Practical Synthesis of MDM2 Antagonist RG7388. Part 2: Development of the Cu(I) Catalyzed [3 + 2] Asymmetric Cycloaddition Process for the Manufacture of Idasanutlin

Article abstract of DOI:10.1021/acs.oprd.6b00319

A concise catalytic asymmetric synthesis of idasanutlin (1) was developed in which the key pyrrolidine core, containing four contiguous stereocenters, was constructed via a Ag/MeOBIPHEP promoted [3 + 2] cycloaddition reaction. Further development of the [

Full text of DOI:10.1021/acs.oprd.6b00319

Subscriber access provided by TUFTS UNIV
Full Paper
Practical Synthesis of MDM2 Antagonist RG7388. Part
: Development of the Cu(I) catalyzed [3+2] asymmetric
cycloaddition process for the manufacture of Idasanutlin
2
Goesta Rimmler, Andre Alker, Marcello Bosco, Ralph Diodone, Dan Fishlock, Stefan Hildbrand,
Bernd Kuhn, Christian Moessner, Carsten Peters, Pankaj D. Rege, and Markus Schantz
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00319 • Publication Date (Web): 31 Oct 2016
Downloaded from http://pubs.acs.org on November 7, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Practical Synthesis of MDM2 Antagonist RG7388.
Part 2: Development of the Cu(I) catalyzed [3+2]
asymmetric cycloaddition process for the
manufacture of Idasanutlin
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
†
Gösta Rimmler, Andre Alker , Marcello Bosco, Ralph Diodone, Dan Fishlock, Stefan
†
Hildbrand, Bernd Kuhn , Christian Moessner, Carsten Peters, Pankaj D. Rege*, Markus
Schantz
Small Molecules Technical Development, F. HoffmannꢀLa Roche, 4070 Basel, Switzerland.
†
Therapeutic Modalities, Roche Pharma Research and Early Developement, Roche
Innovation Center Basel, F. HoffmannꢀLa Roche Ltd., 4070 Basel, Switzerland.
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
GRAPHICS FOR TOC
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CuOAc, (R)-BINAP
Idasanutlin (1)
enantiomeric ratio > 99.9:0.1
overall yield approx. 55%
Exo selective [3+2] asymmetric cycloaddtion
enantiomeric ratio = 99:1
2
ACS Paragon Plus Environment

Page 3 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
KEYWORDS
[3+2]ꢀCycloaddition, Asymmetric, Idasanutlin, Copperꢀcatalyzed, Silverꢀcatalyzed
ABSTRACT
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
A concise catalytic asymmetric synthesis of idasanutlin (1) was developed in which the key
pyrrolidine core, containing four contiguous stereocenters, was constructed via a
Ag/MeOBIPHEP promoted [3+2] cycloaddition reaction. Further development of the [3+2]
cycloaddition reaction resulted in an improvement in diastereoselectivity and
enantioselectivity by changing the catalyst system to Cu(I)/BINAP. While producing
equivalent high quality API, the copper(I) catalyzed process not only increased the overall
yield but also demonstrated benefit with respect to cycle times, waste streams and
processability. The optimized copper(I) catalyzed process has been used to prepare more than
1500 kg of idasanutlin (1).
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Idasanutlin (RG7388, 1, Figure 1) is a mouse double minute 2 (MDM2) homolog protein
antagonist discovered by Roche. MDM2 (also known as E3 ubiquitin-protein ligase
MDM2) is an important negative regulator of the p53 tumor suppressor protein.
Idasanutlin (1) is designed to bind to MDM2 to potentially prevent the p53-MDM2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
interaction, and thereby result in activation of p53. Idasanutlin (1) shows efficacy in
xenograft models and clinical trials of this compound are ongoing. Rapid and durable
responses have been seen in patients suffering from acute myeloid leukemia (AML) in
response to treatment with 1 in combination with cytarabine, and stable disease with
evidence of p53 activation has been observed in early solid tumor studies.
2
CO H
Cl
F
O
N
H
OMe
2
5
3
4
NH
CN
F
Cl
Idasanutlin, RG7388, 1
Figure 1. Left: Structure and absolute stereochemistry of idasanutlin (1). Right: Xꢀray crystal
structure of 1.
As depicted in Figure 1, 1 contains a congested pyrrolidine carboxamide core with four
contiguous stereocenters with all anti relative configuration (2R, 3S, 4R, 5S).
Azomethine ylide-based [3+2] cycloaddition reaction with appropriately substituted
electron-deficient olefins is well established in the literature [2] and provides a logical
synthetic disconnection for 1. The medicinal chemistry synthesis1a employed this
synthetic disconnection in form of racemic [3+2] cycloaddition reaction to construct the
4
ACS Paragon Plus Environment

