Practical Synthesis of a Stable Precursor for Positron Emission Tomography Imaging Agent 18F-GTP1

Article abstract of DOI:10.1021/acs.oprd.0c00301

18F-GTP1 is a deuterated small molecule positron emission tomography (PET) imaging agent used to visualize tau tangles in Alzheimer's disease patients. The first-generation synthesis of 18F-GTP1's nonradiolabeled alkyl tosylate precursor was plagued by low-yielding steps, inefficient chromatographic purifications, and variable product quality. Due to these limitations, a more robust second-generation route was developed and successfully executed on kilogram scale. A reduction with LiAlD4 incorporated geminal deuterium atoms, while an efficient amidation reaction accessed the key acrylamide coupling partner. Moreover, the tricyclic imidazo[1,2-a]pyrimidine core was assembled via a novel, convergent, and highly selective phosphoramidate-directed annulation. The improved synthesis eliminated all chromatography en route to a high-yielding and reproducible acid-promoted tosylation as the final step.

Full text of DOI:10.1021/acs.oprd.0c00301

Subscriber access provided by CARLETON UNIVERSITY
Full Paper
Practical Synthesis of a Stable Precursor for Positron
Emission Tomography Imaging Agent 18F-GTP1
Nicholas A White, Kyle Clagg, Lauren E. Sirois, Ngiap-Kie Lim,
William A Nack, Paul D O'Shea, Haiming Zhang, and Francis Gosselin
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.0c00301 • Publication Date (Web): 28 Jul 2020
Downloaded from pubs.acs.org on August 2, 2020
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 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Practical Synthesis of a Stable Precursor for Positron
Emission Tomography Imaging Agent 18F-GTP1
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
Nicholas A. White ,*,† Kyle Clagg,*,† Lauren E. Sirois,† Ngiap-Kie Lim,† William A. Nack,§ Paul D.
O’Shea,§ Haiming Zhang,† and Francis Gosselin†
†Department of Small Molecule Process Chemistry, Genentech, Inc., 1 DNA Way, South San
Francisco, California 94080, United States
§Exemplify BioPharma, 3000 Eastpark Boulevard, Suite 500-A, Cranbury, New Jersey 08512,
United States
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 27
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
2

Page 3 of 27
Organic Process Research & Development
1
2
3
4
ABSTRACT 18F-GTP1 is a deuterated small molecule positron emission tomography (PET)
5
6
imaging agent used to visualize tau tangles in Alzheimer’s disease patients. The first-generation
7
8
9
18
synthesis of F-GTP1’s non-radiolabeled alkyl tosylate precursor was plagued by low-yielding
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
steps, inefficient chromatographic purifications, and variable product quality. Due to these
limitations, a more robust second-generation route was developed and successfully executed on
kilogram-scale. A reduction with LiAlD4 incorporated the geminal deuterium atoms, while an
efficient amidation reaction accessed the key acrylamide coupling partner. Moreover, the tricyclic
imidazo[1,2,a]pyrimidine core was assembled via a novel, convergent, and highly selective
phosphoramidate-directed annulation. The improved synthesis eliminated all chromatography en
route to a high-yielding and reproducible acid-promoted tosylation as the final step.
KEYWORDS: Alzheimer’s disease, tau, positron emission tomography (PET), deuterium, amide
coupling, annulation, tosylation, imidazo[1,2,a]pyrimidine
INTRODUCTION
Alzheimer’s disease (AD) is a chronic neurodegenerative disorder characterized by the
accumulation in brain tissue of β-amyloid plaques and neurofibrillary tangles consisting of
aggregated hyperphosphorylated tau protein.1 Genentech tau probe 1 (18F-GTP1) is a small-
molecule positron emission tomography (PET) agent for imaging tau aggregates in AD patients as
a means of tracking disease progression.2,3 The structure of 18F-GTP1 features a
benzimidazopyrimidine core and bears a radiolabeled fluorine (18F) adjacent to two geminal
deuterium atoms. The incorporation of the geminal deuterium atoms in 18F-GTP1 is intended to
18
18
reduce metabolism and minimize formation of free F in vivo. Due to the short half-life of F
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 27
1
2
3
4
5
6
7
8
9
(110 minutes), 18F-GTP1 is prepared on an as-needed basis via reaction of K18F with stable
deuterio-tosylate precursor 1D (Eq 1).4 In accordance with this point-of-care synthetic strategy, a
robust route to deuterio-tosylate 1D was required to provide material for clinical purposes.
N
N
N
N
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
D
D
K18
F
D
D
N
N
18F
(1)
OTs
N
N
1D
18F-GTP1
Stable Tosylate Precursor
The first-generation route to access deuterio-tosylate precursor 1D featured an annulation
between α,β-unsaturated trichloromethyl ketone 4 and 2-aminobenzimidazole 5 to furnish the
tricyclic core 6 (Scheme 1A).5 Next, the trichloromethyl group was converted to a hydroxyl group
via an SNAr reaction and then to the corresponding bromide by treatment with POBr3. The
deuterated piperidine fragment 10D was synthesized by reduction of methyl ester 9 with LiAlD4
followed by removal of the N-Boc group using TFA. The resulting salt 10D•TFA was coupled with
the tricyclic heteroaryl bromide 8 in an SNAr reaction to produce advanced intermediate alcohol
11D. The target F-GTP1 precursor 1D was then accessed via reaction with Ts2O under basic
18
conditions and purified by silica gel column chromatography followed by an additional
crystallization. From 1D, 18F could be incorporated via an SN2 reaction with a source of
nucleophilic radiolabeled fluoride. Upon evaluation of the first-generation route, we noted
reproducibility issues during the scale-up of the tosylation step and of the SNAr reaction between
10 and 8. Furthermore, the low yield of the tosylation and the requirement for chromatographic
purification were clear opportunities for process improvement.
Scheme 1. First-Generation Synthesis of 1D and Retrosynthesis for a More Convergent Route
ACS Paragon Plus Environment
4

Page 5 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
A: First-Generation Route to Tosylate 1D
N
O
NH2
N
H
5
Cl3C
Cl
N
N
N
N
N
O
POBr3
1 N NaOH
3
N
N
N
OEt
N
0-25 °C
97%
PhMe, 110 °C
93%
EtO
CCl3
CCl3
OH
MeCN, 50 °C
86%
Br
MeCN, 60 °C
90%
9
8
2
4
6
7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N
N
N
Br
•CF3CO2H
8
Et3N
1. LiAlD4
0 °C
Ts2O
N
N
N
BocN
O
HN
D
D
D
D
D
D
N
N
OH
OTs
N
DMF, 95 °C
76%
pyridine, 20 °C
44%
OMe
OH
2. TFA
DCM, 20 °C
95%, two steps
N
N
10D•TFA
9
11D
1D
B: Proposed Second-Generation Route to Tosylate 1D
N
N
N
N
D
D
D
D
D
D
N
N
N
N
OTs
OH
OR
N
N
N
N
12D
1D
11D
N
NH2
O
N
H
O
EtO
OH
HN
D
D
BocN
O
5
14
EtO
N
D
D
OR
OR
OMe
13D
15D
9
We proposed a more streamlined approach to deuterio-tosylate 1D wherein a modified
condensation/annulation reaction would assemble the entire carbon framework of the molecule in
one step by coupling an α,β-unsaturated amide 13D with 2-aminobenzimidazole 5 (or an activated
derivative thereof) (Scheme 1B). While this disconnect was attractive due to its convergent nature,
we acknowledged that significant development would likely be required, as this type of annulation
between 2-aminobenzimidazole and amides lacked direct precedent. Once the tricyclic core had
been assembled, we proposed intersecting the previous synthesis by cleavage of the protecting
group (e.g. Bz) to reveal alcohol 11D followed by tosylation to provide target 1D. To access key
acrylamide coupling partner 13D we envisioned an amidation of β-ethoxy acrylic acid 14 and
piperidine 15D. In turn, 15D could be prepared by again reducing commercially available piperidine
methyl ester 9 with LiAlD4 followed by protection of the resulting alcohol and cleavage of the N-
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 27
1
2
3
4
5
6
7
8
9
Boc group. In order to streamline our development timelines and minimize cost, we performed the
majority of our small-scale development work for this new route on the hydrogen analogues of the
deuterated compounds.
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
RESULTS AND DISCUSSION
Preparation of β-Ethoxy-acrylamide 13D
Originally, we utilized LiAlD4 to incorporate deuterium into the piperidine fragment 9, as
LiAlH4 reductions of esters have been documented on kilogram or greater scale.6 However, these
reagents are highly flammable in their solid forms, and control of hydrogen gas evolution upon
quenching is another concern with their usage.7 As an alternative to aluminum hydride reagents,
we explored transition metal catalyzed deuterogenation conditions to achieve the reduction of
methyl ester 9.8 This would offer the advantage of using deuterium gas, which is more readily
available than LiAlD4. A brief initial screen of ruthenium catalysts9 and conditions using hydrogen
gas revealed that the Ru-MACHO® complex (1 mol% relative to substrate) fully converted 9 to
16H at 100 °C in the presence of 1  10 mol% NaOMe in either MeOH or toluene as the solvent,
which was in good agreement with the reported literature conditions.10 We also found that pre-
treatment of the starting material with activated charcoal (20 wt% Norit Supra A in MTBE at 40
°C) was crucial to obtain reproducibly high conversion for the reaction under hydrogen gas.
Table 1. Extent of Deuteration Under Ru Catalyzed Reduction of Ester 9 to Alcohol 16
H
Conditions
BocN
O
BocN
R
R
N
PPh2
CO
Ru
OMe
OH
P
R
R
Ph
2 Cl
9
16
R = H or D
Ru-MACHO®
ACS Paragon Plus Environment
6

