Development of an Efficient, Safe, and Environmentally Friendly Process for the Manufacture of GDC-0084

Article abstract of DOI:10.1021/acs.oprd.6b00011

An improved, efficient process with a significantly reduced process mass intensity (PMI) led to the multikilogram synthesis of a brain penetrant PI3K inhibitor GDC-0084. Highlights of the synthesis include a phase transfer catalyzed annulation in water, an efficient Suzuki-Miyaura cross-coupling of a chloropyrimidine with an arylboronic acid using a low palladium catalyst loading, and the development of a controlled crystallization to provide the API. The process delivered GDC-0084 with low levels of both impurities and residual metals.

Full text of DOI:10.1021/acs.oprd.6b00011

Subscriber access provided by NEW YORK UNIV
Full Paper
Development of an Efficient, Safe, and Environmentally
Friendly Process for the Manufacture of GDC-0084
Andreas Stumpf, Andrew McClory, Herbert Yajima,
Nathaniel Segraves, Remy Angelaud, and Francis Gosselin
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00011 • Publication Date (Web): 11 Mar 2016
Downloaded from http://pubs.acs.org on March 12, 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 30
Organic Process Research & Development
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

Organic Process Research & Development
Page 2 of 30
1
2
3
4
5
6
KEYWORDS. Phase Transfer Catalysis. Annulation. Safety Calorimetry. PdꢀCatalyzed Suzukiꢀ
Miyaura CrossꢀCoupling. PMI Reduction. Purine.
7
8
9
ABSTRACT. An improved, efficient process with a significantly reduced process mass intensity
(PMI) led to the multikilogram synthesis of a brain penetrant PI3K inhibitor GDC-0084.
Highlights of the synthesis include a phase transfer catalyzed annulation in water, an efficient
SuzukiꢀMiyaura crossꢀcoupling of a chloropyrimidine with an arylboronic acid using a low
palladium catalyst loading, and the development of a controlled crystallization to provide the
API. The process delivered GDC-0084 with low levels of both impurities and residual metals.
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
3
ACS Paragon Plus Environment

Page 3 of 30
Organic Process Research & Development
1
2
GDC-0084 (Figure 1)1, a potent and selective dual inhibitor of PI3K / mTOR with the ability to
penetrate the bloodꢀbrain barrier, has been discovered in our laboratories for the treatment of
various brain cancers.2 As part of the development program for this drug candidate, we required
multikilogram amounts of API to support human clinical studies.
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 GDC-0084 and ORTEP Representation
O
N
N
N
N
N
N
O
N
NH2
!
GDC-0084
The discovery chemistry synthesis of GDC-0084 served as the starting point for the endeavor
(Scheme 1). The route began with a selective THP protection of 2,6ꢀdichloroꢀ9Hꢀpurine 1,
followed by a regioselective substitution with morpholine to give purine 3. Subsequent
deprotonation with nꢀbutyllithium3 and trapping with acetone furnished tertiary alcohol 4, which
underwent an acidꢀmediated cleavage of the THP group to afford aminoalcohol 5. Alkylation
with 2ꢀbromoethyl acetate, followed by ester hydrolysis, then provided diol 6, which was
subjected to a SN1 ring closure to furnish cyclic ether 7. Finally, a microwaveꢀpromoted Suzuki
crossꢀcoupling with pinacolboronate 8 produced GDC-0084.
4
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 30
1
2
3
4
Scheme 1. Discovery Chemistry Synthesis of GDC-0084
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
At the outset we recognized that this route could be shortened by 2 steps if a direct annulation
between aminoalcohol 5 and a suitable vicinal electrophile could be achieved to forge fused
morpholine 7. Such a strategic change would also potentially lower the amount of an
unproductive elimination side product formed in the SN1 cyclization (9, vide infra). Additionally,
the use of chromatography at several steps, the presence of residual heavy metals in the API, and
the need for a robust API crystallization process all needed to be addressed in the development of
a scalable, phaseꢀappropriate route to GDC-0084.
5
ACS Paragon Plus Environment

Page 5 of 30
Organic Process Research & Development
1
2
3
4
First Generation Synthesis
5
6
7
8
9
In order to accelerate the development timelines, compounds 5 and 84 were selected as starting
materials and outsourced in advance of the first API batch. Compound 5 was manufactured on 30
kg scale in an overall yield of 91% and a purity of 99.5A% by HPLC. Each step of the proposed
synthesis was then optimized in order to improve the yields and enable efficient access to GDC-
0084 with high 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
Stage 1: Annulation of Aminoalcohol 5 to Form Fused Morpholine 7
Our first objective was to establish a direct formation of fused morpholine 7 from
aminoalcohol 5. Indeed, an initial screen of vicinal electrophiles5 identified 1,2ꢀdibromoethane as
a competent partner for annulation with 5 to construct alkylation product 7. The effect of solvent6
and base7 at 90 ºC8 were then examined in an effort to improve the conversion and minimize the
formation of elimination product 9 (Table 1). Conversion was found to be highest in CH3CN and
DMF (entries 1, 2, 5, and 6), with the latter solvent performing slightly better. Cesium carbonate
(entry 1) and potassium carbonate (entry 5) were found to function similarly in terms of
conversion, while the latter afforded a slightly lower amount of elimination side product 9. The
reaction was observed to be exothermic, thus doseꢀcontrol measures were investigated in order to
mitigate the safety risk. Slow addition of 1,2ꢀdibromoethane to a mixture of 5 and K2CO3 (entry
9) did not adversely affect the conversion but did lead to the formation of a new impurity, dimer
11, in 4A% (Figure 2). On the other hand, portionꢀwise addition of substrate 5 to a mixture of
1,2ꢀdibromoethane and K2CO3 (entry 10) suppressed the formation of 11 to <1.0A% without
negatively impacting conversion.
6
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 30
1
2
3
4
Table 1. Optimization of Annulation to Form Fused Morpholine 7
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
Entrya Base
Solvent 5 (A%)d 7 (A%)
9 (A%) 10 (A%)
1
2
Cs2CO3 DMF
1
3
86
84
82
79
87
81
56
36
85
85
13
14
13
15
12
11
7
0
0
Cs2CO3 CH3CN
Cs2CO3 MIBK
3
4
1
4
Cs2CO3
K2CO3
THF
5
1
5
DMF
1
0
6
K2CO3 CH3CN
K2CO3 MIBK
4
4
7
33
46
0
4
8
K2CO3
K2CO3
K2CO3
THF
DMF
DMF
7
11
0
9b
10c
11
11
0
0
aReactions conducted by mixing the reagents (1.0 mmol of 5) in sealed vials with stir bars at
ambient temperature in solvent (1.5 mL) and then heating. b1,2ꢀDibromoethane added dropꢀwise
over 1 h to the heated reaction mixture. c5 added in 2 portions to the heated reaction mixture (1 h
between additions). dHPLC area % after 12 h reaction time.
7
ACS Paragon Plus Environment

