Discovery of a Novel and Selective Indoleamine 2,3-Dioxygenase (IDO-1) Inhibitor 3-(5-Fluoro-1H-indol-3-yl)pyrrolidine-2,5-dione (EOS200271/PF-06840003) and Its Characterization as a Potential Clinical Candidate

Article abstract of DOI:10.1021/acs.jmedchem.7b00974

Tumors use tryptophan-catabolizing enzymes such as indoleamine 2,3-dioxygenase (IDO-1) to induce an immunosuppressive environment. IDO-1 is induced in response to inflammatory stimuli and promotes immune tolerance through effector T-cell anergy and enhanced Treg function. As such, IDO-1 is a nexus for the induction of a key immunosuppressive mechanism and represents an important immunotherapeutic target in oncology. Starting from HTS hit 5, IDO-1 inhibitor 6 (EOS200271/PF-06840003) has been developed. The structure-activity relationship around 6 is described and rationalized using the X-ray crystal structure of 6 bound to human IDO-1, which shows that 6, differently from most of the IDO-1 inhibitors described so far, does not bind to the heme iron atom and has a novel binding mode. Clinical candidate 6 shows good potency in an IDO-1 human whole blood assay and also shows a very favorable ADME profile leading to favorable predicted human pharmacokinetic properties, including a predicted half-life of 16-19 h.

Full text of DOI:10.1021/acs.jmedchem.7b00974

Subscriber access provided by READING UNIV
Article
Discovery of a novel and selective Indoleamine 2,3-dioxygenase (IDO-1)
inhibitor 3-(5-fluoro-1H-indol-3-yl)pyrrolidine-2,5-dione (EOS200271/
PF-06840003) and its characterization as a potential clinical candidate
Stefano Crosignani, Patrick Bingham, Pauline Bottemanne, Hélène Cannelle, Sandra Cauwenberghs, Marie
Cordonnier, Deepak Dalvie, Frederik Deroose, Jun Li Feng, Bruno Gomes, Samantha Greasley, Stephen E.
Kaiser, Manfred Kraus, Michel Négrerie, Karen A. Maegley, Nichol Miller, Brion W. Murray, Manfred Schneider,
James Solowiej, Albert E Stewart, Joseph Tumang, Vince R Torti, Benoit Van den Eynde, and Martin Wythes
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00974 • Publication Date (Web): 07 Nov 2017
Downloaded from http://pubs.acs.org on November 7, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of a novel and selective Indoleamine 2,3-
dioxygenase (IDO-1) inhibitor 3-(5-fluoro-1H-indol-
3-yl)pyrrolidine-2,5-dione (EOS200271/PF-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
06840003) and its characterization as a potential
clinical candidate
*
,§
†
§
§
Stefano Crosignani , Patrick Bingham , Pauline Bottemanne , Hélène Cannelle , Sandra
§
§
†
¥
†
Cauwenberghs , Marie Cordonnier , Deepak Dalvie , Frederik Deroose , Jun Li Feng , Bruno
§
†
†
†
‖
Gomes , Samantha Greasley , Stephen E Kaiser , Manfred Kraus , Michel Négrerie , Karen
†
†
†
§
†
Maegley , Nichol Miller , Brion W Murray , Manfred Schneider , James Soloweij , Albert E
†
†
†
׀
†
Stewart , Joseph Tumang , Vince R Torti , Benoit Van Den Eynde , Martin Wythes *
§
iTeos Therapeutics, Rue des Frères Wright 29, 6041 Gosselies, Belgium
†
Pfizer Global Research and Development, La Jolla Laboratories, 10770 Science Center Dr., San
Diego, CA 92121, USA
׀
Ludwig Institute for Cancer Research, Université Catholique de Louvain, 74 Avenue
Hippocrate, 1200 Brussels, Belgium
¥
Asclepia Outsourcing Solutions, Damvalleistraat 49, Destelbergen, 9070, Belgium
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
‖
Ecole Polytechnique, Unité Inserm 1182 UMR 7645, Route de Saclay, Palaiseau, 91128, France
KEYWORDS Indoleamine-2,3-dioxygenase 1; Cancer immunotherapy.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ABSTRACT. Tumors use tryptophan-catabolizing enzymes such as Indoleamine 2,3-dioxygenase
(
IDO-1) to induce an immunosuppressive environment. IDO-1 is induced in response to
inflammatory stimuli and promotes immune tolerance through effector T-cell anergy and enhanced
Treg function. As such, IDO-1 is a nexus for the induction of key immunosuppressive mechanism
and represents an important immunotherapeutic target in oncology. Starting from HTS hit 5, IDO-
1 inhibitor 6 (EOS200271/PF-06840003) has been developed. SAR around 6 is described and
rationalized using the X-ray crystal structure of 6 bound to human IDO-1, which shows that 6,
differently from most of the IDO-1 inhibitors described so far, does not bind to the heme iron atom
and has a novel binding mode. Clinical candidate 6 shows good potency in an IDO-1 human whole
blood assay, and also shows a very favorable ADME profile leading to favorable predicted human
pharmacokinetic properties, including a predicted half-life of 16-19 hours.
Introduction
Immune checkpoints blockade, as a breakthrough in cancer therapy with anti-PDL1 (Programmed
death-ligand 1), anti-PD1 (Programmed cell death protein 1), and anti-CTLA4 (cytotoxic T-
lymphocyte-associated protein 4), has demonstrated impressive therapeutic effects in multiple
1
clinical trials. However, only a small minority of patients respond to such therapies. Therefore,
there is a need to identify other tumor immunosuppressive and resistance mechanisms.
ACS Paragon Plus Environment

