steven m. weldon, matthew a. cerny, kristina gueneva-boucheva,...

JPET #236463

1

Selectivity of BI 689648 - A Novel, Highly Selective Aldosterone Synthase

Inhibitor: Comparison to FAD286 and LCI699 in Non-Human Primates

Steven M. Weldon, Matthew A. Cerny, Kristina Gueneva-Boucheva, Derek Cogan, Xin

Guo, Neil Moss, Jean-Hugues Parmentier, Jeremy Richman, Glenn A. Reinhart, and

Nicholas F. Brown

CardioMetabolic Diseases Research (S.M.W., J-H. P., J.R.R., G.A.R., N.F.B.) and

Medicinal Chemistry (M.A.C., K.G.B., D.C., X.G., N.M.) Boehringer Ingelheim

Pharmaceuticals Inc, Ridgefield, CT, 06812

This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on August 1, 2016 as DOI: 10.1124/jpet.116.236463

at ASPE

T Journals on February 22, 2021

jpet.aspetjournals.orgD

ownloaded from

http://jpet.aspetjournals.org/

JPET #236463

2

Running Title:

Selectivity of BI 689648, FAD286 and LCI699 in NHPs

Corresponding Author:

Steven M. Weldon

Cardiometabolic Disease Research

Boehringer Ingelheim Pharmaceuticals Inc

175 Briar Ridge Road

Ridgefield, CT 06877

Office: 203-798-4606; Mobile: 203-512-6228; Fax: 203-837-4606

[email protected]

Number of text pages: 28

Number of Tables: 1

Number of figures: 6

Number of references: 40

Number of words in abstract: 247

Number of words in introduction: 709

Number of words in Discussion: 1499


at ASPE



ownloaded from


JPET #236463

3

Non-standard abbreviations:

11-DC, 11-deoxycortisol

11-DOC, 11-deoxycorticosterone

18-OHB, 18-hydroxycorticosterone

ACEI, angiotensin converting enzyme inhibitor

ACTH, adrenocorticotropic hormone

Aldo, aldosterone

ARB, angiotensin receptor blocker

AS, aldosterone synthase

ASI, AS inhibitor

CAM, cynomolgus adrenal mitochondria

Cmax, maximal compound concentration

CMD, cardiometabolic disease

Cort, cortisol

CS, cortisol synthase

Cyno, cynomolgus monkey

DKD, diabetic kidney disease


at ASPE



ownloaded from


JPET #236463

4

DN, diabetic nephropathy

EC10, Effective Concentration for 10% inhibition




LCI, LCI699

MRA, mineralocorticoid receptor antagonist

PD, pharmacodynamic

PK, pharmacokinetic

R-FAD, R-fadrazole; FAD286

RAAS, renin angiotensin aldosterone system

S-FAD, S-fadrazole

SoC, standard of care

Tmax, time of maximal compound concentration

Recommended section assignment: Drug discovery and translational medicine


at ASPE



ownloaded from


JPET #236463

5

Abstract

The mineralocorticoid aldosterone is an important regulator of blood pressure, volume

and electrolyte balance. However, excess aldosterone can be deleterious as a driver of

vascular remodeling and tissue fibrosis associated with cardiometabolic diseases.

Aldosterone synthase (AS) inhibitors (ASI) attenuate the production of aldosterone

directly and have been proposed as an alternative to mineralocorticoid receptor (MR)

antagonists (MRA) for blocking the pathological effects of excess aldosterone.

Discovery of selective ASIs has been challenging because of the high sequence identity

(93%) AS shares with cortisol synthase (CS), and the low identity of rodent AS

compared to human (63%). Using cynomolgus (cyno) monkey-based models we

identified BI 689648, a novel, highly selective ASI that exhibits an in vitro IC50 of 2 nM

against AS and 300 nm against CS (150-fold selectivity) compared with the recently

described ASIs FAD286 (3 nM AS; 90 nM CS; 40-fold) and LCI699 (10 nM AS; 80 nM

CS; 8-fold). Following oral administration in cynos, BI 689648 (5 mg/kg) exhibits a peak

plasma concentration of ~ 500 nM. For in vivo profiling we used an ACTH-challenge

model in which BI 689648 was >20-fold more selective compared with FAD286 and

LCI699. Since both FAD286 and LCI699 failed to provide adequate selectivity for CS

when tested in patients, the desire for more selective molecules to test the ASI

hypothesis remains high. Therefore, highly selective aldosterone synthase inhibitors

such as BI 689648 represent an important step forward towards developing ASIs with

greater potential for clinical success in cardiometabolic diseases.


at ASPE



ownloaded from


JPET #236463

6

INTRODUCTION The mineralocorticoid aldosterone is a principal modulator of electrolyte and water

balance, required to maintain vascular pressure and tissue perfusion. Under normal

conditions aldosterone concentrations are highly regulated, primarily by the renin-

angiotensin aldosterone system (RAAS), plasma potassium concentration and

adrenocorticotropic hormone (ACTH) (Hattangady et al., 2012). Aldosterone generates

its effects through both mineralocorticoid (MR)-mediated and non-MR mediated

pathways (Nguyen Dinh Cat and Jaisser, 2012; Brown, 2013). Increased levels of

aldosterone are associated with cardiovascular and chronic kidney diseases (Gilbert

and Brown, 2010; Siragy and Carey, 2010; Brown, 2013; Namsolleck and Unger, 2014).

In the kidney, excess plasma aldosterone can contribute to damage as a driver of

vascular remodeling, glomerulosclerosis and tubulointerstitial fibrosis, all of which lead

to declining renal function, increased risk of cardiovascular complications, end stage

renal disease and death (Hostetter and Ibrahim, 2003; Briet and Schiffrin, 2010; Gilbert

and Brown, 2010; Siragy and Carey, 2010; Namsolleck and Unger, 2014).

MR-antagonists (MRA) (Pitt et al., 2001; Epstein et al., 2006; Young, 2013; Liu et al.,

2015), ACE inhibitors (ACEI) and angiotensin receptor blockers (ARB) (Unger, 2002)

are current clinical therapies used to antagonize deleterious effects the RAAS in

patients. However, direct aldosterone blockade via MRAs can disrupt critical electrolyte

balance leading to hyperkalemia (Roscioni et al., 2012; Shavit et al., 2012). Also, MRAs

increase plasma aldosterone that may generate harmful effects via non-MR mediated

pathways (Vinson and Coghlan, 2010; Brown, 2013). By comparison, lowering levels of

aldosterone indirectly with angiotensin antagonists often results in only temporary


at ASPE



ownloaded from


JPET #236463

7

aldosterone reduction due to “aldosterone breakthrough”, observed in ~ 30-40% of

patients using ACEIs or ARBs (Bomback and Klemmer, 2007).

