Improved process for azilsartan medoxomil: A new angiotensin receptor blocker

Article abstract of DOI:10.1021/op3002867

An improved process for the active pharmaceutical ingredient of a new angiotensin II AT₁ receptor antagonist, azilsartan medoxomil, has been developed. The results include reinvestigation of the described process as well as its novel modifications. This new process includes transformation of the CN group into amidoxime moiety by aqueous hydroxylamine, its cyclization into the corresponding oxadiazole by treatment with dialkyl carbonates, and the following hydrolysis of the ester and transformation into the medoxomil ester. Several thus far undocumented side products were identified, and some of them were synthesized and duly characterized as potential impurities. Formation and control of possible critical impurities are described.

Full text of DOI:10.1021/op3002867

Subscriber access provided by UNIV OF ALABAMA BIRMINGHAM
Full Paper
Improved Process for Azilsartan Medoxomil:
A New Angiotensin Receptor Blocker
Stanislav Radl, Josef Cerny, Jan Stach, and Zuzana Gablikova
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/op3002867 • Publication Date (Web): 18 Dec 2012
Downloaded from http://pubs.acs.org on December 25, 2012
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.
Organic Process Research & Development 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 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Improved Process for Azilsartan Medoxomil: A
New Angiotensin Receptor Blocker
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
Stanislav Rádl*, Josef Černý, Jan Stach, and Zuzana Gablíková
Zentiva – A Sanofi Company, U kabelovny 130, 102 01 Prague 10, Czech Republic
ABSTRACT - An improved process for the active pharmaceutical ingredient of a new
angiotensin II AT1 receptor antagonist azilsartan medoxomil has been developed. The results
include reinvestigation of the described process as well as its novel modifications. This new
process includes transformation of the CN group into amidoxime moiety by aqueous
hydroxylamine, its cyclization into the corresponding oxadiazole by treatment with dialkyl
carbonates and the following hydrolysis of the ester and transformation into the medoxomil ester.
Several so far undocumented side-products were identified and some of them were synthesized
and duly characterized as potential impurities. Formation and control of critical possible
impurities is described.
1. INTRODUCTION
Angiotensin II AT1 receptor antagonists, called also ARB‘s (Angiotensin Receptor Blockers)
or sartans, are drugs interacting with the renin-angiotensin-aldosterone system.1,2 Most of them
are widely used in the treatment of hypertension and some other cardiovascular diseases. The last
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 30
1
2
3
4
addition to this therapeutic class, a prodrug form of azilsartan (1) azilsartan medoxomil (2), was
5
6
approved in 2011 and marketed under the trade name Edarbi (Figure 1).
7
8
9
N
N
OEt
O
OEt
O
N
O
N
O
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
N
NH
N
NH
O
O
O
HO
O
O
O
azilsartan (1)
azilsartan medoxomil (2)
Figure 1. Strucures of azilsartan (1) and azilsartan medoxomil (2)
When we started development of a generic equivalent of azilsartan medoxomil, only very
limited information of its synthesis was available. Original patents3 and papers4,5 provided only
basic facts on the synthesis of azilsartan (1) and only patents6 described transformation of
azilsartan (1) into azilsartan medoxomil (2). The reported route is schown in Scheme 1.
Scheme 1. Reported Synthetic Scheme for Azilsartan Medoxomil
R
N
N
N
N
N
N
O
OEt
OEt
OEt
ClCOOR
Et3N or Py
OH
N
NH2OH . HCl
MeONa or Et3N
O
N
O
NH2
NH2
CN
MeO
O
MeO
O
MeO
O
3a
4a
5a
xylene, reflux
medoxomil chloride
base
or
medoxomil alcohol
TsCl
N
N
N
OEt
O
OEt
O
NH
OEt
O
N
O
LiOH or NaOH
aq. MeOH
N
O
N
O
N
NH
N
N
NH
O
O
O
HO
O
MeO
O
O
O
azilsartan medoxomil (2)
azilsartan (1)
6a
A literature review revealed only limited variations of the synthesis of azilsartan medoxomil.
Synthetic pathway described in references3-6 is shown in Scheme 1. In the first step, methyl ester
3a is treated with hydroxylamine liberated in situ from its hydrochloride with either sodium
2
ACS Paragon Plus Environment

Page 3 of 30
Organic Process Research & Development
1
2
5
3
methoxide or triethylamine giving 55 % yield of amidoxime 4a (Ref. ). Treatment of 4a with
alkyl chloroformates in the presence of triethylamine or pyridine provided then crude
intermediates 5a, which were without purification thermally cyclized to provide only low yields
of methyl ester 6a. The described yields of 23 % (for R = Et)3 and 52 % (R = 2-ethylhexyl)5 after
chromatography are not acceptable for commercial production. No side products explaining the
low yields of the first steps are mentioned in the references. In the next step, saponification of
ester 6a then provided azilsartan (1) in good yields. Two possible routes of transformation of
azilsartan (1) into azilsartan medoxomil (2) were described in ref.6 The first is based on reaction
of azilsartan (1) with medoxomil chloride (7) using either pre-formed disodium salt of azilsartan
or in situ formed salt with triethylamine. In both cases, only low yields (14-22 %) of azilsartan
medoxomil (2) were reported.6 The second procedure uses medoxomil alcohol (8), which, after
proper activation of azilsartan (1) either with TBz-Cl or TsCl, provided azilsartan medoxomil (2)
in good yields. Of course, no information about possible impurities is given either in the patent
or in other literature sources.
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
O
O
O
O
Cl
HO
O
O
7
8
Figure 2. Strucures of medoxomil chloride (7) and medoxomil alcohol (8)
To overcome the problems associated with the reported processes and to find reasons for the
low yields in critical steps, we planned to study process of Scheme 1 in details. Herein we report
an improved and scalable process for the preparation of azilsartan (1) and azilsartan medoxomil
(2). In addition, an impurity profile study including conditions for the formation and control of
impurities is discussed.
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 30
1
2
3
4
5
6
7
8
9
The main goal of the generic API development is developing an economic, safe and robust
patent non-infringing process, which can be used in the production of commercial batches of the
products. During the development, formed impurities and/or degradation products are identified
and very often their standards are either isolated or synthesized. During the initial phase of the
azilsartan (1) development we have found some inconsistencies in the data given in the above
mentioned references. This paper explains the inconsistencies in the original references,
describes our process of preparation of azilsartan medoxomil (2), as well as synthesis and
physico-chemical characteristics of selected side products.i
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
2. RESULTS AND DISCUSSION
2.1. Synthesis of amidoximes 4. First the published reaction of 3a with hydroxylamine
hydrochloride under the described conditions [DMSO (dimethyl sulfoxide), 90 °C, MeONa or
Et3N] was repeated. We found,7 that after the reaction time indicated in the original references (4
h for MeONa and 15 h for Et3N), the mixtures contained 49.9 % and 35.2 % of the starting
compound 3a, 32.3 % and 44.2 % of the product 4a, and 17.6 % and 19.7 % of a side product,
respectively. The side product was identified7 as 9a which was in accordance with the easy
i
There are some inconsistencies in the Takeda publications. For example, ref.3
describes preparation of 4a using NH2OH.HCl/MeONa in DMSO in 90 % yield after 4h
heating at 90 °C. In our hands, only about 70 % conversion was achieved under
these conditions and beside about 40 % of the product, the mixture contained also
about 30 % of 9a. In all other Takeda publications, the best yields achieved were
55% using Et3N instead of MeONa. Though they obtained in some steps very low
yields, no side products explaining the low yields are mentioned. Two possible
routes of transformation of azilsartan into azilsartan medoxomil were described
in ref.6 The first is based on reaction of azilsartan with medoxomil chloride
using either pre-formed disodium salt of azilsartan or in situ formed salt with
triethylamine. In both cases, only low yields (14-22 %) of azilsartan medoxomil
were reported. The fact, that alkylation on the oxadiazole ring is taking place
is not given.
4
ACS Paragon Plus Environment

