Thomson, Blythe

Abstract Introduction: Rearrangements of the MLL (mixed lineage leukemia) gene on chromosome 11q23 are present in 5-10% of either acute myeloid (AML) or acute lymphoblastic leukemia (ALL), and are associated with a poor prognosis. Fusion proteins involving the MLL histone methyltransferase (HMT) result in recruitment of another HMT, DOT1L, to a multi-protein complex leading to aberrant methylation of Histone H3 lysine 79 (H3K79) at MLL target genes and enhanced expression of leukemogenic genes such as HOXA9 and MEIS1. EPZ-5676 is a small molecule inhibitor of DOT1L with sub-nanomolar affinity and >37,000 fold selectivity against additional HMTs. EPZ-5676 treatment of cultured cells reduced histone H3K79 methylation, decreased MLL target gene expression and resulted in selective killing of MLL-rearranged (MLL-r) leukemia cells by inducing apoptosis and differentiation. EPZ-5676 administration led to tumor regression in an MLL-r leukemia xenograft model. We report preliminary results of an ongoing Phase 1 trial of EPZ-5676 in patients with advanced hematologic malignancies, including acute leukemia with MLL-r. Methods: This is a 2-part phase 1 open label dose escalation study investigating the safety, tolerability, pharmacokinetics (PK), pharmacodynamics (PD) and anti-leukemic activity of EPZ-5676 in adult patients (pts) with relapsed/refractory leukemia (CT.gov: NCT01684150). In the dose-escalation phase, patients with AML, ALL, acute mixed lineage leukemia, myelodysplastic syndrome, myeloproliferative neoplasia (MPN) or chronic myeloid leukemia were eligible. Eligibility in the expansion phase of the study was restricted to pts with MLL-r or partial tandem duplication of MLL (MLL-PTD). EPZ-5676 was administered via continuous intravenous infusion (CIV) for 21 days of a 28 day cycle in the dose escalation phase and CIV for all 28 days of a 28 day cycle in the expansion phase, until disease progression or unacceptable toxicity. Results: Between October 2012 and July 2014, 37 pts have enrolled; 36 are evaluable for safety (received at least 1 dose of EPZ-5676) and 28 pts (19 MLL-r, 4 MLL-PTD) are evaluable for anti-leukemia activity (completed at least one post baseline marrow evaluation). The median age at time of enrollment was 53 years (range: 20 to 81 years); the median number of prior systemic therapies was 2 (range: 1 to 6) and 14 pts had a prior allogeneic stem cell transplant. Of the treated pts, disease characteristics were: 31 pts with AML (21 MLL-r, 5 MLL-PTD), 4 pts with ALL (3 MLL-r) and 1 pt with MPN (MLL-r). Table 1: Number of evaluable pts per cohort Dose (mg/m2/day) 12 24 36 54 80 90 21 day CIV 1 5 4 6 3 - 28 day CIV - - - - - 17 The median time on study was 29 days (range: 5-196 days). Adverse events (AEs) reported in >15% of pts were: anemia, fever/neutropenia, thrombocytopenia, constipation, diarrhea, nausea, chills, fatigue, mucosal inflammation, peripheral edema, hypomagnesemia, dyspnea and sepsis; of which leukocytosis, nausea and hypomagnesemia were assessed by the investigator as drug related. Grade I PR interval prolongation, without associated QTc changes, was observed in 2 pts. There were no deaths attributed to study drug treatment, and there were 2 pts with treatment discontinuation for potentially drug-related AEs. Dose proportional PK was observed with rapid attainment of steady-state plasma concentrations on Day 1 of treatment. Preliminary PD data demonstrated inhibition of H3K79 methylation from baseline in marrow (median: 52%, range 0-81%) and peripheral blood mononuclear cells (median: 43%, range 25-67%) at doses above 36 mg/m2/day. PD effects were observed by day 8. The 21 day CIV pts