An Alternative Focus for Route Design for the Synthesis of Antibody-Drug Conjugate Payloads

Article abstract of DOI:10.1021/acs.joc.8b02876

An analysis of Antibody-Drug Conjugate Payload manufacturing has revealed that the majority of the cost is associated with the use of high-containment facilities for the latter stages of the synthesis. To make a significant reduction in the Cost of Goods

Full text of DOI:10.1021/acs.joc.8b02876

Subscriber access provided by ECU Libraries
Article
An Alternative Focus for Route Design for the
Synthesis of Antibody-Drug Conjugate Payloads
Arnaud Charles Tiberghien, Philip W. Howard, William Robert
Fraser Goundry, Marc McCormick, and Jeremy Stephen Parker
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.8b02876 • Publication Date (Web): 02 Jan 2019
Downloaded from http://pubs.acs.org on January 3, 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 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
An Alternative Focus for Route Design for the
Synthesis of Antibody-Drug Conjugate Payloads
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
Arnaud C. Tiberghien†, Philip W. Howard†, William R. F. Goundry‡, Marc McCormick‡
and Jeremy S. Parker‡*
†Spirogen, QMB Innovation Centre, 42 New Road, E1 2AX London, United Kingdom
‡Early Chemical Development, Pharmaceutical Sciences, IMED Biotech Unit,
AstraZeneca, Macclesfield, United Kingdom
ACS Paragon Plus Environment
1

The Journal of Organic Chemistry
Page 2 of 43
1
2
3
4
5
TOC Graphic
6
7
8
9
O
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
H
N
N
O
O
O
Original Route
8 High Containment Steps
34 Steps, $$$$$
7
O
O
O
O
H
N
NH
N
H
Route Design
O
O
HO
H
N
N
O
O
H
New Route
3 High Containment Steps
30 Steps, $$
N
N
OMe
MeO
O
O
ACS Paragon Plus Environment
2

Page 3 of 43
The Journal of Organic Chemistry
1
2
3
4
5
Abstract
6
7
8
9
An analysis of Antibody-Drug Conjugate Payload manufacturing has revealed that the
majority of the cost is associated with the use of high-containment facilities for the latter
stages of the synthesis. To make a significant reduction in the Cost of Goods (CoGs), a
new approach to route design has been introduced which focusses on minimizing the
number of steps that require high-containment. This approach has been exemplified in
a new synthesis of tesirine, including the first application of a ring-closing copper
(I)/TEMPO aerobic oxidation to the pyrrolobenzodiazepine ring system, affording a 60%
reduction in CoGs.
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
3

The Journal of Organic Chemistry
Page 4 of 43
1
2
3
4
5
Introduction
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
Antibody-Drug Conjugates (ADCs) are an important new class of oncology
therapeutics, combining a tumor-targeting antibody with a cell-killing cytotoxic drug
(payload).1 Significant activity in this new field has resulted in the approval of
brentuximab vedotin (Adcetris)2 in 2011 and trastuzumab emtansine (Kadcyla)3 in 2013,
followed by the 2017 approvals of inotuzumab ozogamicin (Besponsa) and
gemtuzumab ozogamicin (Mylotarg). Extensive clinical programs continue to evaluate
potential commercial candidates, with further launches expected in the next few years.
O
H
N
N
O
O
O
7
O
O
O
O
H
N
NH
N
H
1
tesirine
O
O
HO
H
N
N
O
O
H
B
A
A
B
C
C
N
N
OMe
MeO
O
O
The cytotoxic drugs employed as ADC payloads are typically based on natural
products, exemplified by the auristatins4 (based on dolastatin 10)5 which have been
used in Adcetris, and the maytansinoids6 (based on maytansine)7 used in Kadcyla.
ACS Paragon Plus Environment
4

