A practical, protecting-group-free synthesis of a PI3K/mTOR inhibitor

Article abstract of DOI:10.1021/op500366s

We report a practical and protecting-group-free synthesis amenable to produce multikilogram amounts of PI3K/mTOR inhibitor GDC-0980. The route employed metalation/formylation and reductive amination followed by a metal catalyzed Suzuki-Miyaura cross-coupling. The metalation was performed via triarylmagnesiate intermediates allowing formylation under noncryogenic conditions. 2-Picoline·BH₃ was employed to replace Na(OAc)₃BH in the reductive amination and to eliminate the use of molecular sieves. A concise one-step synthesis was developed for the selective monoamidation of piperazine with (S)-lactate to produce the piperazine lactamide starting material. The boronic acid was produced from 2-amino-5-bromopyrimidine in a one-step and protecting-group-free approach. The final crystallization in 1-propanol and water afforded the API in 59% overall yield in four steps and >99% purity by HPLC.

Full text of DOI:10.1021/op500366s

Subscriber access provided by University of Manitoba Libraries
Full Paper
A Practical, Protecting Group Free Synthesis of a PI3K/mTOR Inhibitor
Qingping Tian, Ursula Hoffmann, Theresa Humphries, Zhigang Cheng, Pirmin Hidber, Herbert Yajima,
Maud Guillemot-Plass, Jane Li, Ulrike Bromberger, Srinivasan Babu, David Askin, and Francis Gosselin
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/op500366s • Publication Date (Web): 13 Feb 2015
Downloaded from http://pubs.acs.org on February 17, 2015
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 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
A Practical, Protecting Group Free Synthesis of a
PI3K/mTOR Inhibitor
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
Qingping Tian,*+ Ursula Hoffmann, # Theresa Humphries, + Zhigang Cheng, + Pirmin Hidber, #
Herbert Yajima, + Maud GuillemotꢀPlass#, Jane Li,+ Ulrike Bromberger,# Srinivasan Babu+,
David Askin+ and Francis Gosselin+
+Small Molecule Process Chemistry, Genentech, Inc., 1 DNA Way, South San Francisco, CA
94080
# Pharma Technical Development PTDCA, F. Hoffmann-La Roche AG, Grenzacherstrasse 124,
CH-4070 Basel, Switzerland
*Corresponding author: qtian@gene.com
Table of contents graphic
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 32
1
2
3
4
ABSTRACT
5
6
7
8
9
We report a practical and protecting group free synthesis amenable to produce multikilogram
amounts of PI3K/mTOR inhibitor GDC-0980. The route employed metalation / formylation and
reductive amination followed by a metal catalyzed SuzukiꢀMiyaura crossꢀcoupling. The
metalation was performed via triarylmagnesiate intermediates allowing formylation under nonꢀ
cryogenic conditions. 2ꢀPicoline•BH3 was employed to replace Na(OAc)3BH in the reductive
amination and to eliminate the use of molecular sieves. A concise oneꢀstep synthesis was
developed for the selective monoꢀamidation of piperazine with (S)ꢀlactate to produce the
piperazine lactamide starting material. The boronic acid was produced from 2ꢀaminoꢀ5ꢀ
bromopyrimidine in a one step and protecting group free approach. The final crystallization in 1ꢀ
propanol and water afforded API in 59% overall yield in 4 steps and > 99% purity by HPLC.
Introduction
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 PI3K/Akt/mTOR pathway plays a central role in cell proliferation, survival, migration and
metabolism. The lipid kinases of the PI3K family are critical for the phosphorylation of the 3'ꢀ
hydroxyl group of phosphatidylinositols, leading to the activation of the serine / threonine protein
kinase Akt and further downstream oncogenes.1 The PI3K pathway is one of the most frequently
activated pathways in tumors, with mutations in one of its components detected in a significant
percentage of human cancers.2 The kinase mTOR is activated downstream of Akt and leads to
increased protein synthesis and growth.3 Thus, considerable attention has been drawn to the
development of compounds that inhibit both kinases.4 GDC-0980 is a novel small molecule
PI3K/mTOR inhibitor pursued at Genentech as an anticancer agent (Figure 1).5 To support the
safety assessment, pharmaceutical development, and clinical studies, we have developed a
practical synthesis suitable for the preparation of the API on large scale.
2
ACS Paragon Plus Environment

Page 3 of 32
Organic Process Research & Development
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
Figure 1. Structure of PI3K/mTOR inhibitor GDC-0980
First-Generation Route to the API.6 The firstꢀgeneration route commenced with the
metalation/formylation of thienopyrimidine 1 which is prepared from commercially available
methyl 3ꢀaminoꢀ4ꢀmethylthiophene 2ꢀcarboxylate drawing inspiration from a similar sequence
we developed for another project (Scheme 1).7 Treatment of thienopyrimidine 1 with nꢀBuLi at –
1
70 °C, followed by warming to –50 °C achieved complete deprotonation as confirmed by H
NMR spectroscopic analysis of aliquots quenched into D2O. Subsequent formylation of the
resulting organolithium compound with DMF at –70 °C, followed by quenching into cold
aqueous HCl afforded aldehyde 2 in 87% yield.
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 32
1
2
3
4
5
Scheme 1. First-Generation Route to GDC-0980
O
O
6
7
8
9
N
HC(OCH3)3
HOAc
Boc
N
N
1) n-BuLi, -70 oC
O
H
S
N
+
S
N
2) DMF, -70 oC
3) 0.25 M aq. HCl
87%
Na(OAc)3BH
88%
N
H
N
Cl
N
Cl
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
Me
Me
2
3
1
O
N
Boc
N
O
Boc
N
O
O
N
B
N
+
N
S
N
PdCl2(PPh3)2/Na2CO3
2-Propanol/H2O
N
S
N
N
N
N
Florisil/Thio-Silica
90%
Me
N
Cl
N
NHBoc
Me
NHBoc
6
4
5
O
Me
HO
O
N
HCl
HN
O
1) L-(+)-Lactic Acid
HOBt H2O, EDC HCl
DIPEA, CH2Cl2
N
N
HCl
N
N
N
S
N
N
S
N
2) MeOH/THF slurry
61%
EtOH
N
N
Me
Me
N
NH2
N
NH2
7
GDC-0980
The reductive amination of aldehyde 2 with Bocꢀpiperazine 3 was performed using trimethyl
orthoformate as the dehydrating agent. Next, the SuzukiꢀMiyaura coupling reaction of
intermediate 4 and the boronate 5 led to intermediate 6. A protecting group was needed for the
boronate 5 to improve the solubility of the SuzukiꢀMiyaura coupling product 6 and facilitate the
removal of the residual Pd. As such, a common Boc protection scheme was introduced for the
boronate 5. After a brief screening of reaction conditions, PdCl2(PPh3)2 and Na2CO3 were
identified as the suitable catalyst and base for the reaction and 1.2 equiv of the boronate and 0.01
equiv of the Pd catalyst were sufficient to drive the reaction to completion. 1,4ꢀDioxane, a ICH
Class 2 solvent, was initially used as the solvent. After a brief survey of solvents, 2ꢀpropanol
was identified as the suitable alternative solvent for this reaction. The reaction proceeded faster
4
ACS Paragon Plus Environment

