Identification and Control of Critical Process Impurities: An Improved Process for the Preparation of Dolutegravir Sodium

Article abstract of DOI:10.1021/acs.oprd.6b00156

A four-stage manufacturing route for the preparation of dolutegravir sodium (1) was assessed and optimized leading to a higher yielding, simpler and scalable process. Key improvements in the process include the development of mild workup procedure by selective derivatization of a difficult to remove a process impurity using tert-butyldimethylsilyl chloride. Metal-based hydrogenation-free O-debenzylation is optimized, and the critical isomeric impurity formed was identified and eliminated from the process by the establishment of a proper control strategy.

Full text of DOI:10.1021/acs.oprd.6b00156

Subscriber access provided by RMIT University Library
Full Paper
Identification and control of critical process impurities: An
improved process for the preparation of dolutegravir sodium
Srimurugan Sankareswaran, Madhavarao Mannam, veerababu
Chakka, Srirami Reddy Mandapati, and Pramod Kumar
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00156 • Publication Date (Web): 18 Jul 2016
Downloaded from http://pubs.acs.org on July 19, 2016
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 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Identification and control of critical process
impurities: An improved process for the
preparation of dolutegravir sodium
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
Srimurugan Sankareswaran, Madhavarao Mannam, Veerababu Chakka, Srirami Reddy
Mandapati, Pramod Kumar*
Micro Labs Ltd., Chemical Research Department, API R&D Centre, BommasandraꢀJigini Link
Road, KIADB INDL Area, Bommasandra, Bangalore 560105, Karnataka, India
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 26
1
2
3
4
TABLE OF CONTENTS GRAPHIC
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
2
ACS Paragon Plus Environment

Page 3 of 26
Organic Process Research & Development
1
2
3
4
ABSTRACT
5
6
7
8
9
A fourꢀstage manufacturing route for the preparation of dolutegravir sodium (1) was assessed
and optimized leading to a higher yielding, simpler and scalable process. Key improvements in
the process include the development of mild workꢀup procedure by selective derivatization of
difficult to remove process impurity using tertꢀbutyldimethylsilyl chloride. Metal based
hydrogenation free Oꢀdebenzylation is optimized and the critical isomeric impurity formed was
identified and eliminated from the process by establishment of proper control strategy.
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
KEY WORDS: Dolutegravir sodium, Oꢀdebenzylation, Silylation, Process impurities.
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 26
1
2
3
4
INTRODUCTION
5
6
7
8
9
Inhibition of HIV integrase, the enzyme that mediates the integration of the viral DNA into the
host cell genome, can halt the further spread of HIV and represents a viable treatment for HIV
infection.1 In this context, integrase strand transfer inhibitors (INSTi’s) have become an
attractive drug target over the past few years.2 Dolutegravir 1 alongside raltegravir and
elvitegravir (Figure 1) are metalꢀbinding INSTi’s approved for the treatment of antiretroviral
therapy.3 Dolutegravir is particularly advantageous due to low 50 mg dosing given once daily
without pharmacokinetic boosting and it’s display of antiviral activity in multiple cellꢀtypes and
cellꢀbased assay formats.4 Synthesis of dolutegravir is well documented in the literature starting
from routes employed during initial structureꢀactivity relationship findings of a series of
carbamoyl pyridone heterocycles to improved process for scaleꢀup.5 The synthetic route based on
a densely functionalized pyridinone core as starting material for constructing dolutegravir is by
far the most impressive synthetic route available compared to tedious route based on inexpensive
maltol.6 Though numerous methods of constructing substituted pyranone and further converting
to dolutegravir is reported, an improved process with detailed impurity profiling and subsequent
control strategy is not yet reported. Herein we report identification of critical process impurities
observed during the laboratory optimization of dolutegravir sodium. An improved process with
appropriate control strategies to reduce the level of these impurities in the final API is described.
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
F
ONa O
N
F
HO
HOOC
HO
Cl
F
F
O
N
N
N
N
H
H
H
N
N
N
N
O
N
O
O
O
O
O
Dolutegravir Sodium
Raltegravir
Elvitegravir
Figure 1 Metalꢀbinding integrase strand transfer inhibitors for antiretroviral therapy
4
ACS Paragon Plus Environment

Page 5 of 26
Organic Process Research & Development
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
9
The synthetic route taken up for lab optimization is adopted from innovator’s patent application
where different schemes are disclosed and demonstrated on milligram scale.7 In a subsequent
priorꢀart, one of these routes is demonstrated at higher scales (Scheme 1) and appears to be
scalable.8 The key feature of this route was selective hydrolysis of diester raw material B to a
halfꢀester C which allows for easier isolation and purification by salt formation with the
carboxylic group in the molecule. The phenolic moiety of the intermediates (B, C and D) was
masked as methyl ether which was liberated to the free phenol at the last stage by the action of
Lewis acids. The alternate easily removable benzyl protection of phenolic group is not opted due
to a possible defluorination sideꢀreaction during deprotection under catalytic hydrogenation. We
observed that this selective hydrolysis route did however did not appear to be reproducible and
highꢀyielding in our hands (TableꢀS1, supporting information). Therefore alternatively reported
monoꢀamidation of diester using 2,4ꢀdifluorobenzylamine was explored and identified to be
comparatively easier, reproducible and scalable. The finalized route of synthesis starting from
benzyl protected pyranone, dimethyl 3ꢀ(benzyloxy)ꢀ4ꢀoxoꢀ4Hꢀpyranꢀ2,5ꢀdicarboxylate 2 and
based on regioselective mono amidation is shown in scheme 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 26
1
2
3
4
5
6
7
8
O
N
O
O
HOOC
OMe
O
ROOC
MeO
OMe
HOOC
MeO
OMe
(i)
(iii)
(ii)
N
COOR
N
COOR
N
9
O
OMe
OMe
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
C
D
R = Me (A) or Et (B)
F
O
O
F
O
O
N
OMe
O
OH
N
(iv)
N
H
H
O
F
N
F
N
N
O
O
E
1
Scheme 1 Reported route for the synthesis of Dolutegravir sodium: (i) LiOH; (ii) (a) MsOH,
AcOH; (b) 3Rꢀaminobutanꢀ1ꢀol; (iii) CDI, 2,4ꢀdifluorobenzylamine; (iv) LiCl or MgCl2
Accordingly, the starting material pyran 2 was ringꢀopened with aminoacetaldehyde dimethyl
acetal and then cyclized back forming pyridinone 3 which reacted selectively with 2,4ꢀ
difluorobenzylamine to form amidoester 4. Acetal hydrolysis followed by reaction with 3Rꢀ
aminobutanꢀ1ꢀol in the presence of acid led to chiral hemiaminal 4b that undergoes critical
substrate controlled diastereoselective cyclization9 to form benzyl protected dolutegravir
intermediate 5. OꢀDebenzylation of 5 under acidic conditions followed by salification using
aqueous sodium hydroxide resulted in dolutegravir sodium.
6
ACS Paragon Plus Environment

