A Phase II study of selinexor plus cytarabine and idarubicin in patients with relapsed/refractory acute myeloid leukaemia

Abstract

Despite treatment advances with new targeted therapies, the majority of patients with acute myeloid leukaemia (AML) will relapse after primary therapy and have a dismal prognosis.1 The best treatment options for patients with relapsed/refractory (R/R) AML to date are cytarabine-based regimens2 with a clinical need for new treatment options. Blocking nuclear export of growth regulatory proteins enhances tumour control, as their anti-neoplastic action requires nuclear localisation. Furthermore, exportin 1 (XPO1) is overexpressed in many cancer types, including AML.3, 4 Selinexor is a selective inhibitor of nuclear export (SINE) specifically blocking XPO1.5 It re-localises topoisomerase IIα to the nucleus resulting in synergism with Topo IIα inhibitors, idarubicin or mitoxantrone.3 Selinexor decreases expression and translation of DNA repair genes RAD51 and checkpoint kinase 1 (Chk1) limiting the repair of double-strand breaks.3 Given the in vitro synergy and promising findings in a Phase I clinical trial,4 we performed a clinical study with selinexor, cytarabine and idarubicin in patients with R/R AML. Patients aged ≥18 years with R/R AML and suitable for R/R standard-dose chemotherapy (‘7 + 3’), with adequate liver, renal and cardiac function were eligible. The main exclusion criteria were: cumulative anthracycline dose (daunorubicin or equivalent) >360 mg/m2, need for immunosuppressive drugs, and presence of central nervous system leukaemia. For additional eligibility criteria and cytogenetic and molecular characteristics refer to the supplements. This multicentre, open-label, non-randomised Phase II trial was conducted at three sites in Germany (ClinicalTrials.gov Identifier: NCT02249091). Patients received standard chemotherapy (‘7 + 3’, continuous infusion of cytarabine 100 mg/m2 on days 1–7 and idarubicin 10 mg/m2 intravenously on days 1, 3 and 5). The primary objective was to determine the efficacy of selinexor in combination with standard chemotherapy in patients with R/R AML, as determined by the rate of complete remission (CR) and morphological CR with incomplete blood count recovery (CRi). Adverse events (AEs) and safety laboratory assessments were monitored from the start of treatment until 30 days after the last dose of study medication. Patients achieving CR or CRi were recommended for stem-cell transplantation (SCT). A total of 42 patients representing a high-risk cohort (see Data S1) were included. Patients demographic and baseline characteristics are listed in Table SI. There were two cohorts: 27 patients received selinexor 40 mg/m2 orally twice-weekly for 4 weeks (Cohort 1). Due to prolonged aplasia, high rates of febrile neutropenia and severe diarrhoea were seen, 15 additional patients were subsequently enrolled receiving 60 mg selinexor absolute twice-weekly for 3 weeks of a 4-week cycle (Cohort 2). In all, 20 patients achieved CR/CRi (including one patient who needed two induction cycles), resulting in an overall response rate (ORR) of 47·6%[95% confidence interval (CI) 33·4–62·3%; for Cohort 1, 55·6% (95% CI 40·0–70·1%); and for Cohort 2, 33·3% (95% CI 15·2–58·3%)] The median (range) time until CR/CRi was 34 (26–85) days (see Table SII and for the subgroup analysis Table SIII). The intention of salvage therapy was to bring patients to SCT. Within the study, allogeneic SCT was performed in 15 of 42 (35·7%) patients. In addition, two previously transplanted patients received donor lymphocyte infusions as consolidation. Of the 15 transplanted patients, six died during follow-up after a median (range) of 106 (11–191) days (two due to progressive disease, two due to graft-versus-host disease, one due to multi-organ failure and one due to pneumonia in aplasia after allogeneic SCT). The median follow-up for the whole group was 8·2 months [median (range) 12·6 (4·2–38·4) months for Cohort 1, and 8·0 (0·5–16·1) months for Cohort 2). The median overall survival (OS) was 8·2 months (12·6 months for Cohort 1 and 8·0 months for Cohort 2). The difference was not statistically significant (P = 0·47) (Fig 1). Predominant AEs in all 42 patients were gastrointestinal and haematological AEs (Table IA). Dose reduction improved vomiting (all grades), Grade 3/4 diarrhoea and febrile