Page 5 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
in 2ꢀpropanol/water and was complete in 2–3 h. The crude product typically contained 200–
1000 ppm residual Pd which was then treated with Florisil®(2.0 wt) and ThioꢀSilica® (0.40 wt) in
dichloromethane for 16 h to reduce residual Pd to < 20 ppm.
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 step of the synthesis featured two chemical transformations; the deprotection of the two
Boc groups and the amidation with Lꢀ(+)ꢀlactic acid as shown in Scheme 1. Deprotection was
readily achieved without any issues by using HCl in ethanol. However, the amidation reaction
turned out to be problematic. Coupling agents, including 2ꢀchloroꢀ4,6ꢀdimethoxyꢀ1,3,5ꢀtriazine
(CDMT), 1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl)carbodiimide hydrochloride (EDC•HCl) and
additive such as 1ꢀhydroxyꢀ1,2,3ꢀbenzotriazole hydrate (HOBt•H2O), have been explored;8
however, no desired product was observed when these coupling reagents were employed
individually in the reaction. It appeared that the combination of EDC•HCl and HOBt•H2O would
work as the reaction proceeded well on lab scale; however, when we performed the reaction on
kilogram scale, it failed to reach completion after 24 h, with 10–20% of intermediate 7 remaining
unconsumed, and required additional amounts of EDC•HCl and Lꢀ(+)ꢀlactic acid to drive the
reaction to completion.
Another issue for the coupling step was the formation of impurities, with the major two being
identified as the desꢀlactate 8 and diꢀlactate 9 (Figure 2), found at 1.8A% and 3.7A% levels by
HPLC, respectively. The crude product was then reꢀslurried in a mixture of methanol (7.5 vol)
and THF (2.5 vol) and the impurities were reduced to 0.53A% and 1.1A%, respectively, with the
overall purity being improved to 97A%.
Residual HOBt (0.89wt%) and N,Nꢀ
diisopropylethylamine (DIPEA) (0.1wt%) were also observed in the crude product, but these
could be readily purged by an acid/base extraction procedure; however, the overall yield of this
step was only 61% due to loss of the product during the above purification procedures.
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 32
1
2
3
4
5
6
Me
O
N
O
N
O
N
HO
HN
7
8
N
N
S
S
N
N
N
N
O
9
N
N
Me
Me
Me
OH
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
N
N
NH2
H
Des-lactate 8
Di-lactate 9
Figure 2. Process Impurities in Final Step of First-Generation Route
While the firstꢀgeneration synthesis was successful for the production of the GLP tox batch and
initial GMP batches, it suffered from several shortcomings that were not ideal for largeꢀscale
implementation. The synthesis was not convergent and also required protection/deprotection
steps. In addition, the yield of the final step was low. Thus, a more convergent and efficient
synthesis was needed to produce large amount of API for the advanced clinical studies.9
Second-Generation Route. Since the reductive amination and SuzukiꢀMiyaura coupling
reactions performed well in the firstꢀgeneration synthesis, we elected to keep both for future
syntheses but strive to overcome the existing problems. As such, we envisioned that the API
could be assembled in a highly convergent manner via SuzukiꢀMiyaura coupling of unprotected
boronic acid 10 and
2ꢀchloroꢀthienopyrimidine 11 (Scheme 2). After the metalation and
formylation of thienopyrimidine 1, piperazine lactamide 12 would be appended via the reductive
amination. By a simple reordering of steps, this would avoid the problematic lateꢀstage
lactamide formation encountered in the firstꢀgeneration synthesis.
6
ACS Paragon Plus Environment

Page 7 of 32
Organic Process Research & Development
1
2
3
4
Scheme 2. Retrosynthetic Analysis of the Second-Generation Route
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
Improved Formylation Process. The cryogenic conditions employed in the first generation
route were not ideal for large scale synthesis. We recently reported that deprotonation of a
thienopyrimidine heterocycle could be achieved under nonꢀcryogenic conditions through use of a
lithium trialkylmagnesiate.7,10 To our delight, we were able to also use this method for
thienopyrimidine 1. It is noteworthy that the presence of the adjacent methyl group significantly
improved the stability of the lithium triarylmagnesiate 13 and thus led to the excellent yield of
aldehyde 2 (Scheme 3).11
Scheme 3. Non-cryogenic Conditions for Metalation and Formylation
Isolation of the aldehyde product proved challenging because of slow filtration of the crude
mixture. We found that removal of THF was beneficial to the filtration. The filtration rate was
also significantly improved through an ripening process. After aging for 1–2 h at 50 °C, the
filtration was about 10× faster as the specific cake resistance was reduced from 1.3×1011 m/kg to
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 32
1
2
4.2×109 m/kg. This is the result of a crystal form change confirmed by DSC and XPRD. We also
noticed that the ripening process afforded larger crystals as indicated by the microscopy data.
Under the optimal conditions, the desired product was reproducibly produced in 94‒98% yield.
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
Reductive Amination. As previously mentioned, the lactamide moiety was incorporated into
the new starting material 12 (Scheme 2). Initially, the HCl salt of 12 was tested in the reaction,12
but a significant amount of the starting material 2 did not react, presumably due to the fact that
the HCl salt was deliquescent under normal lab conditions and introduced water into the reaction.
After screening a variety of acids, we found that the corresponding oxalic acid salt 14, was wellꢀ
behaved, much less hygroscopic and thus easier to handle.13 Therefore, piperazine lactamide
oxalate 14 was chosen as the starting material for the reductive amination (Scheme 4).
Scheme 4. Reductive Amination of the Second-Generation Synthesis
Method A:
Me
HO
O
N
O
N
O
N
O
N
1) NaOAc, HOAc
MS 3Å, ACN
2) Na(OAc)3BH
75%
O
Me
HO
NH
OH
HO
O
H
S
S
N
N
S
+
N
N
+
N
HO
O
Method B:
1) NaOAc, HOAc
MeOH, HC(OCH3)3
2) 2-Picoline BH3
82%
O
N
Cl
N
Cl
N
Cl
Me
Me
Me
2
14
11
15
Initially, Na(OAc)3BH was employed as the reducing reagent (Scheme 4, Method A). With
acetonitrile as solvent, we observed the best conversion and the least amount of the alcohol
impurity 15. After a brief survey of bases, which is required to form the free base of oxalate 14
in situ, we identified sodium acetate as optimal.14 We also found that addition of acetic acid (0.5
equiv) was critical in order to suppress the formation of alcohol impurity 15. This impurity 15
was further controlled by using 1.5 equiv of 14 and adding the Na(OAc)3BH in multiple portions.
8
ACS Paragon Plus Environment

