The discovery of IDX21437: Design, synthesis and antiviral evaluation of 2′-α-chloro-2′-β-C-methyl branched uridine pronucleotides as potent liver-targeted HCV polymerase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2017.08.029

Herein we describe the discovery of IDX21437 35b, a novel R_P D-aminoacid-based phosphoramidate prodrug of 2′-α-chloro-2′-β-C-methyluridine monophosphate. Its corresponding triphosphate 6 is a potent inhibitor of the HCV NS5B RNA-dependent RNA polymerase (RdRp). Despite showing very weak activity in the in vitro Huh-7 cell based HCV replicon assay, 35b demonstrated high levels of active triphosphate 6 in mouse liver and human hepatocytes. A biochemical study revealed that the metabolism of 35b was mainly attributed to carboxyesterase 1 (CES1), an enzyme which is underexpressed in HCV Huh-7-derived replicon cells. Furthermore, due to its metabolic activation, 35b was efficiently processed in liver cells compared to other cell types, including human cardiomyocytes. The selected R_P diastereoisomeric configuration of 35b was assigned by X-ray structural determination. 35b is currently in Phase II clinical trials for the treatment of HCV infection.

Full text of DOI:10.1016/j.bmcl.2017.08.029

Accepted Manuscript
The discovery of IDX21437: design, synthesis and antiviral evaluation of 2’-α-
chloro-2’-β-C-methyl branched uridine pronucleotides as potent liver-targeted
HCV polymerase inhibitors
François-René Alexandre, Eric Badaroux, John P. Bilello, Stéphanie Bot, Tony
Bouisset, Guillaume Brandt, Sylvie Cappelle, Christopher Chapron, Dominique
Chaves, Thierry Convard, Clément Counor, Daniel Da Costa, David Dukhan,
Marion Gay, Gilles Gosselin, Jean-François Griffon, Kusum Gupta, Brenda
Hernandez-Santiago, Massimiliano LaColla, Marie-Pierre Lioure, Julien
Milhau, Jean-Laurent Paparin, Jérôme Peyronnet, Christophe Parsy, Claire
Pierra Rouvière, Houcine Rahali, Rachid Rahali, Aurélien Salanson, Maria
Seifer, Ilaria Serra, David Standring, Dominique Surleraux, Cyril B. Dousson
PII:
S0960-894X(17)30831-4
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.08.029
BMCL 25223
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
20 June 2017
6 August 2017
14 August 2017
Please cite this article as: Alexandre, F-R., Badaroux, E., Bilello, J.P., Bot, S., Bouisset, T., Brandt, G., Cappelle,
S., Chapron, C., Chaves, D., Convard, T., Counor, C., Da Costa, D., Dukhan, D., Gay, M., Gosselin, G., Griffon,
J-F., Gupta, K., Hernandez-Santiago, B., LaColla, M., Lioure, M-P., Milhau, J., Paparin, J-L., Peyronnet, J., Parsy,
C., Pierra Rouvière, C., Rahali, H., Rahali, R., Salanson, A., Seifer, M., Serra, I., Standring, D., Surleraux, D.,
Dousson, C.B., The discovery of IDX21437: design, synthesis and antiviral evaluation of 2’-α-chloro-2’-β-C-methyl
branched uridine pronucleotides as potent liver-targeted HCV polymerase inhibitors, Bioorganic & Medicinal
Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.08.029
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
Leave this area blank for abstract info.
The discovery of IDX21437: Design,
synthesis and antiviral evaluation of 2’-α-
chloro-2’-β-C-methyl
branched
uridine
pronucleotides as potent liver-targeted HCV
polymerase inhibitors
François-René Alexandrea,*, Eric Badarouxa,†, John P. Bilellob,c, Stéphanie Bota, Tony Bouisseta,†, Guillaume
Brandta, Sylvie Cappellea, Christopher Chapronc,†, Dominique Chavesa,†, Thierry Convarda, Clément Counora,†,
Daniel Da Costaa, David Dukhana, Marion Gaya,†, Gilles Gosselind, Jean-François Griffona, Kusum Guptac,†,
Brenda Hernandez-Santiagoc,†, Massimiliano LaCollac,†, Marie-Pierre Liourea,†, Julien Milhaua, Jean-Laurent
Paparina, Jérôme Peyronneta,†, Christophe Parsya,†, Claire Pierra Rouvièrea, Houcine Rahalia, Rachid Rahalia,
Aurélien Salansona, Maria Seiferc,†, Ilaria Serrac,†, David Standringc,†, Dominique Surlerauxa,† and Cyril B.