Page 3 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Indoleamine 2,3-dioxygenase (IDO-1) is a heme-containing dioxygenase that contributes to
metabolic immune regulation by catalyzing oxidative catabolism of the essential amino-acid
tryptophan to N-formyl kynurenine. After this first and rate-limiting step of tryptophan catabolism,
N-formyl kynurenine is subsequently converted to kynurenine and other immunologically active
metabolites via the kynurenine pathway. IDO-1 modifies immune response by producing
biologically active kynurenine-pathway metabolites, natural ligands for the aryl hydrocarbon
receptor (AhR), and by depleting tryptophan to trigger amino-acid sensing signal transduction
pathways.2,3 Given the IDO-1 enzymatic activity-mediated suppression of T-cell responses and
the correlation between IDO-1 expression and poor prognosis and chemoresistance in several
cancer indications,5-7 IDO-1 is a target of interest for cancer immunotherapy. Indeed, potential
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
IDO-1-inhibiting drugs for use in human cancers are now the focus of research and development
efforts.8-9
At the time this project was started, only one IDO-1 inhibitor was in clinical trials,
10
INCB24360. Considering the high attractiveness of the target, as well as the limitations of this
compound, which despite its good potency still requires a relatively high dose to reach full
inhibition of the target, and which has a relatively short half-life in man and requires BID
administration, it was important to discover novel IDO-1 inhibitors, with the potential to become
clinically-relevant therapeutics for oncology immunotherapy.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Some inhibitors of hIDO-1.
Results and Discussion
A library of 178000 diverse compounds was screened using a modification of the published
hIDO-1 enzymatic assay (based on detection of N-formyl kynurenine by measuring absorbance
11
at 320 nm), adapted for 384-well plate format. Among the confirmed hits, the most promising
for subsequent optimization was compound 5. When tested on the p-dimethylamino
benzaldehyde assay (based on detection of kynurenine after formation of an adduct with p-
12
dimethylamino benzaldehyde, PDMAB), this compound showed a moderate potency on hIDO-
1 in the enzymatic assay (IC50 3.0 M) but no activity on tryptophan 2,3-dioxygenase (hTDO-2,
up to 50 M). Given its extremely low molecular weight, the compound had a good ligand
efficiency (LE) of 0.47 highlighting the high efficiency of interaction of the compound with
hIDO-1. Other indole-containing compounds have been reported as hIDO-1 inhibitors, but all
1
3-20
with reported K s of >10 M.
i
ACS Paragon Plus Environment

Page 5 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Of the two enantiomers of 5, only one showed appreciable activity on hIDO-1 in the enzymatic
assay (5a, IC50 1.8 M), whereas the other enantiomer (5b) had only marginal activity on hIDO-
1
(IC50 83 M). The active enantiomer 5a was further profiled to assess its suitability as a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
starting point for further optimization.
Of particular interest, the compound displayed a very attractive ADME profile (Table 1): a very
moderate protein binding, excellent stability in human and mouse microsomes and excellent
permeability in a Caco-2 assay (with no active efflux). Additionally, no significant cytochrome
P450 (CYP) inhibition was observed on five isoforms (IC50s > 50 M on 1A2, 2C19, 2C9, 2D6
and 3A4). In vivo, 5a also displayed an excellent profile in mice, with moderate Clearance (15.6
H
mL/min/kg, 17% Q ), medium volume of distribution and good oral bioavailability (63% and
6
5% respectively at 5 and 30 mg/kg, using a simple suspension administration).
Overall, despite the moderate potency, 5a was considered a good starting point for further
optimization.
Table 1. Profile of hit 5a
hIDO-1 IC50 (M)
1.8
PPB (h/m, Fu)
43% / 55%
Microsomal Clint (h/m; L/min/g)
<10 / <10
-6
Caco-2 Papp (10 cm/s)
28 (efflux ratio 0.77)
CYP450 IC50 (1A2, 2C9, 2C19, 2D6, 3A4; M)
Mouse PK Cl (mL/min/kg)
Mouse PK Vss (L/kg)
>10
15.6
1.2
Mouse PK t1/2 (h)
0.97
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Mouse PK Fz (5 and 30 mg/kg)
63% / 65%
Chemistry. Generally, 3-(3-indolyl)-succinimides could be easily prepared by heating the
corresponding indoles (commercially available or prepared according to published procedures)
with maleimide in acetic acid, either at reflux for several hours, or at higher temperatures in a
microwave apparatus (Scheme 1). For scaling-up, the reaction conditions for this coupling in acetic
acid were found to generate a significant number of by-products, so a milder alternative was
sought. A number of Lewis Acids were tested as catalysts for the reaction, and zinc chloride in
acetonitrile, at 85 °C was found to give excellent conversions and to give a product easy to purify
by a simple recrystallization, at least for the preparation of compound 6. Nitriles 9 and 17 were
obtained from the corresponding bromides 8 and 16, by copper-catalyzed cyanation (Scheme 1).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of Indole Derivatives 1-24.a
aReagents and conditions: (a) Maleimide, acetic acid, 170 °C (microwave irradiation), 2h. (b)
CuCN, NMP, 200 °C (microwave irradiation) 1.5 h
ACS Paragon Plus Environment