Page 5 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
These molecules have high synthetic complexity, and their discovery and development
has required significant contributions from the field of synthetic organic chemistry.
Recent interest in new payloads has been focused on the pyrrolobenzodiazepines
(PBDs),8 which have been employed by Seattle Genetics, Stemcentrx, ADC
Therapeutics and MedImmune in their clinical ADC programs. The leading PBD linker-
payload is tesirine (1, SG3249),9 which is utilized in rovalpituzumab tesirine (Rova-T)10
for the treatment of small cell lung cancer and loncastuximab tesirine (Lonca-T)11 for the
treatment of B-cell non-Hodgkin lymphoma,11 as well as in a range of further programs.
The initial synthesis of tesirine has been described previously,12 as well as a detailed
description of the synthetic development and scale-up of this route.13
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
As part of a detailed review of the route following the delivery of a 60g batch of
tesirine, a full analysis of costs, timings and yields was completed to provide greater
insight into areas for further development. Overall the original route is high yielding and
has been successfully scaled-up, delivering multiple campaigns on multi-gram scales.
However, one area that the analysis highlighted was the cost breakdown between the
chemistry steps performed in high-containment and standard facilities.
High-
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 43
1
2
3
4
5
6
7
8
9
containment is required for molecules that are cytotoxic and these specialist facilities
include features such as controlled air handling (including airlocks); controlled entry and
exit of materials and personnel; specialist containment equipment (gloveboxes for
solids, fume hoods/downflow booths for liquids) and planned cleaning and
decontamination systems. The use of such facilities is expensive and is a major
contributor to any campaign of a cytotoxic linker-payload.
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
For the original route to tesirine (Scheme 1), the final 8 out of the 34 synthetic steps
are performed in a high containment facility (Figure 1). An analysis of the 60g delivery
revealed that this part of the synthetic campaign accounted for 90% of the cost of
campaign, with the initial 26 synthetic steps, performed in standard facilities, only
accounting for 10% of the cost. This cost breakdown is in direct contrast to the
synthesis of a comparable non-cytotoxic molecule prepared using a route of similar
synthetic length and complexity, where the early part of the manufacturing is more
expensive due to the need to use pilot-plant or kilo-lab facilities, whereas the latter steps
can be completed in laboratory glassware.
ACS Paragon Plus Environment
6

Page 7 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Scheme 1. Original synthesis of tesirine detailing chemistry conducted in high-
containment facilities (inside blue box)
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
Alloc
Alloc
OTBS
H
OTBS
H
Williamson
Synthesis
N
N
HO
I
O
N
N
MeO
MeO
7
6
O
O
5 steps
PABC-Ala-Val-Alloc
Alloc
OTBS
TBSO
H
OTBS
H
N
N
TIPSO
O
R
O
O
3 steps
2
3
R = NO2
R = NH2
1
N
N
N
OMe
MeO
O
O
O
8
2 Steps
2 steps (+ coupling)
PABC-Ala-Val-Alloc
PABC-Ala-Val-Alloc
HO
H
HO
Swern
Oxidation
N
OTIPS
HN
OTIPS
N
N
OMe
OMe
4
5
O
O
Figure 1. Schematic of original synthesis of tesirine (each block representing an
intermediate) showing portion conducted in high containment (inside blue box).
ACS Paragon Plus Environment
7

The Journal of Organic Chemistry
Page 8 of 43
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
8

Page 9 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
This analysis clearly demonstrated that in order to make a significant impact on the
cost of manufacturing tesirine, a reduction in the use of high-containment facilities would
be required. To facilitate this, a new approach to route design has been introduced,
focusing on reduction in the number of synthetic steps requiring high-containment
facilities. Route design typically focusses on number of steps, convergency and yield,
but for the large-scale preparation of cytotoxic payloads (and more generally for
cytotoxic molecules), a focus on reduction in the number of steps requiring high-
containment facilities should be a key consideration, both in terms of cost reduction and
improving campaign flexibility and timings.
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
With the completion of the analysis, work was initiated looking to redesign the route
with a focus on reduction in the number of steps requiring high-containment facilities.
Success in this approach offered the potential to afford a new, more efficient route in
terms of cost, time and synthetic efficiency.
ACS Paragon Plus Environment
9

The Journal of Organic Chemistry
Page 10 of 43
1
2
3
4
5
Results and Discussion
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The key structural features imparting cytotoxicity to the PBDs are well understood,14
with a requirement for the A-, B- and C-rings of the molecule to be in place. In the
original synthesis of tesirine, high-containment facilities were required during
construction of the two halves of molecule and for the dimerization and subsequent
stages (Figure 1). For the half of the molecule containing the dipeptide trigger, high
containment was required for the oxidative cyclisation of 4 to 5 to afford the completed
ABC ring system. For the other half of the molecule, a number of the tricyclic
intermediates were tested and found to be non-toxic, but attachment of the 5-carbon
linker to 6 to afford 7 required high-containment. In total, 8 stages needed to be
undertaken in high-containment facilities.
To minimize the number of cytotoxic intermediates and thus the number of steps
requiring high-containment facilities we reasoned that construction of the B-rings should
be moved to as late as possible in the synthesis. To this end, we considered what
would be suitable final steps for the synthesis. Oxidative cyclisation of both B-rings was
ACS Paragon Plus Environment
10

