Novel Androgen Receptor Antagonists for the Treatment of Prostate Cancer - Accounts in Drug Discovery Case Studies in Medicinal Chemistry - page 137

Compared to BMS-641988, BMS-779333 showed higher brain penetration,
with a brain to plasma ratio of 0.7 in mice receiving high oral doses
(4150mg kg 1 ). In mice, a no-effect plasma drug concentration for convulsion
of 220 mM was established for BMS-779333. In a two-week rat study, no con-
vulsions were observed with maximum drug concentrations of 200 mM. In a four-
day dog study, convulsions were noted with repeat dosing (no-effect con-
centration ¼ 170 mM). The margins between the projected maximal exposure at
e cacious dose in humans and the no-effect concentrations for the identified
liabilities in animal testing were 20-fold for seizure and 10-fold for minimal blood
pressure elevation. These values were comparable to BMS-641988, where mar-
gins based on early clinical PK were projected to be 20-fold for seizure due to
mean BMS-949 concentrations and 5-fold for QT prolongation. The advantages
of BMS-779333 were that: (1) minimal blood pressure elevation was viewed as
being less serious than QT prolongation; (2) the margin for seizure was more
certain as it arose from the parent drug instead of a mixture of parent and active
metabolites, including BMS-511, which could not be evaluated for seizurogenic
potential; and (3) BMS-779333 was an AR pan-antagonist against all available
AR mutations, while BMS-641988 was an agonist against certain AR mutations.
In first-in-human enabling dog toxicology studies of longer duration (30
days) and with more animals per group, BMS-779333 unfortunately produced
convulsions at plasma drug concentrations that were less than 10-fold the
projected C max for the ecacious dose in humans. Given the critical importance
of adequate margins to support human testing, 18

the development of BMS-

779333 was discontinued.

6.4.4 Clinical Experience with BMS-641988
The human PK, safety, and ecacy of BMS-641988 was evaluated in Phase I
clinical studies. 20 In vitro incubation experiments had suggested that BMS-
641988 was metabolized by human liver microsomes to BMS-511 via oxidative
N-dealkylation. No metabolites were detected from incubation of BMS-641988
with human hepatocytes. Preclinical studies had established BMS-949 as the
major circulating active metabolite of BMS-641988 in mice, rat, and monkeys
(Table 6.3). Small amounts of BMS-511 were also observed in rats and

Table 6.3 Ratios at steady state of plasma concentrations of the parent drug
and active metabolites upon daily oral administration of BMS-
641988.

Species BMS-641988 BMS-949 BMS-511

mouse 1 9 0

rat 1 0.8 0.2

monkey 1 0.2 0.2

dog 1 0.025 0.02

human 1 2.5 3.3