Structural Correction and Process Improvement for Control of a Critical Process Impurity of Ezetimibe

Article abstract of DOI:10.1021/acs.oprd.9b00024

A new process-related impurity of ezetimibe was identified and characterized. The impurity is critical and common to most of the manufacturing routes of ezetimibe. Structural characterization using HMBC indicated the presence of a six-membered ring rather than a nine-membered ring as proposed by the innovator of ezetimibe. Prominently, the existing pharmacopoeial methods for ezetimibe are not capable of detecting this impurity. A control strategy was established by appropriate process control that is capable of purging the impurity to levels comfortably below the regulatory requirement. The formation of the diastereomer impurity during the demonstration of a scale-up batch under the optimized conditions is attributed to epimerization of ezetimibe induced by thermal degradation of the silylating agent.

Full text of DOI:10.1021/acs.oprd.9b00024

Subscriber access provided by ALBRIGHT COLLEGE
Full Paper
Structural Correction and Process Improvement for
Control of Critical Process Impurity of Ezetimibe
Madhava Rao Mannam, Srimurugan Sankareswaran, venugopal Reddy Gaddam,
Senthilkumar Natarajan, Rajasekhara Prasad Kottapalli, and Pramod Kumar
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00024 • Publication Date (Web): 04 Apr 2019
Downloaded from http://pubs.acs.org on April 4, 2019
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 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 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.
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 16
Organic Process Research & Development
1
2
3
4
5
6
Structural Correction and Process Improvement for Control of
Critical Process Impurity of Ezetimibe
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
*
Madhava Rao Mannam†,‡, Srimurugan Sankareswaran † , Venugopal Reddy Gaddam†,
*
Senthilkumar Natarajan†, Rajasekhara Prasad Kottapalli‡, and Pramod Kumar†
†Micro Labs Ltd., API R&D Centre, Bommasandra-Jigini Link Road, KIADB INDL Area,
Bommasandra, Bangalore 560105, Karnataka, India.
‡Department of Chemistry, Koneru Lakshmaiah Education Foundation, Vaddeswaram –522 502,
Andhra Pradesh, India
ABSTRACT
A new process related impurity of ezetimibe is identified and characterized. The impurity is
critical and common to most of the manufacturing routes of ezetimibe. Structural
characterization using HMBC indicates the presence of a 6-membered ring rather than a 9-
membered ring as proposed by the innovator of ezetimibe. Prominently, the existing
pharmacopoeial methods of ezetimibe are not capable of detecting this impurity. Control strategy
is established by appropriate process control that is capable of purging the impurity to levels
comfortably below the regulatory requirement. Formation of diastereomer impurity during the
demonstration of scale-up batch under optimized condition is attributed to epimerization of
ezetimibe induced by thermal degradation of the silylating agent.
KEY WORDS: Ezetimibe, Process impurity, Improved process, Structure correction,
Epimerization
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 16
1
2
3
4
INTRODUCTION
5
6
7
8
The control of pharmaceutical impurities is currently a critical and challenging issue to the
pharmaceutical industry. Impurities in pharmaceuticals are unwanted components with unwanted
pharmacological or toxicological effects that remain with the active pharmaceutical ingredients
(APIs), or develop upon its aging and influence the efficacy and safety of the pharmaceutical
products.1 Consequently, impurity profiling that describes detection, identification/structure
elucidation and quantitative determination of impurities has become crucial as per various
regulatory requirements.2 Knowing the structure of an impurity is essential for allowing
assessment of its toxicological implications and for understanding its formation mechanisms,
which is critical knowledge for improving the synthetic chemical process and optimizing the
formulation.3 Recently regulatory agencies reported a major issue regarding the detection of a
genotoxic impurity, NDMA (N-nitrosodimethylamine), and subsequently NDEA (N-
nitrosodiethylamine), in valsartan, an API used to manufacture generic angiotensin receptor
blockers (ARBs) that resulted in recalls from the market.4
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
Ezetimibe (marked as Zetia, Ezetrol or Ezedoc) is the first lipid-lowering drug that inhibits
intestinal uptake of dietary and biliary cholesterol without affecting the absorption of fat-soluble
nutrients.5 It may be used alone or together with statins for the treatment of primary
hypercholesterolemia, homozygous sitosterolemia, homozygous familial hypercholesterolemia
and mixed hyperlipidemia. Impurity profiling of ezetimibe is reported in a number of scientific
articles that assist regulatory agencies in ensuring the availability of consistently high quality
API in market.6 In an attempt to develop generic ezetimibe, following the most commonly
adopted synthetic route, we observed the formation of a critical process impurity that was not
disclosed in any of the prior art. The impurity is formed at the late stage of the process and
carried forward to final API. More importantly, available pharmacopoeial methods for impurity
profiling and assay determination using HPLC do not resolve this impurity from the main peak.