Page 7 of 30
Organic Process Research & Development
1
2
3
4
Figure 2. Impurity 11 Formed in Stage 1 Annulation Reaction
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
The safety of the exothermic annulation reaction was further examined through reaction
calorimetry. In the first experiment, the Advanced Reactive System Screening Tool (ARSST)
found a reaction exotherm of 45 ºC and an exothermic decomposition temperature (TD24
temperature at which timeꢀtoꢀmaximumꢀrate is 24 hours) of 185 ºC (Figure 3).
,
Figure 3. ARSST Self Heat Rate (Alkylation Reaction)
!
Exotherm
TD24
In the second calorimetry experiment, integration of the reaction heat flow curve in an
isothermal reaction calorimeter at 90 ºC showed an adiabatic temperature rise of 76.5 ºC, an
8
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 30
1
2
3
4
5
6
7
8
9
exotherm that was doseꢀcontrolled by portionꢀwise addition of 5 (Figure 4). Taken together, the
calorimetry data enabled a criticality index9 of 3 to be assigned to the reaction – a situation with
moderate risk that can be effectively mitigated by employing doseꢀcontrolled conditions.
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 4. RC1 Calorimetry (Alkylation Reaction)
With the Stage 1 reaction conditions in hand the isolation and purification of 7 were
investigated. Addition of water to the heterogeneous reaction mixture failed to induce
crystallization, resulting instead in oilingꢀout of the product. Extraction of the product into
EtOAc and water washes were employed in order to remove DMF. The EtOAc extract was then
solventꢀexchanged to 2ꢀpropanol 10 and a subsequent distillative crystallization successfully
provided 7. The optimized conditions for Stage 1 were ultimately applied to the alkylation of
5.83 kg of aminoalcohol 5, providing 4.27 kg of fused morpholine 7 in 67% yield and a purity of
97.8A% HPLC (9 at 0.40A% HPLC). There were four potential genotoxic impurities present in
the Stage 1 reaction mixture (vinyl bromide, 1,2ꢀdibromoethane, 9, and 10) and a GC method
was developed for their detection. All of these impurities were quantified to be <6 ppm
downstream in the isolated API11, 12
.
9
ACS Paragon Plus Environment

Page 9 of 30
Organic Process Research & Development
1
2
3
4
Stage 2: Cross-Coupling Reaction to Form Crude GDC-0084
5
6
The next objective was to improve the Suzuki crossꢀcoupling reaction between dichloropurine
7
8
9
713 and arylboronate 8 by lowering both the temperature and the catalyst loading. An initial
catalyst screen found PdCl2(dppf)•CH2Cl2 to be superior to PdCl2(PPh3)2, PdCl2(AmPhos), and
Pd(OAc)2 / SPhos14 in terms of conversion. The catalyst loading and solvent were then examined
more closely (Table 2). The catalyst loading was successfully lowered from 5 to 2 mol%, but
conversion was found to be sluggish at 1 mol% (entries 1–3). A brief solvent screen (entries 4–7)
using 2 mol% catalyst showed that aqueous THF was the better system, affording more than 98%
conversion after 2 h (entry 4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Optimization of CrossꢀCoupling Reaction to Form Crude GDC-0084
Entrya Pd (mol%)
Solventb
Conv. (A%)c
1
2
3
4
5
6
7
5
2
1
2
2
2
2
Dioxane, H2O
Dioxane, H2O
Dioxane, H2O
THF, H2O
99
99
88d
98
91
94
93
2ꢀMeTHF, H2O
CH3CN, H2O
2ꢀPropanol, H2O
10
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 30
1
2
aReactions conducted by mixing the reagents (0.9 mmol of 7) in solvent
(3.79 mL) at ambient temperature, evacuating and backfilling with nitrogen, and then heating the
sealed vials. bSolvent ratio 10:1. cAnalyzed by HPLC after 2 h reaction time. dAfter 10 h the ratio
of product/substrate was 99:1
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
Initial isolation of GDC-0084 from the reaction mixture was achieved by cooling to <10 ºC to
complete the precipitation of product, filtration, and washing the resulting filter cake with water.
The optimized reaction conditions for Stage 2 were successfully implemented using 4.10 kg of 7
to provide crude GDC-0084 containing impurity 1215 (0.49A% HPLC) that could be purged in
the subsequent crystallization step (Figure 5).
Figure 5. Impurity 12 Formed in Stage 2 CrossꢀCoupling Reaction
Stage 3 Purification of GDC-0084
The crude GDC-0084 from Stage 2 contained a large amount of residual palladium.
Representative samples containing 2400 ppm Pd were exposed to a variety of scavengers
(20wt% loading) in THF, HOAc (67:33)16 at 55 ºC for 14 h in an effort to reduce the level of
metal (Figure 6). Both SiꢀThiol17 and SiꢀThiourea were found to lower the residual Pd to ≤16
ppm12 in these gramꢀscale experiments, while on kilogramꢀscale a mixture of the 2 scavengers
was employed in order to ensure efficient removal of different Pd species from the API.
11
ACS Paragon Plus Environment