neutropenia (Table IB). Early death before the end of the first 4-week induction cycle occurred in four patients (9·5%, three unrelated). During the follow-up period, 21 additional patients died mainly from underlying disease (Data S1). Cohort 1 Selinexor 40 mg/m2 (N = 27) Cohort 2 Selinexor 60 mg flat dose (N = 15) Total (N = 42) We evaluated treatment with selinexor cytarabine and idarubicin in heavily pre-treated patients with AML. To minimise the toxicity of this regimen and to better evaluate the effect of selinexor, a low cytarabine dose was employed. The chosen ‘7 + 3’ protocol also had the advantage that data on tolerability could be obtained from when the regimen was used as front-line therapy in newly diagnosed patients. The lower dose of Cohort 2 is the recommended dose for Phase III studies due to a better safety profile and we focus here on this cohort. The ORR of 47·6% for the whole cohort lies in the upper range of published studies.1, 2 With the reduced dose regimen, we observed a lower response rate (RR, 33·3% vs. 56%). This might be due to the four unexpected early deaths in Cohort 2 and is still above the predefined limit of 30% for this study. Nevertheless, the difference between the cohorts in event-free survival (EFS), relapse-free survival (RFS) and OS was not statistically significant (EFS: P = 0·87, RFS: P = 0·28, OS: P = 0·47, Fig 1). Although our present results showed a higher median RFS (17·7 months) compared to other published clinical trials with cytarabine-containing regimens (for Cohort 2 the median RFS was not reached at the time of evaluation), the OS of 8·0 months (Cohort 2) in our present study lay within the range of other clinical trials (4·9–9·8 months).2 In all, 35·7% of the patients proceeded to SCT. One reason for non-transplantation might be related to the patient’s age; six of the 12 patients with remission who did not receive a SCT were aged ≥70 years. The most frequent molecular aberration in patients with AML is mutation of the nucleophosmin 1 (NPM1) gene.7 This mutation causes the relocation of the phosphoprotein NPM1 from the nucleus to the cytoplasm, resulting in increased proliferation of myeloid cells in a preclinical zebrafish model.7, 8 Remarkably, in our present study, three of the four patients with NPM1-mutation achieved a CR. This confirms previous findings that confinement of NPM1 to the nucleus by inhibition of nuclear export might be beneficial to patients with NPM1-gene mutation.9 The dose reduction in our present study made the regimen more tolerable. The side-effect profile of Cohort 2 was comparable to the one reported by a Phase 1 study by Wang et al.6 who treated patients with a combination of selinexor, high-dose cytarabine, and mitoxantrone. Selinexor administered as a single-agent to patients with R/R AML showed a similar dose-dependent safety profile.4 Taken together, addition of selinexor at a dose of 60 mg to chemotherapy is feasible in AML. In conclusion, selinexor, cytarabine and idarubicin result in a high remission rate in patients with R/R AML. Limitations of our present study include the small sample size and the short duration of follow-up, as well as that data were censored at SCT. The results should be substantiated in a randomised Phase III study with the recommended dose of 60 mg of selinexor twice a week for 3 weeks of a 4-week cycle. The authors thank the patients who participated in this trial and their families; the co-investigators, study nurses and study co-ordinators at each of the sites; Karyopharm Therapeutics Inc., Newton, MA, USA for supplying Selinexor and provided financial support for the conduct of the study; Oy4Pharma Ltd., Turku, Finland, for the statistical analysis. GSO Global Clinical Research B.V., Amsterdam, The Netherlands, sponsored this study and worked with investigators to design the study, as well as to collect, analyze, and interpret the data. The authors contributed as following: Walter Fiedler (X,Y,Z, A,B), Joerg Chromik (X,Z,A), Stefanie Amberg (B), Maxim Kebenko (X,Z), FelicitasThol (X,Z), Vera Schlipfenbacher (X,Z), Anne Christine Wilke (X,Z), Franziska Modemann (X,Z), Melanie Janning (X,Z), Hubert Serve (X,Z,A), Arnold Ganser (X,Z,A), Carsten Bokemeyer (X,Z,A), Susann Theile (B), Ute Deppermann (B), Anne L. Kranich (Y,X,A,B), Michael Heuser (X,Y,Z,A). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts