A Comparative Pharmacokinetic/Pharmacodynamic (PK/PD) Evaluation of Azacitidine Following Subcutaneous (SC) and Oral Administration in Subjects with Myelodysplastic Syndromes (MDS) or Acute Myelogenous Leukemia (AML), Results From a Phase 1 Study.

Parenteral azacitidine significantly improves overall survival in patients with higher-risk MDS and WHO AML (20-30% blasts) compared to conventional care (Fenaux, Lancet Oncol 2009;10:223). An oral azacitidine formulation is in development. A Phase 1, “3+3”, dose-escalation study was designed to determine the maximum tolerated dose (MTD), dose limiting toxicities (DLTs), safety, PK and PD profiles of SC and increasing doses of orally-administered azacitidine in subjects with MDS or AML (not candidates for other therapies) by WHO criteria. In this study, each subject served as their own control and received SC azacitidine in Cycle 1 followed by oral azacitidine in subsequent cycles (Cycle 2+). PK and PD were evaluated in blood, following SC and oral administration. DNA methylation patterns in blood pre- and post-treatment (SC and oral) were explored to build an understanding of how DNA methylation may relate to azacitidine PK parameters.

Azacitidine was administered SC (75 mg/m²/day x 7 days) in Cycle 1, then orally in Cycle 2 starting at a dose of 120 mg/day x 7 days of every 28-day cycle. PK parameters were derived from azacitidine concentrations in plasma and urine collected after the first and last dose. Samples collected in Cycles 1 and 2 were assayed for DNA methylation using Illumina's Infinium Human Methylation27 BeadArrays, which provide methylation levels for >27,000 CpG loci. A linear models for microarray analysis (LIMMA) was performed to identify loci that were differentially methylated from baseline versus (vs.) each post-treatment timepoint across subjects. To identify loci whose methylation levels were modulated in relation to azacitidine PK parameters, Spearman's correlation coefficients were computed for azacitidine exposure (AUC_inf) or peak plasma concentration (C_max) vs. the fold change in methylation at Day 15 for each locus across subjects.

To date, PK data have been generated for 31 of 45 enrolled subjects. Oral doses evaluated were 120, 180, 240, 300, 360, 480 and 600 mg. Following SC and oral administrations, azacitidine was rapidly absorbed and reached C_max within 0.5 hr (0.2, 1.1) and 1.0 hr (0.5, 2.5) post-dose (median [min, max]), respectively. The concentration vs. time profiles declined in a pseudo bi-phasic manner, with a mean elimination half-life of 1.8 ± 1.1 hr and 0.74 ± 0.28 hr for SC and oral, respectively. For SC azacitidine, mean AUC_inf = 1090 ± 470 hr*ng/mL, mean apparent total clearance (CL/F) = 170 ± 130 L/hr, and mean apparent volume of distribution (Vd/F) = 445 ± 450 L. Following either SC or oral treatments, azacitidine recovery in urine was very small relative to dose (<2%). The median (min, max) relative oral bioavailability was 13% (5%, 35%) and the median (min, max) exposure (AUC_inf) of oral azacitidine compared to SC was 31% (14%, 74%) (except for 1 subject whose oral exposure was 167% SC). CL/F exceeded hepatic blood flow indicating extra-hepatic metabolism; Vd/F was 9- to 11-fold greater than total body water suggesting extensive tissue distribution. To date, PD data have been generated for 8 and 3 subjects during SC and oral treatment, respectively. The numbers of significantly (p<0.010) hypomethylated loci peaked at Day 15 of Cycle 1 (SC) with over 4000 loci hypomethylated and at Day 15 of Cycle 2 (oral) with less than 300 loci hypomethylated vs. either baseline or Day 1 of Cycle 2. Approximately 260 hypomethylated loci are common to both groups. Spearman correlation analyses showed that DNA hypomethylation for 252 and 98 loci were correlated (r>0.8) with Day 1 SC and oral azacitidine AUC_inf and C_max, respectively.

Following oral administration, azacitidine demonstrated increased exposure with increasing dose. At dose levels tested to date, oral exposures are lower than SC. Non-renal clearance (hepatic and extra-hepatic) is the major elimination pathway with renal elimination playing a minor role. Changes in DNA methylation were observed in the blood of MDS/AML subjects following SC and oral azacitidine treatment, and a set of CpG loci was identified for which these changes were associated with AUC_inf and/or C_max. PK/PD correlations identified in these initial subjects will be further explored. The lower exposures and PD effects observed with oral vs. SC azacitidine on a 7-day schedule provide rationale for testing extended schedules and BID dosing in future trials.

MacBeth:Celgene Corporation: Employment, Equity Ownership. Laille:Celgene: Employment, Equity Ownership. Ning:Celgene: Employment, Equity Ownership. Cogle:Celgene: Research Funding, Speakers Bureau; Eisai: Speakers Bureau. Skikne:Celgene: Employment. Gore:Celgene: Consultancy, Equity Ownership, Research Funding; Johnson & Johnson: Research Funding. Ward:Celgene: Employment, Equity Ownership.