Page 7 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
O
F
O
O
O
MeOOC
MeO
OBn
OBn
MeOOC
OBn
(i)
(ii)
N
H
N
COOMe
F
N
COOMe
O
COOMe
MeO
2
4
OMe
3
OMe
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
O
O
N
F
O
O
O
OBn
OBn
COOMe
(v)
(iii)
N
(iv)
N
H
H
O
F
F
N
H
N
OMe
4a
O
4b
F
O
O
N
F
O
O
N
OBn
O
OR
N
(vi)
N
H
H
O
F
F
N
N
O
5
O
R = OH; 6
R = ONa; 1
Scheme 2 Optimized route of synthesis of Dolutegravir sodium: (i) Aminoacetaldehyde dimethyl
acetal, DIPEA, MeOH, 25oC, 85%; (ii) Toluene, AcOH, 2,4ꢀdifluorobenzylamine, 90oC, 85%;
(iii) MsOH, AcOH, ACN, 65oC; (iv) 3Rꢀaminobutanꢀ1ꢀol; 25oC; (v) (a) 65oC; (b) MeOH, 25oC,
78%; (vi) (a) TFA, DCM, 35oC; (b) Toluene, 80oC; 90% (c) aq. NaOH, MeOH, 65oC, 95%.
The yields of all the stages were low to moderate during the initial optimization either due to
lower conversion or due to formation of more impurities. The first stage namely ring opening of
2 with aminoacetaldehyde dimethyl acetal proceeded instantaneously to form a mixture of
acyclic intermediates,10 which cyclised to 3 in varying yields and purity depending upon the
conditions employed. A preliminary screening of solvents (ester, chlorinated, hydrocarbon and
protic solvents) pointed out that the reaction was efficient in alcoholic solvent, especially
methanol (TableꢀS2, supporting information). When performed in methanol at elevated
temperatures the reaction was faster but generated more impurities and those performed at room
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 26
1
2
3
4
5
6
7
8
9
temperature did not go to completion. The use of sodium sulfate (for trapping water) and acid
catalysts showed marginal improvement in yield and purity. Surprisingly addition of base to the
reaction mass conveniently improved the yield and minimized the formation of impurities during
the cyclization of acyclic intermediates. Due to the homogeneous nature of reaction mass and
better control of water content of bases employed, organic bases, preferably
diisopropylethylamine was employed instead of inorganic bases (sodium bicarbonate) in the final
optimized process. Imp-1 and imp-2 (Figure 2) are major impurities formed in stageꢀ1 at
varying levels under different conditions screened (TableꢀS3, supporting information). Under the
optimized conditions, these nonꢀcritical impurities are wellꢀcontrolled and intermediate 3 was
formed in very good yield. Additionally 3 could be conveniently crystallized from the reaction
mass after the regular workꢀup using toluene, xylene, methanol or isopropanol.
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
N
F
O
O
N
F
OBn
O
MeO
OBn
O
OBn
O
HO
N
N
H
OMe
H
N
F
MeO
O
MeO
O
O
OMe
OMe
F
Imp-1
Imp-2
Imp-3
Figure 2 Structures of process impurities formed in stageꢀ1 of the process
However in view of a higher and clean conversion of 3 (typically around 90ꢀ93% area by HPLC)
and to evade material loss due to isolation, an in situ conversion of 3 to 4 was investigated.
Accordingly the reaction mass after workꢀup was distilled off, exchanged with fresh solvent and
reacted with 2,4ꢀdifluorobenzylamine in presence of acetic acid. A preliminary screening of
solvents pointed out the necessity of high temperature for reaction completion and toluene was
identified as the solvent of choice for the transformation. The reaction completion was however
8
ACS Paragon Plus Environment

Page 9 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
observed to be always be inconsistent and required higher mole equivalents of 2,4ꢀ
difluorobenzylamine which in turn generated more of imp-3. Attempts including the use of
higher equivalents of acetic acid and portionꢀwise addition of acetic acid were not beneficial in
driving the reaction to completion. This bottleꢀneck was overcome by portionꢀwise dosing of
2,4ꢀdifluorobenzylamine towards the termination of the reaction. This modified process
terminated the reaction consistently at various scales employed. The next attempt was to improve
the purity of intermediate 4 by effective crystallization (TableꢀS4, supporting information). The
crude reaction mass was highly soluble in most of the solvents and did not crystallize out easily.
After several attempts, effective and practical crystallization of 4 was achieved in isopropyl
alcohol that enabled the removal of many impurities along with imp-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
The next stage involving the construction of the tricyclic ring system was investigated in detail
due to the presence of many in situ intermediates (4a and 4b) and formation of a second chiral
center. Therefore hydrolysis of acetal 4 to aldehyde 4a, its further reaction with 3Rꢀaminobutanꢀ
1ꢀol to form hemiaminal 4b and its diasteroselective cyclisation to benzyl dolutegravir 5 were
studied individually to afford control on chiral and process impurities.
Acid hydrolysis of 4 to aldehyde 4a was faster in aqueous acids and slower with anhydrous
acidic conditions (TableꢀS5, supporting information). The yield in most of the cases was limited
by competitive side reactions.5d In the presence of formic acid and sulphuric acid, hydrolysis was
faster but generated more impurities upon holdꢀup of reaction mass which may turn troublesome
during scaleꢀup. The use of formic acidꢀmethanesulfonic acid in place of sulphuric acid also
produced lower yields. Incomplete reaction and lower yield was observed when using excess of
methanesulphonic acid under anhydrous conditions. Hydrolysis in the presence of sulphonic
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 26
1
2
3
4
5
6
7
8
9
acids like pyridinium pꢀtoluenesulfonate (PPTS) and tosylic acid did not provide encouraging
results. Heteropoly acid like phosphomolybdic acid turned the reaction mass highly colored
without any yield improvement and impurity reduction. The use of a catalytic amount of
methanesulfonic acid in acetonitrile gave a comparatively cleaner reaction at lower temperature
(50ꢀ70oC) but did not go to completion. Raising the reaction temperature to reflux resulted in
completion of the reaction but generated more impurities with lower product yield. In most of the
cases, the major side reaction was identified to be formation of Oꢀdebenzylated hemiacetal or
lactol impurities (imp-4 and imp-5) caused by the internal cyclisation of the hemiacetal or
aldehyde hydrate moiety with the ester functionality. Fine tuning of the reaction parameters
(temperature of reaction mass and mole equivalents of methanesulfonic acid; Refer to the
experimental section) alongside portionꢀwise dosing of methanesulfonic acid afforded clean
conversion of 4 to 4a with minimum of imp-4 and imp-5 (typically less than 1.0%).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
F
O
O
N
F
O
O
N
OR
O
OR
OR
N
F
F
O
O
N
H
N
H
O
H
O
F
N
N
F
HO
O
O
OMe
R = Bn; Imp-6
R = H; Imp-7
OH
Imp-4
(R = H)
Imp-5
(R = H)
Figure 3 Structures of process impurities formed in stageꢀ2 of the process
Isolation of intermediate 4a was not attempted due to unstable nature of the aldehyde
functionality and the accomplishment of a cleaner hydrolysis prompted for in situ reaction with
3Rꢀaminobutanꢀ1ꢀol at ambient temperature. Slow addition of 3Rꢀaminobutanꢀ1ꢀol formed an
imine which converted to hemiaminal 4b instantaneously. Further cyclisation of hemiaminal 4b
with the ester functionality to generate tricyclic core bearing second chiral center of benzyl
dolutegravir 5 was studied by varying the temperature of cyclisation, the nature of reaction
10
ACS Paragon Plus Environment

