Development of an Improved Route for the Synthesis of an Abemaciclib Intermediate

Article abstract of DOI:10.1021/acs.oprd.9b00347

A new synthesis for an intermediate of abemaciclib is described. Keys to this route are the use of inexpensive starting materials, biphasic amine alkylation for mild C-N bond formation, anhydrous coupling of a 2-chloropyridine derivative with LiHMDS to avoid a hydroxy impurity, and neutral, fluoride-free conditions to affect desilylation. Scale-up of the optimized conditions are described on a kilogram scale.

Full text of DOI:10.1021/acs.oprd.9b00347

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Communication
Development of an Improved Route for the
Synthesis of an Abemaciclib Intermediate
Michael P Carroll, Harold Moloney, Olivia Gowran, Aoibheann O'Connor, Eoin M Wilson,
Michael M Murray, Mark A. Pietz, Douglas P. Kjell, Charles Brad Held, and Michael O Frederick
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00347 • Publication Date (Web): 23 Oct 2019
Downloaded from pubs.acs.org on October 23, 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 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Development of an Improved Route for the
Synthesis of an Abemaciclib Intermediate
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
Michael P. Carroll1, Harold Moloney1, Olivia Gowran2, Aoibheann O’Connor1, Eoin M.
Wilson1, Michael M. Murray*1, Mark A. Pietz3, Douglas P. Kjell3, C. Brad Held3, Michael
O. Frederick*3.
1 Technical Services/Manufacturing Science, Eli Lilly Kinsale Limited, Dunderrow,
Kinsale, P17 NY71 Co. Cork, Ireland.
2 Quality Control Laboratories, Eli Lilly Kinsale Limited, Dunderrow, Kinsale, P17 NY71
Co. Cork, Ireland.
3Small Molecule Design and Development, Eli Lilly and Company, Lilly Corporate
Center, Indianapolis, Indiana 46285, United States
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 27
1
2
3
4
5
TOC Figure.
6
7
8
9
1. aq. K2CO3 (1.2 equiv.), THF
2. PdCl2[PhCN]2 (0.5 mol%),
RuPhos (1 mol%),
LiHMDS (1.2 equiv.), Reflux
3. MeOH, reflux
Cl
N
Cl
H2N
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
+
N
N
4. Toluene/MeCN crystallization
HN
ACS Paragon Plus Environment
2

Page 3 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT: A new synthesis for an intermediate of abemaciclib is described. Keys to
this route are: use of inexpensive starting materials; biphasic amine alkylation for mild C–
N bond formation; anhydrous coupling of a 2-chloropyridine derivative with LiHMDS to
avoid a hydroxy impurity; and neutral, fluoride-free conditions to affect desilylation. Scale-
up of the optimized conditions are described on kilogram scale.
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
KEYWORDS: Abemaciclib, CDK 4/6 Inhibitor, Pd-catalyzed C-N coupling, TMS removal
Abemaciclib (brand name Verzenio) is a small molecule CDK4/6 inhibitor approved for
the treatment of hormone-receptor-positive, HER2-negative metastatic or advanced-
stage breast cancer as a monotherapy, in combination with fulvestrant,1 or a non-steroidal
aromatase inhibitor.2
A recent publication has described a route towards the synthesis of abemaciclib (1) by
first breaking the molecule into two roughly equal pieces (2 and 3) which are joined
through a palladium-catalyzed amination (Figure 1).3 Fragment 3 was derived from ethyl
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 27
1
2
3
4
5
piperazine (4) and bromoaldehyde 5 via a Leuckart–Wallach reductive amination4 in flow
6
7
8
followed by a copper-catalyzed Ullmann-type amination.
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
F
Br
N
N
N
F
F
N
N
H
N
F
HN
O
N
N
5
OR
4
2
N
N
Cl
Cl
N
HN
N
H2N
N
N
N
Cl
N
N
3
6
Abemaciclib (1)
Figure 1. Retrosyntheses of abemaciclib (1).
The process is robust and affords high quality 3. The route, however, employs a
relatively expensive starting material 5, requires high-pressure amination conditions, high
temperature/vacuum distillations and is relatively low yielding due to the need for two
crystallizations. Aqueous ammonia is employed for safety reasons and significant yield is
lost because of the formation of alcohol impurity 8 (up to 15%, Scheme 1).3 The high
solubility of 3 in water also complicates product isolation; high temperature/high vacuum
distillations and a hot salt filtration are employed to furnish the desired product.
ACS Paragon Plus Environment
4