Page 11 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
considered as the final step, but based on previous experience this chemistry would be
incompatible with the sensitive maleimide group. It was thus decided to retain the alloc-
deprotection and attachment of the PEG-maleimide linker as the final two steps, and
investigate the oxidative cyclisation of both B-rings as the antepenultimate step
(Scheme 2). Thus, starting from the previously reported mono-protected dimer (9),15 the
alloc-protected Val-Ala-PAB trigger was attached affording 10, following which the TBS
protecting groups were removed affording diol (11).
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 2. Demonstration of the closure of the B-rings as the antepenultimate step.
Cu(MeCN)4)OTf
bpy, TEMPO
NMI, Air
R1
HN
PABC-Ala-Val-Alloc
Alloc
Alloc
NH
HO
H
OH
H
N
N
R2O
OR2
O
O
O
O
O
O
DCM/MeCN, 78%
N
N
OMe
MeO
N
N
O
O
O
O
R1 = H, R2 = TBS
R1 = PABC-Ala-Val-Alloc, R2 = TBS
R1 = PABC-Ala-Val-Alloc, R2 = H
12
9
PABOH-Ala-Val-Alloc
Triphosgene, TEA, 90%
10
11
LiOAc, DMF, 83%
Identifying effective oxidation conditions which would reliably deliver full cyclisation of
both B-rings proved challenging. A wide-ranging initial screen showed low conversion
with methods such as Swern,16 Pfitzner-Moffat17 and TEMPO/bleach,18 but over-
ACS Paragon Plus Environment
11

The Journal of Organic Chemistry
Page 12 of 43
1
2
3
4
5
6
7
8
9
oxidation with Dess_Martin periodinane19 and sulfur trioxide pyridine.20
TEMPO/bis(acetoxy)iodobenzene (BAIB)21 or Stahl conditions ([Cu(MeCN)4]OTf, bpy,
TEMPO, NMI)22 were identified as the most promising options to get clean oxidative
cyclisation of both B-rings. Attempted optimization of the TEMPO/BAIB showed
significant variability in the reaction and in the purity of the product, thus development
was directed to the use of Stahl conditions. Work focused on optimizing the equivalents
of reagents and ensuring that effective air-sparging was scalable. These conditions
gave consistent and selective oxidations, with >95% conversion to the required product
(12). This intermediate then intercepts the original route for the final two steps to
tesirine (Figure 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
This changed route requires 3 steps to be completed in high-containment facilities, in
contrast to the original route which requires 8 steps. To quantify the full benefits of this
revised route in terms of cost reduction, it was decided to develop the required
transformations to allow the newly demonstrated final steps to be used in a
manufacturing campaign.
ACS Paragon Plus Environment
12

Page 13 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
To this end, efficient routes to bis-TBS ether (10) were evaluated starting from the key
intermediate 2. Synthesis via the mono-protected dimer (9) was evaluated in detail
(Scheme 3), where 2 was deprotected and alkylated with 1,5-diiodopentane following
which nitro-reduction afforded a bis-aniline (13). For this approach to demonstrate
synthetic efficiency, an efficient mono-alloc protection of 13 was required. This
transformation was investigated in detail, including varying reagents, stoichiometries
and conditions (including the use of biphasic systems), but no significant improvement
could be made on the statistical ratio of 1:2:1 bis-aniline:mono-alloc protected
product:bis-alloc protected product. Thus, separate preparation of two halves of the
PBD-dimer was targeted as a more efficient method of preparing 10.
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. Approach to bis-TBS ether (10) via bis-aniline (13).
R1
O
R2
NO2
O
O
O
O
RO
O
TBSO
OTBS
OTBS
1,5-diiodopentane
N
N
N
O
O
R1 = R2 = NO2
2
R = TIPS
R = H
Reduction
Deprotection
R1 = R2 = NH2
13
Attempted
monoprotection
R1 = NHAlloc, R2 = NH2
9
ACS Paragon Plus Environment
13

The Journal of Organic Chemistry
Page 14 of 43
1
2
3
4
5
6
7
8
9
Key intermediate (2) was reduced to aniline (3) using zinc and ammonium acetate in
ethanol. Alternative reagents for this transformation were evaluated and the use of
dithionate initially looked promising, but more detailed analytical work showed that
although effective for completing the reduction, the reagent also caused O- to N-
migration of the TIPS group. Thus the original zinc conditions were adopted and the
resultant material used without further purification. From this intermediate (3), work
started on the separate functionalisation of the two halves of the dimer. For the alloc-
protected half of the dimer, the alloc protecting group was introduced using allyl
chloroformate and sodium hydrogen carbonate in aqueous THF and the product (14)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
used without further purification.
Investigation into removal of the TIPS-group
demonstrated that transformation could be achieved using potassium
carbonate/methanol as well as the more standard lithium acetate/DMF, although the
latter conditions were used for manufacturing as they gave a cleaner reaction profile.
The product phenol (15) was purified using a silica plug. Introduction of the 5-carbon
linker was then evaluated using a range of 1,5-dihalopentanes. The concept of using 1-
bromo-5-chloropentane was investigated as it was reasoned that less of this reagent
ACS Paragon Plus Environment
14

