Phase I Clinical Trial of Selinexor in Combination with Daunorubicin

Published OnlineFirst October 21, 2019; DOI: 10.1158/1078-0432.CCR-19-2169

CLINICAL CANCER RESEARCH | CLINICAL TRIALS: TARGETED THERAPY

Phase I Clinical Trial of Selinexor in Combination with Daunorubicin and Cytarabine in Previously Untreated Poor-Risk Acute Myeloid Leukemia Kendra Sweet1, Rami Komrokji1, Eric Padron1, Christopher L. Cubitt2, Joel G. Turner3, Junmin Zhou4, Alan F. List1, David A. Sallman1, Jana L. Dawson5, Daniel M. Sullivan3, Julio Chavez1, Bijal D. Shah1, and Jeffrey E. Lancet1

ABSTRACT ◥ Purpose: Induction chemotherapy results in complete and recommended phase II dose of selinexor were the primary remission (CR) rates of 20% to 50% among patients with endpoints. poor-risk AML. Selinexor is an oral selective inhibitor of Results: Twenty-one patients with poor-risk AML were nuclear export with promising single-agent activity. By inhi- enrolled. All 21 patients were included in the safety evaluations biting the primary export protein, XPO1, selinexor localizes and survival analyses (4 in each of 2 cohorts; 13 in the expansion); and activates tumor suppressor proteins in the nucleus and 8 (53%) of the 19 patients evaluable for response achieved CR/CRi. inhibits DNA damage repair, rationalizing combination with MTD was not reached. Selinexor 80 mg (orally, twice weekly) was DNA-damaging agents. used in the expansion phase. The most common grade 3/4 Patients and Methods: This was a single-arm phase I nonhematologic treatment-emergent adverse events were febrile clinical trial of selinexor combined with cytarabine and dau- neutropenia (67%), diarrhea (29%), hyponatremia (29%), and norubicin (7þ3). Dose escalation was selinexor alone (3þ3) sepsis (14%). At median follow-up (28.9 months), 38% of patients with an expansion at the MTD. Cohorts 1 and 2 received 60 were alive. Median overall survival was 10.3 months. and 80 mg orally, respectively, twice weekly during induction. Conclusions: Selinexor plus 7þ3 is a safe regimen for patients Consolidation cycles ( 2) with selinexor at induction dose with newly diagnosed poor-risk AML and warrants further inves- plus 5þ2 were allowed for patients who achieved CR. MTD tigation in a larger clinical trial.

Introduction the inactivation of TSPs, allowing malignant cells to evade apoptosis (3). XPO1 is overexpressed in AML cells, and increased levels of Acute myeloid leukemia (AML) is a clonal hematopoietic disorder XPO1 are inversely correlated with overall survival (OS) in patients characterized by a block in myeloid differentiation and aberrant with AML (4). proliferation of immature myeloid progenitors. An inherent feature Selective inhibitors of nuclear export (SINE) compounds are orally of these self-renewing leukemia-initiating cells is their resistance to bioavailable small molecules that function by covalently binding to chemotherapy, which contributes to disease relapse (1). Nuclear– cysteine 528 in the cargo-binding pocket of the XPO1 protein. This cytoplasmic protein transport is required to maintain normal intra- prevents XPO1 from binding to the nuclear export signal of proteins, cellular signaling and cell-cycle regulation (2). Exportin 1 (XPO1/ thereby preventing the loading and nuclear export of cargo proteins. CRM1) is the sole nuclear export protein responsible for the cyto- The resulting nuclear retention, accumulation, and activation of TSPs plasmic translocation of nearly all tumor suppressor proteins (TSP) restore the tumor suppressor function, leading to apoptosis and cell- and growth regulators, including p53, p21, p27, forkhead box O3 cycle arrest of tumor cells, as well as forced differentiation of myeloid (FOXO3), and nucleophosmin 1 (NPM1). This nuclear export leads to blasts. These effects are seen in bulk tumor cells and leukemia- initiating cells (1, 5–8). Another client of XPO1 is the protein topoisomerase IIa. Cytoplasmic localization of topoisomerase IIa is a well- 1Department of Malignant Hematology, H. Lee Moffitt Cancer Center and documented mechanism of chemotherapy resistance in multiple Research Institute, Tampa, Florida. 2Department of Translational Research, H. myeloma and AML (9). Selinexor (KPT-330) is a SINE compound Lee Moffitt Cancer Center and Research Institute, Tampa, Florida. 3Department that has been extensively studied in laboratories by using AML cell of Blood and Marrow Transplant and Cellular Immunotherapy, H. Lee Moffitt lines, AML patient cells, and mouse models. Selinexor is reported to 4 Cancer Center and Research Institute, Tampa, Florida. Department of Immu- block XPO1, leading to nuclear retention of cargo proteins, including nology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida. 5Department of Experimental Therapeutics, H. Lee Moffitt Cancer Center and TSPs, and restore their functions. Selinexor also restores topoisom- Research Institute, Tampa, Florida. erase IIa in the nucleus and synergistically sensitizes cells to topoisomerase II inhibitors, such as idarubicin, daunorubicin, etoposide, Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). and mitoxantrone (3, 6, 10). The nuclear retention of topoisomerase IIa causes downregulation of DNA repair proteins, such as Rad51 and Corresponding Author: Kendra L. Sweet, H. Lee Moffitt Cancer Center and Research Institute, 12902 USF Magnolia Drive, Tampa, FL 33612. Phone: 813-745- Chk1, and results in increased lethal DNA double-strand breaks in the 6841; Fax: 813-745-3071; E-mail: kendra.sweet@moffitt.org presence of topoisomerase II inhibitors (10). When given sequentially with chemotherapy, selinexor appears to synergize with many DNA- Clin Cancer Res 2019;XX:XX–XX damaging agents (including idarubicin, daunorubicin, mitoxantrone, doi: 10.1158/1078-0432.CCR-19-2169 etoposide, and fludarabine), especially when the selinexor exposure 2019 American Association for Cancer Research. precedes that of the other drugs. Preclinical studies completed at our

AACRJournals.org | OF1

Sweet et al.