Page 5 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
N
aq. NH3 (20 equiv.), CuO (5%)
2-(methylamino)ethanol (10%)
aq. K2CO3 (20%), EG, 100 °C
N
H2N
HO
N
3
N
N
N
50-60% isolated yield
Br
N
N
7
N
8 (up to 15%)
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 1. Previously reported synthesis of 3 with up to 15% of hydroxy-impurity 8
formed.
The route to 3 described herein joins ethyl piperazine (4) with 2-chloro-5-
chloromethylpyridine (CCMP, 6)5 through an SN2 coupling followed by amination on the
now less activated chloropyridine (9, Scheme 2). Proposed starting material CCMP is
readily available in large quantities and at low cost as it is used in the synthesis of large
volume agricultural chemicals such as imidacloprid.5 As the palladium-mediated
Buchwald–Hartwig amination uses anhydrous conditions,6 the formation of impurity 8
can be reduced or eliminated.
N
cat. Cu or Pd,
NH2 source
HN
SN2
N
N
4
N
N
Cl
N
H2N
N
Cl
9
3
Cl
N
6
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 27
1
2
3
4
5
6
7
8
Scheme 2. Proposed alternative synthesis of 3 starting from CCMP (6) and ethyl
piperazine (4).
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 SN2-coupling between 4 and 6 proceeded well under a variety of conditions. To
settle on optimized conditions, we focused on three areas: 1. high conversion of 6 as it
has a negative impact on color and purity in the subsequent steps; 2. Use of a base that
would be soluble without insoluble by-product salts; and 3. Use of a solvent that would
allow telescoping into the subsequent amination. Table 1 summarizes these efforts and
optimal conditions were found to be a biphasic reaction with THF/H2O (5:3) with 1.20
equiv. of both 4 and K2CO3 at reflux for 6 hours.
Table 1. SN2 reaction between CCMP (6) and ethyl piperazine (4).
Cl
N
Conditionsa
Cl
N
N
N
Cl
HN
N
4
6
9
Entry
Equiv. of 4
1.15
Solvent (ratio)
Toluene
Toluene/H2O (2:1)
Toluene/THF (2:1)
Toluene/THF/H2O (4:1:2)
THF
Temp.
70 °C
70 °C
70 °C
75 °C
Reflux
Reflux
45 °C
Reflux
Time (h) Conversion %a
1
2
3
4
5
6
7
8
6
13
7
74.4
98.8
99.4
90.6
92.3
91.2
95.2
99.6
1.15
1.15
1.15
1.15
1.15
1.15
1.15
8.5
6
THF/H2O (2:1)
THF/H2O (4:1)
THF/H2O (4:1)
7
7
4
ACS Paragon Plus Environment
6

Page 7 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Entry
9
10
11
12
Equiv. of 4
1.15
Solvent (ratio)
THF/H2O (8:1)
THF/H2O (5:1)
THF/H2O (5:3)
THF/H2O (5:3)
THF/H2O (5:3)
Temp.
Time (h) Conversion %a
Reflux
Reflux
Reflux
Reflux
Reflux
3
3
6
6
6
99.5
99.6
91.8
98.5
>99.9
1.15
1.00
1.10
1.20
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
a1.2 equiv. K2CO3 used for all reactions. Numbers listed indicate conversion which were
assessed by HPLC analysis of the organic layer. Reactions conducted on 1 mmol scale.
With conditions to arrive at intermediate 9, attention turned to the amination reaction.
Initial exploration with copper-based systems under an ammonia atmosphere led to very
poor conversion, leading us to explore palladium-catalyzed variants. Screening of various
ligands and additives led us to BrettPhos and t-BuBrettPhos-based ligands in 2-MeTHF.
The results described in Table 2, were promising, but lacking in utility. In addition to
incomplete conversion, bis-adduct 10 was consistently formed. At this stage, screening
with ammonia was halted, as overcoming this side-reaction was deemed difficult.
Table 2. Conditions for the amination of intermediate 9. Reactions conducted on 1 mmol
scale with 4 mol% Pd, 8 mol% ligand.
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 27
1
2
3
4
5
6
7
8
H2N
N
N
N
Pd2dba3, ligand, NH3 (2 bar),
NaOtBu, 2-MeTHF, 80 °C, 18 h
Cl
N
3
N
N
H
N
N
N
N
N
9
N
N
9
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
Entry
Ligand
BrettPhos
Additive
none
none
LiBr
NaBr
KBr
%3
17
55
64
54
74
59
%9 %10a
1
2
3
4
5
6
67
29
23
24
14
24
16
16
13
22
12
17
t-BuBrettPhos
t-BuBrettPhos
t-BuBrettPhos
t-BuBrettPhos
t-BuBrettPhos
ZnBr2
aReactions conducted on 1 mmol scale with 4 mol% Pd2dba3, 8 mol% ligand. Area %
listed are based on HPLC analysis.
To avoid the bis-amination by-product, the use of protected amines was explored.
A 96-reaction screen was set up exploring 8 different protected amines, 4 catalyst
systems, 4 different bases/additives and three different solvents heated to 80 °C for 16
hours (Table 3). Analysis showed multiple hits, but LiHMDS was consistently the best
amine, working well with most every set of conditions explored. Not only was complete
conversion obtained, but it was assumed subsequent TMS cleavage would be easier than
the alternatives.
ACS Paragon Plus Environment
8

