Improved Synthesis of the Nav1.7 Inhibitor GDC-0276 via a Highly Regioselective SNAr Reaction

Subscriber access provided by GUILFORD COLLEGE

Full Paper

Improved Synthesis of the Nav1.7 Inhibitor

GDC-0276 via a Highly Regioselective SNAr Reaction

Andreas Stumpf, Zhigang Cheng, Danial Beaudry, Remy Angelaud, and Francis Gosselin

Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00082 • Publication Date (Web): 31 Jul 2019

Downloaded from pubs.acs.org on July 31, 2019

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted

online prior to technical editing, formatting for publication and author proofing. The American Chemical

Society provides “Just Accepted” as a service to the research community to expedite the dissemination

of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in

full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully

peer reviewed, but should not be considered the official version of record. They are citable by the

Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,

the “Just Accepted” Web site may not include all articles that will be published in the journal. After

a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web

site and published as an ASAP article. Note that technical editing may introduce minor changes

to the manuscript text and/or graphics which could affect content, and all legal disclaimers and

ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or

consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

Washington, DC 20036

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 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

Improved Synthesis of the Nav1.7 Inhibitor GDC-0276 via a Highly Regioselective S

N

Ar Reaction

Andreas Stumpf,^*^,^†Zhigang Ken Cheng, Danial Beaudry, Remy Angelaud and Francis Gosselin

†

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

^†Department of Small Molecule Process Chemistry, Genentech, Inc.,1 DNA Way, South San Francisco, CA 94080,

USA

ACS Paragon Plus Environment

1

Organic Process Research & Development

Page 2 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

TOC

First

Generation

F

t

O Bu

Cl

O

F

O

F

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

H

Regioselective S Ar

N

OH

N

S

O

O O

O

GDC-0276

F

Second

Generation

Cl

ACS Paragon Plus Environment

2

Page 3 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

ABSTRACT

The development of a redesigned and improved second generation synthesis of the Nav1.7 inhibitor GDC-0276

based on experience gained from a fit-for-purpose first generation synthesis will be described. The first generation

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

synthesis proceeded via a regioselective S

difluorobenzoate with 1-adamantanemethanol. In the newly developed second generation synthesis, the much

improved regioselective S Ar reaction was performed on the readily available starting material 1-chloro-2,4-

N

Ar reaction on the advanced starting material t-butyl 5-chloro-2,4-

N

difluorobenzene, followed by installation of the carboxylate group by electrophilic aromatic bromination and a

palladium-catalyzed alkoxycarbonylation. A subsequent Suzuki-Miyaura cross-coupling reaction was then

telescoped directly into a phase transfer catalyzed (PTC) ester hydrolysis. Amidation of the resulting acid

intermediate with 1-azetidine sulfamide in turn provided GDC-0276 in high overall yield and purity on 100

kilogram scale.

KEYWORDS. Nav1.7 inhibitor, S

N

Ar, Suzuki-Miyaura cross-coupling, palladium-catalyzed alkoxycarbonylation,

phase transfer catalysis (PTC)

Introduction

Nav1.7 is a voltage gated sodium channel expressed in pain sensing nerve C-fibers and plays an essential role in

the generation of action potentials and the sensation of pain. The Nav1.7 protein complex has been validated through

1

human genetics, and become a valuable small molecule drug discovery target for the treatment of pain. GDC-0276

2

was developed in our laboratories as a selective small molecule Nav1.7 inhibitor to provide novel and improved

treatment options for pain, and to address shortcomings of existing pain medications, such as addiction and off-

target side effects.

Results and Discussion

First Generation Process

ACS Paragon Plus Environment

3

Organic Process Research & Development

Page 4 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

In order to quickly deliver the first kilogram amounts of active pharmaceutical ingredient (API) to initiate clinical

trials, we set out to develop a first generation process for GDC-0276 based on the retrosynthetic analysis illustrated

in Figure 1. We envisioned that GDC-0276 could be synthesized by amidation of benzoic acid 2 with sulfamide 3.

Suzuki-Miayura cross-coupling with cyclopropylboronic acid (CPBA), and subsequent hydrolysis of the t-butyl

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

N

ester would then furnish penultimate intermediate 2. S Ar reaction of 1 with 1-adamantanemethanol would

regioselectively deliver the C-4 substituted ether because of steric hindrance at C-2 caused by the t-butyl ester.

Figure 1. Retrosynthesis of GDC-0276

S Ar

N

H N

N

2

OH

S

O

F

3

O

F

O O

H

4

2

1

N

t

OH

O Bu

S

Cl

O

O O

B(OH)₂

O

1

2

GDC-0276

We started with esterification of 5-chloro-2,4-difluorobenzoic acid with Boc

DMAP to give the corresponding t-butyl ester 1 (Scheme 1) as oil in quantitative yield and 97.0 A% HPLC purity

after workup. Next, ester 1 was subjected to a regioselective S Ar reaction with 1-adamantanemethanol in DMSO in

the presence of Cs CO to give the expected intermediate 5 from fluoride displacement at C-4.

Scheme 1. Synthesis of Intermediate 5

2

O in THF and a catalytic amount of

N

2

3

1) 1-adamantane-

methanol (96 mol%)

DMSO, Cs CO (191 mol%)

1) Boc O (168 mol%),

F

2

F

2

3

THF, DMAP (11 mol%)

72 °C

t

OH

O Bu

Cl

2

) EtOAc,

2) EtOAc, aq. HCl

3) MeOH

O

4

aq. NaHCO₃

) aq. HCl

1

3

(32 % yield over 2 steps)

9

7.0 A% HPLC

(quant. yield)

O

F

O

F

O

F

O

t

O Bu +

t

+

O Bu

Cl

O

5

6

7

13.8 A% HPLC

3.4 A% HPLC

86.2 A% HPLC

However, under these conditions, significant amounts of adamantyl ester 6 (10─15 A% HPLC) and regioisomer 7

(

3─5 A% HPLC), resulting from substitution of fluoride at C-2, were also generated as major side products.³

ACS Paragon Plus Environment

4

Page 5 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

2 3 2 3

We evaluated a number of bases (Cs CO , K CO , t-BuOK, LiHMDS) and solvents (DMF, DMSO, THF, THF /

DMF, THF / DMSO), but this did not result in any improved overall reaction profile. We therefore proceeded to run

4

the reaction with Cs

2

CO

3

in DMSO; crude 5 could be isolated as a solid in 32% corrected yield and 86.2 A% HPLC

purity after work up with EtOAc and aqueous HCl followed by concentration of the organic layer. Although

undesired regioisomer 7 could be easily removed by trituration with MeOH down to <0.2 A% HPLC, adamantyl

ester 6 could not be purged (13.6 A% HPLC) at this stage.

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

We next investigated the Suzuki-Miyaura cross-coupling of 5 with cyclopropylboronic acid (CPBA), in toluene /

water with K

3

PO

4

as a base, and initiated a catalyst screen (Table 1). A Buchwald second generation palladium

5

precatalyst G2-PdXPhos was identified as the best catalyst from this screen (Table 1, entries 3─5), and the ratio of

G2-PdXPhos / XPhos could be reduced from 1:2 (Table 1, entry 3) to 1:1 without compromising reaction conversion

(

Table 1, entry 4) at 80 °C. However, the stability of CPBA was subsequently examined using differential scanning

calorimetry (DSC) and it was found to decompose at 169 °C (see the Supporting Information). Thus, for safety

reasons, the Suzuki reaction was run at <70 °C. The catalyst loading was further optimized down to 0.5 mol% from

2

mol%, but required an excess of CPBA (140 mol%) for the reaction to go to completion at 65 °C (Table 1, entry

6).

Table 1. Optimization of Suzuki Cross-Coupling of 5 with CPBA

Catalyst,

K PO (300 mol%)

3 4

O

F

O

F

O

F

OH

+

B

t

O Bu

+

O

Cl

toluene (7V),

O

H O (2.4V)

2

O

5

9

8

5

CPBA

Catalyst

(2 mol%)

XPhos Conversion

Entry^a

1

b

(

mmol) (mol%)

(mol%)

N/A

(A%)

0

0.50

1.00

5.00

140

120

140

PdCl

2

(PPh

3

)

2

3

4

Pd(OAc)

2

4

95

100

98

G2-PdXPhos^c

G2-PdXPhos

4

2

5

G2-PdXPhos

4

₆d

_G₂_-_P_d_X_P_h_o_sf

253.00

0.5

100

^aReactions conducted by mixing 5 and CPBA in toluene (7V), and adding K

catalyst and ligand at rt, evacuating and backfilling with N

PO

(300 mol%) in water (2.4V),

3

4

b

2

, and then heating to 80ºC for 14 h. Conversion was

ACS Paragon Plus Environment

5

Organic Process Research & Development

Page 6 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

c

determined by HPLC analysis. Chloro(2-dicyclohexylphosphino-2′,4′,6′-triisopropyl-1,1′-biphenyl)[2-(2′-amino-

d

f

1

,1′-biphenyl)]palladium(II). Heated to 65 ºC for 16 h. G2-PdXPhos (0.5 mol%)

In order to remove residual Pd after aqueous workup, the crude solution of 9 in toluene was treated with Si-

thiol^®⁶and palladium content was reduced from 504 ppm to <10 ppm in the solution after filtration of the resin

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

scavenger. All attempts to isolate intermediate 9 as a solid were unsuccessful, and therefore the resulting solution of

9

was directly subjected to trifluoroacetic acid at 65 °C for 12 h to cleave the t-butyl ester.

Scheme 2. Optimized Suzuki Cross-Coupling and Cleavage of Ester 9

1) G2-PdXPhos (0.5 mol%),

XPhos (0.5 mol%),

O

F

OH

K PO (300 mol%)

3

4

+

B

t

O Bu

toluene, H O

Cl

2

(140 mol%)

2) Si-thiol

O

5

Step 3

O

F

O

F

TFA (230 mol%)

toluene

t

O Bu

OH

O

(

88% yield

9

2

over 2 steps)

Step 4

After distillation of solvent and addition of heptane, the resulting acid 2 was isolated by filtration. Gratifyingly,

adamantyl ester 8 (Table 1), originating from side product 6, was completely purged (<0.1 A% HPLC) and 2 was

obtained in 88% yield over 2 steps and 99.8 A% HPLC purity (Scheme 2).