Page 11 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
conditions (by externally adding acid or base), the water content of reaction mass and the
addition of catalyst (TableꢀS6, supporting information). Optimized condition (refer to
experimental section) afforded 5 containing around 0.8ꢀ1.0% of the unwanted diastreomer.
Further optimization to control diastereomer during the reaction was not explored since this
impurity was removed easily below the desired ICH (The International Council for
Harmonisation of Technical Requirements for Pharmaceuticals for Human Use) limit in the
downstream process. The main process impurity of concern that formed in all of the attempted
experiments was imp-6. For a structurally similar congener, competitive dealkylation of the ether
moiety in ring under acidic condition was reported as main source of this impurity. Assuming
that an acidic nature and temperature of the reaction mass to trigger this sideꢀreaction, different
experimental parameters were explored like neutralizing reaction mass (containing
methanesulfonic acid) with various bases before further reaction with 3Rꢀaminobutanꢀ1ꢀol,
reverse mode addition viz reaction mass added dropwise to 3Rꢀaminobutanꢀ1ꢀol, isolation of in
situ intermediates (4a and 4b) before further reaction and modifying the temperature of reaction
mass. Unfortunately none of the trials controlled the formation of imp-6. The second obvious
strategy opted was purification of the isolated intermediate 5 in various solvent medium (single
and binary combination). Only a marginal reduction of impurity was observed irrespective of
purifications employed. Most importantly, the removal of this impurity was identified as being
critical as it reacted in subsequent stages and the corresponding debenzylated impurity (imp-7)
was carried forward to active pharmaceutical ingredient (API) at same levels without any further
purging. Similarly purification attempts at the API and intermediate stages did not assist in
removing imp-7. After failing with regular methods of removing imp-6, control strategy based
on chemoselective derivatization of the imp-6 was investigated. Imp-6 has a distinct primary
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
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 26
1
2
3
4
5
6
7
8
9
alcohol group which can selectively be functionalised without disturbing the main molecule. The
main challenge was to identify a derivatization method that should be simple enough to
implement on production scale and most importantly the resulting derivative of imp-6 should be
effectively removed from the system by the existing downstream process. Commercial acid
chlorides like benzoyl chloride, acetyl chloride and pivaloyl chloride readily formed the
corresponding ester derivative of imp-6; but further removal did not happen as expected.
Therefore nicotinyl chloride was attempted as a substitute on the basis of assumption that the
nicotinyl ester of imp-6 can be effectively removed by simple acid base workꢀup. Unfortunately
the derivatization reaction was not successful and complete derivatization of the impurity was
found to be highly inconsistent. Finally the more facile silylation of the primary alcohol was
identified as a derivatization strategy. Due to lesser stability of trimethylsilyl chloride,
commercially available tertꢀbutyldimethylsilyl chloride (TBDMSCl) was selected as a
appropriate reagent for selective reaction. The reagent system as expected, demonstrated
successful derivatization followed by complete removal of the impurity formed even at higher
levels.
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
OBn O
N
OBn O
N
F
F
O
O
F
F
O
O
a (or) b
N
N
H
H
N
N
RO
HO
Derivatives of Imp-6
Imp-6
R = -COPh (or) -COCH3 (or)-COC(Me)3
(or) 3-Nicotinoyl (or)TBDMS (imp-6a)
Scheme 3 Attempted derivatization of imp-6 in stageꢀ2: (a) BzCl or AcCl or PivCl or 3ꢀ
Nicotinoyl chloride, TEA; (b) Imidazole, TBDMSCl
The derivatization step was implemented as a part of the workup procedure and integrated with
the main process. Accordingly, the reaction mass was distilled off after the completion of the
12
ACS Paragon Plus Environment