An alternative approach to antagonizing the effects of excess aldosterone in patients

is to directly attenuate its production via selective inhibition of aldosterone synthase

(AS). Clinically, an AS inhibitor (ASI) would be expected to attenuate the pathological

effects of aldosterone mediated through both MR- or non-MR pathways, thereby

providing potentially greater disease modifying benefit compared to MRAs. In addition,

an ASI with an ideal pharmacodynamic profile would reduce excess aldosterone while

allowing for sufficient signalling via MR to maintain electrolyte balance and minimize the

risk of hyperkalemia. Finally, a highly selective ASI should have little to no impact on

cortisol synthase (CS), an enzyme closely related to AS and responsible for generating

cortisol, required for maintaining critical metabolic and immune responses. FAD286

(the R-enantiomer of fadrozole) (Santen et al., 1991; Menard and Pascoe, 2006) and

LCI699 (Amar et al., 2010; Calhoun et al., 2011) are examples of ASIs that have been

evaluated in preclinical models (Fiebeler et al., 2005; Menard et al., 2014) as well as in

patients (Santen et al., 1991; Menard and Pascoe, 2006; Amar et al., 2010; Calhoun et

al., 2011). However, these molecules exhibit only modest selectivity for inhibition of

human AS over CS, and resulted in blunted cortisol responses when evaluated in

patients (Santen et al., 1991; Amar et al., 2010), a likely reason for their clinical failure

as selective AS inhibitors. As such, novel, more selective ASIs are required to define

the ideal level of AS inhibition and selectivity needed to achieve clinical safety and

efficacy.


at ASPE



ownloaded from


JPET #236463

8

Aldosterone synthase is a mitochondrial cytochrome P450 enzyme (CYP11B2) that

catalyzes the final three steps in aldosterone synthesis from 11-deoxycorticosterone

(11-DOC) to aldosterone (Payne and Hales, 2004). Discovery of selective ASI

molecules has been challenging due to a number of factors. First, AS shares 93%

amino acid sequence identity with cortisol synthase, CYP11B1 (Cerny, 2013) and as

mentioned above, attenuating CS/cortisol production is undesirable. Second, the utility

of preclinical rodent models is limited due to the low sequence identity between rodent

and human AS (63%) and the catalytic activity of Cyp11B in rodents, that generates

corticosterone rather than cortisol (van Weerden et al., 1992), making in vivo selectivity

assessments in rodents less applicable to the clinical condition.

Given the limitations of MRAs and angiotensin antagonists for antagonizing

aldosterone in patients and the need for more selective ASIs with greater potential to

achieve clinical success, we used cynomolgus (cyno) monkey-based methods to

identify and profile novel ASI molecules. Herein we describe BI 689648, a novel, highly

selective and orally active ASI that represents a significant step forward in the

development of safe and effective aldosterone antagonists for clinical use.


at ASPE



ownloaded from


JPET #236463

9

Materials and Methods

Reagents and Compounds

In vitro assay reagents 11-DOC, corticosterone, 11-DC, cortisol 18-

hydroxycorticosterone, aldosterone, dimethylsulfoxide (DMSO), and nicotinamide

adenenine diphosphate reduced form (NADPH) were purchased from Sigma-Aldrich

(Milwaukee, WI). All tissue collection procedures were performed according to

protocols approved by the Boehringer Ingelheim Institutional Animal Care and Use

Committee. Cynomolgus monkey adrenal glands were obtained from vehicle-treated

control animals from in-house toxicological studies or from Alpha Genesis, Inc.

(Yemassee, SC), Charles River Laboratories (Reno, NV), and BioChemed (Winchester,

VA). Adrenal glands were flash-frozen immediately after surgical removal and shipped

frozen on dry ice. Aldosterone synthase inhibitor molecules: S-FAD and FAD286 (S-

and R-enantiomers of fadrozole, respectively), 4-(5,6,7,8-tetrahydroimidazo[1,5-

a]pyridin-5-yl)benzonitrile)(Browne et al., 1991; Furet et al., 1993); osilodrostat or

LCI699, ((R)-4-(6,7-dihydro-5H-pyrrolo[1,2-c]imidazol-5-yl)-3-

fluorobenzonitrile)(Meredith et al., 2013) and BI 689648 (6-(5-Methoxymethyl-pyridin-3-

yl)-3,4-dihydro-2H-[1,8]naphthyridine-1-carboxylic acid amide)(Balestra et al., 2015),

were synthesized according to published procedures.

In vitro Cyno Adrenal Homogenate (CAH) Assay

Potency against CYP11B1 and CYP11B2 was assessed using the recently described

CAH assay (Cerny et al., 2015). In brief, homogenized adrenal glands were used to

evaluate test compounds in a 96-well plate format. A mixture of concentrated

homogenate and substrate was added to compound dilutions for analysis. Values for


at ASPE



ownloaded from


JPET #236463

10

the concentration required to inhibit CS (CYP11B1) and AS (CYP11B2) enzyme activity

by 50% (IC50) were calculated from the plot of log [inhibitor] versus % inhibition by curve

fitting using GraphPad Prism 6 (La Jolla, CA).

In vivo Profiling in Conscious Cynomolgus Monkeys

Prior to the in vivo pharmacodynamic evaluation, a separate plasma pharmacokinetic

(PK) profile for each ASI was obtained in normal cynos (n=2-3) following oral

administration at 5 mg/kg. A racemic mixture of R- and S-fadrazole was used for

pharmacokinetic profiling and the two enantiomers are assumed to be in equal amounts

in analysis samples.

The in vivo cynomolgus monkey (Macaca fascicularis) experiments were performed in

collaboration with SNBL-USA (Everett, Washington) in compliance with SNBL and BIPI

approved IACUC protocols. All studies were non-terminal. Healthy, male animals of

approximately 4-8 years of age were housed in a temperature-humidity-controlled

environment with a 12-hour diurnal light cycle. Animals were randomized to study

cohorts based on body weights and previous therapeutic treatment. Cynos were

acclimated to their study cohorts and procedural handling during a 7-14 day period prior

to initiation of dosing during which time they were assessed for behavioral abnormalities

(i.e. stress) that could adversely affect aldosterone-cortisol levels and performance

while on study. Animals were group-housed until the day of study at which time they

were transferred to individual procedural cages to facilitate dosing and sample collection.

On each experimental day, conscious unrestrained animals were pre-treated with

dexamethasone (2 mg/kg, I.M., ~ 6 hrs prior to ACTH) to quench baseline aldosterone

and cortisol levels. Initial ACTH-induced steroid response profiling in normal, non-


at ASPE



ownloaded from


JPET #236463

11

compound treated cynos was conducted in a separate study cohort (n=8). Test

compounds were administered intravenously (brachial-cephalic) or via nasogastric (NG)

gavage (“oral”). Temporary, light sedation (ketamine, 5-10 mg/kg, I.M.) was used only

to facilitate the ASI compound doses administered intravenously via slow bolus. ACTH

(CortrosynTM, 1 ug/kg, I.V.) was administered immediately after ASI compounds given

intravenously. For orally applied test compounds in non-sedated animals, ACTH was

administered (I.V.) at the time of maximal (Tmax) plasma compound concentration (Cmax).