Page 15 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
would be required due to the difference in reactivity between the two ends. However, it
was not possible to develop reproducible conditions using this approach, and the more
standard 1,5-dibromopentane was employed. The reaction conditions and solvent were
evaluated using factorial experimental design, with the optimized conditions using
potassium carbonate in MEK affording >98% conversion. The obtained crude material
was purified by flash column chromatography affording the product (16) with 98% Area
HPLC purity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The half of the PBD dimer with the linker-payload trigger portion was prepared by
coupling 3 with Alloc-Val-Ala-PAB-OH, using CDI as a safer, more convenient
alternative to the standard phosgene conditions to afford the carbamate (17) in high
yields. The TIPS-group was deprotected using lithium acetate/DMF and the resulting
phenol (18) was used without further purification. The two halves of the PBD-dimer
were then coupled together to afford the required bis-TBS ether (10). This reaction was
also optimized using factorial experimental design, with the best conditions utilizing
potassium carbonate in MEK with the addition of tetrabutylammonium iodide (TBAI).
The reaction was slow, typically taking 4-5 days, but a conversion of >98% could be
ACS Paragon Plus Environment
15

The Journal of Organic Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
obtained and the resulting crude material was purified by flash column chromatography
to afford 10 in good purity. Removal of the TBS-groups was achieved simply using
TsOH in aqueous THF, and the resulting diol (11) could be obtained as solid following
precipitation from methyl-THF and heptane. The oxidative cyclisation was then
completed using the previously developed Stahl conditions and the product purified by
reverse-phase HPLC to afford dimer (12). The final steps, common to the original route,
involved alloc-deprotection and coupling of the PEG-maleimide linker, and these utilized
conditions developed in previous campaigns, including purification using reverse-phase
HPLC, to afford tesirine with 98.9% Area HPLC purity. The manufacturing campaign
started with 1.3kg of intermediate 2, and successfully delivered 151g of tesirine
(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. Central section of the revised synthesis of tesirine
ACS Paragon Plus Environment
16

Page 17 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Alloc
NH
Alloc
NH
Allyl
chloroformate
RO
O
Br
O
O
OTBS
OTBS
1,5-Dibromopentane
K2CO3, MEK, 92%
NaHCO3
THF/H2O, 94%
N
N
O
O
R1 = TIPS
R1 = H
LiOAc
DMF/EtOAc/H2O, 93%
14
15
16
K2CO3, TBAI
MEK, 60%
3
10
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
PABC-Ala-Val-Alloc
PABC-Ala-Val-Alloc
PABOH-Ala-Val-Alloc
CDI, DCE, 99%
LiOAc
HN
HN
OH
OTIPS
TBSO
TBSO
DMF, 83%
N
N
O
O
O
O
17
18
With the new route having been demonstrated on a manufacturing scale, an analysis
of the benefit of applying route design with a focus on reduction in the number of steps
requiring high-containment was completed. A direct comparison of Cost of Goods
(CoGs), utilizing $/g for tesirine delivered, showed that the new route afforded material
that was 40% of the cost of the original route. It is worth noting that although the
number of high-containment steps has been minimized to 3 steps (Figure 2, total
number of synthetic chemistry steps in the new route is 30), these steps still represent
50% of the overall cost of the campaign.
ACS Paragon Plus Environment
17

The Journal of Organic Chemistry
Page 18 of 43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Schematic of revised synthesis of tesirine (each block representing an
intermediate) showing portion conducted in high containment (inside blue box).
As noted previously, the concept of applying route design to minimize the number of
steps requiring high-containment is applicable to the synthesis of all cytotoxic
molecules. This concept is of significance for large-scale/commercial synthesis of
cytotoxic molecules, but will be of interest to all chemists working on designing and
developing improved routes to such molecules.
ACS Paragon Plus Environment
18

Page 19 of 43
The Journal of Organic Chemistry
1
2
3
4
5
Conclusions
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In summary, a new approach to route design for the synthesis cytotoxic payloads has
been presented which focusses on the reduction in the number of synthetic chemistry
steps requiring high-containment facilities.
This has been exemplified by the
development of an improved synthesis of the ADC linker-payload tesirine, where
reducing the number of high-containment steps from 8 to 3 has afforded a 60%
reduction in CoGs. This approach has broader applicability to the synthesis of cytotoxic
molecules and provides an important focus for the development of new and improved
routes to molecules with commercial potential.
ACS Paragon Plus Environment
19

