Utilizing o-Quinone Methide Chemistry: Synthesis of d9-Ivacaftor

Article abstract of DOI:10.1021/acs.joc.9b02552

Lead time and cost are important factors for any pharmaceutical API. However, these issues become even more important when the drug substance contains an isotope such as deuterium, which has a natural abundance of only ～0.016% of all hydrogen. Fewer suppliers and logistical barriers both play a role in driving up the cost. These factors can challenge the supply route used to manufacture d₉-ivacaftor (17), requiring investigation into alternative routes. By adapting the work from Pettus et al., a synthetic approach utilizing a transient o-quinone methide allowed access to the deuterium-labeled o-tert-butylphenol moiety. This was developed and proven on pilot scale to significantly reduce the number of deuterated reagents used, leading to an overall reduction in cost by a factor of 10, while also providing the substantial benefit of applying prior process knowledge from the parent, nonisotopically enriched API ivacaftor (7).

Full text of DOI:10.1021/acs.joc.9b02552

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
Utilizing o-Quinone Methide Chemistry: Synthesis of d9-Ivacaftor
Adam Robert Looker, Nathan Wilde, Michael P. Ryan,
Stefanie Roeper, Zhifeng Ye, and Bérénice L. Lewandowski
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b02552 • Publication Date (Web): 17 Dec 2019
Downloaded from pubs.acs.org on December 18, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 21
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Utilizing o-Quinone Methide Chemistry: Synthesis of
d9-Ivacaftor
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
Adam R. Looker,†,* Nathan Wilde,# Michael P. Ryan,‡ Stefanie Roeper,† Zhifeng Ye,§ Bérénice L.
Lewandowski,†
Process Chemistry and Engineering & Materials Sciences, Vertex Pharmaceuticals Incorporated, 50
Northern Avenue, Boston, Massachusetts 02210
* Author for correspondence. E-mail: adam_looker@vrtx.com
† Process Chemistry, Vertex Pharmaceuticals, Inc. Boston, MA
# Process Chemistry, Vertex Pharmaceuticals, Inc. San Diego, CA
‡ Engineering & Materials Sciences, Vertex Pharmaceuticals, Inc. Boston, MA
§ Technical Operations, Vertex Pharmaceuticals, Inc. Boston, MA
ACS Paragon Plus Environment
1

The Journal of Organic Chemistry
Page 2 of 21
Table of Contents Graphic
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Abstract: Lead time and cost are important factors for any pharmaceutical API. However, these issues
become even more important when the drug substance contains an isotope such as deuterium, which has
a natural abundance of only ~0.016% of all hydrogen. Fewer suppliers and logistical barriers all play a
role in driving up the cost. These factors would challenge the supply route used to manufacture d9-
ivacaftor (17), requiring investigation into alternative routes. By adapting the work from Pettus et al., a
synthetic approach utilizing a transient o-quinone methide allowed access to the deuterium-labelled o-
tert-butyl phenol moiety. This was developed and proven on pilot scale to significantly reduce the number
of deuterated reagents used, leading to an overall reduction in cost by a factor of 10, while also providing
the substantial benefit of applying prior process knowledge from the parent, non-isotopically enriched
API ivacaftor (7).
Introduction
Incorporating deuterium into clinical candidates has become more common over recent years. 1 While
substituting hydrogen with deuterium in the proper location has its advantages, 2 there are also significant
associated drawbacks; specifically, there are few suppliers of bulk sources of deuterium, deuterium is
expensive relative to its hydrogen counterpart, and various region/country specific regulations exist,
making it relatively difficult to move raw materials around the world. As a result, lead time and cost of
APIs with deuterium will typically be negatively impacted. For these reasons, it is important to be as
ACS Paragon Plus Environment
2

Page 3 of 21
The Journal of Organic Chemistry
efficient as possible with the source and use of deuterium reagents to mitigate these challenges and prevent
the API price from being too high.
1
2
3
4
5
6
7
8
Ivacaftor (7) was commercialized in 2012 by Vertex Pharmaceuticals Incorporated under the trade name
Kalydeco® for the treatment of Cystic Fibrosis (CF) patients with the G551D mutation.3 It has since been
used in combination with other agents to treat larger populations of CF patients.4 The synthetic route to 7
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
5
is provided in Scheme 1. Commercially available 1 is converted to its methyl carbonate 2, setting the
stage for a selective nitration using AlCl3. Reduction of the nitro moiety, followed by coupling with the
quinoline 5 gives the protected ivacaftor 6, where treatment with sodium methoxide and crystallization
from aqueous CH3CN provides 7. This route has provided ~45 MT since 2012.
Scheme 1. Synthesis of Ivacaftor
d9-Ivacaftor (17) presents potential advantages with respect to 7 that are being clinically evaluated. 6
Several synthetic pathways were identified to install the d9-t-butyl, 7 with a clinical development supply
route identified as shown in Scheme 2. Installation of the d9-t-butyl is accomplished via Friedel-Crafts
Alkylation using d9-t-butanol. To minimize exchange of the deuteriums with protons, 8 is first converted
to 9. In both the protection step and alkylation step, deuterated reagents and solvents were required. The
ACS Paragon Plus Environment
3

