Discovery of Imidazopyridines as Potent Inhibitors of Indoleamine 2,3-Dioxygenase 1 for Cancer Immunotherapy

,3-Dioxygenase 1 for Cancer Immunotherapy

Letter

Discovery of Imidazopyridines as Potent Inhibitors of Indoleamine

Liping Zhang,* Emily C. Cherney,* Xiao Zhu, Tai-an Lin, Johnni Gullo-Brown, Derrick Maley,

Kathy Johnston-Allegretto, Lisa Kopcho, Mark Fereshteh, Christine Huang, Xin Li, Sarah C. Traeger,

Gopal Dhar, Aravind Anandam, Sandeep Mahankali, Shweta Padmanabhan, Prabhakar Rajanna,

Venkata Murali, Thanga Mariappan, Robert Borzilleri, Gregory Vite, John T. Hunt, and Aaron Balog*

Cite This: ACS Med. Chem. Lett. 2021, 12, 494 −501

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: Indoleamine 2,3-dioxygenase 1 (IDO1) has been

identiﬁed as a target for small-molecule immunotherapy for the

treatment of a variety of cancers including renal cell carcinoma and

metastatic melanoma. This work focuses on the identiﬁcation of

IDO1 inhibitors containing replacements or isosteres for the amide

found in BMS-986205, an amide-containing, IDO1-selective inhibitor

currently in phase III clinical trials. Detailed subsequently are eﬀorts

to identify a structurally diﬀerentiated IDO1 inhibitor via the pursuit

of a variety of heterocyclic isosteres, leading to the discovery of highly

potent, imidazopyridine-containing IDO1 inhibitors.

KEYWORDS: Indoleamine 2,3-dioxygenase 1, IDO1, immuno-oncology, amide isostere

he clinical success of monoclonal antibody checkpoint

inhibitors such as Yervoy (Bristol Myers Squibb, BMS),

Figure 2. Putative binding mode of 4 (magenta) in hIDO1, based on

an X-ray cocrystal structure of compound 6 (green).

Figure 1. IDO1 inhibitors evaluated in clinical trials.

Received: January 12, 2021

Accepted: February 26, 2021

Published: March 2, 2021

Keytruda (Merck), and Opdivo (BMS) has received much

attention in the ﬁeld of immuno-oncology (IO). These IO

therapies leverage a patient’s native immune system to reverse

tumor-induced immune suppression and enhance immune

response toward cancer. More recently, the eﬀect of

metabolism, including amino acid catabolism, on immune

2021 American Chemical Society

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

Letter

Scheme 1. Synthesis of Heterocyclic Amide Isosteres and Replacements Starting from Common Intermediate Carboxylic Acid

Reagents and conditions: (a) HATU, NMM, 2-amino-1-(4-chlorophenyl)ethan-1-one (64%); (b) POCl , 100 °C (56%); (c) NH OAc, HOAc,

EtOH, 100 °C (17%); (d) (i) HATU, NMM, tert-butyl carbazate; (ii) HCl (89% over 2 steps); (e) HATU, NMM, 4-chlorobenzoic acid (83%);

(

f) POCl , 90 °C (55%); (g) pyridine, 4-chlorobenzamidine·HCl salt, 110 °C (31%); (h) DIPEA, 1-isocyanato-4-methoxybenzene (93%); (i)

POCl , 90 °C (26%); (j) DPPA, TEA, toluene, 70 °C; (k) 4-methoxybenzohydrazide, DIPEA (54% over 2 steps); (l) POCl , 100 °C (52%); (m)

(i) SOCl , DMF (ii) CH Cl , TEA, 4-ﬂuoro-1,2-phenylenediamine, (n) MsOH, 90 °C (50% over 3 steps).

response in the tumor microenvironment (TME) has provided

a basis for exploring IO targets capable of perturbation by

small molecules. One such strategy for small-molecule cancer

immunotherapy involves the inhibition of indoleamine 2,3-

dioxygenase 1 (IDO1) to decrease local kynurenine levels in

strategy for the re-establishment of immunogenic responses to

cancer.

Several IDO1 inhibitors have entered clinical trials for the

treatment of cancer (Figure 1). Epacadostat (INCB24360)

(

1), an orally active, hydroxyamidine-containing small

molecule developed by Incyte, entered Phase III clinical trials

the TME and restore cancer immunity.

in 2017. Preclinically, epacadostat selectively inhibits IDO1

IDO1 is a monomeric, heme-containing dioxygenase

enzyme that degrades tryptophan by catalyzing the initial,

rate-limiting step of tryptophan metabolism. This step involves

the oxidative cleavage of the indole 2,3-double bond of

tryptophan to give N-formyl kynurenine. Subsequent hydrol-

enzymatic activity and eﬀectively regulates the functions of

various immune cells including T cells, natural killer (NK)

cells, and DCs. In vivo, epacadostat (1) suppressed IDO1

activity in mouse and dog plasma and inhibited tumor growth

,10

in a lymphocyte-dependent manner in mice.