Page 5 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
pyrrolidine core for structure-activity relationship (SAR) studies and initial pre-clinical
supplies.
1
a
The medicinal chemistry route was not suitable for preparation of larger amounts due
3
to the reasons outlined in Part 1 of this work. Therefore, upon selection of 1 as a
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
clinical candidate process research activities commenced to meet the demands of
clinical trial supply, and eventually develop a commercial manufacturing process. These
efforts initially resulted in an asymmetric cycloaddition reaction using Ag(OAc) with
(R)-MeOBIPHEP as a chiral ligand. Although the Ag-catalyzed process gave 1 in good
quality on a 100 kg-scale, a more efficient process was required to support the
increasing API demand. Continuing development activities which were targeted to
address limitations of the Ag-catalyzed process yielded a Cu(I)/(R)-BINAP variant of
the [3+2] cycloaddition reaction. Herein we report the development of a manufacturing
process for 1 which employs a Cu(I)-catalyzed asymmetric [3+2] cycloaddition
reaction. The optimized Cu(I)-catalyzed synthesis represents the industrial process
which to date, has been used to manufacture more than 1500 kg of 1.
RESULTS AND DISCUSSION
Synthesis of the starting materials for the [3+2] cycloaddition reaction: The imine 4 was
prepared by the condensation reaction between 4-(2-amino-acetylamino)-3-methoxy-
benzoic acid ethyl ester hydrochloride (2) and 3,3-dimethyl butyraldehyde (3) (Scheme
1
).4 The reaction was performed in the presence of triethylamine in methyl-tert
butylether (MTBE) at 20 – 25 °C within 6 h. After separation of precipitated NEt •HCl by
filtration, the imine 4 was obtained by crystallization, following a solvent exchange to
-
3
n-
heptane, in a yield of 94% with excellent quality (99.9% purity). Six batches have been
prepared starting with 110 kg of 2 to produce 703 kg of 4.
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 1: Preparation of 4
Et N, MTBE
Cryst from n-heptane
3
94%
2
3
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Stilbene 7 was prepared by Knoevenagel condensation between aldehyde 5 and 2-(4-
chloro-2-fluorophenyl)acetonitrile (6) (Scheme 2). Originally a stoichiometric amount
of sodium methoxide in methanol at 50 °C for 3 hours was used to effect the
condensation. However, use of 5 mol% of NaOH (50% w/w aqueous solution) instead of
sodium methoxide was also effective. Unfortunately, the methoxy substitution of
fluorine (ortho to the chlorine) in the aromatic ring of product 7 was observed when the
reaction mixture was stirred for extended period of time. Moreover this methoxy-
substituted impurity was not depleted in the downstream chemistry. Further
optimization of reaction conditions showed that catalytic amount of 30% solution of
sodium methoxide in methanol facilitated the reaction at ambient temperature within
only 3 hours and the methoxy-substituted impurity was not formed. Thus, upon slow
addition of the 30% methanolic solution of sodium methoxide (5 mol%) to a 1:1
mixture of 5 and 6 in ethanol containing 0.5 %-w/w water, the desired product 7
precipitated from the reaction mixture and was isolated by filtration. The trace amounts
of water in the mixture were found to significantly improve the filterability of 7. The
only observed side product, with levels up to 6%, was the corresponding E-silbene 8
which does not precipitate from the reaction mixture and was efficiently removed to the
mother liquor during the filtration of 7. This optimized manufacturing process for 7
resulted in a yield of 91% with purity of 99.9%. This process was executed on up to 160
kg batch size of 6 to produce 11 batches of 7 (1818 kg in total).
6
ACS Paragon Plus Environment

Page 7 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Scheme 2: Preparation of 7
3
0% NaOMe/MeOH
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
EtOH/H O, 25°C
2
9
1%
5
6
7
Metal-catalyzed asymmetric cycloaddition process. In the original medicinal chemistry
route, the pyrrolidine core in 1 was constructed in a racemic form using stoichiometric
AgF promoted [3+2] cycloaddition reaction.3 The enantiopure 1 was obtained
employing chiral chromatography. Although that route was not suitable to support
upcoming clinical demands, it provided the basis for the initial process development
work. Analogous to the medicinal chemistry route, the pyrrolidine ring in compound 9c
was constructed using an asymmetric version of the Ag-catalyzed [3+2] cycloaddition
reaction. AgOAc with bidentate phosphine ligands was found to provide the best
selectivity amongs all the systems investigated and eventually AgOAc/(
R)-MeOBIPHEP
catalyst system was chosen.5
6
In the presence of 1.0 mol% of AgOAc and 1.1 mol% (
and 7 in MeTHF at 0° C afforded a mixture of pyrrolidine diastereoisomers 9a, 9b and 9c
total 75% area), the ring-opened Michael addition product 10 (11% area) and a
mixture of imine-hydrolyzed Michael addition products 11 (15% area) after 16 – 24 h
Scheme 3).
R)-MeOBIPHEP 12, reaction of 4
(
(
Upon treatment of this reaction mixture with micronized anhydrous lithium hydroxide
the pyrrolidine diastereoisomers 9a and 9b, as well as the Michael addition product 10,
were converted to the desired ester 9c (er 84:16) which was isolated by crystallization
from
n
-heptane/MeTHF. Hydrolysis of 9c with lithium hydroxide in isopropanol
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
followed by the removal of the insoluble racemate (as lithium salt) by filtration
provided enantioenriched filtrate. Subsequent acidification of the filtrate with acetic
acid and isolation by crystallization from isopropanol/water produced 1 with an er of >
9
9:1 in 44% overall yield from 7.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
ACS Paragon Plus Environment