Page 11 of 30
Organic Process Research & Development
1
2
3
4
Figure 6. Pd Removal from Stage 2 Crude GDC-0084
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
In order to increase the solubility of GDC-0084 further experimentation with coꢀsolvents was
carried out (Figure 7)18. A screen of ternary solvent mixtures identified toluene, HOAc, water
(67:24:9) as the best system for solubility, which was further modified (69:30:1) for scaleꢀup.
Figure 7. Solubility of GDC-0084 in Ternary Solvent Systems at 50 ºC
12
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 30
1
2
3
4
5
6
7
8
9
Following filtration to remove the scavenger, GDC-0084 was crystallized by solvent switching
from the toluene, HOAc, water mixture to 2ꢀpropanol.19 A lowꢀlevel of acetylated impurity 13
(0.24A% HPLC) was formed due to the reaction between GDC-0084 and acetic acid during the
high temperature Pd scavenging step (Figure 8). The optimized conditions were ultimately
applied on multikilogram scale, wherein crude GDC-0084 obtained from crossꢀcoupling of 7
(4.10 kg scale) with 8 was scavenged for Pd in a mixture of acetic acid:toluene:H2O (4.3:1.8:0.1)
at 90 ºC. Filtration to remove scavengers and distillative crystallization from 2ꢀpropanol then
provided 3.87 kg of GDC-0084 in 80% yield and 99.4A% HPLC 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
Figure 8. Structure of Impurity 13 Formed During Crystallization
Summary First Generation GMP Synthesis
The synthesis of 3.87 kg of GDC-0084 to initiate clinical studies was successfully carried out
in three steps (Scheme 2). A direct annulation of aminoalcohol 5 with 1,2ꢀdibromoethane was
successfully implemented to replace the discovery chemistry stepꢀwise conversion of 5 to 7. The
subsequent Suzuki crossꢀcoupling reaction provided crude API, which was scavenged for Pd and
purified by distillative crystallization to furnish GDC-0084. Although effective for a first
delivery, some important areas for further improvement were identified. A second generation
synthesis was thus pursued in order to: (1) reduce the level of elimination side product 9 formed
13
ACS Paragon Plus Environment

Page 13 of 30
Organic Process Research & Development
1
2
3
4
in step 1; (2) identify a more efficient catalyst for step 2; and (3) minimize the formation of
5
6
acetamide impurity 13 in step 3.
7
8
9
Scheme 2. Optimized First Generation GMP Route to GDC-0084
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
Second Generation Synthesis
Stage 1 Annulation to Form 7
The wasteful production of side product 9, as well as the use of K2CO3 and DMF in the first
generation process, led to multiple processing steps with extensive workup, solvent exchanges,
and the use of multiple reactors. Therefore, a new stage 1 process using phase transfer catalysis
(PTC)20 in water was investigated for the synthesis of 7. Initial experiments to condense 5 with
1,2ꢀdibromoethane in mixtures of toluene and aqueous base in the presence of catalytic
tricaprylylmethylammonium chloride (aliquat® 336) at 60 ºC resulted in 50% conversion to 7. A
preliminary solvent screen was performed in chlorobenzene (partial conversion), THF, 2ꢀ
14
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 30
1
2
3
4
5
6
7
8
9
MeTHF, DCM, and DCE (no conversion) but they were all inferior to toluene, which better
solubilizes 5. Next aliquat® 336 was substituted with tetrabutylammonium bromide (TBAB). It
was discovered that the reaction went to completion in water without the presence of toluene. A
base screen using various inorganic bases (KOH, NaOH, K2CO3, and NaHCO3) was then
performed at 90 ºC and it was found that all bases showed similar reactivity. KOH was then
selected due to its highest aqueous solubility (121 g / 100 g @ 25 ºC). Optimization of the
reagent stoichiometry helped to drive the reaction to completion. Optimal conditions were found
with equivalent amounts of 1,2ꢀdibromoethane and KOH (300 mol% each, Table 3, entry 6 and
Figure 9) and a catalytic amount of tetrabutylammonium bromide (TBAB, 30 mol%) in water at
90 ºC for 17 hours. The use of excess base and 1,2ꢀdibromoethane was necessary due to the
competitive generation of vinyl bromide, which was detected in the reaction mixture but was not
detected downstream in the API by GC analysis (<6 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
Table 3: PTC Alkylation of 5: Reagent Stoichiometry Study
Entrya BrCH2CH2Br (mol%)
KOH (mol%)
TBAB (mol%) Conv. (A%)b
1
2
3
200
200
400
200
400
200
15
30
30
52
45
68
15
ACS Paragon Plus Environment