The Journal of Organic Chemistry
Page 4 of 21
reaction mixture of the alkylation step contained over alkylated by-product 11, but this can be
circumvented by brominating the mixture with NBS, leading to the bromination of both 10 and 11 to
provide 12. From here, a similar sequence is followed as in Scheme 1 to deliver 17.
Scheme 2. Supply Route for d9-Ivacaftor
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although the initial supply route was successful in delivering the needed quantities of material for
clinical use, the raw material costs associated with it were very high and sourcing the five required
deuterated materials provided significant challenges. In addition, there was a desire to take advantage of
the knowledge gained from the synthesis of 7, leveraging the end-game chemistry and Quality by Design
(QbD) work employed in generating its design spaces and specifications. The question therefore became:
how can we best harmonize the synthesis of d9-ivacaftor with the existing route to 7?
ACS Paragon Plus Environment
4

Page 5 of 21
The Journal of Organic Chemistry
Results and Discussion
Modifying the End-Game Chemistry
1
2
3
4
5
6
7
8
9
In both syntheses (Schemes 1 & 2), an intermediate aniline (4, 15) is produced. It was expected that
there would be no deuterium exchange applying the ivacaftor 7 end-game chemistry starting with an
appropriate deuterated intermediate aniline 15 (Scheme 3). There are significant logistical advantages
with using the commercial chemistry, as the current 7 supply chain of other key intermediates could be
utilized, as well as current procedures and analytical methods with specifications. During the development
of 7, QbD was utilized to establish design spaces for the hydrogenation, coupling and deprotection
chemistry.
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. Modified Supply Route for d9-Ivacaftor
A familiarization run was performed using the center points of the coupling and deprotection steps
starting with aniline 15. After confirming the process was amenable, a series of verification experiments
were performed as part of a comparability protocol with the aim of showing that the design spaces
established for 7 would be applicable to the synthesis of 17. Key to this is evaluating if our models from
the design spaces of the ivacaftor commercial synthesis predict accurately the performance for the
synthesis of d9-ivacaftor 17. Figure 1 shows that the data from verification experiments run on the coupling
reaction of 15 and 5 (Scheme 3) align nicely with the in-process control model generated during enhanced
ACS Paragon Plus Environment
5

The Journal of Organic Chemistry
Page 6 of 21
development studies of the reaction to 7. Similarly, the models for other unit operations were verified,
providing confidence that our established design spaces from the manufacture of 7 could be applied to the
manufacture of 17.
1
2
3
4
5
6
7
8
9
Figure 1. Location and Results of Verification Experiments (Denoted with O) Relative to Coupling
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
Model Mean and Prediction Intervals
This modified supply route was used to produce >200 kg of 17 to support clinical studies. However,
there was still room for improvement, especially with respect to cost effectiveness.
ACS Paragon Plus Environment
6

Page 7 of 21
The Journal of Organic Chemistry
Screening for and Optimizing an Alternative Route
1
2
3
4
5
6
7
8
9
Phenols 18 and 19 were identified as abundant and inexpensive commercially available raw materials
for an alternative route to 15. As Pettus and colleagues have shown, 8 a disconnection involving a transient
o-quinone methide intermediate, that can be used to capture a nucleophile, is an excellent way of
converting an aromatic ester to its t-butyl derivative. Utilizing d6-acetone addition to 18 provides an
alternative pathway to the same o-quinone methide intermediate by way of the tertiary alcohol as disclosed
in this paper. If either was successful, this would allow entry of the d9 version of 2,4-di-t-butylphenol (the
raw material used in the synthesis of 7), allowing us to take full advantage of our knowledge of the
ivacaftor commercial route (Scheme 1). It would also decrease the number of required deuterated reagents
from five (Scheme 2) to one or two (Scheme 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
Scheme 4. Potential Access to d9 Phenol 20 from Commercially Available Raw Materials
Screening studies were initiated to investigate the conversion of 18 to 21, followed by a separate study
examining the conditions required to convert the tertiary alcohol to 22 (Scheme 5). The more conveniently
available MeMgI was used for this study in place of CD3MgI. Using nBuLi (in hexanes) as the lithium
ACS Paragon Plus Environment
7