Page 9 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
A dehydrating reagent was needed for the iminium ion formation in the reaction. When
HC(OCH3)3 was employed, the outcome of the reaction was not reproducible. In some runs,
significant amounts of the unconsumed starting material 2 and alcohol impurity 15 were
observed. We eventually resolved this problem by the use of molecular sieves, with the reaction
proceeding consistently well when powdered 3Å molecular sieves were employed. Use of 100
wt% of sieves suppressed the formation of alcohol impurity 15.
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
Although the reductive amination process using Na(OAc)3BH has been successfully scaled up to
50 kg, it was not desirable to use sieves and add Na(OAc)3BH in multiple portions.
Na(OAc)3BH needs to be added as solid via a special solid dosing unit or as a slurry. To address
these shortcomings, other reducing agents such as sodium borohydride, pyridine•BH3 and 2ꢀ
picoline•BH3 were explored in the reaction.15 A significant amount of the alcohol impurity 15 (7–
10%) occurred when sodium borohydride was used. On the other hand, the reaction with
pyridine•BH3 or 2ꢀpicoline•BH3 proceeded smoothly and fewer impurities formed, although the
level of alcohol impurity 15 was slightly higher than that when Na(OAc)3BH was used.16
We examined the reactions with 2ꢀpicoline•BH3 at different temperatures, and in the presence or
absence of the dehydrating reagent HC(OCH3)3 (Scheme 4, Method B and Table 1). We were
able to achieve the best conversion and the lowest A% of the alcohol impurity 15 when the
o
reaction was run in the presence of HC(OCH3)3 at 50 C (entry 3). It was also shown that
addition of 2ꢀpicoline•BH3 in 3 portions was effective. Other solvents such as ACN, THF, and
EtOH were examined leading to higher amounts of the alcohol impurity 15 (entries 4 ̶ 6). We
determined that 1.2 equiv of 2ꢀpicoline•BH3 and 10 equiv of HC(OCH3)3 would be needed to
drive the reaction to completion (entries 3, 7, 8 and 10). Moreover, we were able to charge 2ꢀ
picoline•BH3 as solutions in methanol and THF eliminating the former problem of solids
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 32
1
2
3
4
addition (entries 9
̶
10). Since 2ꢀpicoline•BH3 in THF is commercially available, it was selected
5
6
for large scale production.
7
8
9
a)
Table 1. Optimization of the Reaction with 2-Picoline•BH3
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
A% of the reaction mixture by
HPLC
Entry Temp
(°C)
2ꢀPicoline•BH3
(equiv)
Solvent
11
15
9.6
2
1
2
3
4
5
6
7
8
9
rt
1.50 (solid)
1.50 (solid)
1.50 (solid)
1.50 (solid)
1.50 (solid)
1.50 (solid)
1.25 (solid)
1.00 (solid)
MeOH
MeOH
MeOH
ACN
84.0
93.0
95.1
84.3
18.6
63.0
94.5
93.8
94.9
6.4
0.0
0.0
2.1
8.7
0.2
0.5
0.6
0.6
rt
7.0
50
50
50
50
50
50
50
4.9
13.6
72.7
36.8
5.0
THF
EtOH
MeOH
MeOH
MeOH
5.6
1.25 (13wt%
solution in MeOH)
4.4
10
50
1.20 (30wt%
solution in THF)
MeOH
96.0
3.5
0.5
a) The reaction of entry 1 was run without HC(OCH3)3. All other reactions were operated in the presence of 10 equiv
of HC(OCH3)3.
The reductive amination reaction using 2ꢀpicoline•BH3 produced cleaner intermediate compared
to the preceding process and has been successfully scaled up to 8.75 kg with a reproducible yield
of 79‒86%.
Synthesis of Oxalate 14. The initial synthesis of oxalate 14, illustrated in Scheme 5, employed 5
steps with an overall yield of 40% (Scheme 5). Several steps were utilized for the protection and
10
ACS Paragon Plus Environment

Page 11 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
deprotection of both the hydroxyl group and piperazine. As the project was advancing to late
stage process development, we developed a more concise and efficient synthesis.
Scheme 5. The Initial Synthesis of Oxalate 14
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) N-Benzylpiperazine Route. We initially envisioned a oneꢀpot amidationꢀdeprotection
process starting from Nꢀbenzylpiperazine with (S)ꢀethyl lactate as solvent at elevated
temperatures (70–100 oC); however, these reactions did not reach completion. (S)ꢀMethyl lactate
was also investigated, as we reasoned the lower boiling point of the methanol byꢀproduct could
aid the reaction conversion, but this reaction also failed to progress to completion and therefore
offered no advantages over the cheaper (S)ꢀethyl lactate. A further issue with these reactions was
the formation of the ester impurity 21 in ~ 13A% (Figure 3).
Figure 3. Ester By-product 21
Next, we decided to concentrate our efforts on reaction conditions that could be carried out at
ambient temperature. The amidation between amines and (S)ꢀethyl lactate in the presence of an
alkoxide base has previously been reported.17 Thus, we explored several alkoxide bases and
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 32
1
2
3
4
5
6
selected NaOMe (25wt% in MeOH) since erosion of enantiomeric ratio (er) was observed when
other bases were employed (Table 2 and Scheme 6).
7
8
9
Table 2. Effect of Alkoxide Bases on Enantiomeric Ratio of the Amidation Reaction
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
Entry
Base
(0.15 equiv)
(S)ꢀEthyl lactate
HPLC assay yield Enantiomeric
(equiv)
(%)
ratio (er)
85.0:15.0
96.7:3.3
1
2
tꢀBuONa (solid)
3.00
1.03
90
tꢀPentONa 1.4 M
solution in THF
61
3
4
NaOEt 21wt%
solution in EtOH
1.03
1.03
37
34
98.9:1.1
99.3:0.7
NaOMe 25wt%
solution in MeOH
Scheme 6. Benzylpiperazine Route to Oxalate 14
Optimization of the reaction conditions were focused on the amount of base and (S)ꢀethyl lactate.
We identified the optimal conditions to be NaOMe (25wt% in MeOH) (0.75 equiv) and (S)ꢀethyl
lactate (3 equiv). The reaction was complete at ambient temperature in ~ 20 h, affording less
than 1% of residual Nꢀbenzylpiperazine and an enantiomeric ratio of 99.2:0.8. We also
12
ACS Paragon Plus Environment