The dosing route of administration (I.V. or N.G.) for vehicle control and test compound

groups was the same on any given study day.

For ASI evaluation, the study cohort consisted of 66 healthy animals. Each separate

study day, 12 cynos were randomized to receive various doses of ASI or Vehicle

Control (n=3/group); animals were reused across studies, allowing a minimum of 2

weeks washout between studies. Aggregation of data across multiple studies was used

to derive in vivo Effective Concentration (EC) values for aldosterone and cortisol by

curve-fitting. Conscious, non-chaired monkeys received vehicle (n=35; where n is equal

to the number of treatments), S-FAD (n=9), FAD286 (n=24), LCI699 (n=36) or BI

689648 (n=26) at doses ranging from 0.003-10 mg/kg. Maximal ACTH-induced

aldosterone and cortisol production occurred quickly, within 15 min following challenge,

at which time blood was collected for plasma aldosterone, cortisol and test compound

concentrations. Baseline values shown are comprised of all animals across all study

days (n=94). Steroid levels for individual ASI-treated cynos were expressed relative to

mean values for the vehicle-treated control groups (Percent of Control; POC responses

or Fold-Change vs Control) and plotted versus log[unbound compound]. This allowed


at ASPE



ownloaded from


JPET #236463

12

aggregation of data across studies to derive in vivo EC values for aldosterone and

cortisol by curve-fitting using a 4-parameter logistic fit with the upper and lower

asymptotes fixed at 100% to 0%; Hill Slope= -1.0 GraphPad Prism 6 software (La Jolla,

CA).

LC/MS Determination of Plasma Steroids and Test Compounds

The quantitation of the five steroids, aldosterone, cortisol, corticosterone, 11-DOC and

11-DC in cyno plasma was performed by LC/MS/MS. An API 5000 triple quadruple

mass spectrometer with Turbo V Ion Source (Applied Biosytems, Toronto, Canada), set

to electrospray negative/positive ionization mode, and Analyst 1.4.2 operating software

was used. Source conditions were as follows: source temperature 500°C, electrospray

capillary voltage 5.5 eV, curtain gas GS1 20, and nebulizing, dissolving gas flow rates

30 and 50 arbitrary units respectively.

For the analytes of interest precursor to product ion transitions were established

through direct infusion of the compounds into the mass spectrometer. Quantitation by

multiple reaction monitoring (MRM) analysis was performed both in negative ion mode

for aldosterone and cortisol and in positive ion mode for corticosterone, 11-DOC and

11-DC. The following ion transitions were used for quantification: aldosterone (359.2-

189.0 m/z, CE -25, DP -150V) cortisol (361.0-331.0 m/z, CE -25, DP -170V),

corticosterone (347.1-97.0 m/z, CE 35, DP 60V), 11-DOC (331.2-109.0 m/z, CE 20, DP

55V) and 11-DC (347.2-121.0 m/z, CE 40, DP 100V). The internal standards used were

aldosterone-d7 (366.2- 193.0 m/z, CE -22, DP -120 V) in negative mode and a

proprietary small molecule in positive mode (424.1-270.1 m/z, CE 50, DP 100). Q1 and


at ASPE



ownloaded from


JPET #236463

13

Q3 were set at unit resolution. The polarity was switched from negative to positive at

6.5 min.

The liquid chromatography system was composed of an Agilent 1200 Series pump

and column oven (Fullerton, CA, USA) and LEAP Technologies HTS-PAL Autosampler

(Carrboro, NC, US). The analytical column used was Phenomenex Synergi Polar RP,

2.1 x 150 mm, 5 um (Torrance, CA, US), and the mobile phase consisted of 10mM

ammonium acetate with 0.1% formic acid in water (A) and Acetonitrile (B). Gradient

was maintained at 5% B for 1 minute, increased to 95% B in 6 min, held at this level for

2 min, and then decreased to the initial 5%B within 1min and held for 1 min. The

column temperature was set at 50 °C, and the flow rate at 0.3 ml/min. The total run time

per injection was 10 min.

The same instrumentation was used to analyze BI 689648, FAD286, S-FAD and

LCI699. The LC column used for the quantitation was Phenomenex Synergi Polar RP

2.1x50, 4u at 400C. A mobile phase consisting of 5 mM ammonium acetate with 0.1%

formic acid in water (A) and acetonitrile (B) was used with a gradient elution 5% B for

0.1 min, increased to 95% B in 1.8 min, held at 95% B for 0.5 min and decreased to 5%

B for 0.2 min and held at 5% B for a total time of 3 min at a flow rate of 0.5 ml/min. The

compound transitions were as follows BI 689648 (299.0-256.0 m/z, CE 28, DP 80), FAD

286 (224.2-82.0 m/z, CE 35, DP 90) S-FAD (224.2-82.0 m/z, CE 35, DP 90) and LCI699

(228.0-81.0 m/z, CE 38, DP 80).

Steroid Sample Preparation: A standard stock solution containing 0.2 mg/ml steroid

mix in methanol was serially diluted in 2.5% bovine serum albumin (BSA) in HBSS

buffer to prepare a standard curve ranging from 0.01 to 1000 ng/ml. Plasma (100 µl),


at ASPE



ownloaded from


JPET #236463

14

standards and blanks were mixed with 40 ul of 0.1% formic acid and 60 ul of 500 ng/ml

aldosterone-d7 diluted in acetonitrile:water (1:9v/v). Samples were vortexed and loaded

on to 100 mg ISOLUTE® SLE plates (Biotage, Charlotte, NC) by applying vacuum. The

steroids were eluted from the matrix by solvent mixture of dichloromethane:t-BTE/ 70:30

4x250 uL . The filtrates were evaporated to dryness under nitrogen using SPE Dry 96

(Jones Chromatography, Lakewood, CO) and reconstituted in 100 ul

acetonitrile:water:formic acid (1:1:0.1%). The reconstituted solution (20 ul) was injected

into the LC/MS/MS for analysis.