Page 7 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Entry
Conditions
%
16 A%b
86.4
D0
D1
D2
D3
D4
≥D2
98
96
78
a
Convb
88
(%)c
0.1
(%)c
2.2
(%)c
95.8
(%)c
0.9
(%)c
1.0
1
2
3
Ru-MACHO® (1 mol%)
NaOMe (1 mol%)
Ru-MACHO® (1 mol%)
NaOMe (3 mol%)
Ru-MACHO-BH® (1
mol%)
97
95.5
0.2
1.7
3.7
94.4
76.1
0.8
1.0
0.9
0.7
100
97.7
20.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4d
5e
Ru-MACHO® (1 mol%)
NaOMe (1 mol%)
LiAlD4 (87 mol%)
74
73.9
97.3
0.0
0.0
0.8
3.1
98.0
95.8
0.3
0.2
0.9
0.9
99
97
100
a Reactions were performed on 2 g (7.8 mmol, entries 3 & 4) or 3 g scale (11.7 mmol, entries 1 & 2) in 10 mL/g
b
toluene over 24 h under 20 atm D2. Defined by 100% - A% of 9 as determined by HPLC analysis at 210 nm UV
detection. c Determined by isotope distribution measured by LCMS subtracted from natural abundance of species. 11
d
Catalyst, base, and solvent were pre-mixed for 0.5 h under nitrogen prior to addition of 9 and D2 gas. e Performed on
2.02 kg scale in 10 mL/g THF at 0 °C over 1 h.
Encouraged by these results, we began evaluating the reaction conditions using deuterium
gas in toluene, but were surprised to observe mixtures of products consistent with both under- and
over-incorporation of deuterium by LCMS analysis (Table 1, entries 1 and 2). We suspected that
the D3 and D4 species resulted from deuterium delivery to the enolate tautomer in equilibrium with
the ester in the presence of base. A similar phenomenon of competing β-deuteration has been
reported during α-deuterations of primary and secondary alcohols by D2O and Ru-MACHO®.12 It
is also possible that scrambling of the deuterium gas could be occurring via a hydrogen source
such as the nearly stoichiometric amount of CH3OD/H generated, or from the a-position of 9.
Abstraction of hydrogen from MeOH to form H2 and CO2 has been reported in the absence of base
with the Ru-MACHO-BH® catalyst (which may be operative in Table 1, entry 3).13 Variations to
the optimized conditions were attempted, but we were unable to achieve consistently ≥97% D2
incorporation using catalytic reducing conditions without sacrificing conversion. The best ≥D2
incorporation result was obtained when the catalyst and base were premixed for 30 min prior to
the addition of substrate, but this example still suffered from >20 A% of 9 remaining after 24 h of
reaction time (Table 1, entry 4). Based on the requirement for ≥97% D2 content and the not-yet-
established tolerance for residual starting material 9, we ultimately chose to use LiAlD4 for our
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 27
1
2
3
4
5
6
scale up production. This allowed us to obtain the desired product 16D in quantitative yield with
≥97% D2 incorporation (Table 1, entry 5) after traditional Fieser work-up conditions.14
7
8
9
With a reliable means of installing the geminal deuterium groups towards 1D established,
we initially envisioned preparing the acrylamide annulation partner through amide coupling and
subsequent protection of the pendant alcohol. To that end, we established proof-of-concept for the
coupling of freebase 10H (from Boc-deprotection of 16H) with (E)-3-ethoxyprop-2-enoic acid (14)
to form amide 17H (Eq 2). However, the preparation and purification of this intermediate proved
to be challenging, as standard amide coupling conditions provided assay yields of only 6080%.
Despite a competing decomposition pathway for propenoic acid 14, the use of T3P® (50 wt%
solution in EtOAc) with non-nucleophilic bases in DCM emerged as the best combination. A slow
addition of T3P® in the presence of Et3N allowed us to isolate the unprotected alcohol 17H in a
moderate yield of 71% after silica gel chromatographic purification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
EtO
OH
14
(1.0 equiv)
T3P® (1.05 equiv)
O
Et3N (3.0 equiv)
HN
H
H
(2)
EtO
N
H
H
OH
DCM, 0 °C, 4 h
71%
OH
10H
17H
In light of the challenges observed, we revisited our strategy. First, we planned to protect
the piperidine alcohol as a benzoyl acid ester. This would allow us to monitor the piperidine
fragment by UV-HPLC throughout the sequence and especially during the acidic annulation
reaction. We also hoped that the aromatic nature of the benzoyl group would impart improved
crystallinity to the intermediates15 and thereby provide additional opportunities for non-
chromatographic purification relative to alcohols 10 and 17. Second, we additionally decided to
ACS Paragon Plus Environment
8