The Journal of Organic Chemistry
Page 20 of 43
1
2
3
4
5
Experimental Section
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
General Procedures. All solvents and reagents were commercially obtained and used
without purification. Advanced intermediates were prepared using previously reported
routes by dedicated Contract Manufacturing Organisations. NMR spectra were
recorded on a Bruker 500MHz spectrometer. Accurate mass spectra analyses were
recorded on a Waters LC-MS ZQ 2000 system in positive electrospray ionization mode
(ESI).
(S)-(2-amino-5-methoxy-4-(triisopropylsilyloxy)phenyl)(2-((tert-
butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrol-1-yl)methanone (3):12 To a
solution of (S)-(2-((tert-butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrol-1-
yl)(5-methoxy-2-nitro-4-(triisopropylsilyloxy)phenyl)methanone (2, 237g, 97.8% w/w,
0.40mol) in EtOH (2400mL) was charged zinc dust (234g, 3.58mol, 8.95eq). A solution
of ammonium acetate (285g, 3.60mol, 9.00eq) in water (360mL) was added at a rate to
control the exotherm, maintaining the reaction between 15 and 30°C. When full
conversion was achieved the mixture was filtered to remove the inorganic solids. The
ACS Paragon Plus Environment
20

Page 21 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
residual solids were washed with EtOH (200ml) and the filtered washings added to the
filtered reaction solution. The solution was then concentrated, removing approximately
1500mL of EtOH affording a two-phase mixture. EtOAc (2000mL) and water (1000mL)
were added and the layers separated. The organic phase was washed with 6% w/w
aqueous sodium hydrogen carbonate solution. The combined aqueous phases were
extracted with EtOAc (1000mL) and the organic phases combined. The combined
EtOAc layers were washed with water (2 × 1000mL) and 10% w/w aqueous sodium
chloride solution (2 × 1000mL) then the solvent was removed in vacuo affording the
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
product (3, 214g, w/w assay by H NMR [CD3OD, dimethylsulfone internal standard]
97.6%, assay corrected yield 209g, 95%) as a clear orange oil. 1H NMR (500 MHz,
DMSO) -0.03 – 0.05 (m, 6H), 0.84 (s, 9H), 1.04 (s, 9H), 1.05 (s, 9H), 1.22 (hept, J = 7.4
Hz, 3H), 1.63 (s, 3H), 2.32 – 2.44 (m, 1H), 2.70 (dd, J = 16.0, 10.4 Hz, 1H), 3.60 (s, 3H),
3.63 – 3.89 (m, 2H), 4.41 – 4.52 (m, 1H), 4.97 (s, 2H), 6.03 – 6.22 (m, 1H), 6.31 (s, 1H),
6.61 (s, 1H).
(S)-allyl
2-(2-((tert-butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-
carbonyl)-4-methoxy-5-(triisopropylsilyloxy)phenylcarbamate (14):12 To a solution of
ACS Paragon Plus Environment
21

The Journal of Organic Chemistry
Page 22 of 43
1
2
3
4
5
(S)-(2-amino-5-methoxy-4-(triisopropylsilyloxy)phenyl)(2-((tert-
6
7
8
9
butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrol-1-yl)methanone (3, 200g,
97.6% w/w, 0.36mol) and sodium hydrogen carbonate (54.6g, 0.65mol, 1.80eq) in THF
(1000mL) and water (500mL) was added allyl chloroformate (43.0g, 0.36mol, 1.00eq) at
a rate to control gas evolution. When full conversion was achieved the phases were
separated and orange organic upper phase dried over sodium sulfate, filtered and the
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
solvent removed in vacuo affording the product (14, 227g, w/w assay by H NMR
[CD3OD, dimethylsulfone internal standard] 93.0%, assay corrected yield 211g, 94%) as
1
a clear orange oil. H NMR (500 MHz, DMSO) -0.06 – 0.1 (m, 6H), 0.76 – 0.88 (m, 9H),
0.98 – 1.1 (m, 18H), 1.11 – 1.29 (m, 3H), 1.59 (s, 3H), 2.3 – 2.41 (m, 1H), 2.59 – 2.76
(m, 1H), 3.6 – 3.75 (m, 4H), 3.75 – 3.86 (m, 1H), 4.36 – 4.56 (m, 3H), 5.16 (dd, J = 10.5,
1.4 Hz, 1H), 5.25 (d, J = 17.4 Hz, 1H), 5.87 (ddt, J = 15.9, 10.6, 5.4 Hz, 1H), 5.99 (s,
1H), 6.77 (s, 1H), 7.17 (s, 1H), 8.95 (s, 1H).
(S)-allyl
2-(2-((tert-butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-
carbonyl)-5-hydroxy-4-methoxyphenylcarbamate (15): To a solution of (S)-allyl 2-(2-
((tert-butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl)-4-
ACS Paragon Plus Environment
22