Page 15 of 30
Organic Process Research & Development
1
2
3
4
5
6
4
5
400
400
400
400
15
30
87
100
7
8
9
6
300
300
30
100
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
aReactions conducted by mixing the reagents (0.34 mmol of 5, KOH, TBAB, 1,2ꢀ
dibromoethane) with 10 mL of solvent at ambient temperature and then heating in vials with
vigorous stirring. 90 ºC, 17 h. bConversion was measured by HPLC analysis.
A study of reaction temperature then revealed that reduction of temperature from 90 ºC to 50
ºC (Figure 9) resulted in a better purity profile (80.9A% HPLC at 90 ºC, 96.0A% HPLC at 50
ºC).
Figure 9. Impact of Temperature on the Purity of 7
98.0
96.0
94.0
92.0
90.0
88.0
86.0
84.0
82.0
80.0
40
50
60
70
80
90
100
Deg C
Reactions conditions: 0.34 mol of 5 / L of water, TBAB (30 mol%), KOH (400 mol%), 1,2ꢀ
dibromoethane (400 mol%), reaction time 22 h.
In all of the PTC reactions the product separated from the aqueous layer as an oil which
contained significant amounts of 1,2ꢀdibromoethane. Addition of 2ꢀpropanol facilitated the
crystallization of the product. Although seeding slightly improved the crystallization, some
oiling out was still observed. This problem was resolved by replacing 2ꢀpropanol with ethanol.
16
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 30
1
2
3
4
5
6
7
8
9
The optimal ethanol:H2O ratio was found to be 1.3:1. The isolation of product 7 was achieved by
direct crystallization from the reaction mixture. The optimized reaction conditions were
demonstrated on 9.40 kg scale, delivering 6.85 kg of 7 in 67% yield and 98.5A% HPLC purity,
thus demonstrating the robustness of this optimized annulation process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Stage 2 Cross-Coupling Reaction to Form Crude GDC-0084
The main issues in the stage 2 process were the high catalyst loading (2 mol%
PdCl2(dppf)•CH2Cl2) for the Suzuki coupling and the large total solvent volume (58 vol).
Toward this end a catalyst screen was performed (Table 4) in THF and water. Experimental
conditions were similar to the first generation process (200 mol% of K3PO4•H2O and 133 mol%
of pinocolboronate 8) but the amount of THF was reduced from 13.6 to 8 volumes. Buchwald’s
second generation XPhos Pd G2 catalyst11 was found to be more active than PdCl2(dppf)•CH2Cl2
(Table 4, entry 1), while PdCl2(amphos) (entry 2) and PdCl2[(tꢀBu2Ph)P]2 (entry 3) were
completely inactive. In the case of Buchwald’s second generation XPhos Pd G2 catalyst, we
eliminated the use of exogenous XPhos ligand (entry 5–6) and the activity of the catalyst was
preserved. This was the lowest catalyst loading the reaction could tolerate as it did not go to
completion when the catalyst loading was further reduced from 0.5 to 0.3 mol%, even in the
presence of an additional equivalent of XPhos ligand (entry 7). Addition of water directly to the
reaction mixture led to the precipitation of the crude GDC-0084. We found that addition of water
at 50 ºC and subsequent aging for 1 h led to a faster filtration of the product. The optimized
reaction conditions are shown in Table 4.
17
ACS Paragon Plus Environment

Page 17 of 30
Organic Process Research & Development
1
2
3
4
Table 4: Pd Catalyst Screen for the Stage 2 Suzuki CrossꢀCoupling Reaction
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
Entrya Catalystc
Pd (mol%) Lig. (mol%) Base (mol%) Conv. (A%)b
1
PdCl2(dppf)•CH2Cl2
1
ꢀ
200
84
2
3
4
5
PdCl2(amphos)
PdCl2[(tꢀBu2Ph)P]2
XPhos Pd G2
5
ꢀ
200
200
200
200
0
0.5
0.5
0.5
ꢀ
0
XPhos/1
XPhos/0.5
100
100
XPhos Pd G2
6
7
XPhos Pd G2
XPhos Pd G2
0.5
0.3
ꢀ
200
200
100
32
XPhos/0.6
a Reactions were performed by mixing 7 (3.0 mmol), THF (8 mL), H2O (1.2 mL), K3PO4•H2O
(6.0 mmol, 200 mol%), and 8 (4.0 mmol, 133 mol%) in round bottom flasks, evacuating and
b
backfilling with nitrogen, and then heating at 65 º C for 5 h. Conversion determined by HPLC
analysis. c See ref 12.
The optimized Suzuki crossꢀcoupling was successfully demonstrated on 6.75 kg of 7 to provide
7.49 kg of crude GDC-0084 (94% yield, 99.4A% HPLC). Overall, replacing
PdCl2(dppf)•CH2Cl2 with XPhos Pd G2 catalyst enabled the catalyst loading to be lowered from
2 to 0.5 mol%.
18
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 30
1
2
3
4
Stage 3 Purification of GDC-0084
5
6
7
8
The main issue with the first generation Stage 3 purification was the formation of acetamide 13
in quantities that approached the maximum allowed level in the API. In order to better
understand the formation of 13, a stress test was performed in which GDC-0084 was treated with
acetic acid, toluene, and water under simulated first generation process conditions. A rapid
increase of the acetamide impurity 13 in the reaction mixture was observed (up to 1.4A% HPLC)
after 30 h of heating at 90 ºC (Figure 10).
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 10. Stability of GDC-0084: Formation of Acetamide 13
Reaction conditions: GDC-0084 (2.6 mmol) in toluene (1.6 mL), HOAc (3.6 mL), H2O (0.01
mL) at 90 ºC.
In the next set of experiments toluene was removed from the process, and the acetic acid to
water ratio was decreased from 98:2 to 1:1 at 90 ºC (Figure 11). It was found that when the
amount of water was increased, the amount of acetamide formed after 6 h was reduced 6ꢀfold
from 0.6% to 0.1%. Having now a good control on the acetamide impurity formation, we
19
ACS Paragon Plus Environment

