Gilteritinib as Post-Transplant Maintenance for AML With Internal Tandem Duplication Mutation of FLT3

Abstract

PURPOSE

Allogeneic hematopoietic cell transplantation (HCT) improves outcomes for patients with AML harboring an internal tandem duplication mutation of FLT3 (FLT3-ITD) AML. These patients are routinely treated with a FLT3 inhibitor after HCT, but there is limited evidence to support this. Accordingly, we conducted a randomized trial of post-HCT maintenance with the FLT3 inhibitor gilteritinib (ClinicalTrials.gov identifier: NCT02997202) to determine if all such patients benefit or if detection of measurable residual disease (MRD) could identify those who might benefit.

METHODS

Adults with FLT3-ITD AML in first remission underwent HCT and were randomly assigned to placebo or 120 mg once daily gilteritinib for 24 months after HCT. The primary end point was relapse-free survival (RFS). Secondary end points included overall survival (OS) and the effect of MRD pre- and post-HCT on RFS and OS.

RESULTS

Three hundred fifty-six participants were randomly assigned post-HCT to receive gilteritinib or placebo. Although RFS was higher in the gilteritinib arm, the difference was not statistically significant (hazard ratio [HR], 0.679 [95% CI, 0.459 to 1.005]; two-sided P = .0518). However, 50.5% of participants had MRD detectable pre- or post-HCT, and, in a prespecified subgroup analysis, gilteritinib was beneficial in this population (HR, 0.515 [95% CI, 0.316 to 0.838]; P = .0065). Those without detectable MRD showed no benefit (HR, 1.213 [95% CI, 0.616 to 2.387]; P = .575).

CONCLUSION

Although the overall improvement in RFS was not statistically significant, RFS was higher for participants with detectable FLT3-ITD MRD pre- or post-HCT who received gilteritinib treatment. To our knowledge, these data are among the first to support the effectiveness of MRD-based post-HCT therapy.

INTRODUCTION

AML is stratified into different molecular subtypes to guide therapy.¹ Internal tandem duplication mutations of FLT3 (FLT3-ITD) are common in AML and confer an increased relapse risk.² Allogeneic hematopoietic stem cell transplantation (HCT) in first remission is considered the standard of care for these patients when feasible.^1,3

CONTEXT

• Key Objective

• To determine if all patients with internal tandem duplication mutation of FLT3 (FLT3-ITD) AML undergoing allogeneic hematopoietic stem-cell transplantation (HCT) benefit from post-HCT maintenance with the FLT3 inhibitor gilteritinib or if benefit is restricted to those patients who have FLT3-ITD measurable residual disease (MRD) at the time of HCT.

• Knowledge Generated

• Patients with AML with FLT3-ITD MRD detectable in the peri-HCT period benefit from post-HCT gilteritinib, whereas those without detectable MRD do not. These prospective results establish FLT3-ITD mutations as essential markers of MRD and illustrate how molecular MRD can be used to guide the therapy of patients with AML undergoing HCT.

• Relevance (C.F. Craddock)

• Post-transplant gilteritinib maintenance represents a significant therapeutic advance in patients allografted for FLT3-ITD AML who have evidence of peri-transplant MRD. MRD-negative patients derive no benefit from gilteritinib maintenance but instead may be exposed to unnecessary toxicity.*

  *Relevance section written by JCO Associate Editor Charles F. Craddock, MD.

Guidelines from the National Comprehensive Cancer Network recommend post-HCT maintenance with FLT3 inhibitors to reduce the risk of relapse⁴ on the basis of results from small randomized trials of sorafenib and midostaurin.^5-8 However, this practice is controversial⁹ as patients in these trials were not treated with FLT3 inhibitors pre-HCT (the current standard practice) and two of the trials^6,8 were nonblinded and allowed only myeloablative conditioning (MAC). Treatment with FLT3 inhibitors can be toxic and often needs to be interrupted or halted because of adverse events (AEs).^8,10-13 For patients treated with current induction standards for FLT3-ITD AML undergoing HCT in first remission, the question remains if the benefits of maintenance with FLT3 inhibition outweigh the risks of toxicity. Despite the risk of post-HCT relapse, at least half of patients with FLT3-ITD AML transplanted in first remission are cured without further treatment,⁴ which means that many patients treated with post-HCT FLT3 inhibition are subjected to an unnecessary therapy.

The presence of measurable residual disease (MRD) pre- or post-HCT is highly predictive of outcomes.^14-17 Because of their apparent instability during the course of the disease, FLT3-ITD mutations have not historically been regarded as useful markers of MRD, but recent data suggest otherwise.^18-20 Highly sensitive assays using sequential polymerase chain reaction (PCR) and next-generation sequencing (NGS) detect low levels of FLT3-ITD mutations in patients in remission, and retrospective studies suggest that the presence of these mutations correlates with relapse.^18,19,21,22

Gilteritinib is a potent, well-tolerated oral FLT3 inhibitor approved as monotherapy for relapsed or refractory FLT3-mutated AML.²³ The randomized, double-blinded, placebo-controlled Blood and Marrow Transplant Clinical Trials Network (BMT CTN) 1506 (MORPHO) trial was designed to determine (1) if post-HCT maintenance with gilteritinib provided benefit for patients with FLT3-ITD AML in first remission undergoing HCT and (2) if FLT3-ITD MRD detection could be used to identify the patients who benefit.