Selinexor was given 2 hours before daunorubicin on days 1 and 3 on the Translational Relevance basis of preclinical studies, suggesting that selinexor sensitizes leuke- This clinical trial studied a group of patients with AML with very mic blasts to chemotherapy (3). During the dose escalation phase, poor-risk disease. Treatment options for these patients are limited patients who did not receive at least 5 doses of selinexor, in the absence and are often ineffective. The preclinical rationale for the combi- of a dose-limiting toxicity (DLT), were replaced. nation of drugs used was strong. This is the first clinical trial Reinduction was permitted for patients with residual leukemia seen completed using this combination in this clinical setting, and the on a day 21 bone marrow biopsy, provided a 50% reduction in the results are encouraging. The safety profile of this treatment com- total blast count from baseline was observed. Reinduction consisted of bination was acceptable for this particular group of patients, and administration of daunorubicin (45 mg/m2/day) on days 1 and 2 and the clinical responses suggest that the combination therapy may be cytarabine (100 mg/m2/day) as a continuous intravenous infusion on more effective than standard chemotherapy regimens alone. The days 1 through 5 (5þ2 regimen). Selinexor was given at the same dose results of this clinical trial justify a larger clinical trial with this as induction on days 1, 3, 8, and 10. On day 1, selinexor was given combination therapy. 2 hours before daunorubicin administration. Patients who achieved complete remission (CR) or CR with incomplete blood count recovery (CRi) were allowed to receive up to 2 cycles of consolidation therapy, consisting of 5þ2 with selinexor given at the institution using AML cell lines and AML patient cells indicated strong same dose as induction on days 1, 3, 8, and 10. synergistic interactions between selinexor and daunorubicin, and these Patients were permitted to discontinue their participation in the data rationalized the development of this clinical trial (Supplementary study and proceed to allogeneic hematopoietic stem cell transplant Figs. S1 and S2; Supplementary Table S1). Several early-phase trials (HSCT) at any time after achieving CR/CRi. Patients who did not have studied selinexor in AML both as a single agent and in combi- undergo HSCT could subsequently receive maintenance therapy nation with chemotherapy and report encouraging results (6, 11–13). consisting of selinexor at the same dose as induction on days 1 and This phase I study (NCT02403310) was conducted to further assess 8 of a 21 day cycle for up to 17 cycles (12 months). the role of selinexor in patients with previously untreated, poor-risk Supportive care measures were established to manage side-effects. AML when administered in combination with cytarabine and dau- These included the use of prophylactic antibiotics, antifungal agents, norubicin. The aim of the study was to identify the MTD and the and antivirals. Nausea was managed with antiemetics as needed and, recommended phase II dose (RP2D) of selinexor and to assess the for some patients, nausea management consisted of empiric treatment safety profile and preliminary signals of efficacy with this combination. with olanzapine (5 mg) taken orally at bedtime beginning 24 to 48 hours before day 1 of induction. Diarrhea was managed with loperamide or tincture of opium as needed, and hyponatremia was Patients and Methods managed with intravenous infusions of sodium chloride and/or Patient selection salt tabs. Adults newly diagnosed with AML (on the basis of World Health Organization 2008 criteria) were eligible if they were considered poor Evaluations, endpoint, and assessments risk in accordance with the National Comprehensive Cancer Network The MTD was defined as the dose at which 0 out of 3 or 1 out of 6 (NCCN) cytogenetic and molecular risk stratification guidelines (14) patients experienced a DLT. If DLTs did not occur in either dose or if they were 60 years of age (15). Key eligibility criteria included cohort, then 80 mg of selinexor taken orally twice weekly was to be Eastern Cooperative Oncology Group (ECOG) performance status of deemed the RP2D. Treatment-emergent adverse events (TEAE) were 2, total bilirubin levels within H. Lee Moffitt Cancer Center and assessed using the National Cancer Institute Common Terminology Research Institution (MCC) institutional normal limits, aspartate Criteria of Adverse Events v4.3. The observation period for a DLT was aminotransferase and alanine aminotransferase levels 2.5 the a minimum of 28 days. Hematologic DLTs were defined as grade 3/4 institutional upper limit of normal, creatinine levels 2 mg/dL, and a neutropenia and/or thrombocytopenia due to bone marrow hypopla- left ventricular ejection fraction of 50%. Patients who had prior sia, without residual leukemic burden, that did not recover to grade 2 AML-directed therapy, with the exception of hydroxyurea, were not by day 56. included. Patients with an antecedent myelodysplastic syndrome were Nonhematologic DLTs were defined as any grade 3/4 drug-related permitted to have had prior hypomethylating agent therapy. A full list toxicity, with the exception of nausea, vomiting, dehydration, or of inclusion/exclusion criteria can be found in the Supplementary diarrhea that was adequately controlled with antiemetics or antidiar- Methods The study was approved by the Chesapeake Institutional rheals; infection or febrile neutropenia adequately controlled with Review Board (IRB) and conducted in accordance with the Declaration antibiotics; liver function abnormalities (without clinical symptoms) of Helsinki. The study was consistent with the International Confer- that recovered to baseline or grade 1 within 7 days; electrolyte or ence on Harmonisation Good Clinical Practice guidelines. Written metabolic laboratory abnormalities that were not considered clinically informed consent was obtained from all patients before enrollment. significant by the treating physician and recovered to baseline or grade 1 within 7 days; and alopecia and dysgeusia. Treatment After induction/reinduction, bone marrow aspirations and biopsies Selinexor was supplied by Karyopharm Therapeutics, Inc. Induc- were performed between days 28 and 72 to assess response. The tion therapy was given in the inpatient setting. All patients received International Working Group criteria were used to define a CR or daunorubicin (60 mg/m2/day) intravenously on days 1 through 3 and CRi (16). Patients who did not achieve CR or CRi with up to 2 cytarabine (100 mg/m2/day) as a continuous intravenous infusion on induction courses were considered treatment failures. Patients who days 1 through 7 (7þ3 regimen). Dose escalation was used for selinexor voluntarily withdrew consent before completing all of the prescribed only. Dose cohorts 1 and 2 received 60 mg and 80 mg of selinexor study treatment for reasons unrelated to toxicity were not considered orally, respectively, during induction on days 1, 3, 8, 10, 15, and 17. evaluable for response.