Scheme 3. Synthesis of Sulfamide 3

1) DCM,

t-BuOH

Cl (100 mol%)

1

) DCM,

TFA

1) azetidine•HCl (110 mol%)

2) aq. citric acid (20 wt%)

H

N

t_B_u_O

(300 mol%)

^tBuO

N

H N

N

C

S

2

S

O

S

O O

3) acetone,

aq. citric acid (0.2 wt%)

O O

2) CPME

2) NEt₃

O

O O

O

O O

(300 mol%)

10

11

12

(65% yield)

3

(69% yield over 2 steps)

9

9.5 A% HPLC

99.6 A% GC

The synthesis of a sulfamoylating agent derived from chlorosulfonyl isocyanate (CSI, 10) had been described

7

8

previously in the literature. Using similar conditions, 10 was reacted with t-BuOH in DCM in the presence of NEt

3

9

to provide zwitterion 11, which was directly quenched with azetidine•HCl to give N-Boc-protected sulfamide 12

(Scheme 3). Crude solid 12 (>81.0 A% HPLC purity) was recrystallized from acetone and dilute aqueous citric acid

ACS Paragon Plus Environment

6

Page 7 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

(

0.2 wt%) to furnish 12 with 99.5 A% purity by HPLC. Deprotection of 12 was then accomplished by treatment with

TFA in DCM at 25 °C, and sulfamide 3 was directly isolated in 65% yield as a white solid from the concentrated

reaction mixture after addition of CPME.

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Scheme 4. Amidation to GDC-0276

1) Toluene,

CDI (125 mol%)

DBU (275 mol%)

2) aq. citric acid

O

F

O

F

H N

N

H

+

2

S

N

OH

S

O O

3) heptane

4) 1-propanol,

activated carbon (40 wt%)

O O

O

(130 mol%)

GDC-0276

(48% yield)

2

3

99.7 A% HPLC

For the final amidation to GDC-0276, acid 2 was first activated with CDI in toluene and DBU, and then reacted with

sulfamide 3 to afford GDC-0276 (Scheme 4). Imidazole byproduct was removed from the reaction mixture by

aqueous acidic washes. After investigating a small subset of acids (including HCl and acetic acid), citric acid was

selected to adjust the pH from 9 to 4, since it offered clean phase separation. Crude GDC-0276 was directly isolated

as a solid from the reaction mixture after addition of heptane, in ~98 A% HPLC purity. Unexpectedly, two major

side products were detected in the API in 0.57 A% and 0.20 A% HPLC, and were later identified by NMR

spectroscopy / LCMS analyses as side products 13 and 14, respectively (Figure 2, see also the Supporting

Information). These impurities were most likely formed from the corresponding azetidine ring-opened precursors

generated during the Boc deprotection of 12 with TFA, and carried through the amidation reaction.¹⁰Furthermore,

GDC-0276 crystallized from toluene was obtained as an undesired solvate¹¹Treatment with activated carbon (40

wt%) and recrystallization from 1-propanol was thus necessary to further upgrade the purity of the API to >99 A%

HPLC.¹²In order to obtain the API with the desired polymorphic form I,¹³GDC-0276 seed crystals (1 wt%) were

added to a supersaturated solution of GDC-0276 in 1-propanol at 70 °C, followed by slow cooling to 25 °C.

Figure 2. Side products 13 and 14

F

O

F

O

F

O

H

N

O

N

S

O

13

14

ACS Paragon Plus Environment

7

Organic Process Research & Development

Page 8 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

Following this route, the first scale-up synthesis of GDC-0276 quickly provided 3.4 kg of API for early clinical

trials, but a number of shortcomings were identified that needed to be addressed before a further pilot plant

N

campaign could be run: 1) S Ar reaction with 1-adamantanemethanol on the t-butyl ester 1 generated significant

amounts of undesired regioisomer 7 and adamantyl ester 6 that contributed an important overall loss of material; 2)

The high level of downstream oligomeric impurities in the API originating from the sulfamide 3 required a carbon

treatment as a penultimate operation; 3) The isolation of an undesired crystal solvate from toluene necessitated

recrystallization in a separate additional step.

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Second Generation Synthesis

We hypothesized that deprotection of Boc sulfamide 12 with TFA resulted in opening of the azetidine ring and

subsequent formation of side products which carried through the amidation reaction. To overcome this issue, we

sought to swap the Boc-protecting group in 12 for a Cbz group, which would allow mild deprotection by

hydrogenolysis. Thus, simply replacing t-butanol with benzyl alcohol during the synthesis of sulfamide 3 provided

8

the corresponding Cbz-protected sulfamide 16 (Scheme 5). The reaction was performed in DCM at 5 °C, and

intermediate 15 was not isolated but telescoped into the next step with Et N and azetidine•HCl to achieve

conversion to 16. Filtering off the precipitated Et N•HCl from the reaction mixture was found to be problematic at

pilot plant scale (~100 kg) due to long filtration times (>10 h). We therefore implemented an extractive aqueous

acidic workup using citric acid,¹⁴and the resulting solution of 16 was swapped to acetoneH

O, followed by

3

2

addition of dilute aqueous citric acid as antisolvent. Crude 16 (97.4 A% HPLC purity) was recrystallized from

acetone and dilute aqueous citric acid (0.2 wt%) to furnish 16 with high purity in 67% overall yield over the 2 steps.

Scheme 5. New Synthesis of 3

1) azetidine•HCl (106 mol%)

aq. citric acid (20 wt%)

2) acetone

aq. citric acid (0.2 wt%)

PhCH OH (101 mol%)

Cl NEt₃(302 mol%)

2

N

C

S

Ph

O

N

O

S

O O

DCM

3) EtOAc, acetone

heptane

O

O O

10

15

(67% yield)

1) Pd / C (0.1 mol%)

H

MeOH, H₂

Ph

O

N

H N

2

N

S

O

O O

16

99.8 A% HPLC

2

) CPME

O O

3

(87% yield)

99.9 A% GC

ACS Paragon Plus Environment

8

Page 9 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

We next investigated the Cbz deprotection of 16 by hydrogenolysis on 3 g scale at 20 ºC using Pd / C as catalyst

(

2

~0.8 mol% Pd) and H (80 psi) in various solvents (2-propanol, EtOH, iPrOAc, MeOH, 2-MeTHF and acetone,

10V). Most reactions went to completion after 4 h, however, product 3 precipitated from the reaction mixtures,

which rendered the removal of the catalyst problematic, unless MeOH or acetone were employed.¹⁵

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Optimizing the Pd / C loading and temperature, we found that the conversion of 16 in MeOH (10V) to 3 was

2

complete after 18 h at 20 ºC when using 0.1 mol% Pd / C at 80 psi H . The catalyst was then filtered off and 3 was

isolated by filtration after addition of CPME antisolvent. This new optimized process was successfully demonstrated

on plant scale and delivered 48 kg of azetidine-1-sulfonamide (3) in 58% overall yield and 99.9 A% purity by GC.

More importantly, a sample from this material was taken forward into the crude API amidation step, and this time

oligomer impurities 13 and 14 could not be detected by HPLC, confirming our hypothesis that they originated from

the original acidic Boc-deprotection.

Figure 3. Retrosynthetic Analysis of 2

CO/Pd

ROH

O

F

S Ar

N

F

O

F

OR

[

Br+] Cl

B(OH)₂

Cl

18

Br

O

2 (R=H)

17

19 (R=Me)

In order to increase the regioselectivity of the S

towards transesterification, a new approach to acid 2 was devised, as illustrated with the retrosynthesis shown in

Figure 3. Installation of the carboxylate group of 18 was set to follow regioselective bromination, which would

N

Ar transformation and to alleviate the reactivity of the t-butyl ester

1

occur after a selective S

N

Ar¹⁶reaction of 1-chloro-2,4-difluorobenzene (17), a commercially and readily available

17

starting material. We first investigated the nucleophilic aromatic substitution of 17 with 1-adamantanemethanol to

18

19

give 20 (Table 2), using different bases and solvents at 25 ºC.

N

Table 2. S Ar of 17 with 1-adamantanemethanol

OH

105 mol%)

F

(

F

O

F

O

+

Cl

Base (220 mol%)

Solvent, 25 ºC, 1h

Cl

20

Cl

17

21

22

Conversion

20

(A%) (A%) (A%)

21^c

22^c

Entry^a

Solvent

Base

b

(A%)

ACS Paragon Plus Environment

9

Organic Process Research & Development

Page 10 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

1

2

THF

t-BuONa

6.4

5.8

6.0

0.4

NA

NaH

4.2 >0.01

3

4

THF

t-BuOK

98

49

84.7

NA

6.1

NA

3.7

NA

0.5

0.8

NA

CH

3

CN

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

5

MTBE

29

6

DMF

DMA

THF

>99

82.1 12.1

81.6 10.9

7

8^d

94.5

3.0

^aReactions conducted in closed vials with stir bars at 25 ºC by mixing 17 (3.4 mmol) and 1-adamantanemethanol in

b

c

solvent (10V), adding bases (220 mol%) as solids. HPLC A% after 1 h reaction time. Proposed structure based on

d

EI-MS analysis (see SI). 17 (135 mmol), 1-adamantanemethanol, solvent (2.6V), base added as solution in solvent

(

7.4V) at 25 ºC.

Among the various bases screened, t-BuOK gave the highest conversion (>98 A% HPLC) to 20 (Table 2, entry 3),

but significant amounts of regioisomer 21 (6.1 A%) and bis-addition impurity 22 (3.7 A%) were also formed. THF

was selected over DMF and DMA²⁰which exhibited similar reaction rates, but resulted in significantly larger

amounts of regioisomer 21 (Table 2, entries 3, 6 and 7). Adding t-BuOK as a 20 wt% solution in THF²¹^,²²instead of

as a solid further improved the reaction profile (Table 2, entry 8). Next, the isolation and purification of 20 by direct

crystallization from the reaction mixture was optimized. The reaction mixture was quenched into water, followed by

2

3

distillation to remove excess THF , and then a mixture of H

product. 20 was thus obtained as a white solid in 91% yield and >97.7 A% purity by HPLC with acceptable levels of

2 (<0.07 A%) and 21 (<1.9 A%), which was further purged down to <0.2 A% in the next step.

2

O / MeOH was dosed in to initiate crystallization of the

2

24

Based on literature precedence , we expected bromination of 20 to occur at the C-4 position of the aromatic ring,

and when 20 was reacted with Br (120 mol%) in DCE (10V) at 50 ºC for 5 h, 18 was isolated in 93% yield and >99

A% HPLC as a single isomer after aqueous Na workup. With proof of concept in hand, we focussed on

2

2 2 3

S O

investigating various less hazardous brominating agents like N-bromosuccinimide (NBS) and N,N-dibromo-

dimethylhydantoin (DBDMH) in various solvent combinations with regards to selectivity and impurity profile

(Table 3).

ACS Paragon Plus Environment

1

0

Page 11 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

Table 3. Bromination of 20

Brominating agent (120 mol%)

F

O

1

2

F

O

_S_o_l_v_e_n_t₍₅_V₎_,₆₀oC, 2 h

Cl

0

Cl

18

4

Br

3

2

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Conversion

With

HBr / HOAc

Conversion

Entry^aSolvent Reagent

b

c

(

A%)

84.0

4.6

b

(

A%)

1

2

DMF

THF

NBS

91.0

4.6

84.0

93.0

2.7

3

CH

3

CN

NBS

0.5

4

DMF

DBDMH

93.0

2.7

5

THF

6

CH

3

CN DBDMH

CN NBS

3.6

76.0

99.2

7^d

N/A

^aReactions conducted by mixing 20 (3.4 mmol) with solvent (5 mL) and brominating agent (120 mol%) in closed

b

c

vials with stir bars, and heating at 60 ºC for 2 h. Determined by HPLC analysis. Addition of HBr as a solution in

d

HOAc (10 mol%, 33 wt % solution). Reaction conducted by mixing 20 (17.0 mmol) with solvent (50 mL) and

brominating agent (120 mol%) and heating at 80 ºC for 1 h.