A book chapter by Schering-Plough research institute discusses the same impurity formation
during their product development.7 Structural elucidation of the impurity indicated that the
correct structure of the impurity is different from the one reported in the article by Schering-
Plough research institute.
ACS Paragon Plus Environment

Page 3 of 16
Organic Process Research & Development
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
Ezetimibe is made by a three stage process starting from 4-(4-fluorobenzoyl)butyric acid 2 as
described earlier (Scheme 1).8 Attachment of chiral auxillary to carboxylic acid group of 2 was
achieved in high yield without the necessity for protection/deprotection of aryl carbonyl group.
This requires activation of carboxylic acid 2 using pivaloyl chloride followed by reaction with
(S)-4-phenyl-2-oxazolidinone. Crystallization in methanol after acid-base work-up affords chiral
auxillary tethered intermediate 3 in good quality and 80% isolation yield.
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
O
O
(i)
(ii)
NH
Ph (S)
O
OH
+
+
O
F
F
2
F
O
O
OH
O
O
O
(iii)
HO
(S)
N
N
O
O
F
N
(S)
(S)
F
Ph
Ph
3
4
OH
OTMS
F
OTMS
O
O
(vi)
(v)
(iv)
(R)
(S)
(R)
1
(S)
(S)
N
N
O
H
N
F
O
(S)
Ph
(S)
OH
F
F
5
Scheme 1 Route of synthesis of ezetimibe: (i) TEA, Pivaloyl chloride, DMAP, DMF, Toluene,
(ii) Con.H2SO4, NaHCO3, Methanol, 82% ; (iii) DCM, BH3.DMS, (R)-CBS, MeOH, H2O2,
H2SO4, 98% ; (iv) DCM, DIPEA, TMSCl, TiCl4, Ti(OPr)4 , DCM, IPA/Methanol, aq. Tartaric
acid, EtOAc, Na2CO3, 65% ; (v) DCM, BSA, TBAF, AcOH, citric acid, 85%; (vi) 2.5N H2SO4,
IPA, Water, 95%.
Asymmetric reduction of 3 using (R)-CBS/BDMS catalyst system affords (S)-alcohol 4 in high
enantioselectivity. Subsequent silyl protection of 4 and its reaction with silyl protected imine 5 in
the presence of mixed titanium catalyst generated in situ results in stereoselective aldol-type
condensed product 5. The selectivity of desired anti isomer to undesired syn isomer was found in
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 16
1
2
3
4
5
6
7
8
9
the ratio 8:2, when performed the coupling reaction at -40oC. The selectivity and yield of the
reaction strongly depends on the nature of titanium species generated in the reaction system. It
was also noted that, the stereoselectivity was not improved further by performing the reaction at
comparatively lower temperatures. Interestingly, the aldol-condensation product was isolated as
free alcohol rather than protected trimethylsilyl derivative as reported by most of the
manufacturers.9 This isolation strategy gives a stable intermediate as required by GMP and helps
in eliminating the undesired diastereomer (syn isomer) completely during the isolation by
filtration. The quality obtained was consistently above 99%.