OF2 Clin Cancer Res; 2019 CLINICAL CANCER RESEARCH

Selinexor Plus 7þ3 Daunorubicin/Cytarabine in Poor-Risk AML

Cytogenetic and molecular characteristics Table 1. Demographics. G-banding cytogenetic analyses were performed on the bone marrow aspirate samples of each patient at the time of diagnosis and, for Variable Total enrollment (N ¼ 21) most patients, at time of response assessment. Next-generation Sex, n (%) sequencing was performed at MCC (Tampa, FL) using the TruSight Male 14 (67) Myeloid-54 gene panel to assess for the presence of common myeloid Female 7 (33) mutations. Age (y; %) 60 18 (86) Pharmacokinetic and correlative studies 70 9 (43) – Pharmacokinetic studies Median (range) 69 (37 77) Race, no. (%) Pharmacokinetic analyses of selinexor were performed on plasma Caucasian 17 (81) samples from patients at hours 0, 1, 2, 4, and 8 after dose adminis- African American 4 (19) tration on days 1 and 15 of induction. At each time point, 2 mL of blood NCCN CMRS, n (%) was collected in an ethylenediaminetetraacetic acid tube, and plasma Intermediate 4 (19) samples were stored at 20 C or below until further analysis. High 17 (81) The samples were assayed using a validated LC/MS-MS by AIT sAML, n (%) 13 (62) Biosciences. The measurement range was 1 to 100 ng/mL. The overall Prior HMA, n (%) 8 (38) accuracy and precision of the quality control samples were both within Selinexor, n (%) 15% of the measurement. 60 mg 4 (19) 80 mg 17 (81) Pharmacokinetic parameters were calculated using a standard ECOG Performance Status, n (%) noncompartmental analysis approach. Protocol nominal sampling 0–1 19 (90) time was used to calculate and summarize pharmacokinetic data. 2 2 (10) Plasma samples that were below the lower limit of quantification before the first quantifiable concentration (i.e., predose samples) were Abbreviations: CMRS, Cytogenetic and Molecular Risk Stratification; HMA, set to zero to conduct pharmacokinetic analyses (17). hypomethylating agent; sAML, secondary AML.

Correlative studies Baseline ex vivo sensitivity to selinexor both alone and in combi- 69 years (range, 37–77). Fourteen patients (67%) were male and 7 nation with daunorubicin and cytarabine was assayed by flow cyto- (33%) were female. Nineteen (90%) patients had an ECOG perfor- þ metry on gated CD34 cells for apoptosis (active caspase-3). An mance status of 0 or 1 and 2 (10%) had an ECOG performance status of additional 20 mL of bone marrow aspirate and 20 mL of blood were 2. On the basis of NCCN cytogenetic and molecular risk stratification, collected from the screening bone marrow biopsies and aspirations for 17 (81%) patients were found to have poor-risk disease, and the exploratory studies. The bone marrow aspirates were processed by remaining 4 (19%) patients were determined to have intermediate- MCC Translational Research Core, and the blood samples were risk disease. These patients were eligible for enrollment as they were processed by MCC Tissue Core to obtain plasma and buffy coats; the 60 years of age or because they had secondary AML with diploid samples were then biobanked (3 or more aliquots each). The bone cytogenetics. Thirteen patients (62%) had secondary AML, 8 (62%) of marrow aspirate samples were enriched for mononuclear cells (MNCs) whom had received a hypomethylating agent before AML progression. using a Ficoll density gradient to evaluate the predictive value of Patients did not receive any AML-directed therapy (with the exception baseline ex vivo sensitivity testing with selinexor alone and in com- of hydroxyurea) before enrollment. All 21 patients received at least 1 bination with daunorubicin and cytarabine. For this assay, the bone dose of selinexor and were included in the safety evaluations and marrow MNCs were incubated for 20 hours in the presence of survival analyses. selinexor with or without daunorubicin or cytarabine. Three patients were removed from the study before the first A cytospin centrifuge was used to deposit 1.0 105 paraformal- response assessment. Two patients withdrew consent before complet- þ dehyde-fixed CD34 cells onto Shandon Double Cytoslides (Thermo ing the planned study treatment (days 3 and 14, respectively), and 1 Fisher Scientific) to perform immunofluorescent microscopy. Rabbit patient died from an adverse reaction to antibiotics on day 24. The 2 polyclonal antibodies (Novus Biologicals, NBP2-38322) were used to patients who voluntarily withdrew consent for reasons other than detect CD34; mouse monoclonal CRM1antibodies (C-1; Santa Cruz toxicity, before completing the study treatment, were not considered Biotechnology, SC74454) were used to detect XPO1; and mouse evaluable for response. Therefore, 19 patients were evaluable for monoclonal Ki-S1 antibodies (Millipore, MAB4197) were used to response. detect topoisomerase IIa. The fluorescent secondary antibodies used Nineteen patients received a full course of induction therapy. Six were anti-rabbit Alexa Fluor 488 and anti-mouse Alexa Fluor 594 patients required reinduction with selinexor plus 5þ2, and 6 patients þ (Invitrogen). Proteins within the nucleus and cytoplasm of the CD34 completed at least 1 cycle of consolidation. Only 1 patient entered the cells were analyzed using the Definiens Tissue Studio v4.4.2 software maintenance phase and completed 5 cycles of maintenance therapy. suite (Definiens). See Supplementary Methods for more details regard- Six patients underwent allogeneic stem cell transplants. ing the methods used. Dose escalation Four patients were enrolled in cohort 1. One patient was Results replaced after consent withdrawal prior to completion of 5 doses Patient demographics of selinexor. Four patients were also enrolled in cohort 2, with one Twenty-one patients were enrolled between June 2015 and 2016. being replaced due to unrelated toxicity prior to completing the Patient characteristics are shown in Table 1. The median age was DLT period.