Page 5 of 30
Organic Process Research & Development
1
2
deethylation of candesartan reported earlier.8 Surprisingly, no corresponding amidoxime 10a was
detected under these conditions (Figure 3). Independent synthesis of these compounds has
already been reported.7
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
H
N
H
N
O
O
N
N
OH
N
NH2
CN
MeO
O
MeO
O
9a
10a
Figure 3. Structures of 2-desethyl side products 9a and 10a
In attempts to optimize the reaction conditions, we tested several solvents [DMSO (dimethyl
sulfoxide), DMF (N,N-dimethylformamide), DMA (N,N-dimethylacetamide), NMP (N-
methylpyrrolidone)] and bases [MeONa, EtONa, tert-BuOK, NaOH, Na2CO3, K2CO3, NaHCO3,
KHCO3,
AcONa,
Et3N,
DIPEA
(N,N-Diisopropylethylamine),
DBU
(1,8-
diazabicyclo[5.4.0]undec-7-ene), proton sponge (1,8-bis(dimethylamino)naphthalene)] at
temperatures ranging from 20 °C to 100 °C with little success. No improvement was achieved by
replacing hydroxylamine hydrochloride with other commercially available hydroxylamine salts
(sulfate, phosphate). Finally we succeeded using commercially available 50 % aqueous
hydroxylamine base in combination with DMSO. The results of this optimization have been
reported recently.7
The heterocyclic moiety present in azilsartan, i.e., 2-ethoxybenzimidazole-7-carboxylic acid, is
also present in candesartan cilexetil (11), a prodrug of candesartan (12) (For a review on
candesartan cilexetil, see ref.9). Since generic candesartan cilexetil is available in some
countries, also intermediates for its production are available. Besides methyl ester 3a used by
Takeda, the corresponding ethyl ester 3b, which is widely used by generic API producers, is also
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 30
1
2
3
4
5
6
commercially available. We found, that using our procedure, both of these potential starting
compounds provided similar results, but yields were consistently better with the ethyl ester.7
7
8
9
N
N
N
N
OEt
OEt
N
N
N N
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
N
NH
N
NH
O
O
HO
O
O
O
O
candesartan cilexetil (11)
candesartan (12)
Figure 4. Structures of candesartan (12) and candesartan cilexetil (11)
2.2. Cyclization of amidoximes 4 with phosgene equivalents. Another problem of
the original synthesis was transformation of amidoxime 4a to oxadiazole 6a via the
corresponding (alkoxycarbonyloxy)carbamimidoyl derivatives 5. In case of the reported thermal
cyclization of 5, the reported yields for ethoxycarbonyl and 2-ethylhexoxycarbonyl derivative
were only 23 and 54 %, respectively. We identified main impurities as the corresponding 2-
desethyl and N-ethyl derivatives of 6a, the fact which was consistent with our observations with
candesartan.8 Based on these findings, we also developed an efficient base-initiated cyclization
of 5 into oxadiazoles 6.7 However, the direct transformation of amidoximes 4 into oxadiazoles 6
was our ultimate goal.
In our attempts to develop shorter and more efficient direct cyclization of 4 into 6, we
suggested using a phosgene equivalent. Due to the high sensitivity of the 2-ethoxy group, it was
evident that such a process would not be compatible with either high temperatures or acid
conditions. Therefore we screened only methods of the 1,2,4-oxadiazol-5(4H)-one synthesis
proceeding under mild conditions. As the phosgene equivalents we considered diphosgene,
triphosgene, 1,1’-carbonyldiimidazole (CDI) and methyl, ethyl and phenyl dicarbonates. Since
solubility of the starting amidoximes is limited, only some dipolar aprotic solvents could be
6
ACS Paragon Plus Environment

Page 7 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
effectively used. We tested common solvents of this type as DMSO, DMF, DMA, NMP and
HMPA, but also some cyclic urea derivatives as 1,1,3,3-tetramethylurea (TMU), 1,3-dimethyl-
3,4,5,6-tetrahydro-2(1H)-pyrimidinone (DMPU) and their mixtures. We also screened a range of
organic and inorganic bases, including amines [triethylamine, diisopropylethylamine (DIPEA)
1,4-diazabicyclo[2.2.2]octane (DABCO), 8-diazabicyclo[5.4.0.]undec-7-en (DBU), 1,8-bis-
(dimethylamino)naphthalene, 1-methylmorpholine, 1-ethylpiperidine], alcoholates (MeONa,
MeOK, EtONa, tert-BuONa, tert-BuOK, tert-AmONa), carbonates (Na2CO3, K2CO3, Cs2CO3),
hydrogen carbonates (NaHCO3, KHCO3), and acetates (NaOAc, KOAc).
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
Very quickly we identified 1,1’-carbonyldiimidazole and dicarbonates as suitable phosgene
equivalents and therefore more detailed screening was done only for these reagents. In case of
combination of 1,1’-carbonyldiimidazole/DBU, the best results were achieved using DMSO,
NMP and DMPU as solvents (Table 1).ii
Table 1. Solvent Screening for Cyclization with 1,1’-carbonyldiimidazole/DBU
Entry
A
Solvent
DMSO
NMP
Yield (%)
HPLC Purity of 6a (%)
85
81
68
77
55
96.6
94.7
97.8
89.9
52.1a
B
C
DMI
D
DMPU
THF
E
ii
A Patent Application (added as ref.15) describing the cyclization using CDI/DBU
in THF giving 68% yield of 6a appeared when our laboratory development was
already finished and this manuscript submitted. We have extensively studied the
reaction using CDI with both methyl (as described in the Patent Application) and
ethyl ester using not only DBU, but also a range of bases. In our hands, THF was
much worse than most of the tested solvents (as you can see in our Table 1).
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 30
1
2
a - 40.4 % of 4a detected (HPLC).
3
4
5
6
7
8
Though only slightly lower yields (5-12 %) were obtained using some other bases (DIPEA, 1-
methylmorpholine, 1-ethylpiperidine) under similar conditions, the present impurity profiles
were different and more complicated. We applied combination of CDI/DBU in DMSO and
obtained compounds 6a or 6b in high yields. Since DBU is quite expensive, we tried to replace it
by a cheap base. Finally we succeeded with simple potassium carbonate. Finally we found the
conditions providing comparable results as using DBU. Under these conditions, 85 % yield of 6a
of HPLC purity higher than 99% was repeatedly obtained.
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
Dialkyl carbonates are considered as cheap and green compounds used even as solvents and
therefore we have paid our attention to their use as reagents for the cyclization. Initial screening
done in DMSO revealed that in case of diphenyl carbonate (DPC) the reaction proceeds with
potassium carbonate, DBU, tert-BuOK and MeONa. On the other hand, dimethyl (DMC) and
diethyl carbonate (DEC) reacted only using tert-BuOK or MeONa. One of the screening
conditions (results are given in Table 2) consisted in simple stirring of the reaction components
in DMSO in vials at room temperature and the following HPLC analysis (in entries C, F, and I
peaks of phenole were not integrated).
Table 2. Reagent and Base Screening for Cyclization in DMSO
Entry
Reagent
Base
HPLC (%)
4a
6a
0.75
0.0
A
B
C
D
E
DMC
DEC
DPC
DMC
DEC
K2CO3
K2CO3
K2CO3
DBU
73.4
76.2
0.0
84.6
0.0
76.6
75.6
DBU
0.0
8
ACS Paragon Plus Environment

