Development of a Robust Synthesis of Dactolisib on a Commercial Manufacturing Scale

Article abstract of DOI:10.1021/acs.oprd.9b00221

The development of the robust synthesis of 2-methyl-2-[4-[3-methyl-2-oxo-8-(quinolin-3-yl)-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-yl]-phenyl]propionitrile (dactolisib) on a commercial scale is described. The key step is a Pd-catalyzed Suzuki coupling of 2-[4-(8-bromo-3-methyl-2-oxo-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-yl)-phenyl]-2-methyl-propionitrile to 3-quinoline boronic acid. A special focus is placed on reducing the amount of Pd catalyst used in the Suzuki coupling and purifying the crude drug substance, including removing traces of Pd.

Full text of DOI:10.1021/acs.oprd.9b00221

Subscriber access provided by Nottingham Trent University
Full Paper
Development of a robust synthesis of dactolisib
at the commercial manufacturing scale
Markus Baenziger, Werner Pachinger, Frederic Stauffer, and Werner Zaugg
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00221 • Publication Date (Web): 15 Aug 2019
Downloaded from pubs.acs.org on August 15, 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 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Development of a robust synthesis of dactolisib at
the commercial manufacturing 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
Markus Baenziger #, *; Werner Pachinger #; Fréderic Stauffer §; Werner Zaugg #.
# Novartis Pharmaceuticals, GDD/TRD/Chemical & Analytical Development (CHAD), 4056
Basel, Switzerland
§ Novartis Institutes for BioMedical Research Basel
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 40
1
2
3
4
5
6
N
N
7
8
9
N
PdCl2(PPh3)2 / PPh3
OH
N
O
O
N
KHCO3 / DMF / H2O
N+
O
N
N
Br
N
O-
N
Br
N
NH2
N
N
OH
B
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
OH
Dactolisib
key steps: Suzuki coupling
purification and
Pd removal
Table of contents graphic
ACS Paragon Plus Environment

Page 3 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT: The development of the robust synthesis of 2-methyl-2-[4-[3-methyl-2-oxo-8-
(quinolin-3-yl)-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-yl]-phenyl]propionitrile (dactolisib) at
the commercial scale is described. The key step is a Pd-catalyzed Suzuki coupling of 2-[4-(8-
bromo-3-methyl-2-oxo-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-yl)-phenyl]-2-methyl-
propionitrile to 3-quinoline boronic acid. A special focus is placed on reducing the amount of Pd
catalyst used in the Suzuki coupling and purifying the crude drug substance, including removing
traces of Pd.
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: Pd-catalyzed Suzuki coupling; large scale; Pd removal; Smopex-234
INTRODUCTION
2-methyl-2-[4-[3-methyl-2-oxo-8-(quinolin-3-yl)-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-
yl]-phenyl]propionitrile (dactolisib; compound pure 9, Scheme 4) is a dual kinase inhibitor of
phosphatidylinositol 3-kinase (PI3K) and mammalian target of rapamycin (mTOR).1 Dactolisib is
currently investigated to treat several solid tumor forms2 and has activity against Plasmodium
falciparum (malaria) 3 and trypanosomiasis (sleeping sickness), as structural analogs of dactolisib
4
have been reported to show efficacy. Several syntheses of the target dactolisib have been
described in the literature. 4,5
Herein, we report and discuss the synthesis development of dactolisib towards a commercial-
scale process that was executed on a several hundred kg scale.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 40
1
2
3
4
5
6
7
8
9
The route to dactolisib starts from the commercially available intermediates 6-bromo-4-
hydroxy-3-nitro quinoline (1)5c and 2-(4-amino-phenyl)-2-methyl propionitrile (3)6. The free base
drug substance was manufactured in 6 steps according to the early phase batch synthesis shown in
scheme 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
In the first early development batch following the research route, intermediate 2 was isolated as
an evaporation residue from toluene. The conversion of intermediate 2 with starting material 3 to
product 4 was then performed in acetic acid, and product 4 was isolated at an 84 % yield over 2
steps. In the next step, the nitro group of intermediate 4 was hydrogenated with a Raney Ni catalyst
at normal pressure to the diamine product 5, which was isolated and purified via column
chromatography at a 69 % yield. Intermediate 5 was converted to product 6 with diphosgene at a
97 % yield. Methylation of intermediate 6 to product 7 with methyl iodide in N,N-
dimethylacetamide and NaH as the base resulted in a 70 % isolated yield. The coupling of
intermediate 7 with 3-quinoline boronic acid (8) required 6 mol % PdCl2(PPh3)2 catalyst. The
product 9 was then extracted with dichloromethane and purified by several precipitation steps from
aqueous media to obtain a purity of > 99 % HPLC area at a 73 % isolated yield (scheme 1).
ACS Paragon Plus Environment