Page 13 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
investigated the effect of temperature on the reaction and found that at elevated temperature (40
°C and 70 °C), the reaction did progress at a faster rate, achieving <1% residual Nꢀ
benzylpiperazine after just 2 h at both temperatures. However, as expected, these elevated
temperatures had a detrimental effect on the chiral purity that after 2 h at 40 °C and 70 °C,
eroded to 98:2 er and 86:14 er, respectively. Thus, the reaction was performed at ambient
temperature.
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
Our attention then turned towards the removal of the residual Nꢀbenzylpiperazine in the
intermediate 23 in order to meet the specification of < 0.2A% of piperazine18 in oxalate 14 as Nꢀ
benzylpiperazine will lead to piperazine upon deprotection. A brief screening of solid phase
scavengers identified activated charcoal to be effective in depleting the residual Nꢀ
benzylpiperazine to <1A%. We observed an acceptable 10–15% decrease in yield, but
pleasingly, no degradation in enantiomeric ratio, with it remaining at 99.3:0.7 er.
Residual sodium salts such as sodium ethoxide and / or sodium methoxide from the reaction
were removed using a mildly acidic cation exchange resin, Amberlite IRCꢀ748, while
maintaining the high reaction yield. This allowed control over the sodium oxalate which would
be formed in the salt formation step and proved difficult to be removed from the final product
oxalate 14.
The benzyl lactamide 23 was then subjected to catalytic hydrogenolysis and oxalate salt
formation to afford 14 in 73% yield with good potency (97.5wt%) and excellent er (99.3:0.7).
The product was obtained within the specifications for residual piperazine (0.03wt%) and sodium
(0.21wt%).
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 32
1
2
3
4
5
6
7
8
9
(2) Piperazine Route. In a continued effort to make the synthesis more efficient, we explored a
oneꢀstep synthesis of oxalate 14 by the amidation of piperazine with (S)ꢀethyl lactate (Scheme 7).
In our initial attempt, using 1.50 equiv of (S)ꢀethyl lactate, the reaction afforded a 55:45 mixture
of the desired product 12 and the bisꢀlactamide 24, and the desired product 12 in 50% assay
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
Scheme 7. Piperazine Route to Oxalate 14
Encouraged by the initial results, we optimized the reaction conditions to improve the selectivity
and yield. First, different amounts of (S)ꢀethyl lactate were examined in the reaction (Table 3).
We found that when more (S)ꢀethyl lactate was used, the assay yield of desired product 12
decreased as more bisꢀlactamide 24 was generated (entries 5–7). Conversely, when less (S)ꢀethyl
lactate was used, the assay yield increased, but more residual piperazine was observed (entries 1–
3). An optimal balance between assay yield and residual piperazine was reached with 1.3 equiv
of (S)ꢀethyl lactate (entry 4). Although the reaction achieved 60% assay yield after only 2 h,
holding the batch to 20 h allowed residual piperazine to be further consumed. It is noteworthy
that the enantiomeric ratio remained constant at 99.4:0.6 even when the reaction time was
extended to 6 days.
14
ACS Paragon Plus Environment

Page 15 of 32
Organic Process Research & Development
1
2
3
4
Table 3. Assay Yields of 12 Using Different Amounts of (S)-Ethyl lactate
5
6
7
8
Entry
(S)ꢀEthyl Lactate
(equiv)
Residual piperazine
(wt%) by HPLC
Yield of Bisꢀ
Assay yield of 12
by HPLC
lactamide 24a
9
1
2
3
4
5
6
7
1.0
1.1
1.2
1.3
1.4
1.5
1.6
63
66
64
63
55
50
45
14
23
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
23
9.2
6.6
5.2
3.4
2.6
26.8
30.4
39.8
46.6
52.4
a) The yield of bisꢀlactamide 24 = 100% – assay yield of 12 (%) – residual piperazine (%)
We discovered that the assay yield was improved to >70% when 3.0 equiv of water was added as
the result of the conversion of the bisꢀlactamide 24 to the desired product 12 presumably by
hydrolysis. 19
Our next task was to isolate the product from the reaction mixture. We thus needed to purge out
the residual sodium salt, present as sodium ethoxide and / or sodium methoxide, residual
piperazine and (S)ꢀethyl lactate, as well as the byꢀproducts, the bisꢀlactamide 24 and lactic acid,
the latter presumably being generated from the hydrolysis of (S)ꢀethyl lactate and the bisꢀ
lactamide 24.
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 32
1
2
3
4
5
6
7
8
9
Since 12 is soluble in water, aqueous work up was not an option to remove the residual sodium.
Our initial attempt was to use Amberlite IRCꢀ748 resin treatment to remove sodium salts.
However, we observed significant loss of product on the resin (~ 20%) even when using the
minimum amount of resin (0.8 equiv) that was needed to sufficiently remove sodium salts. We
next investigated formation of a sodium salt as a way to remove residual sodium, choosing oxalic
acid, since the final product 14 is an oxalate. We found that 0.25 equiv of oxalic acid was able
to generate the diꢀsodium oxalate salt which is insoluble in ethanol and could therefore be readily
removed by filtration.20 Additional oxalic acid was then added to adjust the pH to 7–7.5, the
residual piperazine was precipitated out as the oxalate salt and subsequently removed through a
filtration while the product 14 remained in the mother liquor. Direct treatment of the resultant
filtrate with excess oxalic acid (1.12 equiv) afforded the desired oxalate 14 in 59% isolated yield
and >99% purity.
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 piperazine route to oxalate 14 is significantly more efficient with a 55% reduction in process
mass intensity (PMI).21a Not only was the volume of waste dramatically reduced, but more
importantly, we were able to obviate the use of the hazardous and toxic reagents used in the
enabling synthesis. We were also able to achieve 53% reduction in total solvent used in the
process. More significantly, the undesirable solvent, dichloromethane, heavily used in 3 of the 5
steps of the enabling synthesis, was replaced with preferred solvents and a small amount of THF
(4.3%).21b Another improvement for the piperazine route is a 41% increase in atom economy as
the result of the elimination of the nonꢀvalue adding steps, and thus producing oxalate 14 in a
single chemical step and a protecting group free synthesis.21c
Overall, the piperazine route is more concise, cleaner and safer.
16
ACS Paragon Plus Environment