Page 9 of 27
Organic Process Research & Development
1
2
3
4
protect alcohol 16 prior to removal of the Boc group and peptide coupling, thereby preventing side
5
6
reactions of the nucleophilic alcohol.16
7
8
9
In the forward sense, primary alcohol 16H was smoothly protected using benzoyl chloride
(1.3 equiv) and Et3N(1.5 equiv) in toluene to afford the N-Boc-piperidine benzoate in 99% assay
yield. This reaction mixture, after aqueous workup, was telescoped directly into the subsequent
Boc-deprotection by addition of the toluene process stream to a mixture of HCl in n-PrOH at 0 °C.
The deprotection reaction was then performed at 50 °C for 2 h, at which point the product started
to crystallize. Addition of heptane and cooling to 0 °C allowed for straightforward isolation of
solid 18H•HCl in 79% corrected yield over two steps and with >99.9 A% purity by HPLC (Table
2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Now using 18H•HCl as the peptide coupling partner, we leveraged the T3P® conditions
identified earlier and found that the amidation proceeded in both 2-MeTHF and i-PrOAc (Table 2,
entries 1 and 2). However, conversion plateaued around 92% in i-PrOAc even with excess 14
(Table 2, entry 3). We found that high conversion (>95%) could only be achieved with a
combination of i-PrOH and either Et3N or i-Pr2NEt17 (Table 2, entries 4 and 5). Further
optimization to adjust the stoichiometry of T3P® (from 2.0 to 1.2 equiv) and acid 14 (from 1.0 to
1.3 equiv) as well as the reaction temperature (40 °C) gave consistently 97% conversion (corrected
for relative response factors) (Table 2, entry 6). Isolation of the product was realized by slow
addition of water followed by overnight aging and filtration, thereby providing acrylamide 19H in
94% corrected yield on 99 g scale with >99.0 A% HPLC purity.
Table 2. Protection/Deprotection of 16H and Evaluation of Amide Coupling with 14 using T3P®
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 27
1
2
3
4
5
O
EtO
OH
14
i. BzCl (1.2 equiv)
Et3N (1.5 equiv)
PhMe, 23 °C, 18 h
T3P¨ (2.0 equiv)
Base (5.0 equiv)
•HCl
O
6
BocN
HN
H
H
H H
EtO
N
H
H
ii. AcCl (3.3 equiv)
n-PrOH, 50 °C, 1 h
78% (over two steps)
7
8
OH
OBz
Solvent, 23 °C, 1 h
16H
OBz
18H
19H
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
Entr
ya
1
Base
14
(equiv)
1.0
Solvent
(mL/g)
2-MeTHF
(15)
Conv
71%
19H
(A%)b
74%
Et3N
Et3N
2
1.0
i-PrOAc (5) 91%
75%
3
4
Et3N
Et3N
1.2
1.0
i-PrOAc (5) 90%
66%
91%
i-PrOH (5)
i-PrOH (5)
97%
97%
5
i-
Pr2NEt
Et3N
1.2
1.3
91%
87%
6d
i-PrOH (5)
97%
aPiperidine 18H•HCl (1.9 - 37.1 mmol) was combined with solvent, followed by base, then stirred for 5 minutes after
which was added 14 and lastly T3P®. Determined by HPLC analysis, detection at 220 nm. Performed using 99 g
b
d
(333.5 mmol, corrected for 90.9% qNMR purity) of 18H•HCl, 1.2 equiv T3P®, 40 °C.
Moving forward with the corresponding deuterium-containing intermediates 16D, 20D,
18D•HCl, and 19D, we observed consistent behavior throughout the process while maintaining the
desired ≥95 D2 percentage. The only changes made during scale-up were the addition of catalytic
DMAP (10 mol%) to shorten reaction time for the benzoylation and the incorporation of an
azeotropic drying after this step to remove residual water and Et3N. The improved process allowed
us to produce 1.4 kg of deuterio-19D in 69% overall corrected yield with >99.0 A% purity by
HPLC and no single impurity ≥0.50 A% HPLC (Scheme 2).
Scheme 2. Summary of Improved Process to Deliver β-Ethoxy-acrylamide deuterio-19D
BzCl (1.2 equiv)
Et3N (1.5 equiv)
DMAP (0.1 equiv)
LiAlD4 (0.87 equiv)
THF, 0 °C, 1 h
BocN
O
BocN
BocN
D
D
D D
PhMe, 23 °C, 2 h
OMe
OH
OBz
16D
20D
9
100%
[2 kg]
O
EtO
OH
14
T3P® (1.2 equiv)
O
HN
D
D
NEt3 (5.0 equiv)
AcCl (3.3 equiv)
EtO
N
OBz
18D•HCl
D
D
n-PrOH, 50 °C, 1 h
i-PrOH, 40 °C, 1 h
19D
•HCl
OBz
88% (over two steps)
[2 kg]
88%
[700 g]
ACS Paragon Plus Environment
10

Page 11 of 27
Organic Process Research & Development
1
2
3
4
5
6
Annulation of β-Ethoxy-acrylamide deuterio-19D with 2-Aminobenzimidazole Derivatives
and Synthesis of deuterio-Tosylate 1D
7
8
9
We initially envisioned performing the key annulation reaction between 2-
aminobenzimidazole (5) and acrylamide deuterio-19D. However, conducting the reaction with 5
proved challenging and we subsequently identified phosphoramidate 21 as the optimal coupling
partner to steer the isomeric selectivity of this heterocyclic condensation (vide infra).18 The main
challenge in the synthesis of 21 was controlling the site of N-phosphorylation to favor
functionalization of the exocyclic nitrogen. The choice of base in this reaction was critical in
determining the final distribution of phosphorylated products.19 Through a screen of bases across
a range of pKa’s (Table 3) we found that N-methylimidazole (NMI) provided 21 in the highest
overall yield and selectivity (Table 3, entry 1).
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
We also noted a correlation between the nucleophilicity of the base employed and the final
product distribution. Formation of 21 was favored in the presence of nucleophilic bases such as
NMI, 2,6-lutidine, and pyridine (Table 3, entries 13). When non-nucleophilic bases were used,
the product distribution shifted to favor N2-phosphorylated 22, and by using DBU as base, the
reaction selectivity was fully reversed to 1:99, 21:22 (Table 3, entry 10). An examination of the
NMI-mediated reaction at an earlier time point (1 h) revealed a mixture consisting of both 21 and
22 as well as starting material 5. We propose that species 21 is the thermodynamic product and 22
is the kinetic product, and further hypothesize that in this context, nucleophilic bases could play
the dual roles of both general Brønsted base and phosphoryl transfer reagent (for details, see
Supporting Information). In contrast, when DBU was employed, 22 was the only product observed
over the course of the reaction.
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 27
1
2
3
4
Table 3. Screen of Bases for the Phosphorylation of Aminobenzimidazole 5
5
6
7
8
O
N
OEt
OEt
Base (1.25 equiv)
ClP(O)(OEt)2 (1.25 equiv)
H
N
P
N
N
N
+
NH2
NH2
OEt
N
H
N
H
MeCN (12 mL/g), rt, 16 h
P
OEt
9
O
5
21
22
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
Entr
ya
Conv
21
22
Base
(%)b
(A%)b
(A%)b
21:22
1
NMI
2,6-
lutidine
96
92
0
>99:1
2
3
83
73
73
40
3
3
96:4
93:7
pyridine
4
5
DMAP
TMP
91
84
49
10
25
52
66:44
16:84
6
7
TMG
Et3N
DABCO
TBD
90
82
84
73
99
3
8
3
2
1
51
54
53
45
93
6:94
13:87
5:95
4:96
1:99
8
9
10
DBU
aReactions conducted on 330 mg scale (2.5 mmol). bDetermined by HPLC analysis, detection at 210 nm.
Isolation of phosphoramidate 21 proved to be straightforward as the reaction was
conducted in MeCN and proceeded as a slurry-to-slurry transformation. Upon cooling to 0 °C,
product 21 has a solubility of 12 mg/mL in the reaction medium. This moderately high solubility
led to some loss (12%) to the mother liquor but excellent chemical purity of the isolated solids.
Over the course of development experiments, 21 was consistently isolated in >99.9 A% HPLC
purity (210 nm) and 72-80% yield.
The annulation reaction between acrylamides such as 19 and aminobenzimidazoles was an
unknown transformation at the outset of this project. A number of literature reports outlined the
annulation of α,β-unsaturated ketones (and α,β-unsaturated aldehydes) with aminobenzimidazole,
ACS Paragon Plus Environment
12

