BI-811283

BI 811283, an adenosine triphosphate-competitive, reversible and potent inhibitor of Aurora B kinase, has demonstrated anti-tumour activity in acute myeloid leukaemia (AML) cell lines (Gürtler et al, 2010a,b; Tontsch-Grunt et al, 2010; Mross et al, 2016). Here we report a phase I trial that evaluated the maximum tolerated dose (MTD), safety, efficacy and pharmacokinetics (PK) of BI 811283 with low-dose cytarabine (LDAC) in AML patients considered ineligible for intensive treatment. Patients with previously untreated (except with hydroxycarbamide) or relapsed/refractory AML (excluding acute promyelocytic leukaemia) considered unsuitable for intensive induction or salvage therapy were randomised to one of two schedules, combining BI 811283 (24-h intravenous infusion; 4-week cycle) with LDAC (20 mg twice daily subcutaneously; days 1–10 of a 4-week cycle). In Schedule A, patients received BI 811283 on days 1 and 15. In Schedule B, patients received BI 811283 on day 1. BI 811283 dose escalation followed a 3 + 3 design (BI 811283 starting dose: 5 mg). The primary endpoint was to determine the MTD of the two BI 811283 schedules in combination with LDAC based on the incidence of dose-limiting toxicities (DLTs). Further endpoints included: incidence and intensity of adverse events (AEs), response and PK of cytarabine with BI 811283. Full methodological details are provided in Appendix S1. All patients provided written informed consent and the trial was registered with ClinicalTrials.gov: NCT00632749. Of 68 randomised patients, 64 were treated (Schedule A, n = 28; Schedule B, n = 36). At baseline (Table SI), 11 (17·2%), 37 (57·8%) and 16 (25·0%) patients had an Eastern Cooperative Oncology Group (ECOG) performance score of 0, 1 and 2, respectively. Median age was 73 (range, 49–89) years. Forty-four patients (68·8%) were previously untreated for AML and 20 (31·3%) patients had relapsed/refractory disease. Twenty-two patients (34·4%) had adverse-risk genetics, two patients (3·1%) had favourable-risk genetics, and 20 (31·3%) and six patients (9·4%) had intermediate-risk I and II genetics, respectively. Dose escalations and DLTs are shown in Table 1. In Schedule A, DLTs were seen in the 120 mg dose group; the BI 811283 dose was therefore de-escalated to 100 mg and a further three patients were enrolled. As no DLTs were seen at the 100 mg dose level, the MTD in Schedule A was determined to be 100 mg. In Schedule B, the BI 811283 dose was escalated to 420 mg without the MTD being reached. Due to a strategic decision by the sponsor to halt development of BI 811283, recruitment was stopped at the 420 mg dose. All treated patients had at least one AE during study participation. The most common AEs reported in Schedule A were anaemia (53·6%), nausea (50·0%) and pyrexia, febrile neutropenia and leucopenia (each 39·3%), and in Schedule B were pyrexia (52·8%), thrombocytopenia (50·0%) and anaemia and nausea (each 47·2%). Grade ≥3 AEs occurring in >10% of patients in a treatment group are shown by system organ class in Table 2 and incidence of AEs by dose in Table SII. Overall, 82·8% of patients had AEs that were considered to be drug-related by the investigators. AEs that led to discontinuation were reported in 39·3% of patients in Schedule A and 25·0% of patients in Schedule B. A total of 26 patients (40·6%) had fatal AEs [10 patients (35·7%) in Schedule A and 16 (44·4%) in Schedule B], mainly due to infections and AML progression; only one death (neutropenic sepsis in an 82-year-old female) was considered related to trial medication. The median number of courses completed was 1 (range, 0–3) with Schedule A and 1 (range, 0–15) with Schedule B. The median number of courses was 2 (range, 1–15) for patients with previously untreated AML and 1 (range, 1–6) for patients with relapsed/refractory AML. Best overall response is presented in Table SIII. In Schedule A, two patients (7·1%) achieved a complete remission (CR); 10 patients (35·7%) had no change as their best response, 11 (39·3%) had progressive disease (PD) and five patients (17·9%) had no response assessment available. In Schedule B, five patients (13·9%) achieved a CR. Additionally, one patient (2·8%) had a CR with incomplete haematological recovery (CRi) and one patient (2·8%) had a partial remission (PR). Seventeen remaining patients (47·2%) had no change, eight patients (22·2%) had PD and four patients (11·1%) had no response assessment available. In Schedule A, both patients who achieved CR had previously