recovered H3K79 methylation towards baseline by day 28, and the 28 day CIV pts maintained methyl-mark inhibition throughout cycle 1. To date, best responses for pts with MLL-r are morphologic CR (1 pt.), cytogenetic CR (1 pt), and resolution of leukemia cutis (2 pts). In addition, a treatment-related increase in neutrophils (PMN) and/or monocytes was observed in 6 pts with a median day of onset of 15 days (range 8-28 days). The identification of the MLL-r by split signal FISH in mature PMN suggests a differentiation effect on the leukemic clone. Conclusions: EPZ-5676 is well tolerated and exhibits anti-leukemic activity in MLL-rearranged acute leukemia. These data provide rationale for continued clinical investigation of potential utility of EPZ-5676 in patients with rearrangements of the MLL gene. Disclosures Stein: Janssen Pharmaceuticals: Consultancy. Garcia-Manero:Epizyme, Inc: Research Funding. Rizzieri:Novartis, Inc: Consultancy, Speakers Bureau; Ariad, Inc: Speakers Bureau. Savona:Karyopharm: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees. Altman:Astellas: Membership on an entity's Board of Directors or advisory committees; Ariad: Membership on an entity's Board of Directors or advisory committees; Spectrum: Membership on an entity's Board of Directors or advisory committees; Foundation Medicine: Membership on an entity's Board of Directors or advisory committees. Armstrong:Epizyme, Inc: Consultancy. Pollock:Epizyme: Employment, Equity Ownership. Waters:Epizyme, Inc: Employment, Equity Ownership. Legler:Epizyme, Inc: Employment. Thomson:Epizyme, Inc: Employment. Daigle:Epizyme, Inc: Employment, Equity Ownership. McDonald:Epizyme, Inc: Employment. Campbell:Epizyme, Inc: Employment, Equity Ownership. Olhava:Epizyme, Inc: Employment. Hedrick:Epizyme, Inc: Employment; Pharmacyclics, Inc: Equity Ownership. Copeland:New Enterprise Associates: Ad hoc consultant, Ad hoc consultant Other; Mersana: Membership on an entity's Board of Directors or advisory committees; Epizyme, Inc: Employment, Equity Ownership.

Abstract Introduction: Aberrant fusion proteins involving the MLL histone methyltransferase (HMT) result in recruitment of another HMT, DOT1L, to a multi-protein complex. This leads to abnormal methylation of Histone H3 lysine 79 (H3K79) at MLL target genes and enhanced expression of leukemogenic genes such as HOXA9 and MEIS1 ( Krivstov, 2007). Pinometostat is a small molecule inhibitor of DOT1L with sub-nanomolar affinity and >37,000 fold selectivity against non-MLL HMTs. Pinometostat treatment of MLL-rearranged cells and xenografts reduced histone H3K79 methylation, decreased MLL target gene expression, and induced selective leukemia cell kill (Daigle, 2013). We report the safety, activity, pharmacokinetics (PK) and pharmacodynamics (PD) in the phase 1 trial of pinometostat in adult patients (pts) with relapsed/refractory (R/R) leukemia. Methods: This open label dose escalation/expansion study of pinometostat enrolled pts >18 years (yrs) with R/R leukemia (NCT01684150). In the dose-escalation phase, pts with AML, ALL, mixed lineage leukemia (MLL), myelodysplastic syndrome, myeloproliferative neoplasm or chronic myeloid leukemia were eligible. Eligibility in the two expansion cohorts: 90 mg/m2 (n = 17) and 54 mg/m2 (n = 6), was restricted to pts with MLL-r or MLL-partial tandem duplication (MLL-PTD). Pinometostat was given via continuous intravenous infusion (CIV) for 21 of 28 day cycles in the dose escalation phase and CIV for 28 of 28 day cycles in the expansion phases, until disease progression or unacceptable toxicity. All pts underwent serial collection of PK and peripheral blood mononuclear cells (PBMC) for PD. Leukemic