The Journal of Organic Chemistry
Page 8 of 21
source, and varying equivalents, solvent, co-solvent and the amount of d6-acetone, the results in Table 1
were obtained.
1
2
3
4
5
Scheme 5. Screening the Viability of the d6-Acetone Route
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
Table 1. Screening Studies for the Lithiation and Addition of d6-Acetone
nBuLi
(equiv)
2.0
d6-Acetone
(equiv)
Co-solvent
% Product
Exp #
1
2
3
4
5
6
7
8
9
1.05
1.05
1.05
1.05
1.05
1.05
1.05
1.05
1.5
None
67
71
33
44
70
27
82
65
81
2.0
Et2O
2.0
THF
2.0
2-MeTHF
CPME
2.0
2.0
1,4-Dioxane
MTBE
2.0
2.0a
2.0
MTBE
MTBE
a For entry 8, nBuLi was sourced as a solution in cyclohexane
This initial data set indicated that MTBE provided good results as a co-solvent (entries 1 – 7). Increasing
the equivalents of electrophile had no impact (entry 9). With a strong desire to minimize the d6-acetone
used, we began to investigate the downstream chemistry using the conditions of entry 7. Although
telescoping into the next step made sense to evaluate, for screening purposes, intermediate 21 was isolated
and subjected to further conditions to install the final methyl moiety (Table 2). In each experiment, 2 eq
nBuLi (in hexanes) was used as base with 10 vol solvent as indicated.
ACS Paragon Plus Environment
8

Page 9 of 21
The Journal of Organic Chemistry
Table 2. Screening Studies for Conversion of 21 to 22
1
2
3
4
5
Electrophile Additive
(equiv)
MeMgI %
(equiv) Product
Exp # Solvent
(equiv)
6
7
8
9
1
2
3
THF
THF
THF
PivCl (1.3)
BzCl (1.3)
TFAA (1.0)
None
None
None
1.5
1.5
1.3
43
40
40
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
CuBr-SMe2
(0.1)
4
5
6
THF
THF
THF
TFAA (1.0)
TFAA (1.2)
TFAA (1.2)
1.3
1.5
1.5
1.5
63
52
56
CuBr-SMe2
(0.1)
CuBr-SMe2
(0.5)
CuBr-SMe2
(1.0)
7
8
9
THF
THF
THF
TFAA (1.2)
TFAA (1.2)
TFAA (1.2)
51
9
MgBr2 (0.1) 1.5
Pd(PPh3)4
1.5
27
(0.1)
Pd(PPh3)4
1.5
10
11
12
THF
THF
TFAA (1.2)
TFAA (1.2)
TFAA (1.2)
17
5
(0.5)
DMAP (1.1) 1.5
1:1
None
1.5
2
THF/heptane
13
14
1:1 THF/DCM
MTBE
TFAA (1.2)
TFAA (1.2)
None
None
1.5
1.5
16
9
Trifluoroacetic anhydride (TFAA) furnished the best conversion, but other electrophiles also gave
appreciable product (entries 1 & 2). Several additives were evaluated under the hypothesis they might
stabilize an o-quinone methide. Copper bromide dimethyl sulfide complex did show a positive effect on
the yield (entries 4 – 7). There is some discussion in the literature that an o-quinone methide might react
9
as a diradical, so it was hypothesized that solvent could have an effect. However, efforts to try mixed
solvent systems and/or changing solvent polarity only led to lower yields (entries 12 – 14).
ACS Paragon Plus Environment
9

The Journal of Organic Chemistry
Page 10 of 21
Because of the reactivity of the transient o-quinone methide, the nucleophilic trap needed to be
immediately available, and that was a challenge to this approach using d6-acetone. This raised concerns
about the route’s reproducibility upon scale-up. Thus, attention was turned to screening the process as
described by Pettus starting with ester 19. 8 Toward that end, BOC protection of 19 provided 23 without
incident (Scheme 6). A screening study was setup to evaluate equivalents of CD3MgI, Grignard reagent
solvent choice and co-solvent. The results are shown in Table 3. Note: in all cases, addition of substrate
to Grignard was performed to achieve high concentrations of Grignard to substrate over the course of the
reaction, consistent with the findings of Pettus.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 6. Screening the Viability of the Conversion of 19 to 20
Table 3. Screening Studies for Conversion of 23 to 20
CD3MgI
(equiv)
Grignard
(Reagent Solvent, (volumes)
Molarity)
Co-Solvent
%
Product
Exp #
1
2
3
4
5
6
7
8
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
2-MeTHF (3.4M)
Toluene (1.4M)
Et2O (3.0M)
None
THF (11)
None
62
72
51
73
50
82
48
6
Et2O (3.0M)
THF (10)
MTBE (6)
THF (8)
Et2O (3.0M)
n-Bu2O (2.7M)
n-Bu2O (1.0M)
n-Bu2O (2.1M)
n-Bu2O (10)
Toluene (15)
ACS Paragon Plus Environment
10