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
Final stage of synthesis is the cyclization of diol 5 to azetidinone under fluoride ion catalysis
conditions reported in prior art. Having achieved high purity in the penultimate intermediate (5),
the cyclization was expected to generate final API directly in high purity. However, in contrast to
the expectation, this final stage proved to be troublesome and generated critical process
impurities that are found to carryover to final API. Treatment of 5 with silylating agent, N,O-
bis(trimethylsilyl)acetamide resulted trisilylation of the molecule wherein both diol and amino
groups are silylated as trimethylsilyl groups. Addition of fluoride ion induces cyclization of N-
silyl group to generate desired ezetimibe in the protected form. Acid-catalyzed deprotection of
silyl groups using aqueous sulphuric acid in isopropyl alcohol followed by dilution with water
precipitated ezetimibe as filterable solid. Among three chemical transformations (silyl group
protection, cyclization and deprotection of silyl groups), only cyclization reaction could be
monitored by HPLC. The inherent incompatibility of silyl derivatives with HPLC mobile phase
conditions was a major challenge in development of analytical method. The reaction monitoring
data indicated high conversions during cyclization reaction although the isolated yields were
often lower. In order to understand the missing mass balance of reaction, complete precipitation
of ezetimibe was achieved by addition of excess water to aqueous IPA reaction mass. The
isolated yield was close to theoretical yield with HPLC purity over 99%. The material however
gave lower assay against reference standard, when analyzed by pharmacopoeial HPLC method
(100 min run time). No additional peaks could be observed by HPLC either in purity or assay
method. Surprisingly, in-house assay method by HPLC developed with shorter run time (15 min)
gave the clue for understanding the missing mass balance. A new peak well separated from main
peak, identified as impurity-A and amounted to missing assay was observed during the analysis
of the sample. LC-MS was immediate performed to gain more information on the impurity-A.
ACS Paragon Plus Environment

Page 5 of 16
Organic Process Research & Development
1
2
3
4
5
6
Furthermore, attempts were made to isolate the impurity-A from enriched filtrate mother liquor
employing column chromatography. The isolated impurity-A matched HPLC retention time and
served as an analytical standard.
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
N
OH
H
OH
O
N
OH
O
N
N
O
N
OH
F
O
6-Exo-Tet
F
O
9-Exo-Tet
HO
F
NH
O
O
HO
HO
F
(a)
(b)
F
(5)
F
Figure 1 Plausible structures of impurity-A with m/z 572 formed during cyclization reaction
ESI-MS analysis of impurity-A showed a mass of m/z 572, identical to that of starting material 5.
A different MS/MS pattern obtained during LC-MS analysis ruled out the possibility of
impurity-A to be diastereomer of 5. Anticipating reorganization of 5 during the reaction,
impurity disclosed by Schering-Plough research scientist for the same stage was compared.7 The
article mentioned competitive cyclization involving endo-cyclic carbonyl group of oxazolidinone
resulting in a 9-membered ring as impurity (Fig. 1a). Initially we assumed same structure to
impurity-A and found that the final step of the process generated this impurity to the level as
high as 20-30% and accounted for lower isolation yield. The formation of 9-membered cyclized
impurity at significant levels was expected to be unusual on the basis of Baldwin’s rules for ring
closure. Not many literatures are available that demonstrate facile 9-membered ring formation by
9-exo-tet pathway when more favored 4-exo-tet is possible.
Characterization techniques were utilized to establish the structure of impurity-A. Preliminary
NMR and MS analysis of impurity indicated the presence of ring structure formed by
reorganization of 5 under experimental condition. HMBC together with COSY and HSQC was
used to follow the carbon-carbon connectivity and understand the size of ring present in the
structure of impurity. Interestingly, a triplet proton signal assigned to an exchangeable proton
(confirmed by deuterium exchange with D2O as well as absence of carbon correlation in HSQC)
correlated to methylene carbon C2´ in HMBC by a two bond coupling (marked in Fig. 2 by red
arrow). Such a correlation was difficult to explain with 9-membered structure as the
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 16
1
2
3
4
5
6
7
8
9
exchangeable amide N-H proton gives a doublet correlating methylene carbon C2´ by a three
bond coupling. Additionally, a three bond coupling observed for the same triplet proton signal
with methine carbon C1´ in HMBC could not be correlated with reported structure (marked in
Fig. 2 by blue arrow). In order to explain the missing spectral assignment, an alternative 6-
membered ring structure is proposed for impurity-A (Fig. 1b). A 6-membered ring is expected to
form by the attack of nitrogen anion at the endo-cyclic carbonyl group followed by C-O bond
cleavage rather than C-N bond.11 This revised structure with primary alcohol as leaving group
explains all HMBC spectral behavior of the impurity. Since Baldwin’s rule for tetragonal system
favors all 3 to 7-exo-tet process. Further, mass fragment analyses (Fig. 3) also support the favour
of six membered ring structure, this re-assigned structure also explains the facile formation of
impurity to the level of 20-30% during cyclization process.
