A Prospective Study of Peritransplant Sorafenib for Patients with FLT3-ITD Acute Myeloid Leukemia Undergoing Allogeneic Transplantation

ARTICLE IN PRESS Biol Blood Marrow Transplant 000 (2019) 1
7

Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org

A Prospective Study of Peritransplant Sorafenib for Patients with FLT3-ITD Acute Myeloid Leukemia Undergoing Allogeneic Transplantation Keith W. Pratz1,*, Michelle A. Rudek1, B. Douglas Smith1, Judith Karp1, Ivana Gojo1, Amy Dezern1, Richard J. Jones1, Jackie Greer1, Christopher Gocke2, Maria R. Baer3, Vu H. Duong3, Gary Rosner1, Marianna Zahurak1, John J. Wright4, Ashkan Emadi3, Mark Levis1, on behalf of the ETCTN-8922 study team 1

Johns Hopkins Sidney Kimmel Comprehensive Cancer Center, Baltimore, Maryland Department of Pathology, Johns Hopkins University, Baltimore, Maryland University of Maryland Greenebaum Comprehensive Cancer Center, Baltimore, Maryland 4 IDB/CTEP/NCI, National Cancer Institute, Rockville, Maryland 2 3

Article history: Received 28 June 2019 Accepted 17 September 2019 Keywords: Post-transplant Maintenance FLT3

A B S T R A C T FLT3-ITD-mutated acute myeloid leukemia (AML) remains a therapeutic challenge. FLT3 inhibition in the setting of minimal residual disease and a new immune system via allogeneic transplantation offers a promise of improved survival for these patients. We performed a prospective study of patients with FLT3-ITD AML undergoing allogeneic transplant that was conducted to evaluate the safety, tolerability, and outcome of sorafenib administered peritransplant. Sorafenib dosing was individualized, starting at 200 mg twice a day (BID), and titrated based on tolerability or toxicities until a tolerable dose was identified. Forty-four patients, with a median age of 52 years, undergoing allogeneic transplant were started on sorafenib in the peritransplant period (21 pretransplant). The median duration of post-transplant follow-up was 27.6 months (range, 5.2 to 60.4). Overall survival was 76% at both 24 and 36 months. Event-free survival at 24 and 36 months was 74% and 64%, respectively. Ten patients died in the post-transplant period, with 6 deaths due to relapsed leukemia and 4 from transplant-associated toxicity. Tolerable doses ranged from 200 mg every other day to 400 mg BID with similar exposure. Correlative studies evaluating FLT3 inhibition via a plasma inhibitory activity assay showed consistent inhibition of FLT3 at all tolerability-determined dosing levels. Sorafenib is well tolerated in the peritransplant setting irrespective of the conditioning intensity or the donor source. Our findings indicate that sorafenib dosing can be individualized in the post-transplantation setting according to patient tolerability. This approach results in effective in vivo FLT3 inhibition and yields encouraging survival results. © 2019 Published by Elsevier Inc. on behalf of the American Society for Transplantation and Cellular Therapy

INTRODUCTION FLT3-ITD-mutated acute myeloid leukemia (AML) is rarely cured with chemotherapy alone [1]. Current approaches incorporate allogeneic hematopoietic stem cell transplant (HSCT) in first remission for those patients who have a suitable donor and are medically qualified for transplantation [2]. Specific targeting of the oncogenic FLT3-ITD receptor to maintain remission before and after HSCT represents one avenue yet to be Financial disclosure: See Acknowledgments on page XX. *Correspondence and reprint requests: Keith W. Pratz, The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Division of Hematologic Malignancies, The Bunting-Blaustein Cancer Research Building, 1650 Orleans Street, Room 2M45, Baltimore, MD 21231. E-mail address: [email protected] (K.W. Pratz).

prospectively evaluated to improve this poor prognosis subset. Early in the post-transplant period may be a unique setting to enhance the activity of targeted agents. The minimal residual disease state after allogeneic HSCT provides an opportunity to optimize antileukemia agents because this state has the lowest tumor burden and tumor heterogeneity. Importantly, the FLT3ITD clone appears to be the dominant clone in most relapses of AML with FLT3-ITD. Emerging data suggest that a new, nontolerant, and nonexhausted transplant immune system has the ability to augment the activity of anticancer agents [3,4]. Sorafenib is a multitargeted tyrosine kinase inhibitor with activity against RAF kinase, VEGF receptors, wild-type and ITD-mutated FLT3, Platelet derived growth factor receptors, cKIT, and RET kinase [5]. Sorafenib is approved by the US Food

https://doi.org/10.1016/j.bbmt.2019.09.023 1083-8791/© 2019 Published by Elsevier Inc. on behalf of the American Society for Transplantation and Cellular Therapy