Page 9 of 27
Organic Process Research & Development
1
2
3
4
5
Table 3. Results of a 96-well plate screening amines, additives, ligands, catalysts and
6
7
8
solvents.
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
Conditionsa
Cl
N
R2N
N
N
N
N
N
9
Amine
Base/
XPhos pre-cat
Pd2dba3/Xantphos
(tBu3P)2Pd(0)
PhMe Dioxane
Pd2dba3/CyJohnPhos
Additive
Cs2CO3
Cs2CO3
Cs2CO3
NaOtBu
MeTHF
PhMe
64
Dioxane
MeTHF
PhMe
24
22
0
72
Dioxane
MeTHF
30
0
0
70
MeTHF
PhMe
Dioxane
H2NBoc
Benzamide
diMe-Pyrrole
Allyl amine
Benzophenone
imine
Benzyl amine
LiHMDS
Zn(HMDS)2
67
35
0
34
21
0
20
14
0
38
27
0
42
2
32
0
20
7
9
9
15
8
32
0
14
0
0
0
0
0
0
15
74
88
65
51
89
85
93
NaOtBu
33
10
33
58
23
85
0
1
0
54
28
63
NaOtBu
None
LiCl
56
100
0
27
100
0
28
100
0
98
100
0
88
100
0
100
100
0
89
99
57
84
94
45
83
91
38
98
100
0
93
100
0
100
100
0
aNumbers listed indicate conversion. Reactions conducted at 80 °C for 16 hours with 2
mol% Pd and 4 mol% ligand. Area % listed are based on HPLC analysis.
A subsequent round of optimization was then conducted using LiHMDS as the protected
amine and exploring different palladium catalysts and phosphine ligands (Table 4). The
initial aim of this round of optimization was to retain the excellent conversions observed
in earlier screening while also reducing the palladium loading. Our screening efforts
focused on the Buchwald class of ligands due to literature precedent6 and our earlier
success with these ligands in this transformation (Table 2). Attempting to halve the initial
palladium loading using CyJohnPhos resulted in poor conversion but by employing
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 27
1
2
3
4
5
6
7
8
9
XPhos, up to a 4-fold reduction in catalyst was possible (Entries 1–3). When trying to
rationalize the difference in reactivity between these two ligands it was observed that
XPhos contained substituents in the ortho positions of the lower aryl ring of the ligand
which CyJohnPhos lacked. This led to screening other ligands containing similar
substitution patterns. Applying SPhos and RuPhos in the transformation also resulted in
excellent conversions with RuPhos matching the performance of XPhos (Entries 4 & 5).
Cy3P, was also screened and found to provide excellent conversions (Entry 6). Further
lowering the palladium loading to 0.5 mol %, revealed that RuPhos was the only ligand
capable of giving full conversion at this loading (Entries 7–9). For manufacturing purposes
it was desirable, if possible, to use a Pd(II) source as palladium (0) sources such as
Pd2dba3 have been shown to degrade over time.7 Pd(OAc)2 was screened and although
it showed excellent conversions at higher loadings, reducing the palladium loading to 0.5
mol % resulted in poor conversions (entries 10 – 13). Further it was noted that running
the reaction in toluene resulted in cleaner impurity profile than 2-MeTHF. Toluene also
allows for more efficient water removal via distillation in the previous steps solvent swap,
ensuring the LiHMDS does not get quenched prior to reacting (entry 14). Further
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
10