Page 13 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
but to our knowledge there were no examples involving acrylamides.20 An initial attempt to
perform the condensation of 19H and 5 under basic conditions yielded no desired product. We
hypothesized that activation of the amide to the imidoyl chloride would impart the desired
reactivity. To test this idea, POCl3 was added to a solution of acrylamide 19H and
aminobenzimidazole 5 in MeCN at 0 °C and then heated to 80 °C over 2 h. The desired annulation
product was formed in 63% assay yield as a 75:25 mixture of isomeric products 24H/25H favoring
the desired benzo[4,5]imidazo[1,2-a]pyrimidin-2-amine isomer (Table 4, entry 1). In a key
experiment, we treated 19H and 5 independently with POCl3 and then combined the reaction
streams and observed an improvement in isomeric selectivity to >91:9, again favoring 24H over
the benzo[4,5]imidazo[1,2-a]pyrimidin-4-amine isomer 25H. LCMS analysis of this reaction
mixture indicated multiple N-phosphorylated products from the reaction of 2-aminobenzimidazole
(5) with POCl3. We then set out to identify an isolable analogue of the in situ generated
phosphorylated species that was responsible for the increase in isomeric selectivity.
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
To this end, a series of functionalized aminobenzimidazoles were screened under the
reaction conditions (Table 4). We identified phosphoramidate 21 as the optimal partner, providing
75% assay yield and 98:2 isomeric selectivity for the desired heterocyclic product 24H when the
reaction was conducted in MeCN (Table 4, entry 2). A phenoxy substituted phosphoramidate was
also evaluated and provided product in similar yield and isomeric selectivity (Table 4, entry 4).
We chose to pursue optimization with ethoxy substituted phosphoramidate 21 for better mass
efficiency. Notably, when isomeric phosphoramidate 22 was tested the product selectivity inverted
to favor 25H in a 67:33 ratio (Table 4, entry 3).18
Despite additional optimization, we observed oiling out when the reaction of 21 and 19H
was conducted in MeCN (Table 4, entry 5). In parallel we also evaluated a variety of solvents and
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 27
1
2
3
4
5
6
7
8
9
found that a 1:1 mixture of MeCN/2-MeTHF was optimal in providing product in 95% assay yield
and 98:2 isomeric selectivity for 24H while maintaining a homogenous solution throughout the
course of the reaction (Table 4, entry 6). An inverse quench of the reaction mixture into 25 wt%
K2CO3 at 25 °C provided a safe and controlled workup protocol as well as a clean phase split.21 In
order to streamline our processing operations, the crude quenched reaction mixture was telescoped
directly into the subsequent debenzoylation step.
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 4. Optimization of the Annulation Reaction
O
POCl3
Et3N
N
N
N
N
R1
NH
N
N
H
H
N
N
+
+
EtO
N
H
H
OBz
N
Solvent (10 v)
0–80 °C, 4.5 h
R2
OBz
N
19H
24H
25H
H
H
BzO
Entry
Benzimidazole
(R1, R2)
Equiv
POCl3
24H+25H
a
Equiv Et3N
Solventb
MeCN
(%)c
24H/25H
1
2
5 (H, H)
21 (PO(OEt)2,
H)
22 (H,
PO(OEt)2)
23 (PO(OPh)2,
1.5
1.5
1.5
1.5
2.1
2.1
2.1
2.1
2.1
2.1
0.5
0.5
0.5
0.5
0.5
0.5
0
63
75
40
68
94
95
94
92
94
95
75:25
MeCN
MeCN
MeCN
98:2
33:67
98:2
98:2
98:2
98:2
98:2
98:2
98:2
3
4
H)
21 (PO(OEt)2,
5
H)
MeCN
1:1 2-
MeTHF/MeCN
1:1 2-
MeTHF/MeCN
21 (PO(OEt)2,
6
H)
21 (PO(OEt)2,
7
H)
21 (PO(OEt)2,
8
H)
0.5
0.5
0.5
EtOAc
2-MeTHF
THF
21 (PO(OEt)2,
9
H)
21 (PO(OEt)2,
H)
10
aReactions performed on 100 mg scale (0.30 mmol 19H, 0.33 mmol aminobenzimidazole), POCl3 added to reaction
b
mixture containing acrylamide, benzimidazole, and Et3N. 1:1 MeTHF/MeCN was the only solvent system that
provided a homogeneous mixture over the course of the reaction. cYield determined by HPLC assay (254 nm).
Our first process design for the debenzoylation reaction incorporated a solvent swap from
2-MeTHF/MeCN to EtOH followed by the addition of NaOEt (21 wt% in EtOH, 2.25 equiv).
ACS Paragon Plus Environment
14

Page 15 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Heating this solution of 24H at 70 °C for 1 h provided >99% conversion to alcohol 11H. Solubility
screening indicated that 3:1 H2O:EtOH would be a promising solvent system22 for the
crystallization of 11H, thus after completion of the debenzoylation reaction, the mixture was
concentrated and H2O was added. This isolation protocol initially looked promising, but upon
stress-testing the process we identified a decomposition liability for 24 during the first solvent
swap to EtOH. Although 24 is stable to the acidic conditions encountered at the end of the
annulation reaction (>5 days at 80 °C jacket temperature), after K2CO3 quench, 24 slowly
decomposes at temperatures above 50 °C. We propose that under acidic conditions, the
nucleophilic sites of the heterocycle are protected by protonation, but upon basification they
become reactive towards benzoyl transfer, leading to new unstable intermediate(s) and multiple
destructive side-reactions.23
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To minimize this liability, we redesigned the debenzoylation/isolation protocol to first
accomplish the debenzoylation to 11 before performing the solvent swap, thus avoiding exposure
of 24 to prolonged heating under neutral/basic conditions. To this end, we performed a
development reaction on 50 g scale with deuterated substrate 24D. The crude reaction stream from
the annulation reaction was quenched as above, and then 2.2 equiv NaOEt (21 wt% in EtOH) was
added and the mixture was heated to 70 °C for 1 h. This process provided >99% conversion to 11D
(95 A% by HPLC). After the debenzoylation was complete, a solvent swap to 3:1 H2O:EtOH (10
mL/g) was performed. Finally, the material was crystallized by heating the mixture to 70 °C
followed by cooling to 0 °C (Eq 3). This modified protocol provided 11D in 87% isolated yield
and >99 A% HPLC purity over the two steps.
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 27
1
2
3
4
5
6
7
8
i. NaOEt (2.2 equiv, 21 w% in EtOH)
N
N
N
R
R
2-MeTHF/MeCN (1:1), 70 °C, 1 h
R
R
N
N
OBz
OH
N
N
N
ii. H2O/EtOH (3:1) 70–0 °C, 7 h
(3)
from annulation reaction
24D, R=D
24H, R=H
11D, R=D
11H, R=H
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The primary challenge encountered in the tosylation reaction was the formation of N-
tosylated and bis-tosylated impurities,24 which plagued the first-generation synthesis. Initial
experiments using TsCl under basic conditions led to incomplete conversion and significant
amounts of the N-tosylated and bis-tosylated impurities (Table 5, entry 1). We then explored the
use of Lewis acid and Brønsted acid promoted conditions with Ts2O as the tosylating agent (Table
5, entries 23). Use of Brønsted acid TsOH•H2O completely shut down the unproductive N-
tosylation and bis-tosylation pathways and provided clean conversion to desired product (Table 5,
entries 46). Reaction optimization revealed 1.4 equiv TsOH•H2O and 2.0 equiv Ts2O in 6 mL/g
MeCN to be optimal, providing 97% conversion and 96 A% tosylate 1H based on HPLC analysis
(Table 5, entry 6). We propose that this high chemoselectivity is achieved via in situ protection of
the heterocycle nitrogen atoms via protonation, which then allows only the less basic hydroxyl
group to undergo tosylation. We also found that the order of addition is critical in this reaction and
that it is essential to premix alcohol 11H and TsOH•H2O for a minimum of 10 minutes to ensure
complete protonation of 11H prior to the addition of Ts2O.25
Table 5. Optimization of the Tosylation Reaction of 11 to 1
additive
tosylating agent
N
N
N
N
R
R
R
R
N
N
OH
OTs
N
MeCN (6 mL/g), 16 h
N
11D, R=D
11H, R=H
1D, R=D
1H, R=H
Ent
rya
Tosylating Agent
(equiv)
Additive (equiv)
1 (A %)b
ACS Paragon Plus Environment
16