Page 23 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
methoxy-5-(triisopropylsilyloxy)phenylcarbamate (14, 304g, 72.4% w/w, 0.35mol) in
DMF (290mL) and EtOAc (190mL) was added lithium acetate dihydrate (11.6g,
0.11mol, 0.32eq) in water (40mL). When full conversion was achieved the reaction
mixture was diluted with EtOAc (2200mL), then washed with 0.1M citric acid solution
(2000mL), water (2 × 2000mL) and saturated sodium chloride solution (1000mL). The
solvent was removed in vacuo and the crude material dissolved in MeCN (1500mL)
which was extracted with heptane (6 × 1000mL). The solvent was removed from the
MeCN solution in vacuo and the crude product dissolved in MTBE (500mL) and
petroleum ether (500mL) and purified by filtration through a silica plug (200g of silica),
eluting with 1:1 EtOAc:petroleum ether (6000mL). The solvent was removed in vacuo
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
affording the product (15, 206g, w/w assay by H NMR [CD3OD, dimethylsulfone
internal standard] 75.0%, assay corrected yield 155g, 93%) as a yellow oil. 1H NMR
(500 MHz, DMSO) -0.08 – 0.09 (m, 6H), 0.74 – 0.91 (m, 9H), 1.60 (s, 3H), 2.32 – 2.41
(m, 1H), 2.6 – 2.74 (m, 1H), 3.59 – 3.86 (m, 5H), 4.38 – 4.55 (m, 3H), 5.17 (dd, J = 10.5,
1.4 Hz, 1H), 5.27 (dd, J = 17.5, 1.4 Hz, 1H), 5.89 (ddd, J = 22.5, 10.6, 5.4 Hz, 1H), 6.07
(s, 1H), 6.74 (s, 1H), 7.13 (s, 1H), 8.98 (s, 1H), 9.65 (s, 1H). 13C{1H} NMR (126 MHz,
ACS Paragon Plus Environment
23

The Journal of Organic Chemistry
Page 24 of 43
1
2
3
4
5
6
7
8
9
DMSO) -5.5, -5.4, 17.8, 25.6, 31.2, 36.0, 55.9, 58.5, 62.1, 64.7, 110.1, 112.1, 117.2,
120.4, 125.2, 130.0, 133.2, 143.6, 148.4, 153.2, 164.0. HRMS (ESI/QTOF) m/z: [M+H]+
Calcd for C24H37N2O6Si 477.2416; Found 477.2418.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-allyl 5-(5-bromopentyloxy)-2-(2-((tert-butyldimethylsilyloxy)methyl)-4-methyl-2,3-
dihydro-1H-pyrrole-1-carbonyl)-4-methoxyphenylcarbamate (16): A mixture of (S)-allyl
2-(2-((tert-butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl)-5-
hydroxy-4-methoxyphenylcarbamate (15, 303g, 84.2% w/w, 0.54mol), 1,5-
dibromopentane (714g, 3.11mol, 5.80eq) and potassium carbonate (129g, 0.93mol,
1.74eq) in MEK (2000mL) was stirred at 70°C. After 4 hours, in-process analysis
indicated 54% conversion and additional potassium carbonate (111g, 0.81mol, 1.50eq)
was added. After 6 hours, in-process analysis indicated 84% conversion and additional
potassium carbonate (106g, 0.77mol, 1.43eq) was added. After 8 hours, in-process
analysis indicated >98% conversion and the reaction mixture was cooled and filtered.
The solvent was removed from the MEK solution in vacuo and the crude product
slurried with silica gel (857g) and heptane (1000mL) then concentrated to dryness to
give the adsorbed crude product as a powder. This powder was loaded into a glass
ACS Paragon Plus Environment
24

Page 25 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
sinter funnel and eluted with heptane (50L) to remove the remaining 1,5-
dibromopentane, then 1:3 EtOAc:heptane (32L). The solvent was removed from the
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
1
product factions in vacuo affording the product (16, 352g, w/w assay by H NMR
[CD3OD, dimethylsulfone internal standard] 84.6%, assay corrected yield 298g, 92%) as
1
a yellow oil. H NMR (500 MHz, DMSO) -0.08 – 0.09 (m, 6H), 0.73 – 0.87 (m, 9H), 1.44
– 1.54 (m, 2H), 1.59 (s, 3H), 1.67 – 1.79 (m, 2H), 1.84 (dt, J = 14.4, 6.8 Hz, 2H), 2.3 –
2.4 (m, 1H), 2.54 – 2.77 (m, 1H), 3.52 (t, J = 6.7 Hz, 2H), 3.58 – 3.76 (m, 4H), 3.76 –
3.89 (m, 1H), 3.93 (t, J = 6.4 Hz, 2H), 4.38 – 4.59 (m, 3H), 5.12 – 5.21 (m, 1H), 5.21 –
5.32 (m, 1H), 5.88 (ddd, J = 22.5, 10.6, 5.4 Hz, 1H), 6.00 (s, 1H), 6.76 (s, 1H), 7.22 (s,
1H), 9.01 (s, 1H). 13C{1H} NMR (126 MHz, DMSO) -5.5, -5.4, 17.8, 24.3, 25.62, 27.63,
31.2, 31.9, 35.0, 36.0, 55.8, 58.4, 62.0, 64.7, 68.1, 111.5, 117.2, 120.4, 125.0, 129.5,
133.2, 144.9, 149.2, 153.4, 163.6. HRMS (ESI/QTOF) m/z: [M+H]+ Calcd for
C29H46BrN2O6Si 625.2304; Found 625.2309.
N-{[(prop-2-en-1-yl)oxy]carbonyl}-L-valyl-N-[4-({[(2-[(2S)-2-({[tert-
butyl(dimethyl)silyl]oxy}methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl]-4-methoxy-
5-{[tri(propan-2-yl)silyl]oxy}phenyl)carbamoyl]oxy}methyl)phenyl]-L-alaninamide
(17):
ACS Paragon Plus Environment
25