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
OH
O
N
C ' OH
2
C1'
N
F
O
HO
F
Figure 2: Zoomed HMBC spectrum of impurity-A. Arrows on structure indicates significant
HMBC correlations.
With HPLC method available at hand, optimization of process parameters and identification of
purification methods to minimize and eliminate impurity-A was taken up. Various reaction
parameters like nature of solvent, mode of addition of TBAF, quantity of TBAF and reaction
temperatures were screened to identify the optimum condition for the control of impurity.
ACS Paragon Plus Environment

Page 7 of 16
Organic Process Research & Development
1
2
3
4
5
H+
H+
F
F
HO
O
O
6
7
N
N
Ph
N
N
Ph
8
9
O
O
-H2O
HO
F
HO
F
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
OH
OH
m/z: 555.26
m/z: 573.26
Impurity-A
-H2O
HO
H+
F
O
N
-
Ph
N
Ph
O
F
H+
HO
F
O
N
NH
O
m/z: 537.24
HO
F
OH
-
Ph
H+
F
m/z: 453.15
O
-H2O
N
NH
O
HO
F
m/z: 435.19
Figure 3: MS spectra and fragmentation pathway of Impurity-A for structural confirmation
DCM and toluene afford complete reaction conversion for silyl protection and cyclization. In
both the cases, impurity-A was formed competitively although toluene formed only half the level
compared to similar reaction performed in DCM (Table 1). Excess of TBAF was not beneficial
in reducing the impurity. Dumping of solid TBAF to the reaction mass was adopted as it formed
slurry in toluene. It was found that slow reaction rates at lower temperature resulted slightly
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 16
1
2
3
4
5
6
7
8
9
elevated level of impurity-A compared to reaction performed at ambient temperature.
Conversely, higher temperature accelerated reaction conversion and reduced impurity-A
formation (Table 1, entry 21). Dilution was beneficial in retarding the undesired cyclization
(Table 1, entry 27). Due to generation of diastereomer (vide infra) at higher temperatures, 20-
25oC with 14 volume dilution was identified as the best reaction condition for performing the
cyclization reaction (Table 1, entry 18).
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
Table 1 Screening and optimization of cyclization reaction condition
Reaction Conditions
Impurity-A
Entry
Volume
(x)
7
Reaction
temperature (oC)
40-45
Equivalents of TBAF
(solvent)
(%)
Solvent
1
2
3
4
5
6
7
8
Acetonitrile
TBME
n-Heptane
10 mol% (solid)
10 mol% (solid)
10 mol% (solid)
2.5 mol% (DCM)
2.5 mol% (toluene)
5 mol% (toluene)
10 mol% (toluene)
2.5 mol% (DCM)
10 mol% (solid)
10 mol% (solid)
10 mol% (solid)
2.5 mol% (solid)
10 mol% (solid)
2.5 mol% (solid)
10 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid in lots)
2.5 mol% (DCM)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
2.5 mol% (solid)
--*
--*
7
7
7
7
7
7
7
7
7
7
7
7
7
7
12
12
14
14
14
14
15
15
15
15
15
20
40-45
40-45
25-30
25-30
25-30
25-30
30-35
40-45
40-45
50-55
60-65
60-65
80-85
80-85
20-25
50-55
20-25
20-25
20-25
60-65
0-5
10-15
25-30
60-65
--*
DCM
DCM
DCM
DCM
13.9
15.32
14.15
13.49
18.73
7.41
7.41
7.05
6.46
7.54
5.4
6.06
5.90
4.45
5.18
5.67
5.78
4.13
7.73
6.47
5.06
3.56
4.14
3.09
DCM
9
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
95-100
20-25
* Incomplete reaction and insignificant conversion observed.
Optimization study concluded that process parameters could not completely eliminate the
impurity-A formation and the best condition restricts impurity-A formation to the level of 3-5%.