ARTICLE IN PRESS 2

K.W. Pratz et al. / Biol Blood Marrow Transplant 00 (2019) 1
7

and Drug Administration for the treatment of advanced renal cell cancer, thyroid cancer, and inoperable hepatocellular carcinoma. As a single agent, sorafenib has been studied on an intermittent schedule in refractory AML with or without a FLT3 mutation with a clinical response rate observed in 9 of 16 patients (56%), including all 6 patients with FLT3-ITD [6]. In a previous phase I dose escalation trial of sorafenib in relapsed/ refractory acute leukemias, we demonstrated excellent inhibition of kinase targets ERK and FLT3-ITD, with FLT3-ITD silencing occurring below the maximally tolerated dose [7]. Pharmacokinetic studies aided in the elucidation that an active metabolite, sorafenib n-oxide, was both a more potent and specific FLT3 inhibitor than the parent compound. Additionally, we were able to demonstrate continued FLT3 inhibitory activity in the plasma up to 7 days after dosing was completed due to sorafenib’s long half-life and potency against FLT3-ITD. An exposure-response analysis confirmed a strong link between FLT3 inhibition and exposure to sorafenib and metabolite with varying dosing regimens being simulated [8]. A starting dose of sorafenib 200 mg twice a day (BID) was considered a practical choice to initiate therapy based on simulated FLT3 activity across the spectrum of dosing regimens and desire to improve long-term tolerability. There are several case reports describing compassionate use of sorafenib inducing complete remissions in leukemia in the literature [9-11]. Studies in which sorafenib has been combined with chemotherapy have also been reported [12,13]. The addition of sorafenib to conventional cytarabine and anthracycline induction results in improved event-free survival (EFS) in unselected patients younger than 60 years but not in patients older than 60 years [14]. There are numerous reports of the benefit of sorafenib in combination with an allogeneic effect, but most of these reports are associated with small numbers or short follow-up [15
19]. A randomized placebo-controlled study of sorafenib use for 24 months in the post-transplant setting has been presented in abstract form and suggests a significant benefit in relapse-free survival [20]. We conducted a prospective study of sorafenib administered as peritransplant therapy to patients with FLT3-ITD AML undergoing allogeneic transplant. The primary objective of this study was to develop a practical way to individualize sorafenib administration in a manner that minimized toxicity but maximized the potential benefit. PATIENTS/METHODS Patient Selection The study enrolled patients older than 18 years with AML with a FLT3ITD mutation who were in a complete remission or partial remission (<10% blasts in the marrow), as documented by bone marrow biopsy and for whom allogeneic HSCT was planned. Patients could have received any prior therapy to achieve remission. Sorafenib could be started (or resumed) no sooner than 30 days after induction/consolidation and/or transplantation but no later than 120 days post-transplant if patients achieved count recovery, defined as having an absolute neutrophil count over 500/mm3 and a nontransfused platelet count over 30,000/mm3. Standard end-organ function for phase I investigations was used for inclusion, including bilirubin <2.0 mg/dL, hepatic transaminases <5 times upper limit of normal, and creatinine <1.5 times upper limit of normal. Patients with active graft-versus-host disease (GVHD) or patients who had escalation of GVHD therapies within 21 days were excluded from starting sorafenib. Patients were also ineligible if they had nonengraftment in the bone marrow assessed at day 60, as defined by having less than 90% donor DNA. Treatment Plan Treatment was administered with the starting dose of 200 mg by mouth BID and dose escalated after 7 days. Dosing of sorafenib was based on individual patient tolerance: in tolerant patients, sorafenib could be escalated to 400 mg BID, whereas dose reductions occurred whenever a patient experienced a grade 3 or 4 toxicity in a stepwise manner (200 mg BID, 200 mg daily [QD], or 200 mg every other day [QOD]), until a tolerable dose was identified.