Page 13 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
reaction, taken up in dichloromethaneꢀwater to remove reagents and byꢀproducts and treated with
imidazole and TBDMSCl at ambient temperature for the selective derivatization of imp-6 to
imp-6a. The derivatization was found to be instantaneous and after a second aqueous workꢀup,
the crude reaction mass was crystallized in methanol to afford pure 5 that is free of imp-6a. The
use of other common bases (triethylamine, diisopropylethylamine and potassium carbonate) was
not as effective as imidazole for the derivatization reaction. The identified strategy permitted for
an easy scalable process demonstrating strong impurity control and overall yield of 70%
(conversion of 4 to 5).
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 final chemical stage of the process is the Oꢀdebenzylation of 5 to form the free acid of
dolutegravir. Most of the literature methods use the conventional catalytic hydrogenation in the
presence of Pd/C. Two main drawbacks in implementing catalytic hydrogenation was the poor
solubility of the starting material and product in most of the common solvents employed for
hydrogenation and the potential aromatic defluorination sideꢀreactions generating isomeric
desfluoro impurities that are extremely difficult to remove by purification. Different acidic and
basic reagents were alternatively screened for establishing simple and clean conversion (Tableꢀ
S7, supporting information). Brønsted acids like methanesulphonic acid, tosylic acid and
trifluoroacetic acid (neat) completed the debenzylation reaction but generated more of imp-7
which was difficult to remove by purification. No conversion was observed with mild acid like
PPTS. Lewis acids like LiBr gave quick reaction but formed bromide impurities (by LCMS) as
byꢀproducts. LiCl and MgCl2 reactions were slower and did not go to completion. The reaction
completed in the presence of sodium hydride as reagent and formed dolutegravir sodium directly
but generated more of desfluoro impurity and imp-7. After screening the aboveꢀsaid reagents at
different concentrations, it was found that 5.0 mole equivalent of trifluoroacetic acid (TFA) in
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 26
1
2
3
4
5
6
7
8
9
dichloromethane gave clean conversion (Scheme 4). Attempts to reduce mole equivalents of
TFA again gave incomplete conversion. Reactions at various scales were consistent in terms of
conversion, yield and impurityꢀprofile using 5.0 mole equivalents of TFA (TableꢀS8, supporting
information). Similarly no rise in olefin impurity (imp-7) was observed under these conditions as
against observed using neat TFA. After the regular acid base workup, the byꢀproduct of the
reaction, benzyl trifluoroacetate (7) was removed by recrystallization in toluene and the pure
dolutegravir was isolated in 86% yield.
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
OBn O
F
F
O
O
OCOCF3
F
F
O
O
N
(i)
+
N
H
H
N
N
N
N
O
O
7
6
5
+
Unknown impurity
m/z 509
Scheme 4 OꢀDebenzylation of 5 to 6: (i) (a) TFA, DCM; (b) Toluene.
Though all the known impurities are controlled during the debenzylation stage, one unknown
impurity with m/z 509 (imp-8) matching the molecular weight of benzyl dolutegravir 5 was
observed at a level of 0.06ꢀ0.08% during the purity analysis by HPLC. The impurity formed a
sodium salt and was carried forward to the final API at the same level. The identification of this
unknown impurity therefore became critical to understand the origin and its appropriate control.
The formation of the sodium salt by the impurity indicated a structure similar to API, however a
mass similar to benzyl dolutegravir with a different HPLC retention time indicate that an
additional benzyl group is attached at some part of the API molecule. Initially a rearrangement of
5 was assumed involving a [3,3]ꢀsigmatropic rearrangement of Oꢀbenzyl group to form 8 in the
presence of Bronsted acid, TFA. Though the rearrangement product has identical mass, it does
not have any phenolic group to form a sodium salt. Also the impurity was not formed
14
ACS Paragon Plus Environment

Page 15 of 26
Organic Process Research & Development
1
2
significantly under conditions like Lewis acid (AlCl3, BF3.OEt2) and thermal reaction (250oC)
that are favorable to [3,3]ꢀsigmatropic rearrangement. A pure sample of the impurity isolated
using preparatory HPLC followed by structural characterization revealed that no sigmatropic
rearrangement occurred and indeed the benzyl group was attached to 2,4ꢀdifluorobenzyl moiety
(Figure 4).
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
O
O
F
F
O
O
N
H
[3,3]
N
N
O
O
O
8
F
F
O
O
m/z 509
N
H
N
N
O
OH
O
F
F
O
O
5
N
H
m/z 509.50
N
N
O
Imp-8
m/z 509
Figure 4 Most probable structure of process impurity imp-8
The presence of an additional benzyl group at the aromatic region is a suggestive of mechanism
of imp-8 formation involving electrophilic substitution (FriedelꢀCrafts like alkylation) at the
aromatic part of dolutegravir 6 with byꢀproduct, benzyltrifluoroacetate 7 in the presence of
trifluoroacetic acid. The regioselectivity of electrophilic substitution during the impurity
formation is supported by facile acylation reported for 2,4ꢀdifluorotoluene in the presence of
Lewis acid.11 If the proposed mechanism of impurity formation is valid, suitable process
conditions can be identified to control the formation of imp-8. Since the debenzylation reaction
was performed in dichloromethane, the use of a sacrificial component having a higher affinity to
electrophilic substitution can suppress this side reaction. Accordingly, addition of 10% v/v of
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 26
1
2
3
4
5
6
7
8
9
toluene to the original debenzylation reaction conditions, as expected reduced the level of impꢀ8
was from 0.08% to 0.02% and a separate reaction performed exclusively in toluene did not form
any imp-8 supporting the mechanism of origin of the impurity. The identified solvent system
provided excellent reaction condition for an impurityꢀfree debenzylation. Treatment of 6 with
aqueous sodium hydroxide at reflux temperature afforded highꢀpure dolutegravir sodium 1. The
overall yield of the optimized process was 57%.
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 robust and scalable process for the synthesis of dolutegravir sodium was developed. All
unknown impurities detected during the development of dolutegravir sodium was either
synthesized or isolated via preparative HPLC and identified by NMR and MS. The structural
knowledge of these impurities led to the identification of their root cause of formation and
allowed for establishing a suitable control strategy that can produce high purity API in a
consistent way. The improved process demonstrates consistent control of chiral impurities
(diastereomeric impurity and enantiomeric impurity) to a level of not more than 0.15% w/w in
the final API.
The optimized process was successfully demonstrated and validated at pilot plant scale
matching the proposed lab yield and quality at all the stages (TableꢀS9, supporting information).
EXPERIMENTAL SECTION
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 Mꢀ565 and are uncorrected. IR spectra of samples were
16
ACS Paragon Plus Environment