Results from a

ysis of N-formyl kynurenine yields kynurenine. IDO1

phase I/II combination study with epacadostat and ipilimumab

in melanoma patients showed a 56% overall response rate

among 54 eﬃcacy-evaluable patients and a median pro-

promotes tumoral immune escape from host immune

surveillance and plays an important role in tumor-associated

immunosuppression, leading to tolerance toward tumors. Local

depletion of tryptophan as well as accumulation of kynurenine

have been shown to elicit immunomodulatory activity. Eﬀects

11a

gression-free survival of 12.4 months. However, a phase III

study evaluating the combination of epacadostat and keytruda

for the treatment of metastatic melanoma demonstrated no

include suppression of the T eﬀector (T ) cell immune

11b,c

eff

signiﬁcant improvement in progression-free survival.

response, induction of naive T cell diﬀerentiation into

Navoximod (2), an imidazoisoindoline compound, was

discovered by NewLink Genetics and later licensed to

Genentech as GDC-0919. Rights to this molecule were later

returned to NewLink after phase I clinical trial results were

regulatory T (T ) cells, and both activation of as well as a

reg

decrease in dendritic cell (DC) function. IDO1 is widely

expressed in antigen-presenting cells (DCs, macrophages) and

tumor cells, as well as epithelial and vascular endothelial cells.

Furthermore, IDO1 can be induced in the tumor micro-

environment in response to inﬂammation and T cell activation

disclosed. The third candidate, iTeos/Pﬁzer’s compound PF-

6840003 (3), was dosed as a single agent, once daily, in a

phase I study in patients with malignant gliomas. In January

following immunotherapy, radiotherapy, or chemotherapy.

018, Pﬁzer stopped the development of the compound due to

This observation is indicative of the potential for IDO1

inhibitors in the context of combination therapies. High IDO1

expression in the tumor or tumor-draining lymph nodes is

generally associated with poor prognosis and reduced survival

a lack of eﬃcacy.

In February 2015, BMS expanded its immuno-oncology

pipeline in an agreement with Flexus Biosciences and acquired

linrodostat (4, BMS-986205). Preclinically, compound 4

exhibits potent inhibitory activity against IDO1 (human

in patients. Therefore, inhibition of IDO1 is a promising

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

Letter

Table 1. Eﬀects of Heterocyclic Amide Replacements on

IDO1 Inhibition Potency and Metabolic Stability

Scheme 2. General Schemes for the Synthesis of Acid

Intermediates with Modiﬁed R and Quinoline R

Substitution

Reagents and conditions: (a) Pd(PPh ) , K CO , dioxane, H O, 100

C; (b) Pd−C, MeOH, HCOONH , 80 °C or Pd−C, MeOH, H ;

(

c) LiOH, H O, THF, rt−70 °C; (d) 1,3-dimethyl-3,4,5,6-tetrahydro-

(1H)-pyrimidinone, lithium diisopropylamide, THF, R X, − 78 °C−

rt; (e) NaH, DMSO, 60 °C, 4 h.

structure of compound (6) (a closely related analogue is

Data reported as average test results (N = 2, unless otherwise noted).

14e

shown in green). This model reveals that the chlorophenyl

moiety makes an edge-to-face pi-stacking interaction with

Tyr126. The amide NH forms a key H-bond interaction with

Ser167. The cyclohexyl core serves as a rigid scaﬀold that

correctly positions the quinoline and phenyl group in the

preferred cis-conﬁguration. The quinoline moiety occupies a

hydrophobic pocket, and the quinoline nitrogen could form a

H-bond with Arg343. These observations provided key insight

into the design of new analogues.

See Supporting Information for a description of assay conditions.

Fraction of the parent compound (0.5 μM) remaining after a 10 min

incubation with 1 mg/mL of human, mouse, and rat liver microsomes

(HLM, MsLM, RLM).

HeLa cellular IC = 2 nM and murine M109 cellular IC = 5

nM), with a human whole blood (hWB) IC₅₀potency ranging

between 2 and 42 nM depending on the donor. It also

enhances the proliferation of T_e_f_fcells with an EC₅₀range of

Herein, we describe our eﬀorts toward identifying a

nonamide-containing IDO1 inhibitor while maintaining

potency and improving pharmacologic properties. Heterocyclic

amide isosteres have been extensively studied in the

−7 nM in a human T cell + DC mixed lymphocyte reaction

(

MLR) assay. Linrodostat was advanced into phase I clinical

studies in 2016. Currently, linrodostat is in several clinical

trials including a phase III study in bladder cancer in

combination with nivolumab.

literature. Previous examples have shown that ﬁve-membered

14b

17a,b,18

heterocycles can be eﬀective amide isosteres.

The

More recently, we disclosed the structure of a closely related

success of a given isostere will depend on whether the amide

is simply a spacer or if it is taking part in key interactions that

are critical to molecular recognition. Given the critical nature

of the H-bonding event between Ser167 and compound 6, it

was important to select and tailor an isosteric replacement that

would maintain this interaction.

clinical candidate, BMS-986242 (5), which entered phase I

14c,d

clinical trials in 2017.

