Pharmacokinetics, bioequivalence, tolerability, and effects on platelet counts of two formulations of anagrelide in healthy volunteers and patients with thrombocythemia associated with chronic myeloproliferation.

摘要

BACKGROUND: Anagrelide hydrochloride is an anti-thrombotic agent indicated for the treatment of essential thrombocythemia (ET). In various previously published clinical trials of 2 branded formulations of anagrelide in patients with ET at high risk for thrombohemorrhagic events, the rates of adverse events and discontinuation were strikingly divergent between brands. Because the formulations and manufacturers differed, the differences in tolerability, as well as platelet counts, might have been related to differences in pharmacokinetic properties between the 2 formulations. OBJECTIVES: The present series of investigations (1) determined the pharmacokinetic profile of anagrelide and its metabolites; (2) compared the pharmacokinetic profiles of the test and reference formulations of anagrelide; (3) investigated the in vitro release of anagrelide as a marker of intragastric anagrelide release of the test and reference formulations; and (4) compared the platelet-reducing effects of the test and reference formulations in patients with thrombocythemia in 2 longitudinal studies over 4 weeks. METHODS: A series of 4 in vivo studies and 1 in vitro study were conducted. In a pilot, prospective, singledose study in healthy volunteers, the pharmacokinetic properties (C(max), T(max), and AUC(0-infinity)) of a test formulation of anagrelide were assessed using high-performance liquid chromatography analysis of plasma samples. Based on the results from that study, a single-dose, randomized, double-blind, 2-period crossover study in healthy volunteers was conducted to determine bioequivalence of 2 formulations of anagrelide 2 mg/d (taken as 4 capsules). In vitro dissolution properties of the test or reference formulation containing 0.5 mg anagrelide as the active ingredient were studied in an assay mimicking gastrointestinal release. To test for effects on platelet counts of switching from the reference formulation (previous treatment on stable dose for 3 months) to the test formulation, two 4-week longitudinal trials were conducted: one in patients with ET (in Germany), and one in patients with thrombocythemia associated with chronic myeloproliferative disorders (CMPDs) (in Austria). RESULTS: The pilot pharmacokinetic study of the test formulation in 16 volunteers (10 women, 6 men; mean [SD] age, 20.5 [1.5] years; weight, 69.0 [10.0 kg) suggested that anagrelide was metabolized to 3-hydroxyanagrelide (AUC(0-infinity) 50% compared with anagrelide) and the inactive metabolite 2-amino-5,6-dichloro-,4-dihydroquinazolone. The subsequent bioequivalence study in 24 volunteers (14 women, 10 men; mean [SD] age, 23 [4] years; white, 100%; weight, 67.5 [10.2] kg) found that the test formulation was associated with a significantly lower C(max) (point estimation [PE], 66%; 90% CI, 58%-76%; P < 0.001) and AUC(0-infinity) (PE, 77%; 90% CI, 68%-86%; P = 0.001). T(max) values for anagrelide and 3-hydroxyanagrelide were 1 hour longer with the test formulation compared with the reference formulation. The total number of adverse events with the reference formulation was 46; the test formulation, 29 (P = 0.05). In vitro, anagrelide from the reference formulation was immediately released (89.1% at 5 minutes), whereas there was a delayed release (93.6% at 30 minutes) from the test formulation (P < 0.05). In the last 2 studies, 2 cohorts of white patients (cohort 1, 15 patients with ET; 10 women, 5 men; mean [SD] age, 49.0 [10.7] years [range, 31-66 years]; weight, 73.2 [12.6] kg; cohort 2, 19 patients with thrombocythemia associated with CMPD; 12 women, 7 men; age, 62.6 [12.4] years [range, 38-80 years]; weight, 66.1 [13.3] kg) who had received treatment for > or =3 months with the reference formulation were switched to the same dose of the test formulation and maintained on this dose for 4 weeks. Platelet counts did not change significantly from baseline over 4 weeks and stayed within a predefined margin of 150 x 10(3) cells/microL. CONCLUSIONS: The pharmacokineti