Page 9 of 30
Organic Process Research & Development
1
2
3
4
Entry
Reagent
Base
HPLC (%)
5
6
7
8
4a
0.0
6a
F
G
H
I
DPC
DMC
DEC
DPC
DBU
79.2
24.2
9.1
9
EtONa
EtONa
EtONa
48.4
6.6
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
35.0
For the above mentioned reasons, we first considered diphenyl carbonate (DPC) as a reagent of
choice. When the reaction was done in DMSO using potassium carbonate at room temperature,
the reaction was complete after 4 h. Simple workup consisted in addition of water and
acidification with acetic acid, filtration of the crude product and final crystallization. Though the
yields of the crude product were very good and the purity after crystallization was about 99 %,
the product still contained small amounts of phenol. In addition, in larger scales (up to 50 g), the
yields were gradually lower. Therefore we decided to check the possibility of using dimethyl or
diethyl carbonate in combination with the corresponding sodium alkoxide. If the reaction of
methyl ester 4a with dimethyl carbonate was done in methanol, several days of reflux was
required to complete conversion. In addition, a range of side products was formed. When the
reaction was done at 100 °C in a pressure flask for 24 h, the reaction was complete, but the
mixture contained about 16 % of impurities (HPLC). Surprisingly, when the reaction was done in
DMSO at room temperature, the reaction was complete after 30 minutes and the crude mixture
after simple dilution with water, acidification and filtration contains more than 90 % of 6a. In
this case, the yields of about 90 % of crude 6a with purity of about 98 % were obtained. After
recrystallization from acetone yields of 85-90 % and purity higher than 99 % were routinely
achieved. Even better results were obtained using ethyl ester 4b in combination with diethyl
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 30
1
2
3
4
5
6
7
8
9
carbonate and EtONa providing nearly quantitative yields of 6b. Quite surprisingly, we found
that reaction with ethyl ester 4b in combination of dimethyl carbonate and MeONa provided
methyl ester 6a. The corresponding ethyl ester 6b was present only in amounts ranging from 0 to
5 %, depending on the reaction time. The work-up consisted in simple dilution with water,
acidification and filtration of the product. Since we were interested in the background of the
reaction, we also tested reactions of 4b with dimethyl carbonate/EtONa and diethyl
carbonate/MeONa. In both cases, mixtures of 6a and 6b were obtained.
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
Finally we evaluated the tested possibilities of the synthesis of azilsartan esters 6a and 6b, and
selected for further development compound 4b as the starting substrate (higher yields in the first
step) in combination with dimethyl carbonate (cheap and green reagent) and sodium methoxide
(cheap and commercially available). We have also optimized amount of DMSO with regards of
reaction time at room temperature and purity of the crude product.
2.3. Hydrolysis of esters 6. Esters 6a and 6b were easily hydrolyzed using aqueous
NaOH, KOH, or LiOH. We have not noticed any substantial difference in yields or purity of the
formed azilsartan (1).
2.4. Transformation of azilsartan (1) into azilsartan medoxomil (2). Since
medoxomil chloride (7) is used for the synthesis of olmesartan medoxomil12 and therefore is
commercially available at reasonable price, we first tested possibility of using this compound as
a medoxomil group source. However, under all the conditions tested, both the carboxy group and
the oxadiazole group were alkylated. Therefore the possibility of using medoxomil alcohol (8)
was further studied.
Besides the activation of azilsartan with 2,4,6-trichlorobenzoyl chloride (TBz-Cl) or TsCl
reported in the patent literature,6 we also tested some other known literature methods. However,
10
ACS Paragon Plus Environment

Page 11 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
we found the use of tosyl chloride activation the most appropriate, since the reagent is cheap, and
both the yields and the purity are acceptable. In order to understand the reaction background in
more details, we checked the reaction mixture by HPLC. To our surprise, a very complex
mixture was initially formed, but after some time, the reaction mixture containing about 90 % of
azilsartan medoxomil (2) was obtained. HPLC results of this study are shown in Figure 5. Peak
corresponding to azilsartan (1) quickly disappeared and after 15 min two main peaks, one of
them corresponding to tosyl chloride, were present (See Fig. 5A). After 30 min, the mixture was
even more complex (see Fig. 5B). Composition of the mixture after 1, 1.5 and 2 hours is shown
as Figs 5C, 5D, and 5E, respectively. Finally, after 3 h, the composition of the mixture was much
more reasonable with the only major peak corresponding to the expected product (Fig. 5F).iii
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
Fig. 5A
Fig. 5B
iii
The changes of the HPLC content were so fast that we have not tried to track
them by LCMS. We do not have any explanation but the results were repeatedly
obtained.
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 30
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
Fig. 5C
Fig. 5D
Fig. 5E
Fig. 5F
Figure 5. HPLC changes during the reaction of azilsartan (1) with medoxomil alcohol (8)
We found that quality of the used azilsartan (1) as well as medoxomil alcohol (8) is crucial.
When azilsartan purified by crystallization from an alcohol was used, even after thorough drying,
the corresponding ester of azilsartan was detected as an impurity. Therefore we finally purified
azilsartan (1) by stirring its suspension in acetone. Formation of methyl ester 6a was also
observed using medoxomil alcohol (8) from one supplier and subsequent GC analysis found
methanol as a residual solvent in this material. Since the most medoxomil alcohol (8) producers
probably use the literature procedure13 where the last step is done in methanol, the methanol
content should be carefully checked.
12
ACS Paragon Plus Environment

Page 13 of 30
Organic Process Research & Development
1
2
We also found that the compound forms solvates with a range of solvents14 and some of them,
i.e., solvates with acetone, THF or DMA, could be used for its purification. This method is
suitable for removing most of the impurities, including methyl ester 6 and dimeric impurities 15
and 17. Final desolvating is then done by stirring the slurry with aqueous acetone under reflux.
X-Ray powder diffraction (XRPD) spectra and differential scanning calorimetry (DSC) of
azilsartan medoxomil (2) and its solvate with N,N-dimethylacetamide are shown in Figures 6 and
7, respectively.
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
Desolvated azilsartan medoxomil
Azilsartan medoxomil DMA solvate
Counts
2000
Counts
6000
4000
2000
0
1000
0
10
20
30
40
10
20
30
40
Position [°2Theta] (Copper (Cu))
Position [°2Theta] (Copper (Cu))
Figure 6. XRPD of azilsartan medoxomil and its solvate with N,N-dimethylacetamide
Desolvated azilsartan medoxomil
Azilsartan medoxomil DMA solvate
Figure 7. DSC of azilsartan medoxomil and its solvate with N,N-dimethylacetamide
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 30
1
2
3
4
5
6
7
8
9
2.5. Synthesis of standards of impurities. One of the principal parts of documentation
of any active pharmaceutical ingredient (API) is description of impurities and/or degradation
products which can be present.10 Identified impurities should be included in the specification
when they are present at a level higher than the identification threshold, which is usually 0.10 %.
These impurities must be not only identified but also either isolated or independently
synthesized. Recently, the drug registration authorities are increasingly interested even in
pharmaceutical impurities in the range 0.01–0.1% (Ref.11). To anticipate possible problems
associated with the reported processes, we planned to study reactions of Schemes 1 and 2 in
detail. Consequently, we have tried to synthesize all principal impurities/degradation products
identified during the process development.
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
Having some experience with impurities of similar candesartan cilexetil (11),8 as well as with
impurities formed in the first steps of the original synthesis,7 compounds 13 and 14 should be the
logical impurities of azilsartan medoxomil (Figure 8), which could be even product related
degradation products formed during a long-term storage. Therefore these compounds were also
synthesized. Treatment of azilsartan medoxomil (2) with HCl in acetone provided compound 13.
Compound 14 was prepared by simple alkylation of 2 with ethyl iodide in the presence of
K2CO3.
O
O
H
N
N
N
O
N
O
N
O
O
NH
N
N
O
O
O
O
O
O
O
O
O
O
13
14
Figure 8. Structures of potential impurities of azilsartan medoxomil 13 and 14
14
ACS Paragon Plus Environment