ACS Paragon Plus Environment

Page 9 of 16
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Attention was then focused on downstream isolation procedure to establish strong process
control on the purge of impurity. The behavior of impurity-A removal was found to be critical
with respect to crystallization conditions adopted. Due to high solubility of ezetimibe in organic
solvents, it can be isolated only by precipitation technique induced by addition of an anti-solvent
like water. The most successful condition, as followed by most of the prior art disclosure, was
using IPA as solvent and water as anti-solvent. Table-2 lists the risk-assessment experiments
performed to identify the robust process condition for isolating API in high purity. Slow and
dropwise addition of 5 volumes of water to reaction mass dissolved in 5 volumes of IPA affords
good yield with consistently high purity devoid of impurity-A (Table 2, entry 5). The identified
crystallization condition is capable of removing impurity-A as high as 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
Table 2 Optimization of crystallization solvent for effective removal of impurity-A
Impurity-A
(%) content in
Impurity-A
(%) content
in Ezetimibe
Entry
Purification Conditions
Yield (%)
crude
Ezetimibe
2.56
1
2
3
4
5
6
7
8
9
10 volumes IPA & 8 volumes water
added dropwise
10 volumes IPA & 8.5 volumes
water added dropwise
10 volumes IPA & 10 volumes
water added dropwise
5 volumes IPA & 5 volumes water
dumped
5 volumes IPA & 5 volumes water
added dropwise
5 volumes IPA & 7.5 volumes
water added dropwise
5 volumes IPA & 8 volumes water
added dropwise
5 volumes IPA & 10 volumes water
added dropwise
85.0
88.0
90.0
94.0
94.0
95.0
96.0
97.0
98.0
Not Detected
Not Detected
Not Detected
0.06
2.56
2.56
2.83
2.83
2.83
2.83
2.83
4.52
Not Detected
0.07
0.08
0.23
5 volumes IPA & 12 volumes water
added dropwise
3.66
11
12
5.24
2.58
0.5% IPA in water 6 volumes
5 volumes Methanol & 5 volumes
water
97.0
82.0
4.79
0.06
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 16
1
2
3
4
5
6
7
8
9
The optimized process was transferred to production unit for execution of GMP batches. First
scale-up batch displayed performance similar to R&D batches with respect to level of impurity-
A. Unexpectedly, the batch generated a chiral SSS diastereomer impurity to the level of 2.1%
which was never observed in lab batches. A detailed investigation was immediately carried out
to understand the root cause for the formation of chiral impurity. It was evident that, desired
product underwent partial epimerization under the reaction condition to generate the epimer.
Importantly, under identical operation conditions, R&D batches did not display any
epimerization. The only difference observed in the plant batch was the additional hold up time of
reaction mass at 50-55oC till the release of reaction monitoring results from QC. Upon reviewing
various factors, we finally identified that the silylating agent BSA to be the reason for
epimerization during the scale-up batch. BSA is reported to be thermally unstable and generate
acetonitrile as the degradation product.10 During the additional hold up time (Table 3) at higher
temperature, BSA, (6.0 mol equivalents) underwent gradual degradation to generate acetonitrile
in the reaction mass. To understand the impact of acetonitrile in accelerating epimerization
process, lab batch was performed and spiked with 8% acetonitrile before the addition of TBAF.
The batch unexpectedly generated SSS diastereomer (4.6%) in higher amount. We therefore
concluded, acetonitrile formed by degradation of BSA provided solvent polarity favoring
epimerization process. The identification of valid root cause was further verified by conducting
second trial batch of same size with no hold up time (cooling the reaction mass after the
maintenance time). As expected no epimerization was observed and plant batch trend matched
lab trend for the final stage.
