Novel Androgen Receptor Antagonists for the Treatment of Prostate Cancer - Accounts in Drug Discovery Case Studies in Medicinal Chemistry - page 139

In the clinic, BMS-641988 failed to provide evidence of ecacy despite
achievement of plasma exposures significantly higher than the minimum e-
cacious concentration in xenograft models. Two significant adverse events were
documented: prolongation of QTc interval in a patient receiving 40mg BMS-
641988, and seizure in a patient on a 60mg dose. This combination of safety
and ecacy findings led to the discontinuation of clinical development of BMS-
641988.
Further structural modifications to address metabolism issues, pan-AR
antagonism, and cardiac safety challenges provided BMS-779333, which
exhibited broad spectrum ecacy in four human prostate tumor models
dependent on wt or mutant AR. Despite an encouraging preliminary
safety profile, advancement of BMS-779333 was halted when it induced
seizures in a 30-day dog toxicology study with a margin that was less than
10-fold relative to the exposure needed for ecacy in the tumor xenograft
models.
Several important lessons can be gleaned from this research experience.
Despite a rational approach, the empirical nature of drug discovery can require
navigating an unpredictable and diverse set of liabilities that may not be related
to the biological target, such as convulsions, QT prolongation, thyroid hor-
mone disruption, and blood pressure elevation. Increased investigation of
active metabolites is warranted, especially when the lead compounds suffer
from severe toxicity without premonitory signs, such as seizure. High-dose
testing of animals has a potential to correctly identify a hazard, but the iden-
tified safety margin may be significantly lower in longer duration studies in
animals or in the clinic. Even adjusted for the distinct pattern of human
metabolism relative to that observed in animals, safety margin predictions from
preclinical studies were poorly predictive of clinical margins, perhaps due to
important differences in seizure sensitivity between young, healthy animals and
elderly patients with advanced disease.
The unsatisfactory clinical ecacy of BMS-641988 warrants further inves-
tigation. In the clinic, serum concentrations of PSA decreased for some subjects
after discontinuation of BMS-641988, suggesting that BMS-641988 has partial
AR agonist properties. Anti-androgen withdrawal syndrome has been well
documented with clinical AR antagonists. 5 The BMS AR program relied
heavily on the bicalutamide-resistant CWR-22-BMSLD1 tumor xenograft
model expressing mutant AR. The relevance of this AR mutation in HRPC has
not been established. Recently, development of a novel anti-androgen has been
reported, based on a screen for retention of AR antagonist activity in the
setting of increased AR expression. 22 Such a screen, employing LNCaP human
prostate cancer cells engineered to express higher amounts of wt AR, may be
more representative of the clinical scenario of castration resistant disease.
Translocation of AR to the nucleus is an essential step for downstream tran-
scription of genes involved in proliferation of prostate cells. Therefore, alter-
native novel strategies that involve degradation of AR, or prevent its
translocation to the nucleus, offer significant opportunities for treatment of
HRPC.