Page 19 of 30
Organic Process Research & Development
1
2
3
4
5
6
investigated the solubility of the API in acetic acid, water mixtures in order to run the Pd
scavenging step.
7
8
9
Figure 11. Acetamide 13 Formation in HOAc, Water Mixtures at 90 ºC
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
Reaction conditions: GDC-0084 (2.6 mmol) in solvent mixture (10 mL) at 90 ºC.
By further fineꢀtuning the acetic acid:water ratio it was found that GDC-0084 fully dissolved
in 10 vol of acetic acid:water (3:1) at 90 ºC and crystallized out at 60 ºC. The 30 ºC temperature
width of the metastable zone21 was deemed to be sufficient to perform the required polish
filtration at 90 ºC. Because the Pd loading was reduced in Stage 2 from 2 to 0.5 mol%, it was
found that treatment with only 10wt% of SiꢀThiol was sufficient to reduce residual Pd to below
10 ppm. The final optimized procedure was demonstrated on 7.70 kg scale, producing 6.41 kg of
GDC-0084 in 99.70A% HPLC and 83% yield.
20
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 30
1
2
3
4
Summary of Second Generation Synthesis
5
6
7
8
The second generation synthesis produced 6.41 kg of GDC-0084 in 52% yield over 3 steps and
99.70A% (Scheme 3). The sequence involved an efficient annulation of 5 with 1,2ꢀ
dibromoethane under PTC conditions in water to provide 7 with a subsequent Suzuki crossꢀ
coupling with boronate 8 using only 0.5 mol% of XPhos Pd G2, to provide the crude API. A
final scavenging / recrystallization from acetic acid / water provided GDC-0084 with the
required purity and polymorphic form. The acetamide impurity 13 was reduced from 0.25 to
<0.05A% HPLC by adjusting the solvent composition for crystallization.
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
Scheme 3. Optimized Second Generation Route to GDC-0084
In stage 1, both DMF and EtOAc were eliminated from the process, and the total solvent
volume was reduced by 54%. Furthermore, we eliminated all extractions and solvent exchanges
from this step, and the total unit operations were reduced from 21 to 7. In stage 2, the total
solvent volume was reduced from 58 to 17 vol (71% reduction). In stage 3, toluene was removed
21
ACS Paragon Plus Environment

Page 21 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
from the process and the total solvent volume for recrystallization was lowered from 33 to 21 vol
(37% reduction). Overall, the total solvent volume for the three stages was reduced by 64%
(Figure 12), and as a result the process mass intensity (PMI)22 was reduced from 140 to 70,
which is in the practical range for a commercial process23.
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 12. Elimination and Reduction of Solvents for the Manufacture of GDC-0084
2"Propanol*(a)*
EtOAc*(a)*
EtOH%(a)%
HOAc%(a)%
HOAc*(a)*
THF*(b)*
THF%(b)%
Toluene*(b)*
DMF*(b)*
First Generation Process
504 L Solvent
Second Generation Process
170 L Solvent
aICH class 3 solvent24. bICH class 2 solvent.
Conclusion
In summary, we have developed a streamlined, efficient, and safe synthesis of GDC-0084 that
features a PTC alkylation, Suzuki crossꢀcoupling, and controlled crystallization. Based on the
PMI analysis, the second generation synthesis shows significant improvement relative to the first
generation route and has been demonstrated to produce API on multikilogram scale with
excellent purity.
Experimental Section
General. Unless otherwise noted all reactions were run under a nitrogen atmosphere, solvents
1
and reagents were without further purification. H NMR (500 MHz), 13C NMR (125 MHz) were
22
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 30
1
2
3
4
5
6
7
8
9
recorded on a Bruker Avance 3 spectrometer. Chemical shifts are reported in ppm (δ units)
downfield of internal tetramethylsilane [(CH3)4Si] or residual CHCl3; coupling constants are
reported in hertz (Hz). Multiplicities are as follows: s = singlet, d = doublet, t = triplet, q =
quartet, dd = doublet of doublets, m = multiplet. Melting points were measured by differential
scanning calorimetry (DSC, Büchi Bꢀ540). In process methods for analysis of 5, 7 and GDC-
0084. Diluent: acetonitrile. Mobile phase A: 0.05% TFA/water. Mobile phase B: 0.05%
TFA/acetonitrile. Column: Ace 3 C18 HL column, 3x50 mm 3.0 µm. Column temperature: 35
°C. Detector wavelength: 220 nm. Injection volume: 2 ꢁL. Flow rate: 1 mL/min. Sample
Concentration: 0.5ꢀ1.0 mg/mL in 50% acetonitrile/water. Program: 0.0 min 5.0% B, 0.3 min
5.0% B, 2.0 min 60.0% B, 4.0 min 90% B, 5.0 min 90% B, 5.1 min 5.0% B, 6.5 min 5.0%.
Typical retention times: 5 (RT 2.61 min), 7 (RT 2.20 min), GDC-0084 (RT 2.67 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
2-Chloro-6,6-dimethyl-4-morpholino-8,9-dihydro-6H-[1,4]oxazino[4,3-e]purine
7
(first
generation process). A 100 L reactor was charged with DMF (20.00 L, 3.43 vol), 1,2ꢀ
dibromoethane (7.36 kg, 39.18 mol, 200 mol%), potassium carbonate (6.76 kg, 48.91 mol, 250
mol%), and the first portion of 5 (3.01 kg, 10.11 mol, 52 mol%). The mixture was heated to 103
ºC, held at that temperature for 2 h, and then cooled to 50 ºC. HPLC analysis showed 5 to be
<0.05A%, and 10 to be 3.91A%. The second portion of 5 (2.82 kg, 9.47 mol, 48 mol%) was
charged to the reactor. The mixture was heated to 85 ºC, held at that temperature for 16 h, then
cooled to 53 ºC. HPLC analysis showed 5 and 10 to be <0.05A%. The reaction was cooled to
21 ºC, and then ethyl acetate (19.50 L) and purified water (40.00 L) were added. The aqueous
layer was removed, and then more ethyl acetate (19.50 L) and purified water (20.00 L) were
added to the reactor. The aqueous phase was removed. GC analysis showed DMF to be
0.67wt% in the organic solution. This solution was transferred, along with ethyl acetate rinse
23
ACS Paragon Plus Environment