METHODS

Patients

Eligible patients were adults with FLT3-ITD AML (diagnosed with local mutation testing) who were in continuous first remission achieved with not more than two cycles of intensive therapy (with or without a FLT3 inhibitor and including any investigational regimens) and intended to undergo allogeneic HCT after induction and any consolidation within 1 year of achieving remission. Any donor source, conditioning regimen, and graft-versus-host disease (GVHD) prophylaxis were permitted.

Trial Design and Treatment

Participants were registered before HCT, and a bone marrow (BM) aspirate was obtained to confirm remission and for MRD analysis. Once engrafted (defined by absolute neutrophil count ≥500/mm³, platelet count ≥20,000/mm³, and platelet transfusion–independent) and provided that they were free of grade II-IV GVHD (and requiring not more than 0.5 mg/kg prednisone per day), participants were randomly assigned between days 30 and 90 after HCT to placebo or 120 mg per day gilteritinib for 24 months. Immediately before random assignment, a second BM aspirate was obtained to confirm ongoing remission and for MRD analysis. Random assignment was double-blinded at a ratio of 1:1 between the treatment arm and the placebo arm using permuted blocks of random sizes, stratified by conditioning regimen intensity (myeloablative v reduced intensity/nonmyeloablative), time from transplantation to random assignment (30-60 v 61-90 days), and the presence of FLT3-ITD MRD at a level of 1 × 10⁻⁴ or greater (present v absent/indeterminate) on the basis of the pre-HCT BM aspirate.

MRD Assay

The first 2 mL of any study marrow aspirate was reserved for MRD analysis. For the MRD assay,²¹ 700 ng of genomic DNA was amplified by 25 cycles of PCR using primers flanking exons 14 and 15 of FLT3 and the amplicons were analyzed by NGS. The limit of blank (LOB) was two variant reads, and the lower limit of detection was estimated to be the FLT3-ITD variant allele frequency of 5 × 10⁻⁵. However, any level of FLT3-ITD mutation (minimum of three variant reads) above the LOB (quantified as low as 1 × 10⁻⁶), irrespective of whether it was the same mutation reported at diagnosis, was considered detectable MRD. The pre-HCT level used for stratification was 1 × 10⁻⁴ or higher. Investigators were blinded to the results of MRD analyses.

End Points and Assessments

The primary end point was relapse-free survival (RFS) as assessed by a blinded end point review committee (BERC), measured from the time of random assignment to either morphological relapse or death, using the intention-to-treat (ITT) population. Morphological relapse was defined as BM blasts 5% or higher, any circulating blasts, or any extramedullary blast foci as per published criteria.²⁴ Overall survival (OS) was a key secondary objective. Other secondary objectives included nonrelapse mortality (NRM) and examining the effect of MRD on RFS and OS in the gilteritinib and placebo arms and the effect of gilteritinib versus placebo separately in patients with and without MRD. Additional details on end points and assessments are provided in the Data Supplement (Appendix, online only).

Trial Conduct and Oversight

This trial was conducted in accordance with the Declaration of Helsinki. Institutional review boards at each site approved the trial protocol, and all investigators obtained informed consent from each participant or each participant's guardian. The trial was funded by grant Nos. U10HL069294 and U24HL138660 to the BMT CTN from the National Heart Lung and Blood Institute (NHLBI) and the National Cancer Institute and by Astellas Pharma Global Development, Inc. The trial was designed by the BMT CTN and approved by the NHLBI and Astellas. The Emmes Company monitored North American sites, and Parexel monitored non–North American sites. All investigators and the industry sponsor were blinded to outcomes. Data collection and monitoring procedures are provided in the Data Supplement (Appendix). The investigators had full access to the data at study closure. The study cochairs (M.J.L. and Y.-B. C.) reviewed the data and wrote the manuscript with editorial input from coauthors and without assistance from nonauthors.

Statistical Analysis

The sample size was based on estimates of RFS in the control group of 67% at 1 year, 59% at 2 years, and 55% at 3 years derived from Center for International Blood and Marrow Transplant Research data on participants with FLT3-ITD mutation transplanted in first remission. A total of 122 events would provide 85% power to detect a hazard ratio (HR) of 0.57 (corresponding to a 15% difference in 2-year RFS) with a two-sided significance level of 0.05. The analysis was scheduled for when 122 events were observed or 2.5 years after the last patient was randomly assigned, whichever came first. The primary end point of RFS was summarized using Kaplan-Meier curves and compared between arms using stratified log-rank tests, with the random assignment factors used as stratification variables. A stratified Cox proportional hazards model was used to provide HR estimates and CIs. To maintain the overall two-sided type I error rate at 0.05, formal significance testing of OS using a gatekeeping approach was to be conducted if the RFS comparison was statistically significant. Otherwise, OS analysis would be considered exploratory. OS was analyzed in the ITT population in the same manner as RFS. Competing risk end points (relapse, NRM, acute GVHD [aGVHD], chronic GVHD [cGVHD], eradication or detection of MRD) were summarized using the cumulative incidence function and compared between arms using Gray's test, with subdistribution HRs obtained using the Fine-Gray model. Prespecified subgroup analyses of MRD status were conducted using interaction testing between treatment and subgroup, and forest plots of the treatment effect within subgroups were drawn. No formal multiplicity adjustment for secondary end points or subgroup analyses was used.