ASI Sample Preparation: BI 689648, FAD286, S-FAD and LCI699 standard stock

solution containing 1 mg/mL in methanol were serially diluted in cyno plasma to prepare

an 8- point standard curve ranging from 1 to 5000 ng/mL. Plasma samples (20 µl),

calibration standards and blank plasma samples were deproteinized by precipitation

with 180 µl, 250 ng/ml internal standard (proprietary small molecule) diluted in

acetonitrile: water (85:15). Samples were mixed for 1 min, filtered through AcroPrep

multi-well filter plates (Pall Corporation, Ann Arbor, MI) using Sciclone ALH 3000

Workstation (Caliper Life Sciences, Hopkinton, MA, USA) and transferred into 96-well

injection plate. The filtered solutions (5 µl) were injected into the LC/MS/MS system.


at ASPE



ownloaded from


JPET #236463

15

Results:

In Vitro Selectivity of BI 689648, LCI699 and FADs

The chemical structures for BI 689648, LCI699, R-Fadrozole (FAD286) and S-

Fadrozole (S-FAD) are shown in Figure 1. The in vitro IC50 and selectivity factor for BI

689648 and other ASIs are shown in Table 1. Compared to the FADs and LCI699, BI

689648 was highly selective in vitro providing an IC50 for CYP11B2 of 2.1 nM and a

selectivity factor of 149 over CYP11B1. FAD286 by comparison, showed a similar IC50

for CYP11B2 (2.5 nM) however it’s greater potency against CYP11B1 (94 nM) resulted

in a comparatively modest selectivity factor of 38, approximately 4-fold less than BI

689648. LCI699 showed lower CYP11B2 selectivity (~8-fold) while S-FAD provided no

CYP11B2 selectivity, favoring CYP11B1 by ~ 30-fold over CYP11B2.

In vivo Selectivity of BI 689648 Compared to LCI699 and FADs

Pharmacokinetic (PK) profiles: A PK profile for each ASI was established for normal

cynos (n=2-3) prior to in vivo pharmacodynamic evaluation. S-FAD, FAD286 and

LCI699 exhibited similar PK profiles with maximum plasma concentrations (Cmax) of ~

3000 nM observed ~ 1h post-dose. By comparison, BI 689648 shows a Cmax of ~ 500

nM at 0.3 h post-dose (Figure 2). Over the first hour after dosing, all compounds

evaluated showed plasma Cmax values suitable to achieve > 90% inhibition (IC90) of

CYP11B2 as predicted by the in vitro CAH assay.

Baseline ACTH-responses: To evaluate in vivo efficacy of our ASIs we took advantage

of the ACTH stimulation test which is used clinically (Dorin et al., 2003) to evaluate

stimulated steroid production as an indication of adrenal function. ACTH-challenge


at ASPE



ownloaded from


JPET #236463

16

induced a rapid increase in plasma concentrations of both aldosterone and cortisol,

providing a robust dynamic range for evaluating ASI compounds. Maximal ACTH-

induced aldosterone and cortisol production occurred quickly, within 15 min following

challenge (Figure 3A). Acute ACTH-induced aldosterone levels peaked 15-30 min post-

challenge thereafter returning to baseline values within 2 hrs. The peak cortisol level

was observed at 1h, returning gradually to baseline values by 24 hrs post-challenge.

Pre-challenge (baseline) aldosterone and cortisol levels in cynos showed minimal

variability and were similar to unstimulated levels in normal humans (Hardman, 2001).

Baseline values for the study cohort as well as the ACTH-stimulated responses from the

vehicle control group are shown in Figure 3B. Baseline plasma aldosterone was

0.144±0.016 ng/ml and cortisol was 48±2 ng/ml. Following ACTH-challenge (15 min)

both aldosterone and cortisol levels increased ~4-fold over baseline (aldosterone:

0.486±0.058 ng/ml; cortisol: 208±15 ng/ml). Although, peak levels of aldosterone and

cortisol occurred at different times (presumably reflecting the slower clearance of

cortisol), the apparent maximal rate of synthesis of both occurred quickly, within the first

15 minutes. Therefore, we used the 15 minute values to assess the pharmacodynamic

effect of test compounds.

Pharmacodynamic (PD) profiles: The CAH IC50 values generated for aldosterone in

vitro paralleled the PD effect in vivo (cyno EC50) for all compounds except S-FAD, which

showed ~10-fold higher potency in vivo compared to in vitro (Table 1, Figure 4). All

ASIs exhibited lower potency for cortisol inhibition in vivo compared to in vitro. The

selectivity for each ASI is represented by the black arrow at the center of each curve,

highlighting the “X-Fold” changes in EC50 for aldosterone vs. cortisol. Consistent with


at ASPE



ownloaded from


JPET #236463

17

the rank-order efficacy observed from our in vitro data shown in Table 1, we clearly

show the differentiation for in vivo efficacy between a highly selective AS inhibitor such

as BI 689648, that results in >11000-fold selectivity compared to the relatively modest

effects of FAD286 (530-fold) and LCI699 (309-fold) or S-FAD that exhibits only 8-fold

selectivity for aldosterone over cortisol in vivo.

Selectivity of BI 689648, LCI699 and FADs Applying an Extended Therapeutic

Index (EC75 ALDO-EC10 CORT)

Considering a hypothesis that requires a high degree of aldosterone inhibition (~75%)

while maintaining little to no cortisol inhibition (≤10%) in order to achieve clinical

success, we used the curve fitting method described above to generate EC75 ALDO-EC10

CORT selectivity values representing a postulated, “high bar” therapeutic index (TI) for

safe and effective aldosterone inhibition (Figure 5). The EC75-EC10 data demonstrate

that even when using these more stringent criteria to define CYP11B2 selectivity, BI

689648 still exhibits a considerable TI of 400-fold selectivity whereas FAD286 and

LCI699 exhibit TIs of only 28- and 11-fold, respectively. Furthermore, the small window

of selectivity S-FAD exhibited using the EC50 analysis was completely abolished (<1-

fold) using the more stringent selectivity criteria.

Effect of ASIs on Steroid Precursor Substrates In Vivo

To further evaluate the impact of AS inhibition on acute aldosterone-cortisol balance

in vivo, we examined the effects of the ASIs on respective steroid precursor substrates,

11-DOC, corticosterone, 18-OHB and 11-DC, following ACTH challenge. Compared to

baseline levels, ACTH induced increases of ~ 3-fold in the aldosterone precursor 11-


at ASPE



ownloaded from


JPET #236463

18

DOC (0.1 to 0.3 ng/ml); ~ 12-fold in the cortisol precursor 11-DC (0.5 to 6 ng/ml) and ~

4-fold (data not shown) for both corticosterone and 18-OHB. All of the ASI treatments

increased the aldosterone precursor 11-DOC consistent with increasing compound

concentrations and respective EC50 inhibitions of aldosterone (Figure 6). As expected,

compounds that showed greater potency and selectivity for CYP11B2 (BI

689684>FAD286>LCI699>S-FAD) required much higher plasma concentrations to

modulate 11-DOC and also had a reduced effect on the cortisol precursor 11-DC

compared with less selective compounds. The more selective ASIs showed only

minimal changes in 11-DOC until plasma concentrations reached levels of nearly

complete aldosterone inhibition (≥90%) and excess thereof (Figure 6). For BI 689648

(aldo EC50=2 nM), appreciable changes in 11-DOC were only noted at plasma

concentrations >2000 nM or >1000X its aldo EC50 while FAD286 showed a window of

~100X. By comparison, the least selective ASIs begin to increase plasma 11-DOC at

exposure levels much closer to their respective aldo EC50 values (S-FAD ~2-5X; LCI699

~ 1-20X). Effects on the cortisol precursor 11-DC (Figure 6) were also consistent with

compound selectivity. The less selective molecules exhibited greater increases in 11-

DC and at lower plasma concentrations with respect to EC50 values for cortisol inhibition.