Page 9 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Scheme 3: Silverꢀcatalyzed asymmetric synthesis of 1 (absolute stereochemistry assigned based on Xꢀray crystal structure analysis, vide infra)
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
a
b
c
4
7
1
9
a, exo, RSRR
9b, endo, SSRS
13%, er = 57:43
45%, er = 98:2
9c, RSRS
er = 84:16
1
2, (R)-MeOBIPHEP
9
c, RSRS
10
11%
17% er = 87:13
11
Reagents and Conditions: a) AgOAc (1.0 mol%), 12 (1.1 mol%), MeTHF, not isolated. b) anhydrous LiOH (milled), 60 °C, then n-heptane, 97% yield. c) i. LiOH, i-PrOH-H O, 70 °C; ii. filter-off lithium salt of rac-1; iii. AcOH-H O, 70 °
2
2
C, 45% yield, er > 99:1
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The Ag-catalyzed process outlined above was successfully employed to produce more
than 100 kg of API to support ongoing clinical trials. Despite the successful supply
campaigns, this process was not considered appropriate for commercial manufacturing
for the following reasons: (a) the efficient and complete epimerization can only be
achieved with finely milled lithium hydroxide, making an extra milling step necessary,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(b) silver oxide and other very fine particulates precipitate during the hydrolysis
process leading to serious filter or centrifuge clogging during racemate filtration, (c)
intermediate reactor cleaning is necessary due to formation of silver deposites on the
reactor walls, and (d) the modest enantioselectivity of the cycloaddition reaction leads
to an erosion of the yield due to enantioenrichment via racemate removal.
The limitations of the Ag-catalyzed approach warranted continuation of development
efforts in search of a more robust scalable process for commercial manufacturing. Along
with Ag-based catalysts, Cu-based catalysts are widely employed in asymmetric [3+2]
2
cycloaddition. In an effort to address the disadvantages of the Ag-catalyzed process,
and primarily to improve the diastereo- and enantioselectivity of the cycloaddition
reaction and issues related to precipitation of Ag salts, the use of Cu catalysts was
3
investigated. As reported by Shu et al, the initial focus of this investigation was Cu(II)-
based catalyst systems, and process using Cu(OAc) /(R)-BINAP (1.3% catalyst loading)
2
delivered 1 with an er of approximately 94:6 on a 30 g-scale. Further
enantioenrichement was achieved by filtration of the racemate from EtOAc/THF,
followed by crystallization of idasanutlin (1).
During further development, the Cu(OAc)
2
catalytic system was found to give
inconsistent results when starting materials (4 and 7) of representative quality were
employed. Experiments on 1 kg-scale confirmed the lack of robustness of the
10
ACS Paragon Plus Environment

Page 11 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
cycloaddition step. These reactions were carried out at room temperature with 1.0
mol% of Cu(OAc) and (R)-BINAP in 2-MeTHF. Consumption of stilbene 7 required 16
2
hours, and the contents of diastereoisomers 9a-c in the reaction mixture varied
substantially between single experiments (70 – 80%) along with side products 11 in the
range of 5 – 15%. The ring opened Michael addition product 10 was present in 1 – 2%.
The ratio between 9a-c (10:0.7:0.85) remained fairly constant. The variability of side
product formation, extended reaction times of the cycloaddition reaction and occasional
stalling of the reaction merited more thorough investigation of the Cu-catalyzed
process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
During their studies of Cu(II)-catalyzed systems, Shu et al had noted that Cu(I)-based
catalyst system also provided encouraging results, so CuOAc was tested as an
alternative to Cu(OAc)
CuOAc in combination with 0.53 mol% (
within 6 hours and the overall amount of diastereomers 9a-c was raised to 90 – 93%
2
in the [3+2] cycloaddition reaction. Using only 0.50 mol%
R
)-BINAP led to smooth conversion of stilbene
3
with a fairly constant ratio between the single diastereoisomers 9a (74 – 76%), 9b (5 –
6%) and 9c (7 – 8%) (Scheme 4). Formation of side product diastereoisomers 11 could
be reduced to 2 – 4% and the ring opened Michael addition product 10 was formed in 3
–
4%. In contrast to Cu(OAc) , the CuOAc system did not require any premixing or aging
2
of the metal–ligand complex. The selectivity of the cycloaddition reaction was found to
be very robust, and within the temperature range of 0 – 40 °C no change in selectivity
was observed.
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
Scheme 4: Cu(I)ꢀcatalyzed asymmetric [2+3] cycloaddition reaction.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
a
b
4
7
9
a, exo, RSRR
9b, endo, SSRS
74-76%, er = 99:1
5-6%, er = 40:60
1
, RSRS
er = 93:7
9
c, RSRS
10
7-8% er = 86:14
3-4%
1
3, (R)-BINAP
Reagents and Conditions: a) CuOAc (0.50 mol%), 13 (0.53 mol%), MeTHF, Et N, not isolated.
3
b) i. THF/Ethanol, NaOH(aq). ii. cryst from iPrOH/H O, 78% yield, er 93:7
2
1
2
ACS Paragon Plus Environment