Doussona
aIDENIX an MSD Company, Cap Gamma, 1682 rue de la Valsière, 34189 Montpellier Cedex 4, France
bMerck & Co., Inc., PO Box 4, 770 Sumneytown Pike, West Point, PA 19486, USA
cFormerly at IDENIX Pharmaceuticals, 320 Bent Street - 4th Floor, Cambridge, MA 02139, USA
dFormer employee of the Institut des Biomolécules Max Mousseron (IBMM), UMR 5247 CNRS-Université Montpellier-ENSCM, case courrier 1705,
Place Eugène Bataillon, 34095 Montpellier Cedex 5, France
†Former Idenix employees
450000
IDX21437-TP 6
387000
400000
Sofosbuvir-TP
366956
350000
300000
250000
200000
150000
100000
50000
0
288572
O
O
O
O
O
P
O
P
O
P
O
O
P
N
O
in the liver cells
N
NH
HO
O
O
O
NH
HN
O
OH OH OH
O
O
HO
IC50=0.154 µM
O
Cl
HO
Cl
6
35b
(IDX21437)
BLQ
iCell cardiomyocytes
iCell hepatocytes
Cells were exposed to 100 µM of IDX21437 and 10 µM of sofosbuvir

Bioorganic
journal homepage: www.els evier.com
&
Medicinal
Chemistry
Letters
The discovery of IDX21437: design, synthesis and antiviral evaluation of 2’-α-
chloro-2’-β-C-methyl branched uridine pronucleotides as potent liver-targeted
HCV polymerase inhibitors
François-René Alexandrea,∗, Eric Badarouxa,†, John P. Bilellob,c, Stéphanie Bota, Tony Bouisseta,†,
Guillaume Brandta, Sylvie Cappellea, Christopher Chapronc,†, Dominique Chavesa,†, Thierry Convarda,
Clément Counora,†, Daniel Da Costaa, David Dukhana, Marion Gaya,†, Gilles Gosselind, Jean-François
Griffona, Kusum Guptac,†, Brenda Hernandez-Santiagoc,†, Massimiliano LaCollac,†, Marie-Pierre Liourea,†,
Julien Milhaua, Jean-Laurent Paparina, Jérôme Peyronneta,†, Christophe Parsya,†, Claire Pierra Rouvièrea,
Houcine Rahalia, Rachid Rahalia, Aurélien Salansona, Maria Seiferc,†, Ilaria Serrac,†, David Standringc,†,
Dominique Surlerauxa,† and Cyril B. Doussona
aIDENIX an MSD Company, Cap Gamma, 1682 rue de la Valsière, 34189 Montpellier Cedex 4, France
bMerck & Co., Inc., PO Box 4, 770 Sumneytown Pike, West Point, PA 19486, USA
cFormerly at IDENIX Pharmaceuticals, 320 Bent Street - 4th Floor, Cambridge, MA 02139, USA
dFormer CNRS research director of the Institut des Biomolécules Max Mousseron (IBMM), UMR 5247 CNRS-Université Montpellier-ENSCM, case courrier
1705, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France
†Former Idenix employees
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Herein we describe the discovery of IDX21437 35b, a novel RP D-aminoacid-based
phosphoramidate prodrug of 2’-α-chloro-2’-β-C-methyluridine monophosphate. Its
corresponding triphosphate 6 is a potent inhibitor of the HCV NS5B RNA-dependent RNA
polymerase (RdRp). Despite showing very weak activity in the in vitro Huh-7 cell based HCV
replicon assay, 35b demonstrated high levels of active triphosphate 6 in mouse liver and human
hepatocytes. A biochemical study revealed that the metabolism of 35b was mainly attributed to
carboxyesterase 1 (CES1), an enzyme which is underexpressed in HCV Huh-7-derived replicon
cells. Furthermore, due to its metabolic activation, 35b was efficiently processed in liver cells
compared to other cell types, including human cardiomyocytes. The selected RP
diastereoisomeric configuration of 35b was assigned by X-ray structural determination. 35b is
currently in Phase II clinical trials for the treatment of HCV infection.
Keywords:
Hepatitis C
HCV NS5B polymerase inhibitors
Pronucleotide
Nucleoside
Synthesis and biological evaluation
Liver delivery
2017 Elsevier Ltd. All rights reserved.
———
∗ Corresponding author. Tel.: +33(0)499522252; fax: +33-(0)499522250; francois-rene.alexandre@merck.com

Hepatitis C Virus (HCV) is a single-stranded positive sense
RNA virus belonging to the Flaviviridae family. This virus is
the causative agent of chronic liver diseases affecting around
3% of the world’s population.1 Initially asymptomatic, this
infection can ultimately lead to liver cirrhosis, fibrosis and
hepatocellular carcinoma. In developed countries, HCV is the
primary cause for liver transplants. Introduced in 2001, the
In order to improve the overall therapeutic index and
potency of the parent nucleoside, a prodrug approach was
investigated and applied to pyrimidine and purine 2’-C-
methyl-β-D-ribonucleoside. The aim of this approach was to
target the liver and to bypass the initial monophosphorylation
step, which is known to be rate-limiting in numerous cases.8
From our preclinical studies and based initially on replicon
initial standard of care therapy (SOC) combined
a
activity and in vivo data, we selected IDX184
a
subcutaneous injection of a modified interferon-α (IFN-α)
with an oral administration of ribavirin. However this
treatment was inconvenient, very poorly tolerated and induced
a sustained viral response (SVR) in only 42-82% of patients
depending of the virus genotype.2 In this context development
of orally administered, effective and well-tolerated therapies
became a major stake for industrial researchers. In recent
years, progress in the understanding of the HCV morphology,
virus life cycle, as well as the development of HCV replicon
(RPL) cell-based assays has resulted in a large number of
research groups taking on the rational design of so-called
direct acting antivirals (DAA) as specific inhibitors of the virus
life cycle.3 In 2011, two new DAAs, both HCV NS3/4A
protease inhibitors, boceprevir and telaprevir, became available
to reinforce the existing antiviral arsenal. Albeit bringing clear
improvement over the previous therapies, they possess several
drawbacks that included low efficacy rates in some patient
populations, high pill burdens and serious side effects.3,4
However, the major advance toward an oral, interferon-free
regimen, came in 2014 with the approved combination of
sofosbuvir (NS5B polymerase nucleoside inhibitor (NI)) and
ledipasvir (NS5A replication inhibitor). This combination
demonstrated an improved cure rate up to 99% in certain
patient populations, but remains suboptimal in others.5 To this
end, one critical focus for future HCV therapy is the
identification of a pan-genotypic DAA combination therapy
that could be effective across all HCV patient populations.
phosphoramidate pronucleotide of 2’-C-methyl-guanosine as a
clinical candidate (Chart 1), possessing S-acyl-2-thioethyl
(SATE) and benzylamine moieties.9 This compound was the
first pronucleotide evaluated in the clinic through Phase II
trials to treat HCV. In combination with IFN-α and ribavirin in
genotype 1 treatment-naïve patient in Phase IIa clinical trials, a
4-log HCV RNA reduction by day 14 was observed at a daily
dose of 100 mg.10 Despite those encouraging results, and
although no evidence of cardiac effects in all IDX184 clinical
studies was found, IDX184 was placed on clinical hold by the
FDA due to the cardiac adverse events observed with BMS-
986094 (INX-189), which shares the same 2’-MeGTP
anabolite as IDX184. As a consequence, the program was re-
oriented to other pronucleotides bearing nucleosides other than
2’Me-G.
Replication of HCV occurs predominantly in hepatocytes,
where the genome is translated into a polyprotein that is
subsequently cleaved to produce structural and non-structural
proteins constitutive of the virus. Among the most studied viral
targets, including NS3/4A protease, the NS5A protein and
RdRp NS5B polymerase, the latter is among the most
attractive within the HCV life cycle due to its high pan-
genotypic conservation. Nucleoside inhibitors (NIs) which
have long been demonstrated to inhibit the NS5B polymerase
possess a high barrier to drug resistance and a high genotypic
coverage compared to other DAAs, indicating that they are
likely to remain key components in future treatment regimens.
