Safety and efficacy of vismodegib in relapsed/refractory acute myeloid leukaemia: results of a phase Ib trial

Acute myeloid leukaemia (AML) is the most common acute leukaemia affecting adults (Visser et al, 2012). Leukaemia stem cells (LSCs) have been shown to perpetuate AML (Bonnet & Dick, 1997), and Hedgehog signalling is involved in the maintenance of LSCs (Fukushima et al, 2016). Increased expression of the Hedgehog signalling target transcription factors GLI1 and GLI2 is associated with poor survival rate in AML and is correlated with the FLT3 mutation (Wellbrock et al, 2015). Results of a preclinical study showed that an inhibitor of the receptor Smoothened (SMO) could attenuate the leukaemia-initiating potential of AML cells and sensitise them to cytotoxic chemotherapy with cytarabine (Fukushima et al, 2016). In a phase I dose-finding study of glasdegib (PF-04449913) in patients with myeloid malignancies, one of 28 patients with AML experienced complete remission with incomplete blood count recovery, and four experienced partial remission with incomplete blood count recovery (Martinelli et al, 2015). Vismodegib is a Hedgehog pathway inhibitor (HPI) that binds to and inhibits SMO (Sekulic et al, 2012) and is currently approved for treatment of adults with metastatic or locally advanced basal cell carcinoma inappropriate for surgery or radiotherapy. We report here the results of a completed single-arm, open-label phase Ib study conducted to evaluate the safety and efficacy of vismodegib in adult patients with relapsed/refractory AML without prior HPI treatment. All patients recruited had relapsed/refractory AML; patients with acute promyelocytic leukaemia (M3 subtype) were excluded. Patients were stratified by poor-risk cytogenetics (per National Comprehensive Cancer Network guidelines [NCCN Guidelines®] for AML; O'Donnell et al, 2017), FLT3 mutations with intermediate-risk cytogenetics, or neither. Patients received vismodegib 150 mg orally, once daily until disease progression, intolerable toxicity or withdrawal of consent. Treatment interruption up to 14 days (before Week 8) or up to 4 weeks (after Week 8) was permitted to manage adverse events (AEs). A total of 47 patients were screened for enrolment from 12 study sites in Germany, Canada and the United States, and 38 received at least one dose of vismodegib between September 2013 and September 2014, when the study was terminated by the sponsor because of lack of efficacy. Table 1 shows the baseline characteristics of the overall patient population by prognostic factor subgroup. All enrolled patients had received prior cancer therapy; most had received more than one therapy, including hypomethylating agents, immunomodulators and targeted signalling pathway inhibitors. At the end of the study (data cut-off 3 November 2014), all 38 patients had discontinued treatment with vismodegib, 19 patients (50·0%) because of progressive disease (Supplementary Fig 1). The median follow-up was 113·9 days (range 6–337 days). Vismodegib treatment produced minimal clinical efficacy in patients with relapsed or refractory AML (Table 2). Overall response rate comprised complete response (CR), CR with incomplete blood count recovery (CRi) or partial response (PR), and was 6·1% in the efficacy population. One patient experienced CRi at Day 57 (duration of response, 13 weeks) and one patient experienced PR at Day 26 (duration of response, 6 weeks); neither had poor-risk cytogenetics or FLT3-positive disease. The median duration of overall survival in all treated patients was 3·4 months (95% confidence interval [CI] 2·3–3·9); 33 patients died during the study, mainly because of disease progression (27 deaths). No deaths were considered related to vismodegib. In general, treatment with vismodegib was well-tolerated and most patients achieved a high cumulative dose intensity (median 99·6%; Table SI). Overall, 97·4% of patients experienced at least one AE on study (Table S2). The most common AEs were pyrexia, nausea, dysgeusia and epistaxis (Table S3). Alopecia, frequently associated with HPIs, occurred in 7·9% (three of 38) of patients overall. Grade ≥3 AEs were reported in 84·2% of patients overall. The most common of these were febrile neutropenia (21·1%), anaemia (15·8%), thrombocytopenia (15·8%), hyponatraemia (10·5%), hypoxia (10·5%), lung infection (10·5%) and pyrexia (10·5%). Serious AEs (SAEs) were reported in 26 (68·4%) patients (Table S4). Five