Page 15 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Analysis of crude azilsartan medoxomil (2) revealed only trace amounts of 13 and 14 and
small amount of methyl ester 6a.iv However, the product contained also 2 more critical
impurities. One of them was identified by LC-MS as compound with molecular formula
C55H42N8O13, which could be easily explained by structural formula 15 (Figure 9). Therefore,
content of the corresponding diol 16 in medoxomil alcohol (8) is even more critical than the
methanol content.
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
O
O
HN
NH
O
O
N
N
O
O
N
N
N
O
N
O
HO
OH
O
O
O
O
O
O
O
O
15
16
Figure 9. Structure of potential impurity 15 and the corresponding diol 16
The second critical impurity has molecular formula C54H42N8O10, which can be explained by
structural formula 17. This assumption is supported by the fact, that compound of formula
C50H40N8O7, which corresponds to structure 18, was also identified by MS in crude compound
6a. The corresponding acid of formula C49H38N8O7 was also present in small amounts in
azilsartan (1). Independent synthesis of the mentioned impurities 15, 17 and 18 is underway and
will be published elsewhere.
iv
The amounts lower than 0.03 %, if detected.
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 30
1
2
3
4
N
N
O
NH
O
O
5
6
7
N
O
O
N
N
N
O
O
N
N
N
N
O
O
N
N
N
NH
O
N
8
9
O
O
O
O
MeO
O
O
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
17
18
Figure 10. Structure of potential impurity 17 and the corresponding ester 18
3. SUMMARY
We have developed robust processes for synthesis of azilsartan medoxomil (2) starting from
nitriles 3, consisting of their treatment with aqueous hydroxylamine, followed by transformation
of the formed amidoximes 4 into azilsartan esters 6 using phosgene equivalents, advantageously
dialkyl carbonates. These compounds are then hydrolyzed to azilsartan (1), which is
consequently treated with medoxomil alcohol (8) providing the final compound. The selected
route starting from ethyl ester 3b has been repeatedly checked in amounts of hundreds of grams.
The reactions were routinely done in jacketed laboratory reactors with mechanical stirring
(Radleys) with internal temperature control.
16
ACS Paragon Plus Environment

Page 17 of 30
Organic Process Research & Development
1
2
3
4
Scheme 2. Synthetic Scheme of the Zentiva Process of Azilsartan Medoxomil Production
5
6
7
8
9
N
N
N
N
N
N
OEt
OEt
OEt
O
aq. NH2OH
DMSO, 90 °C
DMC, MeONa
DMSO, RT
O
OH
N
N
NH
NH2
CN
EtO
O
EtO
O
MeO
O
85-90 %
85 %
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
3b
4b
6a
aq. NaOH
50 °C
88-90 %
N
N
N
OEt
O
OEt
O
O
medoxomil alcohol
solvent, TsCl
N
O
N
NH
N
NH
O
O
HO
O
85-90 %
O
O
O
azilsartan medoxomil (2)
azilsartan (1)
We have also detected and/or identified several potential impurities of the final product. Some
of them, e.g., compounds 13 and 14, were also synthesized as standards for analytical
development. All of them were also duly characterized by H NMR, 13C NMR, MS, and
1
elemental analysis. We believe that this information would be of interest of process chemists
working in this area.
The article reports results of our laboratory development and due to the patent expiration time,
the scale-up at Zentiva is not planned in the near future. Of course, additional aspects should be
duly studied during further development. These include safety testing of the particular reaction
steps, conditions for crystallizations of the intermediates as well as the final product (cooling
times, possible seeding, crystal forms, filterability, particle size, etc)v. It must be stressed that the
article describes preparation of the azilsartan medoxomil API, while in the final formulation of
v
Generally the seeding was not necessary and the cooling times were not crucial
to get good yields with acceptable filtration times.
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 30
1
2
3
4
5
6
7
8
9
Takeda, the corresponding potassium salt of azilsartan medoxomil is used. In this case,
preparation of the salt is correctly described in the Takeda patents6 and can be used by generic
companies.
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. EXPERIMANTAL SECTION
Methyl and ethyl 1-((2'-cyanobiphenyl-4-yl)methyl)-2-ethoxy-1H-benzo[d]imidazole-7-
carboxylate (3a) and (3b), respectively, were obtained from Zhejiang Tianyu Pharmaceutical
Company (http://www.tianyupharma.com). Other chemicals used in the synthesis were
purchased from Sigma-Aldrich and were used as supplied.
Melting points were measured on a Kofler block and are uncorrected. NMR experiments were
carried out on a Bruker Avance 500 at 500.13 MHz (1H) and 125.77 MHz (13C). Reference for
1H δ (CDCl3) = 7.26 ppm, for 13C δ (CDCl3) = 77.0 ppm. IR spectra were measured on a FTIR
spectrometer Nicolet Nexus (Thermo, USA) using ZnSe ATR crystal technique by accumulation
of 64 scans with 2 1/cm resolution. The Mass spectra (MS/MS; ionization mode APCI(+)) were
measured on an API 3000 PE machine (Sciex Instruments, Applied Biosystems). XRPD Spectra
were measured on PANalytical's X'Pert PRO Materials Research Diffractometer with graphite
monochromator using CuKα radiation (λ=1.542 Å). DSC measurements were done on a Perkin
Elmer Pyris 1 Differential Scanning Calorimeter. The purity of the prepared substances was
evaluated by TLC on silica gel (FP KG F 254, Merck) and by HPLC system HP Agilent 1050
[column Phenomenex Luna 5ꢀ C18(2) length: 0.25 m, internal diameter 4.6 mm] with UV
detection (240 nm). Gradient elution with mobile phase A (phosphate buffer [1. 2 g NaH2PO4
diluted in 1000 mL of water, pH adjusted to 3.0 with 50% phosphoric acid), and mobile phase B
(methanol) was used. The results are given as LCAP (liquid chromatography area percent). Flash
chromatography was performed on silica gel Merck, particle size 0.04-0.063 mm. Centrifugally
18
ACS Paragon Plus Environment