Chart 1. Sofosbuvir, INX-189 (stopped for cardiac toxicity) and Idenix
pioneered anti HCV clinical molecules: nucleoside NM283 and
pronucleotide IDX184.
Uridine derivatives are generally known to be poorly
recognized by the deoxycytidine kinase (dCK), the enzyme
responsible for the first phosphorylation step en route to
NTP.11 This lack of recognition when associated with a good
activity against the HCV polymerase of the corresponding
NTP, made such nucleosides ideal candidates for
a
monophosphate prodrug approach. We first re-evaluated the
2’-C-methyluridine derivative 1 (NM106), and its triphosphate
2 in the cell-based HCV replicon assay and in the NS5B
polymerase biochemical assay respectively (Table 1).12 While
1 displayed poor activity of 30 µM in the replicon assay (Table
1, Entry 1), the corresponding triphosphate 2 was very potent
at 0.081 µM against the wild type (WT) NS5B polymerase
(Table 1, Entry 4), validating our interest in a prodrug
approach. However, against the HCV S282T mutant
polymerase, the mutant selected by the 2’-methyl nucleosides
class, 2’Me-UTP 2 showed a 102-fold loss in activity
compared to the WT (Table 1, Entry 4).
Our laboratory is engaged in the research of
chemotherapeutics against HCV, with the discovery of the
antiviral properties of the 2’-C-methyl-β-D-ribonucleoside
analogues over 15 years ago. This effort led to the discovery of
NM283, the first nucleoside to enter Phase II clinical trials for
the treatment of chronic hepatitis C (Chart 1).6 NM283,
similarly to other NIs, is metabolized in cells in a three-step
process to its nucleoside monophosphate, diphosphate and
triphosphate consecutively. This nucleoside triphosphate
(NTP) is the active anabolite that binds to the active site of the
NS5B polymerase as a substrate, is incorporated to the
growing chain of viral RNA and acts as a nonobligate chain
terminator by blocking the further production of the viral
RNA. Despite the efficacy shown in clinical trials,
development of NM283 was halted due to gastrointestinal (GI)
adverse events.7
With the aim of improving the mutation profile of such
inhibitors, we investigated the 2’-α-chloro-2’-β-C-methyl
substitution.
The
synthesis
of
2’-α-chloro-2’-β-C-
methylcytidine and 2’-α-chloro-2’-β-C-methyluridine is
depicted in Scheme 1.13

Scheme 1. Reagents and conditions: (a) LDA, Toluene, −78 °C, 5 min; (b) AcOH, 90°C, 17% (2 steps); (c) (1) BzCl, DMAP, CH3CN, (2) Et3N, 84% ; (d) Red-
Al, THF, −35 °C to 0 °C, (α/β 45/55) 96% ; (e) CBr4, PPh3, −20 °C, 66%; (f) N-4-benzoylcytosine, HMDS, SnCl4, 4-chlorobenzene, 70 °C, 89%; (g) 7 N NH3 in
MeOH, RT, 79%; (h) AcOH, H2O, 110 °C, quant. ; (i) benzyloxymethylchloride (BOMCl), DBU, DMF, −5 °C to 5 °C, 80%.
The key protected ribonolactone 11 was synthesized in
three steps from the R-glyceraldehyde acetonide 7. The first
step was an aldol condensation between 7 and 8 to provide 9 as
a diastereomeric mixture. 9 was engaged crude in a cyclization
reaction to give 10a and 10b in a 1:9 ratio, the latter being the
major product, which was isolated in 17% yield by
Table 1. Activities of Nucleosides 1, 3, 5 in the HCV
replicon gen1b assay and of their corresponding nucleoside
5’-triphosphates 2, 4, 6 against the HCV con 1b NS5B
polymerase
a
b
c
d
Entry Cpd EC50
CC50
Entry Cpd
IC50
IC50
FCe
recrystallization from
a
toluene-butanone mixture. The
(µM)
(µM)
WT
(µM)
0.081
S282T
(µM)
8.24
hydroxyl groups of 10b were then protected as benzoates to
obtain the ribonolactone 11 in 84% yield. The lactone was
reduced with Red-Al in THF to afford lactol 12 as a
diastereomeric mixture at the 1’-position, and the resulting
hydroxyl group of 12 was substituted to a bromo group with
carbontetrabromide and triphenylphosphine to afford 13 as
44/56 mixture of α/β anomers. These anomers were readily
separable by chromatography on silica gel, however it was
unnecessary to proceed with this separation as the anomeric
mixture reacted with sylilated N-4-benzoylcytosine and tin
1
2
1
3
5
30
50
57
>100
>100
>100
4
5
6
2
4
6
102
7.7
6.1
0.079
0.154
0.606
0.944
3
a
Inhibition of HCV replication Huh-7 cells containing HCV Con1
subgenomic replicon (GS4.1 cells) with a luciferase read-out. EC50 values
were determined from the 50% inhibition versus concentration data.
CC50’s were calculated as the concentration that caused a 50% of cells
death versus concentration data.
capacity on the 1b HCV NS5B polymerase; IC50 values were determined
from the percent inhibition versus concentration data. Biochemical assay
for inhibition capacity on the S282T HCV NS5B polymerase; IC50 values
b
c
Biochemical assay for inhibition
d
tetrachloride in 4-chlorobenzene affording only the
nucleoside 14 in 89% yield. Subsequent
hydrolysis/deprotection of 14 led to uridine derivative 5 in
β
e
were determined from the percent inhibition versus concentration data.
FC= fold change = IC50 S282T/IC50 WT.
good yields.
Table 2. In vitro inhibition of wild type HCV and cellular
polymerases by nucleoside 5’triphosphate 6.13
IC (µM)a
As expected and similarly to 2’-α-fluoro-2’-β-C-
methyluridine 3, the uridine derivative 5 showed very modest
activity in the RPL cell-based assay (Table 1, Entry 3). The
corresponding triphosphates 6 was synthesized according
known procedures and its activity is shown in Table 1.14
Triphosphate 6 was found to be a potent inhibitor of the HCV
NS5B polymerase. Against the S282T mutant polymerase, 6
showed a 6-fold decrease in activity compared to the wild type,
while a 102-fold decrease was measured for NM106-TP 2. 6
showed similar activity across all genotypes against HCV
polymerases (Table 2, Entries 1−6) and proved to be selective
for HCV polymerases vs human polymerases (Table 2, Entries
7−11). In addition, the half-life of compound 6 was determined
to be over 18h in human hepatocytes which supported the
potential for once daily (QD) dosing. These promising results
prompted us to synthesize and evaluate a series of prodrugs
based on nucleoside 5.