Preclinical biotransformation studies

of BMS-986242 in hepatocytes across several species revealed

the formation of metabolites resulting from quinoline oxidation

and amide bond cleavage. For this reason, we pursued

structurally diﬀerentiated IDO1 inhibitors that no longer

contained an amide (this work) and/or contained a less

metabolically labile replacement for the quinoline. Studies

around the quinoline portion of the molecule will be disclosed

in due course. Notably, other groups have pursued similar

approaches to optimize this chemotype by replacing the amide

All isosteres discussed can be synthesized via the

intermediacy of a carboxylic acid such as previously reported

acid 7 (Scheme 1). To obtain oxazole 8 or imidazole 9,

intermediate 7 is coupled with 2-amino-1-(4-chlorophenyl)-

ethan-1-one. To yield 8, the resulting amide can be heated in

POCl to aﬀect dehydration/cyclization. Alternatively, this

15a,b

in linrodostat with an oxalamide.

amide can be heated with ammonium acetate in the presence

of acetic acid to give imidazole 9. In route to oxadiazole 10 or

triazole 11, intermediate 7 can be treated with t-butyl carbazide

A docking pose of compound 4 (Figure 2, shown in

magenta) in hIDO1 was modeled based on an X-ray cocrystal

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

Letter

Table 2. Eﬀects of R Substitution and Quinoline Modiﬁcations to Amino-oxadiazoles on IDO1 Inhibitory Activity, in Vitro

Metabolic Stability, and PXR Activation

Data reported as average of test results (N = 2 unless otherwise noted). See Supporting Information for a description of assay conditions.

Fraction of the parent compound (0.5 μM) remaining after a 10 min incubation with 1 mg/mL of human, mouse, and rat liver microsomes

(HLM, MsLM, RLM). The pregnane X receptor (PXR) transactivation activity was measured by comparing to activation with rifampicin to assess

the potential for induction of cytochrome P450 (CYP) 3A4 (Et = ethyl, MOM = methoxymethyl, EOM = ethoxymethyl).

followed by treatment with acid to give a hydrazide (not

shown). This hydrazide can be coupled with 4-chlorobenzoic

loss of hIDO1 cellular activity compared to 4. This may be due

to the unfavorable orientation of the amino group NH. The

“ﬂipped” amino-oxadiazole 13 seems to reorient the NH

favorably. Both amino-oxadiazole 13 and benzimidazole 14

demonstrated single-digit nanomolar cellular activity; however,

14 showed inferior metabolic stability.

acid and cyclized with POCl to give 10. Additionally, the

hydrazide can be treated with 4-chlorobenzamidine and heated

to furnish triazole 11. Alternatively, the hydrazide can be

reacted with 4-methoxyphenylisocyanate followed by cycliza-

tion, again with POCl , to provide amino-oxadiazole 12.

Based on the cellular potency of amino-oxadiazole 13 and

benzimidazole 14, both were further pursued. Modular

synthetic approaches were developed to allow for modiﬁcation

Amino-oxadiazole 13 could be made in a three-step sequence

employing a Curtius rearrangement followed by trapping of the

isocyanate intermediate with 4-methoxybenzohydrazide and

of R and quinoline R substitution (see Scheme 2). A vinyl

19−21

ﬁnally heating in POCl to aﬀect dehydration/cyclization to

boronic acid with R groups preinstalled like 15

(Scheme

give 13. Benzimidazole 14 could be synthesized from 7 by ﬁrst

converting the acid to an acyl chloride and treating with 4-

ﬂuoro-1,2-phenylenediamine. Heating the resulting amide with

methanesulfonic acid provided benzimidazole 14.

These various amide isosteres and replacements were

evaluated in HeLa (human) and M109 (murine) cellular

IDO1 inhibition assays where IDO1 activity can be assessed

upon induction with IFNγ (see SI for details). Several ﬁve-

membered heterocycles were found to be eﬀective replace-

ments for the amide in compound 4 (Table 1). Both imidazole

2a) could be joined to quinolinyl halides (16) with Suzuki

coupling. The resulting styrenyl oleﬁn in 17 could be reduced

under either transfer or standard hydrogenation conditions and

hydrolyzed to give 18 as a mixture of cis- and trans-racemates.

Alternatively, the R group could be installed later in the route

by ﬁrst coupling 16 (Scheme 2b) to unsubstituted boronic acid

19 to give 20. The position alpha to the ester could then be

alkylated either before or after reduction of the oleﬁn.

Hydrolysis would again give intermediates of type 18. O-

19,21

Linked quinolines could be made from trans-alcohol 21

and triazole 11 displayed more potent IDO1 inhibitory

employing S Ar on 4-bromoquinoline 22. Hydrolysis provides

activity compared to oxazole 8. One hypothesis that would

account for the greater potency of 9 and 11 is that they

maintain a key H-bond donor NH to interact with Ser167.