Page 5 of 40
Organic Process Research & Development
1
2
3
N
4
5
6
7
O
N+
OH
N
Cl
N
O
N+
N
NH2
3
Br
Br
POCl3 / NPr3
toluene
O-
O-
NH
N
O
N+
Br
O-
acetic acid
y: 84% from 1
8
9
1
2
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
H2,Raney-Ni
MeOH / THF
y: 69%
O
Cl
N
N
Cl
Cl
N
O
O
I
Cl
O
N
NH
N
N
N
Br
Br
NH
NH2
Br
NaH / DMA
y: 70%
DCM / NEt3
y: 97%
N
N
7
6
5
N
PdCl2(PPh3)2 / PPh3
K2CO3 / DMF / H2O
y: 73%
OH
B
OH
8
N
O
N
N
N
N
9
Scheme 1: First generation synthesis
This synthesis route was considered suitable for further development and a subsequent
commercial process. The disconnection strategy to build the structure of dactolisib with three
fragments, 1, 3 and 8, seemed to be logical and straightforward. The synthetic route is atom
economic and does not require protecting groups. All of the intermediates are solids and can be
crystallized, which is another advantage of the syntheses shown in schemes 2 and 4. The major
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 40
1
2
3
4
5
6
7
8
9
anticipated challenge was to obtain the required purity of the drug substance due to the very low
solubility of dactolisib in most organic solvents and to remove the residual Pd to reach the low
level of < 2 ppm required for subsequent commercial processes. We also focused on avoiding any
column chromatography for purification and replacing undesired solvents, e.g., dichloromethane,
or very toxic raw materials, e.g., diphosgene.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Development of the synthesis
In the first step, 6-bromo-4-hydroxy-3-nitro quinoline (1) is converted into the 6-bromo-4-
chloro-3-nitro quinoline (2) with POCl3 as the chlorinating reagent.7 The aim of our development
work was to telescope the two steps from 1 to 4 without isolating unstable intermediate 2 as an
evaporation residue as done in previous research and early tox batch synthesis. The intermediate
(2) is not isolated and directly reacts with starting material 3 to yield product 4.8 Hence, starting
material 1 was reacted in toluene with 2 equiv of POCl3 in the presence of 1.5 equiv
tripropylamine. The reaction was controlled by POCl3 addition, and after the starting material 1
was consumed, toluene and an excess of POCl3 were distilled off. In the first tox batch, 3 equiv of
POCl3 had to be used mainly due to the approximately 20 % H2O present in starting material 1.
Later, the H2O in starting material 1 was controlled to < 1 % (m/m). A reaction temperature of >
90 °C was necessary to ensure complete and addition-controlled conversion from 1 → 2; therefore,
the higher boiling point tripropylamine (b.p. 156 °C) was used instead of triethylamine (b.p. 89
°C). Water quenching of the reaction mixture was not feasible since intermediate 2 would
immediately hydrolyze back to starting material 1. Therefore, an excess of POCl3 was distilled off
ACS Paragon Plus Environment

Page 7 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
after complete conversion of 1 → 2 by partial evaporation and the addition of toluene, followed
by a second partial evaporation. Then, the solvent was switched to acetonitrile. During these
distillations, concentration to dryness was avoided. The residual POCl3 (b.p. 106 °C) was not
quantified in the reaction mixture after the distillations. The thermal stability of the reaction
mixture was carefully examined by the DSC (differential scanning calorimetry) and RC1 (reaction
calorimetry) tests, and a strong thermal decomposition started from 180 °C. To assure safe
handling, the following safety measures were undertaken: Evaporation of the solvent never
proceeded to dryness and was performed at a jacket temperature < 70 °C; in case of an internal
temperature increase, the mixture was diluted with cold toluene or acetonitrile. In the initial tox
batch, the conversion of intermediate 2 → 4 was performed in acetic acid; however, it was found
that intermediate 2 hydrolyzed back to starting material 1 in acetic acid. Intermediate 2 was
sufficiently stable in acetonitrile. During the addition of starting material 3 (1.05 equiv), the
product 4 started to precipitate., However, after complete addition of 3, an HPLC check showed
that approximately 23 % of 3 was still present in the reaction mixture. Therefore, the reaction was
stirred for additional 18 h. The product 4 was isolated after the addition of H2O and NaOH (aq) at
a yield of 90 – 98 % with a purity of 90 – 98 % (HPLC area %) and could be conveniently used in
the next step without further purification. In a resupply campaign, the reaction of intermediate 2
with starting material 3 was performed in toluene after distilling off excess POCl3. After complete
conversion to intermediate 4, a solvent switch to 2-propanol was performed and product 4 was
precipitated by the addition of H2O. At the laboratory scale and in the plant, crust formation and
an inhomogeneous batch product were observed. This process variant was not pursued in
subsequent development batches. The POCl3 in the toluene distillate was safely quenched by
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