Both NBS and DBDMH provided good conversion (>84 A% HPLC) to 18 in DMF at 60 ºC (Table 3, entry 1 and

4

),²⁵but only <4 A% in CH

CN (Table 3, entry 3 and 6). The poor reactivity in CH CN was improved by adding

3 3

catalytic HBr (10 mol% in HOAc), decreasing the concentration, and increasing the temperature from 60 ºC to 80

ºC, which combined to boost conversion up to >99 A% (Table 3, entry 7). On kilogram scale, NBS needed to be

charged to the reactor below the boiling point of the reaction mixture (~85 ºC), which correlated with the addition of

NBS in 10 portions over 1 h at 70 ºC, followed by further heating at reflux until 20 was reduced to <1 A% as

3

ascertained by HPLC analysis. Furthermore, 18 crystallized from the reaction mixture in CH CN, which presented a

path forward to purify and isolate this intermediate. The crystallization was optimized by seeding of the reaction

mixture at 75 ºC, followed by slow cooling to avoid encrustation. In order to safely quench excess brominating

2

6

agent, a 5 wt% aqueous NaHSO

3

was added to the reaction mixture at 5 ºC, which was then warmed to 20 ºC. The

ACS Paragon Plus Environment

1

Organic Process Research & Development

Page 12 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

optimized process was eventually demonstrated at 93 kg scale, delivering 170 kg of 18 as a white solid in 98% yield

and >99.0 A% by HPLC analysis.

For installation of the carbonyl moiety contained in 23, two general pathways were considered: (1) Mg-halogen

exchange, followed by quench of the resulting Grignard with a suitable carbonate and (2) metal-catalyzed

alkoxycarbonylation, both of which have been described extensively in the literature.²⁷Although proof of concept

could be obtained using the Mg-halogen exchange route on gram scale, our efforts focused on the Pd-catalyzed

alkoxycarbonylation of 18, because it avoided a subsequent tedious workup to remove the stoichiometric Mg salts.

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Based on literature precedent,²⁷^ewe selected the Pd(BINAP)Cl

for evaluation (1 mol%, entries 1─3, Table 4), and performed a series of reactions in an autoclave under pressure of

2

2 2 2 2

, PdCl (dppf)•CH Cl and Pd(Xantphos)Cl catalysts

2

8

29

CO (8 bar) at 120 ºC, with MeOH as reaction solvent (~9V), toluene (~3V) as co-solvent, and NEt

this screen, PdCl (dppf)•CH Cl emerged as the catalyst with the fastest conversion (Table 4, entry 3), although a

similar reaction profile was observed across other ligands. The catalyst loading for PdCl (dppf)•CH Cl could be

further reduced from 1 mol% down to 0.05 mol% (Table 4, entry 3─5) while maintaining conversion >99%.

3

as base. From

2

30

Subsequently, PdCl

2

(dppf)•CH

2

Cl

2

was exchanged for the similarly effective PdCl

2

(dppf) at 0.1 mol% (entry 8).

Table 4. Pd Catalyzed Carbonylation of 18

Pd-Cat.,

MeOH (9V),

toluene (3V)

O

F

O

F

O

F

O

F

+

OMe

Cl

O

OMe

N

E

t

(

1

3

0

m

o

l

%

)

Cl

Br

3

,

CO (8 bar)

O

OMe

O

23

20

18

24

120 ºC

Time

h)

Conv

23

20

24^d

_E_n_t_r_ya

Cat (mol%)

Pd(BINAP)Cl

Pd(Xantphos)Cl

PdCl (dppf)•CH Cl

e

(

(A%) (A%) (A%) (A%)

1

2

16

1.45

1.2

3.4

6.4

5.4

2

(1)

>99

4

89

93

92

94

93

2

0.56

0.60

0.47

1.71

1.01

0.11

6.5

2

(1)

-

3

2

4

2

(0.2)

(0.05)

(0.2)

5

2

6^b

2

ACS Paragon Plus Environment

1

2

Page 13 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

7^c

8^c

5.5

8

PdCl

(dppf)•CH

Cl

(0.2)

>99

91

1.29

0.46

1.36

2

PdCl

2

(dppf) (0.1)

96.6

0.69

^aReactions conducted by mixing 18 (31.6 mmol) in MeOH (9.3V), toluene (3.4V) and Et

N (130 mol%) at ambient

3

temperature, evacuating and backfilling with nitrogen 3 times, and then pressurizing with CO (8 bar). The reaction

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

b

c

was then stirred (1000 rpm) at 120 ºC for 15 min. MeOH (0.7V), toluene (12V). MeOH (7.6V), toluene (3.4V).

d

e

Proposed structure based on ESI-MS analysis (see SI). Determined by HPLC analysis.

3

Because of the high solubility of 23 and the low solubility of Et N•HBr in toluene, we thought to precipitate and

filter off salts from the reaction mixture by increasing the amount of toluene relative to MeOH (Table 4, entry 6),

but the reaction itself resulted in a very poor conversion (4%).³¹With respect to a safe pilot plant operation, the

32

filtration for removal of the catalyst needed to be performed at least 20 ºC below the boiling point of the azeotropic

mixture (63 ºC). This required the dilution of the reaction mixture (Table 4, entry 7) with toluene to avoid

33

crystallization of the product in the transfer lines, and subsequent concentration by azeotropic distillation, followed

by addition of MeOH / water at 60 ºC to further increase the solubility of Et N•HBr and allow it to be removed in

3

the supernatant. No seeding was necessary to initiate crystallization of 23, and the reaction mixture was thereafter

cooled to 5 ºC and aged to improve recovery. The optimized process delivered 141 kg of 23 as a white solid in 90 %

yield and with an excellent purity profile of >99.7 A% by HPLC.

Scheme 6. Suzuki-Miyaura Cross-Coupling of 23 with CPBA and PTC Hydrolysis of 19

1

) G2-PdXPhos (0.5 mol%)

XPhos (0.5 mol%),

K PO (300 mol%)

3

O

F

OH

O

F

4

+

B

OMe

Cl

toluene, H O

65 ºC, 17 h

) Si-thiol (10 wt%)

2

(

140 mol%)

O

19

23

2

1) Bu NBr (10 mol%)

4

KOH (297 mol%), H O

7 ºC, 17 h

2

O

F

8

OH

2

) 3M aq. HCl

O

2

(

92% yield over 2 steps)

>

98.9 A% HPLC

Because the new route replaced t-butyl ester 5 with methyl ester 23, we needed to reinvestigate the Suzuki-

Miyaura cross-coupling coupling of 23 with CPBA (Scheme 6). Fortunately, the previously established reaction and

work-up conditions could be applied almost exactly as before, and the resulting solution of crude 19 was directly

ACS Paragon Plus Environment

1

3

Organic Process Research & Development

Page 14 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

telescoped to the hydrolysis. We quickly developed a new protocol for a phase transfer catalyzed (PTC) hydrolysis

2 4

of 19 in toluene in the presence of H O (2.4V), KOH and Bu NBr. After phase separation and addition of aqueous

HCl to adjust to pH <1, intermediate 2 crystallized and was isolated by filtration to give 92 kg of a white solid in

92% yield and purity >98.9 A% by HPLC.

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Finally, we set out to optimize the CDI activated amidation of 2 with sulfamide 3, and directly isolate the API

from the reaction mixture as the desired polymorph. Towards this end, toluene needed to be replaced with another

solvent, because it resulted in an undesired solvate (vide supra). EtOAc and iPrOAc were identified to give the

desired Form I. A base screen was then performed, activating 2 in EtOAc with CDI, and reacting with sulfamide 3

(

Table 5). From this screen it was found that organic bases such as DBU and DIPEA, as well as NaHCO

in low purity reaction profiles (Table 5, entries 1, 2 and 6). K PO was identified as the best base in terms of

conversion to GDC-0276 (Table 5, entry 5) (>98 A% HPLC), and iPrOAc as the best solvent (>99 A% HPLC)

Table 5, entry 7).

3

, resulted

3

4

(

Table 5. Base Screen of Amidation of 2 with 3 in EtOAc

CDI (125 mol%)

Solvent (10V)

Base (200 mol%)

O

F

O

F

H

OH

N

S

H N

N

O O

O

2

O

S

2

GDC-0276

O O

3

(150 mol%)

2

A%)

6.7

GDC-0276

(A%)

Entry^a

1

Base

DBU

Solvent

c

(

EtOAc

iPrOAc

9

2

3

4

5

DIPEA

CO

PO

NaHCO

PO

4.2

2

24

K

2

3

97.4

92.6

98.4

5.3

Cs

2

3

5.6

1

K

3

4

6

3

2.7

<1

₇b

K

3

4

99.8

ACS Paragon Plus Environment

1

4

Page 15 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

^aReactions conducted by mixing 2 (3 mmol) with solvent (10V) and CDI (125 mol%) in closed vials with stir bars,

_h_e_a_t_i_n_g_a_t₇₀_º_C_f_o_r₁_h_,_a_d_d_i_n_g_b_a_s_e₍₂₀₀_m_o_l_%₎_,₃₍₁₅₀_m_o_l_%₎_a_n_d_h_e_a_t_i_n_g_a_t₇₀_º_C_f_o_r₁₈_h_.b 2 (145 mmol),

c

solvent (7V), CDI (130 mol%), base (200 mol%) at 70 ºC for 18 h. Determined by HPLC analysis.

3 4

For workup, K PO was filtered off and the resulting solution washed with aqueous HCl. Unfortunately, the

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

recovery of API after addition of heptane anti-solvent was only moderate because of high solubility in the iPrOAc /

heptane mixture (~85% yield).³⁴Therefore, the solvent was swapped to 1-propanol, GDC-0276 seed crystals (1

wt%) were added to the supersaturated solution at 70 °C, and this was followed by slow cooling to ~10 °C. The

optimized amidation (Scheme 7) was then implemented at 92 kg scale, delivering 108 kg of GDC-0276 as a white

solid in 88% yield and >99.7 A% by HPLC (no single impurity >0.05 A%).