RESULTS

Participants

Between August 17, 2017, and July 8, 2020, 620 patients at 122 centers in 16 countries were screened for eligibility, 488 participants were registered, and 356 were randomly assigned, 178 in each arm (Fig 1). The last participant finished treatment in July 2022. The primary analysis is based on a data cutoff on January 7, 2023 (2.5 years after the last participant was randomly assigned). Of 488 participants registered, 132 (27%) participants were not randomly assigned for the following reasons: 68 (51.5%) failed to meet random assignment criteria (including GVHD and failure to engraft); 26 (19.7%) for patient/physician decision; 16 (12.1%) for early death; 10 (7.6%) for relapse; and 12 (9.1%) for other reasons. The safety analysis set (SAF) comprised 355 participants (178 in the gilteritinib arm and 177 in the placebo arm) who took at least one dose of study drug (one participant randomly assigned to placebo received gilteritinib, and one participant randomly assigned to gilteritinib did not take study drug). The most common reasons for early discontinuation were an AE in the gilteritinib arm (17.4%) and relapse (23%) in the placebo arm (Fig 1).

FIG 1.

Screening, registration, random assignment, and reasons for discontinuing study treatment. GVHD, graft-versus-host disease; HCT, hematopoietic cell transplantation.

Participant characteristics are displayed in Table 1. There were more than 30 unique conditioning regimens used worldwide. NPM1 mutations were reported in 34.6% of participants. Information on other comutations or FLT3-ITD allelic ratio was not available, and so classification according to the European LeukemiaNet 2022 system was not possible.¹ Marrow aspirates for MRD analysis were available from 350 of 356 (98%) participants pre-HCT and 347 of 356 (97.5%) post-HCT (before random assignment). MRD was detected at the stratification level (1 × 10⁻⁴ or higher) in 75 of 356 (21.1%) participants and at a level of 1 × 10⁻⁶ or higher in 164 of 356 (46.1%) pre-HCT. Post-HCT, MRD was detected at a level of 1 × 10⁻⁶ or higher in 71 of 356 (19.9%), including 16 (4.5%) participants with detectable MRD post-HCT but not pre-HCT. Therefore, a total of 180 ([164 + 16 of 356]; 50.6%) participants had detectable MRD in the peri-HCT period.

TABLE 1.

Participant Characteristics at Baseline (ITT population)

Parameter	Gilteritinib (n = 178)	Placebo (n = 178)
Age, years, median (range)	53 (20-78)	53 (18-76)
Sex, No. (%)
Male	91 (51.1)	92 (51.7)
Female	87 (48.9)	86 (48.3)
Race, No. (%)
White	114 (64)	106 (59.6)
African American	6 (3.4)	3 (1.7)
Asian	47 (26.4)	56 (31.5)
Other/missing	11 (6.2)	13 (7.3)
Geographic, No. (%)
North America	77 (43.3)	77 (43.3)
Europe	49 (27.5)	43 (24.2)
Asia/Pacific	52 (29.2)	58 (32.6)
Genetic results at AML diagnosis, No. (%)
Favorable karyotype	9 (5.1)	4 (2.2)
Intermediate karyotype	119 (66.9)	90 (50.6)
Adverse karyotype	7 (3.9)	7 (3.9)
Unknown	29 (16.3)	51 (28.7)
Other	14 (7.9)	26 (14.6)
FLT3 inhibitor pre-HCT, No. (%)	110 (61.8)	103 (57.9)
HCT-specific comorbidity index, No. (%)
0	79 (44.4)	70 (39.3)
1-2	49 (27.5)	51 (28.7)
3+	49 (27.5)	57 (32)
Conditioning regimen intensity, No. (%)
MAC	106 (59.6)	107 (60.1)
RIC/nonmyeloablative	72 (40.4)	71 (39.9)
Stem-cell donor, No. (%)
Matched sibling	55 (30.9)	48 (27)
Haploidentical	22 (12.4)	38 (21.3)
Matched unrelated	71 (39.9)	65 (36.5)
Mismatched unrelated	15 (8.4)	17 (9.6)
Cord blood	11 (6.2)	8 (4.5)
Stem-cell source, No. (%)
Peripheral blood	140 (78.7)	140 (78.7)
Marrow	27 (15.2)	30 (16.9)
Cord blood	11 (6.2)	8 (4.5)
GVHD prophylaxis, No. (%)
Calcineurin inhibitor + methotrexate	98 (55.1)	96 (53.9)
Calcineurin inhibitor + mycophenolate mofetil	43 (24.2)	51 (28.7)
Other	37 (20.8)	30 (16.9)
Missing	0 (0)	1 (0.6)
Time from HCT to random assignment, No. (%)
30-60 days	95 (53.4)	97 (54.5)
61-90 days	83 (46.6)	81 (45.5)
MRD, No. (%)
Pre-HCT MRD ≥ 10−4	39 (21.9)	36 (20.2)
Pre-HCT MRD ≥ 10−6	82 (46.1)	82 (46.1)
Pre- or post-HCT MRD ≥ 10−6	89 (50)	91 (51.1)

Abbreviations: GVHD, graft-versus-host disease; HCT, hematopoietic cell transplantation; ITT, intention-to-treat; MAC, myeloablative conditioning; MRD, measurable residual disease; RIC, reduced-intensity conditioning.