Page 11 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
palladium (II) salts were then screened with RuPhos (Entries 15–17) and pleasingly, it
was found that PdCl2(PhCN)2 was capable of providing complete conversion at 0.5 mol
% loading.8 The Pd-allyl RuPhos complex, Pd-1719 was also screened (entry 18) and
allowed a further reduction in catalyst loading to 0.25 mol %, however the
RuPhos/PdCl2(PhCN)2 was chosen going forward due to its wide availability from multiple
sources. Finally prompted by a recent report,10 the nickel catalyst NiCl2(PPh3)2 was
screened under the reaction conditions but was found be ineffective (entry 19).
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 amination with LiHMDS.
TMS
N
Pd source, ligand, solvent
Cl
N
N
LiHMDS, reflux, 2h
N
TMS
N
N
N
9
11
Cl
Pd
P
R,R' = H, CyJohnPhos
R,R' = iPr, XPhos
R = OMe, R' = H, SPhos
R = OiPr, R' = H, RuPhos
Cy
PCy2
R
OiPr
R
Cy
OiPr
R'
Pd-171
Entry
Pd source
Ligand
Solvent
% conversiona
1
2
3
4
5
6
7
8
Pd2dba3 1 mol%
Pd2dba3 1 mol%
CyJohnPhos 4 mol%
XPhos 4 mol%
XPhos 2 mol%
SPhos 2 mol%
RuPhos 2 mol%
Cy3P 2 mol%
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
33
96
94
87
94
99
100
95
Pd2dba3 0.5 mol%
Pd2dba3 0.5 mol%
Pd2dba3 0.5 mol%
Pd2dba3 0.5 mol%
Pd2dba3 0.25 mol%
Pd2dba3 0.25 mol%
RuPhos 1 mol%
XPhos 1 mol%
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 27
1
2
3
4
5
6
7
8
9
Entry
9
Pd source
Ligand
Cy3P 1 mol%
RuPhos 3 mol%
XPhos 3 mol%
RuPhos 1.5 mol%
XPhos 1.5 mol%
RuPhos 1 mol%
RuPhos 1 mol%
RuPhos 1 mol%
RuPhos 1 mol%
-
Solvent
% conversiona
Pd2dba3 0.25 mol%
Pd(OAc)2 1 mol%
Pd(OAc)2 1 mol%
Pd(OAc)2 0.5 mol%
Pd(OAc)2 0.5 mol%
Pd2dba3 0.25 mol%
PdCl2 0.5 mol%
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
Toluene
Toluene
Toluene
Toluene
Toluene
31
92
99
65
16
100
1
68
100
100
2
10
11
12
13
14
15
16
17
18
19b
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
PdCl2(MeCN)2 0.5 mol%
PdCl2(PhCN)2 0.5 mol%
Pd-171 0.25 mol%
NiCl2(PPh3)2 1 mol%
Toluene
-
aDetermined by HPLC. b Reaction was run for 3.5 hours.
Quenching the amination reaction with water led to emulsions and slow hydrolysis of
11 to 3, which was subsequently lost to the aqueous layer. Numerous additives were
evaluated to improve layer separation (Table 5).
Table 5. Qualitative assessment of different aqueous solutions impacts on layer
separations.
Entry
1
Aqueous solutiona
Water
Rag Layer/Emulsion
Yes
Yes
Yes
Yes
Yes
No
2
3
4
5
6
7
8
9
Benzalkonium chloride
Benzyltributylammonium chloride
Hexadecylpyridinium chloride
Sodium dodecyl sulfate
Potassium Pyrophosphate
Ammonium phosphate
Potassium sodium tartrate
Potassium chloride
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
10
11
12
13
14
15
Sodium chloride
Potassium carbonate
Ammonium acetate
Ammonium chloride
Ammonium formate
Potassium phosphate dibasic
ACS Paragon Plus Environment
12