Page 7 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
-(1-Naphthyl)-succinimide 25 was prepared starting from 1-napthyl-boronic acid, by
[
RhOH(cod)] -catalyzed coupling with maleimide (Scheme 2). Substituted 3-(1-Naphthyl)-
2
succinimides were prepared from the corresponding 1-nephthalene-amines, via the respective
diazo salts and then succinic acids (Scheme 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Synthesis of Naphthalene Derivatives 25-29.a
aReagents and conditions: (a) Maleimide, [RhOH(cod)]
NaNO , H O, 0 °C-RT, 1h; (c) Maleic anhydride, TiCl
, Et
N, dioxane, 50 °C, 2.5 h; (b) HBF ,
2
3
4
2
2
3
, water/acetone; (d) urea, 180 °C, 1h.
Biological activity. Systematic SAR was performed on both halves of the molecule (Table 2). It
soon became apparent that the succinimide could not be modified without very significant drops
in activity (list of molecules with full biological data in Supporting Information). Removal of
either or both of the carbonyl groups to give either the regioisomeric pyrrolidinones or the
corresponding pyrrolidine, or ring expansion to the corresponding glutarimide (either
regioisomer) all gave compounds with IC50 > 30 M. Similarly, substitution with a methyl group
on either the succinimide nitrogen or on either of the two succinimide aliphatic carbon atoms
(with any stereochemistry) resulted in loss of more than an order of magnitude of activity. Also,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
replacing the succinimide CH
2
group with either a nitrogen or oxygen atom resulted in
compounds with IC50 > 50 M. Finally, inserting a CH linker between the succinimide and
2
indole groups also resulted in loss of activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. SAR of Indolyl and Naphthyl Succinimides
Compound
X
R1
R2
IC50 hIDO-1
a
(
M)
3.0
1.8
83
5
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
5
5
ab
bc
H
6
F
0.15
0.12
54
6
ab
bc
F
6
F
7
7ab
7bc
8
Cl
Cl
Cl
Br
CN
OMe
Me
0.83
2.2
>50
0.37
1.2
53
9
1
1
1
0
1
2
53
CF
3
9.7
ACS Paragon Plus Environment

Page 9 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
3
4
5
6
7
NH
NH
H
H
H
H
H
H
F
F
Cl
Br
CN
OMe
Me
F
3.4
2.1
NH
0.42
1.7
NH
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NH
>50
54
18
19
20
NH
NH
1.8
NH
F
Cl
Cl
Cl
Br
Br
Br
Me
H
0.49
0.58
>50
0.29
0.62
8.0
2
0ab
0bc
NH
F
2
NH
F
2
1
NH
F
2
1ab
1bc
NH
F
2
NH
F
2
2
3
4
5
NH
F
46
2
2
2
CH=CH
CH=CH
CH=CH
CH=CH
CH=CH
H
F
18
H
4.6
H
Cl
H
F
1.7
26
27
H
>50
>50
Cl
a
b
c
average of n≥2 independent experiments. First enantiomer (by elution order in chiral HPLC). Second
enantiomer (by elution order in chiral HPLC).
Exploration of the indole group also quickly showed that some parts of the molecule were not
free for substitution: methylation of the indole NH or of the C2 resulted in significant drop in
activity, whereas substitution in positions C4 and C7 with either methyl, chlorine or methoxy
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
also gave compounds with IC50s > 50 M (full list of compounds with biological data in
Supporting Information). On the other hand, substitution on positions C5 and C6 of the indole
turned out to be more tolerated, and in some cases beneficial. In position 5, a clear preference
was identified for halogens, with increase in activity for the fluoro (6, IC50 0.15 M), chloro (7,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
IC50 0.83 M) and bromo (8, IC50 0.37 M) substitutions, whereas substitution with methyl (11,
IC50 53 M), methoxy (10, IC50 53 M) and CF (12, IC50 9.7 M) groups led to significant loss
3
of activity. In position 6, substitution with halogen was neutral for potency (F, 13, IC50 3.4 M);
Cl, 14, IC50 2.1 M) or advantageous (Br, 15, IC50 0.42 M), whereas other substitutions led to
loss of potency, with the exception of the nitrile substituent (16, IC50 1.7 M).
Having identified 6 as the most potent mono-substituted compound, we tried a small number of
doubly-substituted compounds based on this scaffold: as expected, adding a methyl group in
position 6 (22, IC50 46 M) led to a significant drop in activity, but unfortunately adding either a
fluorine (19, IC50 1.8 M), chlorine (20, IC50 0.49 M) or bromine (21, IC50 0.29 M) in
position 6 did not yield any improvement in potency compared to 6.
Finally, several replacements of the indole core with alternative heterocycles were attempted.
Replacement with either indazole, benzimidazole, all four regioisomeric azaindoles, different
isoquinoline and quinoline regioisomers, different indole regioisomers as well as substituted
benzofurans and benzothiophenes were all found to be significantly less active (IC50 > 20 M,
full list of compounds with biological data in Supporting Information). The only replacement
which maintained a significant potency on the target was found to be the 1-napthalene compound
2
3 (IC50 18 M). It was thus decided to explore if appropriate substitution with halogens might
yield potent compounds. Unfortunately, the most potent compound in this sub-series was 24,
ACS Paragon Plus Environment