Page 19 of 30
Organic Process Research & Development
1
2
accelerated axial chromatography was done using CyclographTM instrument (Analtech) with
silica gel pre-scraped rotors.
3
4
5
6
7
8
9
Methyl
2-Ethoxy-1-((2'-(N'-hydroxycarbamimidoyl)biphenyl-4-yl)methyl)-1H-
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
benzo[d]imidazole-7-carboxylate (4a). A mixture of 3a (10 g, 2.4 mmol), DMSO (75 mL),
and 50 % aqueous hydroxylamine (5 mL) was stirred at 90 °C for 18h. Then the mixture was
poured into water (250 mL), the mixture was stirred for 30 min, the insoluble portion was filtered
off and washed with water providing, after drying, 10.6 g of white precipitate containing
according to HPLC 91.5 % 3a. The solid was crystallized from 2-propanol to give 8.0 g (75 %)
of white crystals; Mp: 203-206 °C (ref.5 Mp: 207-209 °C). HPLC purity 97.5 %. HRMS for
C25H25N4O4 (M+H)+ Calcd: 445.1876, found: 445.1992. 1H NMR (DMSO) δ (ppm): 9.18 (s, 1H,
OH), 7.70 (dd, J = 7.9, 1.2 Hz, 1H, Ar), 7.46 (dd, J = 7.9, 1.2 Hz, 1H, Ar), 7.45-7.33 (m, 3H,
Ar), 7.35 (d, J = 8.3 Hz, 2H, Ar), 7.29 (dd, J = 7.6, 1.5 Hz, 1H, Ar), 7.19 (t, J = 7.9 Hz, 1H, Ar),
6.94 (d, J = 8.3 Hz, 2H, Ar), 5.55 (bs, 2H, NH2), 5.51 (s, 2H, N-CH2-Ar), 4.62 (q, J = 7.1 Hz,
13
2H, OCH2CH3), 3.71 (s, 3H, OCH3), 1.42 (t, J = 7.1 Hz, 3H, OCH2CH3). C NMR (DMSO) δ
(ppm): 166.22, 158.33, 151.96, 141.63, 139.85, 139.74, 135.54, 133.20, 130.83, 130.02, 129.85,
128.85, 128.52, 126.90, 125.92, 122.85, 121.55, 120.81, 115.56, 66.62, 52.32, 46.29, 14.40.
Anal. Calcd. for C25H24N4O4: C, 67.55; H, 5.44; N, 12.60. Found: C, 67.27; H, 5.72; N, 12.87.
Ethyl
2-Ethoxy-1-((2'-(N'-hydroxycarbamimidoyl)biphenyl-4-yl)methyl)-1H-
benzo[d]imidazole-7-carboxylate (4b). 50 % Aqueous hydroxylamine (240 mL) was added
to a stirred suspension of 3b (400 g, 941 mmol) in DMSO (2400 mL) and the mixture was stirred
at 90 °C for 15 h. Then the mixture was diluted with water (400 mL) and slowly cooled down to
15 °C under stirring and then stirred at this temperature for 1 h. The insoluble portion was
filtered off and washed with i-PrOH (5 x 1 L). The crude product was dried in a vacuum drier
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 30
1
2
3
4
(35 °C/50 mbar) to provide 386.4 g (90 %) of white precipitate containing according to HPLC
5
6
97.8 % 4b; Mp: 209-211 °C (2-propanol). HRMS for C26H27N4O4 (M+H)+ Calcd: 459.2032,
7
8
9
1
found: 459.2189. H NMR (DMSO) δ (ppm): 9.19 (bs, 1H, OH), 7.70 (dd, J = 7.9, 1.2 Hz, 1H,
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
Ar), 7.46 (dd, J = 7.9, 1.2 Hz, 1H, Ar), 7.42-7.34 (m, 5H, Ar), 7.27 (dd, J = 7.6, 0.9 Hz, 1H, Ar),
7.19 (dd, J = 7.6, 1.5 Hz, 1H, Ar), 6.92 (d, J = 8.4 Hz, 2H, Ar), 5.57 (bs, 2H, NH2), 5.53 (s, 2H,
N-CH2-Ar), 4.62 (q, J = 7.1 Hz, 2H, OCH2CH3), 4.21 (q, J = 7.1 Hz, 2H, OCH2CH3), 1.42 (t, J =
7.1 Hz, 3H, OCH2CH3), 1.17 (t, J = 7.1 Hz, 3H, OCH2CH3). 13C NMR (DMSO) δ (ppm):
165.74, 158.26, 151.92, 141.58, 139.83, 139.78, 135.53, 133.20, 130.77, 130.03, 129.82, 128.85,
128.54, 126.91, 125.91, 122.84, 121.48, 120.78, 115.89, 66.62, 61.05, 46.20, 14.41, 13.93. IR:
ν(N-H) 3515, 3407, ν(O-H) 3254, ν(C-H) 2986, ν(C=O) 1703, ν(C=C) + ν(C=N) 1634, 1611,
1545, ν(C-O) 1284, 1256, 1136 cm-1. Anal. Calcd. for C26H26N4O4: C, 68.11; H, 5.72; N, 12.22.
Found: C, 67.78; H, 5.88; N, 12.54.
Methyl
2-Ethoxy-1-((2'-(5-oxo-4,5-dihydro-1,2,4-oxadiazol-3-yl)biphenyl-4-
yl)methyl)-1H-benzo[d]imidazole-7-carboxylate (6a).
Method A) A mixture of 4a (8 g, 18 mmol), DMSO (70 mL), 1,1’-carbonyldiimidazole (3.5
g, 21.6 mmol), and DBU (3.5 g, 23 mmol) was stirred at room temperature for 4 h. The mixture
was poured into water (300 mL) and the solution was acidified with 5% HCl to pH about 4-5.
The formed precipitate was filtered off, washed with water and dried on air to provide 8.4 g of
white precipitate. Its crystallization from acetone provided 7.7 g (91 %) of white crystals was
obtained; Mp: 192-196 °C. HPLC purity 99.1 %.
Method B) A mixture of 4a (20 g, 45 mmol), DMSO (130 mL), 1,1’-carbonyldiimidazole (8
g, 50 mmol), and K2CO3 (9 g, 65 mmol) was stirred at room temperature for 3 h. The mixture
was poured into water (500 mL) and the solution was acidified with 5% HCl. The formed
20
ACS Paragon Plus Environment