Page 13 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Entry
16
17
18
19
Aqueous solutiona
Ammonium sulfate
Lithium sulfate
Sodium sulfate
Sodium sulfateb
Rag Layer/Emulsion
Yes
Yes
No
No
a*Solutions were made up to 80% saturation of the respective salt or surfactant. b40% saturated
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
solution used.
Of the salts and surfactants screened, three (potassium pyrophosphate, ammonium
phosphate and sodium sulfate) were able to provide an improvement in separation over
using water alone. Of these 3, sodium sulfate was chosen for further development due
to its cost, stability and availability.12 For operational reasons it was desirable to use a
more dilute solution of sulfate to prevent any salt precipitation at lower temperatures
and employing a more dilute solution (40% saturated) of sodium sulfate maintained a
clean phase separation.
Having developed satisfactory conditions to obtain the protected amine 11 we next
examined a method for its deprotection. Commonly, the removal of silyl groups from an
amine has been accomplished through the use of aqueous acid.6 In our case we wished
to avoid using aqueous solutions owing to the high solubility of 3 in water.11
Furthermore, due to the presence of several basic nitrogen atoms in the molecule, large
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 27
1
2
3
4
5
6
7
8
9
amounts of acid would likely generate a salt of 3 which would require an additional
freebasing, followed by back extraction into an organic solvent from which to crystallize.
In past cases where acid sensitive functional groups were present tetrabutylammonium
fluoride (TBAF) has also been used.6a Large scale use of TBAF or other fluoride
sources is undesirable due to the amount of fluoride waste which would be produced.
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
These considerations prompted us to seek a method of deprotection which could be
run in an anhydrous homogeneous organic system. It has previously been
demonstrated that silyl groups can also be cleaved from amines through the use of an
acid catalyst in the presence of an alcohol13 or by simply stirring in an alcohol solvent.14
Wishing to avoid long reaction times we first examined the combination of an acid
catalyst and alcohol combination in the desilylation reaction of 11. MeOH was chosen
as the alcohol because the side product generated, methoxytrimethylsilane 11, is
volatile (b.p. 57 °C) and could be removed by distillation.
ACS Paragon Plus Environment
14

Page 15 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
During our initial screen monitoring by 1H NMR revealed that the deprotection
proceeds in a stepwise manner first producing a partially deprotected mono-silylated
intermediate 11 which then fully deprotects 12 to give 3 (Scheme 3).
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
TMS
N
TMS
HN
N
N
H2N
N
TMS
N
N
N
+
Si
N
N
N
O
11
3
12
Scheme 3. Stepwise deprotection of intermediate 11 giving amine 3 via mono-silylated
intermediate 12.
Both heterogenous and homogenous acids and bases were examined. While bases
(Table 6, Entries 1–2) were shown to be capable of affecting the deprotection, overall,
they were inferior to the acids. A number of acids were screened with the most effective,
acetic and trifluoroacetic acid (Entries 7 & 8), able to provide complete conversion in 1
hour at room temperature. Although reported previously for this transformation,13 p-
1
toluene sulfonic acid (Entry 6) and silica gel (Entry 4) were less effective. H NMR
analysis of a sample of 3 isolated after using 5 mol% acetic acid and MeOH to deprotect
revealed the presence of approximately 5 mol% of an acetate salt of 3.
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 27
1
2
3
4
5
Table 6. Initial screening of conditions in the desilylation reaction.
6
7
8
9
TMS
N
N
H2N
N
Toluene
TMS
N
N
N
N
22 °C, 1 h
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
11
3
Entry
Conditions
% conversion
after 1 h at room
temperatutea
1
2
3
4
5
6
7
8
Potassium Carbonate (3 equiv.)
TMG (5 mol%) MeOH (3 equiv.)
Bentonite clay (5 mol%) MeOH (3 equiv.)
Silica Gel (3.5 wt%)
PPTS (5 mol%) MeOH (3 equiv.)
PTSA (5 mol%) MeOH (3 equiv.)
Acetic Acid (5 mol%) MeOH (3 equiv.)
24
37
30
25
54
72
100
100
Trifluoroacetic Acid (5 mol%) MeOH (3 equiv.)
aDetermined by 1H NMR.
In an effort to circumvent salt formation phenols were screened in the deprotection
reaction. To improve the rate of reaction the amount of methanol used was increased
from 3 equivalents to 5 volumes and the reactions were run at 55 °C. A baseline was
first established by running the deprotection reaction in 5 volumes of methanol at 55 °C
without any catalyst present. Pleasingly, in the absence of any other catalyst this
reaction went to 72% conversion after 2 hours and full conversion after 4 hours (See
Figure 2 for kinetics of some systems).
ACS Paragon Plus Environment
16