Similar to that of 11-DOC, BI 689648 exhibited minimal impact on 11-DC and only at

very high plasma concentrations (~10 uM). Effects of ASIs on plasma corticosterone

and 18-OHB levels were unremarkable, showing a slight elevation above control levels

with no clear dose-response relationship (data not shown).


at ASPE



ownloaded from


JPET #236463

19

Discussion:

Clinical benefit of targeting the harmful effects of aldosterone in cardiovascular

disease has been demonstrated using the MRAs eplerenone (Pitt et al., 2001) and

spironolactone (RALES, 1996) while finerenone, representing the next generation of

non-steroidal MRAs is currently under clinical evaluation (Bakris et al., 2015). However,

the use of MRAs presents specific limitations that could in principle be mitigated by

direct inhibition of aldosterone synthesis. First, blocking the MR leads to an increase in

circulating aldosterone levels. Since deleterious effects of aldosterone can be mediated

through both MR-dependent and MR-independent pathways (Nguyen Dinh Cat and

Jaisser, 2012; Brown, 2013), MRAs will block only MR-dependent pathways, whereas

an ASI would attenuate all aldosterone-dependent pathways. Second, treatment with

MRAs can disrupt electrolyte balance leading to hyperkalemia (Roscioni et al., 2012;

Shavit et al., 2012). An ASI with an ideal pharmacokinetic profile would be expected to

reduce excess aldosterone exposure while allowing adequate signalling via MR to

maintain electrolyte balance and minimize the risk of hyperkalemia.

The angiotensin pathway is an important contributor to cardiometabolic disease (CMD)

pathologies, including diabetic kidney disease (DKD). Angiotensin converting enzyme

inhibitors and ARBs are part of the current standard of care for DKD but angiotensin

antagonists provide only limited protection against the progressive loss in renal function

observed in these patients (Lewis et al., 1993; Brenner et al., 2001; Lewis et al., 2001).

Consequently, additional therapeutic agents are needed to combat alternative

pathological pathways such as aldosterone, associated with cardiometabolic diseases.

In addition, although ACEIs and ARBs have an indirect effect of reducing aldosterone


at ASPE



ownloaded from


JPET #236463

20

levels, a substantial number of patients on ACEIs or ARBs experience “aldosterone

breakthrough”, in which initially reduced aldosterone levels return to pretreatment

values or higher (Bomback and Klemmer, 2007). Therefore, in CMDs such as DKD,

ASIs would provide additional disease modifying benefit on top of the angiotensin

antagonists by inhibiting pathological effects of aldosterone mediated through both MR-

and non-MR pathways.

Discovery of selective ASIs has been challenging due to the high protein sequence

identity between human CYP11B1 and CYP11B2 and limitations of current preclinical

model systems, in vitro and in vivo (Cerny et al., 2015). Addressing these challenges,

we used complementary in vitro and in vivo, primate-based models for identifying and

profiling novel, highly selective ASIs. To identify molecules suitable for optimization, we

developed the CAH assay a tissue-derived assay that is physiologically relevant to the

intact in vivo system regarding protein homology, enzyme-cofactors and activities

(Cerny et al., 2015). For in vivo selectivity profiling we used conscious cynos subjected

to the ACTH stimulation test, used clinically to evaluate adrenal steroid production

(Dorin et al., 2003) and as such, highly relevant for evaluating the pharmacodynamic

effects of ASIs. The use of primate-based models for preclinical profiling of ASIs is

advantageous to rodent models for defining selectivity and the potential to better predict

compounds likely to achieve clinical success. Following this discovery strategy we

identified BI 689648, a potent and highly selective ASI that exhibits an in vitro IC50 of 2

nM (Table 1) and 11000-fold in vivo selectivity (Figure 4), almost 4 times more selective

than LCI699, a recent ASI evaluated in patients. This is an important finding since the

selectivity of LCI699 in preclinical studies (~3-fold), in conjunction with its


at ASPE



ownloaded from


JPET #236463

21

pharmacological profile in early clinical trials, were considered suitable at the time for

subsequent Phase II clinical evaluation. However, this compound ultimately failed to

show acceptable selectivity for aldosterone vs cortisol inhibition in humans (Amar et al.,

2010; Azizi et al., 2013).

Our CAH data showed IC50’s for S-FAD, LCI699 and FAD286 that were slightly lower

but in general, comparable to those previously published (Roumen et al., 2007; Roumen

et al., 2010) although, a limitation of our data is that for LCI699 and BI 689648, only a

single experiment was performed. However, we recently reported good reproducibility

for the CAH assay using structurally diverse compounds, different CAH preparations;

and including male and female tissue showing there are no gender differences

associated with CYP11B activities in cynos (Cerny et al., 2015). As a predictor of in

vivo efficacy (plasma aldosterone inhibition), we found that the CAH assay provided

accurate and useful rank-order results and was highly predictive for both potency and

selectivity in vivo. Our in vitro-in vivo (IC50-EC50) correlations were similar for CYP11B2-

aldosterone, but for CYP11B1-cortisol, all compounds were consistently less potent in

vivo (Table 1; Figure 4) but the lower CYP11B1 in vitro potencies did not impact the

predictive value of the CAH assay. Comparing compound selectivity based on IC50-

EC50 values we showed the rank-order selectivity of BI 689648 > FAD286 > LCI699 >

S-FAD is directly in line for both profiling modalities. As all compounds tested are

believed to be competitive inhibitors, there was no expectation that the absolute

measure of potency under both in vitro and in vivo assay conditions should coincide

since the enzymatic reactions (and inhibition thereof) are dependent upon prevailing


at ASPE



ownloaded from


JPET #236463

22

local substrate concentrations and as such, not expected to be the same for both

experimental conditions.