Page 21 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
precipitate was filtered off, washed with water and dried to provide 22.2 g of white precipitate.
Its crystallization from acetone provided 18.1 g (85 %) of white crystals was obtained; Mp: 194-
198 °C. HPLC purity 99.3 %.
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
Method C) Diphenyl carbonate (0.32 g, 1.5 mmol) was added to a mixture of 4a (0.44 g, 1
mmol) and potassium carbonate (0.2 g, 1.4 mmol) in DMSO (10 mL) and the mixture was stirred
at room temperature for 4 h. The mixture was poured into water (25 mL), acidified with acetic
acid and the formed precipitate was filtered off, washed with water and dried providing 0.43 g
(91 %) of white precipitate of HPLC purity 93.2 %. Its crystallization from ethyl acetate
provided 0.33 g (74 %) of white crystals; Mp: 194-197 °C. HPLC purity 98.8 %.
Method D) A 30% (w/w) solution of MeONa in MeOH (31 mL) was added dropwise (10
minutes) to a solution of 4a (36 g, 81 mmol) and dimethyl carbonate (20.5 mL = 22 g, 244
mmol) in DMSO (205 mL) stirred at temperature of about 10 °C. The cooling bath was removed
and the mixture was stirred for 20 minutes (temperature of about 21 °C). The mixture was added
into cold water (1 L, 5 °C) and acidified with 5% aqueous HCl. The formed precipitate was
filtered off, washed with cold water and dried to provide 38.7 g of white precipitate. Its
crystallization from acetone provided 33.1 g (87 %) of white crystals was obtained; Mp: 195-198
°C. HPLC purity 98.7 %.
Method E) A 30% (w/w) solution of MeONa in MeOH (190 mL) was added dropwise (15
minutes) to a mixture of 4b (190 g, 414 mmol) and dimethyl carbonate (112 g, 1243 mmol) in
DMSO (900 mL) stirred under cooling at inner temperature not exceeding 25 °C. The cooling
bath was removed and the mixture was stirred for 2 h (temperature of about 20-25 °C). Cold
water (5 °C, 3.6 L) was added, the mixture was cooled down to 15 °C and an aqueous HCl
(prepared from 72 mL of concd. HCl and 72 mL of water) was added during 1 h. The formed
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 30
1
2
3
4
5
6
7
8
9
mixture was stirred at 15 °C for 1 h, the precipitate was filtered off, washed with water (3 x 300
mL) and dried in a vacuum drier (35 °C/50 mbar) to provide 201.5 g of white precipitate. The
crude product of HPLC purity of 88.4 % containing also 4.1 % of ethyl ester 6b and 4.4 % of
azilsartan (1) was used for the next step (hydrolysis) without further purification. A sample
purified by crystallization from acetone ( HPLC purity 98.3 %) melted at 194-198 °C. HRMS for
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
C26H23N4O5 (M+H)+ Calcd: 471.1668, found: 471.1688. H NMR (DMSO) δ (ppm): 12.39 (s,
1H, NH), 7.70 (dd, J = 7.9, 1.1 Hz, 1H, Ar), 7.68-7.63 (m, 2H, Ar), 7.55 (td, J = 7.4, 1.1 Hz, 1H,
Ar), 7.46 (m, 2H, Ar), 7.24 (d, J = 8.1, 2H, Ar ), 7.19 (t, J = 7.9 Hz, 1H, Ar), 7.00 (d, J = 8.1
Hz, 2H, Ar), 5.54 (s, 2H, N-CH2-Ar), 4.62 (q, J = 7.1 Hz, 2H, OCH2CH3), 3.69 (s, 3H, OCH3),
1.39 (t, J = 7.1 Hz, 3H, OCH2CH3). 13C NMR (DMSO) δ (ppm): 166.15, 159.45, 158.32, 158.26,
141.64, 140.78, 137.73, 136.73, 131.85, 130.90, 130.60, 130.16, 128.81, 128.80, 126.24, 122.91,
122.18, 121.58, 120.82, 115.50, 66.61, 52.18, 46.38, 14.35. Anal. Calcd. for C26H22N4O5: C,
66.37; H, 4.71; N, 11.91. Found: C, 66.22; H, 4.93; N, 11.66.
Ethyl
2-Ethoxy-1-((2'-(5-oxo-4,5-dihydro-1,2,4-oxadiazol-3-yl)biphenyl-4-
yl)methyl)-1H-benzo[d]imidazole-7-carboxylate (6b). Using the procedure described for
the synthesis of 6a from 4a (Method A) and starting from ethyl ester 4b, 98 % yield of 6b was
obtained; Mp: 179-182 °C (acetone). HPLC purity 97.7 %. HRMS for C27H25N4O5 (M+H)+
1
Calcd: 485.1825, found: 485.1798. H NMR (DMSO) δ (ppm): 12.40 (s,1H, NH), 7.69 (dd, J =
7.9, 1.2 Hz, 1H, Ar), 7.50 (m, 1H, Ar), 7.44 (dd, J = 7.9, 1.2 Hz, 1H, Ar), 7.43-7.40 (m, 4H, Ar),
7.35 (d, J = 7.7 Hz, 1H, Ar), 7.19 (t, J = 7.9 Hz, 1H, Ar), 6.95 (d, J = 8.3 Hz, 2H, Ar), 5.57 (s,
2H, N-CH2-Ar), 4.60 (q, J = 7.1 Hz, 2H, COOCH2CH3), 4.18 (q, J = 7.1 Hz, 2H, OCH2CH3),
13
1.39 (t, J = 7.1 Hz, 3H, COOCH2CH3), 1.16 (t, J = 7.1 Hz, 3H, OCH2CH3). C NMR (DMSO)
δ (ppm): 165.70, 159.46, 158.26, 158.23, 141.60, 140.76, 137.76, 136.74, 131.84, 130.86,
22
ACS Paragon Plus Environment

Page 23 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
130.60, 130.18, 128.84, 127.80, 126.23, 122.92, 122.17, 121.51, 120.79, 115.83, 66.60, 61.01,
46.30, 14.35, 13.85. IR: ν(C-H) 2979, ν(C=O) 1771, 1713, ν(C=C) + ν(C=N) 1610, 1544, ν(C-O)
1275, 1124 cm-1. Anal. Calcd. for C27H24N4O5: C, 66.93; H, 4.99; N, 11.56. Found: C, 66.69; H,
4.73; N, 11.27.
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
2-Ethoxy-1-((2'-(5-oxo-4,5-dihydro-1,2,4-oxadiazol-3-yl)biphenyl-4-yl)methyl)-1H-
benzo[d]imidazole-7-carboxylic Acid, Azilsartan (1). A mixture of crude 6a obtained by
Method E (200 g, 425 mmol) and aqueous NaOH (100 g in 2 L of water, 2.5 mol) was stirred at
50 °C for 3 h (HPLC checking). The mixture was diluted with acetone (1 L) and acidified with
acetic acid (about 150 mL) at 45 °C. Then the mixture was diluted with water (700 mL), slowly
cooled down to 20°C (4 h) and stirred for additional 1 h at this temperature. The formed
precipitate was filtered off, washed with a mixture of acetone - water (200 mL, 1 : 2 v/v) and
dried in dark in a vacuum drier (35 °C/50 mbar) overnight to provide 185.6 g of off white
precipitate (HPLC purity 95.7 %). This product was suspended in acetone (360 mL) and stirred
under reflux for 1 h. After cooling to room temperature, the insoluble material was filtered off to
get 173.2 g (89 %) of azilsartan (1); Mp: 208-211 °C (ref.5 m. p. 212-214 °C). HPLC purity 98.9
1
%. HRMS for C25H21N4O5 (M+) Calcd: 457.1512, found: 457.1597. H NMR (DMSO) δ (ppm):
13.17 (bs, 1H, OH or NH), 12.42 (bs, 1H, OH or NH), 7.70-7.60 (m, 3H, Ar), 7.57-7.50 (m, 2H,
Ar), 7.50-7.44 (m, 1H, Ar), 7.23 (d, J = 8.3 Hz, 2H, Ar), 7.18 (t, J = 7.9 Hz, 1H, Ar), 7.05 (d, J
= 8.3 Hz, 2H, Ar), 5.68 (s, 2H, N-CH2-Ar), 4.58 (q, J = 7.1 Hz, 2H, OCH2CH3), 1.38 (t, J = 7.1
Hz, 3H, OCH2CH3). 13C NMR (DMSO) δ (ppm): 167.51, 159.49, 158.27, 158.26, 141.63,
140.70, 137.71, 137.16, 131.87, 131.26, 130.65, 130.20, 128.85, 127.81, 126.61, 123.47, 122.13,
121.45, 120.69, 116.57, 66.48, 46.34, 14.36. Anal. Calcd. for C25H20N4O5: C, 65.78; H, 4.42; N,
12.27. Found: C, 65.55; H, 4.66; N, 11.92.
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 30
1
2
3
4
5
6
7
8
9
(5-Methyl-2-oxo-1,3-dioxol-4-yl)methyl 2-Ethoxy-1-((2'-(5-oxo-4,5-dihydro-1,2,4-
oxadiazol-3-yl)biphenyl-4-yl)methyl)-1H-benzo[d]imidazole-7-carboxylate,
Azilsartan Medoxomil (2), solvate with N,N-dimethylacetamide (1:1)
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
Potassium carbonate (23 g, 167 mmol), 4-toluenesulfonyl chloride (27.5 g, 144 mmol) and
DMAP (2.0 g, 16 mmol) was added to a solution of azilsartan (1; 50 g, 110 mmol) and
medoxomil alcohol (8; 18.8 g, 144 mmol) in N,N-dimethylacetamide (500 mL) at 30 °C and the
mixture was stirred for 3 h at this temperature. The mixture was diluted with water (300 mL),
acidified with concentrated aqueous hydrochloric acid to pH about 5 and a seed (0.1 g) was
added. The formed white crystals were filtered off to give 63.6 g (88.5 %) of crude product as a
solvate with N,N-dimethylacetamide after drying in vacuo at 45°C for 5 h. Mp: 111-114 °C; mp
1
(onset temperature) by DSC: 110 °C. HPLC purity 99.2 %. H NMR (500 MHz, DMSO)
δ (ppm): 12.37 (bs, 1H, NH), 7.73 (dd, J = 7.9, 1.2 , 1H, Ar), 7.69-7.61 (m, 2H, Ar), 7.55 (dd, J
= 7.6, 1.2, 1H, Ar), 7.52 (dd, J = 7.9, 1.2 , 1H, Ar), 7.50-7.44 (m, 1H, Ar), 7.22 (d, J = 8.2 Hz,
2H, Ar), 7.21 (t, J = 7.7 Hz, 1H, Ar), 7.00 (d, J = 8.2 Hz, 2H, Ar), 5.55 (s, 2H, N-CH2-Ar), 5.12
(s, 2H, COOCH2-R), 4.60 (q, J = 7.0 Hz, 2H, OCH2CH3), 2.94 (s, 3H, CH3CON(CH3)2), 2.78 (s,
3H, CH3CON(CH3)2), 2.16 (s, 3H, CH3), ), 1.95 (s, 3H, CH3CON(CH3)2), 1.38 (t, J = 7.0 Hz,
3H, OCH2CH3). 13C NMR (500 MHz, CDCl3) δ (ppm): 169.54, 165.04, 159.43, 158.38, 158.21,
151.77, 141.70, 140.73, 140.35, 137.81, 136.66, 133.12, 131.83, 131.17, 130.60, 130.12, 128.82,
127.82, 126.36, 123.42, 122.17, 122.14, 120.87, 114.48, 66.67, 54.67, 46.43, 37.41, 34.45, 21.37,
14.32, 8.83.
(5-Methyl-2-oxo-1,3-dioxol-4-yl)methyl
2-Ethoxy-1-((2'-(5-oxo-4,5-dihydro-1,2,4-
oxadiazol-3-yl)biphenyl-4-yl)methyl)-1H-benzo[d]imidazole-7-carboxylate,
Azilsartan Medoxomil (2)
24
ACS Paragon Plus Environment