Efficacy

Among the 270 participants who survived at data cutoff, the median follow-up was 43.8 months. A total of 103 RFS events (by BERC) were observed in the primary analysis, which led to an approximate reduction in power to 78.6% instead of 85.0%. Longer follow-up would not have increased the number of events measurably because of very low event rates beyond 2 years post-HCT. While there was improved RFS in the gilteritinib arm compared with that in the placebo arm (Fig 2A), the difference did not meet the predetermined threshold for significance (HR, 0.679 [95% CI, 0.459 to 1.005]; two-sided P = .0518). The 2-year RFS rate by BERC (95% CI) was 77.2% (CI, 70.1 to 82.8) for participants receiving gilteritinib and 69.9% (CI, 62.4 to 76.2) for those receiving placebo. OS (Fig 2B) was analyzed by ITT in the primary analysis (which included a total of 86 deaths) and did not show a statistically significant difference (HR, 0.846 in favor of gilteritinib [95% CI, 0.554 to 1.293]; two-sided P = .4394). The incidence of relapse was lower and NRM was higher in the gilteritinib arm compared with the placebo arm (Figs 2C and 2D). Of 47 participants who relapsed in the placebo arm, 20 (42.6%) were treated with a FLT3 inhibitor (gilteritinib-13, quizartinib-4, sorafenib-3) after relapse. The cumulative incidence of relapse by geographic region is displayed in the Data Supplement (Fig 1).

FIG 2.

Survival, relapse, and nonrelapse mortality (ITT population). (A) Relapse-free survival, (B) overall survival for the gilteritinib and placebo groups, (C) cumulative incidence of relapse for the gilteritinib group versus placebo group, and (D) cumulative incidence of nonrelapse mortality (defined as death without documentation of morphological relapse). HR, hazard ratio; ITT, intention-to-treat; OS, overall survival; RFS, relapse-free survival.

MRD at a level of 1 × 10⁻⁶ or greater was associated with decreased RFS and OS (Figs 3A and 3B) irrespective of the treatment arm. Subgroup analysis of RFS and OS performed on MRD and other prespecified subgroups is displayed in the Data Supplement (Figs 2 and 3). Participants with detectable MRD pre- or post-HCT had a significantly improved RFS if they were on gilteritinib compared with the placebo arm, whereas MRD-negative participants in both arms had similar RFS (Figs 3C and 3D). This was the case for participants with detectable MRD pre-HCT (P = .0105), post-HCT (P = .0143), or either pre- or post-HCT (P = .0065). Similarly, participants with pre- or post-HCT MRD at a level of 1 × 10⁻⁶ or greater had improved OS when treated with gilteritinib (Data Supplement, Fig 4) although this did not reach statistical significance (P = .0731).

FIG 3.

The impact of measurable residual disease on relapse-free survival (ITT population). (A) Relapse-free survival, (B) overall survival for all participants irrespective of the treatment arm according to whether any (eg, FLT3-ITD variant allele frequency of 1 × 10⁻⁶ or above) MRD was detectable peri-HCT, (C) relapse-free survival in participants with any (eg, FLT3-ITD variant allele frequency of 1 × 10⁻⁶ or above) detectable peri-HCT MRD according to the treatment arm, and (D) relapse-free survival in participants with no detectable peri-HCT MRD, according to the treatment arm. FLT3-ITD, internal tandem duplication mutation of FLT3; HCT, hematopoietic cell transplantation; HR, hazard ratio; ITT, intention-to-treat; MRD, measurable residual disease; OS, overall survival; RFS, relapse-free survival.

For participants who received a FLT3 inhibitor pre-HCT (60%), gilteritinib conferred a RFS benefit compared with placebo (HR, 0.598; P = .0436) although there was no difference between those who did and did not receive pre-HCT FLT3 inhibition in the rate of detectable pre-HCT MRD (48.3% v 52.1%). Participants who received MAC had improved OS compared with those who received reduced-intensity conditioning (RIC) (HR, 0.529; P = .0027), irrespective of MRD status (Data Supplement, Fig 5). The effect of gilteritinib versus placebo in participants receiving MAC and RIC separately is shown in the Data Supplement (Fig 6).

Subgroup analysis revealed differences in outcomes according to the geographic region. Gilteritinib was beneficial in North America, was of minimal benefit in Asia/rest of world (ROW), and had a mildly negative effect in Europe (Fig 4). However, there were distinct geographic differences in study populations and practice patterns, such as the time from diagnosis to HCT, number of induction and consolidation courses, pre-HCT FLT3 inhibitor use, conditioning regimen, and concomitant azole use (Data Supplement, Table 1).