During cardiometabolic diseases in which aldosterone is implicated as a key mediator,

the aldosterone synthesis pathway is expected to be increased, but the level of

aldosterone inhibition required to exhibit clinical safety and efficacy has yet to be

determined. Considering the efficacy (aldo EC50) profiles for the ASIs shown in Figure 4,

one might predict that the selectivity observed for LCI699 (~300-fold) or FAD286 (~ 500-

fold) would be sufficient to provide an adequate therapeutic window for aldosterone-

cortisol inhibition. However, that was not the case when these compounds were

evaluated in humans (Santen et al., 1991; Azizi et al., 2013). We postulated a higher

therapeutic index may be required to achieve clinical success with an ASI; assuming

greater aldosterone inhibition (75%) in combination with little or no inhibition of cortisol

(≤10%). We applied this postulated therapeutic index of EC75 ALDO vs. EC10 CORT as a

more likely indicator of clinical reality compared to a conventional EC50 ALDO-CORT ratio

and in doing so we saw quite a different selectivity profile for each compound. LCI699

and FAD286, both of which appeared to provide favorable selectivity using the EC50

criteria became <30-fold selective when using the EC75 ALDO vs EC10 CORT formula while

the selectivity of S-FAD was reduced to less than zero. Only BI 689648, the most

selective ASI we tested maintained a substantial degree of selectivity (~400-fold) under

the more stringent criteria (Figure 5). Whether or not such criteria will be beneficial for

successful clinical projections going forward will need to be proven.

Since cortisol and aldosterone are the terminal products of multi-step enzymatic

processes mediated by CYP11B1 and CYP11B2, respectively, we evaluated the effects


at ASPE



ownloaded from


JPET #236463

23

of ASIs on aldosterone and cortisol pathway intermediates 11-DOC, corticosterone, 18-

OHB and 11-DC. CYP11B2 catalyzes three successive reactions first converting 11-

DOC to corticosterone, then corticosterone to 18-OHB and finally 18-OHB to

aldosterone. CYP11B1 is primarily responsible for converting 11-DC to cortisol but is

also capable of converting 11-DOC to corticosterone (Payne and Hales, 2004).

Consistent with mechanism of action, we expected that all ASIs would increase the

aldosterone precursor 11-DOC concomitant with increasing CYP11B2 inhibition and,

compounds that showed greater selectivity for CYP11B2 vs CYP11B1 would exhibit

less effect on the cortisol precursor 11-DC. We observed the expected effect on both

intermediates with all ASIs, albeit at distinctly different levels of CYP11B inhibition. The

most robust changes in 11-DOC and 11-DC were noted for the least selective ASIs (S-

FAD; LCI699) and occurred at respective plasma concentrations expected to induce

significant CYP11B1 inhibition (i.e. cortisol EC50) while compounds with higher AS

selectivity (FAD286 and BI 689648) showed less impact on 11-DC (Figure 6). The

larger changes observed for the less selective ASIs are indicative of their greater impact

on CYP11B1 inhibition since both 11-DOC and 11-DC are substrates for CYP11B1.

Since both CYP11B1 and B2 are capable of metabolizing 11-DOC to corticosterone and

there is an inherently higher concentration of endogenous CYP11B1 compared to

CYP11B2 (Guengerich et al., 2011), there is likely a greater role for CYP11B1 vs

CYP11B2 in the conversion of 11-DOC to corticosterone resulting in higher 11-DOC

accumulation during combined CYP11B1-B2 inhibition as would be the case following

treatment with the less selective ASIs. There were no meaningful changes in either

corticosterone or 18-OHB following ASI treatment.


at ASPE



ownloaded from


JPET #236463

24

In summary, we employed cynomolgus monkey based models to identify and prioritize

novel ASIs as potential therapeutic agents based on potency and selectivity. These

primate-based models are highly relevant to adrenal steroid biology in humans and

provide a more applicable model for assessing ASI selectivity compared to those in

rodents. Herein, we report the identification of a novel, orally active AS inhibitor BI

689648 that is highly selective and exhibits minimal to no effect on cortisol or other

steroid pathway intermediates compared to the recently described ASIs FAD286 and

LCI699. We conclude that highly selective aldosterone synthase inhibitors such as BI

689648 represent an important step toward the discovery of safe and effective

aldosterone modulators for clinical use in cardiometabolic diseases including CKD and

diabetic nephropathy.


at ASPE



ownloaded from


JPET #236463

25

Acknowledgements:

The authors would like to acknowledge the Scientific Service staff at SNBL-USA: Narine

Lalayeva, Megumi Bailey, Motti Dabadi and Tiana Spencer for their technical expertise

performing the in-life procedures for the cynomolgus monkey studies.


at ASPE



ownloaded from


JPET #236463

26

Authorship contributions and disclosures:

Participated in research design: Steven M. Weldon, Matthew A. Cerny, Derek Cogan,

Neil Moss, Jean-Hugues Parmentier, Jeremy Richman, Nicholas F. Brown

Conducted experiments: Steven M. Weldon, Matthew A. Cerny

Contributed new reagent or analytic tools: Derek Cogan, Xin Guo, Neil Moss, Jean-

Hugues Parmentier

Performed data analysis: Steven M. Weldon, Matthew A. Cerny, Kristina Gueneva-

Boucheva, Neil Moss, Jeremy Richman, Nicholas F. Brown

Wrote or contributed to writing of the manuscript: Steven M. Weldon, Matthew A.

Cerny, Kristina Gueneva-Boucheva, Glenn A. Reinhart, and Nicholas F. Brown


at ASPE



ownloaded from


JPET #236463

27

References

Amar L, Azizi M, Menard J, Peyrard S, Watson C and Plouin PF (2010) Aldosterone synthase

inhibition with LCI699: a proof-of-concept study in patients with primary aldosteronism.

Hypertension 56:831-838.

Azizi M, Amar L and Menard J (2013) Aldosterone synthase inhibition in humans. Nephrol Dial

Transplant 28:36-43.

Bakris GL, Agarwal R, Chan JC, Cooper ME, Gansevoort RT, Haller H, Remuzzi G, Rossing P,

Schmieder RE, Nowack C, Kolkhof P, Joseph A, Pieper A, Kimmeskamp-Kirschbaum N

and Ruilope LM (2015) Effect of Finerenone on Albuminuria in Patients With Diabetic

Nephropathy: A Randomized Clinical Trial. Jama 314:884-894.

Balestra M, Burke J, Chen Z, Cogan D, Fader L, Guo X, McKibben B, Marshall D, Richard N,

Peter A and Yu H (2015) US9181272 -Aldosterone synthase inhibitors in USPTO

(USPTO ed), Boehringer Ingelheim International GmbH (Ingelheim am Rhein, DE),

United States.

Bomback AS and Klemmer PJ (2007) The incidence and implications of aldosterone

breakthrough. Nat Clin Pract Nephrol 3:486-492.

Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, Remuzzi G, Snapinn

SM, Zhang Z and Shahinfar S (2001) Effects of losartan on renal and cardiovascular

outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med 345:861-869.

Briet M and Schiffrin EL (2010) Aldosterone: effects on the kidney and cardiovascular system.

Nat Rev Nephrol 6:261-273.

Brown NJ (2013) Contribution of aldosterone to cardiovascular and renal inflammation and

fibrosis. Nat Rev Nephrol 9:459-469.