This hypothesis is consistent with the modest potency of

oxadiazole 10. It is of note that 10 is more potent than 8

possibly due to either stronger H-bond acceptor ability or

containing two potential H-bond-accepting nitrogens as

opposed to one. Analogue 12, which has an amino linkage

between the oxadiazole and aryl ring, displayed a signiﬁcant

racemic acid 24. Acids of type 18 and 24 can be converted to

amino-oxadiazoles, benzimidazoles, or closely related imidazo-

pyridines as outlined in Scheme 1. All compounds synthesized

by the above general methods and discussed in Tables 2 and 3

have been chirally resolved to obtain enantiopure material. For

all ﬁnal compounds discussed in Tables 2 and 3, only the most

potent enantiomer is shown. To elucidate the relative cis-/trans-

stereochemistry of ﬁnal compounds or synthetic precursors, H

NMR, C NMR, 2D COSY, 2D NOESY, and H− C−Dept-

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

Letter

Table 3. Eﬀects of R Substitution and Quinoline Modiﬁcation of Benzimidazoles and Imidazopyridines on IDO1 Inhibitory

Activity and in Vitro Metabolic Stability

Absolute stereochemistry is R-. Data reported as average test results (N = 3, unless otherwise noted). See Supporting Information for a

description of assay conditions. hWB = human whole blood; data reported as average test results (N = 2, unless otherwise noted). See Supporting

Information for a description of assay conditions. Fractions of the parent compound (0.5 μM) remaining after a 10 min incubation with 1 mg/mL

of human, mouse, and rat liver microsomes (HLM, MsLM, RLM).

Table 4. Compound 36 in Vitro Proﬁling

parameter

compound 36

met. stability CYP (T₁

min)

54 (H), 17 (M), 60 (R), 41 (D), 21 (C)

met. stability UGT (T₁

min)

95 (H), >120 (M), 107 (R), 105 (D), >120

(C)

PAMPA cosolvent (pH 7.4) pH 5.5: 2380 nm/s

pH 7.4: 2630 nm/s

caco (a-b:b-a)

PXR-TA EC₅₀(μM)

113: 66 nm/s

1.19 (32% Y_m_a_x

1A2: 2.55

)

human rCYP panel IC

(μM)

D6: 4.28

C8: 1.47

C9: 0.242

C19: 1.13

A4: 1.93

Table 5. PK−PD Study of Imidazopyridine 36 vs

Linrodostat (4) in a Human SKOV3 Xenograft Tumor

Mouse Model

Figure 3. Putative binding mode of compound 36 in hIDO1, based

on an X-ray cocrystal structure of compound 6 (green).

dose (QDX5)

(mg/kg)

PD AUEC0−24h

treatment

AUC

%[Kyn]↓

0−24h

linrodostat 4

compound 36

34.9

43.6

HSQC were performed (see SI for details). Absolute

stereochemistry was not conﬁrmed unless speciﬁcally noted.

Amino-oxadiazoles related to 13 are discussed in Table 2.

Introducing an ethyl substituent at the position alpha to the

ester (25) maintained potency and proved to be beneﬁcial in

terms of metabolic stability. Analogue 25 also showed potent

hWB activity. Since quinolines are a known potential site of

PK is AUC (0−24 h) in tumor μM*h. PD is percent kynurenine

AUEC (0−24 h) reduction. % Kyn reduction was measured at a

steady state after the 5th dose in the tumor, calculated as the area

under the Kyn concentration−time curve from 0 to 24 h and

compared with that of vehicle control.

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

The Supporting Information is available free of charge at

Letter

metabolic oxidation, we turned our attention to modeling

the phenyl ring of 6, it still is in a position to make a productive

edge-to-face pi-stacking interaction with Tyr126. Imidazopyr-

idine 36 possessed the best overall in vitro proﬁle in terms of

potency and stability for this series and was selected for more

extensive proﬁling.

22b

substituent eﬀects on the oxidation potentials of quinolines.

Reduction potentials of nitrogen-containing compounds have

been reported to correlate with the lowest unoccupied

23,24

molecular orbital (LUMO) energies.

Modeling calcula-

tions suggested that electron-withdrawing group(s) on the

quinoline ring could reduce N-oxidation potential. For

Further in vitro proﬁling (Table 4) showed the compound

had modest oxidative metabolic stability; however, this

metabolic stability proﬁle was superior to both linrodostat

(4) (T₁_/₂human = 53, mouse = 4, rat = 20, dog = 17, cyno =

7) and BMS-986242 (5) (T₁_/₂human = 14, mouse = 4, rat =

10, dog = 10, cyno = 2). Compound 36 had good intrinsic

permeability in a PAMPA assay and Caco-2 cells (a-b = 113

nm/s; eﬄux ratio = 0.6). Unfortunately, compound 36 showed

modest PXR activation (PXR EC50 = 1.2 μM (30% Ymax)) and

CYP inhibition in several human isoforms including potent

inhibition of CYP 2C9.

Compound 36 was further examined in the SKOV3 human

ovarian carcinoma xenograft model (Table 5). In this model,

immune-compromised nu/nu nude mice were implanted with

SKOV3 cells, and the resulting tumors were allowed to grow

for 2 weeks. On day 14, tumor-bearing mice were dosed with

imidazopyridine 36 QD at 20 mg/kg for 5 days. On day 18 at

2, 6, and 24 h time points, tumor kynurenine concentration

(PD) and compound 36 concentration (PK) were measured in

the tumor. The reduction of kynurenine levels, when compared

to a vehicle control, was used as a pharmacodynamic marker.