FIG 4.

Relapse-free survival by treatment arm according to the geographic region: (A) North America (United States and Canada), (B) Europe (Greece, Belgium, France, Spain, Italy, United Kingdom, Denmark, Poland, Germany), (C) Asia Pacific and ROW (Japan, Korea, Taiwan, Australia, New Zealand), and (D) Relapse-free survival of placebo arms only from the three regions (NA, EU, and ROW). EU, Europe; HR, hazard ratio; NA, North America; ROW, rest of world.

Safety

The SAF consisted of 178 gilteritinib and 177 placebo participants. In the gilteritinib arm, 94 of 178 (52.8%) participants completed 24 months of maintenance compared with 96 of 178 (53.9%) on placebo. Treatment-emergent grade II-IV aGVHD occurred in 33 of 178 (18.5%) participants on gilteritinib versus 36 of 177 (20.3%) on placebo (P = .6157), whereas treatment-emergent cGVHD occurred in 93 of 178 (52.2%) on gilteritinib versus 75 of 177 (42.4%) on placebo (P = .181).

Treatment-emergent AEs (TEAEs) ≥grade 3 occurred in 146 of 178 (82%) participants on gilteritinib compared with 94 of 177 (53.1%) on placebo. Both treatment-emergent myelosuppression and infection were more common in the gilteritinib arm compared with placebo, and myelosuppression was the most common reason for early withdrawal from study treatment. Table 2 lists grade 3 or greater TEAEs occurring in 5% or more of participants, and TEAEs leading to drug interruption, dose reduction, or withdrawal from treatment are summarized in the Data Supplement (Table 2). TEAEs leading to drug discontinuation by geographic region are displayed in the Data Supplement (Table 3).

TABLE 2.

Grade 3 or Greater Treatment-Emergent Adverse Events Occurring in 5% or More of Participants (SAF population)

Adverse Event	Gilteritinib (n = 178)	Placebo (n = 177)	Total (n = 355)
	No. of Patients (%)
Hematologic
Neutrophil count decreased	64 (36)	23 (13)	87 (24.5)
Platelet count decreased	38 (21.3)	20 (11.3)	58 (16.3)
Anemia	17 (9.6)	14 (7.9)	31 (8.7)
WBC count decreased	18 (10.1)	3 (1.7)	21 (5.9)
Nonhematologic
ALT increased	11 (6.2)	8 (4.5)	19 (5.4)
AST increased	11 (6.2)	6 (3.4)	17 (4.8)
Hypertension	11 (6.2)	6 (3.4)	17 (4.8)
Creatine phosphokinase elevation	14 (7.9)	1 (0.6)	15 (4.2)

Abbreviation: SAF, safety analysis set.

Because of a previously noted association between azole use, gilteritinib trough levels, and myelosuppression,²⁵ we examined gilteritinib pharmacokinetics using plasma collected at regular intervals. A total of 67.8% of participants were treated with concomitant azoles (fluconazole, itraconazole, posaconazole, voriconazole, and isavuconazonium), with considerable geographic variation. Concomitant azole use was associated with higher median gilteritinib concentrations, but there was wide interparticipant variability (Data Supplement, Fig 7A). Concomitant azole use was more common outside of North America (Data Supplement, Fig 7B).

DISCUSSION

These data show that the improvement in RFS conferred by gilteritinib over placebo did not reach the predetermined level of significance. However, in secondary analysis, consistent with the pretrial hypothesis, participants with FLT3-ITD AML who undergo HCT in first remission with peri-HCT detectable FLT3-ITD MRD benefit from post-HCT gilteritinib. By contrast, participants in deep remissions did not benefit from maintenance gilteritinib and were therefore exposed unnecessarily to its potential toxicity.

Our data suggest that FLT3 inhibition during induction and/or consolidation may select for participants who are more likely to benefit from post-HCT FLT3 inhibition, which was somewhat unexpected. It is possible that in many cases, pre-HCT FLT3 inhibition serves to control, but not eliminate, FLT3-driven AML clones, and continuous inhibition is necessary until an allogeneic effect can eradicate the disease. In the absence of FLT3 inhibition during induction, many participants with these FLT3-driven clones presumably relapse before HCT.

Although FLT3-ITD mutations detected by standard PCR have generally been considered unreliable markers of MRD,²⁶ recent studies have established the value of PCR-NGS FLT3-ITD MRD.^18-20 Using that assay (currently available in the United States),²¹ we found a high correlation between detection of a FLT3-ITD mutation (at any level) and benefit from a drug specifically targeting that mutation. A post hoc analysis of a recent study using a similar MRD assay suggested that a level of 10⁻⁴ was an important survival discriminator, but this was postinduction rather than peri-HCT.²⁰ Our prospective findings establish FLT3-ITD mutations as reliable and actionable markers of MRD in the peri-HCT setting.