Page 17 of 27
Organic Process Research & Development
1
2
3
4
5
1
2
Et3N (5.0)
TsCl (1.05)
Ts2O (1.2)
55
Yb(OTf) (0.15)
trace
6
7
8
9
Yb(OTf)/TFA
(0.15/1.2)
3
Ts2O (1.2)
Ts2O (1.2)
47
80
4
5
6
7
TsOH•H2O (1.2)
TsOH•H2O (1.2)
TsOH•H2O (1.4)
TsOH•H2O (1.4)
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
Ts2O (1.4)
Ts2O (2.0)
90
96
TsCl (1.05)
80
aReactions conducted on 300 mg scale (1.0 mmol) 11. bDetermined by HPLC analysis,
UV detection at 210 nm.
The first-generation synthesis of deuterio-1D relied on chromatographic purification and
provided material of variable quality, which led to inconsistencies in the downstream SN2
fluorination. Initial screening indicated that deuterio-1D has limited solubility in the majority of
solvents at 25 °C. 26 Due to the presence of the alkyl tosylate, we could not heat deuterio-1D above
40 °C without observing decomposition. Furthermore, we were unable to identify a water-
immiscible organic solvent suitable for an aqueous workup to remove the p-toluenesulfonic acid
byproduct. Therefore, we opted to pursue a procedure wherein excess Ts2O was first quenched by
the addition of H2O (1 mL/g) followed by direct neutralization/crystallization via slow addition of
the crude MeCN reaction stream to aqueous 20 wt% K3PO4. This procedure led to the smooth
crystallization of 1, but also resulted in a triphasic mixture. In order to avoid multiphasic liquid
layers, we explored adding a water-miscible organic co-solvent (6 mL/g) to the reaction mixture
prior to the crystallization.27 Ultimately, we identified NMP as the co-solvent of choice as it
consistently provided deuterio-1D in >99.0 HPLC A% purity and >90% isolated yield as a free-
flowing crystalline solid.
CONCLUSION
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 27
1
2
3
4
5
6
7
8
9
In summary, we developed an improved synthesis of the deuterio-tosylate precursor to 18F-
GTP1 via a 7-step synthetic sequence from commercially available materials (Schemes 2 and 3).
Deuterium was incorporated via reduction of ester 9 with LiAlD4, then the alcohol product
deuterio-16D was further elaborated to acrylamide deuterio-19D via T3P® coupling with 14
following protecting group manipulations. The key step of the route featured a novel and highly
selective annulation between acrylamide deuterio-19D and phosphoramidate 21 to deliver the
tricyclic benzo[4,5]imidazo[1,2-a]pyrimidine core of 18F-GTP1 in a 98:2 isomeric ratio. This
intermediate was then debenzoylated using NaOEt, and isolation via crystallization provided
alcohol deuterio-11D. Finally, tosylation under acidic conditions (TsOH/Ts2O), delivered the target
deuterio-1D in 89% yield and 99.3 A% HPLC purity. This new synthesis improved the overall
yield of 1D from 23% to 49% and provided high-quality deuterio-tosylate substrate that can
undergo reproducible radiolabeling with 18F to produce PET imaging agent 18F-GTP1.
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. Overview of Final Synthetic Route to Tosylate deuterio-1D
O
BocN
O
EtO
N
D
D
4 steps
69% yield
OMe
OBz
9
19D
POCl3 (2.2 equiv)
Et3N (0.5 equiv)
N
N
D
NaOEt (21 wt% in EtOH, 2.25 equiv)
D
N
OBz
N
MeCN/MeTHF, 70 °C, 1 h
71% (two steps), [1.0 kg]
MeCN/MeTHF, 0–80 °C, 4 h
ClP(O)(OEt)2 (1.35 equiv)
NMI (1.35 equiv)
O
P
N
OEt
OEt
24D
H
N
N
NH2
N
H
MeCN, 20 °C, 16 h
73% [1.0 kg]
N
H
5
21
TsOH•H2O (1.4 equiv)
Ts2O (2.0 equiv)
N
N
N
D
D
D
D
D
D
N
N
N
N
18F
OH
OTs
N
N
N
N
N
MeCN, 20°C, 16 h
89% [690 g]
11D
1D
18F-GTP1
Experimental Section
2-(Piperidin-4-yl)ethyl-1,1-d2 benzoate hydrochloride (18D). To a reactor (Reactor 1) under nitrogen equipped with
overhead agitation was charged anhydrous THF (20 L) and tert-butyl 4-(2-methoxy-2-oxoethyl)piperidine-1-
ACS Paragon Plus Environment
18