Page 23 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(1.0 L), from the 100 L reactor to a 50 L reactor. Distillation was carried out to minimum stir
volume (9 L), and then 2ꢀpropanol (12.00 L) was charged to the reactor. Distillation was again
carried out to minimum stir volume (9 L), and then 2ꢀpropanol (30.00 L) was added. GC
analysis showed ethyl acetate to be 0.06wt%. The 2ꢀpropanol suspension (39 L) was then heated
to 65 ºC, held at that temperature for 2 h, and then cooled to 5 ºC over 1 h. The mixture was held
for 2 h. HPLC analysis indicated the product concentration in supernatant to be 14 mg / g. The
mixture was filtered on an Aurora filter and the cake was washed with 2ꢀpropanol (16.00 L).
After no more filtrate could be collected from the filter, the cake was dried on the filter at 50 ºC
(jacket temperature) under house vacuum with a nitrogen purge, until GC analysis showed 2ꢀ
propanol to be 0.91wt%. 2ꢀChloroꢀ6,6ꢀdimethylꢀ4ꢀmorpholinoꢀ8,9ꢀdihydroꢀ6Hꢀ[1,4]oxazino[4,3ꢀ
e]purine 7 was obtained as a light yellow solid (4.27 kg, 67% yield, 97.8A% HPLC).
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-Chloro-6,6-dimethyl-4-morpholino-8,9-dihydro-6H-[1,4]oxazino[4,3-e]purine 7 (second
generation process). A 100 L reactor was charged with water (45.10 kg, 4.8 vol), potassium
hydroxide (5.40 kg, 96.24 mol, 300 mol%), 5 (9.40 kg, 31.57 mol, 100 mol%), tetraꢀnꢀ
butylammonium bromide (2.92 kg, 9.06 mol, 29 mol%), and 1,2ꢀdibromoethane (17.70 kg, 94.22
mol, 298 mol%). The mixture was heated to 47 ºC and then held at that temperature for 20 h.
HPLC analysis showed 5 to be 3.1A%. Ethanol (46.2 kg, 6.2 vol) and 7 seeds (96.6 g, 1wt%) in
ethanol (0.60 kg, 0.06 vol) were added. The contents of the reactor were aged for 2 h, then
cooled to 5 ºC over 2 h and aged for 1 h. The mixture was transferred into a filter dryer and the
filter cake was washed with water (27.4 kg, 2.9 vol). After no more filtrate could be collected
from the filter, the cake was dried on the filter at 60 ºC (jacket temperature) under house vacuum
with a nitrogen purge. 2ꢀChloroꢀ6,6ꢀdimethylꢀ4ꢀmorpholinoꢀ8,9ꢀdihydroꢀ6Hꢀ[1,4]oxazino[4,3ꢀ
e]purine 7 was obtained as a light yellow solid (6.85 kg, 67% yield, 98.5A% HPLC, all
24
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 30
1
2
3
4
5
6
impurities were detected at low levels: 5 at 0.43A% HPLC, 9 at 0.22A%, 10 at 0.23A%, 11 at
0.10A%25): mp 147 ºC; 1H NMR (500 MHz, DMSOꢀd6) δ 4.75ꢀ4.00 (m, 8H), 3.73ꢀ3.71 (m, 4H),
7
13
8
9
1.56 (s, 6H); C NMR (125 MHz, DMSOꢀd6) δ 152.9, 151.7, 151.6, 151.5, 117.3, 73.6, 66.0,
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
57.6, 45.3, 41.6, 27.2. HRMS [M+H]+ calcd for C14H18ClN5O2 324.1222; found 324.1225.
5-(6,6-Dimethyl-4-morpholino-8,9-dihydro-6H-[1,4]oxazino[4,3-e]purin-2-yl)pyrimidin-2-
amine GDC-0084 (first generation process). A solution of K3PO4•H2O (5.84 kg, 24.80 mol,
196 mol%) in purified water (8.45 L) was sparged with N2 for 1 h. A 100 L reactor was charged
with the aqueous K3PO4 solution, 7 (4.10 kg, 12.66 mol, 100 mol%), 8 (3.36 kg, 15.20 mol, 120
mol%), and THF (55.80 L). The contents were sparged with N2 for 45 min, and then
PdCl2(dppf)•CH2Cl2 (0.207 kg, 0.253 mol, 2 mol%) was charged. The contents were sparged
with N2 for 10 min. The mixture was heated to 61 ºC, and then held at that temperature for 4 h.
HPLC analysis showed 7 to be 0.027 mg / mL. Purified water (45.80 L) was charged. The
reaction was cooled to 7 ºC, then held at that temperature for 1 h. The mixture was filtered and
the cake was washed with purified water (4 x 30.00 L). After no more filtrate could be collected
from the filter, the cake was dried on the filter at 70 ºC (jacket temperature) under house vacuum
with a nitrogen purge. Drying was continued for 8 h. HPLC analysis showed 12 to be 0.4A%.
Crude GDC-0084 was transferred to a 100 L reactor, along with SiꢀThiol (2.22 kg), SiꢀThiourea
(2.22 kg), acetic acid (17.50 L), toluene (7.50 L), and purified water (0.25 L). The mixture was
heated to 90 ºC and held at that temperature for 3 h. The mixture was transferred, along with
acetic acid (7.50 L) / toluene (3.50 L) rinses (2 x), to a filter, and the hot filtrate (~70 ºC) was
passed through an inꢀline filter directly into a 50 L reactor. Metal analysis showed residual Pd to
be ≤3 ppm. Distillation was carried out to minimum stir volume (10 L), and then the contents
were heated to 70 ºC. 2ꢀPropanol (40.00 L) was charged to the reactor through an inꢀline filter
25
ACS Paragon Plus Environment