Page 11 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
which was still significantly less potent than 6 (IC50 4.6 M), while having significantly higher
cLogP, and was thus not further pursued.
Compound 6 was separated into its pure enantiomers, 6 and 6b. As observed for 5, only one of
the two enantiomers 6a was responsible for the inhibition of IDO-1 in the enzymatic assay. The
absolute configuration of the active enantiomer of 6 was determined by X-ray crystallography
and found to be (R).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Subsequent to the discovery of 6a further structural insights to its IDO-1 inhibition binding mode
were sought, by X-ray and spectroscopic methods. IDO-1 has an active site heme cofactor which
enables mechanistic characterization of inhibitor binding interactions. Heme is a porphyrin ring
with an iron at its center which absorbs light in UV-vis spectrum maximally at a wavelength
around 400 nm depending on the oxidation and coordination states of its iron – the Soret band.22
Inhibitors that coordinate the heme iron will shift the Soret band λmax. Inhibitors that do not
directly coordinate the iron will only modulate the Soret band because the heme is highly-
23
sensitive to changes in the polarity of its surroundings. The observed Soret bands for the
ferrous IDO-1 (λmax 427 nm) and ferric IDO-1 (λmax 405 nm) are in-line with values previously
2
+
3+
24
reported (Fe form, 429; Fe form 404 nm). The IDO-1 inhibitors did not shift the Soret λmax
for either ferric or ferrous forms of IDO-1 which is not consistent with direct binding to the heme
iron (Figure 2). The binding affinity of the IDO-1 inhibitors was determined by measuring the
decrease in the Soret band absorbance. 6a is shown to bind weakly to both the ferric and ferrous
forms of IDO-1 in the presence of oxygen with apparent binding constants of 9 and 6 M,
respectively. Under conditions where the oxygen content is depleted, 6a binds to the ferric form
of IDO-1 with an apparent binding constant of 0.16 μM. This value is in-line with the IC50 value
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
determined kinetically and suggests that the ferric form, without oxygen bound at the heme, is
the form of the enzyme that the inhibitor binds to during catalysis. Further analysis of the ferric
app
form with depleted oxygen and added tryptophan showed that the affinity (K
d
0.15 μM) is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
similar to the value in the absence of tryptophan (Table 3). Since tryptophan does not bind this
form of the enzyme tightly, 6a is unlikely to be competitive with tryptophan. These findings are
consistent with 6a being a tryptophan non-competitive, non-heme binding IDO-1 inhibitor.
Similar results were shown for the racemic compound, 6 indicating that the behavior of the
racemic mixture and the pure enantiomer are comparable.
Table 3. Biochemical affinity analysis for different forms of IDO-1. Inhibition was determined
by a decrease of the Soret band intensity (ferrous form λmax = 427 nm, ferric form λmax = 405
nm). Data are represented as geometric mean in μM, plus 95% confidence interval in
parentheses determined from two independent experiments. Kdapp ferrous are derived from
titrating the inhibitors into the ferrous form of IDO-1. Kdapp ferric is derived from titrating the
inhibitors into the ferric form of IDO-1 without measures to remove oxygen. Kdapp ferric
(depleted O2) is derived from titrating the inhibitors into the ferric form of IDO-1 following
oxygen depletion. Kdapp ferric (depleted O
, +Trp) is derived from titrating the inhibitors into the
2
ferric form of IDO-1 in the presence of tryptophan following oxygen depletion. Kdapp is the
apparent dissociation constant. NT = not tested.
IDO-1 inhibitor binding to different forms of hIDO1 (M)
app
app
app
app
IDO-1 Inhibitor
K
d
Ferrous
K
d
Ferric
K
d
Ferric
K
d
Ferric
(depleted O
2
)
(depleted O2, +Trp)
6
14 (12-16)
6 (3-12)
22.3 (22.1-22.6)
7 (5-11)
0.32 (0.27-0.38)
0.16 (0.13-0.19)
NT
6
a
0.15 (0.12-0.19)
ACS Paragon Plus Environment