Page 19 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
carboxylate (9, 2.02 kg, 7.85 mol). Reactor 1 was then cooled to 0 °C over 2 h and LiAlD4 (0.29 kg, 6.90 mol, 0.87
equiv) was slowly charged in a portion-wise fashion (Caution: gas evolution!). Upon complete charging of the reagent,
the reaction was stirred at 0 °C for 1 h. After reaction completion was confirmed by HPLC (9/16D A% ≤1.0%), the
reaction was quenched by slow addition of H2O (0.29 L) while maintaining the internal temperature to ≤15 ºC
(Caution: gas evolution!). Reactor 1 was then warmed to 20 °C and stirred for 30 min. Aqueous 4N NaOH (0.29 kg,
0.25 L) was slowly charged to Reactor 1 and the mixture was stirred at 20 °C for 30 min. H2O (0.86 L) was charged
to the Reactor 1 and the mixture was stirred at 20 °C for 30 min. The organic layer was separated. Na2SO4 (2.50 kg)
was charged to Reactor 1 containing the organic layer and the mixture was stirred at 20 °C for 2 h. The suspension
was then filtered through a short pad of Celite® (1.50 kg) then the pad was washed with three separate portions of
THF (10 L). After cleaning Reactor 1, the washes and organic layer were combined and transferred back into Reactor
1, then concentrated under vacuum distillation at 55 °C jacket temperature to afford 16D as a yellow oil. An additional
portion of 16D (0.30 kg) was then charged to Reactor 1. To Reactor 1 containing tert-butyl 4-(2-hydroxyethyl-2,2-
d2)piperidine-1-carboxylate (16D, 2.11 kg, 9.13 mol) was charged toluene (16 L) followed by Et3N (1.90 L, 13.6 mol,
1.5 equiv), and DMAP (0.11 kg, 0.90 mol, 0.10 equiv). Reactor 1 was then cooled to 5 ºC until the internal
temperature achieved <5 °C. Benzoyl chloride (1.38 L, 11.9 mol, 1.3 equiv) was charged slowly over 1 h while
maintaining the internal temperature <5 °C. The reaction mixture was warmed to 25 °C and stirred for 2 h. After
completion of the reaction was confirmed (16D/20D A% ≤1.0%), aqueous 5 w% NaHCO3 (11.8 L) was charged to
Reactor 1 and the mixture was stirred for 10 min. The layers were separated and the aqueous layer was drained. H2O
(11.8 L) was then charged to Reactor 1 and the mixture was stirred for 10 min. The layers were separated and the
aqueous layer was drained. The organic layer was then concentrated and dried via azeotropic vacuum distillation with
65 °C jacket temperature to an overall volume of 4.8 L. After cooling to 23 °C, toluene (11.2 L) was added. The
resulting tert-butyl 4-(2-(benzoyloxy)ethyl-2,2-d2)piperidine-1-carboxylate (20D) was obtained as a toluene solution
(16.0 L) in Reactor 1. Concurrently, to a second reactor (Reactor 2) under nitrogen equipped with overhead agitation
was charged 1-propanol (5.66 L). Reactor 2 was cooled to 0 °C then acetyl chloride (2.14 L, 30.1 mol, 3.3 equiv) was
charged slowly while maintaining the internal temperature <5 °C (Caution: exothermic!). The contents were stirred
between 05 °C for 30 min then tert-butyl 4-(2-(benzoyloxy)ethyl-2,2-d2)piperidine-1-carboxylate (20D, 3.06 kg, 9.13
mol) dissolved in toluene (16.0 L total volume) was transferred from Reactor 1 to Reactor 2. Reactor 1 and its transfer
line were rinsed forward to Reactor 2 with toluene (4.0 L). The mixture was then stirred at 50 °C for 1 h. Upon
confirmation of reaction completion (20D / (20D + 18D) A% ≤2.0% by HPLC analysis), heptane (16.5 L was charged
to Reactor 2 at 50 °C over 30 min to the self-seeded mixture. The contents were aged for 2 h, and cooled to 0 °C over
1.5 h, and stirred at 0 °C for 1 h. The slurry was then filtered and washed with a pre-chilled 1-propanol: heptane (12.0
L, 15:85) solution at 0 °C. The solids were then dried under vacuum at 25 °C for 48 h to afford 2-(piperidin-4-yl)ethyl-
1,1-d2 benzoate hydrochloride (18D) as an off-white solid (2.19 kg, 88% over three steps). HPLC: 99.7% 18D; mp
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
178.86−179.98 °C; H NMR (400 MHz, DMSO-d6) δ 9.348.83 (m, 2H), 8.027.92 (m, 2H), 7.707.61 (m, 1H),
7.597.48 (m, 2H), 3.21 (br d, J = 12.5 Hz, 2H), 2.902.75 (m, 2H), 1.84 (br d, J = 13.4 Hz, 2H), 1.801.69 (m, 1H),
1.681.63 (m, 2H), 1.501.35 (m, 2H); 13C NMR (101 MHz, DMSO-d6) δ 166.2, 133.8, 130.2, 129.6, 129.2, 62.2,
43.3, 34.5, 30.8, 28.6. HRMS: calcd for C14H18D2NO2 [M+H]+ : 236.1614; found 236.1606.
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 27
1
2
3
4
5
6
7
8
9
(E)-2-(1-(3-ethoxyacryloyl)piperidin-4-yl)ethyl-1,1-d2 benzoate (19D). To a reactor under nitrogen equipped with
overhead agitation was charged 2-(piperidin-4-yl)ethyl-1,1-d2 benzoate hydrochloride (18D, 0.70 kg, 2.60 mol), (E)-
3-ethoxyprop 2-enoic acid (14, 0.39 kg, 3.40 mol, 1.3 equiv), 2-propanol (3.5 L), and Et3N (1.80 L, 13.0 mol, 5.0
equiv). The contents were then stirred at 25 °C for 15 min. T3P® (50 w% solution in EtOAc, 1.85 L, 3.09 mol, 1.2
equiv) was charged slowly over 15 min while maintaining the temperature <30 °C. The reaction was then heated to
40 °C and stirred for 1 h. Upon confirmation of reaction completion (18D /(18D+ 19D) A% ≤5.0% by HPLC analysis),
H2O (10.5 L) was charged to the reactor over 2 h. The mixture was cooled to 25 °C over 2 h, aged for 12 h, filtered
and the resulting cake was washed with H2O:2-propanol (4.0 L 2, 3:1). The washed cake was then dried at 45 °C for
44 h to afford (E)-2-(1-(3-ethoxyacryloyl)piperidin-4-yl)ethyl-1,1-d2 benzoate (19D) as a white solid (0.73 kg, 85%).
HPLC: 99.3% 19D; mp 106.18−108.78 °C; 1H NMR (400 MHz, DMSO-d6, 60 °C) δ 7.97 (d, J = 7.3 Hz, 2H), 7.697.62
(m, 1H), 7.567.50 (m, 2H), 7.37 (d, J = 12.0 Hz, 1H), 5.82 (d, J = 11.8 Hz, 1H), 4.17 (br s, 2H), 3.94 (q, J = 6.9 Hz,
2H), 2.78 (br s, 2H), 1.811.64 (m, 5H), 1.25 (t, J = 7.0 Hz, 3H), 1.161.01 (m, 2H); 13C NMR (101 MHz, DMSO-
d6, 60 °C) δ 166.2, 165.4, 160.9, 133.6, 130.5, 129.5, 129.2, 96.7, 66.9, 62.5, 43.7, 34.9, 33.3, 32.5, 14.9. HRMS:
calcd for C19H24D2NO4 [M+H]+ : 334.1982; found 334.1989.
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
Diethyl (1,3-dihydro-2H-benzo[d]imidazol-2-ylidene)phosphoramidate (21). To a 20 L reactor under nitrogen at
20 °C was charged 1H-benzo[d]imidazol-2-amine (5, 0.788 kg, 6.002mol) followed by MeCN (5.3L, 4.2 kg). To this
mixture was then added N-methylimidazole (0.669 kg, 8.10 mol, 1.35 equiv) followed by MeCN (0.46 L, 0.40 kg) to
rinse the reaction vessel. This slurry was then stirred at 20 °C for a minimum of 15 min. Next, ClP(O)(OEt)2 (1.41 kg,
8.15 mol, 1.35 equiv) was added dropwise over 1 h while keeping the internal temperature below 30 °C. MeCN (0.40
kg, 0.46 L) was added to rinse the reaction vessel. This slurry was then stirred at 20 °C for 12 h. Upon reaction
completion, (5 /21 A% ≤6.0% by HPLC analysis) the mixture was cooled to 0 °C over 1 h and then stirred for 2 h.
The mixture was filtered and the wet cake was washed with MeCN (3.2 L  2). The solids were dried under vacuum
at 40 °C for 12 h to afford compound 21 as an off-white solid (1.21 kg, 73%, 97.0 w%). HPLC: 99.5% 21; mp 229 °C
(DSC); FTIR (cm-1, neat) 2979, 2987, 2685, 2146, 2012, 1646, 1313; 1H NMR (400 MHz, Methanol-d4) δ 7.20 (dt, J
= 7.6, 3.8 Hz, 2H), 7.11 (dd, J = 5.9, 3.2 Hz, 2H), 4.07 (p, J = 7.2 Hz, 4H), 1.32 (t, J = 7.1 Hz, 6H); 13C NMR (101
MHz, Methanol-d4) δ 123.4, 111.2, 63.42, 63.4, 16.6, 16.6; 31P NMR (162 MHz, Methanol-d4) δ 7.31. HRMS: calcd
for C11H17N3O3P [M+H]+ : 270.1002; found: 270.1004.
2-(1-(Benzo[4,5]imidazo[1,2-a]pyrimidin-2-yl)piperidin-4-yl)ethan-1,1-d2-1-ol (11D). To a 20 L reactor (Reactor
1) under nitrogen at 20 °C was charged MeCN (3.9 kg, 5.0 L) followed by 19D (1.00 kg, 3.00 mol) and 21 (889 g, 3.30
mol, 1.1 equiv). MeTHF (4.3 kg, 5.0 L) was charged followed by Et3N (152 g, 1.50 mol, 0.5 equiv). The contents of
Reactor 1 were adjusted to 0 °C. POCl3 (1.01 kg, 6.60 mol, 2.2 equiv) was added over 30 min while maintaining the
reaction temperature at 0 °C. The contents of Reactor 1 were heated to 80 °C over 2 h and stirred for an additional 2
h (reaction complete upon 19D/24D) A% ≤4.0% by HPLC analysis). Then the reaction was cooled to 20 °C.
Concurrently, a second reactor (Reactor 2) was charged with aqueous 25 wt% K2CO3 (5 L). The contents of Reactor
ACS Paragon Plus Environment
20