Chronic grade 2 toxicities resulted in dose de-escalations at the discretion of the treating provider. Pretransplant dosing was held 3 days before the start of the bone marrow transplant preparative regimen. Once the dose was reduced, patients remained at the reduced dose for 30 days, at which point the dose could be maintained or escalated. Dose escalation occurred one dose level at a time at the discretion of the investigator. Patients could have their dose de-escalated 3 times before being removed from the trial. Patients could stop sorafenib at 24 months post-transplant or continue at the discretion of the investigator. Protocol and consent form were approved by the Johns Hopkins and University of Maryland Schools of Medicine Institutional Review Boards. All patients gave informed consent according to the Declaration of Helsinki. The study was registered at ClinicalTrials.gov under NCT01578109. Evaluation of Response Standard post-transplant follow-up was maintained with serial bone marrow biopsies occurring 60 days, 6 months, 12 months, and 2 years after transplant. At the discretion of the investigator, bone marrow biopsies were performed for evaluation of cytopenias at earlier time points. Standard response criteria for AML were used to evaluate for relapse according to the International Working Group [21]. GVHD was graded by the investigator according to modified Keystone criteria [22]. Toxicity was monitored and graded according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (version 4) and was reported for all cycles of treatment. Pharmacokinetic and Pharmacodynamic Studies Plasma was collected at each monthly visit and was to be a trough sample (»12 or 24 hours after the last dose depending on QD or BID regimen and before the next dose). Pharmacokinetic assessment of sorafenib and sorafenib-N-oxide was performed according to previously described methods [23,24]. Adjusted sorafenib concentrations were determined for each point accounting for the active metabolite (sorafenib + [14.59*sorafenib N-oxide]) as previously described [7]. The samples were also analyzed to assess FLT3ITD inhibition according to our previously described plasma inhibitory activity (PIA) assay [7]. FLT3 inhibition was expressed as the percentage of the baseline FLT3 phosphorylation, with baseline considered 100%. Statistical Methods Event-time distributions for EFS and overall survival (OS) were estimated using the method of Kaplan and Meier and compared using the log-rank statistic or the proportional hazards regression model. EFS was defined as time from transplant to relapse, nonrelapse mortality (NRM), or last follow-up date. Estimates and confidence intervals for the cumulative incidences of NRM, relapse, and acute GVHD were obtained using the proportional subdistribution hazard regression model for competing risks [25]. Proportions are reported with exact 95% binomial confidence intervals. Statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC) and R version 3.0 or JMP statistical discovery software v7 (SAS Institute). Individual adjusted sorafenib concentrations and FLT3 inhibition values were compared across dose levels deemed to be tolerable. To account for the correlation among samples obtained from the same patient, generalized estimating equations were used (assuming a compound symmetry correlation structure) for model estimation. In all cases, P < .05 was considered statistically significant. An inhibitory Emax model was used to assess correlations between FLT3 inhibition and adjusted sorafenib concentration using Phoenix WinNonlin version 6.3 (Pharsight A Certara Company, Cary, NC).

RESULTS The study accrued 44 patients from February 2012 through May 2017. Patient characteristics are listed in Table 1. Twentyone patients received sorafenib before allogeneic HSCT as part of the study. Transplant conditioning regimens and donor sources were determined independently of this study. Sixteen patients underwent myeloablative conditioning and received a transplant from a matched related donor (n = 9), a matched unrelated donor (n = 3), or a haploidentical donor (n = 4). Twenty-eight patients underwent nonmyeloablative conditioning and transplant using a haploidentical donor (n = 15), a matched sibling donor (n = 6), a matched unrelated donor (n = 4), or an umbilical cord blood donor (n = 3). In the 43 patients who met the criteria to start sorafenib post-transplant, the median sorafenib start day post-transplant was 65 days (range, 30 to 119). One patient, who received sorafenib in the pretransplant setting, did not recover platelet count

ARTICLE IN PRESS K.W. Pratz et al. / Biol Blood Marrow Transplant 00 (2019) 1
7

Table 1 Demographics Characteristic

common grade 3/4 toxicity was elevated hepatic enzymes (aspartate aminotransferase/alanine aminotransferase/Alkaline phosphatase) seen in 10 patients (grade 4 in 3 patients). There were no cases of veno-occlusive disease of the liver observed on study. The most common hematologic toxicity was grade 3/4 thrombocytopenia seen in 7 patients. Other grade 4/5 toxicities included multiorgan system failure in 3 patients, all related to infections in the post-transplant period. One patient death was attributed to post-transplant pulmonary syndrome/pulmonary GVHD 335 days post-transplant after only a 10-day exposure to sorafenib in the immediate post-transplant period.

Value

Age, median (range), yr

52 (19-72)

Sex: male, n (%)

20/44 (45)

Cytogenetics, n (%) Normal

28/44 (64)

Abnormal

16/44 (36)

NPM1 mutation, n (%)

25/42 (59)

FLT-3 ITD mutation allelic ratio, n (%) Low (1%-24% AR)

6/35 (17)

Intermediate (25%-50% AR)

8/35 (23)

High (>50% AR)

21/35 (60)

Complete remission 2 or greater, n (%)

Tolerability of Sorafenib Dosing started at 200 mg BID in all patients post-transplant and continued according to patient tolerance. Most patients on trial were unable to escalate the dose of sorafenib or could not tolerate the dose of 400 mg BID in the post-transplant period. The dose was able to be escalated and maintained at the 400-mg BID dose in only 4 patients on study. Twenty-two patients were able to tolerate 200 mg BID, 14 patients were maintained on 200 mg QD, and 3 patients could tolerate only 200 mg QOD. Seven patients stopped the drug during therapy due to toxicity or intolerance. The median duration of sorafenib dosing was 679 days in the post-transplant setting.