Page 13 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 2, panel A. Evaluation of 6. (A) Ferrous hIDO1. (B) Ferric hIDO1 in the presence of
oxygen. (C) Ferric hIDO1 after oxygen depletion (D) Ferric hIDO1 with tryptophan after
oxygen depletion.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2, panel A. Ferrous hIDO1
4
2 7 n m
[
In h ib ito r],  M
0
0
.7
.6
0
0
2
.6
4
1
0
1
5
0 .5
0 .4
0 .3
4 0 0
4 2 0
4 4 0
4 6 0
4 8 0
W a v e le n g th (n m )
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 2, panel B. Ferric hIDO1 in the Presence of Oxygen
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
.8
4 0 5 n m
[
Inh ibitor],  M
0
1
0 .6
4
1
0
3
0
0
.4
0 .2
3
5 0
4 0 0
W a v e le n g th (n m )
4 5 0
ACS Paragon Plus Environment

Page 15 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 2, panel C. Ferric hIDO1 after Oxygen Depletion
4
0 5 n m
[
In hib ito r,  M ]
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0 .0 4
0
.0 8
0 .1 2
.2
.8
0
0
0
.8
.6
.4
0
1
3
.8
5
.8
9
.8
3 8 0
4 0 0
4 2 0
4 4 0
W a v e le n g th (n m )
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 2, panel D. Ferric hIDO1 plus Tryptophan, after Oxygen Depletion
4
0 5 n m
0
.9
[
Inh ibitor],  M
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0 .0 4
.0 8
0
0
.1 2
0 .8
0 .7
0 .6
0 .5
0 .2
0
.6
3 6 0
3 8 0
4 0 0
4 2 0
4 4 0
W a v e le n g th (n m )
To confirm this binding mode hypothesis, and to map the interactions between 6a and hIDO-1, an
X-ray of the 6a-IDO-1 complex was obtained (Figure 3, PDB accession code 5WHR). This
structure shows that the succinimide ring of 6a sits parallel to the heme, at a short distance from
it, displacing the apical water molecule, while the indole ring lies in the lipophilic pocket. There
is no direct interaction between 6a and the iron atom, despite the close spatial proximity, which
justifies the absence of an apical water molecule. Remarkably, all four heteroatoms of 6a
participate in hydrogen-bonding. The indole NH interacts with side chain of Ser167, while the
succinimide NH hydrogen bonds to the heme carboxylic acid which adopts a different rotomer
compared to other IDO-1 structures in the PDB. The two carbonyls interact to the main chain NH
of Ala264 and Thr379. Indeed, residues 376-383 adopt a unique conformation in the 6a bound
structure by forming a loop that folds up to close off the binding pocket and positions Thr379 to
ACS Paragon Plus Environment

Page 17 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
interact with the ligand (Figure 3, loop residues highlighted in dark blue C trace). In addition to
the hydrogen bond interactions, several aromatic groups (Tyr126, Phe163, Phe164) are
appropriately positioned to give π-π interactions with the indole aromatic core. Additional
lipophilic residues also are in close contact with the indole group (Leu234, Val130, Cys129).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The very tight packing of the molecule within the IDO-1 active site clearly explains the lack of
activity of most of the analogues bearing larger substituents around the indole, as well as the
challenge to change any of the succinimide features. Quaternarization of the chiral carbon, which
would avoid racemization, is also prevented by its closeness to the heme.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A
B
C
Figure 3. 2.28Å X-ray structure of the complex hIDO-1-6a. Protein is represented in light blue
C trace with key interacting residues in orange and green stick representation. The ligand, 6a, is
drawn in magenta and the heme in cyan stick representation. (A) Shows the key hydrogen bond
interactions; (B) Connolly surface of the protein is shown in grey highlighting the closed binding
pocket as a result of the unique ordering of loop A376-D383 (dark blue C trace). (C)
Superposition of the crystal structure of IDO-1 bound to 6a versus PDB IDO-1 structure 5EK3
(
grey C trace with grey heme in stick representation, ligand removed for clarity). The unique
conformation of residues 376-383 which form a loop in the 6a bound structure are shown in dark
ACS Paragon Plus Environment