The Journal of Organic Chemistry
Page 26 of 43
1
2
3
4
5
6
7
8
9
To a solution of (S)-(2-amino-5-methoxy-4-(triisopropylsilyloxy)phenyl) (2-((tert-
butyldimethylsilyloxy)methyl)-4-methyl-2,3-dihydro-1H-pyrrol-1-yl)methanone (3, 350g,
90.2% w/w, 0.57mol) in DCE (2500mL) at 23°C was added CDI (113g, 0.68mol,
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
1.18eq).
The mixture was stirred for 30 min then allyl (S)-1-((S)-1-(4-
(hydroxymethyl)phenylamino)-1-oxopropan-2-ylamino)-3-methyl-1-oxobutan-2-
ylcarbamate (241g, 98.3% w/w, 0.63mol, 1.09eq) was added. The reaction mixture was
then heated to 80°C and stirred for 5 days at which time 98% conversion was observed.
The reaction was cooled to ambient temperature and filtered through silica gel (250g).
The silica gel was washed with DCE (500mL), the solutions combined and the solvent
1
removed in vacuo to afford the product (17, 650g, w/w assay by H NMR [CDCl3,
dimethylsulfone internal standard] 84.0%, assay corrected yield 546g, 99%) as an
orange foam. 1H NMR (500 MHz, DMSO) -0.02 – 0.09 (m, 6H), 0.89 (s, 9H), 0.91 –
1.13 (m, 27H), 1.35 (d, J = 6.8 Hz, 3H), 1.64 (s, 3H), 1.94 – 2.11 (m, 1H), 2.36 – 2.51
(m, 1H), 2.63 – 2.79 (m, 1H), 3.78 (d, J = 8.0 Hz, 3H), 3.91 – 3.96 (m, 1H), 3.98 (t, J =
5.8 Hz, 1H), 4.35 – 4.41 (m, 1H), 4.42 – 4.58 (m, 4H), 5 – 5.13 (m, 2H), 5.21 (d, J = 10.5
Hz, 1H), 5.27 – 5.41 (m, 1H), 5.95 (ddt, J = 15.8, 10.5, 5.3 Hz, 1H), 6.03 – 6.26 (m, 1H),
ACS Paragon Plus Environment
26

Page 27 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
7.25 (d, J = 1.1 Hz, 1H), 7.3 – 7.32 (m, 1H), 7.35 (dd, J = 8.4, 3.3 Hz, 2H), 7.64 (d, J =
8.5 Hz, 2H), 8.22 (d, J = 7.0 Hz, 1H), 8.93 – 9.13 (m, 1H), 10.08 (s, 1H). 13C{1H} NMR
(126 MHz, DMSO) -5.5, -5.4, 17.7, 17.8, 18.1, 19.2, 25.6, 30.3, 31.9, 36.2, 44.2, 49.0,
55.9, 59.0, 59.9, 62.1, 64.4, 65.5, 116.9, 118.89, 118.91, 119.6, 121.0, 127.9, 128.56,
128.62, 131.2, 131.3, 133.6, 134.9, 137.2, 137.6, 138.75, 138.81, 156.0, 164.0, 171.0,
171.1. HRMS (ESI/QTOF) m/z: [M+H]+ Calcd for C49H78N5O10Si2 952.5282; Not Found
(only mass for 18 found).
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-{[(prop-2-en-1-yl)oxy]carbonyl}-L-valyl-N-(4-{[({2-[(2S)-2-({[tert-
butyl(dimethyl)silyl]oxy}methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl]-5-hydroxy-
4-methoxyphenyl}carbamoyl)oxy]methyl}phenyl)-L-alaninamide (18): To a solution of N-
{[(prop-2-en-1-yl)oxy]carbonyl}-L-valyl-N-[4-({[(2-[(2S)-2-({[tert-
butyl(dimethyl)silyl]oxy}methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl]-4-methoxy-
5-{[tri(propan-2-yl)silyl]oxy}phenyl)carbamoyl]oxy}methyl)phenyl]-L-alaninamide
(17,
641g, 84.0% w/w, 0.57mol) in DMF (3200mL) and water (64mL) at ambient temperature
was added LiOAc (46.0g, 0.45mol, 0.80eq). When full conversion was achieved the
reaction mixture was diluted with MTBE (3000mL), then washed water (3000mL then
ACS Paragon Plus Environment
27