12/44 (27)

Transplant type, n (%) Myeloablative

16/44 (36)

Nonmyeloablative

28/44 (64)

Stem cell source, n (%) Bone marrow

26/44 (59)

PBSC

15/44 (34)

Umbilical cord blood

3/44 (7)

Pretransplant sorafenib, n (%)

21/44 (48)

MRD at time of transplant, n (%)

17/44 (37)

Donor sex disparity, n (%)

13/44 (30)

3

AR indicates allelic ratio; PBSC, peripheral blood stem cell; MRD, minimal residual disease.

Effect of Pretransplant Sorafenib Nearly half of the patients (21/44) on this study received sorafenib as maintenance therapy before transplantation as this was available to patients in remission waiting more than 4 weeks after count recovery to proceed to transplant. The median duration of pretransplant sorafenib use was 31 days on study. Of the 21 patients who received pretransplant sorafenib, all achieved donor engraftment as documented in unsorted donor chimerism at day 60 post-transplant with no episodes of veno-occlusive disease.

above 30,000/mm3 at day 120, despite full donor engraftment, and was not treated with sorafenib in the post-transplant setting. Sorafenib-Associated Toxicities Treatment-emergent adverse events deemed possibly related to sorafenib are summarized in Table 2. The most Table 2 Treatment-Emergent Adverse Events Category

Grade

400 mg BID, n

200 mg BID, n

200 mg QD, n

200 mg QOD, n

Total, n (%)

ALT/AST

2

3

3

0

0

6/44 (14)

ALT/AST

3/4

1

5

3

1

10/44 (23)

Platelet count decreased

2

1

7

1

0

9/44 (20)

Platelet count decreased

3/4

0

4

3

0

7/44 (16)

Neutrophil count decreased

3/4

0

4

3

0

7/44 (16)

Hepatic

Hematologic

Metabolic Hypophosphatemia

3

0

1

0

0

1/44 (2)

Hyperkalemia

3

1

1

1

0

3/44 (7)

Elevated creatinine

2

0

1

1

0

2/44 (5)

Hypertension

2

2

6

1

0

9/44 (20)

Hypertension

3

0

2

1

0

3/44 (7)

Diarrhea

2

3

1

1

0

5/44 (11)

Nausea/anorexia

2

1

6

1

0

8/44 (18)

Nausea/anorexia

3

0

1

0

0

1/44 (2)

Palmar-plantar erythrodysesthesia

2

1

8

2

0

11/44 (25)

Palmar-plantar erythrodysesthesia

3

0

2

0

0

2/44 (5)

Rash maculopapular

2

1

5

1

0

7/44 (16)

Cardiac

Gastrointestinal

Skin

ALT indicates alanine aminotransferase; AST, aspartate aminotransferase.

ARTICLE IN PRESS 4

K.W. Pratz et al. / Biol Blood Marrow Transplant 00 (2019) 1
7

Table 3 Transplant-Specific Outcomes Transplant Type

N

GVHD Grade III or IV, n

Relapsed Disease, n

NRM, n

Event Free, n

Myeloablative

16

4

3

1

12

Matched sibling donor*

9

1

1

1

7

Matched unrelated donory

3

2

1

0

2

4

1

1

0

3

28

8

5

3

20

Haploidenticalz Nonmyeloablative Haploidentical donorx

15

5

3

1

11

Matched sibling donor║

6

0

1

0

5

Matched unrelated donor{

4

3

0

2

2

Double umbilical cord blood#

3

0

1

0

2

* Busulfan (Bu)-Cytoxan (Cy)-Post transplant Cytoxan (PtCy) (3), Bu-Fludarabine (Flu) (2), Cy-Total body irradiation (TBI) (2), Bu-Flu-Clofarabine (Clo) (1), and Bu-Cy (1). y Bu-Cy-PtCy (3). z Bu-Cy-PtCy (4). x Flu-Cy-TBI-PtCy (15/15). ║ Flu-Cy-TBI-PtCy (3), Flu-Bu-TBI (2), and Bu-Flu-Anti-thymocyte globulin (ATG) (1). { Flu-Cy-TBI-PtCy (2), Flu-Melphalan (Mel)-ATG (1), and Bu-Flu-TBI-ATG (1). # Flu-Cy-TBI (3/3).