Page 13 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
To better understand the the Cu-catalyzed reaction, Cu-chiral ligand catalyst systems
were analyzed using 31P NMR. These experiments showed that the complexation of
8
Cu(I) with (
Cu(II) system showed significantly less complexation of the chiral phosphine ligand
with the metal, even after 80 hours. Since the CuOAc/( )-BINAP system provided
robust conditions and desired selectivity and ( )-BINAP is a moderately priced
commercially available ligand, a repetition of the chiral ligand screening was not
initiated and CuOAc/( )-BINAP system was therefore chosen for further development.
R)-BINAP occurs immediately at room temperature. On the other hand,
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
R
R
R
Selectivity in [3+2] cycloaddition reaction. Copper-based catalysts are reported to
deliver exo-selective [3+2] cycloaddition products of metalated azomethine ylides that
lead to 4,5-trans substituted pyrrolidines.2 Indeed, the relative stereochemistry
between electron withdrawing group (—CN) at C4 and the neopentyl group at C5 in the
main product 9a from the cycloaddition reaction is trans. Therefore, the [3+2]
cycloaddition reaction under our conditions can be presumed to proceed analogously as
reported by Garner et al.2d The endo-product 9b was observed in 5 – 6%, and another
product observed in minor amounts 7 – 8% is the diastereomer 9c. The latter can be an
outcome of a non-concerted step-wise mechanism (Michael addition followed by
Mannich ring closure) via a zwitterionic intermediate as postulated by Garner et al.2d
Further evaluation of the stereoisomeric outcome of this cycloaddition addressed the
configuration of the dipolarophile. The Z- and E-stilbenes, 7 and 8, would lead to two
distinct series of diastereoisomers differing in the relative stereochemistry of the two
aromatic rings in 3,4-position of the cycloaddition product (Scheme 5). The Z-stilbene 7
generates pyrrolidine stereoisomers 9 with the desired trans orientation of the
9
aromatic rings, whereas the corresponding E-stilbene 8 leads to product 14 with
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
undesired 3,4-cis orientation. It could be demonstrated that the Z-stilbene 7 is
configurationally stable under the reaction conditions, and as a consequence no isomers
with 3,4-cis configuration were detected in the reaction mixture. Therefore, the
formation of only four diastereomers must be considered.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 5: Zꢀ and Eꢀ Stilbene, 7 and 8, and resulting cycloaddition product mixtures.
cycloaddition
7, Z-Stilbene
3,4-anti aryl-9
(and enantiomer)
hν
benzophenone (cat)
CH Cl (50% yield)
2
2
cycloaddition
8
, E-Stilbene
3
,4-syn aryl-14
(and enantiomer)
Furthermore, for unequivocal structural assignment diastereoisomers 9a-c and ring-
opened product 1010 (Scheme 4) were isolated by preparative chromatography from
the cycloaddition mixture, and characterized by NMR. The isolated diastereoisomers
were further subjected to preparative chiral HPLC separation and then each enantiomer
was fully characterized by NMR and chiral HPLC. In addition, the absolute configuration
for at least one of the two enantiomers of each diastereoisomer 9a, 9b and 9c were
determined by X-ray crystallography (see Supporting Information).11
14
ACS Paragon Plus Environment

Page 15 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
It was found that the main product 9a was formed with high enantioselectivity of 99:1
(with RSRR as the major isomer), while negligible enantioselectivity (40:60) was
observed for the endo-product 9b. The isomer 9c bearing the same configuration as
idasanutlin was formed with moderate enantioselectivity of 86:14 (with RSRS as the
major isomer). The enantiomeric purity of the ring-opened product 10 could not be
determined at the ester stage, but 10 was also converted to idasanutlin (1) during
hydrolysis/isomerization (vide infra) with er >99:1, indicating that the stereocenter at
C3 was formed with a high enantioselectivity in the cycloaddition.
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 undesired constitutional isomer (ethyl ester of 15) could only be generated in trace
amounts (< 1%) by applying catalyst-free, strongly basic conditions (sodium ethoxide
in 2-MeTHF at – 20 °C) (Figure 2). After hydrolysis of the ester, the resulting acid 15
was used as analytical marker to prove the absence of this regioisomer in the
idasanutlin process.
rac-15
Figure 2: Constitutional isomer racꢀ15
Hydrolysis / Isomerization. The main product, 9a, of the Cu(I)-catalyzed reaction differs
from the minor product 9c in the configuration of the C5 position. The product 9c
possesses the same relative configuration as 1. Compound 9b differs from 9c in the
configuration of the C2 position. We knew from the Ag-catalyzed process that C5 and C2
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
positions in 9a and 9b respectively can be epimerized under basic conditions. During
epimerization of compounds 9a and 9b concomitant hydrolysis of the ester bond occurs
to produce 1. This observation suggests that 1 might be the thermodynamically most
stable diastereomer of the four possible diastereomers.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To evaluate our hypothesis, the relative internal energies (in gas phase) of the four
diastereoisomers (9a’, 9b’, 9c’ and 9d’) of a simplified pyrrolidine core (Ar = phenyl) of
1
were calculated quantum mechanically and the results are shown in Figure 3.12 The
relative energy differences of the pyrrolidine core favor the formation of the 9c´
configuration under thermodynamic control. Since the fourth potential diastereoisomer,
which would have the same relative stereochemistry as 9d´ with SSRR/RRSS
configuration, was never observed and considering its high relative internal energy, it
might be concluded that it is either not formed in the concerted cycloaddition or, if
formed, undergoes a rapid retro-Mannich ring opening reaction to give an intermediate
such as 10.
9
c'
9a'
9b'
9d'
RSRS/SRSR RSRR/SRSS
SSRS/RRSR
SSRR/RRSS
ꢀE (kcal/mol)
2.1
5.1
0
.0
0.4
Figure 3: Quantum mechanically calculated relative internal energy differences of the four
model diastereoisomers in gas phase (see Supporting Information for details on the
theoretical method).
When the reaction mixture from Cu(I)-catalyzed [3+2] cycloaddition reaction is
subjected to aqueous sodium hydroxide, following an aqueous workup, compounds 9a–
c and 10 are all converted to 1. Under these conditions, compound 10 rapidly undergoes
16
ACS Paragon Plus Environment