Entry
Species
Enzyme
50
1
2
3
4
5
Genotype 1a
Genotype 2a
Genotype 3a
Genotype 4a
Genotype 5a
Genotype 6a
0.14
0.25
0.16
0.15
0.23
0.10
HCV polymerases
6
7
8
9
10
DNA pol alpha
DNA pol beta
DNA pol gamma
RNA pol II
>500
>500
95
>500
>500
Human polymerases
11
Mitochondrial RNA pol
a
Biochemical assay for inhibition capacity on various polymerases; IC50
values were determined from the percent inhibition versus concentration
data.
Based on our previous experience in the IDX184 program,
we selected several classes of phosphorylated moieties from
our prodrug collection for evaluation.9 We initially synthesized
SATE derivative phosphoramidate prodrugs 20 and 21
(Scheme 2). BOM protected 15, obtained from 5 in 80% yield
(Scheme 1), reacted with 16 and 17 (synthesized according
known methods)9,15 to afford intermediate phosphonic acid
derivatives, which underwent Atherton−Todd oxidative
amination with benzylamine in carbon tetrachloride to give 18
and 19. Trityl (for 18) and BOM protecting groups were
removed by reaction with BBr3 at low temperature followed by
treatment with formic acid in acetonitrile and water to remove
Chart 2. Evaluated nucleosides and corresponding NTPs

residual
hydroxymethyl
resulting
from
incomplete
determined to be the RP isomer and 45b as the SP isomer. The
phosphorus stereochemistry of the other prodrugs was assigned
by analogy to the 31P NMR chemical shifts with those of 35b,
45a, 45b and similar structures reported in the literature.17,18
deprotection. This sequence furnished 20 and 21 as mixtures of
diastereomers, which were separated into their SP and RP
isomers by preparative HPLC. Cyclic phosphate triesters
(CPO) 26 and 27, derived from SATE prodrugs, were obtained
from the reaction of protected uridine 15 and
phosphorodiamidite 22 and 23, followed by oxidation and
deprotection (Scheme 2). As previously, the diastereomers
were separated by preparative HPLC.
Scheme 2. Synthesis of SATE prodrug derivatives.13 Reagents and
conditions: (a) Pivaloyl chloride, pyridine, RT; (b) BnNH2, CCl4, DCM,
RT, 46-63% (2 steps); (c) BBr3, DCM −80 °C; (d) HCO2H, ACN, H2O,
50°C, 15-26% (2 steps); (e) Pyridine, tetrazole (0.45 M in ACN), −5°C to
RT; (f) tert-butylhydroperoxide, 0 °C to RT, 32-46% (2 steps).
Figure 1. Chemical structure and ORTEP drawing of 35b showing RP
configuration at phosphorous (hydrogens were removed for clarity).
An initial set of SATE derivatives 20−21 and L-aminoacid
phosphoramidates 29−35 was evaluated in the HCV replicon
assay and results are reported in Table 3. SATE derivatives
20a−b and 21a−b displayed modest activities in the
micromolar range (Entries 1−4) with no significant difference
between the diastereomers as observed previously for this class
of prodrug.15,19 L-aminoacid phosphoramidate protides SAR is
well documented and our results were in accordance with
reported data in other nucleoside series.20 A brief study of the
amino acid side chain of phosphoramidate protides showed
that compounds displayed excellent to good potencies when R3
=Me (29a−b, Table 2, Entries 5−6) and R1= CH2CHMe2
(30a−b, Table 2, Entries 7−8) and are completely inactive in
the case of R1= CHMe2 (31a−b, Table 2, Entries 9−10).
Among the diastereomers, the SP isomers 29a, 30a, 32a, 33a
and 34a demonstrated activities 2.6- to 11-fold more active
than their RP analogues, and 29a, which bears the same
phosphoramidate moiety as sofosbuvir, was 1.8-fold more
active that the latter. Terminal carboxylic acid ester variations
in the case of alanine side chains indicated that the most
lipophilic esters were the most potent with Bn > iPr > Et
(Table 2, Entries 5−6 and 13−16). For leucine, ethyl esters
32a-b were more active than isopropyl analogues 30a-b.
Finally, D-amino acid phosphoramidates 35a-b showed poor
activities with high EC50 values (Table 2, Entries 17−18), in
line with what has been reported previously in other series.20
Scheme 3. Synthesis of phosphoramidate, phosphorodiamidate and cyclic
phosphoramidate prodrug derivatives 29-45.13 Reagents and conditions: (a)
tert-butyl magnesium chloride 1M in THF, THF, RT, 8-56% ; (b) DBU,
ACN, THF, RT, 11-67%.
A series of linear amino acid phosphoramidate (PON)
prodrugs was also prepared by reaction of 5 with variety of
phosphoramidate reagents 28 in the presence of tert-butyl
magnesium chloride in THF (Scheme 3).16 Again, the
diastereomers were separated by preparative HPLC or by
column chromatography on silica gel. Two related
phosphorodiamidates (PDN), 41 and 42, were synthesized
from 5 and 40 using the same protocol. Finally, cyclic
phosphoramidates (CPN) 44 and 45 were prepared from
phosphoramidate reagent 43 using a known protocol.17 In the
case of 35b, 45a and 45b, single crystals were isolated and X-
ray structure determination was performed allowing us to
assign compound 35b as the RP isomer (Figure 1). 45a was
In our initial screening cascade, the replicon activity was
the first filter, and was intended to be indicative of the ability
of prodrugs to deliver high triphosphate levels in the target
cells.21 Thus, to verify this assumption and to further select the
compounds of interest based on their pharmacokinetics
properties, an in vivo evaluation of liver levels of triphosphate
6 and plasma levels of parent nucleoside 5 in mice at a single
oral dose of 10mg/kg was performed on an initial set of
compounds (Table 2). The highest liver exposures were
obtained with RP phosphoramidates prodrugs 29b, 33b and

34b which delivered 1.4- to 2.1-fold more NTP than their SP
counterparts 29a, 33a, and 34a, and 2.7- to 3.9-fold more than
sofosbuvir (Entry 37). This difference between diastereomers
is significant compared to the case of sofosbuvir (SP) and its RP
analogue PSI-7976. Indeed despite having a 7.3-fold difference
in EC50, both showed the same levels of NTP in the liver
(Table 2, Entries 37−38). Interestingly, the weakest RP
phosphoramidate compounds in terms of replicon activities
provided the highest levels of liver NTP demonstrating that
antiviral activity observed in the replicon system did not
correlate with amounts of NTP produced in the mouse liver.