Imidazopyridine 36 demonstrated a robust proﬁle. At a 20 mg/

kg dose, it achieved 56% reduction in tumor kynurenine levels

and tumor exposure of 43.6 μM*h AUC. This proﬁle compares

well with linrodostat (4): at a 60 mg/kg dose, 4 achieved a

61% reduction in kynurenine levels and a tumor AUC of 34.9

μM*h.

In summary, structurally diﬀerentiated IDO1 inhibitors were

identiﬁed. Heterocyclic amide isosteres and replacements were

investigated. Amino-oxadiazoles, such as compound 25,

demonstrated potent cellular and hWB potency but led to

PXR activation. Optimization of benzimidazole 14 led to the

identiﬁcation of imidazopyridine 36. Lead compound 36

possessed potent cellular and hWB activity as well as improved

^m^e^t^a^b^o^lⁱ^c^s^t^a^bⁱ^lⁱ^t^y⁽^T1/2). Additionally, it had a suitable

permeability proﬁle. Based on these ﬁndings, compound 36

was advanced into an in vivo human SKOV3 xenograft tumor

model in mice. Compound 36 demonstrated a robust PK/PD

proﬁle with improved exposure and comparable PD eﬀects to

linrodostat (4). In contrast to linrodostat (4), which showed

^l^e^s^s^P^X^R^a^c^tⁱ^v^a^tⁱ^oⁿ⁽^P^X^R^E^C50 ^>⁵⁰^μ^M⁽¹³^%^Ymax)) and a

cleaner rCYP panel proﬁle, compound 36 demonstrated more

signiﬁcant PXR activation and CYP inhibition across several

isoforms; therefore, compound 36 was not investigated further.

example, 6-CF quinoline displays higher predicted reduction

22,23

potential E_r_e_d= 2.12 than that of 6-F-quinoline E_r_e_d= 1.93.

Gratifyingly, the 6-CF -substituted quinoline 26 oﬀered potent

cellular activity as well as improved metabolic stability but had

a 3−4-fold loss of hWB potency compared to analogue 25.

Analogue 27, with a 6,8-diﬂuoro-quinoline moiety, unexpect-

edly showed less stability, especially in mouse LMs. Further

proﬁling revealed that all three cis-isomers (25−27) led to PXR

activation. As a consequence of this ﬁnding, eﬀorts were

focused on improving PXR activity. We were pleased to ﬁnd

that O-linked trans-isomer 28 did not activate PXR, but

unfortunately it showed not only a 10-fold disparity between

human and mouse IDO1 activity, which would hinder in vivo

studies, but also a signiﬁcant drop in hWB potency. It is worth

noting that the trans-isomer of C-linked compound 25 (not

shown) also did not activate PXR; however, IDO1 inhibitory

activity was very poor. We then turned our attention to alpha-

substituent modiﬁcation and found that α-MOM-substituted

analogue 29 and α-EOM-substituted analogue 30 both

demonstrated signiﬁcant improvement in PXR activity

compared to compounds 25−27. While 29 and 30 maintained

good hIDO1 activity in cells, they had more modest hWB

potency (IC₅₀= 0.031 μM and 0.096 μM, respectively, vs

.002−0.042 μM for 4) and metabolic stability compared to

lead compound 4 (see Table 1). Since PXR activation could

not be remedied while maintaining suitable hWB and mouse

cellular activity, this series was not progressed.

We then focused on the benzimidazole series. Although

benzimidazole 14 displayed poor metabolic stability, it had

very potent cellular activity (see Table 1). Therefore,

identiﬁcation of a more stable benzimidazole was of primary

interest. Addition of a chloro substituent on the phenyl ring of

the benzimidazole in combination with incorporation of a

nitrogen atom into the ring yielded imidazopyridine 32 (Table

), which exhibited a signiﬁcant improvement in stability

compared to 31 (H/M/R = 67/22/17) while maintaining

hWB activity (IC₅₀= 0.039 μM). As was previously observed

in the amino-oxadiazole series, introduction of ethyl

substitution at the alpha-position generally increased potency

and improved metabolic stability. Analogue 33 displayed

potent hWB activity and improved stability, while imidazopyr-

idine 34 revealed unexpectedly poor mouse metabolic stability

(

H/M/R = 78/7/35). Consistent with trends observed in the

ASSOCIATED CONTENT

sı Supporting Information

https://pubs.acs.org/doi/10.1021/acsmedchemlett.1c00014.

amino-oxadiazole series (see 28, Table 2), incorporating an O-

linked trans-isomer (35) was found to increase metabolic

stability. Once again, however, mouse cellular activity suﬀered.

Combining the more potent alpha-ethyl group with the more

■

stable imidazopyridine and the 6-CF -quinoline (see 26, Table

Biological assay protocols, in vivo pharmacokinetic−

pharmacodynamic study protocols, experimental proce-

dures, and analytical data for all ﬁnal compounds (PDF)

, vide supra) led to compound 36. Docking models of 36,

based on the X-ray cocrystal of compound 6 and hIDO1 in

Figure 2, indicated that the imidazopyridine likely maintains

the same key interactions observed for the amide (Figure 3).