The Journal of Organic Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
9
2000mL). The aqueous washes were back extracted with MTBE (1500mL) and the
combined organic portions filtered through silica gel (190g). The silica gel was washed
with MTBE (1000mL) and the organic portions combined. The solvent was removed in
vacuo and the crude material dissolved in MeCN (1250mL) which was washed with
petroleum ether (5 × 1250mL). The solvent was removed from the MeCN solution in
vacuo affording the product (18, 503g, w/w assay by 1H NMR [CD3OD, dimethylsulfone
internal standard] 75.0%, assay corrected yield 377g, 83%) as an orange waxy solid.
1H NMR (500 MHz, DMSO) -0.05 – 0.07 (m, 6H), 0.81 – 0.92 (m, 15H), 1.31 (d, J = 7.1
Hz, 3H), 1.62 (s, 3H), 1.98 (dq, J = 13.4, 6.7 Hz, 1H), 2.38 (d, J = 15.5 Hz, 1H), 2.62 –
2.72 (m, 1H), 3.65 – 3.77 (m, 4H), 3.77 – 3.86 (m, 1H), 3.86 – 3.94 (m, 1H), 4.36 – 4.56
(m, 4H), 4.97 – 5.07 (m, 2H), 5.17 (d, 1H), 5.31 (dd, 1H), 5.91 (ddt, J = 15.8, 10.5, 5.3
Hz, 1H), 6.11 (s, 1H), 6.7 – 6.85 (m, 1H), 7.19 (s, 1H), 7.25 (d, J = 8.7 Hz, 1H), 7.31 (d,
J = 8.5 Hz, 2H), 7.60 (d, J = 8.5 Hz, 2H), 8.17 (d, J = 7.0 Hz, 1H), 9.02 (s, 1H), 9.69 (s,
1H), 10.01 (s, 1H). 13C{1H} NMR (126 MHz, DMSO) -5.5, -5.4, 17.80, 17.82, 18.1, 19.2,
25.6, 30.3, 36.0, 49.0, 55.9, 58.5, 59.9, 62.1, 64.4, 65.5, 112.1, 116.9, 118.9, 120.5,
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
ACS Paragon Plus Environment
28

Page 29 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
125.2, 128.6, 130.1, 131.2, 138.8, 143.5, 148.4, 153.4, 156.0, 164.0, 171.0, 171.1.
HRMS (ESI/QTOF) m/z: [M+H]+ Calcd for C40H58N5O10Si 796.3948; Found 796.3949.
N-{[(prop-2-en-1-yl)oxy]carbonyl}-L-valyl-N-(4-{[({2-[(2S)-2-({[tert-
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
butyl(dimethyl)silyl]oxy}methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl]-5-[(5-{4-
[(2S)-2-({[tert-butyl(dimethyl)silyl]oxy}methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-
carbonyl]-2-methoxy-5-({[(prop-2-en-1-yl)oxy]carbonyl}amino)phenoxy}pentyl)oxy]-4-
methoxyphenyl}carbamoyl)oxy]methyl}phenyl)-L-alaninamide (10): To a solution of N-
{[(prop-2-en-1-yl)oxy]carbonyl}-L-valyl-N-(4-{[({2-[(2S)-2-({[tert-
butyl(dimethyl)silyl]oxy}methyl)-4-methyl-2,3-dihydro-1H-pyrrole-1-carbonyl]-5-hydroxy-
4-methoxyphenyl}carbamoyl)oxy]methyl}phenyl)-L-alaninamide (18, 236g, 82.4% w/w,
0.24mol) and (S)-allyl 5-(5-bromopentyloxy)-2-(2-((tert-butyldimethylsilyloxy)methyl)-4-
methyl-2,3-dihydro-1H-pyrrole-1-carbonyl)-4-methoxyphenylcarbamate
(16,
214g,
76.7% w/w, .0.26mol, 1.07eq) in MEK (2500mL) was added potassium carbonate
(34.6g, 0.25mol, 1.03eq) and tetrabutyl ammonium iodide (90.6g, 0.23mol, 0.96eq) and
the mixture was stirred at 50°C. After 29 hours, additional potassium carbonate (17.4g,
0.13mol, 0.52eq) was added. After an additional 43 hours, further potassium carbonate
ACS Paragon Plus Environment
29