Page 17 of 26
Organic Process Research & Development
1
2
1
3
4
recorded on Shimadzu IR AffinityꢀI FTꢀIR spectrophotometer. H NMR spectra (400 MHz) and
5
6
7
8
13CNMR spectra (100 MHz) were recorded in CDCl3, DMSOꢀd6, on a Bruker 400 MHz NMR
(ASCEND, 5 mm PABBO) instrument and mass spectra were determined on Velos Pro Ion Trap
Mass Spectrophotometer (Thermo Scientific)
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
Methyl 5-(2,4-difluorobenzylcarbamoyl)-3-(benzyloxy)-1,4-dihydro-1-(2,2-dimethoxyethyl)-
4-oxopyridine-2-carboxylate (4) Dimethyl 3ꢀ(benzyloxy)ꢀ4ꢀoxoꢀ4Hꢀpyranꢀ2,5ꢀdicarboxylate 2
(150 g, 0.471 mol) was suspended in methanol (1500 mL) in a 3ꢀneck RB flask at room
temperature. Aminoacetaldehyde dimethyl acetal (54.5 g, 0.518 mol) was added dropꢀwise
followed by addition of diisopropylethylamine (60.9 g, 0.471 mol) and the resulting clear
solution was stirred untill the consumption of starting material was observed. After the
completion of the reaction (16ꢀ18 h), the pH of reaction mass was adjusted to around 6.0ꢀ7.0
using aqueous citric acid (50% solution, 300 mL) and solvent was distilled off completely under
vacuum. The residue was taken up in dichloromethane (1200 mL) and water (300 mL), separated
organic layer and distilled off the dichloromethane layer to afford product as crude solid.
Toluene (1350 mL) and 2,4ꢀdifluorobenzylamine (67.45 g, 0.471) was added to the crude
reaction mass at room temperature. Acetic acid (11.32 g, 0.189 mol) was then added dropꢀwise
and the reaction mass was heated to 90ꢀ95oC, two more portions of 2,4ꢀdifluorobenzylamine
(16.86 g, 0.118 mol) after every 3.0 h. After the consumption of the starting material, the
reaction mass was cooled to room temperature and washed the toluene layer sequentially with
aqueous citric acid (25% solution, 300 mL) and water (300 mL). The toluene layer was separated
and distilled under reduced pressure to give product 4 as a transparent viscous oil. Isopropyl
alcohol (300 mL) was added and heated the reaction mass to 55ꢀ60oC. Cooled the reaction mass
was cooled to 0ꢀ5oC gradually and maintained for 2.0 h, filtered and dried under vacuum to give
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 26
1
2
210 g (86%) of 4 as an offꢀwhite solid. Characterization data of 3: 1H NMR (400 MHz, CDCl3) δ
8.1(s, 1H), 7.4 (m, 2H), 7.3 (m, 3H), 5.3 (s, 2H), 4.4ꢀ4.5 (t, J=4.8Hz, 1H), 3.9 (m, 5H), 3.8 (s,
3H), 3.3 (s, 6H); 13C NMR (400 MHz, CDCl3) δ 171.0, 165.5, 162.2, 149.2, 146.1,136.9, 133.9,
128.7, 128.2, 128.1, 118.1, 102.7, 74.1, 56.7, 55.7, 53.0, 52.3; IR (KBr, cmꢀ1): 3041, 2999, 2954,
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
1
2835, 1722, 1616, 1602. ESIꢀMS m/z 406.17; Characterization data of 4: H NMR (400 MHz,
CDCl3) δ 10.43ꢀ10.45 ( t, J=5.6Hz, 1H), 8.44(s, 1H), 7.33ꢀ7.42 (m, 6H), 6.83ꢀ6.87 (m, 2H),
4.66ꢀ4.67 (d, J=6.0Hz, 2H), 4.47ꢀ4.50 (t, J=4.8Hz, 1H), 4.03ꢀ4.04 (d, J=4.8Hz, 2H), 3.81 (s,
1H), 3.38 (s, 6H); 13C NMR (400 MHz, CDCl3) δ 173.2, 164.1, 162.1, 160.9, 161.0, 163ꢀ3, 163ꢀ
5, 159.5, 159.6, 162.0, 162.1, 148.3, 144.5, 136.7, 135.0, 130.6, 130.8, 128.5, 128.3, 128.2,
121.4ꢀ121.6, 119.3, 111.0, 111.2, 103.5, 104.0, 102.6, 74.4, 56.9, 55.6, 53.1, 36.4, 36.5; IR (KBr,
cmꢀ1): 3172, 3039, 2956, 2875, 2841, 1734, 1658, 1608, 1546, 1132; ESIꢀMS m/z 517.34.
(4R,12aS)-N-(2,4-Difluorobenzyl)-7-benzyloxy-4-methyl-6,8-dioxo-3,4,6,8,12,12a-hexahydro
-2H-pyrido[1',2':4,5]pyrazino[2,1-b][1,3]oxazine-9-carboxamide (5) A solution of 4 (210 g,
0.406 mol) in acetonitrile (1470 mL) was treated with acetic acid (146.5 g, 2.440 mol) under
nitrogen atmosphere in a 3ꢀnecked RB flask. A solution of methanesulphonic acid (7.81 g, 0.081
mol) in acetonitrile (210 mL) was added slowly at ambient temperature and the reaction mass
heated to 60ꢀ65oC. A second lot of methanesulphonic acid (3.90 g, 0.041 mol) was added and
continued heating until the completion of the reaction. The reaction was cooled to room
temperature and treated with a solution of 3Rꢀaminobutanꢀ1ꢀol (50.74 g, 0.569 mol) in
acetonitrile (210 mL). Maintained the reaction mass for 2 h at same temperature and again heated
to 60ꢀ65oC until the consumption of 4b. The reaction mass was completely distilled under
vacuum and the residue was takenꢀup in dichloromethane (1680 mL). The organic layer was
washed sequentially with 10% sodium carbonate solution (1050 mL) and water (840 mL). The
18
ACS Paragon Plus Environment