The principal toxicity observed in this study was myelosuppression, a known effect of potent FLT3 inhibitors.^23,27 The mechanism is likely inhibition of wild-type FLT3 on multipotent progenitor cells.²⁸ A study of gilteritinib combined with intensive chemotherapy reported an association between higher gilteritinib plasma concentrations and concomitant azole use and myelosuppression.²⁵ Azole use was much more common outside North America, and given that myelosuppression led to drug interruption, reduction, or withdrawal, variations in azole use might have contributed to the geographic variation in efficacy we observed.

A single cause of the observed regional differences was not identified in efficacy end points. Participants in the placebo arm in North America, in contrast to those in Europe or Asia/ROW, displayed a 2-year RFS very close to the 59% that was predicted from Center for International Blood and Marrow Transplant Research data used in the statistical analysis plan (Fig 4D). In contrast to the other participants, most North American participants received FLT3 inhibitors pre-HCT and, in general, were bridged more rapidly to HCT (Data Supplement, Table 1). FLT3-ITD AML is a molecularly heterogeneous disease, with responsiveness to FLT3 inhibition clearly influenced by comutations.^29,30 It is possible that, outside of North America, patients with disease in which FLT3 was a more prominent driver were less likely to remain in remission long enough to enroll on this study because of lack of FLT3 inhibition, a longer time from diagnosis to transplant, or both. These differences might have selected for a different patient population in North America, one more likely to benefit from post-HCT FLT3 inhibition. At the 110 different centers on this study, the variation in number and intensity of induction and consolidation regimens, azole use, availability of FLT3 inhibitors, time to transplantation, conditioning regimens, and GVHD prophylaxis platforms all were reflections of local clinical practice. They might have contributed to such regional differences, but no single practice or group of practices explaining the differences could be identified in multivariate regression models.

We conducted this study to challenge the assumption that all patients with FLT3-ITD AML worldwide, regardless of those variations, should receive a FLT3 inhibitor post-HCT, and our results have indeed invalidated that assumption. In summary, we found that post-HCT maintenance with gilteritinib does confer a benefit for patients with FLT3-ITD AML, but only for those with peri-HCT FLT3-ITD MRD. At the same time, we have validated the utility of FLT3-ITD mutations as useful markers of MRD with clear implications for intervention. These findings are practice-changing, and further study of the data from this trial is likely to yield more insights into the biology and management of this disease.

APPENDIX

TABLE A1.