Page 11 of 21
The Journal of Organic Chemistry
1,4-Dioxane
1
2
3
4
5
6
7
8
9
5.0
5.0
5.0
n-Bu2O (2.1M)
6
(15)
Toluene/THF
3:1 (11.6)
10
11
n-Bu2O (2.7M)
n-Bu2O (2.7M)
40
70
Toluene/THF
1:1 (11.6)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
12
13
14
15
16
17
3.2
3.5
4.1
4.3
4.6
6.0
n-Bu2O (2.5M)
n-Bu2O (2.7M)
n-Bu2O (2.4M)
n-Bu2O (2.7M)
n-Bu2O (2.5M)
n-Bu2O (2.5M)
THF (3)
THF (6)
44
61
75
74
77
79
THF (10)
THF (9)
THF (10)
THF (11)
While the Grignard could be sourced in several solvents, ethereal solvents were desired due to
commercial availability. THF as a co-solvent led to higher yields and more consistent results (entries 2, 4
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
and 6). A clear shift in the Schlenck equilibrium occurred with THF as co-solvent, as a MgI2-THF
insoluble complex formed, leading to a beneficial system containing the more reactive (CD3)2Mg. 11 The
volume of THF was important, though, because too little THF created a problematic thick slurry. Other
additional solvents that were tried did not prove as fruitful. In choosing between Et2O and n-Bu2O as the
solvent source of the Grignard solution (for example, entries 4 & 6), n-Bu2O was more attractive due to
less risk with flammability.
While both alternative routes provide a viable pathway to 20, the approach as described in Scheme 6
using the screening conditions from Table 3 entry 6 were taken forward for further development. Prior to
conducting a proof of concept on pilot scale, a few key questions remained, namely around the ideal
temperature of the reaction and whether the process could be telescoped into the downstream chemistry.
An evaluation of reaction temperature was warranted to understand what temperature the reaction
requires and understand if the impurity profile can be impacted. A Mettler Toledo React-IR 15 instrument
running iC IR software version 7.0297 was used to effectively evaluate in a single experiment the
ACS Paragon Plus Environment
11

The Journal of Organic Chemistry
Page 12 of 21
temperature at which the reaction takes place. Figure 2 outlines the experiment and key changes in the IR
as the temperature was stepped up from -60 °C to -20 °C. In this case, the entire substrate solution was
added to the Grignard at -60°C and monitored for changes in the IR spectrum as the temperature was
raised in stages. The iC IR software has a feature to automatically identify wavenumbers of interest that
correlate to statistical variation in the spectra over time during the experiment. This feature, built upon the
ConcIRT algorithm, was utilized to identify several trends in the spectra during a period at the end of the
experiment where substrate was reacting at a constant temperature and concentration (>1.5 hours). Once
identified, the useful wavenumbers (1625 cm-1 is believed to correlate with the product and 1762 cm-1
with the substrate) could be tracked throughout the experiment. It was noted that the reaction does start to
proceed around -40°C. Separate experiments were performed at selected temperatures from -30°C to
+20°C to evaluate differences in impurity profile. All reactions provided similar product and impurity
profiles, with major impurities analogous to those previously described by Pettus (tertiary alcohol 24 and
styrene 25, Figure 3). Due to simplicity and logistical preference, a temperature of 15 – 25°C was selected
avoiding the need for low-temperature conditions.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
12

Page 13 of 21
The Journal of Organic Chemistry
Figure 2. React-IR Profile in which a Half-portion of Ester 23 was Added and the Reaction Warmed
1
2
3
in Stages Prior to Adding the Remaining Substrate.
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 3. Major Impurities Formed During Grignard Reaction
A simple quench/workup sequence was identified. Addition of the reaction to 6N HCl, followed by
aqueous washes and a thiosulfate wash, produced the crude product, which was assessed in the
downstream chemistry without further purification. Towards this end, subjection of crude 20 to the
ACS Paragon Plus Environment
13