Page 21 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
1 were transferred to Reactor 2 over 15 min while keeping the internal temperature of Reactor 2 below 30 °C. The
mixture in Reactor 2 was then stirred for an additional 15. The layers were allowed to separate the bottom layer was
discharged. 21 wt% NaOEt in EtOH (1.90 kg, 6.00 mol, 2.0 equiv) was added to the reactor and the solution was
heated to 70 °C and stirred for 1 h. The reaction was concentrated to ~10 L via a vacuum distillation (pressure: ~210
mbar, internal temp: 30-45 °C). EtOH (10.0 L, 7.9 kg) was added to the reactor over the course of a constant-volume
distillation while maintaining an internal temperature below 55 °C (pressure: ~210 mbar, internal temp: 30-45 °C).
Then H2O (7.5 L) was added to the reactor over the course of a constant volume distillation while maintaining an
internal temperature below 60 °C (pressure: ~130 mbar, temp: 30-45 °C). An additional 21 wt% NaOEt in EtOH (300
g, 0.75 mol, 0.25 equiv) was added, and the reaction was heated to 75 °C for 1.5 h (reaction complete upon 24D/11D)
A% ≤1.0% by HPLC analysis).. The contents of Reactor 2 were cooled to 5 °C over 7 h, and stirred at 5 °C for 2 h.
The mixture was then filtered and the wet cake was rinsed with H2O:EtOH (4.0 L, 3:1) at 15 °C and H2O (4.0 L) at
20 °C. The solids were dried under vacuum at 40 °C for 12 h to afford 11D as a light-yellow solid (684 g, 71%, 92.3
w/w%). HPLC 97.9% 11D; mp 186 °C (DSC); FTIR (cm-1, neat) 3236, 2926, 2905, 2842, 1640, 1608, 1444, 1375,
1219, 1196; 1H NMR (400 MHz, DMSO-d6) δ 8.93 (d, J = 7.7 Hz, 1H), 7.93 (d, J = 8.0 Hz, 1H), 7.50 (d, J = 8.0 Hz,
1H), 7.28 (t, J = 7.6 Hz, 1H), 7.13 (t, J = 7.6 Hz, 1H), 6.86 (d, J = 7.8 Hz, 1H), 4.55 (s, 2H), 4.36 (s, 1H), 2.99 (t, J =
12.7 Hz, 2H), 1.82 – 1.69 (m, 3H), 1.38 (d, J = 6.2 Hz, 2H), 1.12 (ddt, J = 15.6, 11.8, 6.0 Hz, 2H); 13C NMR (101
MHz, DMSO-d6) δ 157.8, 152.1, 144.1, 134.2, 127.7, 124.0, 118.9, 117.0, 110.1, 95.8, 44.6, 38.7, 32.1, 31.8. HRMS:
calcd for C17H19D2N4O [M+H]+ : 299.1835; found: 299.1836.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-(1-(Benzo[4,5]imidazo[1,2-a]pyrimidin-2-yl)piperidin-4-yl)ethyl-1,1-d2 4-methylbenzenesulfonate (1D). To a
20 L reactor (Reactor 1) under nitrogen at 20 °C was charged 2-(1-(benzo[4,5]imidazo[1,2-a]pyrimidin-2-yl)piperidin-
4-yl)ethan-1,1-d2-1-ol (11D, 0.687 kg, 2.30 mol) followed by TsOH•H2O (0.531 kg, 2.76 mol, 1.2 equiv). MeCN (2.7
kg, 3.4 L) was added and the reaction was stirred for 15 min. Ts2O (1.50 kg, 4.61 mol, 2.0 equiv) was added followed
by MeCN (0.49 kg, 0.63 L) to rinse the reaction vessel. This slurry was then stirred at 20 °C for 16 h. Upon completion
of the reaction (reaction complete upon 11D/1D) A% ≤2.0% by HPLC analysis)., H2O (0.69 kg, 0.69 L) was added
while keeping the internal temperature below 30 °C. NMP (4.3 kg, 4.1 L) was added while keeping the internal
temperature below 30 °C. The reaction mixture was then stirred for 15 min. Meanwhile, to a separate reactor (Reactor
2) at 20 °C was added aqueous 20 wt% K3PO4 (6.9 L). The contents of Reactor 1 were added to Reactor 2 over 2 h.
The contents of Reactor 2 were then agitated for 1 h. The mixture was filtered and the wet cake was washed with H2O
(6.9 L 2). The solids were dried under vacuum at 20 °C for 2 h followed by drying under vacuum at 35 °C for of 12
h to afford 1D as an off-white solid (0.888 kg, 89%, 96.7 w/w%). HPLC: 99.3% 1D; mp 151°C (DSC); FTIR (cm-1,
neat) 1639, 1610, 1438, 1355, 1264, 1204, 1173, 1125; 1H NMR (400 MHz, Chloroform-d) δ 8.23 (d, J = 7.8 Hz, 1H),
7.84 – 7.77 (m, 2H), 7.71 (d, J = 8.1 Hz, 1H), 7.55 (d, J = 8.0 Hz, 1H), 7.37 (dd, J = 7.6, 5.1 Hz, 3H), 7.23 – 7.14 (m,
1H), 6.40 (d, J = 7.8 Hz, 1H), 4.58 (s, 2H), 2.94 (t, J = 12.9 Hz, 2H), 2.46 (s, 3H), 1.73 (s, 1H), 1.60 (d, J = 6.5 Hz,
2H), 1.18 (qd, J = 12.4, 4.1 Hz, 2H); 13C NMR (101 MHz, Chloroform-d) δ 157.8, 145.1, 132.7, 130.0, 128.0, 127.8,
124.8, 119.8, 118.7, 108.8, 108.8, 95.5, 95.4, 67.9, 45.2, 35.2, 32.5, 31.7, 21.8. HRMS: calcd for C24H25D2N4O3S
[M+H]+ : 453.1924; found: 453.1913.
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 27
1
2
3
4
5
AUTHOR INFORMATION
6
7
8
Corresponding Authors
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
*Email: white.nicholas@gene.com
*Email: clagg.kyle@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.
Funding Sources
Any funds used to support the research of the manuscript should be placed here (per journal style).
Notes
Any additional relevant notes should be placed here.
ACKNOWLEDGMENT
We thank Dr. Remy Angelaud (Genentech, Inc.) for many helpful discussions, Lady Mae Alabanza
(Genentech, Inc.) for performing reaction calorimetry for compounds 1D, 11D, and 21, Dr. Kenji
Kurita (Genentech, Inc.) for analysis and determination of deuterated species in the reduction step,
Dr. Antonio DiPasquale (Genentech, Inc.) for X-ray crystallographic analysis of compounds 1D
and 11H, Mark Nguyen (Genentech, Inc.) for assistance in the scale-up for compounds 1D, 11D,
and 21, and Tina Nguyen (Genentech, Inc.) for analytical support.
ABBREVIATIONS
ACS Paragon Plus Environment
22