Page 19 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
dichloromethane layer was separated, dried over sodium sulphate and treated with TBDMSꢀCl
(12.26 g, 0.081 mol) and imidazole (11.07 g, 0.163 mol) in a fresh RB flask under nitrogen
atmosphere. After stirring for 2 h, 15% sodium chloride solution (840 mL) was added to the
reaction mass and the layers are separated. The organic layer was taken in a fresh RB flask and
distilled out completely under reduced pressure. Methanol (420 mL) was added to the
concentrated residue and heated the reaction mass to 55ꢀ60oC for 1.0 h. The solid formed was
cooled gradually to 5ꢀ10oC, filtered and dried under vacuum to afford 162.0 g (78%) of 5 as
paleꢀyellow solid. Characterization data of 5: 1H NMR (400 MHz, CDCl3) δ 10.4 (t, J=6Hz, 1H),
8.3 (s, 1H), 7.6 (d, J=7.2Hz, 2H), 7.2ꢀ7.4 (m, 4H), 6.8 (m, 2H), 5.2ꢀ5.3 (dd, J=16.4Hz and 10Hz,
2H), 5.1 (m, 1H), 4.9ꢀ5.0 (m, 1H), 4.6 (d, J=6Hz, 2H), 4.2 (dd, J=13.2Hz and 3.2Hz, 1H), 4.0ꢀ
4.1 (dd, J=13.2Hz and J=6Hz, 1H), 3.9 (m, 2H), 2.1ꢀ2.2 (m, 1H), 1.4ꢀ1.5 (dd, J=14Hz and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
J=2Hz, 1H), 1.3 (d, J=6.8Hz, 3H); C NMR (400 MHz, CDCl3) δ 174.5, 164.0, 160.9, 161.0,
163.4, 163.5, 159.4, 159.5, 161.9, 162.0, 155.5, 153.2, 142.0, 136.6, 130.5, 130.6, 129.4, 128.9,
128.2, 128.1, 121.3, 121.5, 118.7, 111.0, 111.2, 103.5, 104.4, 76.0, 74.4, 62.5, 53.5, 44.5, 36.4,
36.5, 29.3, 15.9; IR (KBr, cmꢀ1): 3180, 3061, 2974, 2941, 2870, 1656, 1606, 1546, 1093; ESIꢀ
MS m/z 510.17
(4R,12aS)-N-(2,4-Difluorobenzyl)-7-hydroxy-4-methyl-6,8-dioxo-3,4,6,8,12,12a-hexahydro-
2H-pyrido[1',2':4,5]pyrazino[2,1-b][1,3]oxazine-9-carboxamide (6) A solution of 5 (135 g,
0.265 mol) in dichloromethane (2430 mL) was heated to reflux in a 3ꢀnecked RB flask. A
solution of trifluoroacetic acid (151 g, 1.325 mol) in dichloromethane (270 mL) was slowly
added to the reaction mass at reflux over a period of 1.0ꢀ2.0 h and continued stirring until the
completion of the reaction. The reaction mass was cooled to 10ꢀ15oC, carefully treated with
water (540 mL) and pH (~7.0ꢀ9.0) was adjusted using aq. ammonia (114 mL). The reaction mass
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 26
1
2
3
4
5
6
7
8
9
was warmed to room temperature, the layers were separated and washed sequentially with water
(270 mL), 10% citric acid solution (270 mL) and brine (540 mL). The dichloromethane layer
was taken in a fresh RB flask and distilled out completely under reduced pressure. Toluene (405
mL) was added to the concentrated residue and heated the reaction mass to 60ꢀ65oC for 1.0 h.
The solid formed was cooled gradually to 10ꢀ15oC, filtered, washed with toluene (67.5 mL) and
dried under vacuum to afford 100 g (90%) of 6 as offꢀwhite to paleꢀyellow solid.
Characterization data of 6: 1H NMR (400 MHz, CDCl3) δ 12.4 (s, 1H), 10.3 (t, J=5.6Hz), 8.3 (s,
1H), 7.3 (m, 1H), 6.7ꢀ6.8 (m, 2H), 5.2 (m, 1H), 4.9ꢀ5.0 (m, 1H), 4.6 (d, J=6.0Hz, 2H), 4.2ꢀ4.3
(dd, J=13.2Hz and 4Hz, 1H), 4.1 (dd, J=13.2Hz and 6Hz), 4.0 (m, 2H), 2.1ꢀ2.2 (m, 1H), 1.5 (dd,
J=14Hz and 2.0Hz, 1H), 1.3ꢀ1.4 (d, J=7.2Hz, 3H); 13C NMR (400 MHz, CDCl3) δ 171.3, 164.1,
162.5, 160.9ꢀ161.0, 163.3ꢀ163.4, 159.4, 159.5, 161.9, 162.0, 156.0, 140.1, 130.2, 130.4, 121.4,
121.6, 116.6, 115.8, 111.0, 111.2, 103.4, 103.9, 76.3, 62.7, 52.4, 44.7, 36.5, 36.5, 29.2, 15.5; IR
(KBr, cmꢀ1): 3442, 3190, 3068, 2974, 2949, 2885, 1660,1629, 1544, 1087; ESIꢀMS m/z 420.17
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
Sodium(4R,12aS)-9-{[(2,4-difluorophenyl)methyl]carbamoyl}-4-methyl-6,8-dioxo-
3,4,6,8,12,12a-hexahydro-2H-pyrido[1`,2`:4,5]pyrazino[2,1-b][1,3]oxazol-7-olate (1) A slurry
of 6 (85 g, 0.203 mol) in methanol (765 mL) was heated to reflux in a 3ꢀnecked RB flask. A
solution of sodium hydroxide (8.5 g, 0.213 mol) in water (25.5 mL) was slowly added to the
reaction mass at reflux over a period of 30ꢀ45 min and further continued stirring for a period of
2.0 h. The solid formed was cooled gradually to 25ꢀ30oC, filtered, washed with methanol (170
mL) and dried under vacuum to afford 85.0 g (95%) of 1 as offꢀwhite to paleꢀyellow solid.
1
Characterization data of 1: H NMR (400 MHz, DMSOꢀd6 ) δ 10.6ꢀ10.7 (t, J=6.0Hz, 1H), 7.8
(s, 1H), 7.3 (dd, J=8.4Hz and 7.2Hz, 1H), 7.1ꢀ7.2 (m, 1H), 7.0 (t, J=8.0Hz, 1H), 5.1 (bs, 1H),
4.7ꢀ4.8 (m, 1H), 4.5 (d, J=5.6Hz, 2H), 4.2ꢀ4.3 (d, J=11.2Hz, 1H), 4.1 (m, 1H), 3.9 (m, 1H), 3.7ꢀ
20
ACS Paragon Plus Environment