Page 25 of 30
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(polish filtration). The suspension was heated to 71 ºC, held at that temperature for 1 h, and then
cooled to 18 ºC over 4 h. The mixture was held for 9 h. HPLC analysis indicated the product
concentration in supernatant to be 1.5 mg / g. The mixture was transferred into a filter dryer and
the filter cake was washed with 2ꢀpropanol (48.00 L). After no more filtrate could be collected
from the filter, the product was dried at 70 ºC (jacket temperature) under house vacuum with a
nitrogen purge. GDC-0084 was obtained as a light yellow solid (3.87 kg, 80% yield, 99.4A%
HPLC, impurities: 12 at 0.08A% HPLC, 13 at 0.24A% HPLC).
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-(6,6-Dimethyl-4-morpholino-8,9-dihydro-6H-[1,4]oxazino[4,3-e]purin-2-yl)pyrimidin-
2-amine GDC-0084 (second generation process). A 100 L reactor was charged with water
(8.00 kg, 1.2 vol), THF (39.20 kg, 6.5 vol), K3PO4•H2O (9.51 kg, 40.42 mol, 194 mol%), 7 (6.75
kg, 20.85 mol, 100 mol%), and 8 (5.50 kg, 24.88 mol, 119 mol%). The reactor was cycled from
vacuum to nitrogen 3 times, and then XPhos Pd G2 (82.0 g, 0.104 mol, 0.5 mol%) was charged.
The mixture was degassed by vacuumꢀnitrogen purge cycles (3x). The mixture was heated to 67
ºC for 5 h. HPLC analysis indicated completed conversion. Purified water (48.10 kg, 7.1 vol)
was charged and aged for 1 h at 50 ºC. The reaction was cooled to 20 ºC over 2 h, held for 2 h at
20 ºC, then cooled to 5 ºC and held at that temperature for 2 h. The mixture was transferred into
a filter dryer and the filter cake was washed with water (37.10 kg, 5.5 vol). After no more
filtrate could be collected from the filter, the cake was dried on the filter at 60 ºC (jacket
temperature) under house vacuum with a nitrogen purge. Crude GDC-0084 was obtained as an
offꢀwhite solid (7.49 kg, 94% yield, 99.4A% HPLC).
Recrystallization
of
crude
5-(6,6-dimethyl-4-morpholino-8,9-dihydro-6H-
[1,4]oxazino[4,3-e]purin-2-yl)pyrimidin-2-amine GDC-0084 (second generation process). A
100 L reactor was charged with water (7.75 kg, 1 vol), acetic acid (60.80 kg, 7.5 vol), crude
26
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 30
1
2
3
4
5
6
7
8
9
GDC-0084 (7.70 kg, 20.13 mol, 100 mol%), and SiꢀThiol (770 g, 10wt%). The mixture was
heated to 90 ºC and then held at that temperature for 3 h. The contents were filtered through an
Aurora filter to remove the metal scavenger and then through a 1 ꢁm inꢀline filter, and the
Aurora filter was rinsed with hot acetic acid (7.10 kg, 0.9 vol). The solution was then cooled to
77 ºC and GDCꢀ0084 seeds (82 g, 1.1wt%) were added as a slurry in acetic acid (69 g) and water
(87 g). The contents were aged for 1 h at 68 ºC. Purified water (12.00 kg, 1.6 vol) was charged
and the contents cooled to 45 ºC, aged at 45 ºC for 1 h, cooled to 20 ºC over 2 h, aged at 20 ºC
for 6 h, cooled to 5 ºC over 2 h, and held for 2 h at 5 ºC. The mixture was transferred into a filter
dryer and the cake was washed with water (69.90 kg, 9.1 vol). After no more filtrate could be
collected from the filter, the cake was dried on the filter at 60 ºC (jacket temperature) under
house vacuum with a nitrogen purge. GDC-0084 was obtained as an offꢀwhite solid (6.41 kg,
83% yield, 99.70A% HPLC): mp 211 ºC; 1H NMR (500 MHz, DMSOꢀd6) δ 9.09 (s, 2H), 7.03 (s,
2H), 4.32ꢀ4.17 (m, 4H), 4.17ꢀ4.04 (m, 4H), 3.84ꢀ3.65 (m, 4H), 1.58 (s, 6H); 13C NMR (125
MHz, DMSOꢀd6) δ 163.8, 157.6, 154.2, 152.5, 151.3, 151.0, 120.3, 117.3, 73.7, 66.2, 57.8, 45.2,
41.5, 27.3. HRMS [M+H]+ calcd for C18H22N8O2 383.1938; found 383.1945.
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. Spectroscopic data are available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Eꢀmail: stumpf.andreas@gene.com
*Eꢀmail: mcclory.andrew@gene.com
27
ACS Paragon Plus Environment