Page 25 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The isolated crystals of solvate (63 g) were suspended in a mixture of acetone (200 mL) and
water (400 mL) and the formed suspension was stirred at 40 °C for 3 h. Then the mixture was
cooled down to 25°C, stirred at this temperature for 1 h and the insoluble portion was filtered off,
washed with water (200 mL) and dried under reduced pressure at 40 °C to provide 51.2 g (93.7
%) of the final desolvated product. HPLC purity 99.5 %. Residual solvents (acetone: 100 ppm;
N,N-dimethylacetamide: 50 ppm). Mp 177 - 179 °C; mp (onset temperature) by DSC 176 °C.
HRMS for C30H25N4O8 (M+H)+ Calcd: 569.1594, found: 569.1599. 1H NMR (500 MHz,
DMSO) δ (ppm): 12.37 (bs, 1H, NH), 7.73 (dd, J = 7.9, 1.2, 1H, Ar), 7.69-7.62 (m, 2H, Ar), 7.55
(dd, J = 7.6, 1.2, 1H, Ar), 7.52 (dd, J = 7.9, 1.2, 1H, Ar), 7.50-7.44 (m, 1H, Ar), 7.22 (d, J = 8.2
Hz, 2H, Ar), 7.24-7.19 (m, 1H, Ar), 7.00 (d, J = 8.2 Hz, 2H, Ar), 5.55 (s, 2H, N-CH2-Ar), 5.12
(s, 2H, COOCH2-R), 4.60 (q, J = 7.0 Hz, 2H, OCH2CH3), 2.16 (s, 3H, CH3), 1.38 (t, J = 7.0 Hz,
3H, OCH2CH3). 13C NMR (500 MHz, CDCl3) δ (ppm): 165.04, 159.43, 158.38, 158.21, 151.77,
141.70, 140.73, 140.35, 137.81, 136.66, 133.12, 131.83, 131.17, 130.60, 130.12, 128.82, 127.82,
126.36, 123.42, 122.17, 122.14, 120.87, 114.48, 66.67, 54.67, 46.43, 14.32. Anal. Calcd. for
C28H20N4O8: C, 71.42; H, 4.79; N, 11.10. Found: C, 71.42; H, 4.79; N, 11.10.
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
(5-Methyl-2-oxo-1,3-dioxol-4-yl)methyl
2-Oxo-3-((2'-(5-oxo-4,5-dihydro-1,2,4-
oxadiazol-3-yl)biphenyl-4-yl)methyl)-2,3-dihydro-1H-benzo[d]imidazole-4-
carboxylate (13). A mixture of 1 (0.57 g, 1 mmol), acetone (5 mL) and concentrated aqueous
HCl (0.1 mL) was refluxed for 1 h. After cooling, the insoluble portion was filtered off, washed
with water and dried to give 0.48 g of 13 (HPLC purity 97.4 %). Crystallization from i-PrOH
provided 0.40 g (73 %) of 13 of HPLC purity 98.6 %. Mp: 164-168 °C. HRMS for C28H21N4O8
1
(M+H)+ Calcd: 541.1359, found: 541.1473. H NMR (500 MHz, DMSO) δ (ppm): 12.38 (bs,
1H, NH), 11.51 (bs, 1H, NH), 7.67 (ddd, J = 7.6, 1.4 Hz, 1H, Ar), 7.65-7.62 (m, 1H, Ar), 7.55
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 30
1
2
3
4
5
6
(ddd, J = 7.6, 1.2 Hz, 1H, Ar), 7.50-7.45 (m, 1H, Ar), 7.30 (dd, J = 7.9, 1.2 Hz, 1H, Ar), 7.28
(dd, J = 7.9, 1.2 Hz, 1H, Ar), 7.22 (d, J = 8.2 Hz, 2H, Ar), 7.09 (t, J = 7.9 Hz, 1H, Ar), 7.04 (d, J
7
8
9
13
= 8.2 Hz, 2H, Ar), 5.36 (s, 2H, N-CH2-Ar), 5.04 (s, 2H, COOCH2-R), 2.08 (s, 3H, CH3). C
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
NMR (500 MHz, CDCl3) δ (ppm): 164.95, 159.46, 158.24, 154.90, 151.75, 140.73, 140.33,
137.62, 136.75, 133.05, 131.83, 130.63, 130.14, 129.98, 128.76, 128.04, 127.78, 126.46, 122.35,
122.10, 120.76, 113.52, 112.93, 54.69, 44.68, 8.83. Anal. Calcd. for C28H20N4O8: C, 62.22; H,
3.73; N, 10.37. Found: C, 61.91; H, 3.89; N, 10.03.
(5-Methyl-2-oxo-1,3-dioxol-4-yl)methyl 2-Ethoxy-1-((2'-(4-ethyl-5-oxo-4,5-dihydro-
1,2,4-oxadiazol-3-yl)biphenyl-4-yl)methyl)-1H-benzo[d]imidazole-7-carboxylate
(14). A mixture of 1 (0.57 g, 1 mmol), acetone (10 mL), K2CO3 (0.25 g, 1.8 mmol) and ethyl
iodide (0.2 mL) was stirred in a closed flask for 24 h. The insoluble portion was filtered off
through a celite pad, washed with acetone and the filtrate was evaporated. The residue after
evaporation was mixed with water (15 mL) and extracted with dichloromethane (3 x 20 mL).
The extract was washed with water and dried with MgSO4. The filtrate was evaporated (0.68 g)
to give an oily product, which was purified by crystallization from ethyl acetate to give 0.35 g
(59 %) of white crystals. Mp: 164-170 °C (HPLC purity 96.8). HRMS for C32H29N4O8 (M+H)+
Calcd: 597.1985, found: 597.1736. 1H NMR (500 MHz, DMSO) δ (ppm): 7.78-7.70 (m, 2H, Ar),
7.68-7.64 (m, 1H, Ar), 7.63-7.56 (m, 2H, Ar), 7.52 (dd, J = 7.9, 1.2 , 1H, Ar), 7.26 (d, J = 8.3
Hz, 2H, Ar), 7.21 (t, J = 7.9 Hz, 1H, Ar), 7.01 (d, J = 8.3 Hz, 2H, Ar), 5.53 (s, 2H, N-CH2-Ar),
5.13 (s, 2H, COOCH2-R), 4.58 (q, J = 7.1 Hz, 2H, OCH2CH3), 2.93 (q, J = 7.2 Hz, 2H,
NCH2CH3), 2.16 (s, 3H, CH3), 1.36 (t, J = 7.1 Hz, 3H, OCH2CH3), 0.58 (t, J = 7.2 Hz, 3H,
NCH2CH3). 13C NMR (500 MHz, CDCl3) δ (ppm): 165.00, 159.06, 158.40, 158.20, 151.75,
26
ACS Paragon Plus Environment