Page 17 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Suzuki-Miyaura Cross-Coupling. We next turned our attention to the SuzukiꢀMiyaura crossꢀ
coupling reaction (Scheme 8). As previously illustrated in Scheme 2, a boronic acid 1022 was
used to replace the expensive boronate 5 employed in the firstꢀgeneration route. But unlike the
firstꢀgeneration synthesis, no protecting group was employed for boronic acid 10. In addition,
boronic acid 10 was prepared directly from 2ꢀaminoꢀ5ꢀbromopyrimidine in a oneꢀstep and
protecting group free synthesis.23
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
Scheme 8. Suzuki-Miyaura Cross-Coupling Reaction of the Second-Generation Synthesis
Me
O
N
Me
O
N
O
N
O
N
B(OH)2
PdCl2(PPh3)2
HO
HO
K3PO4, n-PrOH
N
S
+
N
S
N
N
N
N
85 oC
81%
N
N
NH2
N
Cl
Me
Me
N
NH2
10
11
crude GDC-0980
Several solvents including 2ꢀpropanol, 1ꢀpropanol and acetonitrile were evaluated for the
reaction. 1ꢀPropanol was chosen as the reaction solvent due to the better solubility of the product
GDC-0980 and its higher boiling point which allowed us to run the reaction at a higher
temperature. A variety of bases had been investigated in the reaction when boronate ester 5 was
used. We found that in the presence of Na2CO3 or Cs2CO3, a significant amount of amide
hydrolysis byproduct desꢀlactate 8 (7–17%, Figure 2) was generated.
To our delight, with
K3PO4 no desꢀlactate 8 was detected in the reaction with the boronate ester 5 or when we later
made the switch to the corresponding boronic acid 10. Different amounts of boronic acid were
examined in the reaction. We found that 1.18 equiv of boronic acid 10 were sufficient for the
reaction.
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 32
1
2
3
4
5
6
7
8
9
Expensive scavengers were employed for the residual Pd removal in the firstꢀgeneration route,
which would not be practical on large scale. To resolve this issue, we reduced the Pd catalyst
loading to 0.15 mol% permitted by the faster reaction at higher temperature with 1ꢀpropanol as
the solvent, in combination with an activated carbon cartridge filtration of crude API solution in
1ꢀpropanol and water, to reduce residual Pd to 20–40 ppm.
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
Three major impurities, desꢀlactate impurity 8 (Figure 2) in ~ 0.60A%, homoꢀcoupling impurity
25 (Figure 4) in ~ 0.65A% and alcohol impurity 26 (Figure 4) in ~ 0.25A%, stemming from
alcohol impurity 15 in the precursor, were generated in the reaction. The removal of these
impurities proved challenging in particular for 25. We found that all three impurities could be
depleted to < 0.20A% through a crystallization process in 25:75 w/w mixture of 1ꢀpropanol and
water. The product was produced on 16 kg scale reproducibly in 79‒83% yield.
Figure 4. Homo-coupling Impurity 25 and Alcohol Impurity 26
API Crystallization. We developed the API crystallization to further deplete residual Pd and
process impurities and achieve the desired crystalline form and particle size distribution. We
found that the API has relatively low solubility in organic solvents; however, the API did show
good solubility in a mixture of water and 1ꢀpropanol, with maximum solubility in 40:60 w/w
(Figure 5). The solubility and superꢀsaturation curves in this solvent composition are displayed
in Figure 6.
18
ACS Paragon Plus Environment

Page 19 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
140
120
100
80
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
5wt% API
7wt% API
11wt% API
60
0
10
20
30
40
50
60
70
80
Water in 1-Propanol (wt%)
Figure 5. Solubility Curves of Various Concentrations of GDC-0980 as a Function of
Water/1-Propanol Composition
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 32
1
2
3
4
5
6
7
8
9
17
15
13
11
9
clear point
cloud point
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
7
5
45
50
55
60
65
70
75
80
85
Figure 6. Clear and Cloud Curves of GDC-0980 in Water/1-Propanol (40:60 w/w)
The boiling point of the water/1ꢀpropanol mixture (40:60 w/w) was 87–88 °C. Under normal
pressure the maximum achievable solubility is ~14 wt% as shown in Figure 6. For practical
purposes and to ensure that the API would not spontaneously precipitate out during the polish
filtration, we set the concentration for dissolution at 11.7wt%, resulting in a final concentration
of 10wt% after the polish filtration and rinses. The solution was then cooled slowly to ‒10 oC
while spontaneous nucleation started at a temperature between 60 °C and 65 °C. The suspension
was filtered and dried to afford API of > 99.0% by HPLC, all identified impurities within the
specifications and all unidentified impurities < 0.15A% (Scheme 9). The crystallization process
has been scaled up reproducibly in multiple batches (13.9 kg to 15.7 kg) in 89 ̶ 96% yields.
Scheme 9. API Crystallization
20
ACS Paragon Plus Environment

Page 21 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Me
HO
O
N
O
N
Me
HO
O
N
O
N
1) n-PrOH/Water
88 oC
N
S
N
N
S
2) Cool to -10 oC
92%
N
N
N
N
N
Me
Me
N
NH2
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
NH2
crude GDC-0980
GDC-0980
Conclusion
We have developed a practical and protecting group free synthesis for GDC-0980 (Scheme 10).
Noncryogenic conditions were employed in the formylation of 1 via a triarylmagnesiate
intermediate. The synthesis of the key intermediate 11 was achieved through a reductive
amination in the presence of 2ꢀpicoline•BH3. A concise oneꢀstep synthesis was developed for the
direct monoꢀlactamidation of piperazine with (S)ꢀlactate to afford the key starting material,
oxalate 14. The Boc protected boronate 5 was replaced with boronic acid 10 which was
prepared in one step directly from 2ꢀaminoꢀ5ꢀbromopyrimidine obviating the use of a protecting
group. We developed the crystallization process to produce the API in 59% overall yield in four
steps and >99A% HPLC purity on 10 kg scale.
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 32
1
2
3
4
Scheme 10. Manufacturing Scale Synthesis of GDC-0980
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
Experimental
General. All reactions were performed under a nitrogen (or Argon) atmosphere. Melting points
were measured by differential scanning calorimetry (DSC). HPLC method for purity and assay
analysis are listed below.
HPLC method for compounds 1 and 2: column, Water XBridge Phenyl (150 mm × 4.6 mm, 3.5
o
µm); temperature: 10 C; mobile phase A, water; mobile phase B, CH3CN; gradient (17 min)
60:40 A/B to 40:60 A/B over 10 min, then change to 20:80 A/B over 2 min, then change to 60:40
A/B in 0.1min, hold for 5 min; flow rate = 1.2 mL/min; detection, 210 nm; injection volume, 10
µL; tR of 1 = 6.12 min, tR of 2 = 6.95 min.
22
ACS Paragon Plus Environment