Page 17 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Mannich ring closure to form compound 9c (62%) along with 22% of 9a. The formation
of 9c in the reaction mixture, and rapid conversion of 10 to 9c supports co-existence of a
step-wise mechanism along with concerted mechanism in this Cu(I)-catalyzed [3+2]
cycloaddition reaction. Furthermore, epimerization at C5 position in 9a to form 9c with
aqueous sodium hydroxide can be an outcome of retro-Mannich ring opening followed
by Mannich ring closure cascade. The epimerization at the C2 position in 9b can be
envisioned as deprotonation/protonation to the thermodynamically more stable
diastereomer 9c. Ester hydrolysis was relatively slow with 8 hours required for
complete conversion of 9c to 1. Based on the stereoselectiviy of the cycloaddition
reaction, the anticipated amount of undesired enantiomer SRSR acid after the
hydrolysis/isomerization step is 5 − 8% and the results obtained on manufacturing
scale were well in line with this theoretical value. The step product is consistently
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
3
obtained with 93:7 er after isolation by crystallization from isopropanol/water. All the
undesired isomers along with other related compounds are efficient purged in the
mother liquor at this stage.14 No impurity above 0.1% is detected in the isolated
product. Therefore this stage establishes the primary control of the purity profile of
idasanutlin.
Upon complete hydrolysis/isomerization, the pH of the reaction mixture is adjusted
with acetic acid and the product is crystallized by distillative solvent exchange from
THF/ethanol to isopropanol/water. During early development, this solvent exchange
was conducted in a defined fashion, wherein approximately 25% of total amount of
isopropanol was added in one portion followed by distillation to a desired volume. The
distillation was continued at constant volume by feeding the remaining isopropanol.
Water was then added to the mixture to obtain the final crystallization mixture. On
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
laboratory scale, the acetic acid as well as the first portion of isopropanol were added
without time target by simple temperature control. However, on the manufacturing
scale this procedure led to an extremely poor filterability of the product due to
formation of very fine particles. To address this issue, the concept of the mesomixing
time 15 was applied to adapt the addition times to the manufacturing scale to ensure
good filterability of the product.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To predict the addition times on various scales and manufacturing equipment, a model
for mesomixing time(s) was built. To construct a reliable model, along with different
stirrer types and filling levels, other important parameters such as the density of the
reaction mixture were also included in the model. Using this model (see Supporting
Information for details), addition times for acetic acid and isopropanol were defined
such that the corresponding mesomixing time are within the same order of magnitude
on laboratory and manufacturing scale. The addition times predicted by the model — 3
±
±
1 hours for acetic acid (mesomixing time approx. 0.1 sec, stirrer speed 40 rpm) and 6
1 hours for first portion of isopropanol (mesomixing time approx. 0.13 sec, stirrer
speed 40 rpm), were successfully demonstrated on 100 kg-scale. To date 15 batches
have been prepared to produce 2000 kg of enantioenriched 1 using the process outlined
above.
Preparation of idasanutlin from enantiomerically enriched 1 (er 93:7). The undesired
enantiomer SRSR acid can be completely removed as a racemic mixture from
enantioenriched 1 by exploiting significant difference in solubilities of the racemate and
enantiopure 1. Thus, enantioenriched 1 (er 93:7) was triturated in a mixture of ethyl
acetate/THF to quantitatively remove the insoluble racemic mixture by filtration.16, 17
The filtrate so obtained was distilled and ethyl acetate/THF was exchanged to
18
ACS Paragon Plus Environment

Page 19 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
acetonitrile for crystallization of a sparingly soluble acetonitrile solvate of 1. After
isolation and drying, the material desolvates to provide enantiopure 1 (er > 99.9:0.1) in
7
9% yield. Absolute stereochemistry of 1 has been confirmed by single crystal X-ray
analysis (see Supporting Information for details).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although the enantiomeric excess in compounds 9a, 9b and 9c obtained via the Ag-
catalyzed process is similar to the Cu(I)-catalyzed process, significantly less racemate is
formed in the Cu(I)-catalyzed process due to the better diastereoselectivity in the [3+2]
cycloaddition reaction. This results in an improvement in the overall yield from 44% to
6
1%. Furthermore the aqueous waste streams generated in the Cu-process are
biodegradable, thus reducing the environmental impact.
Final recrystallization of idasanutlin. Several polymorphs, pseudo-polymorphs, hydrates
and an amorphous form of 1 are known, and the most relevant forms are summarized in
Table 1. The access to the different polymorphic forms is mainly determined by the
solvate present in the crystallization process. The use of acetonitrile in the
crystallization process of 1 (vide supra) results in the formation of a corresponding low
soluble acetonitrile solvate. The acetonitrile in the solvate can easily be removed by
drying, resulting in the solvate free Form III.
Table 1. Summary of the most important polymorphs and pseudoꢀpolymorphs of 1 and
thermal behavior.
Form
Thermal events
Solid Form
Form I
Endothermic solidꢀsolid conversion around 270 °C followed Polymorph
by melting as Form III
Form II
Form III
Form V
Around 280 °C (melting)
Around 282 °C (melting)
Between 25 °C and 140 °C loss of weight (dehydration ) and Hydrate
conversion to Form I
Polymorph
Polymorph
Form IX
Desolvation and conversion into Form III followed by Acetonitrile
melting of Form III
solvate
Amorphous
Glass transition around 146 °C
Amorphous
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The final recrystallization process is the following: a water-acetonitrile mixture (1:2.5
%
-w/w) is added to a polish-filtered THF solution of 1 (14 %-w/w). Crystallization of 1
starts after addition of approx. 1/3 of the aqueous mixture. The isolated acetonitrile
solvate (Form IX) of 1 is dried at 80 °C under vacuum to obtain 1 in 93% yield . The
acetonitrile in the solvate is easily removed by drying, resulting in the solvate free Form
III. As listed in Table 1, Form III is a slightly hygroscopic polymorph of 1 and is reliably
produced during clinical supply. The commercial manufacturing process for 1 is shown
in Scheme 6.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
20
ACS Paragon Plus Environment