This observation may indicate a different metabolic pathway or
differentiated kinetics of cleavage between the diastereomers
as it was observed in the 2’β-C-Me-2’-α-C-F-uridine series.22
In the case of leucine 32, liver TP levels were lower than with
alanine derivative 34; however 32b, which displayed only
2.5µM activity, gave same levels of liver NTP as 32a (Table 2,
Entry 12). Similarly prodrugs such as SATE 20a and 20b were
also evaluated in vivo, despite their poor EC50, in order to
provide a comparison to IDX184. In these cases, significant
levels of liver NTP were observed (Table 2, Entries 1−2 and
39) within 1.8 to 2.6-fold less compared to IDX184. These
unexpected results prompted us to further evaluate compounds
in vivo that exhibited modest to poor EC50 to make sure we
would base our selection on the liver delivery propensity of the
prodrug rather than its in vitro replicon activity. Compounds
35a and 35b were therefore evaluated in vivo and very high
NTP levels of 3750 and 6200 pmol•h/g respectively (Table 2,
Entries 17−18), were observed, significantly above the levels
resulting from sofosbuvir administration.
and in vivo (Table 2). As mentioned previously, increasing
lipophilicity by incorporation of the carboxylic ester with n-
butyl 36 and benzyl 37 improved EC50 values and provided
slightly higher liver NTP levels. Based on the potential
liabilities found by others in the literature,20c, 23 n-butyl ester
derivative 36b was abandoned and 35b, namely IDX21437,
was selected for further preclinical studies.
While having selected 35b as a preclinical candidate, our
efforts
to
evaluate
other
prodrugs
continued.
Phosphorodiamidates bearing two L-alanine units 41 or two D-
alanine units 42 displayed 12 µM and >100 µM EC50,
respectively. As the D-alanine moiety provided high liver TP
levels in the phosphoramidate prodrug series, and although 42
is inactive in the HCV replicon system, it was evaluated in vivo
and good levels of NTP were detected in the liver. This
additional result demonstrated clearly that it is very likely that
a different metabolism, specific to the D-amino acid moeity,
occurred in the liver compared to the HCV replicon cells
system.
Continuing our SAR of D-amino-acid derived prodrugs,
CPN bearing L-alanine 44 and D-alanine 45 were studied.
Similarly to the linear phosphoramidates, the D-alanine
derivatives 45a and 45b produced high levels of liver NTP
whereas L-alanine derivatives 44a and 44b showed lower
levels with a dramatic difference of 11-fold between the
diastereomers. On the basis of the CPN results, we investigated
cyclic SATE prodrugs 26−27. Dramatic improvements in EC50
and in liver NTP levels were observed with 26 compared to 20.
However concerns around potential toxic metabolites such as
episulfide led us to discontinue further evaluation of 26 and 27,
despite the lack of conclusive evidence of its formation in vivo
in the case of IDX184. Despite 35b having already being
selected for further preclinical evaluation, 45a and 45b were
also selected as preclinical candidates and their biological
profiles will be reported in due course.
We therefore changed our screening cascade by the placing
in vivo mouse study as the primary filter for our prodrug
selection as it was apparent that the in vitro cell assay and in
vivo cell displayed differentiate metabolic activity. To this end,
a
new series of prodrugs consisting of D-amino acid
phosphoramidates 36−39, phosphorodiamidates 41−42, CPOs
26−27 and CPN 45 were synthesized and evaluated in vitro
Table 2. HCV replicon activity, selectivity index and pharmacokinetic parameters for pronucleotides 20−4513
2’-C-MeUTP Liver
DNAUC0-24h
2’-C-MeU
AA
P
EC50
Entry
Cpd
Series
R1
R3
SId
Plasma DNAUC0-24h
seriesa
conf.b
(µM)c
(pmol•h/g)e
(pmol•h/mL)e
1
2
20a
20b
21a
21b
29a
29b
30a
30b
31a
31b
32a
32b
SATE
/
/
OH
OH
NHEtoc
NHEtoc
iPr
/
SP
RP
SP
RP
SP
RP
SP
RP
SP
RP
SP
RP
3.61
3.99
>28
>25
>79
>86
>633
>105
>90
>8
96
590
930
ND
ND
1300
970
ND
ND
ND
ND
2100
940
"
/
150
ND
ND
560
840
ND
ND
ND
ND
270
220
3
"
/
/
1.269
1.159
0.158
0.953
1.108
12.596
>100
>100
0.291
2.486
4
"
/
/
5
PON
L
L
L
L
L
L
L
L
Me
6
"
"
"
"
"
"
"
iPr
Me
7
iPr
CH2CHMe2
CH2CHMe2
CHMe2
CHMe2
CH2CHMe2
CH2CHMe2
8
iPr
9
iPr
>1
10
11
12
iPr
>1
Et
>344
>40
Et

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
33a
33b
"
L
L
L
L
D
D
D
D
D
D
D
D
D
D
L
D
L
L
D
D
/
Bn
Bn
Me
SP
RP
SP
RP
SP
0.071
0.185
0.257
0.845
9.6
>1408
>541
>389
>118
>10
850
1200
400
2500
1900
1100
1100
1050
1400
1700
1700
1700
1400
ND
"
Me
34a
"
Et
Me
34b
"
Et
Me
840
35a
"
iPr
Me
3750
6200
7200
8400
6900
5600
ND
*
RP
35b
"
iPr
Me
56.8
>2
36a
"
nBu
nBu
Bn
Me
SP
0.528
2.18
>189
>46
36b
"
Me
RP
SP
37a
"
Me
0.819
4.283
2.548
8.136
>100
37.4
>122
>23
37b
"
Bn
Me
RP
SP
38a
"
Et
Me
>39
38b
"
Et
Me
RP
SP
>12
ND
ND
39a
"
Et
CH2CHMe2
>1
ND
ND
39b
"
Et
CH2CHMe2
RP
NA
NA
RP
SP
>3
ND
ND
41
PDN
iPr
Me
Me
Me
Me
Me
Me
/
12.03
>100
0.508
4.022
31.834
42.299
0.248
0.417
0.351
0.073
0.284
2.08
>8
ND
ND
42
"
iPr
>1
490
210
44a
CPN
iPr
>197
>25
21
570
44b
"
"
"
iPr
520
1100
770
*
RP
45a
iPr
>3
2400
7500
5000
3600
ND
*
SP
45b
iPr
>2
910
CPO
SATE
26a
OH
OH
NHEtoc
NHEtoc
iPr
SP
RP
SP
RP
>403
>240
>285
>1370
>352
>48
1300
710
26b
"
/
/
27a
"
"
/
/
ND
27b
/
/
3300
310
1600
330
*
SP
Sofosbuvir
PSI-7976
IDX184
PON
"
L
L
Me
Me
iPr
RP
310
320
SATE
0.203
>370
275
45
a AA = aminoacids; b P conf. = phosphorus configuration assumed and assigned by 31P NMR chemical shift comparison with compounds marked with *
indicating that X-ray structure determination was performed; c Inhibition of HCV replication Huh-7 cells containing HCV Con1 subgenomic replicon (GS4.1
cells) with a luciferase read-out. EC50 values were determined from the 50% inhibition versus concentration data; d SI= selectivity index = CC50/EC50, CC50’s
were calculated as the concentration that caused a 50% of cells death versus concentration data, all compounds had CC50 >100 µM; e DNAUC: Dose Normalized
AUC= AUC/mg/kg

Scheme 4: Proposed metabolic pathway for 35b (IDX21437) in liver cells.