Page 21 of 26
Organic Process Research & Development
1
2
13
3
3.8 (m, 1H), 1.8 (m, 1H), 1.3 (d, J=13.2Hz, 1H), 1.2 (d, J=6.8Hz, 3H); C NMR (400 MHz,
DMSOꢀd6) δ 177.9, 167.0, 166.0, 161.0, 159.9, 160.0, 162.4, 162.5, 158.6, 158.8, 161.1, 161.2,
134.2, 130.4, 130.5, 122ꢀ8, 123.0, 114.8, 111.0, 111.3, 108.6, 103.3, 103.8, 75.5, 61.8, 53.1,
42.9, 35.3, 29.1, 15.3; IR (KBr, cmꢀ1): 3165, 3072, 2974, 2941, 2873, 1643, 1539, 1504, 1101;
ESIꢀMS m/z: 418.17
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
1-(2,2-Dimethoxyethyl)-5-(benzyloxy)-6-(methoxycarbonyl)-4-oxo-1,4-dihydropyridine-3-carboxylic
acid (imp-1); 1H NMR (400 MHz, CDCl3) δ 15.0 (s, 1H), 8.4 (s, 1H), 7.3ꢀ7.4 (m, 5H), 5.3 (s,
13
2H), 4.4 (t, J=4.8Hz, 1H), 4.0 (d, J=4.4Hz, 2H), 3.8 (s, 3H), 3.4 (s, 6H); C NMR (400 MHz,
CDCl3) δ 174.8, 165.8, 161.4, 147.4, 145.1, 136.4, 136.1, 128.7, 128.5, 128.4, 116.5, 102.2, 74.6,
57.4, 55.8, 53.4; IR (KBr, cmꢀ1): 3076, 3043, 2999, 2954, 2835, 1724, 1620; ESIꢀMS m/z 392.13.
Methyl
5-(2,2-dimethoxyethylcarbamoyl)-3-(benzyloxy)-1,4-dihydro-1-(2,2-dimethoxyethyl)-4-
oxopyridine-2-carboxylate (imp-2): 1H NMR (400 MHz, CDCl3) δ 10.1 (t, J=5.2Hz, 1H), 8.4 (s,
1H), 7.3ꢀ7.4 (m, 5H), 5.2 (s, 2H), 4.5(t, J=5.2Hz, 1H), 4.4 (t, J=4.4Hz, 1H), 4.0 (d, J=4.4hz, 2H),
3.7 (s, 3H), 3.6 (t, J=6.0Hz, 2H), 3.4 (s, 6H), 3.3 (s, 6H) ; 13C NMR (400 MHz, CDCl3) δ 173.2,
164.2, 162.1, 148.2, 144.4, 136.7, 134.8, 128.5, 128.3, 128.1, 119.3, 102.7, 102.4, 74.3, 56.8,
55.6, 53.9, 53.1, 40.8; IR (KBr, cmꢀ1): 3201, 3045, 2949, 2835, 1734, 1662, 1610, 1548; ESIꢀMS
m/z 479.26
N-(2,4-Difluorobenzyl)-2,8-dihydro-9-benzyloxy-2-(4-hydroxybutan-2-yl)-1,8-dioxo-1H-
1
pyrido[1,2-a]pyrazine-7-carboxamide (imp-6): H NMR (400 MHz, DMSOꢀd6 ) δ 10.6 (t,
J=5.6Hz, 1H), 8.9 (s, 1H), 7.6 (d, J=7.2Hz, 2H), 7.3ꢀ7.4 (m, 5H), 7.2 (m, 1H), 7.0ꢀ7.1 (m, 1H),
7.0 (d, J=6.4Hz, 1H), 5.0ꢀ5.1 (dd, J=12.8Hz and 10.4Hz), 4.9 (m, 1H), 4.57ꢀ4.59 (d, J=5.6Hz,
2H), 4.51ꢀ4.5 (t, J=4.8Hz,1H), 3.3ꢀ3.4 (dd, J=11.2Hz and 1.5Hz, 2H), 1.7ꢀ1.8 (m, 2H), 1.2 (d,
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 26
1
2
J=6.8Hz, 3H); 13C NMR (400 MHz, DMSOꢀd6) δ 171.8, 163.1, 160.2ꢀ160.3, 162.6, 162.7,
158.8, 158.9, 161.3, 161.4, 153.5,151.7, 138.2, 137.4, 130.9, 131.0, 129.8, 128.1, 127.9, 127.6,
122.0, 122.2, 119.7,115.3,111.6, 111.2, 111.5, 103.5, 104.0, 73.2, 57.6, 47.5, 36.8, 35.8, 35.8,
18.5; IR (KBr, cmꢀ1): 3419, 3192, 3059, 2933, 2881, 1660, 1604, 1543, 1095; ESIꢀMS m/z
510.31
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
N-(2,4-Difluorobenzyl)-2,8-dihydro-9-benzyloxy-2-(4-(tert-butyldimethylsilyloxy)butan-2-
yl)-1,8-dioxo-1H-pyrido[1,2-a]pyrazine-7-carboxamide (imp-6a): 1H NMR (400 MHz,
CDCl3) δ 10.6 (t, J=6.0Hz, 1H), 8.6 (s, 1H), 7.6 (d, J=7.2Hz, 2H), 7.2ꢀ7.3 (m, 4H), 6.7ꢀ6.8 (m,
2H), 6.7 (d, J=6.0Hz, 1H), 6.3 (d, J=6.4Hz, 1H), 5.2 (dd, J=15.6Hz and 10Hz), 5.0 (m, 1H), 4.6
13
(d, J=6.0Hz, 2H), 3.6 (t, J=6.0Hz, 2H), 1.8ꢀ1.9 (m, 2H), 1.3 (d, J=6.8Hz, 3H), 0.0 (s, 6H); C
NMR (400 MHz, CDCl3) δ 172.8, 163.8, 160.9ꢀ161.0, 163.3, 163.4, 159.3ꢀ159.5, 161.8, 161.9,
153.7, 152.1, 137.6, 136.7, 130.4, 130.6, 129.5, 128.9, 128.2, 128.0, 121.2, 121.4, 120.5, 115.2,
111.0, 111.2, 110.8, 103.4, 10.3ꢀ9, 74.3, 59.8, 48.6, 37.4, 36.5, 25.9, 19.3, 18.1, ꢀ5.5; IR (KBr,
cmꢀ1): 3192, 3059, 2953, 2856, 1664, 1604, 1543, 1095, 839; ESIꢀMS m/z 624.42
N-(2,4-Difluorobenzyl)-2,8-dihydro-9-hydroxy-2-(4-hydroxybutan-2-yl)-1,8-dioxo-1H-
1
pyrido[1,2-a]pyrazine-7-carboxamide (imp-7): H NMR (400 MHz, DMSOꢀd6) 12.2 (s, 1H),
10.5ꢀ10.6 (t, J=6.0Hz, 1H), 8.8 (s, 1H), 7.5 (d, J=6.4Hz, 1H), 7.3ꢀ7.4 (dd, J=15.6Hz and 6.8Hz,
1H), 7.2 (m, 1H), 7.05ꢀ7.09 (m, 1H), 7.01(d, J=6.4Hz, 1H), 4.8ꢀ4.9 (m, 1H), 4.5 (m, 3H), 3.3ꢀ3.4
(m, 2H), 1.7ꢀ1.9 (m, 2H), 1.2 (d, J=6.8Hz, 3H); 13C NMR (400 MHz, DMSOꢀd6) δ 168.1, 163.4,
160.5, 160.2ꢀ160.3, 162.6, 162.7, 158.8, 158.9, 161.3, 161.4, 153.0, 134.8, 130.7, 130.9, 122.0,
122.2, 117.6, 117.2, 114.7, 113.5, 111.2, 111.4, 103.5, 104.0, 57.4, 47.4, 36.5, 35.8, 18.3; IR
22
ACS Paragon Plus Environment