Scheme 7. Optimized Second-Generation Synthesis of GDC-0276

1) iPrOAc,

CDI (125 mol%)

K3PO4 (200 mol%)

O

F

O

>

F

H N

N

2

+

S

H

OH

N

O O

S

2

) 3M aq. HCl

3) 1-propanol

O O

O

2

3

GDC-0276

(88% yield)

99.7 A% HPLC

Conclusion

We have developed an improved large-scale synthesis of GDC-0276 via a newly designed six-step synthetic

sequence from readily available 1-chloro-2,4-difluorobenzene. A key step of the revised route was the highly

regioselective

N

S Ar fluoride displacement of 1-chloro-2,4-difluorobenzene with 1-adamantanemethanol

(regioselectivity >98:2), followed by regioselective bromination and an efficient subsequent palladium-catalyzed

methoxycarbonylation of aryl bromide 18 to give methyl ester 23. This intermediate was then subjected to a Suzuki-

Miyaura cross-coupling with CPBA to give 19, and telescoped into the PTC ester hydrolysis to furnish 2. The

optimized CDI-activated amidation of 2 with sulfamide 3 delivered GDC-0276 as the desired crystal form from 1-

propanol in >99.7 A% HPLC purity. This second generation manufacturing process represented a significant

improvement in term of costs and efficiency over the first generation route: 1) the difluorochlorobenzene starting

material was about four times less expensive (in molar ratio) than the original more functionalized starting material

4

; 2) overall yield was improved to 64% from 12% over the 6-step sequence; 3) Most of the intermediates were

ACS Paragon Plus Environment

1

5

Organic Process Research & Development

Page 16 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

isolated via direct crystallization which considerably reduced the number of unit operations by eliminating

extractive workups.

Experimental Section

General. Unless otherwise noted all reactions were run under a nitrogen atmosphere, solvents and reagents were

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

13

used without further purification. H NMR (300 MHz, 500 MHz, 600 MHz), C NMR (125 MHz) were recorded on

a Bruker Avance 3 spectrometer. Chemical shifts are reported in ppm ( units) downfield of internal

3 4 3

tetramethylsilane [(CH ) Si] or residual CHCl ; 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 B-540). Compound 1 (CAS No. 1354961-13-9), 4 (CAS

No. 130025-33-1) and 17 (CAS No. 1435-44-5) are commercially available. In process method for analysis of 2, 3,

5

, 7, 9, 19, 23 and GDC-276. 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. Signal 220 nm Bandwidth 4 nm, Reference off. Injection volume: 2 μL. Flow rate: 1.5

mL/min. Sample Concentration: 0.5─1.0 mg/mL in 50% acetonitrile / water. Program: 0.0 min 30.0% B, 0.3 min

30.0% B, 5.0 min 90.0% B, 6.5 min 5%. Typical retention times: 2 (RT 2.6 min), 3 (RT 5.4 min), 5 (RT 4.0 min), 7

(RT 5.3 min), 9 (RT 5.2 min), 19 (RT 2.8 min), 23 (RT 3.6 min), GDC-0276 (RT 4.2 min).

Method for analysis of 17, 18, 20, 21, 22, 23, 24. Diluent: acetonitrile. Mobile phase A: 0.05% TFA / water. Mobile

phase B: 0.05% TFA / acetonitrile. Column: XBridge C18, 150 x 4.6 mm. Column temperature: 40 °C. Detector

wavelength: 220 nm. Signal 220 nm Bandwidth 4 nm, Reference off. Injection volume: 5 μL. Flow rate: 1.5

mL/min. Sample Concentration: 0.5─1.0 mg/mL in 50% acetonitrile / water. Program: 0.0 min 60.0% B, 5.0 min

8

5.0% B, 14.0 min 92.5% B, 16.0 min 100%, 16.5 min 60%, 22.0 min 60%. Typical retention times: 17 (RT 10.7

min), 18 (RT 9.1 min), 20 (RT 14.2 min), 21 (RT 14.5 min), 22 (RT 17.6 min), 23 (RT 15.1 min), 24 (RT 9.5 min).

Injection volume 5 µl.

First Generation Synthesis

tert-Butyl-5-chloro-2,4-difluorobenzoate (1)

A reactor was charged with 5-chloro-2,4-difluorobenzoic acid (10.5 kg, 54.5 mol, 100 mol%), 4-

dimethylaminopyridine (0.73 kg, 6.0 mol, 11 mol%) and THF (60 kg) at 20 ºC, and to the solution was added

(

Boc)

2

O (20 kg, 91.6 mol, 170 mol%) as solids in portions over 1 h. The resulting mixture was heated to 35 ºC for

ACS Paragon Plus Environment

1

6

Page 17 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

1

8 h. HPLC analysis indicated that 5-chloro-2,4-difluorobenzoic acid 4 <2 A%. The mixture was concentrated to 35

L, diluted with EtOAc (40 kg), washed with saturated NaHCO

3

(3 x 25 L), 10% aqueous HCl (3 x 6 L) and brine (3

x 25 L), dried over Na SO (5 kg) and concentrated under vacuum at 40 ºC to give 14.0 kg of 1 as a light-brown oil

2

4

(100% uncorrected yield, 97.0 A% HPLC), which was used in the next step without any further purification.

tert-Butyl 4-((-adamantan-1-yl)methoxy)-5-chloro-2-fluorobenzoate (5)

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

A reactor was charged with 1 (14.0 kg, 56.3 mol, 100 mol%), DMSO (77 kg, 5V), 1-adamantanemethanol (9.0 kg,

5

4.1 mol, 96 mol%) and Cs

2

CO

3

(35 kg, 107.4 mol, 191 mol%). The resulting mixture was heated to 72 ºC for 20 h,

®

cooled to 25 ºC and filtered through a pad of Celite (4 kg). The filtered cake was washed with EtOAc (140 kg). The

2 4

combined filtrates were washed with 1M aqueous HCl (170 kg) and brine (170 kg), dried over Na SO (10 kg) and

concentrated to ~30 L, which was triturated with MeOH (30 kg). The solid was collected and dried at 45 ºC under

vacuum overnight to give crude 5, which was triturated with MeOH (11 kg) for 2 h at 20 ºC. The white solid was

collected, washed with MeOH (11 kg) and dried under vacuum to give 8.3 kg of 5 (32% corrected yield, 86.2 A%

1

HPLC purity; by-product 17: 13.6 A% HPLC): H NMR (600 MHz, CDCl

3

) δ 7.85 (d, 1H), 7.61 (d, 1H), 3.55 (s,

_H₎_,₂_.₀₁_─₂_.₁₅₍_m_,₃_H₎_,₁_.₆₅_─₁_.₈₁₍_m_,₁₂_H₎_,₁_.₅₅₍_s_,₉_H₎_;13C NMR (105 MHz, CDCl

₎_δ₁₆₁_.₅₍_d_,₂₆₀_H_z₎a_,

3

2

a

162.e (d, 150 Hz) , 158.94 (d, 105 Hz) , 132.5, 117.9, 112.4, 101.5, 89.8,.77.2, 39.3, 37.0, 33.9, 33.4, 28.2, 28.2,

a

13

19

+

28.1, 28.1, 28.1, 28.1. Splitting due to C─ F coupling. HRMS [M+H] calcd for C22

3

H28ClFO 339.1158; found

339.1145.

tert-Butyl (azetidin-1-ylsulfonyl)carbamate (12)³⁵

A reactor was charged with t-butanol (4.77 kg, 64.4 mol, 100 mol%), anhydrous dichloromethane (190 kg) and

cooled to 0 ºC. Chlorosulfonyl isocyanate (9.16 kg, 64.4 mol, 100 mol%) was added slowly. The resulting mixture

was stirred at 0 ºC for 1 h. Et

followed by azetidine hydrochloride (6.65 kg, 71.0 mol, 110 mol%) in four portions. The resulting mixture was

stirred for 15 h at 20 ºC. Et N•HBr was filtered off and the filtrate was washed with 20 wt% aqueous citric acid (102

3

N (19.55 kg, 193.2 mol, 300 mol%) was added drop-wise to the mixture at 5─10 ºC,

3

kg, 106 mol, 160 mol%) (pH = 4─5) below 5 ºC, brine (2 x 135 kg) and dried over anhydrous sodium sulfate. The

solid was filtered and the organic fraction was concentrated under vacuum to afford 13.3 kg of crude 12 as a pale

pink solid. The crude product was dissolved in acetone (10.8 kg), and the mixture was treated with 0.2 wt% aqueous

citric acid (106 kg, 1 mol, 2 mol%) at 0─5 ºC for 1 h. The product was filtered and washed with water (2 x 25 kg).

The cake was collected and dried under vacuum to afford 10.5 kg of 12 as a solid (69% yield, 99.5 A% HPLC): mp

ACS Paragon Plus Environment

1

7

Organic Process Research & Development

Page 18 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

1

=

83 ºC; H NMR (500 MHz, DMSO-d

_s_,₉_H₎_;13C NMR (125 MHz, DMSO-d ₎_δ₁₅₁_.₃₀_,₈₂_.₁₅_;₅₁_.₈₆_,₂₈_.₁₃_,₁₅_.₁₁_._H_R_M_S_[_M₊_H_]+ calcd for

6

) δ 10.95 (br s, 1H), 3.94 (t, J = 7.7 Hz, 4H), 2.15 (p, J = 7.8 Hz, 2H), 1.45

(

C

8

H

16

N

2

O

4

S 237.0904; found 237.0905.

Azetidine-1-sulfonamide (3)³⁴

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

A reactor was charged with dichloromethane (96.73 kg), 12 (7.30 kg, 30.9 mol, 100 mol%), TFA (10.57 kg, 92.7

mol, 300 mol%) and the mixture was held at 25 ºC for >14 h. HPLC analysis showed 12 to be <3 mg/mL. The

contents of the reactor were distilled under vacuum until a volume of 45 L was obtained while maintaining the

internal batch temperature at less than 25 ºC. Cyclopentyl methyl ether (CPME, 31.39 kg) was charged to the reactor

and the mixture was aged for a minimum of 30 min at 25 ºC. The contents of the reactor were filtered through an

Aurora filter and washed twice with CPME (5.18 kg each time). After no more filtrate could be collected from the

filter, the cake was dried on the filter at 50 ºC (jacket temperature) under vacuum with a nitrogen purge for ≥24 h.