Page 19 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
blue Ctrace. Also shown is the different rotomer adopted by the heme carboxylic acid to interact
with the succinimide NH of 6a.
To confirm activity of 6 and 6a on the hIDO-1 enzyme, a new assay was developed, using a
more sensitive high-throughput mass spectrometry assay which monitors tryptophan and
kynurenine. The new assay format allowed us to use lower concentrations of enzyme (1.5 nM)
and to have better reproducibility. The IC50s of 6 and 6a in this assay were 0.40 and 0.20 M,
respectively; whereas the other enantiomer 6b had much weaker affinity for hIDO1 with an IC50
of 38 M. Compounds 6 and 6a were found to be less active on the mouse enzyme (IC50 1.5 and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
.3 M, respectively), whereas the activity on dog IDO-1 was more similar to human IDO-1
(IC50 0.59 and 0.20 M). Compound 6 was found to have very weak activity against hTDO-2,
with an IC50 of 140 M.
1
0
In cellular assays, 6 and 6a were first tested in the classical hIDO-1 HeLa assay: HeLa cells are
well known to produce hIDO-1 after stimulation with interferon gamma (INF). Incubation of
the cells for 24 h after stimulation produces significant levels of kynurenine. In addition to the
HeLa cell system, 6 and 6a were also tested using a second cellular assay based on THP1 (acute
monocytic leukemia) cells, to test the effect on a cell line derived from a myeloid-derived cell
line. In these cellular assays, 6 and 6a showed activity both in the HeLa assay (IC50 1.8 and 1.0
M respectively), as well as in the LPS/INF-stimulated THP1 cells (IC501.7 and 1.1 M).
Table 3. Summary of In Vitro Pharmacology Data for 6 and 6a
6
6a
0.20
hIDO-1 IC50 (LCMSMS method, M)
0.41
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mIDO-1 IC50 (LCMSMS method, M)
1.5
7.3
dIDO-1 IC50 (LCMSMS method, M)
HeLa cell IC50 (M)
0.59
1.8
0.20
1.0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
THP1 cell IC50 (INF/LPS, M)
hTDO-2 IC50 (M)
1.7
1.1
>50
>50
Whole Blood IC50 (M)
4.7 (total)
2.5 (total)
1
.1 (unbound)
Interestingly, we also observed a significant activity of enantiomer 6b in the HeLa assay (IC50 13
M), with only a 13-fold difference in potency between the two enantiomers, compared to a 190-

fold difference in activity in the enzymatic assay. Considering that the chiral center of 6 is an
enolizable carbon, this was interpreted as a sign that significant racemization of the compound
could be happening in the conditions of the cellular assay, but not of the enzymatic assay.
1
6
As reported in more detail elsewhere compounds 6 and 6a were also tested in a human whole
blood assay: in this assay, whole blood from healthy volunteers, in presence of different
concentrations of inhibitors, is treated with INF and LPS to induce IDO-1. After 24h, the
kynurenine produced can be measured by LCMSMS. The presence of full plasma and red blood
cells makes this assay more physiologically relevant, as plasma protein binding and red blood
cell distribution will affect the potency of the compound, so that only the unbound fraction of
each compound can really inhibit the target. The assay was repeated on 9 different donors, to
account for differences due to any possible difference in immune status or metabolism of each
individual donor. Despite the moderate potency in isolated cell systems, the low protein binding
makes the 6 and 6a maintain a good efficiency in the whole blood assay, with IC50 of
ACS Paragon Plus Environment

Page 21 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
respectively 4.7±2.4 M and 2.5±1.5 M. Compound 6 was tested against hIDO2 and it is
inactive up to 200 M. The individual enantiomers were not tested due to very limited enzyme
supply.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
5
As reported in more detail elsewhere compounds 6 was evaluated for target selectivity,
cardiovascular effects, and genotoxic potential in a series of in vitro assays. To assess the off-
target binding potential 6 was tested at a concentration of 200 M in a CEREP Wide Ligand
Profile Screen, where it showed good selectivity.15 The potential for cardiovascular impact,
specifically QT prolongation, was tested using the hERG assay, where 6 showed less than a 50%
inhibition of the hERG channel up to 300 µM
ADME. Although 6a showed optimal ADME properties similar to those of 5a, it was rapidly
converted to the inactive enantiomer 6b in buffer and plasma of preclinical species and humans.
Studies of 6a with phosphate buffer (pH 7.4) and plasma from animal species were conducted to
assess the degree of inter-conversion of 6a to 6b (Table 4). While the conversion of 6a to 6b was
minimal in phosphate buffered saline (pH of 7.4), racemization seemed to be rapid in plasma of
preclinical species and humans (Table 4, Figures 4a and 4b). For example, 65% of 6b was
produced in 6 hr, when 6a was incubated with plasma from humans (Figure 4b). Rapid inter-
conversion in the plasma relative to phosphate buffer suggested a possibility of an enzyme
catalyzed process.
Interestingly, the ratio of the two enantiomers in human plasma at 48 h was not at the theoretical
level of 1:1 irrespective of which enantiomer was present at T0. This was attributed to higher free
fraction of 6a (fup 0.449) relative to 6b (fup 0.286). When corrected for protein binding, the ratio
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of the free fractions of the two enantiomers after 48 h incubations were reasonably close to the
theoretical 1:1 ratio.
Table 4. Racemization Experiment: Formation of 6a/6b After Incubation for 48 h of 6a in Various
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Biological Media
Matrices
6a Remaining (%)
6b Formed (%)
Phosphate buffered saline (pH
8
9
11
7
.4)
Mouse Plasma
Dog Plasma
35
30
65
70
76
6.6
Rat Plasma
27
Acidified Plasma1
Dog plasma was acidified
93.5
1
1
0 0
H u m a n
M
o u s e
8
0
D o g
6 0
P B S
D o g ( A c id if ie d )
4
2
0
0
0
0
1 2
2 4
T i m
3 6
4 8
e
( h r )
ACS Paragon Plus Environment