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
Table 3 Reaction mass holding time study for formation SSS isomer
S.No
Reaction mass
maintenance
time (hr) at
52.52.5 °C
Reaction mass
holding
time (hr) at
SSS isomer
content in
Reaction
monitoring
1.02
SSS isomer
content after
isolation
SSS isomer
after
purification
52.52.5 °C
1
2
3
4
5
1.0
1.0
1.0
1.0
1.0
0
1
2
3
4
0.18
0.35
0.44
0.62
0.77
0.04
0.07
0.13
0.19
0.25
1.37
1.52
2.10
3.27
ACS Paragon Plus Environment

Page 11 of 16
Organic Process Research & Development
1
2
3
4
5
6
EXPERIMENTAL SECTION
7
8
9
All materials were purchased from commercial suppliers. Unless specified otherwise, all
reagents and solvents were used as supplied by manufacturers. Melting points were determined
by open air capillary with Buchi M-565 and are uncorrected. IR spectra of samples were
recorded on Shimadzu IR Affinity-I FT-IR spectrophotometer. 1H NMR spectra (400 MHz) and
13CNMR spectra (100 MHz) were recorded in CDCl3, DMSO-d6, on Bruker 400 MHz NMR
(ASCEND, 5 mm PABBO) instrument and mass spectra were determined on Velos Pro Ion Trap
Mass Spectrophotometer (Thermo Scientific)
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
(3R,4S)-1-(4-Fluorophenyl)-3-[(S)-3-(4-fluorophenyl)-3-hydroxypropyl]-4-(4-hydroxy-
phenyl)azetidin-2-one (Ezetimibe) (1)
Compound 5 (80g, 0.139 mol) was added to toluene (1120 mL) under nitrogen atmosphere in a
3-neck RB flask to obtain a slurry. The reaction mass was heated to 50-55oC and BSA (170.5 g,
0.838 mol) was added slowly over a period of 1.0-1.5 h under nitrogen atmosphere. The reaction
was maintained for 1 h at the same temperature, cooled down to room temperature and
monitored for reaction completion. Sodium sulfate (5.95 g, 0.041 mol) was added to the reaction
mass and cooled down to 20-25oC. TBAF.3H2O (1.1 g, ) was added to the reaction mass as solid
and stirred for a period of 1 h under nitrogen atmosphere. After the completion of reaction, citric
acid solution (320 mL) was added slowly to the reaction mass and the layers were separated. The
toluene layer was washed with water (160 mL) and distilled completely under vacuum to obtain
viscous oil. IPA (400 mL) was added to the reaction mass followed by slow addition of dilute
sulphuric acid (40 mL). The reaction mass was maintained at the same temperature for 1 h and
then treated with slow addition of water (640 mL). The solid obtained was maintained for a
period of 1 h, filtered and washed with aqueous IPA. Wet solid was further dried under vacuum
to afford 46 g (81 %) of 1 as off-white to white solid. Characterization data of 1: MR: 164-166
1H NMR (400 MHz, DMSO-D6) δ 9.54, (s, -OH), 7.29-7.33 (m, 2H), 7.20-7.24 (m, 4H),
7.09-7.16 (m, 4H), 6.75-6.78 (d, J=8.4Hz, 2H), 5.29-5.30 (d, J=4.8 Hz, 1H), 4.80-4.81
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 16
1
2
3
4
5
6
7
8
9
(d, J=2.4Hz, 1H), 4.48-4.52 (m, 1H), 3.07-3.10 (m, 1H), 1.71-1.87 (m, 4H); 13C NMR
(400 MHz, DMSO-D6) δ 167.41, 159.91-162.31 (d, J=240 Hz), 156.89-159.28 (d,
J=238.7 Hz), 157.50, 142.18-142.21 (d, J=2.8Hz), 134.05-134.07 (d, J=2.5Hz), 127.96,
127.54,127.62, 118.25-118.33 (d, J=7.9Hz), 115.79, 115.74-115.96 (d, 22.5 Hz),
114.61-114.82 (d, J=20.9Hz), 71.16, 59.69, 59.50, 36.45, 24.60 ; IR (KBr, cm-1): 3263,
2963, 2912, 1716, 1614, 1592, 1404, 1220, 1157, 1066; ESI-MS m/z 408.38.