The Journal of Organic Chemistry
Page 14 of 21
manufacturing process according to Scheme 1 led to nitro aromatic 3 meeting all specifications. While
using crude 20 was proven viable, a purification of 20 by distillation was developed to establish a clear
breakpoint in the supply chain. It was advantageous to our supply chain to source the phenol and ship this
to our desired manufacturing locations as needed, rather than back integrate the nitration chemistry where
20 was being made.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Pilot Scale Proof of Concept and a Key Finding
With a process in hand, an evaluation was sought at pilot scale to assess robustness, produce material
for required project studies and ensure viability for future production. Starting with 21.6 kg of 19, the
protection chemistry to manufacture 23 ran smoothly, producing a THF solution containing 31.3 kg 23
(97.7% yield) that met specifications. This solution was charged into the mixture of CD3MgI/n-Bu2O (5.0
eq at 2.7 M) and THF (8 vol), achieving the desired conversion (no detected 23) and impurity profile
(>80% product 20) at the end of reaction. After quenching and aqueous washes, the product was distilled
to provide 13.9 kg 20 (64% yield). Analytical data of the purified product (two fractions via the
distillation) showed an overall combined purity of >95% with a d9 incorporation in the t-butyl of >98%,
as determined by mass spectrometry.
Examination of the 1H NMR led to an interesting discovery. The integration of the proton ortho to the
phenol was lower than expected. Further investigation showed that in fact, we had ~20% deuterium
incorporation (compound 27). This is likely due to exchange chemistry occurring during the distillation.
12 The source of the deuterium is styrene impurity 24, produced during the conversion of 23 to 20 by way
of a [1,5] sigmatropic shift competing with the desired addition of the third methyl Grignard. During the
high-temperature distillation, deuterium can be incorporated into the system by way of additional [1,5]
shifts involving styrene impurity 24 (Scheme 7), providing deuterium incorporation on the phenol of 26
by way of o-quinone methide intermediate 25. 13 Any deuterium incorporated on the exchangeable phenol
of 26 or its isotopologues would also be distributed among the phenol and ortho positions of 20 via
exchange chemistry.
ACS Paragon Plus Environment
14

Page 15 of 21
The Journal of Organic Chemistry
Scheme 7. [1,5] Sigmatropic Shift of Styrene Impurity During Purification Allowing Deuterium
1
2
3
into the System
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
Due to this discovery, an HCl wash sequence was installed as the first unit operation of the next step in
the downstream chemistry (refer to Scheme 1, preparation of the methyl carbonate) to remove the
exchangeable deuterium from 27, converting back to 20. This process was proven to provide 20 with no
detectable ortho or phenolic deuterium.
Conclusion
An alternative route to d9-ivacaftor 17 was identified and developed to reduce the number of required
deuterated reagents from five to one, while harmonizing the synthetic routes to 17 and 7 to start from the
analogous 2,4-di-t-butylphenol. The cost of API production was reduced by a factor of 10. The new
synthetic route took advantage of work by Pettus et al., using a transient o-quinone methide to produce
the deuterium-labelled o-tert-butyl phenol. The product 20 contained high levels of deuterium
incorporation (>98% d9) in the t-butyl, with no detectable deuterium elsewhere in the molecule once
understanding was gained of the exchange that occurs during the distillation. This project was a good
reminder of the exchangeability of protons and alkyl groups under acidic conditions (and especially during
a Friedel-Crafts alkylation) with electron rich aromatics. This should be kept in mind regarding stable
label API’s and preparation of deuterated reference standards.
ACS Paragon Plus Environment
15

The Journal of Organic Chemistry
Page 16 of 21
1
2
3
Experimental
4
5
6
7
8
9
General. HPLC conditions: Analyses were performed using a Waters Symmetry Shield RP-18, 50 x
4.6mm, 3.5µm column (Part #186000177). Mobile phase consisted of a gradient method using A: 0.1%
(v/v) phosphoric acid (85%) / water and B: 0.1% (v/v) phosphoric acid (85%) / acetonitrile. Assays were
determined using high purity reference standards. NMR spectra were collected using a Bruker 400 MHz
spectrometer. Solvents and reagents were obtained from commercial sources and used as is without
further purification, unless otherwise noted. All reactors are standard multi-purpose equipment, either
glass-lined or stainless steel, unless otherwise noted. All reactions were carried out under an atmosphere
of nitrogen, unless otherwise noted.
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
4-(Tert-butyl)-2-(2-hydroxypropan-2-yl-1,1,1,3,3,3-d6)phenol (21). To an oven dried J-Kem culture
tube was charged a magnetic stir bar. The tube was capped with a septa. The reactor was purged with
nitrogen and dropped in a -78°C bath. To the reactor was charged nBuLi (5.0 ml, 8.0 mmol, 1.6 M in
hexanes), followed by a solution of 18 (0.94 g, 4.0 mmol) in MTBE (2.5 ml). The mixture was stirred for
2 hrs. To the reaction was charged d6-acetone (0.46 g, 8.0 mmol) dropwise over 30 min. The reaction was
allowed to slowly warm to ambient temperature overnight. The reaction was quenched and washed
sequentially with saturated NH4Cl, water and then saturated NaCl. The organic phase was concentrated to
dryness to provide an oil that was a mixture of products. This mixture was purified by SFC
chromatography to provide the title compound as an oil with a yield of 70% (0.62 g (99.1% purity): 1H
NMR (400 MHz, CDCl3): δ 8.74 (s, 1H), 7.22 (dd, J = 8.5, 2.5 Hz, 1H), 7.10 (d, J = 2.4 Hz, 1H), 6.84 (d,
J = 8.4 Hz, 1H), 2.40 (s, 1H), 1.31 (s, 9H). C{1H} NMR (100 MHz, CDCl3): δ 153.3, 142.1, 130.1,
13
125.8, 121.9, 117.0, 76.0, 34.1, 31.6, 29.3 (m, 2CD3). HRMS (ESI/Q-TOF) m/z: [M - H]- Calcd for
C13H13D6O2 213.1767; Found 213.1840.
4-(Tert-butyl)-2-(2-methylpropan-2-yl-1,1,1,3,3,3-d6)phenol (22). To a 250 ml round bottom flask
equipped with a magnetic stir bar, temperature probe and nitrogen bubble was charged 21 (4.3 g, 20.06
ACS Paragon Plus Environment
16