untreated AML, and in Schedule B, all but one patient who achieved CR, CRi or PR had previously untreated AML. No dose-related trend in response rate was observed and responses occurred after a median of 1 cycle of therapy (range, 1–7). For the eight patients overall who had a best response of CR or CRi, the median remission duration was 263 days (range, 15–1492). BI 811283 plasma PK profiles suggested no effect of the dosing schedule on the PK of BI 811283, and in both treatment schedules, suggested close to dose-proportional PK (Figure S1). Cytarabine plasma concentration profiles were similar in each BI 811283 dose group (Figure S2) and the PK parameters of cytarabine did not vary with BI 811283 dose (data not shown), indicating that cytarabine PK was not influenced by BI 811283. In summary, this phase I trial demonstrated an acceptable safety profile with BI 811283 administered according to two different treatment schedules in combination with LDAC. Other Aurora kinase inhibitors, including the Aurora B kinase inhibitor barasertib, have demonstrated encouraging activity as both monotherapy and in combination, in phase I/II trials in patients with AML (Bavetsias & Linardopoulos, 2015). However, the results of this study do not indicate improved anti-leukaemic activity of BI 811283 in combination with LDAC compared with historical results for LDAC alone. The clinical development of BI 811283 was stopped for strategic reasons, but it has been hypothesized that Aurora kinase inhibition in combination with inhibition of other key oncogenic drivers in AML (e.g., FMS-like tyrosine kinase receptor-3) may provide improved outcomes and impede tumour resistance. Therefore, further clinical exploration of Aurora B kinase as a therapeutic target in AML might be warranted. This study was funded by Boehringer Ingelheim. Medical writing assistance, supported financially by Boehringer Ingelheim, was provided by Laura Winton of GeoMed, an Ashfield company, part of UDG Healthcare plc, during the preparation of this manuscript. HD designed the study, analysed and interpreted the data, and provided study materials and patient data. CM-T, ML, WF, AK, JW, GB, RFS and OGO collected and assembled data. UK collected and assembled data, and provided study materials and patient data. RGG designed the study, assembled data, and analysed and interpreted the data. JH and TT designed the study, and analysed and interpreted the data. All authors contributed towards the writing and approval of the manuscript for submission. HD has received honoraria for consultation from: AbbVie, Agios, Amgen, Astex Pharmaceuticals, Celator, Celgene, Janssen, Jazz, Novartis, Seattle Genetics and Sunesis; and research support from: AROG Pharmaceuticals, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Novartis, Pfizer and Sunesis. WF has received royalties, travel expenses and research funding, and acted in advisory and consulting roles for Amgen; travel expenses from Daiichi Sankyo, Gilead, GSO, Jazz Pharma and Teva; research funding from Pfizer; and has acted in an advisory role for Ariad/Incyte, Jazz Pharma, Novartis and Pfizer. AK has received support for work outside the submitted work from Bayer, Merck and Teva. JW has received honoraria, research funding and travel expenses from Amgen, Bristol-Myers Squibb, Celgene, Daiichi Sankyo and Novartis. GB has received honoraria from Boehringer Ingelheim. RFS has received support for work outside the submitted work from Boehringer Ingelheim, Celgene, Novartis, Pfizer and Teva. UK reports receiving research funding from Boehringer Ingelheim; lecture fees from Celgene and Jazz Pharma; and has participated in advisory boards for Boehringer Ingelheim, Celgene, Gilead Sciences and Roche. R-GG and TT are employees of Boehringer Ingelheim. JH was an employee of Boehringer Ingelheim at the time of study conduct and is currently an employee of Applied Biomath, LLC. All other authors declare no potential conflicts of interest. Appendix S1. Methods. Table SI. Patient demographic and baseline disease characteristics. Table SII. Incidence of AEs by dose. Table SIII. Best overall response. Figure S1. Geometric mean plasma concentration–time profiles of BI 811283 during and after the 24-h infusion of BI 811283. (A) Day 1, cycle 1, Schedule A. (B) Day 15, cycle 1, Schedule A. (C) Day 1, cycle 1, Schedule B. Figure S2. Geometric mean plasma concentration–time profiles of cytarabine. (A) Schedule A. (B) Schedule B. 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.