Page 17 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Several phenols were screened with these conditions (Table 7). A trend became
apparent whereby the more electron deficient (and thus more acidic) the phenol
resulted in a faster rate of deprotection (Entries 1–5). Interestingly in the presence of the
electron rich and sterically hindered phenol, butylated hydroxytoluene (BHT, Entry 6),
the reaction rate slowed down and was only 46% complete after 2 hours. Both AcOH
and TFA (Entries 8–10) were further screened with reduced loadings under these
conditions and were still found to give full conversion after 2 hours. Finally 2-tert-butyl-
1,1,3,3-tetramethylguanidine (TMG, Entries 11–12) was screened under these
conditions at two loadings (0.5 and 5 mol %) and could not give full conversion after 2
hours even at the higher loading of 5 mol% further suggesting that the base catalyzed
deprotection is slower.
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 7. Optimization of conditions in the desilylation reaction.
TMS
Conditions
55 °C, 2 h
N
N
H2N
N
TMS
N
N
N
N
11
3
Entry
Conditions
% conversion
after 2 hoursa
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 27
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
Phenol (0.5 mol %) MeOH (5 vols.)
4-(Trifluoromethyl)phenol (0.5 mol %) MeOH (5 vols.)
4-Nitrophenol (0.5 mol %) MeOH (5 vols.)
2-Nitrophenol (0.5 mol %) MeOH (5 vols.)
2-Fluoro-4-nitrophenol (0.5 mol %) MeOH (5 vols.)
BHT (0.5 mol %) MeOH (5 vols.)
MeOH (5 vols.)
80
97
100
100
100
46
100b
100
100
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
AcOH (0.5 mol %) MeOH (5 vols.)
TFA (0.5 mol %) MeOH (5 vols.)
AcOH (0.5 mol %)
TMG (0.5 mol %) MeOH (5 vols.)
TMG (5 mol %) MeOH (5 vols.)
10
11
12
82
91
aConversion determined by GC. bReaction was run for 5 hours.
100
90
80
70
60
50
40
30
20
10
0
4-CF3Phenol (0.5 mol%) & MeOH (5 vol)
Phenol (0.5 mol%) & MeOH (5 vol)
4-NitroPhenol (0.5 mol%) & MeOH (5 vol)
4-nitro-2-fluorophenol (0.5 mol%) MeOH
(5 vol)
2-nitrophenol (0.5 mol%) & MeOH (5 vol)
MeOH (5 Vol) Baseline
BHT (0.5 mol%) MeOH (5 vol)
0
20
40
60
80
100
120
Time (min)
Figure 2. Reaction conversion vs time in the deprotection with various phenols.
When using TMG as a catalyst instead of acid, a change in the amount of mono-
silylated intermediate which accumulated during the reaction was observed. When the
deprotection was carried out using 5 mol% of TMG the amount of the mono-silylated
ACS Paragon Plus Environment
18

Page 19 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
intermediate observed never rose above 5 area % by GC (Figure 3). When an acid
catalyst was used, such as 4-trifluoromethylphenol (Figure 4), the mono-silylated
intermediate 12 would rise as high as 48 area % relative to 3 and 11. This would
suggest that there might be different mechanisms at work in the deprotection reaction
depending on whether an acid or base is used as the catalyst; this is in agreement with
previous reports of the mechanism of desilylation of silyl amines.14 Desilylating with
MeOH led to a similar buildup of 12 during the reaction which could imply that the
mechanism of desilylation via methanolysis is similar to that of acid catalyzed
desilylation.
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
TMG 5 mol%
100
80
60
40
20
3
12
11
0
0
20
40
60
80
100
120
Time (mins)
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 27
1
2
3
4
5
Figure 3. Product, bis-silyl and mono-silyl % during the base catalyzed deprotection.
6
7
8
4-Trifluoromethylphenol
9
100
80
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
60
40
20
3
12
11
0
0
20
40
60
80
100
120
Time (min)
Figure 4. Product 3, bis-silyl 11 and mono-silyl 12 % during the acid catalyzed
deprotection.
ACS Paragon Plus Environment
20