Page 21 of 36
Organic Process Research & Development
1
2
3
4
5
Scheme 6: Commercial manufacturing process for idasanutlin (1).
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
Et N, MTBE
3
Cryst from n-heptane
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
94%
2
3
Step 1'
4
CuOAc, (R)-BINAP,
2-MeTHF, Et N
3
Exo selective
[3+2] cycloaddition
Step 2a
3
0% NaOMe/MeOH
9
a, exo, RSRR
74-76%, er = 99:1
and minor isomers
EtOH/H O, 25°C
2
91%
5
6
7
Step 1
THF, EtOH, aq. NaOH
Cryst from 2-PrOH/H O
Hydrolysis/
2
Isomerization
83% over 2 steps
Step 2b
THF/EtOAc
racemate filtered)
(
THF (polish filtration)
Cryst from CH CN/H O
Cryst from CH CN/H O
3
2
3
2
93%
79%
Recrystallization
Step 4
Racemate removal
Step 3
1
, RSRS
Enantiopure 1
er > 99.9:0.1
Idasanutlin (1), Form III
er = 93:7
2
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Genotoxic impurity control. Accompanying process development activities, starting
materials and their observed impurities, reagents, solvents, process intermediates,
observed impurities, degradation products and reasonably expected reaction by-
products involved in the synthesis of 1 were assessed for genotoxicity risk. The
assessment included in silico evaluation (using two complementary methods, DEREK
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0.0.2 for Windows and Leadscope Model, according to ICH M7 guidance: Assessment
and Control of DNA Reactive (Mutagenic) Impurities in Pharmaceuticals to Limit
Potential Carcinogenic Risk), Ames testing and analytical testing. Two compounds, 16
and its ethyl ester 17, produced structural alerts for mutagenicity during in silico
evaluation and were subsequently subjected to Ames testing. As outlined in Scheme 7,
there are at least two potential pathways for formation of 16 and 17. Compound 16 is
not only a by-product during the saponification reaction but also a degradation product
of 1 formed by hydrolysis of the amide bond, and ethyl ester 17 is a potential
degradation product of imine 4 and a by-product in the hydrolysis/isomerization
reaction.
Scheme 7: Pathways for formation of genotoxic impurities 16 and 17.
Hydrolysis/
isomerisation
Racemate
removal
Re-
crystallisation
Cycloaddition
4 + 7
9a-c, 10
1
1
1
(
enantioenriched)
(enantiopure)
(recrystallised)
hydrolysis
degradation
degradation
degradation
degradation
hydrolysis
1
7
16
The saponification of 17 under the reaction conditions was found to be significantly
slower as compared to the saponification of intermediate products 9a-c. Therefore, both
1
7 and 16 are typically detected in the reaction mixture after the hydrolysis in only up
22
ACS Paragon Plus Environment

Page 23 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
to 2% and 1%, respectively. These amounts are effectively purged during the
crystallization of enantioenriched 1 to below the reporting limit of the analytical
method.18 In the racemate removal step, the main purging effect was observed during
the crystallization of 1. The filter cake containing racemic 1 did not show significant
levels for both impurities. Furthermore, the final recrystallization step provides good
purging for both genotoxic compounds. Analysis of enantiopure 1 consistently proved
both the compounds to be below 10 ppm.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Despite the low levels routinely obtained during manufacturing, the purging power of
both crystallizations were investigated to correlate the experimental purge factors and
1
9
the theoretical purge factors calculated following the procedure by Teasdale et al. The
experimental purge factors were determined using data from spiking experiments. For
the crystallization of enantioenriched 1, both compounds were spiked to the mixture
after the hydrolysis just before the neutralization. Purging factors of > 150 for 16 and >
6
0 for 17 were observed. In the racemate removal step as well as the final
crystallisation, both compounds were spiked to a level of 1000 ppm to 1. As anticipated
the ethyl ester 17 is significantly better depleted than 16 in the racemate removal step
showing a purge factor of 50 compared to 15 for 16. The final recrystallization step
proved to have a very high purging power for 17 due to the presence of water in the
crystallization mixture. The overall purge factors (experimental) were determined to be
5
5
>
7 x 10 and > 2.4 x 10 for compounds 16 and 17 respectively. As shown in Table 2,
these values are significantly higher than the theoretical purge factors of 1000 and 3000
for compound 16 and 17 respectively. Therefore, in the case provided here, the
theoretical purge factor underestimates the real purging effect of the crystallization
process and provides a worst case estimate for the impurity control.
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2: Theoretical and measured purge factors for 16 and 17.
Theoretical
Purge Factor
Measured
Purge Factor
> 150
GTI Step No. Reactivity Solubility Volatility
2
a/2b
3
4
1
1
1
10
10
10
1
1
1
10
10
10
1000
30
10
> 15
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
> 330
> 7 x 10
5
Process
Process
2a/2b
3
4
3
1
1
10
10
10
1
1
1
> 60
> 50
> 80
1
7
10
3000
5
> 2.4 x 10
24
ACS Paragon Plus Environment

