Abstract

Background: In a previously reported Phase 2 randomized study of patients with acute myeloid leukemia (AML), addition of the investigational agent glasdegib (PF-04449913) to low-dose cytarabine (LDAC) improved overall survival (OS) when compared with LDAC alone. In a non-randomized study arm, glasdegib together with 7+3 chemotherapy was well tolerated and associated with clinical activity. We used a comprehensive biomarker analysis, evaluating gene expression, circulating cytokine levels, and gene mutations, to identify molecular drivers that predict overall response (OR) and OS.

Methods: In this Phase 2 multicenter study (NCT01546038), patients with AML who were suitable for non-intensive therapy were randomized (2:1) to LDAC + glasdegib 100 mg QD or LDAC alone, and patients suitable for intensive therapy were assigned 7+3 plus glasdegib 100 mg QD. Whole blood, serum, and bone marrow aspirate samples were collected at baseline, and used to assess 19 genes for expression analysis, 38 analytes for circulating cytokine levels, and 109 genes for mutation analysis. Gene expression was analyzed using TaqMan Low Density Array Cards (TLDCs), cytokine levels were analyzed using quantitative, multiplexed immunoassays (Myriad RBM), and mutation analysis was performed using the Illumina® MiSeq instrument (San Diego, CA). All correlations were performed either for OS or for OR. For gene expression and cytokine analysis, a cut-off value above or below the median expression level for each treatment arm was used to separate samples into two subgroups (< or ≥ the median value) to explore the relationship of expression levels with OS data. Criteria for significance in the non-intensive cohort required one subgroup to have a p-value of <0.05 in the between-treatment arms comparison and the HR difference between the two subgroups to be ≥2 fold. Responses were defined as patients with a complete remission (CR), CR with incomplete blood count recovery (CRi), morphologic leukemia-free state, partial remission (PR), or PRi. For response correlations, genes or cytokines were considered to be differentially expressed if they had a p-value <0.05 and were differentially expressed by ≥2-fold.

Results: Within the non-intensive arm (LDAC + glasdegib, n=68; LDAC alone, n=30), expression levels of several genes correlated with improved OS with glasdegib plus LDAC. Lower levels of expression of FOXM1 and MSI2, and higher expression levels of BCL2 and CCND2 correlated with improved OS with the combination. Additionally, lower levels of the cytokines 6CKINE (CCL21), ICAM-1, MIP-1α, and MMP-3 correlated with improved OS. An analysis of correlations of gene expression and cytokine levels with OR could not be completed due to the low number of responders in the LDAC only group (n=2). In the intensive treatment arm (glasdegib and 7+3, n=59), higher PTCH1 expression correlated with improved OS (p=0.0219, median OS 10.8 versus 39.5 months). In this cohort, lower levels of IL-8 (p=0.0225) and MIP-3β (p=0.0403) correlated with lower OS. Expression levels of no genes or cytokines significantly correlated with OR in this arm. We also examined correlations between gene mutation status and OS in both study arms. In the non-intensive arm (LDAC + glasdegib, n=58; LDAC alone, n=25), no genes mutated in at least 5 patients correlated with OS. In the intensive treatment arm (n=47), mutations in FLT3, TP53, CEP170, NPM1, and ANKRD26 correlated with OS (all p<0.05). Patients in this arm with FLT3 mutations responded better than patients with wild type FLT3 (p=0.0336, median OS of 13.1 months versus unreached for FLT3 mutant).

Conclusions: In this biomarker analysis, we found that expression levels of a select number of genes and circulating cytokines implicated in AML correlated with OS in the non-intensive and the intensive arms. The improved response for patients with FLT3 mutations and high PTCH1 expression levels in the intensive arm deserves further investigation. These findings need to be verified in larger controlled studies, which are ongoing.

Disclosures

Cortes:Novartis: Consultancy, Research Funding; Daiichi Sankyo: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; Astellas Pharma: Consultancy, Research Funding; Arog: Research Funding. Pollyea:Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Consultancy; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Research Funding; Curis: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Heuser:Astellas: Research Funding; Daiichi Sankyo: Research Funding; Sunesis: Research Funding; Tetralogic: Research Funding; Bayer Pharma AG: Consultancy, Research Funding; StemLine Therapeutics: Consultancy; Janssen: Consultancy; Pfizer: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Honoraria, Research Funding; BergenBio: Research Funding; Karyopharm: Research Funding. Chan:Pfizer: Employment, Equity Ownership. Wang:Pfizer: Employment, Equity Ownership. Ching:Pfizer Inc: Employment, Equity Ownership. Johnson:Pfizer Inc: Employment, Equity Ownership. O'Brien:Pfizer Inc: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.