Page 23 of 32
Organic Process Research & Development
1
2
3
4
HPLC method for compounds 14: column, Sielc Primesep 100 C18 (4.6 mm × 150 mm, 5 um);
5
6
o
temperature: 40 C; mobile phase A, 0.5% H3PO4 in water; mobile phase B, CH3CN; gradient
7
8
9
(18 min) 50:50 A/B for 8 min, then change to 10:90 A/B over 4 min, hold for 3 min, then change
to 95/5 A/B in 0.1 min, hold for 3 min, then change to 50/50 A/B in 0.1 min; flow rate = 1.2
mL/min; detection, 210 nm; injection volume, 5 µL; tR of 14 = 7.0 min.
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
HPLC method for compound 11: column, Water XBridge C18 (150 mm × 4.6 mm, 3.5 µm);
o
temperature: 40 C; mobile phase A, water, 0.1% (v/v) 25% aqueous NH4OH; mobile phase B,
CH3CN, 0.1% (v/v) 25% aqueous NH4OH; gradient (43 min) 95:5 A/B to 80:20 A/B over 5 min,
then change to 71:29 A/B over 20 min, then change to 16:84 A/B over 8 min, then hold for 5
min, then change to 95/5 A/B in 0.1 min, hold for 5 min; flow rate = 1.4 mL/min; detection, 270
nm; injection volume, 10 µL, tR of 11 = 25.0 min, tR of 2 = 27.5 min.
HPLC method for GDC-0980: column, Water XBridge C18 (150 mm × 4.6 mm, 3.5 µm);
temperature: 40 oC; mobile phase A, 1000 ml of water, 560 µL of 25% aqueous NH4OH, pH 9.2
(adjusted with HOAc); mobile phase B, CH3CN; gradient (43 min) 95:5 A/B to 80:20 A/B over
5 min, then change to 71:29 A/B over 20 min, then change to 16:84 A/B over 8 min, then hold
for 5 min, then change to 95/5 A/B in 0.1 min, hold for 5 min; flow rate = 1.2 mL/min; detection,
270 nm; injection volume, 15 µL, tR of GDC-0980 = 16.9 min.
2-Chloro-7-methyl-4-morpholinothieno[3,2-d]pyrimidine-6-carbaldehyde (2). A mixture of
4ꢀ(2ꢀchloroꢀ7ꢀmethylthieno[3,2ꢀd]pyrimidinꢀ4ꢀyl)morpholine (1) (9.45 kg, 35.0 mol) and
anhydrous THF (94.5 L) was cooled to below –8 °C. A 20wt% solution of iꢀPrMgCl in THF
(8.64 kg, 16.8 mol) was added, followed by addition of a 25wt% solution of nꢀBuLi in heptane
(9.84 kg, 38.4 mol) while maintaining the temperature below –8 °C. The mixture was stirred at
below –8 °C for 2 h. Anhydrous DMF (5.11kg, 69.9 mol) was added while keeping the
temperature below –8 °C. The reaction mixture was stirred for 1 h, transferred to a mixture of
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 32
1
2
3
4
5
6
7
8
9
acetic acid (23.6 kg), 37% aqueous hydrochloric acid (9.10 kg), isopropanol (94.4 L) and water
(91.7 L). The resulting slurry was heated to 50−55 °C and stirred for 1 h. The suspension was
concentrated under reduced pressure to remove THF. The suspension was then cooled to rt,
filtered and rinsed with water (2 × 20 L). The cake was dried under reduced pressure at 40−60 °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
o
1
to afford 2 as a yellow solid (10.0 kg, 96% yield): mp 198.0 C; H NMR (300 MHz, CDCl3)
13
10.37 (s, 1H), 4.16–3.97 (m, 4H), 3.93–3.70 (m, 4H), 2.75 (s, 3H); C NMR (75 MHz, CDCl3)
183.83, 162.20, 158.96, 157.50, 142.26, 140.95, 116.37, 66.59, 46.55, 11.50; HRMS calcd for
C12H12CN3O2S [M+H] 298.0412, found 298.0408.
(S)-2-hydroxy-1-(piperazin-1-yl)propan-1-one (14) (Benzylpiperazine Route, Scheme 6). A
flask charged with benzyl piperazine (100 g, 0.57 mol 1.00 equiv) was cooled to 10 °C. (S)ꢀethyl
lactate (22) (201 g, 1,67 mol, 3.00 equiv) was added at a rate to maintain the temperature below
20 °C. Sodium methoxide (25 wt% in MeOH) (97.3 mL, 0.43 mmol, 0.75 equiv) was added
while maintaining the temperature below 20 °C. The reaction mixture was allowed to warm to
ambient temperature and was aged for 16 h. The mixture was diluted with ethanol (500 mL) and
treated with Amberlite® IRCꢀ748 resin (Na+ form 630 g, 1.8meq/g, 2 equiv, preꢀconditioned to
H+ form using 5% aq HCl). The suspension was stirred at ambient temperature for 2 h. The resin
was removed by filtration through a pad of Celite® (70 g) and the pad was washed with EtOH (2
× 677 mL). The filtrate and washes were combined and concentrated under reduced pressure to
1000 mL. To the solution was added activated charcoal (70 g), the suspension was stirred at
ambient temperature for 18 h, then filtered through a pad of Celite® (70 g) and the pad was
washed with EtOH (2 × 677 mL). The filtrate and washes were combined and concentrated under
vacuum to afford a residue containing (S)ꢀ1ꢀ(4ꢀBenzylpiperazinꢀ1ꢀyl)ꢀ2ꢀhydroxypropanꢀ1ꢀone
(23).
24
ACS Paragon Plus Environment