Page 25 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
CONCLUSION
A short and high-yielding manufacturing process for idasanutlin (1) was developed. The
salient features of the process include the convergent route using readily accessible
non-complex starting materials, the exceptionally selective and consistent
cycloaddition/isomerization sequence using a cheap chiral Cu(I) catalyst system, the
overall robustness and the high throughput in production, and the low environmental
impact. The process ensures control of impurity formation and purging (including
isomers and genotoxic impurities), and the crystallization of a desired polymorphic
form. This process has been utilized to produce more than 1500 kg of the idasanutlin
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) to date to support ongoing clinical studies.
EXPERIMENTAL DETAILS
General remarks. Reagents and solvents were used as received from commercial
suppliers. All reactions were carried out in standard pilot plant equipment consisting of
a setup of three connected 2500 L reactors with variable speed agitator, - 50 to 150 °C
jacket temperature range, and a 20 mbar vacuum to 1050 mbar pressure rating. Solids
were separated on a 100-cm basket centrifuge. All reactions were carried out under a
nitrogen atmosphere. Nitrogen was also routinely used to break vacuums for safety
reasons.
The four starting materials 2, 3, 5 and 6 and the intermediates 4, 7 and 9c were tested in
the central safety laboratory of F. Hoffmann-La Roche AG and the chemical
transformations were investigated via reaction calorimetric measurements. Common
calorimetric methods such as differential scanning calorimetry (DSC), accelerating rate
calorimetry (ARC), and reaction calorimetry were used to determine safety of reaction
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mixtures, handling of reactants, reagents and solvents. For each step reported in this
publication, the processes were deemed safe based on data from safety analysis.
All the GC and HPLC data were generated using calibrated equipment. Unless otherwise
specified all reported purity and assay values are measured using HPLC methods. MS
and NMR spectra were measured by the central analytical service of F. Hoffmann-La
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
13
Roche AG. The H NMR and C NMR spectra were measured on Bruker 600 MHz NMR
spectrometers at 600 MHz, 400 MHz and 150 MHz, respectively. The relative chemical
shifts are reported in ppm relative to TMS. The spectroscopical data of 1 and 7 are
1a
already published. For compounds 9a-c, 10, 15, 30a and 30b, the spectroscopical data
are listed in the Supporting Information.
Preparation of Imine (4)
To a suspension of 4-(2-amino-acetylamino)-3-methoxy-benzoic acid ethyl ester
hydrochloride (2) (110.0 kg; 381.0 mol) in MTBE (733 kg) at 20 – 25 °C 3,3-dimethyl
butyraldehyde (3) (42.0 kg; 419.3 mol) was added within 10 min, and the addition
vessel was rinsed with MTBE (74 kg). To the suspension was added triethylamine (42.8
kg; 423.0 mol) at 20 – 25 °C, and the addition vessel rinsed with MTBE (74 kg). The
resulting suspension was stirred for 6 h at 23 – 27 °C, when IPC indicated completion of
reaction. Triethylammonium hydrochloride was removed by centrifugation and the
cake rinsed with MTBE (370 kg). The combined filtrates were subjected to solvent
switch under reduced pressure at 40 – 50 °C to
constant. The solvent exchange was deemed to be completed when the residual level of
MTBE was below 4.0%. In total, 1467 kg of -heptane were used for the solvent
n-heptane while keeping the volume
n
exchange. The product partially precipiated during the solvent exchange. The mixture
was then heated to 60 – 65 °C to afford an almost clear solution, which was subjected to
26
ACS Paragon Plus Environment