The NH of the imidazopyridine nicely overlays with the NH of

the amide bond in 6 which would correctly position it to make

the key hydrogen bond with Ser167. While the pyridine

portion of the imidazopyridine is positioned a little lower than

AUTHOR INFORMATION

Corresponding Authors

■

Liping Zhang − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

Complete contact information is available at:

Letter

States; orcid.org/0000-0002-1797-7227; Phone: 609-

https://pubs.acs.org/10.1021/acsmedchemlett.1c00014

52-3087; Email: liping.zhang@bms.com

Emily C. Cherney − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States ; Phone: 609-252-6064; Email: emily.cherney@

bms.com

Aaron Balog − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States ; Phone: 609-252-4632; Email: aaron.balog@

bms.com

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

We thank Jay Markwalder, Weifang Shan, and the Richard

Rampulla/BBRC group for providing intermediates. We thank

■

Dr. Shana Posy for assistance with Figure .

Authors

ABBREVIATIONS

■

Xiao Zhu − Bristol Myers Squibb Research and Development,

Princeton, New Jersey 08543-4000, United States;

orcid.org/0000-0001-7647-4390

Tai-an Lin − Bristol Myers Squibb Research and Development,

Princeton, New Jersey 08543-4000, United States

Johnni Gullo-Brown − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

Derrick Maley − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

Kathy Johnston-Allegretto − Bristol Myers Squibb Research

and Development, Princeton, New Jersey 08543-4000, United

States

Lisa Kopcho − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

Mark Fereshteh − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

Christine Huang − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

Xin Li − Bristol Myers Squibb Research and Development,

Princeton, New Jersey 08543-4000, United States

Sarah C. Traeger − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

HATU, 1-[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo-

[4,5-b]pyridinium 3-oxide hexaﬂuorophosphate; NMM, N-

methylmorpholine; DIPEA, N,N-diisopropylethylamine; DMF,

N,N-dimethylformamide; TEA, triethylamine; DPPA, diphe-

nylphosphorylazide; MsOH, methanesulfonic acid; H, human;

M, mouse; R, rat; D, dog; C, cyno

REFERENCES

■

1) Adams, J. L.; Smothers, J.; Srinivasan, R.; Hoos, A. Big

opportunities for small molecules in immuno-oncology. Nat. Rev.

Drug Discovery 2015, 14, 603−622.

(2) Austin, C. J. D.; Rendina, L. M. Targeting key dioxygenases in

tryptophan-kynurenine metabolism for immunomodulation and

cancer chemotherapy. Drug Discovery Today 2015 , 20 , 609 −617.

3) Munn, D. H.; Mellor, A. L. Indoleamine 2,3-dioxygenase and

tumor-induced tolerance. J. Clin. Invest. 2007 , 117 , 1147 −1154.

4) Munn, D. H.; Mellor, A. L. Indoleamine 2,3-dioxygenase and

metabolic control of immune responses. Trends Immunol. 2013, 34,

37−143.

5) Munn, D. H.; Mellor, A. L. IDO in the tumor microenviron-

(6) Prendergast, G. C.; Malachowski, W. P.; DuHadaway, J. B.;

ment: inflammation, counter-regulation, and tolerance. Trends

Immunol. 2016, 37, 193−207.

Muller, A. J. Discovery of IDO1 inhibitors: from bench to bedside.

Cancer Res. 2017, 77, 6795−6811.

(

7) Yue, E. W.; Sparks, R.; Polam, P.; Modi, D.; Douty, B.; Wayland,

B.; Glass, B.; Takvorian, A.; Glenn, J.; Zhu, W.; Bower, M.; Liu, X.;

Leffet, L.; Wang, Q.; Bowman, K. J.; Hansbury, M. J.; Wei, M.; Li, Y.;

Wynn, R.; Burn, T. C.; Koblish, H. K.; Fridman, J. S.; Emm, T.;

Scherle, P. A.; Metcalf, B.; Combs, A. P. INCB24360 (epacadostat), a

highly potent and selective indoleamine-2,3-dioxygenase 1 (IDO1)

inhibitor for immunooncology. ACS Med. Chem. Lett. 2017, 8, 486−

491.

(8) Combs, A. P.; Yue, E. W. Preparation of 4-amino-N’-hydroxy-

,2,5-oxadiazole-3-carboximidamides and related compounds as

modulators of indoleamine 2,3-dioxygenase for inhibiting immuno-

suppression and treating various disorders. WO 2006122150 A1,

November 16, 2006.

Gopal Dhar − Biocon Bristol Myers Squibb R&D Center,

Bengaluru, Karnataka 560099, India

Aravind Anandam − Biocon Bristol Myers Squibb R&D

Center, Bengaluru, Karnataka 560099, India

Sandeep Mahankali − Biocon Bristol Myers Squibb R&D

Center, Bengaluru, Karnataka 560099, India

Shweta Padmanabhan − Biocon Bristol Myers Squibb R&D

Center, Bengaluru, Karnataka 560099, India

Prabhakar Rajanna − Biocon Bristol Myers Squibb R&D

Center, Bengaluru, Karnataka 560099, India

Venkata Murali − Biocon Bristol Myers Squibb R&D Center,

Bengaluru, Karnataka 560099, India

Thanga Mariappan − Biocon Bristol Myers Squibb R&D

Center, Bengaluru, Karnataka 560099, India

Robert Borzilleri − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

Gregory Vite − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

John T. Hunt − Bristol Myers Squibb Research and

Development, Princeton, New Jersey 08543-4000, United

States

(

9) Liu, X.; Shin, N.; Koblish, H. K.; Yang, G.; Wang, Q.; Wang, K.;

Leffet, L.; Hansbury, M. J.; Thomas, B.; Rupar, M.; Waeltz, P.;

Bowman, K. J.; Polam, P.; Sparks, R. R.; Yue, E. W.; Li, Y.; Wynn, R.;

Fridman, J. S.; Burn, T. C.; Combs, A. P.; Newton, R. C.; Scherle, P.