Page 21 of 27
Organic Process Research & Development
1
2
3
4
5
6
MeOH
100
90
80
70
60
50
40
30
20
10
0
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
3
12
11
0
20
40
60
80
100
120
140
160
180
Time (min)
Figure 5. Product 3, bis-silyl 11 and mono-silyl 12 % during the desilylation using
MeOH.
Having screened several successful conditions for the desilylation of 11 it was
decided to proceed with desilylation in 5 volumes of MeOH alone. Although slower than
the acid catalyzed process (1 h vs 4 h) this route avoided the formation of any product
salts and used one less raw material for the process. Intrigued by the simplicity of the
methanolysis approach we screened other silyl amines to see if they could be as easily
and rapidly desilylated under similar conditions (Table 8). Refluxing several silylated
amines in a solvent mixture similar to our process stream resulted in complete
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 27
1
2
3
4
5
6
7
8
9
desilylation within 3 hours for all substrates. From this screen and our earlier work, it
can be shown that methanolysis is a viable method to desilylate silylamines under
anhydrous conditions and in short reaction times.
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 8. Comparison of the desilylation rates for different substrates with refluxing
MeOH (5 volumes).
Toluene (5 Vol), THF (3.5 Vol)
Si
N
MeOH (5 Vol), reflux
NH2
+
Si
R
Si
O
R
Entry
R
Time to 100% conversion
1
2
3
4
Hexyl
Benzyl
Phenyl
11
1 h
1.5 h
2.5 h
4 h
The deprotected product 3 could now be obtained in a process stream containing
large quantities of MeOH, THF, Toluene and methoxytrimethylsilane (MTMS). It was
found that THF and MeOH needed to be removed from the stream to minimize yield
loss as 3 is highly soluble in these solvents.11 In order to ensure that the MeOH would
ACS Paragon Plus Environment
22

Page 23 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
always be sufficiently removed a 5 volume toluene add back followed by a second
distillation was added to the process.
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
Crystallization of 3 from the toluene solution was achieved initially through addition of
antisolvent. Use of heptane as antisolvent provided poor impurity rejection. While use of
MTBE provided acceptable impurity rejection, optimization to prevent occasional oiling
was unsuccessful. Use of MeCN as co-solvent provided a system with acceptable
impurity rejection and yield which also was not prone to oiling.
The final crystallization solvent system chosen was a 2:1 MeCN/toluene as a cooling
crystallization and possessed excellent rejection capabilities. At the saturation point, the
metastable zone width is 12 °C, and given the potential batch-batch variability at scale,
considerable levels of supersaturation could exist within the process stream and lead to
a lack of control in the crystallization. The chosen seed point limits the maximum
supersaturation of the solution prior to crystal growth and mitigates against the risk of
oiling. During development and 30 L scale runs, crystals have been observed at the
seed point and thus have not required seeding. No impact has been observed to the
quality of 3 obtained in unseeded versus seeded crystallizations.
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 27
1
2
3
4
5
6
7
8
9
In conclusion, a new set of conditions were developed for the synthesis of abemaciclib
intermediate 3. The new process for the production of 3 has been performed on
kilogram scale as a telescoped three step process involving two bond forming steps,
one deprotection step, three extractions, two distillations and product isolation by
crystallization (Scheme 4). The improvements addressed shortcomings of the previous
route (eliminating formation of hydroxy impurity 8 and the use of a simpler, higher
yielding process) and make for a good alternative to existing chemistry.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1. aq. K2CO3 (1.2 equiv.), THF
2. PdCl2[PhCN]2 (0.5 mol%),
RuPhos (1 mol%),
LiHMDS (1.2 equiv.), Reflux
3. MeOH, reflux
Cl
N
Cl
H2N
N
6
+
N
N
N
4. Toluene/MeCN crystallization
3
HN
75% yield
4
Scheme 3. Summary of the optimized process for the synthesis of 3.
Supporting Information
An optimized procedure for the synthesis of 3 is provided in the Supporting
Information file. In addition to procedures, characterization of new compound (11) is
also included.
ACS Paragon Plus Environment
24