AACRJournals.org Clin Cancer Res; 2019 OF3

Sweet et al.

Table 2. Nonhematologic adverse events occurring in 10% of tazobactam) acute tubular necrosis in the setting of febrile neutrope- patients during induction (N ¼ 21). nia. No other deaths occurred within 60 days of beginning induction. Thirteen (62%) patients died during the treatment or follow-up phases Grade, no. (%) Total, no. of the study. The majority of deaths were related to relapsed or Toxicity 1 2 3 4 (%) refractory AML. There were no deaths related to selinexor. Diarrhea 8 (38) 3 (14) 6 (29) — 17 (81) The median number of days in the hospital for all 21 patients was 37 – Hyponatremia 11 (52) — 6 (29) — 17 (81) (range, 3 82). Among the 10 patients who achieved CR/CRi, the Febrile neutropenia 1 (5) — 8 (38) 6 (29) 15 (71) median time to a platelet count 50,000 was 35 days (range, 25– Nausea 10 (48) 5 (24) ——15 (71) 77) and the median time to absolute neutrophil count (ANC) 500 Vomiting 6 (29) 1 (5) 1 (5) — 8 (38) was 26 days (range, 18–45). Anorexia 5 (24) 2 (10) ——7 (33) None of the 6 patients who completed at least 1 cycle of consol- Abdominal pain 6 (29) ———6 (29) idation experienced diarrhea or hyponatremia during this phase of — Fatigue 3 (14) 1 (5) 2 (10) 6 (29) treatment. The only grade 3/4 AEs during consolidation were dyspnea Constipation 4 (19) 1 (5) ——5 (24) (17%) and headache (17%). For the 1 patient who received mainte- Gastrointestinal 3 (14) ———3 (14) disorders nance therapy, the only grade 3 AE was fatigue, which prompted the Muscle weakness — 3 (14) ——3 (14) patient to voluntarily withdraw from the study while in remission. Oral mucositis 1 (5) 2 (10) ——3 (14) Rash 1 (5) 2 (10) ——3 (14) Efficacy Sepsis ——1 (5) 2 (10) 3 (14) All 21 patients were evaluated for survival. The 2 patients who withdrew consent did not undergo a formal response assessment and were, therefore, not included in the efficacy analysis. At a median DLTs did not occur in either dose-escalation cohort and the MTD follow-up of 28.9 months for all 21 patients, 8 (38%) patients were was not established. Further dose escalations did not occur because of alive, and the median OS was 10.3 months [95% confidence interval concerns of cerebellar toxicity observed at higher doses in concurrent (CI), 3.74–NR; Fig. 1A]. selinexor trials at other institutions (6). An additional 13 patients Eight patients (42%) achieved CR and 2 (11%) achieved CRi, received 80 mg of selinexor in a safety expansion phase to better resulting in an overall response rate of 53%. The median time to understand the safety profile of this drug combination. The RP2D was confirmation of CR/CRi for these 10 patients was 42 days (range, 31– 80 mg of selinexor taken orally twice weekly. 77). Nine patients (47%) were considered treatment failures due to refractory AML (Table 3). Of the 6 patients who received a second Adverse events induction, 1 (17%) achieved CRi. Among the 19 patients evaluable for All 21 patients were included in the adverse event (AE) assessment. response, the median follow-up time was 26.9 months, with a median Every patient experienced at least 1 TEAEs during the induction OS of 10.3 months. At 12 months, 47.4% of evaluable patients were period, and 20 (95%) had at least one grade 3/4 AEs. Table 2 lists alive (Fig. 1B). Response rates were less encouraging among those all the nonhematologic TEAEs observed in 10% of patients. The patients who had previously received a hypomethylating agent for an most common nonhematologic TEAEs (all grades) were diarrhea antecedent hematologic disorder, with only 13% of those patients (81%), hyponatremia (81%), which typically occurred between days achieving CR/CRi. Among the 17 patients who were considered poor- 8 and 21 of induction, nausea (71%), and febrile neutropenia (71%). risk on the basis of NCCN criteria, one was not eligible for response. The most common grade 3/4 nonhematologic TEAEs that occurred in Of the 16 response eligible patients in this subgroup, 8 (50%) achieved 10% of patients included febrile neutropenia (67%), diarrhea (29%), CR/CRi and 8 (50%) were treatment failures. Among the 4 patients hyponatremia (29%), and sepsis (14%). Diarrhea was managed with who were eligible for this study on the basis of being 60 years of age or antidiarrheal medications; nausea was controlled with antiemetics because they had secondary AML with diploid cytogenetics, 3 were (including prophylactic olanzapine in some cases); and hyponatremia eligible for response. Of those 3 patients, 2 (67%) achieved CR/CRi and was managed with high-salt diets, salt tablets, and intravenous fluids. 1 (33%) was a treatment failure. No patients experienced symptoms associated with hyponatremia. Among the 10 responding patients, the median OS was not reached, There were no cases of cardiac events, heart failure, decreased ejection and 6 of these patients were alive at the time of this report (September fraction, or arrhythmias. 2018; median follow-up 28.9 months). Six patients proceeded to One patient (4.8%) died within 30 days of beginning the study undergo an allogeneic stem cell transplant in first remission, 5 of treatment from antibiotic-related (vancomycin and piperacillin/ whom remain in remission. Among the 5 patients who relapsed, the