GVHD Sorafenib initiation criteria required stability of acute GVHD for at least 21 days. Flares of GVHD to grade III or IV levels during the first 4 weeks of sorafenib dosing requiring escalation of systemic immunosuppression occurred in 12 of 44 patients (27%). GVHD did not appear linked to donor type or conditioning regimen (Table 3). Cumulative incidence of grade III to IV acute GVHD at days 100, 200, and 365 were 14% (95% confidence interval [CI], 3% to 24%), 23% (95% CI, 10% to 35%), and 25% (95% CI, 12% to 38%). Due to the various transplant regimens, stem cell sources, and post-transplant immunosuppression strategies, the higher rates of grade III/IV GVHD seen in matched unrelated transplant donors (5/7 patients) are not clearly attributable to sorafenib. Outcome of Patients on Sorafenib Eight patients have had recurrence of AML in the posttransplant period (Figure 1), with 6 patient deaths attributed to relapsed disease. Five patient relapses occurred while the patient was off sorafenib for more than 1 week for toxicity. One patient relapsed with FLT3-ITD leukemia 115 days after stopping sorafenib at the 24-month planned stopping time and is now back in remission after resuming sorafenib. One

relapse occurred at the sorafenib dose of 200 mg QD, and 1 relapse occurred at a dose of 200 mg QOD. Molecular drivers of relapse included isolated D835 relapse in a patient with concurrent ITD and D835 at diagnosis, D835 and ITD relapse in 1 patient, ITD and NRAS mutation in 1 patient, and non-FLT3 relapse in 1 patient. With a median follow-up of 27.6 months post-transplant, the median OS for the 44 patients on study has not been reached. OS at 24, 36, and 48 months was 76% (95% CI, 63% to 91%), 76% (95% CI, 63% to 91%), and 57% (95% CI, 31% to 91%), respectively. Median EFS has not been reached. The 24-, 36-, and 48-month EFS rates were 74% (95% CI, 62% to 90%), 64% (95% CI, 48% to 85%), and 64% (95% CI, 48% to 85%), respectively. NRM at 3 years post-transplant was 10% (95% CI, 1% to 20%). Outcomes based on transplant conditioning regimen and stem cell source appear similar, but the study was not large enough to draw formal conclusions (Table 3). Other leukemia features, such as NPM1 positivity or low FLT3 allelic ratio, were not statistically examined formally due to small sample size. Sorafenib Pharmacokinetics There was a large degree of interpatient variability in steady-state sorafenib trough levels (Figure 2). The average

Figure 1. Clinical outcomes: EFS and OS estimated by Kaplan-Meier method.

ARTICLE IN PRESS K.W. Pratz et al. / Biol Blood Marrow Transplant 00 (2019) 1
7

5

FLT3 Inhibition Correlative studies evaluating FLT3 inhibition using a PIA (Figure 3A) assay showed consistent inhibition of FLT3 at all tolerability-determined dosing levels, suggesting that dosing can be safely adjusted in patients having adverse events without loss of FLT3 inhibition (Figure 3B). Mean suppression of FLT3 ITD as measured by PIA was 95.8%, 93.8%, 91.8%, and 89.8% at 400-mg BID, 200-mg BID, 200-mg QD, and 200-mg QOD dosing levels, respectively. There was a statistically significant difference in suppression of FLT3 ITD via PIA across the dose levels (P = .01). The clinical significance of this statistical difference is unclear, as prior studies of FLT3 inhibitors have correlated favorable clinical outcomes with suppression of FLT3 ITD as measured by PIA at levels greater than 80% that was achieved in this trial. Figure 2. Individual adjusted sorafenib concentrations by dose level at the time of sample collection.

adjusted sorafenib trough concentration exposure was 10,738, 8046, 5354, and 2662 ng/mL at 400-mg BID, 200-mg BID, 200-mg QD, and 200-mg QOD dosing levels, respectively. There was a statistically significant difference in exposure across dosing levels (P < .0001).