Page 17 of 21
The Journal of Organic Chemistry
mmol), followed by THF (86 ml). The mixture was cooled to -70°C in a dry ice/acetone bath. To the
reactor was charged dropwise nBuLi (25.01 ml, 40.1 mmol, 1.6 M in hexanes), and stirred for 20 min. To
the reactor was charged dropwise trifluoroacetic anhydride (2.80 ml, 20.1 mmol), and stirred for 35 min.
To the reactor was charged CuBr DMS (0.41 g, 2.00 mmol), followed by dropwise addition of CH3MgI
(8.69 ml, 26.1 mmol, 3.0 M in Et2O). The reaction mixture was allowed to warm slowly to ambient
temperature. The reaction was quenched with saturated NH4Cl and transferred to a separatory funnel. The
mixture was extracted with IPAc (86 ml), and the aqueous phase back extracted with an additional IPAc
(86 ml). The combined organics were washed with water, then saturated NaCl and dried over Na2SO4.
After filtration, the mixture was concentrated to dryness to provide an oil that was a mixture of products.
This mixture was purified by SFC chromatography to provide the title compound as an oil with a yield of
51% (2.2 g (98.0% purity): 1H NMR (400 MHz, CDCl3): δ 7.29 (d, J = 2.5 Hz, 1H), 7.08 (dd, J = 8.2, 2.5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hz, 1H), 6.60 (d, J = 8.2 Hz, 1H), 4.59 (bs, 1H), 1.41 (s, 3H), 1.30 (s, 9H). C{1H} NMR (100 MHz,
13
CDCl3): δ 151.9, 143.1, 135.3, 124.2, 123.7, 116.0, 34.4, 31.8, 29.6, 28.8 (m, 2CD3). HRMS (ESI/Q-TOF)
m/z: [M - H]- Calcd for C14H15D6O 211.1974; Found 211.2033.
Methyl 2-((tert-butoxycarbonyl)oxy)-5-(tert-butyl)benzoate (23). 19 (21.6 kg, 103.7 mol) was
charged to a reactor, followed by 4-dimethylaminopyridine (62.1 g, 0.5 mol) and dichloromethane (170.3
kg). The reaction was stirred to dissolve while maintaining a temperature of 25 °C +/- 5.0 °C. Di-t-
butyldicarbonate (23.1 kg, 105.8 mol) as a solution in dichloromethane (53.9 kg) was charged over 1 hour
maintaining a temperature of 25 °C +/- 5.0 °C. The reaction was stirred for 1 hour. IPC confirmed
complete conversion (residual 19 NMT 0.50% area). Ammonium chloride (6.7 kg, 125.3 mol) as a
solution in water (58.1 kg) was charged to the reaction, and the mixture was stirred for 1 hour. The layers
were separated, and the organic retained. To the organic layer was charged water (64.1 kg), and the
mixture was stirred for 1 hour. The layers were separated, and the organic retained. This water wash was
repeated a 2nd time. The organic layer was solvent swapped to THF with successive distillations and
additions of THF. A final solution (84.7 kg) was obtained in THF containing the title compound (31.3 kg,
97.7% yield and 99.2% area purity) which was used as-is in the next reaction: 1H NMR (400 MHz, CDCl3):
ACS Paragon Plus Environment
17