Page 27 of 30
Organic Process Research & Development
1
2
3
4
Author Contributions
5
6
7
8
9
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript. ‡These authors contributed equally.
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
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors would like to thank the following Genentech colleagues: Haiming Zhang for
document review and helpful suggestions, Christine Gu for HRMS analysis, Jeffrey Stults for
obtaining Xꢀray data of GDC-0084, and Keena Green for melting point analysis.
REFERENCES
1 The structure of GDC-0084 was unambiguously confirmed by Xꢀray analysis. See supporting
information for data.
2
The Discovery of Clinical Development Candidate GDCꢀ0084, a Brain Penetrant Inhibitor of
Class I Phosphoinositide 3ꢀKinases (PI3K) and mTOR. Heffron, T; Ndubaku, C.; Salphati, L.;
Alicke, B.; Cheong, J.; Drobnick, J.; Edgar, K.; Gould, S.; Lee, L.; Lesnick, J.; Lewis, C.;
Nonomiya, J.; Pang, j.; Plise, E.; Sideris, S.; Wallin, J.; Wang, L.; Zhang, X.; Olivero, A., ACS
Med. Chem. Lett. 2016, DOI: 10.1021/acsmedchemlett.6b00005.
3 (a) Purine Derivatives Useful as PI3 Kinase Inhibitors. Goldsmith, P. Hancox, T. C.; Hudson,
A.; Pegg, N. A.; Kulagowski, J. J.; Nadin, A. J.; Price, S.PCT Int. Appl. WO 2009053716 A1
Apr 30, 2009. (b) Preparation of Purine Derivatives with PI3K Inhibitory Activity and Methods
of Use Thereof. Castanedo, G.; Chuckowree, I.; Folkes, A.; Sutherlin, D. P.; Wan, N. C. PCT Int.
Appl. WO 2009146406 A1 Dec 03, 2009.
4 Available from Boron Molecular, Inc. (Catalog #: BM382, CAS #: 402960ꢀ38ꢀ7).
5 Reaction with 1,2ꢀdichloroethane did not proceed, while using 1ꢀbromoꢀ2ꢀchloroethane led to
extensive buildꢀup of the corresponding monoalkylated alkyl chloride intermediate.
6 NMP and DMSO were also investigated but were inferior in terms of both reaction rate and
impurity formation.
28
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 30
1
2
3
4
5
7 Other bases (K3PO4, EtONa and DBU) were also investigated but gave inferior results.
6
7
8
8 The reaction was studied at lower temperatures (50 ºC and 70 ºC) in an effort to increase the
selectivity, but no improvement was observed.
9
9 For a recent example of safety assessment using criticality index: (a) Abele, S.; Uppelli, S. B.;
Schwaninger, M.; Fierz, H.; Funel, J. A.; Stossel, F. Org. Process Res. Dev. 2012, 16, 2015–
2020.
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
10 A dried representative crude mixture of 7 and 9 was subjected to trituration with various
solvents, and 2ꢀpropanol was found to achieve the highest purging of 9.
11
For more information on genotoxic impurities: Snodin, D. J. Org. Process Res. Dev.
2011, 15, 1243–1246.
12 ICH Q7 Guideline: Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients
Current version dated 10 June 2015
13 Further examples for SuzukiꢀMiyaura crossꢀcouplings of chloropyrimidines: (a) Faldu, V. J.;
Talpara, P. K.; Shah, V. H. Tetrahedron Lett. 2014, 55, 1456–1460. (b) Mao, S. L.; Sun, Y.; Yu,
G. A.; Zhao, C.; Han, Z. J.; Yuan, J.; Zhu, X.; Yang, Q.; Liu, S. H. Org. Biomolecular
Chem. 2012, 10, 9410–9417.
14
SPhos = 2ꢀDicyclohexylphosphinoꢀ2′,6′ꢀdimethoxybiphenyl, PdCl2(dppf)•CH2Cl2 = [1,1′ꢀ
Bis(diphenylphosphino)ferrocene]dichloropalladium(II) CH2Cl2 (1:1), PdCl2(amphos) = Bis(diꢀ
tertꢀbutyl(4ꢀdimethylaminophenyl)phosphine)dichloropalladium(II),
PdCl2[(tBu2Ph)P]2
=
Dichlorobis(diꢀtertꢀbutylphenylphosphine)palladium(II), XPhos Pd G2
dicyclohexylphosphinoꢀ2′,4′,6′ꢀtriisopropylꢀ1,1′ꢀbiphenyl)[2ꢀ(2′ꢀaminoꢀ1,1′ꢀ
biphenyl)]palladium(II):
=
Chloro(2ꢀ
XPhos = 2ꢀDicyclohexylphosphinoꢀ2′,4′,6′ꢀtriisopropylbiphenyl.
15 Presumably formed by oxidative dimerization of pinacolboronate 8.
16 The poor solubility of GDC-0084 in organic solvents such as dichloromethane, toluene, ethyl
acetate precluded the use of aqueous scavenger washes. Additionally, acetic acid and elevated
temperature both proved to be critical for obtaining good solubility of GDC-0084.
29
ACS Paragon Plus Environment

Page 29 of 30
Organic Process Research & Development
1
2
3
4
5
17 SiꢀThiol and SiꢀThiourea are proprietary solidꢀsupported resins available from Silicycle.
6
7
8
18 Adding 10 % water to the organic solvent / HOAc mixtures increased the solubility of GDCꢀ
0084.
9
19 The solubility of pure GDC-0084 in 2ꢀpropanol at 23 ºC was measured to be 3 mg / g.
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
20 Further examples of PTC reactions: (a) Solanki, P. V.; Uppelli, S. B.; Pandit, Bhushan S.;
Mathad, V. T. Org. Process Res. Dev. 2014, 18, 342–348. (b) Tahira, S.; Vineet, K.; Satya, P.
Syn. Comm. 2014, 44, 620ꢀ632. (c) Attarian, O. S.; Matsoyan, S. G.; Martirosyan, S. S.
Chemistry of Heterocyclic Compounds (New York, NY, United States) 2005, 41, 452–455.
o
21 The solubility of GDCꢀ0084 in acetic acid:water (3:1) was found to be: 14 mg/mL at 50 C, 2
mg/mL at 20 oC, 0.5 mg/mL at 5 oC.
22 Concepcion, J.; Ponder, C. S.; Broxterman, Q. B. ; Manley, J. B. Org. Process Res. Dev. 2011,
15, 912−917.
23
ACS GCIPR. Henderson, R. K.; Kindervater, J.; Manley, J. B. Lessons Learned through
Measuring Green Chemistry Performance: The Pharmaceutical Experience. American Chemical
Society, Green Chemistry Institute, Pharmaceutical Roundtable: www.acs.org/greenꢀ chemistry,
2008.
24
ICH Guideline Q3C (R5) on Impurities, Guideline for Residual Solvents, Current version
dated September 2015.
25 In contrast to the first generation conditions, RC1 calorimetry showed the second generation
PTC conditions to be only mildly exothermic. Vinyl bromide is generated in the synthesis and
therefore the exhaust gases are scrubbed in cold ethanol. The following genotoxic impurities
have been analyzed for in GDC-0084 but were not detected: vinyl bromide, 1,2ꢀdibromethane, 9,
and 10.
30
ACS Paragon Plus Environment