Page 27 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
141.72, 140.54, 140.35, 137.28, 137.24, 133.08, 132.47, 131.25, 131.18, 130.36, 128.66, 128.14,
126.83, 123.50, 122.22, 121.17, 120.90, 114.38, 66.63, 54.65, 46.43, 37.40, 14.30, 11.86, 8.82.
Anal. Calcd. for C32H28N4O8: C, 64.42; H, 4.73; N, 9.39. Found: C, 64.77; H, 4.81; N, 9.07.
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
AUTHOR INFORMATION
Corresponding Author
E-mail: stanislav.radl@zentiva.cz. Fax: +420-2-67243741.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGEMENTS
The authors’ thanks are also due to Drs Lukáš Plaček, Jan Srbek and Jiří Břicháč for the LC-
MS and HRMS measurements and Dr. Tomáš Pekárek for the IR spectra measurement.
ABBREVIATIONS
CDI,
1,1’-carbonyldiimidazole;
DABCO,
1,4-diazabicyclo[2.2.2]octane;
DBU,
8-
diazabicyclo[5.4.0.]undec-7-en; DEC, diethyl carbonate; DIPEA (N,N-Diisopropylethylamine);
DMA, N,N-dimethylacetamide; DMC, dimethyl carbonate; DMF, N,N-dimethylformamide;
NMP, N-methylpyrrolidone; DMPU, 1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone;
DMSO, dimethyl sulfoxide; DPC, diphenyl carbonate; DSC, differential scanning calorimetry;
HPLC, high-performance liquid chromatography; TBz-Cl, 2,4,6-trichlorobenzoyl chloride;
TLC, thin layer chromatography; TMU, 1,1,3,3-tetramethylurea; TsCl, 4-toluenesylfonyl
chloride; XRPD, X-ray powder diffraction.
.
27
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 30
1
2
3
4
REFERENCES
5
6
7
8
(1) Aulakh, G. K.; Sodhi, R. K.; Singh, M. J. Life Sci. 2007, 81, 615-639.
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
(2) Miura, S.; Karnik, S. S.; Saku, K. J. Renin Angiotensin Aldosterone Syst. 2011, 12, 1-7.
(3) a) Naka, T.; Inada, Y. (Takeda Chem. Ind.), U.S. Patent 5,583,141, 1996; b) Naka, T.;
Inada, Y. (Takeda Chem. Ind.), Eur Pat. Appl. EP 0,520, 423, 1992.
(4) Kohara, Y.; Imamiya, E.; Kubo, K.; Wada, T.; Inada, Y.; Naka, T. Bioorg. Med. Chem.
Lett. 1995, 5, 1903-1908.
(5) Kohara, Y.; Kubo, K.; Imamiya, E.; Wada, T.; Inada, Y.; Naka, T. J. Med. Chem. 1996,
39, 5228-5235.
(6) a) Kuroita, T.; Sakamoto, H.; Ojima, M. (Takeda Chem. Ind.), U.S. Pat. Appl. 0,187,269,
2005; b) Kuroita, T.; Sakamoto, H.; Ojima, M. (Takeda Chem. Ind.), U.S. Pat. Appl. 7,157,584,
2007.
(7) Rádl, S.; Černý, J.; Stach, J.; Holec, J.; Píša, O.; Gablíková, Z. J. Heterocycl. Chem.
2012, 49, in press.
(8) Havlíček, J.; Mandelová, Z.; Weisemann, R.; Střelec, I.; Plaček, L.; Rádl, S. Collect.
Czech. Chem. Commun.. 2009, 74, 347-362.
(9) a) Kubo, K.; Kohara, Y.; Imamiya, E.; Sugiura, Y.; Inada, Y.; Furukawa, Y.; Nishikawa,
K.; Naka, T. J. Med. Chem. 1993, 36, 2182-2195; b) Hansson, L. Drugs of Today 1999, 35, 117-
126; c) Julius, S.; Nesbitt, S. D.; Egan, B. M.; Weber, M. A.; Michelson, E. L.; Kaciroti, N.;
28
ACS Paragon Plus Environment

Page 29 of 30
Organic Process Research & Development
1
2
3
4
5
6
Black, H. R.; Grimm, R. H. Jr.; Messerli, F. H.; Oparil, S.; Schork, M. A. N. Engl. J. Med. 2006,
354, 1685-1697.
7
8
9
(10) International Conference on Harmonisation, ICH (2003), Q3A Impurities in New Drug
Substances.
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
(11) Rahman, N.; Azmi, S. N. H.; Wu, H.-F. Accred. Qual. Assur. 2006, 11, 69-74.
(12) Yet, L. In: The Art of Drug Synthesis (Johnson, D. S.; Li, J. J.; Eds) Wiley Interscience,
2007; pp.129-141.
(13) Alpegiani, M.; Zarini, F.; Perrone, E. Synth. Commun. 1992, 22, 1277-1282.
(14) Sharmarke, M.; Ridvan, L.; Rádl, S.; Černý, J.; Dammer, O.; Krejčík, L. (Zentiva) Czech
Pat. Appl. CZ 2012-274, 2012.
(15) Bansai, D.; Mishra, H.; Dubey, S. K.; Choudhary, A. S.; Vir, D.; Agarwal, A. (Jubilant
Life Sci.) WO2102/107814.
GRAPHICAL ABSTRACT
Improved Process for Azilsartan Medoxomil: A New Angiotensin Receptor
Blocker
Stanislav Rádl, Josef Černý, Jan Stach, and Zuzana Gablíková
29
ACS Paragon Plus Environment