accordance with a previous report.23 These results revealed a
clear difference between the metabolic pathways of sofosbuvir
and IDX21437 (35b) and confirmed that the weak in vitro anti-
HCV activity of IDX21437 in the commonly used Huh-7-
derived replicon cells was the result of under-expression of
CES1.
It has been reported that preferred phosphoramidates of the
2’β-C-methyl-2’-α-C-F-uridine series contain L-amino acid
moieties. L-Alanine phosphoramidates, such as sofosbuvir and
its diastereomer PSI-7976, are metabolized to their
intermediate monophosphate through the action of cathepsin A
(CatA), and carboxylesterase 1 (CES1), that are responsible of
the ester cleavage. The second metabolic step, which is the
cleavage of the P-N bond, is promoted by the histidine triad
nucleotide binding protein 1 (Hint-1).21 It was also reported
that CES1 is poorly expressed in Huh-7 cells containing the
HCV con1 genomic replicon which explains the difference in
activity between the sofosbuvir (SP) and PSI-7976 (RP), the
former preferentially cleaved by CatA and the latter by CES1,
despite both being substrates of the two enzymes. Unlike L-
amino acid phosphoramidates, little has been reported about
the metabolic activation of D-amino acid-containing
phosphoramidate pronucleotides. As such, the metabolic
activation of 35b was investigated. Our initial hypothesis was
that D-amino-acid phosphoramidates would be largely cleaved
by CES1 which would rationalize the lack of activity in the
HCV replicon assay. Furthermore, as CES1 is predominantly
expressed in the liver and intestine,24 while CatA is abundantly
present in many organs e.g. in human heart tissue,25 a prodrug
mainly activated by CES1 would be able to deliver high level
of active NTP in the liver as it is the case for the D-alanine
phosphoramidates.
Additional experiments (data not shown) indicated that
cytochromes 3A4/5 (CYP 3A4/5) were also able to process
35b to 46 (Scheme 4), such CYP-mediated activation via
isopropyl oxidation of esters was observed by other groups in
CPO series.26 It demonstrated that the metabolism of 35b is
more complex than original hypothesis, and that the formation
of cytidine metabolites, as shown elsewhere,27 was not studied
in this article but was further observed and will be discuss in a
separate report.
Table 3. In vitro processing of 35a−b and sofosbuvir by
CES1 and CatA purified enzymes at indicated times.
% metabolized
PD
Entry
Cpd
type
CES1
3h
CES1
21h
CatA
18h
1
2
3
35a
35b
D-SP
D-RP
L-SP
ND
23
4.5
94
15
0
0
SOF
12
100
ND= not determined.
To this end, a transient HCV replicon assay overexpressing
CES1 was set up. In this assay, 35b had an EC50 of 0.323µM
while it had an EC50 of 56.8 µM in the wild-type cell assay
(Table 2, Entry 18), indicating that CES1 plays an important
role in the metabolic activation of 35b. Sofosbuvir, being
efficiently hydrolyzed by CatA, remained largely unaffected
by transfection with CES1, with its activity slightly decreasing
in the transient HCV replicon assay (EC50=0.552 µM) versus
the wild type assay (EC50= 0.284 µM).
In order to be an efficient inhibitor of HCV replication, 35b
must be metabolized not only transiently to its monophosphate
49, but also to its corresponding active triphosphate anabolite
6. In order to measure the effective in vitro formation of NTP
6, a study in cultured human cells was performed using
multiple different types of cells, including: hepatoma cell line
Huh-7, which are commonly used in the HCV replicon model
system, fresh iCell hepatocytes (human stem cells derived),
and iCell cardiomyocytes to assess metabolic specificity of
IDX21437. These cells were exposed to 100 µM IDX21437
and then analyzed for the formation of NTP 6. The cells were
also exposed to 10 µM sofosbuvir for comparison and
analyzed for sofosbuvir-TP formation. As shown on Figure 2,
IDX21437 and sofosbuvir were both readily metabolized to the
corresponding triphosphates in hepatocytes. As expected
IDX21437 was inefficiently processed in Huh-7 cells to NTP 6
and yielded an AUC of 437 pmol•h/106cells, confirming that
the lack of CES1 expression in Huh-7 cells explains its weak
in vitro anti-HCV activity in the Huh-7-derived replicon cells.