Page 27 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
hot polish filtration. After rinsing the filter with hot n-heptane (290 kg), the combined
filtrates were cooled to 50 °C and seed crystals (550 g) were charged. A thin suspension
formed and the crystallization was completed by cooling the reaction mixture to −12 °C
to −8 °C at a rate of 18 °C/h. The crystals were isolated by centrifugation, washed with
n-heptane (220 kg) and dried under reduced pressure at 60 °C for 7 h, to furnish 119.6
kg of 4-{2-[3,3-dimethyl-but-(E)-ylideneamino]-acetylamino}-3-methoxy-benzoic acid
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
ethyl ester (4) (94% yield, 99.9% w/w, GC assay) as a white solid.
1
Data for 4: H NMR (600 MHz, CDCl
3
) ὁ ppm 9.45 (br s, 1 H), 8.52 (d,
= 8.4, 1.7 Hz, 1 H), 7.53 – 7.56 (m, 1 H), 4.36 (q,
= 7.2 Hz, 3 H), 1.03
) ὁ 14.41, 29.73 (3C), 31,31, 49.93, 55.95, 60.96,
J
=8.5 Hz, 1 H),
7
.82 - 7.89 (m, 1 H), 7.69 (dd,
J
J
= 7.2
Hz, 2 H), 4.22 (d,
J = 2 H), 3.93 (s, 3H) 2.26 – 2.29 (m, 2 H), 1.39 (t, J
(
s, 9 H). 13C NMR (150 MHz, CDCl
3
63.78, 110.74, 118.68, 123.27, 125.55, 131.63, 147.48, 168.17, 168.69. HRMS calcd. For
334.18871; found: 334.18690.
C
18
H
26
N
O
2 4
Preparation of Stilbene (7)
To a solution of 3-chloro-2-fluoro-benzaldehyde (5) (100 kg; 631 mol) and 4-chloro-2-
fluoro-phenyl acetonitrile (6) (107 kg; 631 mol) in ethanol (1722 kg) was added water
(11 kg) to adjust the water content to 0.5 %-w/w. At 25 °C, a solution of sodium
methoxide (30 %-w/w in methanol, 6.5 kg; 36.1 mol) and ethanol (150 kg) were added
over 90 minutes. The product begins to precipite during the addition. The lines were
rinsed with ethanol (40 kg) and the suspension was stirred at 25 °C for 4 h. When the
IPC indicated completion of the reaction, the crystals were isolated by centrifugation
and washed with a mixture of ethanol (332 kg) and water (47 kg) followed by ethanol
(368 kg). The product was dried under reduced pressure at 60 °C for 10 h, to afford 178
27
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
kg of (Z)-3-(3-chloro-2-fluoro-phenyl)-2-(4-chloro-2-fluoro-phenyl)-acrylonitrile (7)
(91% yield, 99.9% GC purity and 99.1 %-w/w) as a white solid.
Preparation of idasanutlin (1) via AgOAc catalyzed cycloaddition:
Preparation of ester (9c).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To a stirred suspension of 4-2-[3,3-dimethyl-but-(E)-ylideneamino]-acetylamino-3-
methoxy-benzoic acid ethyl ester (4) (58.1 kg; 173.7 mol) and (Z)-3-(3-chloro-2-fluoro-
phenyl)-2-(4-chloro-2-fluoro-phenyl)-acrylonitrile (7) (49.0 kg; 158.0 mol) in 2-MeTHF
(220 kg) at 4 °C, a previously prepared thin suspension of (
R)-MeOBIPHEP 12 (1.01 kg;
1
.73 mol) and AgOAc (0.26 kg; 1.56 mol) in 2-MeTHF (8.6 kg ) was added and the
feeding tank rinsed with 2-MeTHF (4 kg) into the vessel. Afterwards the reaction
mixture was stirred at 2 – 4 °C for 16 h. After successful IPC, the reaction mixture was
warmed to 23 °C and diluted by addition of 2-MeTHF (36 kg). Then a suspension of
previously wet milled LiOH 20 in 2-MeTHF was added and the reaction mixture was
heated to 65 °C and stirred at this temperature for 19 h. After complete conversion, n-
heptane (460 kg) was added within 50 minutes and afterwards the suspension was
cooled to 15 °C at a rate of 0.15 °C/min and stirred at target temperature for 4 h. The
product was collected by centrifugation and washed with a mixture of
kg) and 2-MeTHF (98 kg). It was dried to constant weight (55 °C, 7 mbar, 5 h) to furnish
103.6 kg of enantiomerically enriched 4-{[(2 ,3 ,4 ,5 )-4-(4-chloro-2-fluoro-phenyl)-
-(3-chloro-2-fluoro-phenyl)-4-cyano-5-(2,2-dimethyl-propyl)-pyrrolidine-2-
carbonyl]-amino}-3-methoxy-benzoic acid ethyl ester (9c) (94.9 %-w/w, er: 78.5:21.5,
7% yield) as an off-white solid.
Peparation of Idasanutlin (1) from Ester (9c).
n-heptane (156
R
S R S
3
9
28
ACS Paragon Plus Environment

Page 29 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The reactor was charged with 4-{[(2R,3S,4R,5S)-4-(4-chloro-2-fluoro-phenyl)-3-(3-
chloro-2-fluoro-phenyl)-4-cyano-5-(2,2-dimethyl-propyl)-pyrrolidine-2-carbonyl]-
amino}-3-methoxy-benzoic acid ethyl ester (9c) (128.0 kg; 198.6 mol) followed by
isopropanol (447 kg) at 20 – 25 °C and to the resulting suspension a previously
prepared solution of lithium hydroxide (0.97 kg; 40.5 mol) in water (109 kg) was added
followed by addition of water (19 kg) for rinsing of the feed vessel. The thin suspension
was heated to 63 – 67 °C and stirred at this temperature for 6 h. After completion of the
reaction, the reaction mixture was transferred to a crystallization vessel and the reactor
and lines were rinsed with isopropanol (45 kg). The thin suspension was cooled within
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
h to 15 – 20 °C and stirred for an additional 8 h at this temperature. The solids were
collected by centrifugation and the filter cake was rinsed with isopropanol (507 kg).
The combined filtrates were afterwards polish filtered and warmed to 63 – 67 °C. At this
temperature, glacial acetic acid (26.4 kg; 439.6 mol) was added followed by water (8.2
kg). After adjusting the pH to < 7, water (1075 kg) was added within 100 minutes and
then the suspension cooled to 8 – 12 °C within 4 h and stirred at this temperature for
additional 4 h. After collecting the solids by centrifugation, the wet cake was rinsed with
a mixture of isopropanol (294 kg) and water (372 kg) and dried to constant weight (55
°C, 10 mbar, 5 h) to furnish 55.4 kg 4-[(2R,3S,4R,5S)-4-(4-chloro-2-fluoro-phenyl)-3-(3-
chloro-2-fluoro-phenyl)-4-cyano-5-(2,2-dimethyl-propyl)-pyrrolidine-2-carbonyl]-
amino-3-methoxy-benzoic acid (1) (99.8 %-w/w, 45% yield) as a white solid with an
enantiomeric ratio of 99.7:0.3 and < 100 ppm of Ag.
Preparation of idasanutlin (1) via CuOAc catalyzed cycloaddition:
Preparation of enantiomerically enriched acid 1
29
ACS Paragon Plus Environment