Page 23 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 4a. Time-dependent racemization of 6a in plasma across species and phosphate-buffered
saline (PBS).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
0 0
8 0
%
%
6 a r e m a i n in g
6 b f o r m e d
6
4
0
0
2 0
0
0
1 2
2 4
T i m
3 6
4 8
e
( h r )
Figure 4b. Time-dependent racemization of 6a in human plasma. Concentrations of 6a (empty
triangles) and 6b (filled circles) are measured using a chiral LC/MS-MS method
The low inter-conversion barrier negated the benefits of using single enantiomer for further in vivo
biological testing and characterization of the pure enantiomer (6a). Advancing the racemate
instead of the pure enantiomer was further justified since conversion of 6a to 6b was also observed
in vivo in all species following an IV dose of 6a (Table 5). In fact the exposure of 6a after oral
administration of either 6a or 6 to cynomolgus monkeys suggested only a 20% difference in the
exposure of 6a as depicted in Figure 5. Rapid racemization of 6a both in vitro and in vivo also
justified the advancement of the racemate 6 rather than the active enantiomer 6a into humans.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 5. Exposure (AUCinf) of 6a and 6b after intravenous administration (dose, 1 mg/kg) of the
active enantiomer 6a in rat, dog and monkey.
Rat
Dog
696
Monkey
6808
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
in vivo AUC 6a
2274
(h*ng/ml)
in vivo AUC 6b
153
124
3516
0.52
(h*ng/ml)
in vivo ratio AUC 6b
0
.067
0.22
/
AUC 6a
A
ACS Paragon Plus Environment

Page 25 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
B
Figure 5. Total concentrations of 6a (blue diamonds) and 6b (red squares) dosing of either 6
panel A) or 6a (panel B) in monkey (PO dosing at 10 mg/kg)
(
Consequently, the pharmacokinetics of the active enantiomer (6a) was evaluated after single IV
administration of the racemic mixture 6 in CD 1 mice, Wistar Han rats and beagle dogs followed
by quantitation of the active enantiomer using a chiral LC-MS/MS assay. After a single 1 mg/kg
IV dose of 6, the mean CLp, Vss, and t½ of 6a was 7.2 mL/min/kg, 0.79 L/kg, and 1.3 hours,
respectively, in mouse and 3.4 mL/min/kg, 0.64 L/kg, and 2.2 hours, respectively, in rat. In dogs,
the CLp, Vss, and t½ of 6a following a 1 mg/kg IV dose of 6 was 20 ml/min/kg, 0.96 L/kg, and
0
.80 hours, respectively (Table 6; Figure 6).
Table 6: Mean Pharmacokinetics of 6a in Mouse, Rat, Dog, and Monkey Following Intravenous
Administration of 6.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Dose
AUCinf
CL
p
V
ss
T
1/2
Species
Sex/n
(ng-hr/kg)
(
mg/kg)
(mL/min/kg) (L/kg)
(hr)
CD-1
Mouse
Wistar-Han
Rat
Male/3
Male/3
1
1160
7.21
0.787
2.5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
2520
423
3.44
20.0
0.639
0.957
2.22
Beagle Dog Male/3
0.802
A
B
1 0 0 0 0
1
0 0 0
1 0 0
1 0
1
0 0 0
1 0 0
1
0
1
M o u se
M o u se
R a t
R a t
1
D o g
D o g
0
.1
0
.1
0
2
4
6
8
0
2
4
6
8
T im e (h r)
T im e (h r)
Figure 6: Plasma profile of 6a following A) an intravenous dose (1 mg/kg) and B) oral dose (3
mg/kg) of the racemate 6 to CD-1 mouse, Wistar-Han rat and Beagle dog.
These results suggested that 6a exhibited a low plasma clearance (CLp) in the mouse and rat, and
moderate CLp in the dog with a low volume of distribution (Vss) in these species (Table 6). Low
clearance of 6a in mouse and rat and moderate to high clearance in the dog was consistent with
the projected hepatic clearance values estimated from in vitro incubation of 6 with hepatocytes
from the preclinical species (Table 7).
Table 7. Projection of Hepatic Clearance (CLhep) of 6 Following Incubation with Hepatocytes
ACS Paragon Plus Environment