1

The process gave 2.72 kg of 3 (65% yield; 99.6 A% by GC) as an off white solid: mp = 129 ºC; H NMR (500 MHz,

DMSO-d

for C

6

) δ 6.84 (s, 2H), 3.67 (t, 4H), 2.07 (p, 2H); ¹³C NMR (125 MHz, DMSO-d

) δ 50.38, 14.92; HRMS calcd

6

+

3

H

8

N

2

O

2

S [M+H] 137.0379, found m/z 137.0380.

t-Butyl 4-((adamantan-1-yl)methoxy)-5-chloro-2-fluorobenzoate (9)

A reactor was charged with water (19.44 kg), K PO (13.06 kg, 61.5 mol, 300 mol%), toluene (49.09 kg), 5 (8.10

3

4

kg, 20.5 mol uncorrected, 100 mol%), cyclopropylboronic acid (2.47 kg, 28.7 mol, 140 mol%) and XPhos (49 g, 103

mmol, 0.5 mol%). The reactor was inerted with nitrogen by cycling from full vacuum to nitrogen three times and

then sparged with nitrogen for minimum 10 min. Chloro(2-dicylcophexylphosphino-2’,4’,6’-triisopropyl-1,1’-

biphenyl)[2-(2’-amino-1,1’-biphenyl)]palladium(II) (81 g, 103 mmol, 0.5 mol%) was then charged and sparged for

another 20 min with nitrogen. The mixture was heated to 65 ºC and held at that temperature for >14 h. HPLC

analysis showed 5 to be <1.0 A%. The contents of the reactor were cooled to 40 ºC. Water (16.20 kg) was charged

to the reactor at 40 ºC and the contents were stirred for a minimum of 5 min. Agitation was stopped and the bottom

aqueous layer was drained to waste. A brine solution was prepared from NaCl (4.86 kg) and water (32.40 kg), added

to the reactor and the contents stirred for a minimum of 5 min. Agitation was stopped and the bottom aqueous layer

was drained to waste. Silica-thiol (0.81 kg) was charged to the reactor and the mixture heated to 65 ºC and held at

that temperature for >14 h. ICP MS analysis showed Pd to be <10 ppm. The contents of the reactor were cooled to

4

0 ºC and filtered through a Nutsche filter. The reactor and Nutsche filter were rinsed with toluene (21.06 kg) and

ACS Paragon Plus Environment

1

8

Page 19 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

combined with the filtrate in another reactor and the crude solution of 9 (100% corrected yield, 7.1 kg, 17.7 mol,

1

3

1.7 wt% in toluene) and telescoped to the subsequent t-butyl ester cleavage. H NMR (300 MHz, CDCl )  7.37 (d,

J = 8.4 Hz, 1H), 6.47 (d, J = 12.9 Hz, 1H), 3.49 (s, 2H), 2.05─1.95 (m, 4H), 1.78─1.61 (m, 12H), 1.55 (s, 9H),

0.91─0.84 (m, 2H), 0.64─0.58 (m, 2H).

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

4-((Adamantan-1-yl)methoxy)-5-cyclopropyl-2-fluorobenzoic acid (2)

TFA (4.68 kg, 41 mol, 232 mol%) was charged to the toluene solution from the previous Suzuki-Miyaura coupling

(7.1 kg of 9, 17.7 mol, 11.3 wt%) in the reactor, the mixture was heated to 65 ºC and then held at that temperature

o

for ≥12 h. HPLC analysis showed 9 to be ≤2.0 A%. The contents of the reactor were distilled at 65 C under vacuum

until a volume of 65 L was obtained (35 L distillate removed). Heptane (11.10 kg) was charged to the reactor and

the mixture was heated to 65 ºC and held at that temperature for ≥30 min. The mixture was cooled over 5 h to 20 ºC

and aged at that temperature for a minimum of 4 h. The mixture was filtered through a filter dryer and the cake was

washed twice with heptane (11.10 kg each). After no more filtrate could be collected from the filter, the cake was

dried on the filter at 60 ºC (jacket temperature) under vacuum with a nitrogen purge. Drying was continued for ≥24

1

h. The process gave 5.36 kg of 2 (88% yield; 99.8 A% HPLC purity) as an off-white solid: mp = 298 ºC; H NMR

(

500 MHz, DMSO-d

6

)  12.79 (br s, 1H), 7.33 (d, J = 8.4 Hz, 1H), 6.87 (d, J = 13.2 Hz, 1H), 3.64 (s, 2H), 2.03 (dt,

13

1

H), 2.00 (t, 3H), 1.72 (m, 6H), 1.66 (s, 6H), 0.91 (m, 2H), 0.60 (m, 2H). C NMR (125 MHz, DMSO-d

6

)  165.38,

a

1

62.62 (d, 255 Hz) , 160.50, 128.46, 127.77, 110.19, 100.71, 78.70, 39.30, 37.04, 33.95, 27.97, 9.62, 7.52. Splitting

13

19

+

3

due to C─ F coupling. HRMS [M+H] calcd for C21H25FO 345.1861; found 345.1863.

4-(Adamantan-1-ylmethoxy)-N-(azetidin-1-ylsulfonyl)-5-cyclopropyl-2-fluorobenzamide, GDC-0276

A reactor was charged with toluene (31.90 kg), acid 2 (5.27 kg, 15.3 mol, 100 mol%), 1,1-carbonyldiimidazole

(3.10 kg, 19.1 mol, 125 mol%) and the mixture was heated to 50 ºC and held at that temperature for >2 h. HPLC

o

analysis showed 2 to be ≤2.0 A%. DBU (2.33 kg, 15.3 mol, 100 mol%) was charged to the reactor at 50 C. A

second reactor was charged with toluene (13.68 kg), sulfonamide 3 (2.71 kg, 19.9 mol, 130 mol%), DBU (3.49 kg,

o

2

2.9 mol, 150 mol%) and the mixture was transferred at 25 C during 5 min into the first reactor. The second reactor

was rinsed with toluene (5.00 kg) into the first reactor. The contents of the reactor were heated to 50 ºC and held at

that temperature for a minimum of 2 h, then cooled to 20 ºC. A solution of citric acid (8.23 kg, 42.8 mol, 280 mol%)

in water (30.00 kg) was added and agitated for a minimum of 5 min. The bottom aqueous layer was drained to waste

and the reactor contents were filtered through a Nutsche filter with a subsequent toluene rinse (5.00 kg) into another

ACS Paragon Plus Environment

1

9

Organic Process Research & Development

Page 20 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

reactor in order to remove any unreacted acid. The contents were distilled under vacuum until a volume of 30 L was

obtained while maintaining the internal batch temperature of less than 65 ºC. Heptane (21.63 kg) was charged to the

o

reactor at 50 C and the mixture was aged for a minimum of 30 min at 50 ºC. The contents of the reactor were

cooled to 20 ºC and aged at that temperature for a minimum of 8 h. The content was filtered through an Aurora filter

followed by a rinse with heptane (8.66 kg). The cake was washed three times with heptane (7.21 kg each) and after

no more filtrate could be collected from the filter, the cake was dried on the filter at 60 ºC (jacket temperature) under

vacuum with a nitrogen purge. Drying was continued under vacuum for ≥24 h. The process gave 7.26 kg of crude

GDC-0276 (100% uncorrected yield, 98.20 A% HPLC purity) as an off-white solid.

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Recrystallization of GDC-0276

A reactor was charged with 1-propanol (32.58 kg), GDC-0276 (4.50 kg, 15.3 mol, 100 mol%), activated carbon

(

Darco^T^MG-60, 1.80 kg, 40 wt%) and the contents were heated to 95 ºC and held at that temperature for >1 h. HPLC

analysis showed 13 ≤0.45 A% and 14 ≤0.15 A%. The reactor contents were transferred through a Nutsche (pre-

rinsed with 1-propanol at 90 ºC, 16.07 kg) followed by a 1µm polish filter into a clean reactor. The equipment was

rinsed with 1-propanol (3.60 kg) and combined with the mother liquors. The reactor content was heated to 85 ºC

until a clear solution was obtained, and then cooled to 70 ºC. A suspension of GDC-0276 seed crystals (45.0 g, 1

mol%) in 1-propanol (0.36 kg) was transferred to the reactor and aged for >0.5 h at 70 ºC. The reactor contents were

cooled over a minimum of 1 h to 60 ºC and held at that temperature for a minimum of 1 h, then cooled to 25 ºC

during a minimum of 2 h and held at that temperature for ≥2 h. The content was filtered through a filter dryer

followed by a rinse with 1-propanol (7.20 kg). The cake was washed two times with 1-propanol (7.20 kg each) and

after no more filtrate could be collected from the filter, the cake was dried on the filter at 75 ºC (jacket temperature)

under vacuum with a nitrogen purge. Drying was continued for ≥24 h. Head-Space GC showed 1-propanol ≤0.4

wt%. The process gave 3.43 kg of GDC-0276 (48% yield, 99.7 A% HPLC purity) as an off-white solid: mp = 151

ºC; 1H NMR (600 MHz, benzene-d6)  8.85 (s, NH), 7.84 (d, J = 8.96 Hz, 1H), 6.12 (d, J = 14.23 Hz, 1H), 3.98 (t, J

=

7.75 Hz, 4H), 3.06 (s, 2H), 1.94 (m, 3H), 1.92 (m, 1H) 1.69 (m, 12.36 Hz, 3H), 1.62 (m, 11.46 Hz, 3H), 1.54 (m,

2

.58 Hz, 6H), 1.42 (m, 7.73 Hz, 2H) 0.66 (m, 2.74 Hz, 2H), 0.52 (m, 3.22 Hz, 2H); 13C NMR (150 MHz, benzene-

a

d6)  163.54 (d, 120 Hz) , 161.52─159.89 (245 Hz) , 161.21─161.19 (30 Hz) , 129.85 (d, 30 Hz) , 129.10 (d, 165

a

Hz) , 128.04, 110.20 (d, 105 Hz) , 99.10 (d, 285 Hz) , 78.79, 51.43, 39.58, 37.24, 33.97, 28.49, 15.01, 9.87, 7.29.

Splitting due to 13C─19F coupling. HRMS [M+H]+ calcd for C24 S 462.5773; found 463.2045.

H

2 4

31FN O

ACS Paragon Plus Environment

2

0

Page 21 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

Second Generation Synthesis

-((2-Chloro-5-fluorophenoxy)methyl)adamantane (20)

1

A reactor was charged with 1-adamantanemethanol (29.4 kg, 177 mol, 105 mol%), 17 (25.0 kg, 168 mol, 100 mol%)

and THF (71.2 kg). To the reaction mixture was added a 20% solution of t-BuOK in THF (208 kg, 371 mol, 221

mol%) at 20 °C during 1 h. After 1 h stirring, HPLC analysis showed 17 to be ≤1.0%. The reaction mixture was

added onto purified water (286 kg) at 20 °C during 30 min. The biphasic solution was distilled under vacuum

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

(250─500 mbar, 40─60°C) down to 8.4V. Then a mixture of purified water (375 kg) and MeOH (59 kg) was

charged at 53°C during 30 min. The reaction mixture was cooled to 20 °C during 2 h, stirred for 1 h, filtered, and the

cake washed successively with a 1:1 mixture of water and MeOH (44 L), and twice with purified water (2 x 150 L).

The cake was dried in vacuum at 53 °C to give 45.0 kg of 20 as a white solid (91% yield, 97.7 A% HPLC purity, 21

1

=

1.85 A% HPLC): mp = 119 ºC; H NMR (500 MHz, DCM-d

2

)  7.33 (dd, 1H), 6.71 (dd, 1H), 6.64 (td, 1H), 3.57

s, 3H), 2.06 (m, 3H), 1.79 (m, 6H), 1.74 (d, 6H). ¹³C NMR (125 MHz, DCM-d

)  162.02 (d, 244 Hz) , 155.92,

a

(

2

30.19, 117.92, 107.08, 101.27, 79.36, 39.28, 37.01, 33.90, 28.31. ^aSplitting due to ¹³C─ F coupling. HRMS

19

1

+

[

M+H] calcd for C17

H20ClFO 294.1187; found 294.1190.

