM, EP, JC, MAS, HD, SA, JW, AEP, MJL, and HPE) designed the study. All authors contributed to the critical review and revision of the article and approved the decision to submit it for publication. Data were collected by the investigators and monitored by Daiichi Sankyo. All authors had full access to all the data in the study. LL and RFS directly accessed and verified the data. RFS, AEP, MJL, and HPE wrote the first draft of this article with professional medical writing and editing support. Daiichi Sankyo and all authors were responsible for data analysis and interpretation.

Funding

This study was sponsored and supported by funding from Daiichi Sankyo, Inc. Medical writing support was funded by Daiichi Sankyo, Inc.

Acknowledgments

We would like to thank the patients, their families, and caregivers for their participation in the QuANTUM-First study. We would further like to thank the QuANTUM-First steering committee members, the investigators, the study staff, and independent review committee and data monitoring committee members for their important contributions. Medical writing support was provided by Mohamed Abdelmegeed, MD, PhD, CMPP, and Francesca Balordi, PhD, CMPP, of The Lockwood Group (Stamford, CT, USA), in accordance with Good Publication Practice (GPP 2022) guidelines.