Pharmacokinetic-Pharmacodynamic Link The adjusted sorafenib concentration correlated with FLT3 inhibition over all tolerable dose levels (r2 > 0.97; Figure 3C). Improved FLT3 inhibition was seen with increasing dose: an estimated additional
2.0% per level (95% CI,
3.6% to 0.5%). Factors affecting pharmacokinetic/pharmacodynamic relationships of FLT3 inhibitors on the FLT3 suppression such as high

Figure 3. (A) Western blot showing representative results of PIA. Upper gel is phosphorylated FLT3 and lower gel is total FLT3 for each time point. Pre represents plasma before sorafenib dosing, and time points beyond are at monthly intervals and labeled by dosing of sorafenib taken that cycle. (B) Scatterplot of individual PIA results on study divided by dosed level demonstrating relative suppression of baseline FLT3 phosphorylation. (C) Inhibitory Emax model of degree of FLT3 inhibition and adjusted sorafenib concentrations plotted with each individual assessment. The dose level at the time of sample collection was denoted by an open diamond (400 mg BID), an open triangle (200 mg BID), an open square (200 mg QD), or an open circle (200 mg QOD).

ARTICLE IN PRESS 6

K.W. Pratz et al. / Biol Blood Marrow Transplant 00 (2019) 1
7

levels of FLT3 ligand observed in the chemotherapy setting [26] were not readily observed in the post-transplant setting, although direct measurements of FLT3 ligand were not done. DISCUSSION Our study represents the largest cohort of prospectively treated patients using sorafenib in the post-transplant setting for FLT3-ITD AML. We have established that sorafenib can be given safely during the pretransplant period and does not appear to impair engraftment. We have also confirmed, in concordance with the findings of other groups, that it can be given safely after engraftment and is associated with a favorable EFS and OS compared with historical outcomes and in 1 study against a placebo control. Sorafenib is an effective FLT3 inhibitor, but its use in patients with AML has been limited by an unfavorable side effect profile. As a drug that is metabolized by the cytochrome P450 system, steady-state levels of sorafenib have a high degree of interpatient variability, and the “standard” dose of 400 mg BID is not tolerated by most patients with AML. However, we have shown here that when sorafenib-induced side effects are used as a guide, adequate suppression of FLT3 activity in vivo is consistently achieved, at the maximum dose tolerated by each patient. Our results therefore can serve as a guide to managing patients with FLT3-ITD AML on sorafenib as follows: (1) begin dosing at 200 mg BID; (2) if there are no side effects, increase the dose to a maximum of 400 mg BID; and (3) at the first dose level causing side effects, decrease the dose to the next lower level. Using this approach of symptom-guided dosing, we have achieved favorable longterm survival while maintaining quality of life for the patient. With this strategy, we have been able to increase patient compliance and maintain patients in remission for periods of time significantly longer compared with other studies [17]. The importance of long-term compliance was suggested by the low rate of recurrence of AML while on therapy (2 cases relapsed while on therapy), whereas those patients who stopped for toxicity had relapses often shortly after therapy cessation. Sixteen patients required dosing holds and 6 (38%) of those patients relapsed while off therapy (median time off sorafenib 16 days [13 to 115]), reinforcing the need for longterm compliance. Lastly, the duration of maintenance with our approach is unknown, as 1 patient stopped protocol-recommended therapy at 2 years post-transplant and relapsed 115 days later. Others have reported relapses late after discontinuation of sorafenib [27]. The administration of sorafenib after engraftment may be associated with increased GVHD, which may actually contribute to the overall therapeutic effect. In the absence of a placebo control, we cannot be certain that sorafenib had an effect on GVHD due to potential reporting bias in our study. However, wild-type FLT3 is expressed on dendritic cells, and its activation contributes to their proliferation [28]. Dendritic cells, in turn, stimulate the proliferation of T-regulatory cells [29]. Therefore, it is possible that FLT3 inhibition by sorafenib suppresses dendritic cell and T-regulatory cell populations, which might cause an increase in GVHD. Future studies of FLT3 inhibition post-transplant could potentially address this question using T cell immunophenotyping during therapy. Lastly, it has been demonstrated that sorafenib synergizes with the T-cellmediated graft-versus-leukemia effect via increased IL-15 production in FLT3-ITD AML cells [30]. In summary, FLT3 inhibition may offer improved survival following allogeneic transplant for patients with FLT3-ITD AML. Well-designed randomized studies will be necessary to establish this approach, and such studies will likely involve the