BMT-CTN 1506/MORPHO Study Investigators

Investigator	Institution
Ed Agura	Baylor University Research Institute
Jessica Altman	Northwestern Medicine
Achiles Anagnostopoulos	General Hospital of Thessaloniki “G. Papanikolaou”
Sarah Anand	University of Michigan
Andrew Artz	University of Chicago
Walter Aulitzky	Robert-Bosch-Krankenhaus GmbH
Sophia Balderman	Roswell Park Cancer Institute
Karen Ballen	University of Virginia
Michael Becker	University of Rochester Medical Center
Yves Beguin	CHU de Liege
Leanne Berkahn	Auckland Hospital
Zwi Berneman	UZ Antwerpen
Vijaya Bhatt	University of Nebraska Medical Center
Ian Bilmon	Westmead Hospital
Francesca Bonifazi	A.O.di Bologna Policl.S.Orsola
Adrienne Briggs	Cancer Transplant Institute at Virginia G. Piper Cancer Center
Benedetto Bruno	Universita di Torino
Claudio Brunstein	University of Minnesota
Michael Byrne	Vanderbilt University Medical Center
Jenny Byrne	Nottingham City Hospital
Monica Cabrero	Hospital Universitario de Salamanca
Roberto Cairoli	Ospedale Metropolitano Niguarda
George Carrum	Baylor College of Medicine
Jan Cerny	University of Massachusetts Memorial Medical Center
Yi-Bin Chen	Massachusetts General Hospital
June-Won Cheong	Severance Hospital in Yonsei University Health System
Fabio Ciceri	Ospedale San Raffaele
Mercedes Colorado	H.U.Marq.Valdecilla
Rachel Cook	Oregon Health & Science University
Daniel Couriel	University of Utah, Huntsman Cancer Institute
Charles Craddock	Queen Elizabeth Hospital Birmingham
Lloyd Damon	University of California, San Francisco
Abhinav Deol	Karmanos Cancer Institute
Yohan Desbrosses	Hopital Jean Minjoz
Steve Devine	Ohio State University Hospital
Carmela Di Grazia	Diparitmento di Malattie Infettive, IRCCS San Martino IST
Antonio Di Stasi	University of Alabama at Birmingham
Ajoy Dias	Beth Israel Deaconess Medical Center
Kathy Dorritie	University of Pittsburgh Cancer Institute
James Essell	Oncology Hematology Care, Inc
Tetsuya Eto	KKR Hamanomachi Hospital
Sherif Farag	Indiana University
Edouard Forcade	Hopital Haut Leveque
Olga Frankfurt	Northwestern Medicine
Shinichiro Fujiwara	Jichi Medical University Hospital
Takahiro Fukuda	National Cancer Center Hospital
Kentaro Fukushima	Osaka University Hospital
Sabine Furst	Institut Paoli-Calmettes
Tatsunori Goto	Japanese Red Cross Aichi Medical Center Nagoya Daiichi Hospital
Aric Hall	University of Wisconsin Hospital & Clinics
Shunsuke Hatta	National Hospital Organization Sendai Medical Center
Yosr Hicheri	Hopital Saint-Eloi
Mitchell Horwitz	Duke University Health System
Hsin-An Hou	National Taiwan University Hospital
Jonathan How	McGill University Health Centre
Dianna Howard	Wake Forest Baptist Health
Wei-Hsun (Blake) Hsu	Christchurch Clinical Studies Trust Ltd
Anne Huynh	I.U.C.T-O
David Irvine	Beatson West of Scotland Cancer Centre
Takayuki Ishikawa	Kobe City Medical Center General Hospital
Katarzyna Jamieson	University of North Carolina Chapel Hill
Wieslaw Jedrzejczak	MTZ Clinical Research Sp. z o.o.
Yogesh Jethava	Indiana Blood and Marrow Transplant
Antonio Jimenez	University of Miami University of Miami Hospital and Clinics
Chul Won Jung	Samsung Medical Center
Junya Kanda	Kyoto University Hospital
Dimitrios Karakasis	Evangelismos Hospital
Jun Kato	Keio University Hospital
Natasha Kekre	Ottawa Hospital
Nandita Khera	Mayo Clinic—Phoenix, AZ
Hee-Je Kim	Seoul St Mary's Hospital
Andreas Klein	Tufts Medical Center
Guido Kobbe	Universitätsklinikum Düsseldorf, Klinik für Nephrologie
Brian Kornblit	Rigshospitalet
Vamsi Kota	Augusta University, Georgia Regents University
Silvy Lachance	Maisonneuve-Rosemont, Université de Montréal
Brian Leber	Hamilton Health Sciences/Juravinski Cancer Centre
Catherine Lee	University of Utah, Huntsman Cancer Institute
Je Hwan Lee	Asan Medical Center
Mark J. Levis	Johns Hopkins University
Tung-Liang Lin	Chang Gung Medical Foundation-Linkou Branch
Mark Litzow	Mayo Clinic—Rochester
Ta-Chih Liu	Kaohsiung Medical University Hospital
Maurizio Martelli	Università degli Studi di Firenze
Carmen Martinez	Hospital Clinic de Barcelona
Kenichi Matsuoka	Okayama University Hospital
John McCarty	Virginia Commonwealth University, Massey Cancer Center
Lourdes Mendez	Beth Israel Deaconess Medical Center
Fotios Michelis	Princess Margaret Cancer Centre
Jan-Henrik Mikesch	Universitatsklinikum Muenster
Shin Mineishi	Penn State Hershey Medical Center
Asmita Mishra	H. Lee Moffitt Cancer Center
Mohamad Mohty	Hopital Saint-Antoine
Ine Moors	UZ Gent
Gabriela Motyckova	LDS Hospital, Intermountain BMT
Lutz Mueller	Universitatsklinik und Poliklinik fuer Innere Medizin IV
Lori Muffly	Stanford University
Yuho Najima	Tokyo Metropolitan Komagome Hospital
Hirohisa Nakamae	Osaka Metropolitan University Hospital
Nobuaki Nakano	Imamura Bun-in Hospital
Sunita Nathan	Rush University Medical Center
Emma Nicholson	Royal Marsden NHS Foundation
Maxim Norkin	University of Florida
Yoshiaki Ogawa	Tokai University Hospital
Gitte Olesen	Aarhus University Hospital
Olalekan Oluwole	Vanderbilt University Medical Center
Masahiro Onozawa	Hokkaido University Hospital
Jeremy Pantin	Augusta University, Georgia Regents University
Esperanza B. Papadopoulos	Memorial Sloan-Kettering Cancer Center
Kristjan Paulson	CancerCare Manitoba
Lucy Pemberton	Dunedin Hospital
Travis Perera	Wellington Hospital
Alexander E. Perl	University of Pennsylvania
Beata Piatkowska-Jakubas	Szpital Uniwersytecki w Krakowie
Xavier Poire	Cliniques Universitaires Saint-Luc
Rachel Protheroe	University Hospitals Bristol NHS Foundation Trust
Alessandro Rambaldi	Ospedale Papa Giovanni XXIII
David Ritchie	Royal Melbourne Hospital
Kelly Ross	West Virginia University Medicine
Marie-Therese Rubio	CHRU Brabois—Service Hématologie et Medecine Interne
Stella Santarone	Ospedale Civile Santo Spirito
Jaime Sanz Caballer	H. U. Politecnico La Fe
Masashi Sawa	Anjo Kosei Hospital
Dale Schaar	Rutgers Cancer Institute
Christoph Scheid	Medical University of Cologne
Jeffrey Schriber	Cancer Transplant Institute at Virginia G. Piper Cancer Center
Stuart Seropian	Yale University
Nilay Shah	West Virginia University Medicine
Nirav Shah	Medical College of Wisconsin
Tsiporah Shore	NYP/Weill Cornell Medical Center
Jorge Sierra Gil	Hospital de la Santa Creu i Sant Pau
Anurag Singh	The University of Kansas Health System
Ronald Sobecks	Cleveland Clinic Foundation
Gerard Socie	Hopital Saint Louis
Robert Soiffer	Dana Farber Cancer Institute
Melhem Solh	Northside Hospital
Kellie Sprague	Tufts Medical Center
Alexandros Spyridonidis	University General Hospital of Patras
Matthias Stelljes	Universitatsklinikum Muenster
Patrick Stiff	Loyola University Medical Center
Robert Stuart	Medical University of South Carolina
Masatsugu Tanaka	Kanagawa Cancer Center
Anand Tandra	Indiana Blood and Marrow Transplant
Eleni Tholouli	Central Manchester University Hospital NHS Foundation Trust
Xavier Thomas	Centre Hospitalier Lyon Sud
Kirsty Thomson	University College London Hospital NHS Foundation Trust
Mario Tiribelli	Azienda Ospedaliero-Universitaria di Udine
Benjamin Tomlinson	University Hospitals Cleveland Medical Center
Panagiotis Tsirigotis	University Hospital Attikon
Dimitrios Tzachanis	University of California San Diego
Naoyuki Uchida	KKR Toranomon Hospital
Masumi Ueda	Fred Hutchinson Cancer Research Center
Celalettin Ustun	University of Minnesota
Geoffrey L. Uy	Washington University in St Louis
David Valcarcel Ferreiras	Hospital Universitario Vall D'Hebron
Sumithra Vasu	Ohio State University Hospital
Eva Wagner	Johannes-Gutenberg-Universitat, Universitätsklinik Mainz
Edmund K. Waller	Emory University
Anne-Marie Watson	Liverpool Hospital
Daniel Weisdorf	University of Minnesota
John R. Wingard	University of Florida
Christine Wolschke	Universitatsklinikum Hamburg-Eppendorf
Tomasz Wrobel	Uniw Szpital Kliniczny im Jana Mikulicza-Radeckieg we Wrocla
Ibrahim Yakoub-Agha	CHRU de Lille
Takuji Yamauchi	Kyushu University Hospital
Jean Yared	University of Maryland Medical Center
Su-Peng Yeh	China Medical University Hospital
Sung-Soo Yoon	Seoul National University Hospital
Satoshi Yoshihara	Hyogo College of Medicine, College Hospital