Figure 1. Overall survival. A, OS among entire cohort (n ¼ 21). Median OS was 10.3 months (95% CI, 3.74–NR). B, OS among 19 evaluable patients. Median OS was 10.3 months (95% CI, 3.74–NR).

OF4 Clin Cancer Res; 2019 CLINICAL CANCER RESEARCH

Selinexor Plus 7þ3 Daunorubicin/Cytarabine in Poor-Risk AML

Table 3. CR rates and cytogenetic data results.

CR rate results Evaluable patients, no. Age 60, no. Age 70, no. sAML, no. Prior HMA, no. (%; n ¼ 19) (%; n ¼ 16) (%; n ¼ 8) (%; n ¼ 12) (%; n ¼ 8)

CR þ CRi 10 (52.6) 9 (56) 3 (38) 4 (33) 1 (13) CR 8 (42.1) 7 (44) 3 (38) 3 (25) 1 (13) CRi 2 (10.5) 2 (13) 0 (0) 1 (8) 0 (0) Treatment failure 9 (47.4) 7 (44) 5 (6%) 8 (67) 7 (87)

Cytogenetic data results at baseline and follow-up in response-evaluable patients Chr Abnormalities Diploid Complex Chr 5 Chr 7 Other Chr Not assessed

Responders (n ¼ 10), no. Baseline 4 3 0 1 2 0 Follow-up 8 1 0 1 0 0 Treatment failure (n ¼ 9), no. Baseline 3 4 1 0 1 0 Follow-up 2 1 1 0 0 5

Abbreviations: Chr, chorosome; CR, complete remission; CRi, complete remission with incomplete blood count recovery; HMA, hypomethylating agent; sAML, secondary acute myeloid leukemia. median time to relapse was 9.2 months. The characteristics of the Correlative studies assays responding patients are detailed in Table 4. Flow cytometry was performed on bone marrow aspirates obtained A total of 16 patients evaluable for efficacy were treated at the RP2D from 10 patients to conduct ex vivo treatment studies. Of these, 2 of selinexor, 8 (50%) of whom achieved CR/CRi. None of the 16 samples were eliminated from the analysis. One sample had low þ patients required a dose reduction of selinexor. viability and the other had low representation CD34 cells (<5%). For the 8 remaining samples, the MNCs were plated at a concentration Pharmacokinetics of 3 million cells per mL. The cells were then treated with titrated Oral selinexor was absorbed at a moderate rate on day 1, with a concentrations of selinexor in combination with either 40 mmol/L T m median max of 2.0 hours at both 60 mg and 80 mg. For the 4 patients of cytarabine or 1 mol/L of daunorubicin. Selinexor concentrations C > m treated with 60 mg of selinexor, the mean max was 457 ng/mL 6.25 mol/L had an increased active caspase-3 compared with the m (1.0316 mol/L) and AUC0–8h was 1,853 ng/h/mL. For the 17 patients untreated controls. Adding cytarabine or daunorubicin to the titrated C who received 80 mg of selinexor, the mean max was 601 ng/mL selinexor resulted in increased apoptosis, above the levels observed in m (1.3567 mol/L) and AUC0–8h was 2,590 ng/h/mL (Table 5). The individual drug-treated cells. Contrary to preclinical observations, the pharmacokinetics of selinexor exhibited moderate to high variability effects of the combination appeared to be additive, and no synergy was (Table 5). noted between selinexor and daunorubicin. Considering the relatively high intersubject variability, exposure Immunofluorescent microscopy was used to evaluate topoisomer- (based on AUC) appeared to be similar on days 1 and 15. On day 15, ase IIa and XPO1 relative protein amounts and locations in 8 patients. T a delay was observed in the absorption of selinexor (mean max of These data were used to compare patients who had a complete 4 hours); as a result, the day 15 Cmax values were lower than those on response and those determined to be treatment failures, with 4 patients day 1 (Table 5). in each group. No significant difference was found between the complete response and treatment failure groups when comparing topoisomerase IIa protein concentrations (mean fluorescence; P ¼ Table 4. Demographics of the responding patients (n ¼ 10). 0.425) or nuclear/cytoplasmic ratios (P ¼ 0.444). Similarly, no significant difference was found between the XPO1 amount (P ¼ Age, no. (%) P ¼ 0.303) and nuclear/cytoplasmic ratios ( 0.330; refer to Supple- 60 9 (90) fi 70 3 (30) mentary Fig. S3). Lack of signi cance may be due to the small Median, y (range) 69 (58–77) sample size. NCCN CMRS, n (%) These ex vivo results did not provide data predictive of response to Poor-risk 8 (80) this treatment combination. Intermediate-risk 2 (20) Alive, n (%) 6 (60) Relapsed, n (%) 5 (50) Discussion n Underwent Allo-SCT, (%) 6 (60) In this trial, oral selinexor was safely combined with a standard 7þ3 Relapsed after Allo-SCT 1 (16) daunorubicin and cytarabine regimen for treatment of patients with sAML, n (%) 4 (40) Prior HMA therapy, n (%) 1 (10) newly diagnosed, poor-risk AML, resulting in an overall response rate of 53%. The median OS was 10.3 months, and the 12-month OS was Abbreviations: CMRS, cytogenetic and molecular risk stratification; HMA, hypo- 47.4%. Of note, the responses were durable, with 60% of responders methylating agent; sAML, secondary acute myeloid leukemia; Allo-SCT, alloge- being alive at a median follow-up of 28.9 months. The MTD was not neic stem cell transplant. reached, and the RP2D of selinexor in combination with daunorubicin