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
Purification of ezetimibe
Crude ezetimibe (100 g) having impurity-A 2.83% was dissolved in IPA (500 mL) to obtain
clear solution at 25-30oC. Water (500 mL) was added slowly to the reaction mass during which
solid was precipitated. The solid obtained was maintained for a period of 1 h, filtered and washed
with 1:1 aqueous IPA. Wet solid was further dried under vacuum to afford 94 g (94 %) of pure
ezetimibe as off-white to white anhydrous crystalline solid and found impurity-A not detected
(5R,6S)-1-(4-Fluorophenyl)-5-[(3S)-3-(4-fluorophenyl)-3-hydroxypropyl]-3-[(1S)-2-
hydroxy-1-phenylethyl]-6-(4-hydroxyphenyl)-1,3-diazinane-2,4-dione (Impurity-A)
Filtrate mother liquor obtained during the isolation of 6 was concentrated and extracted with
DCM. The DCM layer was distilled under vacuum and the residue was chromatographed on
silica gel and the pure fractions corresponding to impurity-A are collected and concentrated to
generate a analytical standard of impurity-A. Characterization data of Impurity-A: MR: 144.7-
1
145.0 °C; H NMR (400 MHz, DMSO-d6) δ 9.47 (s, -OH), 7.12-7.33 (m, 15H), 6.66-6.68
(d, J=8.4 Hz, 2H), 5.73-5.77 (t, J=7.2Hz, 1H), 5.37-5.38 (d, J=4.4 Hz, 1H), 5.10-5.13 (t,
J=5.2Hz, 1H), 4.79(s, 1H), 4.57-4.58 (m, 1H), 4.03-4.22 (m, 1H), 2.68-2.69 (m, 1H),
1.62-1.81(m, 1H); 13C NMR (100MHz, DMSO-d6) δ 171.42, 159.88-162.28(d, J=240
Hz), 159.07-161.49 (d, J=242 Hz), 156.99, 151.75, 142.06-142.09 (d, J=2.8 Hz),
ACS Paragon Plus Environment

Page 13 of 16
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
138.83, 138.17-138.20 (d, J=2.8Hz), 128.69, 128.91, 128.02, 127.37, 127.56, 126.94,
115.59-115.81 (d, J=22.5 Hz), 115.42, 114.56-114.77 (d, J=20.9 Hz), 71.26, 62.35,
60.68, 57.02, 49.98, 36.50, 27.46; IR (KBr, cm-1): 3391, 3069, 2949, 1713, 1659, 1603,
1435, 1435, 1220, 1155, 1045; ESI-MS m/z 573.28
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
CONCLUSION
A critical process impurity of ezetimibe formed at the last chemical stage is identified and
characterized. Structural characterization of impurity indicated 6-membered ring structure rather
than 9-membered ring structure proposed in prior-art. Detection method by HPLC was
developed as the impurity does not show specificity in the existing pharmacopoeia methods.
Careful selection of process parameters minimize the competitive side reaction which in
combination with improved isolation condition leads to complete purge of impurity to levels
much below the regulatory requirement.
ASSOCIATED CONTENT
Supporting information
Copies of IR, ESI-MS, 1H-NMR and 13C-NMR of compounds 2, 5, 1 and impurity impurity-A
AUTHOR INFORMATION
*Corresponding Authors
Sankareswaran Srimurugan
E-mail: Srimurugan@microlabs.in, Tel.: 08110-415647, ext. 274
Pramod Kumar
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 16
1
2
3
4
5
E-mail: pramodkumar@microlabs.in, Tel.: 08110-415647, ext. 245
6
7
8
Conflict of interest
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
The author declared no conflict of interest.
ACKNOWLEDGMENT
The authors are thankful to the management of Micro Labs Ltd., API Division Centre,
ML-27, Bangalore. We also wish to thank Mr. Vijay Kumar and Mr. Mohit Jain for their
valuable analytical support and cooperation.
REFERENCES:
1 International Conference On Harmonization Guidelines, Q3A (R2): Impurities In New
Drug Substances (Revised Guideline) 2006.
2 Warad, T.A., Bhusnure, O.G., Gholve S.B., “Impurity profile of pharmaceuticals
ingradient,” J. Pharm. Res. 2016, 10, 523.
3 (a) Qiu, Fenghe, Daniel L. Norwood., “Identification of pharmaceutical impurities,”
J. Liq. Chromatogr. Relat. Technol. 2007, 30, 877.
(b) Görög, Sándor., “Drug safety, drug quality, drug analysis,” J. Pharm. Biomed. Anal.
2008, 48, 247.