In the same assay, and at a 10-fold lower dose than IDX21437,
sofosbuvir was efficiently processed and an AUC of 35748
pmol•h/106cells was measured (Figure 2). More notably, only
Following these preliminary results, an in vitro metabolism
study of D-Alanine phosphoramidate pronucleotides 35a−b
and sofosbuvir by purified CES1 and CatA enzymes was
undertaken, and the results are reported in Table 3. CatA did
not hydrolyse the D-alanine phosphoramidates 35a and 35b
(IDX21437) even after 18 hours of incubation, while
sofosbuvir was completely metabolized under the same
conditions. CES1 displayed a clear substrate preference for
35b (RP) with 94% of the compound hydrolyzed after 21 hours
and only 4.5% of the SP diastereomer 35a hydrolyzed during
the same time. Further, sofosbuvir hydrolysis by this enzyme
did not progress in a time-dependent manner (Entry 3, Table 3)
with 12% processed at 3 hours and 15% at 21 hours, in

5. Afdhal, N.; Reddy, K. R.; Nelson, D. R.; Lawitz, E.; Gordon, S.
C.; Schiff, E.; Nahass, R.; Ghalib, R.; Gitlin, N.; Herring, R.;
Lalezari, J.; Younes, Z. H.; Pockros, P. J.; Di Bisceglie, A. M.;
Arora, S.; Subramanian, G. M.; Zhu, Y.; Dvory-Sobol, H.;
Yang, J. C.; Pang, P. S.; Symonds, W. T.; McHutchison, J. G.;
Muir, A. J.; Sulkowski, M.; Kwo, P. N. Engl. J. Med. 2014,
370, 1483.
6. Pierra, C.; Amador, A.; Benzaria, S.; Cretton-Scott, E.;
D’Amours, M.; Mao, J.; Mathieu, S.; Moussa, A.; Bridges, E.
G.; Standring, D. N.; Sommadossi, J.-P.; Storer, R.; Gosselin,
G. J. Med. Chem. 2006, 49, 6614.
traces of NTP 6 were found in human cardiomyocytes after
exposure for 72 h and 96 h to IDX21437, whereas sofosbuvir
was efficiently processed to its triphosphate analogue, again at
a 10 times lower concentration. These data and further study28
suggested that, contrary to sofosbuvir, the metabolism of
pronucleotide 35b occurs efficiently in hepatocytes with little
uptake and/or metabolism in cardiac cells such as
cardiomyocytes or CES1 deficient cells such as Huh-7.
7. Coats, S. J.; Garnier-Amblard, E. C.; Amblard, F.; Ehteshami,
M.; Amiralaei, S.; Zhang, H.; Zhou, L.; Boucle, S. R. L.; Lu,
X.; Bondada, L.; Shelton, J. R.; Li, H.; Liu, P.; Li, C.; Cho, J.
H.; Chavre, S. N.; Zhou, S.; Mathew, J.; Schinazi, R. F.
Antiviral Res. 2014, 102, 119.
8. Golitsina, N. L.; Danehy, F. T. J.; Fellows, R.; Cretton-Scott,
E.; Standring, D. N. Antiviral Res. 2010, 85, 470.
9. Sizun, G.; Pierra, C.; Peyronnet, J.; Badaroux, E.; Rabeson, C.;
Benzaria-Prad, S.; Surleraux, D.; Loi, A. G.; Musiu, C.; Liuzzi,
M.; Seifer, M.; Standring, D.; Sommadossi, J.-P.; Gosselin, G.
Future Med. Chem. 2015, 7, 1675.
10. Sofia, M. J. Antiviral Chem. Chemother. 2011, 22, 23.
11. Murakami, E.; Niu, C.; Bao, H.; Micolochick Steuer, H. M.;
Whitaker, T.; Nachman, T.; Sofia, M. J.; Wang, P.; Otto, M. J.;
Furman, P. A. Antimicrob. Agents Chemother. 2008, 458.
12. Idenix unpublished results but activities of NM106 1 and 2
were reported by: Kirschberg, T. A.; Mish, M. R.; Zhang, L.;
Squires, N. H.; Wang, K.-Y.; Cho, A.; Feng,J. Y.; Fenaux, M.;
Babusis, D.; Park, Y.; Ray, A. S.; Kim, C. U. Bioorg. Med.
Chem. Lett. 2015, 25, 1040.
13. Gosselin, G.; C. Parsy, C.; Alexandre, F.-R.; Rahali, H.;
Griffon, J.-F.; Milhau, J.; Surleraux, D.; Dousson, C.; Pierra,
C.; Moussa, A.; Mayes, B.WO2014058801A1, 2014.
14. Gillerman, I.; Fischer, B. Nucleosides Nucleotides Nucleic
Acids 2010, 29, 245.
Figure 2: Measured AUC of NTP 6 and sofosbuvir-TP after
incubation of cells with 100µM of 35b (IDX21437) or 10µM of
sofosbuvir for 24 h to 72 h in Huh-7 cells, 48 h to 96 h in
hepatocytes, or 24 h to 168 h in cardiomyocytes.
Based on its overall favorable profile and its high level of
NTP delivered in the liver, 35b (IDX21437) was selected as
clinical development candidate and is currently in phase II
clinical trials as MK-3682 (uprifosbuvir).29
15. Mayes, B. A.; Wang, J.; Arumugasamy, J.; Arunachalam, K.;
Baloglu, E.; Bauer, D.; Becker, A.; Chaudhuri, N.; Glynn, R.;
Latham, G. M.; Li, J.; Lim, J.; Liu, J.; Mathieu, S.; McGarry, F.
P.; Rosinovsky, E.; Soret, A. F.; Stewart, A.; Moussa, A. Org.
Process Res. Dev. 2015, 19, 520.
In conclusion, we have described the discovery of 35b
(IDX21437), a novel RP D-amino acid-based phosphoramidate
monophosphate prodrug of 2’-α-chloro-2’-β-C-methyluridine.
Despite showing very weak activity in the in vitro replicon cell
assay, 35b demonstrated high levels of active triphosphate 6 in
mouse liver and in human hepatocytes. Furthermore, due to its
important CES-1 metabolic activation, 35b displayed high
level of NTP in the liver cells compared to other cell types
such as CES1-deficient Huh-7 cells and human
cardiomyocytes. 35b is currently in phase II clinical
development for the treatment of HCV infection.
16. (a) Ross, B. S.; Ganapati Reddy, P.; Zhang, H.-R.; Rachakonda,
S.; Sofia, M. J. J. Org. Chem. 2011, 76, 8311. (b) Pradere, U.;
Garnier-Amblard, E. C.; Coats, S. J.; Amblard, F.; Schinazi, R.
F. Chem. Rev. 2014, 114, 9154.
17. Jain, H. V.; Kalman, T. I. Bioorg. Med. Chem. Lett. 2012, 22,
4497.