blasts were isolated from PBMCs using flow cytometry and quantified for dimethylation of H3K79 (H3K79-me2) by ChIP-Seq. Results: As of 28-June-2015, 49 pts have enrolled in the dose escalation and expansion phases. Pts receiving 21/28 day infusions: 12 (n=1), 24 (n=5), 36 (n=4), 54 (n=6) and 80 mg/m2/day (n=3). Pts receiving 28/28 day infusions: 54 (n=6) and 90 mg/m2/day (n=24). Table 1. Patient Characteristics n (%) Median age, yrs (range) 51 (19 - 81) Sex (M / F) 27/22 Diagnosis MLL-r* 29 (59) AML MLL-PTD** 5 (9) MLL-wt 7 (14) ALL MLL-r* 5 (10) MLL-wt 1 (2) MLL MLL-r 1 (2) CMML MLL-r 1 (2) # of prior therapeutic regimens 1 - 2 29 (59) 3 - 4 18 (36) >4 2 (4) Prior allogeneic hematopoietic cell transplant 20 (41) * centrally confirmed by karyotype/FISH ** centrally confirmed by NGS (next generation sequencing) Adverse events (AEs) reported in >15% of pts regardless of attribution were: nausea, constipation, vomiting, abdominal pain, diarrhea, hypocalcemia, hypokalemia, hypomagnesemia, fatigue, fever, peripheral edema, mucositis, febrile neutropenia, leukocytosis, anemia, cough, dyspnea, and pneumonia. Grade ≥3 non-hematologic related toxicities include: hypophosphatemia (n=1), decreased ejection fraction (n=3), or elevated transaminases (n=1). Nine patients had leukocytosis (absolute monocyte and neutrophil 50% above baseline and above upper limit of normal) or differentiation. The median days of pinometostat treatment was 35 days (range 3-189 days). To date, objective responses observed are morphologic CR (1 pt), cytogenetic CR (MLL negative by FISH) (1 pt), PR (1 pt) and resolution of leukemia cutis (3 pts). Dose proportional PK was observed with rapid attainment of steady-state plasma concentrations (Css) on Day 1 of treatment. Plasma Css correlated with inhibition of global H3K79-me2 in PBMCs. H3K79-me2 ChIP-Seq demonstrated pinometostat induced reductions in methylation at MLL -r target genes HOXA9 and MEIS1 (median inhibition = 61%: range = 13-91%) in all 9 pts analyzed to date from the 90 mg/m2 expansion cohort. Inhibition of H3K79-me2 in leukemic blasts is consistent with DOT1L suppression. PK-PD relationships in both expansion cohorts using both free and total plasma Css are being explored. Conclusions: Pinometostat administered as a CIV to adults with R/R leukemia has an acceptable safety profile. Clinical activity as demonstrated by both marrow responses and resolution of leukemia cutis were observed. In addition, analysis of H3K79-me2 by ChIP-Seq demonstrated PD reductions in the methylation of MLL-r target genes following pinometostat exposure, as expected from DOT1L inhibition. Relationships between PK exposure, reductions in pinometostat induced H3K79-me2 levels and clinical response are being interrogated. Disclosures Stein: Seattle Genetics: Consultancy; Agios Pharmaceuticals: Consultancy. Rizzieri:Teva: Other: ad board, Speakers Bureau; Celgene: Other: ad board, Speakers Bureau. Berdeja:Novartis: Research Funding; Takeda: Research Funding; BMS: Research Funding; Curis: Research Funding; MEI: Research Funding; Abbvie: Research Funding; Onyx: Research Funding; Array: Research Funding; Janssen: Research Funding; Celgene: Research Funding; Acetylon: Research Funding. Altman:Novartis: Other: Advisory board; Spectrum: Other: Advisory board; Ariad: Other: Advisory board; Seattle Genetics: Other: Advisory board; BMS: Other: Advisory board; Astellas: Other: Participation in an advisory board December 2013. Thomson:Epizyme, Inc: Employment. Blakemore:Epizyme: Employment. Daigle:Epizyme, Inc: Employment. Fine:Epizyme: Employment. Waters:Epizyme, Inc: Employment. Armstrong:Epizyme, Inc: Consultancy. Ho:Epizyme, Inc: Employment.