use of next-generation FLT3 inhibitors. One such study that recently began accrual is BMT-CTN1506 (“MORPHO”), a trial randomizing patients with FLT3-TID AML who have undergone HSCT to post-transplant gilteritinib versus placebo (NCT02997202). However, sorafenib is immediately available off-label in much of the world. Until randomized trials are completed, and until these newer inhibitors are available, sorafenib dosed according to patient tolerance represents an important therapeutic option for FLT3-ITD AML. ACKNOWLEDGMENTS Financial disclosure: The project described was supported in part by National Cancer Institute (NCI) Cooperative Agreement U01CA070095, P01 CA015396, and UM1CA186691 and by the Analytical Pharmacology Core of the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins (National Institutes of Health [NIH] grants P30CA006973 and UL1TR001079) and the University of Maryland Comprehensive Cancer Center Support Grant NCI P30CA134274. Grant UL1TR 001079 is from the National Center for Advancing Translational Sciences (NCATS), a component of the NIH, and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of the Johns Hopkins Institute for Clinical and Translational Research, NCATS, or NIH. Conflict of interest statement: K.W.P. received grant/research support from AbbVie, Agios, Astellas, and Millenium/Takeda and provided consulting for AbbVie, Astellas, and Boston Biomedical. M.A.R. received the following support: employment: Novavax (spouse); stock or other ownership: Novavax (spouse); research funding: Celgene (institution) and Taiho Pharmaceutical (institution); honoraria: UpToDate; travel, accommodations, expenses: Expert Medical Events; and advisory committee: National Comprehensive Cancer Network. B.D.S. received grant/research support from Novartis, Tolero, and Io Therapeutics and provided consulting for Celgene, Jazz, Pfizer, and Bristol Myers Squibb. J.K.: data safety monitoring committee’s for Tolero Pharmaceuticals; Imago; Leukemia Lymphoma Society Beat AML. I.G. received research support from Merck and Amgen and is on the advisory board for Abbvie and Jazz. A.E.D. received grant/research support from Acceleron and Celegene. G. R. is a consultant for Novartis and has stock ownership in Johnson & Johnson. R.J., J.G., C.G., V.H.D., J.J.W., M.B., M.Z., and A.E. have nothing to disclose. M.L. received grant/research support from FujiFilm, Novartis, Astellas, and Takeda and provided consulting for Amgen, Agios, Daiichi-Sankyo, Novartis, Astellas, and FujiFilm. Authorship statement: K.W.P., J.K., R.J.J., G.R., J.J.W., and M.L. designed the study. B.D.S., I.G., A.D., J.G., M.R.B., V.H.D., and A.E. conducted the study and treated study participants. C.G. performed FLT3 and engraftment studies. G.R. and M.Z. performed statistical analysis. M.A.R. performed pharmacokinetic analysis. K.W.P. performed pharmacodynamic analysis and prepared the manuscript and figures for publication. All authors analyzed the data, reviewed the manuscript, and agreed to its submission for publication. REFERENCES 1. Levis M, Small D. FLT3: ITDoes matter in leukemia. Leukemia. 2003;17 (9):1738–1752. 2. Pratz KW, Levis M. How I treat FLT3-mutated AML. Blood. 2017;129 (5):565–571. 3. Bouchlaka MN, Redelman D, Murphy WJ. Immunotherapy following hematopoietic stem cell transplantation: potential for synergistic effects. Immunotherapy. 2010;2(3):399–418. 4. Choi J, Ritchey J, Prior JL, et al. In vivo administration of hypomethylating agents mitigate graft-versus-host disease without sacrificing graft-versusleukemia. Blood. 2010;116(1):129–139.