1

-((4-Bromo-2-chloro-5-fluorophenoxy)methyl)adamantane (18)

In a reactor 20 (45.0 kg, 153 mol, 100 mol%) was dissolved in acetonitrile (540 kg) and heated to 75 °C. A solution

of 33 wt% HBr in acetic acid (3.7 kg, 15 mol, 10 mol%) was added, followed by NBS (32.6 kg, 183 mol, 120

mol%) in 10 portions over 1 h. Subsequently, the mixture was heated to 85 °C (reflux) and stirred for 1 h until

deemed complete by HPLC (20 ≤1.0 A%). The contents of the reactor were cooled to 77 °C, seeded, further cooled

to 53 °C during 1 h, aged for 1 h, and then cooled to 3 °C during 2 h. The reaction mixture was quenched with a 5

wt% aqueous sodium bisulfite solution (22.0 kg, 11 mol, 7 mol%) and heated to 20 °C during 30 min. The

suspension was filtered at 20 °C, washed with a 3:1 mixture of water and acetonitrile (270 kg), and the cake dried

1

under vacuum at 48 °C to give 56.3 kg of 18 as a white solid (98% yield, 99.9 A% HPLC purity): mp = 116 ºC; H

1

3

NMR (500 MHz, DCM-d

2

)  7.56 (d, 1H), 6.79 (d, 1H), 3.56 (s, 2H), 2.06 (m, 3H), 1.79 (m, 6H), 1.73 (m, 6H). C

a

NMR (125 MHz, DCM-d

2

)  158.20 (d, 246 Hz) , 155.43, 132.85, 118.89, 102.02, 98.02, 79.71, 39.97, 33.92,

a

13

19

+

2

8.27. Splitting due to C─ F coupling. HRMS [M+H] calcd for C17H19BrClFO 372.03; not observed.

Methyl 4-((adamantan-1-yl)methoxy)-5-chloro-2-fluorobenzoate (23)

ACS Paragon Plus Environment

2

1

Organic Process Research & Development

Page 22 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

3

In a reactor 18 (42.2 kg, 113 mol, 100 mol%) was dissolved in toluene (210 kg), heated to 55 °C and Et N (15.0 kg,

1

48 mol, 131 mol%) was charged. The solution was diluted with MeOH (311 kg) and then transferred to an

2

autoclave. A solution of Pd(dppf)Cl (84 g, 0.11 mol, 0.1 mol%) in MeOH (19 kg) was added to the reaction mixture

followed by a rinse with MeOH (24 kg). The reactor was inerted with nitrogen by cycling from vacuum to nitrogen

three times, pressurized with carbon monoxide (98%) to 7.80 bar, heated to 120 °C and stirred for 8 h. The reaction

mixture was then cooled to 57 °C and the autoclave was vented. The reactor was inerted with nitrogen by cycling

from vacuum to nitrogen three times. The contents of the reactor were filtered through a polish filter, followed by a

rinse with toluene (129 kg). The solution was concentrated at 70 °C under atmospheric pressure (removal of 527 L

distillate). Afterwards MeOH (34 kg) was added and another 1.0V of solvent (42 L) was removed. Then MeOH (269

kg) was charged to the reactor, followed by a 5:1 mixture of MeOH and water (104 kg). The suspension was cooled

to 60 °C during 30 min, cooled to 5 °C during 3 h, then aged for 90 min at this temperature. The suspension was

filtered and the cake washed with a 5:1 mixture of MeOH and water (90 kg), followed by MeOH (43 kg). The cake

was dried under vacuum (21 mbar) at 50 °C to give 36.6 kg of 23 as a white solid (92% yield, 99.8 A% HPLC

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

purity): mp = 128 ºC; H NMR (500 MHz, DMSO-d )  9.00 (s, 1H), 7.87 (d, 1H), 7.22 (d, 1H), 3.82 (s, 3H), 3.73 (s,

6

_H₎_,₁_.₉₉₍_m_,₃_H₎_,₁_.₇₀₍_m_,₆_H₎_,₁_.₆₄₍_s_,₆_H₎_.13C NMR (125 MHz, DMSO-d

₎_₁₆₃_.₂₃_,₁₆₀_.₇₀₍_d_,₂₅₉_H_z₎a_,

2

6

a

13

19

1

59.44, 132.14, 117.50, 110.65, 103.39, 79.68, 52.74, 39.07, 36.98, 33.94, 27.89. Splitting due to C─ F coupling.

+

3

HRMS [M+H] calcd for C19H22ClFO 353.1315; found 353.1300.

Benzyl (azetidin-1-ylsulfonyl)carbamate (16)

A reactor was charged with DCM (1378.0 kg), benzyl alcohol (66.2 kg, 612 mol, 101 mol%), and the content of the

reactor was cooled to -5 °C. Chlorosulfonyl isocyanate (85.3 kg, 603 mol, 100 mol%) was added at -5 °C, and the

mixture was stirred for 3 h at this temperature until deemed complete by HPLC analysis (derivatization with

3

dibenzylamine, chlorosulfonyl isocyanate <0.3 A%, see SI). Et N (184.6 kg, 1824 mol, 302 mol%) and azetidine

hydrochloride (60.0 kg, 641 mol, 106 mol%) was added to the reaction mixture at -5 °C, and stirred for 3 h at -5 °C

until the reaction was deemed complete by HPLC (derivatization with dibenzylamine, 15 <1 A%, see SI). Aqueous

2

0% citric acid (328 kg) was added to adjust to pH=4.5. The DCM layer was drained into another reactor and the

aqueous layer was extracted with DCM (671.0 kg) followed by addition of water (170 kg). The DCM layer was

drained into the reactor containing the first DCM extract, and the mixture was washed with water (520 kg), then

acetone (680.9 kg, 686.7 kg, 681.8 kg, 100 kg) was added to the reactor in four portions while continuously

ACS Paragon Plus Environment

2

Page 23 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

distilling off DCM (residual DCM of final solution = 0.2 wt% by head-space GC). Aqueous 0.2% citric acid (1038

kg) and seed crystals of 16 (40 g) were added to the reactor to initiate crystallization. The mixture was stirred for 30

min at 10 °C and cooled to 0 °C during 4 h. The suspension was centrifuged and the cake washed with aqueous

0.2% citric acid in acetone (72.0 kg, 75.8 kg, 76.0 kg) to give after drying at 40 °C, 133.90 kg of crude 16. The

crude 16 was dissolved in EtOAc (489.8 kg), acetone (7.0 kg), and heptane (916.8 kg) was added followed by seed

crystals of 16 (40.0 g). The mixture was stirred for 30 min at 25 °C, cooled to 0 °C during 3 h and stirred for 2 h at 0

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

°

C. The suspension was centrifuged and the cake washed with EtOAcheptane (92.0 kg : 88 kg), and then dried for

1

6 h at 40 °C in vacuum to give 110 kg of 16 (67% yield, 99.8 A% HPLC purity) as a white solid: mp = 110 ºC; H

NMR (500 MHz, DMSO-d

6

) δ 11.43 (s, 1H), 7.39 (m, 5H), 5.18 (m, 2H), 3.95 (t, 4H), 2.12 (p, 2H); ¹³C NMR (125

MHz, DMSO-d

6

) δ 152.40, 136.12, 128.97, 128.76, 128.56, 67.57, 51.91, 15.03; HRMS calcd for C11 NaO S

H

14

N

2

4

+

[

M+H] 293.0572, found m/z 293.0568.

Azetidine-1-sulfonamide (3)

An autoclave was charged with 16 (109.0 kg, 403 mol, 100 mol%), Pd / C (Noblyst P1070, 803 g, 0.377 mol, 0.09

mol%, 10 wt% Pd / C dry basis, ~50 wt% water) and MeOH (823.9 kg) at 20 ºC. The reactor was inerted with

nitrogen by cycling from vacuum (-0.8 bar) to nitrogen three times, pressurized with hydrogen to 5.4 bar and then

stirred at this temperature for 18 h. HPLC analysis showed 16 to be not detected. The reactor was inerted with

nitrogen by cycling from vacuum (-0.8 bar) to nitrogen two times. The contents of the reactor were filtered through a

1.2 µm polish filter, followed by a rinse with MeOH (196.2 kg). The filtrate was distilled under vacuum (200 mbar)

until a volume of 100 L was obtained. The contents of the reactor were cooled to -10 ºC, CPME (215.0 kg) was

charged over 20 min and the mixture was aged for 11 h at -10 ºC. The contents of the reactor were filtered over a

filter dryer and washed with CPME (43.0 kg). After no more filtrate could be collected from the filter dryer, the cake

was dried at 50 ºC under vacuum (20 mbar) for 14 h. The process gave 47.7 kg of 3 (87% yield, 99.9 A% GC purity)

as a white solid.

4

-((Adamantan-1-yl)methoxy)-5-cyclopropyl-2-fluorobenzoic acid (2)

A reactor was charged with water (84.0 kg,), K PO (63.0 kg, 297 mol, 300 mol%), toluene (378.0 kg), 23 (35.0 kg,

9 mol, 100 mol%), cyclopropyl boronic acid (11.9 kg, 139 mol, 140 mol%) and XPhos (240 g, 0.503 mol, 0.5

3

4

9

mol%). The reactor was sparged with nitrogen for 15 min, chloro(2-dicylcophexylphosphino-2’,4’,6’-triisopropyl-

1,1’-biphenyl)[2-(2’-amino-1,1’-biphenyl)]palladium(II) (390 g, 0.496 mol, 0.5 mol%) was added. The mixture was

ACS Paragon Plus Environment

2

3

Organic Process Research & Development

Page 24 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

heated to 65 ºC and held at that temperature for 17 h. HPLC analysis showed 23 to be not detected. The contents of

the reactor were cooled to 20 ºC, agitation was stopped and the bottom aqueous layer was drained to waste. Silica-

thiol (3.5 kg, 10 wt%) was charged to the reactor, the mixture heated to 70 ºC and held at that temperature for 18 h.

4

ICP MS showed Pd to be <10 ppm. The contents of the reactor were cooled to 21 ºC and a solution Bu NBr (3.20

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

kg, 9.9 mol, 10 mol%) and KOH (19.4 kg, 294 mol, 297 mol%, pellets 85 wt%) in water (84 kg) was added,

followed by toluene (34.7 kg). The contents of the reactor were heated to 87 ºC and held at that temperature for 17 h,

then cooled to 70 ºC. HPLC analysis showed 19 to be <0.2 A%. Agitation was stopped and the bottom aqueous layer

was drained to waste. The contents of the reactor were heated to 46 ºC, filtered over a 1µm polish filter and rinsed

with toluene (26.0 kg). The combined filtrate was heated to 45 ºC, aqueous 3M HCl (50.5 kg, 144 mol, 145 mol%)

was added (pH ≤1), aged for 1 h, then cooled over 2 h to 25 ºC and aged for 10 h. The content was filtered over a

filter dryer followed by a rinse with toluene (65.0 kg). The cake was dried at 60 ºC for 135 h. The process gave 31.4

kg of 2 (92% yield, 99.6 A% HPLC purity) as an off white solid.