Page 25 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
AUTHOR INFORMATION
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
Corresponding Authors
*Michael M. Murray: murray_michael@lilly.com
*Michael O. Frederick: frederickmo@lilly.com
Author Contributions
The manuscript was written through contributions of all authors. All authors have given
approval to the final version of the manuscript.
REFERENCES
1) Goetz, M. P.; et al. MONARCH 3: abemaciclib as initial therapy for advanced breast
cancer. J. Clin. Oncol 2017, 35 (32), 3638-3646.
2) Johnston, S.; et al. MONARCH 3 final PFS: a randomized study of abemaciclib as
initial therapy for advanced breast cancer. NPJ Breast Cancer 2019, 5 (5), 1-8.
3) Frederick, M. O.; Pietz, M. A.; Kjell, D. P.; Richey, R. N.; Tharp, G. A.; Touge, T.;
Yokoyama, N.; Kida, M.; Matsuo, T. Development of a Leuckart-Wallach reaction in flow
for the synthesis of abemaciclib. Org. Process Res. Dev. 2017, 21, 1447.
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 27
1
2
3
4
5
6
7
8
9
4) a) Frederick, M. O.; Kjell, D. P. A synthesis of abemaciclib utilizing a Leuckart-Wallach
reaction. Tetrahedron Lett. 2015, 56, 949; b) Leuckart, R. Tribenzylamine. Ber. Dtsch.
Chem. Ges 1886, 18, 2341; c) Leuckart, R.; Bach, E. Action of ammonium formate on
benzaldehyde and benzophenone. Ber. Dtsch. Chem. Ges. 1886, 19, 2128; d) Wallach,
O. Menthylamine. Ber. Dtsch. Chem. Ges. 1891, 24, 3992.
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) a) Zhang, T. Y., Scriven, E. F. V. Cyclocondensation process for the preparation of 2-
halo-5-(halomethyl)pyridines. U.S. Patent, 5229519 A 19930720, 20 Jul 1993; b) Shao,
X.; Liu, Z.; Xu, X.; Li, Z., Qian, X. Overall status of neonicotinoid insecticides in China:
production, application and innovation. J. Pestic. Sci. 2013, 38(1), 1.
6) a) Huang, X.; Buchwald, S. L. New ammonia equivalents for the Pd-catalyzed amination of aryl
halides. Org. Lett., 2001, 3, 3417. b) Lee, S.; Jørgensen, M.; Hartwig, J. F. Palladium-catalyzed
synthesis of arylamines from aryl halides and lithium bis(trimethylsilyl)amide as an ammonia
equivalent. Org. Lett. 2001, 3, 2729
7) Zalesskiy, S. S.; Ananikov, V. P. Pd2(dba)3 as a precursor of soluble metal complexes
and nanoparticles: determination of palladium active species for catalysis and synthesis.
Organometallics, 2012, 31, 2302.
8) The superior performance of PdCl2(PhCN)2 over the other palladium (II) chloride salts
is believed to be due to its greater solubility in the reaction solvent.
9) DeAngelis, A. J.; Gildner, P. G.; Chow, R.; Colacot, T. J, Generating active “L-Pd(0)”
via neutral or cationic π-allylpalladium complexes featuring biaryl/bipyrazolylphosphines:
synthetic, mechanistic, and structure-activity studies in challenging cross-coupling
reactions. J. Org. Chem. 2015, 80, 6794.
10) Martinez, G. E.; Nugent, J. W.; Fout, A. R., Simple nickel salts for the amination of
(hetero)aryl
bromides
and
iodides
with
lithium
bis(trimethylsilyl)amide.
Organometallics 2018, 37, 2941.
ACS Paragon Plus Environment
26

Page 27 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
11) The solubility of 3 in water was measured to be 72mg/mL at room temperature.
Solubility of 3 in THF was measured to be 83 mg/mL at room temperature and in MeOH
was measured to be 127 mg/mL at room temperature.
9
12) Analysis of the aqueous layers revealed an additional benefit from using the sodium
sulfate solution as the yield loss to the aqueous layer decreased from 2% using water to
0.5% using the sulfate solution. This decrease in yield loss is likely due to the Hoffmeister
“salting out” effect from using a sodium sulfate solution. See Hyde, A. M.; Zultanski, S. L.;
Waldman, J. H.; Zhong, Y.; Shevlin, M.; Peng, F. General principles and strategies for
salting-out informed by the Hofmeister series. Org. Process Res. Dev. 2017, 21, 1355 for
a paper describing this effect.
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
13) Morimoto, T.; Takahashi, T.; Sekiya, M. “N,N-Bis(trimethylsilyl)methoxymethylamine
as a convenient synthetic equivalent for +CH2NH2: primary aminomethylation of
organometallic compounds” J. Chem. Soc., Chem. Commun, 1984, 794.
14) Boe, B. The mechanisms of solvolysis of organosilicon compounds containing bonds
from silicon to oxygen, nitrogen, and carbon. J. Organometal. Chem. 1976, 107, 139.
ACS Paragon Plus Environment
27