Page 23 of 26
Organic Process Research & Development
1
2
(KBr, cmꢀ1): 3450, 3236, 3101, 3049, 2941, 2885, 1658, 1624, 1558, 1504, 1099; ESIꢀMS m/z
3
4
5
6
420.16
7
8
9
(4R,12aS)-N-(5-Benzyl-2,4-difluorobenzyl)-7-hydroxy-4-methyl-6,8-dioxo-3,4,6,8,12,12a-
hexahydro-2H-pyrido[1',2':4,5]pyrazino[2,1-b][1,3]oxazine-9-carboxamide (imp-8): 1H
NMR (400 MHz, DMSOꢀd6) δ 12.3 (m, 1H), 10.2 (s, 1H), 8.2 (s, 1H), 7.0ꢀ7.3 (m, 6H), 6.6ꢀ6.7
(t, J=9.6Hz, 1H), 5.1 (bs, 1H), 4.8ꢀ4.9 (m, 1H), 4.5 (m, 2H), 3.5ꢀ4.0 (m, 6H), 2.0ꢀ2.1 (m, 1H), 1.4
(m, 1H), 1.3 (m, 1H); 13C NMR (400 MHz, DMSOꢀd6) δ 171.2, 164.0, 162.4, 158.7, 158.6,
161.2, 161.1, 160.6, 160.5, 158ꢀ1, 158.0, 155.9, 140.2, 139.5, 131.4, 128.5, 128.4, 126.1, 123.7,
123ꢀ8, 121.2, 121.3, 116.4, 115.7, 103.3, 103.8, 75.7, 62.6, 52.3, 44.6, 36.6, 34.4, 29.2, 15.4; IR
(KBr, cmꢀ1): 3464, 3242, 3059, 2972, 2926, 1658, 1633, 1541, 1504, 1091; ESIꢀMS m/z 508.16
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
ASSOCIATED CONTENT
SUPPORTING INFORMATION. Copies of IR, ESIꢀMS, 1HꢀNMR and 13CꢀNMR of compounds
3, 4, 5, 6, 1 and impurities imp-1, imp-2, imp-6, imp-6a, imp-7 and imp-8. LCMS of imp-8,
HPLC chromatogram of isolated intermediates 3 and 4 along with reaction monitoring for the
conversion of 2 to 3 in the presence and absence of water is provided. Results of optimization
studies for preparation of intermediates and API are also provided.
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 26
1
2
3
4
5
AUTHOR INFORMATION
6
7
8
*Corresponding Author
9
Pramod Kumar
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
Eꢀmail: pramodkumar@microlabs.in
Author Contributions
All authors contributed equally to this paper and have given approval to the final version of the
manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors greatly appreciate financial support for this work from Micro Labs Ltd., API
Division Centre, MLꢀ27, Bangalore. We thank our group colleges for their appreciated
contribution. We also wish to thank Mr. Ajay Thakur, and Dr. S.G. Hiriyanna for their valuable
analytical support and cooperation.
24
ACS Paragon Plus Environment

Page 25 of 26
Organic Process Research & Development
1
2
3
4
REFERENCES:
5
6
7
8
9
(1) For reviews on HIV integrase inhibitors, see: (a) Pommier, Y.; Johnson, A. A.; Marchand, C.
Nat. Rev. Drug Discovery 2005, 4, 236. (b) Savarino, A. Expert Opin. Invest. Drugs 2006, 15,
1507.
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) (a) Reinke, R.; Lee, D., Robinson, W. Antimicrob. Agents Chemother. 2002, 46, 3301. (b)
Hazuda, D.; Felock, P.; Witmer, M.; Wolfe, A.; Stillmock, K.; Grobler, J. Science 2000, 287,
646.
(3) (a) Pendri, A.; Meanwell, N. A.; Peese, K. M.; Walker, M. A. Expert Opin. Ther. Pat. 2011,
21, 1173. (b) Johns, B. A.; Svolto, A. C. Expert Opin. Ther. Pat. 2008, 18, 1225. (c) Kawasuji,
T.; Yoshinaga, T.; Sato, A.; Yodo, M.; Fujiwara, T.; Kiyama, R. Bioorg. Med. Chem. 2006, 14,
8430. (d) Kawasuji, T.; Fuji, M.; Yoshinaga, T.; Sato, A.; Fujiwara, T.; Kiyama, R. Bioorg. Med.
Chem. 2006, 14, 8420.
(4) (a) Lennox, J. L.; DeJesus, E.; Lazzarin, A.; Pollard, R. B.; Madruga, J. V. R.; Berger, D. S.;
Zhao, J.; Xu, X.; WilliamsꢀDiaz, A.; Rodgers, A. J.; Barnard, R. J. O.; Miller, M. D.; DiNubile,
M. J.; Nguyen, B.ꢀY.; Leavitt, R.; Sklar, P. Lancet 2009, 374, 796. (b) McCormack, P. L. Drugs
2014, 74, 1241. (c) Hare, S.; Smith, S. J.; Metifiot, M.; JaxaꢀChamiec, A.; Pommier, Y.; Hughes,
S. H.; Cherepanov, P. Mol. Pharmacol. 2011, 80, 565.
(5) (a) Kawasuji, T.; Johns, B. A.; Yoshida, H.; Taishi, T.; Taoda, Y.; Murai, H.; Kiyama, R.;
Fuji, M.; Yoshinaga, T.; Seki, T.; Kobayashi, M.; Sato, A.; Fujiwara, T. J. Med. Chem. 2012, 55,
8735. (b) Kawasuji, T.; Johns, B. A.; Yoshida, H.; Weatherhead, J. G.; Akiyama, T.; Taishi, T.;
Taoda, Y.; MikamiyamaꢀIwata, M.; Murai, H.; Kiyama, R.; Fuji, M.; Yoshinaga, T.; Seki, T.;
Kobayashi, M.; Sato, A.; Garvey, E. P.; Fujiwara, T. J. Med. Chem. 2013, 56, 1124. (c) Johns, B.
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 26
1
2
3
4
5
6
7
8
9
A.; Kawasuji, T.; Taishi, T.; Taoda, Y.; Polycyclic carbamoylpyridone derivative having HIV
integrase inhibitory activity. WO 2006116764 A1. (d) Wang, H.; Kowalski, M. D.; Lakdawala,
A. S.; Vogt, F. G.; Wu, L Org. Lett. 2015, 17, 564.
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
(6) Yoshida, H.; Taoda, Y.; Johns, B. A.; Kawasuji, T.; Nagamatsu, D. US8754214B2
(7) Sumino, Y.; Masui, M.; Yamada, D.; Ikarashi, F.; Okamoto, K. US20140011995
(8) Goodman, S. N.; Wang, H.; Mans, D.; Kowalski, M. US9120817B2
(9) Johns, B. A.; Kawasuji, T.; Weatherhead, J. G.; Taishi, T.; Temelkoff, D. P.; Yoshida, H.;
Akiyama, T.; Taoda, Y.; Murai, H.; Kiyama, R.; Fuji, M.; Tanimoto, N.; Jeffrey, J.; Foster, S.
A.; Yoshinaga, T.; Seki, T.; Kobayashi, M.; Sato, A.; Johnson, M. N.; Garvey, E. P.; Fujiwara,
T. J. Med. Chem., 2013, 56, 5901.
(10) Obydennov, D. L.; Röschenthaler, G.;. Sosnovskikh, V. Y. Tetrahedron Lett. 2013, 54, 545.
(11) Boivin, J.; Ramos, L.; Zard, S. Z. Tetrahedron Lett. 1998, 39, 6877.
26
ACS Paragon Plus Environment