4 FDA Press Release, “FDA announces voluntary recall of several medicines containing
valsartan following detection of an impurity,” fda.gov, July 17, 2018.
ACS Paragon Plus Environment

Page 15 of 16
Organic Process Research & Development
1
2
3
4
5 (a) Bays, Harold., “Ezetimibe” Expert Opin. Investig. Drugs 2002, 11, 1587.
5
6
7
(b) Van Heek, M., H. Davis., “Pharmacology of ezetimibe,” Eur Heart. J Suppl. 2002, 4,
J5-J8.
8
9
6 (a) Gajjar, Anuradha K., Vishal D. Shah., “Isolation and structure elucidation of major
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
alkaline degradant of Ezetimibe,” J. Pharm. Biomed. Anal. 2011, 55, 225.
(b) Raman, Bhanu, Brajesh A. Sharma, Rahul Butala, Pradeep D. Ghugare, Ashok
Kumar., “Structural elucidation of a process-related impurity in ezetimibe by LC/MS/MS and
NMR,” J. Pharm. Biomed. Anal. 2010, 52, 73.
(c) Guntupalli, Srikanth, Uttam Kumar Ray, N. Murali, P. Badrinadh Gupta, Vundavilli
Jagadeesh Kumar, D. Satheesh, Aminul Islam., “Identification, isolation and characterization of
process related impurities in ezetimibe,” J. Pharm. Biomed. Anal. 2014, 88, 385.
(d) Ren, Yun, RenJun Li, Yong Deng, Mei Guan, Yong Wu, Li Hai., “First synthesis and
characterization of SRR/RSS-ezetimibe,” Tetrahedron Lett. 2013, 54, 6443.
(e) Ren, Yun, Yan-Jun Duan, Ren-Jun Li, Yong Deng, Li Hai, Yong Wu., “First
synthesis and characterization of key stereoisomers related to ezetimibe,” Chin. Chem. Lett.
2014, 25, 1157.
7 Thiruvengadam, T. K., Sudhakar, A. R., Wu, G. Process Chemistry in the
Pharmaceutical Industry, Vol.1, chapter 13, 221.
8 Thiruvengadam, T. K., Fu, X., Tann, C, H., McAllister, T. L., Chiu, J. S., Colon, C.,
“Process for the synthesis of azetidinones,” US6207822 B1, Mar.27, 2001
9 (a) Hesk, D., G. Bignan, J. Lee, J. Yang, K. Voronin, C. Magatti, P. McNamara.,
3
“Synthesis of H, 14C and 13C6 labelled Sch 58235,” J. Labelled comp. Radiopharm. 2002, 45,
145.
(b) Yong, W., Lizhong, L., Zhiqiang, S., Jianqiang, Y., Jin, W., Heyun, Y., “Molecular
sieve catalytic synthesis method of ezetimibe intermediate,” CN105622661, Jun. 1, 2016
(c) Baoquin, W., Lizhong, L., Zhiqiang, S., Jianqiang, Y., Jin, W., Heyun, Y., “Slurry bed
continuous production method of ezetimibe intermediate,” CN105566374 , May.11, 2016
10 (a) Smith, Edgar D., “Gas Chromatographic Analysis of Bis (Trimethylsilyl)
Acetamide,” J. Chromatogr. Sci. 1972, 10, 34.
(b) Shinohara, T., Kudo, M., Ueno, S., Maruyama, M., N,O-Bis(Trimethylsilyl)
Acetamide stabilization,” US5264601, Nov.23, 1993.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 16
1
2
3
4
5
6
11 Please refer supporting information for detailed assignment and spectral
characterization of Impurity-A.
7
8
9
GRAPHICAL ABSTRACT
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
F
OH
OH
NH
O
(S)
(R)
(S)
(S)
(R)
(S)
Cyclization/Deprotection
N
N
F
O
O
R = Silyl, Benzyl protecting groups
RO
F
O
(S)
Ezetimibe
OR
F
F
OH
OH
O
OH
(S)
(S)
O
N
(S)
(R)
(S)
NH
(S)
HO
F
N
O
(R)
(S)
HO
N
O
O
F
F
Impurity-Reported structure
Impurity- Reassigned structure
ACS Paragon Plus Environment