AACRJournals.org Clin Cancer Res; 2019 OF5

Sweet et al.

Table 5. Pharmacokinetic parameters.

Day 1 Day 15

Selinexor dose (mg) Patients, no. Cmax (ng/mL) Tmax (h) AUC0-8h (ngh/mL) Patients, no. Cmax (ng/mL) Tmax (h) AUC0–8h (ngh/mL)

60 4 457 196 2 3.2 1853 972 3 392 290 4 1.2 1714 1030 80 17 601 302 2 2.2 2590 1109 13 482 288 4 2.5 2095 1078

and cytarabine was 80 mg twice weekly. Furthermore, induction form of salt tabs or intravenous infusions of 0.9% sodium chloride. therapy with selinexor plus 7þ3 for patients with previously untreated, Nausea, vomiting, and anorexia were also commonly reported. In most poor-risk AML appears to be safe and tolerable. cases, these AEs were grade 1/2 and were successfully managed with Several studies have been conducted using selinexor either as a supportive measures, including the use of prophylactic olanzapine for single agent or in combination with chemotherapy to treat patients some patients. with AML. Garzon and colleagues (13) recently published data from a The noncompartmental analysis for pharmacokinetic studies indi- phase I dose-escalation trial of single-agent selinexor for patients with cated that the plasma concentration of selinexor was similar to plasma advanced hematologic malignancies. This trial enrolled 95 patients concentrations observed in previous studies (unpublished data). The with AML with a median age of 70 years. Of the 81 patients evaluable moderate intersubject variability (approximately 50%–70% CV) in C for response, 14% achieved a CR/CRi. The most common AEs exposure parameters (AUC, max) is probably due to the limited included fatigue, nausea, vomiting, anorexia, thrombocytopenia, and sampling schedule and differences in actual sampling time among hyponatremia. patients. No correlation was identified between the plasma concen- Given the molecular complexity of AML, the efficacy of selinexor tration of selinexor and overall response. may be greater when it is used in combination with other therapies. As The responses to selinexor and 7þ3 appeared durable in many such, many investigator-initiated trials have been designed combining cases, as evidenced by the fact that 5 of 10 responding patients selinexor with cytotoxic chemotherapy. In another phase I trial, 18 remained in CR/CRi at a median follow-up time of over 2 years; all pediatric patients with relapsed or refractory acute leukemia were 5 of these patients underwent allogeneic stem cell transplant in first treated with a combination of selinexor plus fludarabine and cytar- remission. abine. Fifteen of these patients were evaluable for response after In addition to demonstrating single-agent activity, selinexor has combination therapy, 47% of whom achieved CR/CRi, and 5 patients been shown to demonstrate promising activity in combination with tested negative for measurable residual disease (6). A larger, phase II chemotherapy. CPX-351 (liposomal daunorubicin plus cytarabine) trial of 42 adults with relapsed or refractory AML (many of whom has emerged as frontline therapy for secondary AML, with reported relapsed after a HSCT) combined selinexor (40 mg/m2, twice weekly) response rates similar to those reported in our current trial, among a with idarubicin (10 mg/m2, days 1, 3, and 5) and cytarabine similar group of patients (20). Therefore, a future trial of selinexor in (100 mg/m2, days 1–7; ref. 12). Two selinexor dose cohorts were combination with CPX-351 is likely warranted (20). Further emerging studied, and the CR/CRi rate ranged from 45% to 55%. The side-effect novel therapies (including venetoclax and other targeted agents) could profile of our trial and theirs demonstrated strong similarities. warrant novel combinations. The CR/CRi rate of 53% observed in our cohort of older poor-risk In summary, SINE compounds offer a unique mechanism to patients was encouraging. The CR/CRi rate (13%) was lowest among increase lethal DNA damage by retaining topoisomerase IIa in the the 8 patients who had previously been treated with hypomethylating nucleus in proximity to its DNA substrate and by localizing and agents for an antecedent hematologic disorder. Sixteen (84%) of the 19 activating TSPs in the nucleus. Although selinexor is reported to have patients who were evaluable for response after induction therapy had modest single-agent activity in AML, available data suggest that NCCN-defined poor-risk AML. Although 3 of the evaluable interme- selinexor in combination with other cytotoxic chemotherapy agents diate-risk patients had a clinical diagnosis of secondary AML, they leads to improved efficacy (6, 11–13). Our preclinical data suggest were classified as intermediate-risk per NCCN criteria. strong synergy between selinexor and daunorubicin. However, our The CR/CRi rates with 7þ3 alone that have been reported in other correlative studies, which were conducted using bone marrow samples trials with patient populations similar to ours range from 30% to from enrolled patients, could not reproduce these results. Further- 50% (18–20). A TP53 mutation was detected in 7 evaluable patients more, neither the baseline levels of topoisomerase IIa nor the nuclear/ at the time of diagnosis; this mutation typically correlates with a low cytoplasmic ratios of topoisomerase IIa or XPO1 appeared to correlate CR rate and an overall poor prognosis (18). Six of these patients with response to treatment. However, this analysis was limited to 8 also had a complex karyotype. In our study, the CR/CRi rate among patient samples. this very high-risk subset of patients was 43% with 7þ3 plus selinexor. Our data indicate that 80 mg of selinexor can safely be given in Encouragingly, the early death rate observed in our phase I study combination with a 7þ3 induction regimen in older adults with newly was low, comparing favorably with expected mortality rates among diagnosed poor-risk AML. Future randomized studies are needed to similar patient populations (20). The side-effect profile observed was definitively identify the benefit of selinexor as a component of induc- consistent with the well-established side effect profile, including tion therapy in AML. gastrointestinal toxicity, fatigue, hyponatremia, and cytopenias (13). Neither new safety events nor major organ toxicities were noted. The Disclosure of Potential Conflicts of Interest most frequently observed TEAEs were diarrhea and hyponatremia. K. Sweet reports receiving other commercial research support from Karyopharm Diarrhea was controlled with antidiarrheal medications. Hyponatre- Therapeutics. R. Komrokji is an employee/paid consultant for DSI, Pfizer, JAZZ, mia is a well-documented side effect of selinexor, with an unclear Celgene, Novartis and Agios, and reports receiving commercial research grants from etiology (13). It can be managed with sodium supplementation in the Celgene, and reports receiving speakers bureau honoraria from Jazz Pharmaceutical