A. Selective inhibition of IDO1 effectively regulates mediators of

antitumor immunity. Blood 2010, 115, 3520−3530.

(10) Koblish, H. K.; Hansbury, M. J.; Bowman, K. J.; Yang, G.;

Neilan, C. L.; Haley, P. J.; Burn, T. C.; Waeltz, P.; Sparks, R. B.; Yue,

E. W.; Combs, A. P.; Scherle, P. A.; Vaddi, K.; Fridman, J. S.

Hydroxyamidine inhibitors of indoleamine-2,3-dioxygenase potently

suppress systemic tryptophan catabolism and the growth of IDO-

expressing tumors. Mol. Cancer Ther. 2010, 9, 489−498.

(

11) (a) Hamid, O.; Gajewski, T. F.; Frankel, A. E.; Bauer, T. M.;

Olszanski, A. J.; Luke, J. J.; Balmanoukian, A. S.; Schmidt, E. V.;

Sharkey, B.; Maleski, J.; Jones, M. J.; Gangadhar, T. C. Epacadostat

plus pembrolizumab in patients with advanced melanoma: phase 1

and 2 efficacy and safety results from ECHO-202/KEYNOTE-037.

ACS Med. Chem. Lett. 2021, 12, 494−501

ACS Medicinal Chemistry Letters

Rittmann, B. E. Rittmann. Accelerating quinoline biodegradation and

Letter

Ann.Oncol 2017, 28, v428−v448. (b) NCT 02752074, ECHO-301/

KEYNOTE-252. (c) Mullard, A. IDO takes a blow. Nat. Rev. Drug

Discovery 2018, 17, 307.

D. K.; Guo, W. Immunoregulatory agents. WO 2016073738, A2, May

12, 2016.

(21) Beck, H. P.; Jaen, J. C.; Osipov, M.; Powers, J. P.; Reilly, M. K.;

Shunatona, H. P.; Walker, J. R.; Zibinsky, M.; Balog, J. A.; Williams,

D. K.; Markwalder, J. A.; Cherney, E. C.; Shan, W.; Huang, A..

Immunoregulatory agents. WO 2016073770, A1, May 12, 2016.

(22) (a) Bai, Q.; Yang, L.; Li, R.; Chen, B.; Zhang, L.; Zhang, Y.;

oxidation with endogenous electron donors. Environ. Sci. Technol.

(12) Burris, H. A.; Gordon, M. S.; Hellmann, M. D.; LoRusso, P.;

Emens, L. A.; Hodi, F. S.; Lieu, C. H.; Infante, J. R.; Tsai, F. Y.-C.;

Eder, J. P.; Cleary, J. M.; Jelovac, D.; Tsuhako, A. L.; Mueller, L.; Lin,

R.; Morrissey, K.; Mahrus, S.; Morley, R.; Pirzkall, A.; Davis, S. L. A

phase Ib dose escalation study of combined inhibition of IDO1

GDC-0919) and PD-L1 (atezolizumab) in patients with locally

advanced or metastatic solid tumors. J. Clin. Oncol. 2017, 35, 105−

05A.

13) ClinicalTrials.gov Identiﬁer: NCT 02764151: First in patient

015, 49, 11536−11542. (b) Modeling methods for calculating N-

oxidation potential: Molecular structures of each model compound

are constructed using the Maestro v2015-4 software (Schrodinger,

(

015). QM calculations are performed using the Jaguar software

study for PF-06840003 in Malignant gliomas: http://www.

iteostherapeutics.com/news/20180104-iteos-therapeutics-regains-

worldwide-rights-to-clinical-stage-IDO1-inhibitor.

(Bochevarov, 2013). Each model compound is subjected to geometry

optimization at the B3LYP/6-31G** level of theory and single-point

energy calculation at the B3LYP/cc-pVTZ(-f) level of theory, from

which the LUMO energy term is extracted. LUMO energy is

converted to unscaled reduction potential (eV) via E_r_e_d= −27.2107 *

LUMO (hartree) (ref 23) such that higher E_r_e_dsignifies a lower

propensity for oxidation.

(14) (a) Balog, A.; Lin, T.-A.; Maley, E.; Gullo-Brown, J.; Hamza

Kandoussi, E.; Zeng, J.; Hunt, J. T. Preclinical characterization of

Linrodostat mesylate, a novel, potent, and selective oral indoleamine

,3-dioxygenase 1 inhibitor. Mol. Cancer Ther. 2020, mol-

canther.0251.2020. (b) A Study of chemo only versus chemo plus

nivo with or without BMS-986205, followed by post-surgery therapy

with nivo or nivo and BMS-986205 in patients with MIBC.