patients (13·2%) experienced a total of six SAEs that were considered related to vismodegib: grade 3 pyrexia (two cases), grade 2 oesophageal pain and grade 4 mucosal inflammation (both in one patient), ongoing grade 1 abdominal pain and grade 3 nausea (one case each). The AE profile was probably driven by patients’ underlying disease. The most common AEs are typical of patients with AML or are consistent with those seen in studies of vismodegib in patients with advanced basal cell carcinoma (BCC) (Lacouture et al, 2016). The most common SAEs and grade ≥3 AEs were infection and blood disorders, both of which are common in patients with relapsed or refractory AML, as is pyrexia. No new vismodegib safety signals were identified in this study. In summary, vismodegib as a single agent produced minimal clinical efficacy in patients with relapsed/refractory AML, similar to other preliminary studies of HPI treatment in patients with myeloid malignancies with abnormal Hedgehog pathway signalling (Martinelli et al, 2015; Tibes et al, 2015). Single-drug treatment is unlikely to produce significant durable responses in the absence of chemotherapy to clear the bulk leukaemia cell population (Pollyea et al, 2014). In the current study, detection of biomarkers of LSC response (e.g., changes in expression of Hedgehog ligands or target genes, such as GLI1 [Martinelli et al, 2015], was not possible; therefore, an effect of vismodegib on LSCs cannot be ruled out. The two patients who responded were not FLT3 mutation positive. FLT3 mutation is associated with increased GLI1 and GLI2 expression and poor survival outcomes (Wellbrock et al, 2015). If the constitutive receptor tyrosine kinase activity of mutant FLT3 induces GLI transcription factor expression through cross-talk between signalling pathways, inhibiting Hedgehog signalling through SMO is unlikely to have an effect. Further investigation of combined targeted therapies may be warranted. For future studies, combination treatments that target the bulk leukaemia population and LSC-targeted therapy, such as HPIs, are more likely to show clinical benefit than LSC-targeted agents alone. Given the role of Hedgehog signalling in maintaining LSCs, HPIs may be suited to maintenance therapy after induction of a response by chemotherapy. This study was funded by F. Hoffman-La Roche, Ltd. Medical writing support was provided by Lucy Smithers, PhD (ApotheCom, London, UK), which was funded by F. Hoffmann-La Roche Ltd. DC and BS participated in the study design. DB, RN, TLL, JC, JK, KY, BCM, AK, SA, WF, ND, BS, TR and DC participated in the collection, interpretation and analysis of data, and the development and review of the manuscript draft. All authors approved the final manuscript for submission. JK has received research funding from F. Hoffman-La Roche Ltd. TLL has received research funding from F. Hoffman-La Roche Ltd and Pfizer Inc. and honoraria from Jazz Pharmaceuticals. JC has received research funding from F. Hoffman-La Roche Ltd, Amphivena Therapeutics Inc., Astellas Pharma, Daiichi Sankyo, Janssen Pharmaceutica, Jazz Pharmaceuticals, Novartis and Pfizer Inc. and consultancy fees from Amphivena Therapeutics Inc., Astellas Pharma, Daiichi Sankyo, Janssen Pharmaceutica, Jazz Pharmaceuticals, Novartis and Pfizer Inc. KY has received research funding from Agensys, Inc., Astex Pharmaceuticals, GlaxoSmithKline, F. Hoffman-La Roche Ltd and Oncoethix SA, and has served on advisory boards for Celgene, Novartis, Pfizer Inc. and Tolero Pharmaceuticals, Inc. BCM has received research support and honoraria from F. Hoffman-La Roche Ltd. SA has served on advisory boards for Ariad Pharmaceuticals, Inc./Incyte Corporation, Bristol-Myers Squibb, Novartis and Pfizer Inc. and has received research funding from Novartis and F. Hoffman-La Roche Ltd. WF has received research funding from Amgen and Pfizer Inc.; royalties from Amgen; meeting and travel expenses from Amgen, Gilead Sciences, Jazz Pharmaceuticals and Teva Pharmaceutical Industries; and has served on advisory boards for Amgen, Ariad Pharmaceuticals, Inc./Incyte Corporation, Pfizer Inc. and Jazz Pharmaceuticals. BS, TR and DC are employees of Genentech Inc., and TR holds stock in F. Hoffmann-La Roche/Genentech, Inc. ND is an employee of Roche Products Ltd. DB, RN and AK have no conflicts of interest to disclose. 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.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.