OF6 Clin Cancer Res; 2019 CLINICAL CANCER RESEARCH

Selinexor Plus 7þ3 Daunorubicin/Cytarabine in Poor-Risk AML

and Novartis. D.M. Sullivan reports receiving commercial research grants from Administrative, technical, or material support (i.e., reporting or organizing data, Karyopharm Therapeutics. B.D. Shah reports receiving commercial research grants constructing databases): K. Sweet from Incyte and Jazz Pharma, and is an advisory board member/unpaid consultant for Study supervision: K. Sweet Kite/Gilead, Celgene/Juno, Novartis, Adaptive, AstraZeneca, Pharmacyclics, and Spectrum/Acrotech. J.E. Lancet is an employee/paid consultant for Jazz Pharmaceu- Acknowledgments fi fl ticals, Daiichi Sankyo, Agios Pharmaceuticals, and P zer. No potential con icts of We thank Paul Fletcher and Daley Drucker (H. Lee Moffitt Cancer Center and interest were disclosed by the other authors. Research Institution) for editorial assistance. They were not compensated beyond their regular salaries. This work has been supported in part by the Analytic Authors’ Contributions Microscopy Core Facility at the H. Lee Moffitt Cancer Center and Research Institute; Conception and design: K. Sweet, R. Komrokji, D.M. Sullivan, J.E. Lancet an NCI-designated Comprehensive Cancer Center (P30-CA076292). The clinical trial Development of methodology: K. Sweet, C.L. Cubitt, J.G. Turner was funded by Karyopharm. Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): K. Sweet, R. Komrokji, E. Padron, C.L. Cubitt, J.G. Turner, The costs of publication of this article were defrayed in part by the payment of page J.L. Dawson, J. Chavez, B.D. Shah, J.E. Lancet charges. This article must therefore be hereby marked advertisement in accordance Analysis and interpretation of data (e.g., statistical analysis, biostatistics, with 18 U.S.C. Section 1734 solely to indicate this fact. computational analysis): K. Sweet, C.L. Cubitt, J.G. Turner, J. Zhou, A.F. List, D.M. Sullivan Writing, review, and/or revision of the manuscript: K. Sweet, R. Komrokji, Received July 2, 2019; revised August 21, 2019; accepted October 1, 2019; C.L. Cubitt, A.F. List, D.A. Sallman, J.L. Dawson, D.M. Sullivan, B.D. Shah, J.E. Lancet published first October 21, 2019.