The Journal of Organic Chemistry
Page 18 of 21
δ 8.02 (d, J = 2.5 Hz, 1H), 7.58 (dd, J = 8.5, 2.6 Hz, 1H), 7.12 (d, J = 8.5 Hz, 1H), 3.90 (s, 3H), 1.59 (s,
9H), 1.34 (s, 9H). 13C{1H} NMR (100 MHz, CDCl3): δ 165.5, 151.9, 149.1, 148.3, 130.9, 128.7, 123.0,
122.7, 83.6, 52.3, 34.6, 31.2, 27.7. HRMS (HESI-QEHF) m/z: [M + H]+ Calcd for C17H25O5 309.16965;
Found 309.1685.
1
2
3
4
5
6
7
8
9
4-(Tert-butyl)-2-(2-(methyl-d3)propan-2-yl-1,1,1,3,3,3-d6)phenol (20). THF (212.5 kg) was charged
to a reactor. CD3MgI in dibutyl ether (2.7M, 202.2 kg, 482.0 mol) was charged over 4 hours while
maintaining a temperature of 25 °C +/- 5.0 °C. 23 as a solution in THF from the previous reaction (32.0
kg, 103.7 mol) was charged over 5 hours while maintaining a temperature of NMT 30 °C, rinsing the
source drum for 23 with THF (6 kg). The reaction was stirred for 2 hours. IPC confirmed complete
conversion (residual 23 NMT 0.20% area). The reaction was cooled to 10 °C +/- 5.0 °C. t-Butanol (14.4
kg, 194.3 mol) in dibutyl ether (23.1 kg) was charged over 1.5 hours maintaining a temperature of 10 °C
+/- 5.0 °C and stirred for 20 minutes. To the reaction was charged 6N HCl (163.4 kg) over 2 hours while
maintaining a temperature of NMT 20 °C. The reaction was stirred for 30 minutes. The layers were
separated, drumming off both the lower aqueous and upper organic layers. The aqueous layer was charged
back into the reactor and back extracted with dibutyl ether (137.0 kg), stirring for 30 minutes. The layers
were separated, and the lower aqueous layer removed. The previously drummed organic phase was
charged to the reactor to combine both organics. To the reactor was charged water (27.6 kg), and the
mixture was stirred for 45 minutes. The layers were separated, and the organic retained. This water wash
was repeated a 2nd time. To the reactor was charged 5% sodium thiosulfate (126 kg), and the mixture was
stirred for 45 minutes. The layers were separated, and the organic retained. To the reactor was charged
water (27.6 kg), and the mixture was stirred for 45 minutes. The layers were separated, and the organic
retained. The organic was distilled to minimal volume and transferred to a reactor setup to perform
fractional distillation. Purification was performed using fractional distillation, collecting the product at a
pot temperature of 135 – 155 °C and a head temperature of 118 – 120 °C. The title compound was isolated
as a low melting solid in two batches with a total yield of 64% (11.0 kg (96.8% purity) and 2.9 kg (89.0%
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
1
purity)): H NMR (400 MHz, CDCl3): δ 7.35 (d, J = 2.4 Hz, 1H), 7.14 (dd, J = 8.3, 2.5 Hz, 1H), 6.64 (d,
ACS Paragon Plus Environment
18

Page 19 of 21
The Journal of Organic Chemistry
J = 8.2 Hz, 1H), 1.36 (s, 9H). C{1H} NMR (100 MHz, CDCl3): δ 151.8, 143.0, 135.2, 124.1, 123.6,
115.9, 34.3, 31.7, 28.6 (m, 3CD3). Deuterated purity by LC-MS: >98% d9. HRMS (HESI-QEHF) m/z:
[M]+ Calcd for C14H13D9O 215.22301; Found 215.2226.
13
1
2
3
4
5
6
7
8
9
Supporting Information
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 Supporting Information is available free of charge on the ACS Publications website at DOI:
•
1H and 13C NMR of compounds 20, 21, 22 and 23
• HRMS data for compounds 20, 21, 22 and 23
Acknowledgment: Special thanks to Eduard Luss, Joseph Snodgrass and Yanan Peng for their
analytical data, contributions and discussions related to the work described.
References:
(1)
(a) Notman, Nina. “Heavy Drugs Gaining Momentum”. Chemistry World. 2016-06-10. (b)
Ray, P.C.; Pawar, Y.D.; Singare, D.T.; Deshpande, T.N.; Singh, G.P. “Novel Process for
Preparation of Tetrabenazine and Deutetrabenazine”. Org. Process Res. Dev. 2018, 22,
520.
(2)
(a) Liu, J.; Harbeson, S.; Brummel, C.; Tung, R.; Silverman, R.; Doller, D. “A Decade of
Deuteration in Medicinal Chemistry”. Ann. Rep. Med. Chem. 2017, 50, 519. (b)
Katsnelson, A. “Heavy drugs draw heavy interest from pharma backers”. Nat. Med. 2013,
19, 656. (c) Gant, T. “Using Deuterium in Drug Discovery: Leaving the Label in the Drug”.
J. Med. Chem. 2014, 57, 3595.
(3)
“FDA Approves KALYDECOTM (ivacaftor), the First Medicine to Treat the Underlying
Cause of Cystic Fibrosis” (Press release). Cambridge, Massachusetts: Vertex
Pharmaceuticals. 2012-01-31.
ACS Paragon Plus Environment
19