18. Meppen, M.; Pacini, B.; Bazzo, R.; Koch, U.; Leone, J. F.;
Koeplinger, K. A.; Rowley, M.; Altamura, S.; Di Marco, A.;
Fiore, F.; Giuliano, C.; Gonzalez-Paz, O.; Laufer, R.; Pucci, V.;
Narjes, F.; Gardelli, C. Eur. J. Med. Chem. 2009, 44, 3765.
19. Cretton-Scott, E.; Hernandez-Santiago, B.; Larsson, M.; Gupta,
K. US2010003217A1 (2010).
20. (a) McGuigan, C.; Perrone, P.; Madela, K.; Neyts, J. Bioorg.
Med. Chem. Lett. 2009, 19, 4316 (b) Gardelli, C.; Attenni, B.;
Donghi, M.; Meppen, M.; Pacini, B.; Harper, S.; Di Marco, A.;
Fiore, F.; Giuliano, C.; Pucci, V.; Laufer, R.; Gennari, N.;
Marcucci, I.; Leone, J. F.; Olsen, D. B.; MacCoss, M.; Rowley,
M.; Narjes, F. J. Med. Chem. 2009, 52, 5394. (c) Sofia, M. J.;
Bao, D.; Chang, W.; Du, J.; Nagarathnam, D.; Rachakonda, S.;
Reddy, P. G.; Ross, B. S.; Wang, P.; Zhang, H.-R.; Bansal, S.;
Espiritu, C.; Keilman, M.; Lam, A. M.; Micolochick Steuer, H.
M.; Niu, C.; Otto, M. J.; Furman, P. A. J. Med. Chem., 2010,
53, 7202.
21. (a) Ma, H.; Jiang, W.-R.; Robledo, N.; Leveque, V.; Ali, S.;
Lara-Jaime, T.; Masjedizadeh, M.; Smith, D. B.; Cammack, N.;
Klumpp, K.; Symons, J. J. Biol. Chem. 2007, 282, 29812. (b)
Chang, W.; Bao, D.; Chun, B.-K.; Naduthambi, D.;
Nagarathnam, D.; Rachakonda, S.; Reddy, P. G.; Ross, B. S.;
Zhang, H.-R.; Bansal, S.; Espiritu, C. L.; Keilman, M.; Lam, A.
M.; Niu, C.; Steuer, H. M.; Furman, P. A.; Otto, M. J.; Sofia,
M. J. ACS Med. Chem. Lett. 2011, 2, 130.
22. Murakami, E.; Tolstykh, T.; Bao, H.; Niu, C.; Micolochick
Steuer, H. M.; Bao, D.; Chang, W.; Espiritu, C.; Bansal, S.;
Lam, A. M.; Otto, M. J.; Sofia, M. J.; Furman, P. A. J. Biol.
Chem. 2010, 285, 34337.
Acknowledgments
We thank many colleagues from our HCV development
team and discovery at Idenix and MSD. We thank also Dr Arie
Van der Lee from the Institut Européen des Membranes (IMR
n°5635-CNRS, ENSCM, UMII) for single crystal X-ray
studies. Finally, F.-R. Alexandre thanks Dr Izzat Raheem for
his invaluable assistance in reviewing this manuscript.
References and notes
1.
2.
Hepatitis C. Fact sheet #164; WHO, updated April 2017;
retrieved from:
http://www.who.int/mediacentre/factsheets/fs164/en/
Ghany, M. G.; Strader, D. B.; Thomas, D. L.; Seeff, L. B.
Hepatology 2009, 49, 1335.
3.
4.
Kwong, A. D. ACS Med. Chem. Lett. 2014, 5, 214.
(a) Lim, T. R.; Tan, B. H.; Mutimer, D. J. Int. J. Antimicrobial
Agents 2014, 43, 17. (b) Hayes, C. N.; Chayama, K. J.
Formosan Medical Association 2015, 114, 204.
23. Cho, A.; Zhang, L.; Xu, J.; Lee, R.; Butler, T.; Metobo, S.;
Aktoudianakis, V.; Lew, W.; Ye, H.; Clarke, M.; Doerffler, E.;
Byun, D.; Wang, T.; Babusis, D.; Carey, A. C.; German, P.;

Sauer, D.; Zhong, W.; Rossi, S.; Fenaux, M.; McHutchison, J.
G.; Perry, J.; Feng, J.; Ray, A. S.; Kim, C. U. J. Med. Chem.
2014, 57, 1812.
24. Zhu, H. J. ; Brinda, B. ; Froehlich, T. E.; Markowitz, J. S.
Pharmacogenetics Genomics 2012, 22, 215.
25. Jackman, H. L.; Massad, M. G.; Sekosan, M.; Tan, F.;
Brovkovych, V.; Marcic, B. M.; Erdös, E. G. Hypertension
2002, 39, 976.
26. (a) Niu, C.; Tolstykh, T.; Bao, H.; Park, Y.; Babusis, D.; Lam,
A. M.; Bansal, S.; Du, J.; Chang, W.; Reddy, P. G.; Zhang, H.-
R.; Woolley, J.; Wang, L.-Q.; Chao, P. B.; Ray, A. S.; Otto, M.
J.; Sofia, M. J.; Furman, P. A.; Murakami, E. Antimicrob.
Agents Chemother. 2012, 56, 3767 (b) Jonckers, T. H. M.;
Tahri, A.; Vijgen, L.; Berke, J. M.; Lachau-Durand, S.; Stoops,
B.; Snoeys, J.; Leclercq, L.; Tambuyzer, L.; Lin, T.-I.; Simmen,
K.; Raboisson, P. J. Med. Chem. 2016, 59, 5790.
27. Pinho, P.; Kalayanov, G.; Westerlind, H.; Rosenquist, Å.;
Wähling, H.; Sund, C.; Almeida, M.; Ayesa, S.; Tejbrant, J.;
Targett-Adams, P.; Eneroth, A.; Lindqvist, A. Bioorg. Med.
Chem. Lett. 2017, 27, 3468.
28. Lagrutta, A.; Regan, C. P.; Zeng, H.; Imredy, J. P.; Koeplinger,
K.; Morissette, P.; Liu, L.; Wollenberg, G.; Brynczka, C.;
Lebrón, J.; DeGeorge, J.; Sannajust, F. Scientific Reports 2017,
7, 44820.
29. Press Release 22 April 2017
http://investors.merck.com/news/press-release-
details/2017/Merck-Announces-New-Phase-2-Data-on-
Investigational-Triple-Combination-Therapy-MK-3682B-for-
Chronic-Hepatitis-C/default.aspx