4

-((Adamantan-1-yl)methoxy)-N-(azetidin-1-ylsulfonyl)-5-cyclopropyl-2-fluorobenzamide, GDC-0276

A reactor was charged with acid 2 (47.2 kg, 137 mol, 100 mol%), and iPrOAc (278.0 kg). The mixture was heated to

40 ºC and held at that temperature for 25 min. A suspension of CDI (27.8 kg, 171 mol, 125 mol%) in iPrOAc (87.0

kg) was added to the reactor in three portions and stirred for 15 min between additions. The contents of the reactor

o

were then heated to 50 C and stirred for 50 min at this temperature. HPLC showed acid 2 to be not detected. K

3

PO

4

(58.2 kg, 274 mol, 200 mol%) and 3 (24.3 kg, 178 mol, 130 mol%) was charged to the reactor, the mixture was

o

heated to 70 C and held at this temperature for 19 h. The contents of the reactor were cooled to 12 ºC and aqueous

o

3

M HCl (309.0 kg, 862 mol, 630 mol%) was added. The mixture was then heated to 58 C and agitated for 20 min.

The bottom aqueous layer was drained to waste. The organic layer was washed twice with water (235.0 kg, 95.0 kg)

and the aqueous layer was drained to waste. The contents of the reactor were filtered over a 1 µm polish filter at 55

^oC and rinsed with iPrOAc (26.0 kg). The filtrate was distilled under vacuum (280 mbar) until a volume of 140 L

was obtained. 1-Propanol (144.0 kg) was added to the reactor and distilled under vacuum to a volume of 140 L. 1-

o

Propanol (376.0 kg) was added to the reactor and distilled under vacuum to a volume of 470 L at 97 C. The

o

contents of the reactor were cooled to 72 C over 2 h. A suspension of GDC-0276 seed crystals (0.64 kg) in 1-

o

propanol (1.8 kg) was added to the mixture and aged at 70 C for 1 h, then cooled to 11 C over 8 h, aged for 13 h at

this temperature and then filtered over a filter dryer, followed by a rinse with 1-propanol (94.0 kg). The cake was

ACS Paragon Plus Environment

2

4

Page 25 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

dried under vacuum at 65 ºC for 65 h. The process gave 55.6 kg of GDC-0276 (88% yield, 99.8 A% HPLC purity)

as a white solid (NMR in accordance with previous procedure).

ASSOCIATED CONTENT

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Supporting Information. NMR spectra for 2, 3, 5, 6, 12, 13, 14, 16, 18, 20, 23, GDC-0276, solubility data for 4

and 16, experimental information about derivatization of CSI (10) and 15, DSC data for cyclopropylboronic acid,

and LCMS spectra of 6, 21, 22, 24 are available free of charge via the Internet at http://pubs.acs.org .

AUTHOR INFORMATION

Corresponding Author

*

E-mail: stumpf.andreas@gene.com

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version

of the manuscript.

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENT

The authors would like to thank the following Genentech colleagues: Dr. C. J. Venkatramani, Ila. Patel, Dr.

Nathaniel Seagraves, Dr. Sarah Robinson, Tristan Maschmeyer, Rohit Ranjan for analytical support, and Dr. Lauren

Sirois for helpful suggestions on the manuscript; Dr. Markus Stöckli and Dr. Roland Weixler (Dottikon Exclusive

Synthesis) for technical support.

REFERENCES

1

(a) Emery, E. C.; Luiz, A. P.; Wood, J. N. Nav1.7 and other Voltage-Gated Sodium Channels

as Drug Targets for Pain Relief. Expert Opin. Ther. Targets 2016, 20, 975983. (b) Weiss, M.

M.; Dineen, T. A.; Marx, I. E.; Altmann, S.; Boezio, A.; Bregman, H.; Chu-Moyer, M.;

DiMauro, E. F.; Bojic, E. F.; Foti, R. S.; Gao, H.; Graceffa, R.; Gunaydin, H.; Guzman-Perez,

ACS Paragon Plus Environment

2

5

Organic Process Research & Development

Page 26 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

A.; Huang, H.; Huang, L.; Jarosh, M.; Kornecook, T.; Kreiman, C. R.; Ligutti, J.; La, D. S.; Lin,

M.-H. J.; Liu, D.; Moyer, B. D.; Nguyen, H. N.; Peterson, E. A.; Rose, P. E.; Taborn, K.;

Youngblood, B. D.; Yu, V.; Fremeau, R. T. Sulfonamides as Selective Nav1.7 Inhibitors:

Optimizing Potency and Pharmacokinetics While Mitigating Metabolic Liabilities. J. Med.

Chem. 2017, 60, 5969−5989. (c) Graceffa, R. F.; Boezio, A. A.; Able, J.; Altmann, S.; M. Berry,

L.; Boezio, C.; Butler, J. R.; Chu-Moyer, M.; Cooke, M.; DiMauro, E. F.; Dineen, T. A.; Bojic,

E. F.; Foti, R. S.; Fremeau, R. T.; Guzman-Perez, A.; Gao, H.; Gunaydin, H.; Huang, H.; Huang,

L.; Ilch, C.; Jarosh, M.; Kornecook, T.; Kreiman, C. R.; La, D. S.; Ligutti, J.; Milgram, B. C.;

Lin, M.-H. J.; Marx, I. E.; Nguyen, H. N.; Peterson, E. A.; Rescourio, G.; Roberts, J.; Schenkel,

L.; Shimanovich, R.; Sparling, B. A.; Stellwagen, J.; Taborn, K.; Vaida, K. R.; Wang, J.;

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Yeoman, J.; Yu, V.; Zhu, D.; Moyer, B. D.; Weiss, M. M. Sulfonamides as Selective Na 1.7

v

Inhibitors: Optimizing Potency, Pharmacokinetics, and Metabolic Properties to Obtain

Atropisomeric Quinolinone (AM-0466) that Affords Robust in Vivo Activity. J. Med. Chem.

2

017, 60, 5990−6017. (d) King, G. F.; Vetter, I. No Gain, No Pain: Na 1.7 as an Analgesic

v

Target. ACS Chem. Neurosci. 2014, 5, 749−751. (e) Browne, L.; Lidster, K.; Al-Izki, S.;

Clutterbuck, L.; Posada, C.; Chan, A. W. E.; Riddall, D.; Garthwaite, J.; Baker, D.; Selwood, D.

L. Imidazol-1-ylethylindazole Voltage-Gated Sodium Channel Ligands Are Neuroprotective

during Optic Neuritis in a Mouse Model of Multiple Sclerosis. J. Med. Chem. 2014, 57,

2942−2952. (f) Swain, N. A.; Batchelor, D.; Beaudoin, S.; Bechle, B. M.; Bradley, P. A.; Brown,

A. D.; Brown, B.; Butcher, K. J.; Butt, R. P.; Chapman, M. L.; Denton, S.; Ellis, D.; Galan, S. R.

G.; Gaulier, S. M.; Greener, B. S.; de Groot, M. J.; Glossop, M. S.; Gurrell, I. K.; Hannam, J., M.

S.; Lin, Z.; Markworth, C. J.; Marron, B. E.; Millan, D. S.; Nakagawa, S.; Pike, A.; Printzenhoff,

D.; Rawson, D. J.; Ransley, S. J.; Reister, S. M.; Sasaki, K.; Storer, R. I.; Stupple, P. A.; West,

C. W. Discovery of Clinical Candidate 4‑[2-(5-Amino‑1H‑pyrazol-4-yl)-4-chlorophenoxy]-5-

chloro-2-fluoro ‑ N ‑ 1,3-thiazol-4-ylbenzenesulfonamide (PF-05089771): Design and

Optimization of Diaryl Ether Aryl Sulfonamides as Selective Inhibitors of Nav1. J. Med. Chem.

2

017, 60, 7029−7042. (g) Bagal, S. K.; Brown, A. D.; Cox, P. J.; Omoto, K.; Owen, R. M.;

Pryde, D. C.; Sidders, B.; Skerratt, S. E.; Stevens, E. B.; Storer, R. I.; Swain, N. A. Ion Channels

as Therapeutic Targets: A Drug Discovery Perspective. J. Med. Chem. 2013, 56, 593−624. (h)

Blass, B. E. Acyl Sulfonamides Nav1.7 Blockers Useful for the Treatment of Pain. ACS Med.

Chem. Lett. 2018, 9, 161−162. (i) Ali, S. R.; Liu, Z.; Nenov, M. N.; Folorunso, O.; Singh, A.;

Scala, F.; Chen, H.; James, T. F.; Alshammari, M.; Panova-Elektronova, N. I.; White, M. A.;

Zhou, J.; Laezza, F. Functional Modulation of Voltage-Gated Sodium Channels by a FGF14-

Based Peptidomimetic. ACS Chem. Neurosci. 2018, 9, 976−987. (j) de Lera Ruiz, M.; Kraus, R.

L. Voltage-Gated Sodium Channels: Structure, Function, Pharmacology, and Clinical

Indications. J. Med. Chem. 2015, 58, 7093−7118.

2

(a) Hemeon; I. W.; Safina, B.; Sutherlin, D. Preparation of Fluorinated Aromatic Ethers as

Sodium Channel Blockers Useful in Treatment of Diseases. PCT Int. Appl. (2017),

WO2017035271 A1 20170302. (b) Chowdhury, S.; Dehnhardt, C. M.; Martin, C.; Hasan, A.;

Hemeon, I. W.; Jia, Q.; Jun, L.; Liu, Z.; Ortwine, D. F.; Safina, B.; Sutherlin, D. Preparation of

Substituted Benzamide Compounds. as Sodium Channel Inhibitors. U.S. Pat. Appl. Publ. (2016)

US20160340309 A1 20161124. (c) Ahuja, S.; Mukund, S.; Deng, L.; Khakh, K; Chang, E.; Ho,

H.; Shriver, S.; Young, C.; Lin, S.; Johnson Jr., J. P.; Wu, P.; Li, J.; Coons, M.; Tam, C.;

Brillantes, B.; Sampang, H.; Mortara, K.; Bowman, K. K.; Clark, K. R.; Estevez, A.; Xie, Z.;

Verschoof, H.; Grimwood, M.; Dehnhardt, C.; Andrez; J.-C.; Focken, T.; Sutherlin, D. P.;

ACS Paragon Plus Environment

2

6

Page 27 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

Safina, B. S.; Starovasnik, M. A.; Ortwine, D. F.; Franke, Y.; Cohen, C. J.; Hackos, D. H.; Koth,

C. M.; Payandeh J. Structural Basis of Nav1.7 Inhibition by an Isoform-Selective Small-

Molecule Antagonist. Science 2015, 350, aac5464-1-9. (d) Focken, T.; Liu, S.; Chahal, N.;

Dauphinais, M.; Grimwood, M. E.; Chowdhury, S.; Hemeon, I.; Bichler, P.; Bogucki, D.;

Waldbrook, M.; G. Bankar, G.; Sojo, L. E.; Young, C.; Lin, S.; Shuart, N.; Kwan, R.; Pang, J.;

Chang, J. H.; Safina, B. S.; Sutherlin, D. P.; Johnson Jr., J. P.; Dehnhardt, C. M.; Mansour, T. S.;

Oballa, R. M.; Cohen, C. J.; Robinette, C. L. Discovery of Aryl Sulfonamides as Isoform-

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

Selective Inhibitors of Na 1.7 with Efficacy in Rodent Pain Models. ACS Med. Chem. Lett. 2016,

v

7

, 277−282.