Page 25 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The residue was diluted with EtOH (450 mL) and the solution was degassed three times through
a nitrogen and vacuum cycle. Palladium (10% on activated carbon, 56.14% water wetted) (41.4
g, 0.017 mol, 0.03 equiv) was added and the mixture was degassed five times through a nitrogen
and vacuum cycle. The mixture was heated to 55 °C and cyclohexene (233 g, 2.84 mol, 5.00
equiv) was slowly added. The reaction mixture was heated to reflux and stirred for 2.5 h. After
cooling to ambient temperature, the mixture was filtered through a pad of Celite® (70 g) and
washed with EtOH (2 × 675 mL). The filtrate and wash were combined and concentrated under
reduced pressure to afford a residue containing (12) as an oil which was dissolved in a 50:50
(v/v) mixture of EtOH/THF (352 mL each) and cooled to 10 °C. A solution of oxalic acid
dihydrate (143 g, 1.23 mol, 2.00 equiv) in a 50:50 (v/v) mixture of EtOH/THF (282 mL each)
was slowly added. The suspension was allowed to warm to ambient temperature and aged for 13
h. The suspension was cooled to 10°C and the solid was collected by filtration. The solid was
washed with cold EtOH (2 x 550 mL) and dried under vacuum at 50 °C for 24 h to afford oxalate
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
1
salt 14 as a white solid (103 g, 73%): mp 115.2 C; H NMR (300 MHz, D2O) 4.73–4.51 (m,
13
1H), 3.93–3.59 (m, 4H), 3.26 (dd, J = 8.8, 4.0 Hz, 4H), 1.27 (d, J = 6.7 Hz, 3H). C NMR (75
MHz, D2O) 174.48, 165.55, 64.36, 43.02, 42.94, 41.96, 38.91, 19.17. HRMS calcd for
C7H14N2O2 [M+H] 159.1128, found 159.1122.
(S)-2-hydroxy-1-(piperazin-1-yl)propan-1-one (14) (Piperazine Route, Scheme 7). A flask
charged with piperazine (100 g, 1.16 mol) and (S)ꢀethyl lactate (22) (178 g, 48 mmol, 1.30 equiv)
was cooled to 10 °C. Sodium methoxide (25 wt% in MeOH) (128 g, 0.59 mol, 0.50 equiv) was
added while maintaining the temperature below 20 °C. The reaction mixture was allowed to
warm to ambient temperature and aged for 19 h. Water (62.3 g, 3.46 mol, 3.0 equiv) was added
and the mixture was aged for 16 h. The mixture was diluted with EtOH (400 mL) and
concentrated under reduced pressure to half of the volume. The residue was flushed with EtOH
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 32
1
2
3
4
5
6
7
8
9
(3 × 300 mL) to remove the water. The resulting solution (300 mL) was treated with a solution
of oxalic acid dihydrate (37 g, 0.29 mol, 0.25 equiv) in EtOH (180 mL) and the resulting
suspension was aged for 1 h. The suspension was cooled to < 10 °C and a solution of oxalic acid
dihydrate (37 g, 0.29 mol, 0.25 equiv) in EtOH (180 mL) was added to adjust pH to 7.0–7.5. The
suspension was cooled to 0–5 °C, aged for 1 h, filtered through a pad of Celite® (404 g) and
washed with cold EtOH (2 × 350 mL). The filtrate and washes were combined and filtered
through a 0.45 ꢁm membrane filter, concentrated to 1300 mL. The solution was cooled to 10 °C
and a solution of oxalic acid dihydrate (161 g, 1.28 mol, 1.12 equiv) in EtOH (750 mL) was
slowly added. The suspension was allowed to warm to ambient temperature and stirred for 12 h.
The suspension was cooled to 5 °C, aged for 1 h and the solid was collected by filtration, washed
with cold EtOH (2 × 180 mL) and dried under reduced pressure at 50 °C for 24 h to afford
oxalate 14 as a white solid (161 g, 59%).
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
(S)-1-(4-((2-Chloro-7-methyl-4-morpholinothieno[3,2-d]pyrimidin-6-yl)methyl)piperazin-1-
yl)-2-hydroxypropan-1-one (11) 2ꢀChloroꢀ7ꢀmethylꢀ4ꢀmorpholinothieno[3,2ꢀd]pyrimidineꢀ6ꢀ
carbaldehyde (2) (8.75 kg, 29.4 mol) was charged to a reactor, followed by MeOH (165 L), (S)ꢀ2ꢀ
hydroxyꢀ1ꢀ(piperazinꢀ1ꢀyl)propanꢀ1ꢀone oxalate (14) (10.9 kg, 43.9 mol), sodium acetate (7.21
kg, 87.9 mol), glacial acetic acid (1.76 kg, 29.3 mol) and trimethylorthoformate (31.2 kg, 294
mol). The slurry was heated to 55−60 °C and stirred for 4 h. A 30% solution of 2ꢀ
picoline•borane in THF (12.6 kg, 35.3 mol) was added and the slurry was stirred for 1 h. The
reaction mixture was partially concentrated under reduced pressure. Toluene (134 L) was added
and the reaction mixture was concentrated under reduced pressure. Toluene (66.7 L) was added
and the reaction mixture was again partially concentrated under reduced pressure. To the residue
was added toluene (113 L) and the mixture was cooled to 20−30 °C. Water (253 L) was added
and the pH was adjusted to 7.5–8.5 with 10% aqueous sodium carbonate solution. The organic
26
ACS Paragon Plus Environment