The Journal of Organic Chemistry
Page 20 of 21
(4)
(a) “FDA Approves ORKAMBITM (lumacaftor/ivacaftor) – the First Medicine to Treat the
Underlying Cause of Cystic Fibrosis for People Ages 12 and Older with Two Copies of
the F508del Mutation” (Press release). Boston, Massachusetts: Vertex Pharmaceuticals
2015-07-02. (b) “FDA Approves SYMDEKOTM (tezacaftor/ivacaftor and ivacaftor) – to
Treat the Underlying Cause of Cystic Fibrosis for People Ages 12 and Older with Certain
Mutations in the CFTR Gene” (Press release). Boston, Massachusetts: Vertex
Pharmaceuticals 2018-02-12.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(5)
DeMattei, John; Looker, Adam R.; Neubert-Langille, Bobbianna; Trudeau, Martin;
Roeper, Stefanie; Ryan, Michael P.; Yap, Dahrika Milfred Lao; Krueger, Brian R.;
Grootenhuis, Peter D. J.; Van Goor, Frederick F. “Process for Making Modulators of
Cystic Fibrosis Transmembrane Conductance Regulator”. WO 2010108162A1, 2010.
(6)
(7)
“Vertex and Concert Pharmaceuticals Complete Asset Purchase Agreement for CTP-656”
(Press release). Boston, Massachusetts: Vertex Pharmaceuticals. 2017-07-25.
LeMercier Lewandowski, Berenice; Lewis, Robert; Looker, Adam; Morgan, Adam;
Roeper, Stefanie; Ryan, Michael; Solsona Rocabert, Juan G.; Wilde, Nathan. “Processes
for Making Modulators of Cystic Fibrosis Transmembrane Conductance Regulator”.
WO 2019109021A1, 2019.
(8)
(9)
Jones, R.; Van De Water, R.; Lindsey, C.; Hoarau, C.; Ung, T.; Pettus, T. “A Mild Anionic
Method for Generating o-Quinone Methides: Facile Preparations of Ortho-Functionalized
Phenols”. J. Org. Chem. 2001, 66, 3435.
Bai, W-J.; David, J.; Feng, Z-G.; Weaver, M.; Wu, K-L.; Pettus, T. “The Domestication of
ortho-Quinone Methides”. Acc. Chem. Res. 2014, 47, 3655.
ACS Paragon Plus Environment
20

Page 21 of 21
The Journal of Organic Chemistry
(10) Schlenk, W.; Schlenk, W. Jr. “Über die Konstitution der Grignardschen
Magnesiumverbindungen”. Chem. Ber. 1929, 62, 920.
1
2
3
4
5
6
7
8
(11) (a) Aston, J.G.; Bernhard, S.A. “Composition of the Grignard Reagent”. Nature 1950, 165,
485. (b) Anteunis, M. “Studies of the Grignard Reaction. II. Kinetics of the Reaction of
Dimethylmagnesium with Benzophenone and of Methylmagnesium Bromide-Magnesium
Bromide with Pinacolone”. J. Org. Chem. 1962, 27, 596.
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
(12) (a) Junk, T.; Catallo, W. “Hydrogen isotope exchange reactions involving C-H (D, T)
bonds”. J. Chem. Soc. Rev. 1997, 26, 401. (b) Atzrodt, J.; Derdau, V.; Fey, T.;
Zimmermann, J. “The Renaissance of H/D/ Exchange”. Angew. Chem., Int. Ed. 2007, 46,
7744.
(13) (a) Hansen, H.J. “Aromatic Sigmatropic Hydrogen-Shifts in 2-Vinyl- and 2-Allyl-
phenols”. Helv. Chim. Acta 1977, 60, 2007. (b) Note: another pathway to deuterium
incorporation in the ring is via 18c during the reaction sequence. Conversion of 18c to
known impurity styrene 24, can allow “free” deuterium into the system during reaction.
However, with the aqueous workup of the reaction it is believed that this deuterium is
exchanged prior to distillation, rendering this pathway an in-significant contributor.
ACS Paragon Plus Environment
21