3

Replacing the t-Butyl ester group with the corresponding Me ester or adamantanemethyl ester

turned out to be unsuccessful, as they produced large amounts of the transester impurity (~25

A% HPLC) and the 2-regioisomer (~20 A% HPLC) respectively.

4

o

Reaction was performed safely at 72 C, but safety concerns have been raised for the use of

DMSO as a process solvent under certain conditions: Wang, Z.; Richter, M.S.; Gates, B.D.;

Grieme, T.A. Safety Concerns in a Pharmaceutical Manufacturing Process Using Dimethyl

Sulfoxide (DMSO) as a Solvent. Org. Process Res. Dev., 2012, 16, 1994–2000.

5

Kinzel, T; Zhang, Y.; Buchwald, S. L. A New Palladium Precatalyst Allows for the Fast

Suzuki−Miyaura Coupling Reactions of Unstable Polyfluorophenyl and 2-Heteroaryl Boronic

Acids. J. Am. Chem. Soc. 2010, 132, 14073–14075.

6

Si−thiol is a proprietary solid-supported resin available from Silicycle.

7

(a) Burgess, E. M.; Penton, H. R.; Taylor, E. A. Thermal Reactions of Alkyl N-

Carbomethoxysulfamate Esters. J. Org. Chem. 1973, 38, 2631. (b) Winum, J-Y.; Toupet, L.;

Barragan, V.; Dewynter, G.; Montero, J-L. N-(tert-Butoxycarbonyl)-N-[4-

(dimethylazaniumylidene)-1,4-dihydropyridin-1-ylsulfonyl]azanide: A New Sulfamoylating

Agent. Structure and Reactivity toward Amines. Org. Lett., 2001, 3, 2241–2243.

8

Although DCM is not a desired reaction solvent, its use was acceptable at this stage of

development. For selecting sustainable solvents: (a) Prat, D.; Pardigon, O.; Flemming, H-W.;

Letest, S.; Ducandas, V.; Isnard, P.; Guntrum, E.; Senac, T.; Ruisseau, S.; Cruciani, P.; Hosek, P.

Sanofi’s Solvent Selection Guide: A Step Toward More Sustainable Processes. Org. Process

Res. Dev., 2013, 17, 1517–1525. (b) Diorazio, L. J.; Hose, D. R. J.; Adlington, N. K. Toward a

More Holistic Framework for Solvent Selection. Org. Process Res. Dev. 2016, 20, 760–773.

9

For synthesis of azetidine: Hillier, M. C.; Chen, C-Y. A One-Pot Preparation of 1,3-

Disubstituted Azetidines. J. Org. Chem. 2006, 71, 7885–7887.

1

0

The precursor impurities have not been identified and characterized in isolated 4 since they

were not detected by the GC method.

1

Several polymorphic forms have been identified in a screen: Form I, Form II, and Form III,

along with solvates of IPA, DMSO, DMF, dioxane, THF, DCM, toluene, and acetone. See ref 13

for further details.

ACS Paragon Plus Environment

2

7

Organic Process Research & Development

Page 28 of 29

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

12

1

3: 0.25 A% HPLC; 14: <0.03 A% HPLC.

1

3

Chakravarty, P.; DiPasquale, A. G.; Stumpf, A.; Lubach, J. W.; Nagapudi, K. Understanding

Phase Behavior of Nearly Energetically Equivalent Polymorphs To Achieve Controlled

Crystallization for a Nav1.7 Pain Inhibitor Compound. Molecular Pharmaceutics 2018, 15,

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

5

0725080.

1

4

Aqueous HCl was also effective in removing excess NEt , but citric acid gave better phase

3

separation, and was therefore implemented.

1

5

A solubility screen of 16 and 3 (see SI) confirmed that either solvent could be used at more

concentrated conditions (>10 mg/mL) without 3 coming out of solution prior removal of Pd / C

by filtration.

1

6

Nucleophilic substitution reactions on 1-bromo-2,4-difluorobenzene with alcohols proceeded

in excellent regioselectivity: Ouellet, S. G., Bernardi, A., Angelaud, R., O’Shea, P. D.

Regioselective S Ar Reactions of Substituted Difluorobenzene Derivatives: Practical Synthesis

of Fluoroaryl Ethers and Substituted Resorcinols. Tetrahedron Letters, 2009, 50, 37763779.

N

1

7

Starting material 17 is available in bulk and about 4 fold less expensive than starting material 4

based on molar ratio.

1

8

Other bases screened: K CO , Cs CO showed poor conversion (<0.01 A% HPLC), KOCH

3,

2

3

2

3

NaHMDS, KHMDS resulted in low purity reaction profiles containing various side products

(<45 A% HPLC).

1

9

o

When the reaction was performed at 50 C significant amounts of a byproduct, presumably the

t-butoxide adduct, was formed.

2

0

Phase transfer catalysis was also attempted with NaOH (220 mol%) in biphasic solvent

mixtures (toluene / water; MTBE / water). Different catalysts (Bu NCl, BnEt NCl), catalyst

4

3

concentrations (10130 mol%), increased temperatures were investigated. No significant product

formation occurred.

2

1

Water content in THF used for the pilot plant batches was controlled (<0.05 wt% H O by Karl

2

Fischer) to avoid quenching of t-BuOK, which impeded reaction conversion.

2

Addition of t-BuOK THF solution was exothermic and was therefore dose controlled over 1 h

to maintain T <25 ºC.

2

3

The solvent composition of the ternary reaction mixture (t-butanol, H O and THF) impacted

2

solubility of 20, and needed to be controlled by headspace GC (THF <89.1 wt% of distillate) in

order to achieve reproducible quality and yield.

2

4

(a) Grillo, M.; Li, A-R.; Liu, J.; Medina, J.C.; Su, Y.; Wang, Y.; Jona, J.; Allgeier, A.; Milne,

J.; Murry, J.; Payack, J.F.; Storz, T. Preparation of Benzeneacetic Acid Derivatives as

Inflammation Modulators. PCT Int. Appl., 2009085177, 09 Jul 2009. (b) Kromann, J. C.; Jensen,

J. H.; Kruszyk, M.; Jessing, M.; Jørgensen, M. Fast and Accurate Prediction of the

ACS Paragon Plus Environment

2

8

Page 29 of 29

Organic Process Research & Development

1

2

3

4

5

6

7

8

9

Regioselectivity of Electrophilic Aromatic Substitution Reactions. Chem. Sci. 2018, 9, 660-665.

(c) http://regiosqm.org/.

2

5

No conversion of 20 to 18 occurred at 25 ºC.

2

6

A potassium iodide test was performed to check for effective quenching of bromine.

1

2

3

4

5

6

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

7

(a) Transition Metal Catalyzed Carbonylation Reactions:Carbonylative Activation. Beller, M.;

Wu X-F. Springer 2013, ISBN 978-3-642-39016-6. (b) Barnard, C. F. J. Palladium-Catalyzed

Carbonylation A Reaction Come of Age. Organometallics 2008, 27, 54025422. (c) Martinelli,

J. R.; Watson, D. A.; Freckmann, D. M. M.; Barder, T. E.; Buchwald, S. L. Palladium-Catalyzed

Carbonylation Reactions of Aryl Bromides at Atmospheric Pressure: A General System Based

on Xantphos. J Org Chem. 2008, 73, 71027107. (d) Natte, K.; Dumrath, A.; Neumann, H.;

Beller, M. Palladium‐Catalyzed Carbonylations of Aryl Bromides using Paraformaldehyde:

Synthesis of Aldehydes and Esters. Angew. Chem. Int. Ed. 2014, 53, 1009010094. (e) Almeida,

A. M.; Andersen, T. L.; Linhardt, A. T.; Almeida, M. V.; Skrydstrup, T. General Method for the

Preparation of Active Esters by Palladium-Catalyzed Alkoxycarbonylation of Aryl Bromides. J.

Org. Chem. 2015, 80, 19201928.

H.

Carbon

Monoxide

Toxicity.

https://www.ncbi.nlm.nih.gov/books/NBK430740/.

29

Solubility of 18 in toluene: 209 mg/mL. Solubility of 23 (25 ºC) in toluene: 209 mg/mL; in

MeOH: 3.4 mg/mL. Solubility of Et N•HBr (25 ºC) in MeOH: 294 mg/mL; MeOH / toluene

3

(69:31): 238 mg/mL.

3

0

PdCl (dppf)•CH Cl was difficult to source on large quantities and had a higher cost than

2

PdCl (dppf).

2

3

1

The carbonylation in MeOH (7.6V) and THF (1.7V) as an alternative co-solvent which would

allow to run the reaction under more concentrated conditions due to the higher solubility of 18 in

THF was also tested and performed similarly to toluene. However, all attempts to crystallize 23

from the MeOH / THF reaction mixture after concentration and addition of antisolvents (MeOH,

1

-propanol, H O, CH CN) resulted in poorly filterable solids and inferior product purity

2 3

compared to material isolated from the MeOH / toluene solvent system.

3

2

Pd content in isolated 23 <5 ppm as measured by ICP-MS.

3

Azeotropic solvent composition measured by Headspace GC: MeOHtoluene (69:31), bp = 63

ºC.

3

4

Solubility of GDC-0276 in iPrOAc / heptane (1:1) at 25 ºC: 20 mg/mL.

3

5

Buckman, B.; Nicholas, J. B.; Beigelman, L.; Serebryany, V.; Stoycheva, A. D.; Thrailkill, T.;

Seiwert, S. D. Reparation of Macrocyclic Peptides, Especially Proline-Containing Peptides, as

Inhibitors of Hepatitis C Virus Replication for Treating Hepatitis C Infection and Liver Fibrosis.

PCT Int. Appl. (2011), WO 2011038293 A1 20110331.

ACS Paragon Plus Environment

2

9

Article Doi

Improved Synthesis of the Nav1.7 Inhibitor GDC-0276 via a Highly Regioselective SNAr Reaction

DOI: 10.1021/acs.oprd.9b00082

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.oprd.9b00082

Full text of DOI:10.1021/acs.oprd.9b00082

Products guided by the article

R&D Labs maybe for 1354961-13-9

Relevant to this article

Hot Product