Browne LJ, Gude C, Rodriguez H, Steele RE and Bhatnager A (1991) Fadrozole hydrochloride:

a potent, selective, nonsteroidal inhibitor of aromatase for the treatment of estrogen-

dependent disease. J Med Chem 34:725-736.


at ASPE



ownloaded from


JPET #236463

28

Calhoun DA, White WB, Krum H, Guo W, Bermann G, Trapani A, Lefkowitz MP and Menard J

(2011) Effects of a novel aldosterone synthase inhibitor for treatment of primary

hypertension: results of a randomized, double-blind, placebo- and active-controlled

phase 2 trial. Circulation 124:1945-1955.

Cerny MA (2013) Progress towards clinically useful aldosterone synthase inhibitors. Curr Top

Med Chem 13:1385-1401.

Cerny MA, Csengery A, Schmenk J and Frederick K (2015) Development of CYP11B1 and

CYP11B2 assays utilizing homogenates of adrenal glands: Utility of monkey as a

surrogate for human. J Steroid Biochem Mol Biol 154:197-205.

Dorin RI, Qualls CR and Crapo LM (2003) Diagnosis of adrenal insufficiency. Ann Intern Med

139:194-204.

Epstein M, Williams GH, Weinberger M, Lewin A, Krause S, Mukherjee R, Patni R and

Beckerman B (2006) Selective aldosterone blockade with eplerenone reduces

albuminuria in patients with type 2 diabetes. Clin J Am Soc Nephrol 1:940-951.

Fiebeler A, Nussberger J, Shagdarsuren E, Rong S, Hilfenhaus G, Al-Saadi N, Dechend R,

Wellner M, Meiners S, Maser-Gluth C, Jeng AY, Webb RL, Luft FC and Muller DN (2005)

Aldosterone synthase inhibitor ameliorates angiotensin II-induced organ damage.

Circulation 111:3087-3094.

Furet P, Batzl C, Bhatnagar A, Francotte E, Rihs G and Lang M (1993) Aromatase inhibitors:

synthesis, biological activity, and binding mode of azole-type compounds. J Med Chem

36:1393-1400.

Gilbert KC and Brown NJ (2010) Aldosterone and inflammation. Curr Opin Endocrinol Diabetes

Obes 17:199-204.

Guengerich FP, Sohl CD and Chowdhury G (2011) Multi-step oxidations catalyzed by

cytochrome P450 enzymes: Processive vs. distributive kinetics and the issue of carbonyl

oxidation in chemical mechanisms. Arch Biochem Biophys 507:126-134.


at ASPE



ownloaded from


JPET #236463

29

Hardman JL, L.; Gilman, A. ed (2001) Goodman and Gillman's The Pharmacological Basis of

Therapeutics McGraw-Hill, New York.

Hattangady NG, Olala LO, Bollag WB and Rainey WE (2012) Acute and chronic regulation of

aldosterone production. Mol Cell Endocrinol 350:151-162.

Hostetter TH and Ibrahim HN (2003) Aldosterone in chronic kidney and cardiac disease. J Am

Soc Nephrol 14:2395-2401.

Lewis EJ, Hunsicker LG, Bain RP and Rohde RD (1993) The effect of angiotensin-converting-

enzyme inhibition on diabetic nephropathy. The Collaborative Study Group. N Engl J

Med 329:1456-1462.

Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, Ritz E, Atkins RC, Rohde R

and Raz I (2001) Renoprotective effect of the angiotensin-receptor antagonist irbesartan

in patients with nephropathy due to type 2 diabetes. N Engl J Med 345:851-860.

Liu LC, Schutte E, Gansevoort RT, van der Meer P and Voors AA (2015) Finerenone : third-

generation mineralocorticoid receptor antagonist for the treatment of heart failure and

diabetic kidney disease. Expert Opin Investig Drugs 24:1123-1135.

Menard J and Pascoe L (2006) Can the dextroenantiomer of the aromatase inhibitor fadrozole

be useful for clinical investigation of aldosterone-synthase inhibition? J Hypertens

24:993-997.

Menard J, Rigel DF, Watson C, Jeng AY, Fu F, Beil M, Liu J, Chen W, Hu CW, Leung-Chu J,

LaSala D, Liang G, Rebello S, Zhang Y and Dole WP (2014) Aldosterone synthase

inhibition: cardiorenal protection in animal disease models and translation of hormonal

effects to human subjects. J Transl Med 12:340.

Meredith EL, Ksander G, Monovich LG, Papillon JP, Liu Q, Miranda K, Morris P, Rao C, Burgis

R, Capparelli M, Hu QY, Singh A, Rigel DF, Jeng AY, Beil M, Fu F, Hu CW and LaSala

D (2013) Discovery and in Vivo Evaluation of Potent Dual CYP11B2 (Aldosterone

Synthase) and CYP11B1 Inhibitors. ACS Med Chem Lett 4:1203-1207.


at ASPE



ownloaded from


JPET #236463

30

Namsolleck P and Unger T (2014) Aldosterone synthase inhibitors in cardiovascular and renal

diseases. Nephrol Dial Transplant 29 Suppl 1:i62-i68.

Nguyen Dinh Cat A and Jaisser F (2012) Extrarenal effects of aldosterone. Curr Opin Nephrol

Hypertens 21:147-156.

Payne AH and Hales DB (2004) Overview of steroidogenic enzymes in the pathway from

cholesterol to active steroid hormones. Endocr Rev 25:947-970.

Pitt B, Williams G, Remme W, Martinez F, Lopez-Sendon J, Zannad F, Neaton J, Roniker B,

Hurley S, Burns D, Bittman R and Kleiman J (2001) The EPHESUS trial: eplerenone in

patients with heart failure due to systolic dysfunction complicating acute myocardial

infarction. Eplerenone Post-AMI Heart Failure Efficacy and Survival Study. Cardiovasc

Drugs Ther 15:79-87.

RALES (1996) Effectiveness of spironolactone added to an angiotensin-converting enzyme

inhibitor and a loop diuretic for severe chronic congestive heart failure (the Randomized

Aldactone Evaluation Study [RALES]). Am J Cardiol 78:902-907.

Roscioni SS, de Zeeuw D, Bakker SJ and Lambers Heerspink HJ (2012) Management of

hyperkalaemia consequent to mineralocorticoid-receptor antagonist therapy. Nat Rev

Nephrol 8:691-699.

Roumen L, Peeters JW, Emmen JM, Beugels IP, Custers EM, de Gooyer M, Plate R, Pieterse K,

Hilbers PA, Smits JF, Vekemans JA, Leysen D, Ottenheijm HC, Janssen HM and

Hermans JJ (2010) Synthesis, biological evaluation, and molecular modeling of 1-

benzyl-1H-imidazoles as selective inhibitors of aldosterone synthase (CYP11B2). J Med

Chem 53:1712-1725.