ClinicalTrials.gov. Identifier: NCT03661320. (c) Cherney, E. C.;

Zhang, L.; Nara, S.; Zhu, X.; Gullo-Brown, J.; Maley, D.; Lin, T.;

(23) Assary, R. S.; Brushett, F. R.; Curtiss, L. A. Reduction potential

predictions of some aromatic nitrogen-containing molecules. RCS

Advances 2014, 4, 57442−57451.

(24) Bochevarov, A. D.; Harder, E.; Hughes, T. F.; Greenwood, J.

R.; Braden, D. A.; Philipp, D. M.; Rinaldo, D.; Halls, M. D.; Zhang, J.;

Friesner, R. A. Jaguar A high-performance quantum chemistry

software program with strengths in life and materials sciences. Int. J.

Quantum Chem. 2013, 113, 2110−2142.

Hunt, J.; Huang, C.; Yang, Z.; DArienzo, C.; Discenza, L.;

Ranasinghe, A.; Grubb, M.; Ziemba, T.; Traeger, S.; Li, X.;

Johnston, K.; Kopcho, L.; Fereshteh, M.; Foster, K. A.; Stefanski,

K.; Fargnoli, J.; Swanson, J.; Brown, J.; Delpy, D.; Seitz, S.; Borzilleri,

R.; Vite, G.; Balog, A. Discovery and preclinical evaluation of BMS-

86242, a potent, selective inhibitor of indoleamine-2,3-dioxygenase

. ACS Med. Chem. Lett. 2021, 12, 288. (d) An investigational

immunotherapy study of experimental medication BMS-986242 given

in combination with nivolumab in patients with advanced cancer.

ClinicalTrials.gov. Identifier: NCT03351231. (e) Nelp, M. T.; Kates,

P. A.; Hunt, J. T.; Newitt, J. A.; Balog, A.; Maley, D.; Zhu, X.; Abell,

L.; Allentoff, A.; Borzilleri, R.; Lewis, H. A.; Lin, Z.; Seitz, S. P.; Yan,

C.; Groves, J. T. Immune-modulating enzyme indoleamine 2,3-

dioxygenase is effectively inhibited by targeting its apo-form. Proc.

Natl. Acad. Sci. U. S. A. 2018, 115, 3249−3254.

(15) (a) Steeneck, C.; Kinzel, O.; Anderhub, S.; Hornberger, M.;

Pinto, S.; Morschhaeuser, B.; Albers, M.; Sonnek, C.; Czekanska, M.;

Hoffmann, T. Discovery and optimization of substituted oxalamides

as novel heme-displacing IDO1 inhibitors. Bioorg. Med. Chem. Lett.

021, 33, 127744. (b) Kinzel, O.; Steeneck, C.; Anderhub, S.;

Hornberger, M.; Pinto, S.; Morschhaeuser, B.; Albers, M.; Sonnek, C.;

Wang, Y.; Mallinger, A.; Czekanska, M.; Hoffmann, T. Discovery of

highly potent heme-displacing IDO1 inhibitors based on a spirofused

bicyclic scaffold. Bioorg. Med. Chem. Lett. 2021 , 33 , 127738.

16) Meanwell, N. A. Synopsis of some recent tactical application of

bioisosteres in drug design. J. Med. Chem. 2011, 54, 2529−2591.

17) (a) Kumari, S.; Carmona, A. V.; Tiwari, A. K.; Trippier, P. C.

Singh, G.; Sharova, N.; Lichinchi, G.; Dang, J.; Stevenson, M.; Rana,

Amide bond bioisosteres: strategies, synthesis, and successes. J. Med.

Chem. 2020, 63, 12290−12358. (b) Mohammed, I.; Kummetha, I. R.;

T. M. 1,2,3-triazoles as amide bioisosteres: discovery of new class of

potent HIV-1 vif antagonists. J. Med. Chem. 2016, 59, 7677−7682.

(18) Di Francesco, M. E.; Dessole, G.; Nizi, E.; Pace, P.; Koch, U.;

Fiore, F.; Pesci, S.; Di Muzio, J.; Monteagudo, E.; Rowley, M.;

Summa, V. Novel macrocyclic inhibitors of hepatitis C NS3/4A

protease featuring a 2-amino-1,3-thiazole as a P4 carbamate

replacement. J. Med. Chem. 2009, 52, 7014−7028.

(19) Beck, H. P.; Jaen, J. C.; Osipov, M.; Powers, J. P.; Reilly, M. K.;

Shunatona, H. P.; Walker, J. R.; Zibinsky, M.; Balog, J. A.; Williams,

D. K.; Markwalder, J. A.; Seitz, S. P.; Cherney, E. C.; Zhang, L.; Shan,

W.; Guo, W.; Huang, A. Preparation of immunoregulatory agents.

WO 2016073774, May 12, 2016.

(20) Beck, H. P.; Jaen, J. C.; Osipov, M.; Powers, J. P.; Reilly, M. K.;

Shunatona, H. P.; Walker, J. R.; Zibinsky, M.; Balog, J. A.; Williams,