References 1. Etchin J, Montero J, Berezovskaya A, Le BT, Kentsis A, Christie AL, et al. Activity trone (Mito) for remission induction in acute myeloid leukemia (AML) is feasible of a selective inhibitor of nuclear export, selinexor (KPT-330), against AML- and tolerable. Blood 2016;128:212. initiating cells engrafted into immunosuppressed NSG mice. Leukemia 2016;30: 12. Fiedler W, Heuser M, Chromik J, Thol F, Bokemeyer C, Theile S, et al. Phase II 190–9. results of Ara-C and idarubicin in combination with the selective inhibitor of 2. Gravina GL, Senapedis W, McCauley D, Baloglu E, Shacham S, Festuccia C. nuclear export (SINE) compound selinexor (KPT-330) in patients with relapsed Nucleo-cytoplasmic transport as a therapeutic target of cancer. J Hematol Oncol or refractory AML. Blood 2016;128:341. 2014;7:85. 13. Garzon R, Savona M, Baz R, Andreeff M, Gabrail N, Gutierrez M, et al. A phase 1 3. Turner JG, Dawson J, Cubitt CL, Baz R, Sullivan DM. Inhibition of CRM1- clinical trial of single-agent selinexor in acute myeloid leukemia. Blood 2017;129: dependent nuclear export sensitizes malignant cells to cytotoxic and targeted 3165–74. agents. Semin Cancer Biol 2014;27:62–73. 14. National Comprehensive Cancer Network. Clinical practice guidelines in oncol- 4. Kojima K, Kornblau SM, Ruvolo V, Dilip A, Duvvuri S, Davis RE, et al. ogy chronic myeloid leukemia; 2017 (Version 1.2018). Available from: https:// Prognostic impact and targeting of CRM1 in acute myeloid leukemia. Blood www.nccn.org/professionals/physician_gls/pdf/cml.pdf. 2013;121:4166–74. 15. National Comprehensive Cancer Network. NCCN Guidelines Version 3.2015 5. Ranganathan P, Yu X, Na C, Santhanam R, Shacham S, Kauffman M, et al. acute myeloid leukemia; 2015. www.NCNN.org. Preclinical activity of a novel CRM1 inhibitor in acute myeloid leukemia. Blood 16. Cheson BD, Bennett JM, Kopecky KJ, Buchner T, Willman CL, Estey EH, et al. 2012;120:1765–73. Revised recommendations of the international working group for diagnosis, 6. Alexander TB, Lacayo NJ, Choi JK, Ribeiro RC, Pui CH, Rubnitz JE. Phase I study standardization of response criteria, treatment outcomes, and reporting stan- of selinexor, a selective inhibitor of nuclear export, in combination with dards for therapeutic trials in acute myeloid leukemia. J Clin Oncol 2003;21: ﬂudarabine and cytarabine, in pediatric relapsed or refractory acute leukemia. 4642–9. J Clin Oncol 2016;34:4094–101. 17. Chou TC. Theoretical basis, experimental design, and computerized simulation 7. Brunetti L, Gundry MC, Sorcini D, Guzman AG, Huang YH, Ramabadran R, of synergism and antagonism in drug combination studies. Pharmacol Rev 2006; et al. Mutant NPM1 maintains the leukemic state through HOX expression. 58:621–81. Cancer Cell 2018;34:499–512. 18. Granfeldt Ostgard LS, Medeiros BC, Sengelov H, Norgaard M, Andersen MK, 8. Gu X, Ebrahem Q, Mahfouz RZ, Hasipek M, Enane F, Radivoyevitch T, et al. Dufva IH, et al. Epidemiology and clinical signiﬁcance of secondary and therapy- Leukemogenic nucleophosmin mutation disrupts the transcription factor hub related acute myeloid leukemia: a National population-based cohort study. J Clin that regulates granulomonocytic fates. J Clin Invest 2018;128:4260–79. Oncol 2015;33:3641–9. 9.TurnerJG,DawsonJ,EmmonsMF,CubittCL,KauffmanM,ShachamS, 19. Hulegardh E, Nilsson C, Lazarevic V, Garelius H, Antunovic P, et al. CRM1 inhibition sensitizes drug resistant human myeloma cells to Rangert Derolf A, et al. Characterization and prognostic features of topoisomerase II and proteasome inhibitors both in vitro and ex vivo. secondary acute myeloid leukemia in a population-based setting: a J Cancer 2013;4:614–25. report from the Swedish Acute Leukemia Registry. Am J Hematol 10. Ranganathan P, Kashyap T, Yu X, Meng X, Lai TH, McNeil B, et al. XPO1 2015;90:208–14. inhibition using selinexor synergizes with chemotherapy in acute myeloid 20. Lancet JE, Uy GL, Cortes JE, Newell LF, Lin TL, Ritchie EK, et al. leukemia by targeting DNA repair and restoring topoisomerase IIalpha to the CPX-351 (cytarabine and daunorubicin) liposome for injection versus nucleus. Clin Cancer Res 2016;22:6142–52. conventional cytarabine plus daunorubicin in older patients with newly 11. Wang AY, Weiner HL, Green M, Larson RA, Odenike O, Artz A, et al. diagnosed secondary acute myeloid leukemia. J Clin Oncol 2018;36: Combination of selinexor with high-dose cytarabine (HiDAC) and mitoxan- 2684–92.

AACRJournals.org Clin Cancer Res; 2019 OF7

Phase I Clinical Trial of Selinexor in Combination with Daunorubicin and Cytarabine in Previously Untreated Poor-Risk Acute Myeloid Leukemia

Kendra Sweet, Rami Komrokji, Eric Padron, et al.

Clin Cancer Res Published OnlineFirst October 21, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-19-2169

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/10/19/1078-0432.CCR-19-2169.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2019/11/19/1078-0432.CCR-19-2169. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.