Page 27 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In Vivo
Species
Clint app
CLhep
CLblood1
6
µL/min/10 cells (mL/min/kg) (mL/min/kg)
Rat
Dog
Mouse
2.3
10.3
2.4
2.9
16.6
4.5
3.7
24.3
8.4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
. In vivo CLblood was estimated from the ratio of CLplasma of 6a (Table 6) and blood to plasma
ratio (B/P) for mouse rat and dog which was 0.863, 929 and 0.824, respectively.
Absorption of 6a was assessed after oral administration of the racemate 6 to CD-1 mice, Wistar-
Han rat and Beagle dog. Compound 6a was rapidly absorbed in all three preclinical species when
the racemate 6 was dosed at 3 mg/kg as indicated by the time to reach maximum concentration
(Tmax) ranging from 0.25 to 1.3 hours. Oral bioavailability in the mouse and rat was 59 and 94%
respectively, and 19% in the dog (Table 8).
Table 8: Mean Pharmacokinetics of 6a in Mouse, Rat, Dog, and Monkey Following Oral
Administration of 6.
Dose
C
max
T
max
AUCinf
t
½
F
Species
Sex/n
(%)
(
mg/kg) (ng/mL) (hour) (ng•hr/mL) (hour)
CD-1
Mouse
Male/3
Male/3
3
538
0.58
2170
2.48
2.42
59.3
94.0
Wistar Han
Rat
3
3
985
219
1.30
0.25
6940
237
Beagle Dog Male/3
0.840 18.6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The product of fraction absorbed and fraction of 6a escaping the gut (fa × fg) was calculated using
the equation 1;
퐶퐿
퐹 = 푓 푥 푓 푥 (1 −
)
Equation 1
푎
푔
푄
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
where F, bioavailability; fa, the fraction absorbed from the GI tract; fg, the fraction escaping
metabolism in the intestine; CL, the blood clearance which was estimated from CLp and blood to
plasma ratio in the appropriate species and Q, the hepatic blood flow (mouse, 90 ml/min/kg; rat,
70 ml/min/kg and dog, 40 ml/min/kg) and ranged from ~0.5 to 0.99 (mean fa x fg was computed
to be 0.70). Good absorption in vivo was consistent with the in vitro permeability assessment of
6
in Ralph-Russ canine kidney (RRCK) cells which demonstrated high intrinsic permeability with
-6
an apical to basal (Papp A-B) value of 25.5 x 10 cm/sec. Further in vitro studies in Madin –Darby
canine kidney (MDCK) cells transfected with the multi-drug resistance gene (MDCK-MDR) that
encodes for human Pgp showed minimal evidence of efflux [(basolateral to apical/apical to
basolateral (BA/AB) efflux ratio <1.2 up to concentrations of 10 µM] indicating that 6 (or 6a) was
a very weak substrate Pgp substrate in humans. Based on these in vivo and in vitro results, the
absorption of 6a in humans was anticipated to be good after dosing oral administration of the
racemate.
Compound 6a was weakly bound to plasma proteins and was equally distributed between red blood
cells and plasma. The unbound fraction (fu) of 6a following incubation of the racemate with
plasma from preclinical species and humans ranged from 0.25 in the rat to 0.45 in humans while
the blood to plasma ratio of 6a ranged from 0.824 to 1.1 when 6 was allowed to equilibrate between
blood and plasma. Assessment of CNS exposure of 6a in the cerebral spinal fluid (CSF) and brain
of rats after oral administration of the racemate 6 indicated that 6a was a brain penetrant with total
brain to plasma ratio (Kp = 0.51) and ratio of free concentrations in the brain (B ) to free
u
ACS Paragon Plus Environment

Page 29 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentrations in the plasma (P
CSF/P = 0.49).
u
), B
u
/P = 0.20. The ratio of CSF to free plasma concentrations
u
(
u
Investigation into clearance mechanisms indicated that the percentage of unchanged 6a excreted
in the urine and bile of rat and the urine of dog was limited (< 1%) following dosing of 6 suggesting
that 6a was primarily cleared via metabolism in preclinical species. Preliminary in vitro
metabolism studies with 6 in hepatocytes from mouse, dog, and humans indicated oxidation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(hydroxylation) and hydrolysis as the primary routes of metabolism. All metabolites detected in
vitro in human hepatocytes were also present in the hepatocytes evaluated from the preclinical
species. CYP phenotyping studies with 6a in cryopreserved human hepatocytes following
incubation with and without 1-aminobenzotriazole (ABT), a non-specific CYP enzyme inhibitor,
suggested that CYP450 dependent metabolism accounted for approximately 61% of the metabolic
turnover and non-CYP450 dependent hydrolytic metabolism accounted for 39%. Incubation with
specific inhibitors of the major CYP isoforms indicated that CYP1A2 was the primary isozyme
that was responsible for oxidative metabolism of 6a.
Inhibition and induction studies with primary CYP isoforms and major efflux transporters were
conducted to assess the risk of pharmacokinetic drug interactions. These studies revealed that 6
was a very weak inhibitor of CYPs with IC50 values greater than 100 µM for most major CYP
isozymes except 2C19 (IC50 value was 78 µM). Additionally, 6 did not exhibit metabolism
dependent (time-dependent and NADPH-dependent) or time dependent (NADPH-independent)
inhibition of the major CYP enzymes investigated. Likewise, in vitro experiments to investigate
the potential for 6 to inhibit P-gp or BCRP mediated transport of digoxin and pitavastatin as known
P-gp and BCRP substrates in MDCK-MDR or MDCK-BCRP cells indicated an IC50 of >300 M
and 65.4 M, respectively, suggesting weak inhibition of these transporters as well. Similarly, 6
ACS Paragon Plus Environment