ARTICLE IN PRESS K.W. Pratz et al. / Biol Blood Marrow Transplant 00 (2019) 1
7

5. Wilhelm SM, Carter C, Tang L, et al. BAY 43-9006 exhibits broad spectrum oral antitumor activity and targets the RAF/MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis. Cancer Res. 2004;64(19):7099–7109. 6. Zhang W, Konopleva M, Shi YX, et al. Mutant FLT3: a direct target of sorafenib in acute myelogenous leukemia. J Natl Cancer Inst. 2008;100(3): 184–198. 7. Pratz KW, Cho E, Levis MJ, et al. A pharmacodynamic study of sorafenib in patients with relapsed and refractory acute leukemias. Leukemia. 2010;24 (8):1437–1444. 8. Liu T, Ivaturi V, Sabato P, et al. Sorafenib dose recommendation in acute myeloid leukemia based on exposure-FLT3 relationship. Clin Transl Sci. 2018;11(4):435–443. 9. Safaian NN, Czibere A, Bruns I, et al. Sorafenib (NexavarÒ) induces molecular remission and regression of extramedullary disease in a patient with FLT3-ITD+ acute myeloid leukemia. Leuk Res. 2009;33(2):348–350. 10. Metzelder S, Wang Y, Wollmer E, et al. Compassionate use of sorafenib in FLT3-ITD-positive acute myeloid leukemia: sustained regression before and after allogeneic stem cell transplantation. Blood. 2009;113(26): 6567–6571. 11. Sammons SL, Pratz KW, Smith BD, Karp JE, Emadi A. Sorafenib is tolerable and improves clinical outcomes in patients with FLT3-ITD acute myeloid leukemia prior to stem cell transplant and after relapse post-transplant. Am J Hematol. 2014;89(9):936–938. 12. Ravandi F, Cortes JE, Jones D, et al. Phase I/II study of combination therapy with sorafenib, idarubicin, and cytarabine in younger patients with acute myeloid leukemia. J Clin Oncol. 2010;28(11):1856–1862. 13. Serve H, Krug U, Wagner R, et al. Sorafenib in combination with intensive chemotherapy in elderly patients with acute myeloid leukemia: results from a randomized, placebo-controlled trial. J Clin Oncol. 2013;31 (25):3110–3118. 14. Rollig C, Serve H, Huttmann A, et al. Addition of sorafenib versus placebo to standard therapy in patients aged 60 years or younger with newly diagnosed acute myeloid leukaemia (SORAML): a multicentre, phase 2, randomised controlled trial. Lancet Oncol. 2015;16(16):1691–1699. 15. Sora F, Chiusolo P, Metafuni E, et al. Sorafenib for refractory FMS-like tyrosine kinase receptor-3 (FLT3/ITD+) acute myeloid leukemia after allogenic stem cell transplantation. Leuk Res. 2011;35(3):422–423. 16. Metzelder SK, Schroeder T, Finck A, et al. High activity of sorafenib in FLT3-ITD-positive acute myeloid leukemia synergizes with allo-immune effects to induce sustained responses. Leukemia. 2012;26(11):2353–2359.

7

17. Chen YB, Li S, Lane AA, et al. Phase I trial of maintenance sorafenib after allogeneic hematopoietic stem cell transplantation for fms-like tyrosine kinase 3 internal tandem duplication acute myeloid leukemia. Biol Blood Marrow Transplant. 2014;20(12):2042–2048. 18. Metzelder SK, Schroeder T, Lubbert M, et al. Long-term survival of sorafenib-treated FLT3-ITD-positive acute myeloid leukaemia patients relapsing after allogeneic stem cell transplantation. Eur J Cancer. 2017;86:233–239. 19. Brunner AM, Li S, Fathi AT, et al. Haematopoietic cell transplantation with and without sorafenib maintenance for patients with FLT3-ITD acute myeloid leukaemia in first complete remission. Br J Haematol. 2016;175(3):496–504. 20. Burchert A, Bug G, Finke J, et al. Sorafenib as maintenance therapy post allogeneic stem cell transplantation for FLT3-ITD positive AML: results from the randomized, double-blind, placebo-controlled multicentre SORMAIN trial. Blood. 2018;132(suppl 1):661. 21. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recommendations of the International Working Group for Diagnosis, Standardization of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol. 2003;21(24):4642–4649. 22. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus conference on acute GVHD grading. Bone Marrow Transplant. 1995;15(6):825–828. 23. Villarroel MC, Pratz KW, Xu L, Wright JJ, Smith BD, Rudek MA. Plasma protein binding of sorafenib, a multi kinase inhibitor: in vitro and in cancer patients. Invest New Drugs. 2012;30(6):2096–2102. 24. Li L, Zhao M, Navid F, et al. Quantitation of sorafenib and its active metabolite sorafenib N-oxide in human plasma by liquid chromatographytandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci. 2010;878(29):3033–3038. 25. Fine J. A proportional hazards model for the subdistribution of a competing risk. J Am Statist Assoc. 1999;94:496–509. 26. Sato T, Yang X, Knapper S, et al. FLT3 ligand impedes the efficacy of FLT3 inhibitors in vitro and in vivo. Blood. 2011;117(12):3286–3293. 27. Gerull S, Tschan-Plessl A, Mathew R, Nair G, Passweg JR, Halter JP. Late relapse after stopping sorafenib in allogeneic hematopoietic stem cell transplant recipients. Bone Marrow Transplant. 2019;54(5):769–771. 28. Maraskovsky E, Daro E, Roux E, et al. In vivo generation of human dendritic cell subsets by Flt3 ligand. Blood. 2000;96(3):878–884. 29. Klein O, Ebert LM, Zanker D, et al. Flt3 ligand expands CD4+ FoxP3+ regulatory T cells in human subjects. Eur J Immunol. 2013;43(2):533–539. 30. Mathew NR, Baumgartner F, Braun L, et al. Sorafenib promotes graft-versus-leukemia activity in mice and humans through IL-15 production in FLT3-ITD-mutant leukemia cells. Nat Med. 2018;24(3):282–291.