Page 27 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
phase was separated, cooled to 0−5 °C and extracted with a mixture of water (105.7 L) and 96%
sulfuric acid (3.55 kg). The aqueous phase was separated and toluene (69.6 L) was added. The
pH was adjusted to 7.5−8.5 with 10% aqueous sodium carbonate solution (60.4 L) at 0−5 °C.
The mixture was warmed to 20 °C and the organic phase was separated. The organic phase was
diluted with toluene (58.6 L) and concentrated under reduced pressure to its original volume
(approximately 80 L). The solution was warmed to 53−57 °C, and nꢀheptane (13.5 L) was added.
The solution was seeded with (S)ꢀ1ꢀ(4ꢀ((2ꢀchloroꢀ7ꢀmethylꢀ4ꢀmorpholinothieno[3,2ꢀd]pyrimidinꢀ
6ꢀyl)methyl)piperazinꢀ1ꢀyl)ꢀ2ꢀhydroxypropanꢀ1ꢀone (11) (30 g) and the suspension was stirred at
53−57 °C for 30 min. nꢀHeptane (48.2 L) was slowly added and the resulting slurry was cooled
to 0−5 °C, filtered and washed with a mixture of toluene (6.9 L) and nꢀheptane (14.7 L) and
subsequently with nꢀheptane (29.4 L). The cake was dried at 30−45 °C under reduced pressure to
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
1
afford 11 (10.6 kg, 82% yield): mp 145.8 C; H NMR (300 MHz, DMSOꢀd6) 4.84 (d, J = 6.9
Hz, 1H), 4.45–4.37 (m, 1H), 3.90–3.84 (m, 4H), 3.79–3.63 (m, 4H), 3.58–3.41 (m, 4H), 3.30 (s,
13
2H), 2.59–2.35 (m, 4H), 2.24 (s, 3H), 1.17 (d, J = 6.5 Hz, 3H). C NMR (75 MHz, DMSOꢀd6)
172.06, 162.19, 157.90, 155.55, 145.57, 128.67, 111.59, 65.79, 63.98, 54.15, 52.90, 52.46, 45.86,
44.42, 41.37, 20.60, 10.83; HRMS calcd for C19H26ClN5O3S [M+H] 440.1518, found 440.1505.
(S)-1-(4-((2-(2-Aminopyrimidin-5-yl)-7-methyl-4-morpholinothieno[3,2-d]pyrimidin-6-
yl)methyl)piperazin-1-yl)-2-hydroxypropan-1-one (GDC-0980). (S)ꢀ1ꢀ(4ꢀ((2ꢀchloroꢀ7ꢀmethylꢀ
4ꢀmorpholinothieno[3,2ꢀd]pyrimidinꢀ6ꢀyl)methyl)piperazinꢀ1ꢀyl)ꢀ2ꢀhydroxy propanꢀ1ꢀone (11)
(16.5 kg, 37.5 mol) was charged to a reactor, followed by nꢀpropanol (143 L), 2ꢀaminopyrimidinꢀ
5ꢀylboronic acid (10) (6.22 kg, 44.8 mol) and potassium phosphate (15.9 kg, 74.9 mol). The
resulting mixture was degassed by vacuum/argon purge three times. Bis(triphenylphosphine)ꢀ
palladium (II) chloride (395 g, 56.3 mmol) was added and the slurry was again degassed by
vacuum/argon purge three times. The mixture was heated within 2 h to 85 °C and stirred for 30
27
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 32
1
2
3
4
5
6
7
8
9
min. The reaction mixture was cooled to rt, water (152 L) was added and the pH was adjusted to
6.0–8.0 with 37wt% aqueous hydrochloric acid solution (6.1 kg). The biphasic mixture was
heated to 80 °C and stirred for 1 h. The organic phase was separated and filtered over a preheated
pressure filter loaded with an activated carbon pad. The filter unit was washed with a warm (80
°C) mixture of nꢀpropanol (35.0 L) and water (18.5 L). The filtrate was concentrated under
reduced pressure while keeping the volume constant by addition of water (98 L). The resulting
slurry was cooled to 26–36 °C, filtered and rinsed with a mixture of nꢀpropanol (20.0 L) and
water (75.5 L). The cake was dried under reduced pressure at 45 °C to afford the crude product as
a yellowish white solid (15.1 kg, 81%).
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
API Crystallization. The crude product (15.7 kg) was charged to a reactor, followed by nꢀ
propanol (108 L) and water (56.0 L). The suspension was heated to 85 °C and stirred to afford a
clear solution. The solution was filtered over a preheated polishing filter unit and rinsed with a
hot mixture of nꢀpropanol (23.1 L) and water (11.0 L). The filtrate was cooled to −10 °C, aged
for 1 h and filtered. The filter cake was washed with cold nꢀpropanol (110 L) and dried under
reduced pressure at 60−70 °C to afford GDC-0980 as an offꢀwhite solid (14.5 kg, 92%): mp
o
1
232.8 C; H NMR (300 MHz, DMSOꢀd6) 9.15 (s, 2H), 7.07 (s, 2H), 4.86 (d, J = 7.0 Hz, 1H),
4.49–4.32 (m, 1H), 4.00–3.89 (m, 4H), 3.84 (s, 2H), 3.81–3.71 (m, 4H), 3.64 3.48 (m, 4H), 2.52
̶
̶
2.49 (m, 4H), 2.34 (s, 3H), 1.17 (d, J = 6.5 Hz, 3H); 13C NMR (75 MHz, DMSOꢀd6) 172.07,
164.04, 161.48, 157.76, 157.30, 156.36, 143.30, 129.59, 120.19, 110.91, 65.99, 64.03, 54.39,
52.89, 52.46, 45.80, 44.44, 40.33, 20.61, 10.95; HRMS calcd for C23H30N8O3S [M+H] 499.2234,
found 499.2220.
1
Supporting Information. Copies of H and 13C spectra for all the compounds listed in the
experimental section.
http://pubs.acs.org
This material is available free of charge via the Internet at
28
ACS Paragon Plus Environment

Page 29 of 32
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Acknowledgment. We thank Mr. Hong Lin and Ms. Stefanie Gee for analytical support, Dr.
Alan Deese for help with NMR analysis, and Dr. Christine Gu for HRMS analysis. We also
thank Dr. Reinhard Reents and Dr. Mark Reynolds for their valuable suggestions and support.
References
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) (a) Cantley, L. C. Science 2002, 296, 1655ꢀ1657. (b) Engelman, J. A. Nature Rev. Cancer
2009, 9, 550−562.
(2) Luo, J.; Manning, B. D.; Cantley L. C. Cancer Cell, 2003, 4, 257−262.
(3) Laplante, M.; Sabatini, D.M.J. Cell. Sci. 2010, 122, 3589–3594.
(4) (a) Maira, S. M.; Stauffer, F.; Brueggen, J.; Furet, P.; Schnell, C.; Fritsch, C.; Brachmann, S.;
Chene, P.; De Pover, A.; Schoemaker, K.; Fabbro, D.; Gabriel, D.; Simonen, M.; Murphy, L.;
Finan, P.; Sellers, W.; GarciaꢀEcheverria, C. Mol. Cancer Ther. 2008, 7, 185−1863. (b) Knight,
S. D.; Adams, N. D.; Burgess, J. L.; Chaudhari, A. M.; Darcy., M. G.; Donatelli, C. A.; Luengo,
J. I.; Newlander, K. A.; Parish, C. A.; Ridgers, L. H.; Sarpong, M. A.; Schmidt, S. J.; Van Aller,
G. S.; Carson, J. D.; Diamond, M. A.; Elkins, P. A.; Gardiner, C. M.; Garver, E.; Gilbert, S. A.;
Gontarek, R. R.; Jackson, J. R.; Kershner, K. L.; Luo, L.; Raha, K.; Sherk, C. S.; Sung, C.ꢀM.;
Sutton, D.; Tummino, P. J.; Wegrzyn, R. J.; Auger, K. R.; Dhanak, D. ACS Med. Chem. Lett.
2010, 1, 39–43. (c) Dehnhardt, C. M.; Venkatesan, A. M.; Santos, E. D.; Chen, Z.; Santos, O.;
AyralꢀKaloustain, S.; Brooijmans, N.; Mallon, R.; Hollander, I.; Feldberg, L.; Lucas, J.;
Chaudhary, I.; Yu, K.; Gibbons, J.; Abraham, R.; Mansour, T. S. J. Med. Chem. 2010, 53, 798–
810.
(5) Sutherlin, D.P.; Bao, L.; Berry, M., Castanedo, G.; Chuckowree, I.; Dotson, J.; Folkes, A. J.;
Friedman, L.; Goldsmith, R.; Gunzner, J.; Heffron, T.; Lesnick, J.; Lewis, C.; Mathieu, S.;
29
ACS Paragon Plus Environment