Page 23 of 27
Organic Process Research & Development
1
2
3
4
REFERENCES
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
Catalyst screening for the reduction of 9, proposal for the formation for 21, experimental details
of non-deuterated analogues, solubility data for compounds 1H and 11H, NMR data of all
compounds, X-ray crystallographic analysis of compounds 1D and 11H.
REFERENCES
(1) Binder, L. I.; Guillozet-Bongaarts, A. L.; Garcia-Sierra, F.; Berry, R. W. Tau, Tangles, and
Alzheimer’s Disease. Biochem. Biophys. Acta, 2005, 1739, 216223.
(2) Bohórquez, S. S.; Marik, J.; Ogasawara, A.; Tinianow, J. N.; Gill, H. S.; Barret, O.;
Tamagnan, G.; Alagille, D.; Ayalon, G.; Manser, P.; Bengtsson, T.; Ward, M.; Williams, S-.P.;
Kerchner, G. A.; Seibyl, J. P.; Marek, K.; Weimer, R. M. [18F]GTP1 (Genentech Tau Probe 1), A
Radioligand for Detecting Neurofibrillary Tangle Tau Pathology in Alzheimer’s Disease. Eur. J.
Nucl. Med. Mol. Imaging, 2019, 46, 20772089.
(3) (a) Chien, D. T.; Bahri, S.; Szardenings, A. K.; Walsh, J. C.; Mu, F.; Su, M.-Y.; Shankle, W.
R.; Elizarov, A.; Kolb, H. C. Early Clinical PET Imaging Results with the Novel PHF-Tau
Radioligand [F-18]-T807. J. Alzheimers Dis. 2013, 34, 457-468, (b) Chien, D. T.; Szardenings,
A. K.; Bahri, S.; Walsh, J. C.; Mu, F.; Xia, C.; Shankle, W. R.; Lerner, A. J.; Su, M.-Y.;
Elizarov, A.; Kolb, H. C. Early Clinical PET Imaging Results with the Novel PHF-Tau
Radioligand [F18]-T808. J. Alzheimers Dis. 2013, 38, 171-184.
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 27
1
2
3
4
5
6
7
8
9
(4) Fowler, J. S.; Wolf, A. P. National Research Council. 1982. The synthesis of Carbon-11,
Fluorine-18 and Nitrogen-13 Labeled Radiotracers for Biomedical Applications. Washington, DC:
The National Academies Press. https://doi.org/10.17226/19636.
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) The corresponding chloride, bromide, and mesylate precursors were also evaluated. Tosylate
precursor 1D was selected as it provided the optimal combination of stability and performance in
the downstream fluorination step.
(6) Magano, J.; Dunetz, J. R. Large-Scale Carbonyl Reductions in the Pharmaceutical Industry.
Org. Process Res. Dev. 2012, 16, 1156−1184.
(7) Anthes, R.; Bello, O.; Benoit, S.; Chen, C.-K.; Corbett, E.; Corbett, R. M.; DelMonte, A. J.;
Gingras, S.; Livingston, R.; Sausker, J.; Soumeillant, M. Kilogram Synthesis of a Selective
Serotonin Reuptake Inhibitor. Org. Process Res. Dev. 2008, 12, 168−177.
(8) Kuriyama, W.; Matsumoto, T.; Ogata, O.; Ino, Y.; Aoki, K.; Tanaka, S.; Ishida, K.;
Kobayashi, T.; Sayo, N.; Saito, T. Catalytic Hydrogenation of Esters. Development of an Efficient
Catalyst and Processes for Synthesizing (R)-1,2-Propanediol and 2-(l-Menthoxy)ethanol. Org.
Process Res. Dev. 2012, 16, 166−171.
(9) See Supporting Information for full list of catalysts and conditions examined.
(10) Werkmeister, S.; Junge, K.; Beller, M. Catalytic Hydrogenation of Carboxylic Acid Esters,
Amides, and Nitriles with Homogeneous Catalysts. Org. Process Res. Dev. 2014, 18, 289−302.
(11) A mixture of species bearing 1, 2, 3, and 4 deuterium atoms over the α- and β-carbons were
observed. We did not characterize these products beyond noting the total number of deuterium
atoms incorporated.
(12) Chatterjee, B.; Gunanathan, C. Ruthenium Catalyzed Selective α- and α,β-Deuteration of
Alcohols Using D2O. Org. Lett. 2015, 17, 4794−4797.
ACS Paragon Plus Environment
24

Page 25 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(13) Monney, A.; Barsch, E.; Sponholz, P.; Junge, H.; Ludwig, R.; Beller, M. Base-Free
Hydrogen Generation from Methanol using a Bi-catalytic System. Chem. Commun. 2014, 50,
707709.
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
(14) Fieser, L. F.; Fieser, M. P.; "Reagents for Organic Synthesis," John Wiley & Sons, 1967; p.
584.
(15) Thakuria, R.; Nath, N. K.; Saha, B. K. The Nature and Applications of π−π Interactions: A
Perspective. Cryst. Growth Des. 2019, 19, 523−528.
(16) Croxall, W. J.; Van Hook, J. O.; Luckenbaugh, R. Transetherification Reactions. Alcohols
with Certain β-Alkoxy Esters. J. Am. Chem. Soc. 1949, 71, 27362740.
(17) We later found that isolations involving i-Pr2NEt had a propensity to gum unless
stoichiometry and addition rates were carefully controlled, whereas isolations utilizing Et3N did
not exhibit this same tendency.
(18) For full details of the optimization and scope of this annulation reaction see: White, N. A.;
Clagg, K.; Sirois, L. E.; Mack, K. A.; Al-Sayah, M. A.; Nack, W. A.; O’Shea, P. D.; Zhang, H.;
Gosselin, F. Phosphoramidates as Steering Elements for Highly Selective Access to
Complementary Imidazo[1,2-a]pyrimidine Isomers. Org. Lett. 2019, 21, 95279531.
(19) Bandyopadhyay, P.; Sathe, M.; Tikar, S. N.; Yadav, R.; Sharma, P.; Kumar, A.; Kaushik,
M. P. Synthesis of Some Novel Phosphorylated and Thiophosphorylated Benzimidazoles and
Benzothiazoles and their Evaluation for Larvicidal Potential to Aedes albopictus and Culex
quiqquefasciatus. Bioorg. Med. Chem. Lett. 2014, 24, 29342939.
(20) (a) Bajwa, J. S.; Sykes, P. J. Synthesis and Structure of Some Azolo[a]pyrimidines, 2,6,7,8-
Tetrahydro-1H-cyclopenta[e]azolo[a]-pyrimidines,
azolo[a]pyrimidines,
6,7-Dihydro-5H-cyclopenta[f]-
7,8-Dihydro-6H-cyclopenta[f]-s-triazolo[4,3-b]pyridazine,
5,6,7,8-
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 27
1
2
3
4
5
6
7
8
9
Tetrahydro-azolo[b]quinazolines, 6,7,8,9-Tetrahydroazolo[a]quinazolines, and 7,8,9,10-
Tetrahydro-s-triazolo[3,4-a]phthalazine. J. Chem. Soc., Perkin Trans. 1, 1979, 30853094. (b)
Künstlinger, M.; Breitmaier, E. Imidazo[1,2-a]pyrimidine from 2-aminoimidazoles and 3-
alkoxyacroleins. Synthesis, 1983, 2, 161-162. (c) Tseng, S.-S.; Epstein, J. W.; Brabander, H. J.;
Francisco, G. A Simple Regioselective Synthesis of Pyrimido[1,2-a]benzimidazoles. J.
Heterocycl. Chem. 1987, 24, 837843. (d) Gao, M.; Wang, M.; Zheng, Q.-H. Concise and High-
Yield Synthesis of T808 and T808P for Radiosynthesis of [18F]-T808, a PET Tau Tracer for
Alzheimer’s Disease. Bioorg. Med. Chem. Lett. 2014, 24, 254-257.
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
(21) Achmatowicz, M. M.; Thiel, O. R.; Colyer, J. T.; Hu, J.; Elipe, M. V. S.; Tomaskevitch, J.;
Tedrow, J. S.; Larsen, R. D. Hydrolysis of Phosphoryl Trichloride (POCl3): Characterization, in
Situ Detection, and Safe Quenching of Energetic Metastable Intermediates. Org. Proccess Res.
Dev. 2010, 14, 14901500.
(22) 86 mg/mL at 70 °C vs 2 mg/mL at 0 °C. See Supporting Information for additional solubility
data for 11H.
(23) Proposed structure of an unstable degradation species for 24H, having undergone benzoyl
transfer to a heterocyclic nitrogen (consistent with LCMS data from crude reaction mixtures).
(24) Proposed structures of N-tosylated and bis-tosylated impurities formed during the base-
promoted tosylation of 11D (consistent with LCMS data from crude reaction mixtures).
ACS Paragon Plus Environment
26

Page 27 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(25) Choudary, B. M.; Chowdari, N. S.; Kantam, M. L. Montmorillonite Clay Catalyzed
Tosylation of Alcohols and Selective Monotosylation of Diols with p-Toluenesulfonic Acid: An
Enviro-Economic Route. Tetrahedron 2000, 56, 72917298.
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
(26) See Supporting Information for solubility data for 1H.
(27) See Supporting Information for details on the co-solvents evaluated for the crystallization
of 1.
ACS Paragon Plus Environment
27