DATA SHARING STATEMENT

Researchers may request access to anonymized participant-level data, trial-level data, and protocols from Astellas-sponsored clinical trials at www.clinicalstudydatarequest.com. For the Astellas criteria on data sharing, see https://clinicalstudydatarequest.com/Study-Sponsors/Study-Sponsors-Astellas.aspx.

AUTHOR CONTRIBUTIONS

Conception and design: Mark J. Levis, Mehdi Hamadani, Brent Logan, Richard J. Jones, Anurag K. Singh, Alexander E. Perl, Robert Soiffer, Edmund K. Waller, Guido Marcucci, Nahla Hasabou, Matt Rosales, Jason Hill, Mary M. Horowitz, Yi-Bin Chen

Administrative support: Jan-Henrik Mikesch, Guido Marcucci, Heather Wittsack, Mary M. Horowitz

Provision of study materials or patients: Richard J. Jones, Mark Litzow, John R. Wingard, Alexander E. Perl, Robert Soiffer, Geoffrey L. Uy, Edmund K. Waller, Sumithra Vasu, Hee-Je Kim, Jan-Henrik Mikesch, Masahiro Onozawa, Kirsty Thomson, Guido Marcucci, Yi-Bin Chen

Collection and assembly of data: Mark J. Levis, Mehdi Hamadani, Anurag K. Singh, Mark Litzow, John R. Wingard, Alexander E. Perl, Robert Soiffer, Geoffrey L. Uy, Edmund K. Waller, Sumithra Vasu, Hee-Je Kim, Jan-Henrik Mikesch, Yuho Najima, Masahiro Onozawa, Kirsty Thomson, Guido Marcucci, David Delgado, Matt Rosales, Jason Hill, Denise King, Heather Wittsack, Mary M. Horowitz, Yi-Bin Chen

Data analysis and interpretation: Mark J. Levis, Mehdi Hamadani, Brent Logan, Anurag K. Singh, Mark Litzow, John R. Wingard, Esperanza B. Papadopoulos, Alexander E. Perl, Robert Soiffer, Celalettin Ustun, Masumi Ueda Oshima, Geoffrey L. Uy, Edmund K. Waller, Sumithra Vasu, Melhem Solh, Asmita Mishra, Lori Muffly, Hee-Je Kim, Jan-Henrik Mikesch, Masahiro Onozawa, Arnon Nagler, Andrew H. Wei, Guido Marcucci, Nancy L. Geller, Nahla Hasabou, David Delgado, Matt Rosales, Jason Hill, Stanley C. Gill, Rishita Nuthethi, Steven M. Devine, Mary M. Horowitz, Yi-Bin Chen

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Gilteritinib as Post-Transplant Maintenance for AML With Internal Tandem Duplication Mutation of FLT3

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).