Key Points Pinometostat demonstrates first evidence of DOT1L target inhibition and clinical responses in a subset of MLL-r advanced leukemia patients. The observed safety profile of pinometostat shows potential for exploration of combination therapies in leukemia.

Abstract Pinometostat is a highly selective first in class DOT1L inhibitor currently in Phase 1 clinical trials in adult and pediatric leukemia patients (pts) with MLL rearrangements (MLL-r or MLL-PTD). Preliminary results of the adult trial have demonstrated clinical activity including complete remissions in a subset of patients (Stein, 2014). Investigation and identification of candidate molecular correlates of pinometostat response in both pt samples and cell lines are reported. RNA and DNA were isolated from PBMCs and/or bone marrow collected prior to treatment from 18 pts enrolled in the adult pinometostat Phase 1 study (CT.gov: NCT01684150), at the following doses 24 (n = 2), 36 (n = 3), 54 (n = 6), 80 (n = 3) and 90 mg/m2/day doses (n = 4). mRNA transcript abundance was assessed using whole genome RNASeq and DNA variants were determined using a 194 gene panel, MyAML (Genection Inc.). Correlations of transcript abundance and DNA variants detected with categorical (responder = CR or PR [n = 3], or no response [n = 16]) and continuous response parameters (time on study [TOS], mean = 59 days: range = 8-196 days) were performed. For cell lines, whole genome RNASeq data was generated from 14 cell lines (MLL-r or MLL-PTD) with a range of in vitro sensitivity to pinometostat (cell proliferation IC50 2 nM to > 10 μM) and RNA transcripts identified as correlated with IC50 were submitted for pathway analysis. Univariate analyses revealed no DNA variants to be associated with either response category (FDR adjusted P < 0.05), however, a statistical association (unadjusted P = 0.05) was identified between increased TOS and those pts harboring t(11:19). Indicating that specific MLL fusion partners may elicit differential sensitivity to pinometostat therapy (2/2 CR pts had t(11:19)). Analysis of the baseline PBMC RNASeq data revealed 201 genes as significantly correlated with TOS (FDR adjusted P<0.05), these data were further analyzed using the Selventa causal modelling platform to identify pathways associated with TOS. Increased activity in pathways leading to differentiation, oxidative stress, inflammation, ras activation and decreased activities of PPAR-γ, DNA methylation and stem cell renewal were associated with increased TOS. Independent analysis of the cell line response data identified pathways that significantly overlapped with those observed in pt samples, with 4/7 of the pt derived mechanisms identified (PPAR-γ activity, inflammation, stem cell renewal and ras activation). In all cases directionality of pathway activity relative to pinometostat sensitivity were concordant between pt and cell line data. Based on these results the effect of a PPAR-γ agonist e.g. rosiglitazone, would be predicted to be antagonistic to pinometostat activity. Results of experiments investigating the effects of combination of rosiglitazone and pinometostat in MOLM-13 cell lines demonstrated that the combination was significantly less effective at inhibiting cell proliferation than pinometostat alone (IC50 shift of 2 fold (n = 2)). MLL-r fusion partner (e.g. t(11:19)) may influence clinical response to pinometostat. In addition RNASeq based characterization of patient samples and cell lines revealed candidate pathways that may cooperate with or antagonize pinometostat activity that warrant further investigation. Citation Format: Scott Daigle, Alice McDonald, Ty M. Thomson, David A. Drubin, Michael Maria, A. Carson, Brad Patay, Jeff Keats, Christine Klaus, Alejandra Raimondi, G. Garcia-Manero, D. A. Rizzieri, Raoul Tibes, Jesus Berdeja, Eytan M. Stein, Blythe Thomson, Stephen J. Blakemore. Identification of biomarkers and pathways associated with response to the DOT1L inhibitor Pinometostat (EPZ-5676) in MLL-r leukemia. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C12.