Roumen L, Sanders MP, Pieterse K, Hilbers PA, Plate R, Custers E, de Gooyer M, Smits JF,

Beugels I, Emmen J, Ottenheijm HC, Leysen D and Hermans JJ (2007) Construction of

3D models of the CYP11B family as a tool to predict ligand binding characteristics. J

Comput Aided Mol Des 21:455-471.


at ASPE



ownloaded from


JPET #236463

31

Santen RJ, Demers LM, Lynch J, Harvey H, Lipton A, Mulagha M, Hanagan J, Garber JE,

Henderson IC, Navari RM and et al. (1991) Specificity of low dose fadrozole

hydrochloride (CGS 16949A) as an aromatase inhibitor. J Clin Endocrinol Metab 73:99-

106.

Shavit L, Lifschitz MD and Epstein M (2012) Aldosterone blockade and the mineralocorticoid

receptor in the management of chronic kidney disease: current concepts and emerging

treatment paradigms. Kidney Int 81:955-968.

Siragy HM and Carey RM (2010) Role of the intrarenal renin-angiotensin-aldosterone system in

chronic kidney disease. Am J Nephrol 31:541-550.

Unger T (2002) The role of the renin-angiotensin system in the development of cardiovascular

disease. Am J Cardiol 89:3A-9A; discussion 10A.

van Weerden WM, Bierings HG, van Steenbrugge GJ, de Jong FH and Schroder FH (1992)

Adrenal glands of mouse and rat do not synthesize androgens. Life Sci 50:857-861.

Vinson GP and Coghlan JP (2010) Expanding view of aldosterone action, with an emphasis on

rapid action. Clin Exp Pharmacol Physiol 37:410-416.

Young MJ (2013) Targeting the mineralocorticoid receptor in cardiovascular disease. Expert

Opin Ther Targets 17:321-331.


at ASPE



ownloaded from


JPET #236463

32

Footnotes:

Funding: This work was funded by Boehringer Ingelheim Pharmaceuticals Inc.

Address for communication and reprint requests:

Steven M. Weldon

Cardiometabolic Disease Research

Boehringer Ingelheim Pharmaceuticals Inc

175 Briar Ridge Road

Ridgefield, CT 06877

[email protected]


at ASPE



ownloaded from


JPET #236463

33

Figure Legends:

Figure 1. Chemical structures: S-fadrazole (S-FAD), R-Fadrazole (FAD286),

LCI699 and BI 689648

Figure 2. Pharmacokinetic profiles of ASIs in cynomolgus monkeys. Plasma

compound concentrations of aldosterone synthase inhibitors R-S-FAD (n=2), LCI699

(n=3) and BI 689648 (n=3) were evaluated following oral administration at 5 mg/kg in

cynomolgus monkeys. A racemic mixture of R- and S-fadrazole was used for PK

profiling of FAD286 and S-FAD. The two enantiomers R- and S-fadrazole are assumed

to be in equal amounts in analysis samples and presented here as a single curve. All

data plotted as Mean±SD.

Figure 3. ACTH-induced steroid responses in cynomolgus monkeys. Control

responses in non-compound treated monkeys. (A) Maximal ACTH-induced aldosterone

and cortisol production occurs within 15 min following ACTH challenge (shaded grey

area). (B) Baseline aldosterone and cortisol levels for the study cohort (n=94

treatments; see Concise Methods) and ACTH-stimulated steroid responses from the

vehicle control groups (n=32 treatments). Baseline vs. ACTH responses were

compared using the t test and p values <0.05 were considered statistically significant.

Figure 4. In vivo selectivity of BI 689648, LCI699 and FADs. Animals were pre-

treated with dexamethasone approximately 6 hrs prior to administration of test

compound, vehicle and ACTH. ACTH was administered immediately after test


at ASPE



ownloaded from


JPET #236463

34

compound given intravenously or at the respective Cmax/Tmax for orally applied

compounds, to ensure peak compound exposure upon sampling (15 min post-ACTH).

Conscious monkeys received compound doses of 0.003-10 mg/kg to generate

exposure-response relationship curves for (A) S-FAD, n=9 (B) LCI699, n=36; (C)

FAD286, n=24 or (D) BI 689648, n=26. n=number of treatments. Data represents

plasma aldosterone, cortisol and compound levels 15 min following ACTH

administration. Steroid levels are expressed relative to mean values (Percent of Control;

POC) for the Vehicle-treated control groups (n=35) and plotted against plasma

compound (unbound) concentration. Each data point represents individual animal

responses for either aldo or cortisol. Group data were aggregated across studies to

derive Effective Concentrations to achieve 50% inhibition (EC50) of aldosterone and

cortisol calculated by curve-fitting; 95% confidence intervals are shown in parentheses

next to respective EC50’s. Selectivity (X-Fold; double-headed arrow) was calculated

using EC50 CORT divided by the EC50 ALDO.

Figure 5. Selectivity of AS inhibitors based on EC75 ALDO vs EC10 CORT. AS inhibitor

EC data for aldosterone and cortisol (described in Figure 4) was used to calculate a

hypothetical therapeutic index for clinical efficacy based on a postulated level of

aldosterone inhibition of 75% inhibition compared to ≤10% inhibition of plasma cortisol.

Dotted lines represent 25- and 90-Percent of Control Responses and correspond to

interpolated EC75 and EC10 values, respectively. Selectivity values (X-Fold; double

headed arrow) were calculated using EC10 CORT divided by the EC75 ALDO.


at ASPE



ownloaded from


JPET #236463

35

Figure 6. Effect of ASIs on steroid precursor substrates in vivo. Consistent with

the methods described in Figure 4; effect of ASI treatment on ACTH-induced 11-DOC

and 11-DC. Levels are expressed as Fold-Change vs Control responses. Dotted line

represents aldo EC50; solid line represents aldo EC90; dashed line represents cort EC50.


at ASPE



ownloaded from


JPET #236463

36

Table 1. In Vitro Selectivity of BI 689648, LCI699 and FADs. In vitro IC50 values and

selectivity of AS inhibitors determined using the cynomolgus monkey adrenal

homogenate (CAH) assay. Data represent average results from n=1-3 preparations.

Assay Compound

CAH IC50 (nM) S-FAD LCI699 FAD286 BI 689648

Cortisol Synthase a

(CYP11B1) 8 77 94 310

Aldosterone Synthase b

(CYP11B2) 230 10 2.5 2.1

Selectivity c

0.04 7.7 38 149

a S-FAD (n=2); LCI699 (n=1); FAD286 (n=2); BI 689648 (n=1)

b S-FAD (n=3); LCI699 (n=1); FAD286 (n=3); BI 689648 (n=1)

c Calculated as: CAH IC50 CYP11B1

CAH IC50 CYP11B2


at ASPE



ownloaded from



at ASPE



ownloaded from


steven m. weldon, matthew a. cerny, kristina gueneva-boucheva,...

Documents