Organic Process Research & Development
Page 8 of 40
1
2
3
4
5
6
adding the distillate to hot water at 60 °C and stirring for 3 h at 60 °C, followed by neutralization
with an aqueous NaOH solution.
7
8
9
For the hydrogenation of the nitro group in compound 4, the following catalysts were screened:
Pt/Al2O3, Pt/C, Rh/Al2O3, and Raney-Ni. These catalysts showed similar selectivity; however, the
least expensive catalyst, Raney Ni,9 was selected. With this catalyst, formation of only 0.5 – 2 %
of desbromo impurity 10 was observed (HPLC area % at 210 nm; scheme 3). This amount of
impurity 10 can be tolerated and still guarantees the proper quality of product pure 9. Impurity 10
can be cleanly synthesized by Pd/C hydrogenolysis of intermediate 5 in 2-methyl-THF in the
presence of triethylamine. Moreover, the daughter impurity of impurity 10 in subsequent
downstream processes leads to impurity 11, which can be tolerated up to 10 % in intermediate
crude 9 to obtain qualified product pure 9. Due to catalyst poisoning by the residual NaH2PO4
(originating from the residual POCl3 in the reaction of 2 → 3) present in intermediate 4, Na2CO3
was added to neutralize the sodium dihydrogen phosphate in the hydrogenation mixture. Under
these conditions, hydrogenation ran smoothly. This nitro group hydrogenation could either be
performed at normal pressure or up to a pressure of 5 bar in THF. The water that formed in the
nitro group reduction and that was introduced with the Raney Ni catalyst had to be removed via
azeotropic distillation in the presence of the reaction solvent THF and monitored by an in-process
steering control via Karl Fisher titration with a limit of < 1000 ppm. The goal for future
development was to avoid column chromatography and not to isolate the Ames positive
intermediate 5. In the first generation research synthesis, diphosgene5a was used to generate the
cyclic urea intermediate 6. Diphosgene is toxic, was used in chemical weapons in World War I
and is not readily commercially available at a large scale. Moreover, the transport regulations of
large quantities of diphosgene are quite strict. We found that the nontoxic phosgene equivalent
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

Page 9 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
carbonyldiimidazole10 worked nicely in this transformation. The use of triphosgene was not
evaluated further. After azeotropic distillation of water, carbonyldiimidazole (1.9 – 2 equiv) was
added; product 6 precipitated from the reaction mixture and was isolated by filtration at a yield of
75 – 85 %. Adding a sacrificial amount of carbonyldiimidazole to remove H2O was not tested.
Methylation of 6 to 7 was performed in THF11. In the first generation research synthesis, NaH
was used as a base. On pilot plant scale, the base was changed to a solution of KOtBu in THF,
which was easier to handle than solid NaH. Methyl iodide was used as a methylating agent in the
first development batches, and precipitated product 7 was isolated by filtration after the addition
of H2O at a 70 – 75 % yield. Product 7 was then carried onwards in the synthesis as a crude product.
In the late phase of development, the methylating agent was changed to dimethyl sulfate, which
has the advantage of being less volatile than methyl iodide; however, both methylating agents are
carcinogens. Crude product 7 was recrystallized in late phase development batches. The
methylation could be performed at room temperature using both methylating agents. Based on
RC1 experiments, the reaction was assessed as non-critical. No further optimization of the addition
rate of dimethyl sulfate (standard rate 2.5 h) was investigated. The quaternization of the quinoline
nitrogen was not monitored. Dimethyl carbonate could also be used for this methylation as a
methylating agent in DMF at 130 °C for 3 h in the presence of K2CO3; however, a significant
amount of impurity 12 (up to 1 area % by HPLC at 210 nm; scheme 3) was formed under these
conditions. This impurity 12 and its corresponding daughter impurity after Suzuki coupling with
3-quinoline boronic acid 8 did not meet the specification of < 0.1 area % by HPLC in pure 9.
Furthermore, use of dimethyl carbonate as a methylating agent yielded dark-colored intermediate
7, and even the color of pure 9 derived from this material did not meet our internal specifications.
Therefore, intermediate 7 originating from the dimethyl carbonate process had to be purified,
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

Organic Process Research & Development
Page 10 of 40
1
2
3
4
5
6
7
8
9
which was achieved using a slurry in pyridine, followed by recrystallization from anisole at an 84
% yield. In this purification step, the critical impurity 12 (scheme 3) was reduced to 0.1 %. Due to
these findings and the necessity of performing an extra laborious purification of intermediate 7,
we decided to use dimethyl sulfate as a methylating agent for subsequent production batches. At a
large scale, the 6 → 7 step can be performed with an average yield of 70 – 75 % after
recrystallization from DMF. The critical impurity 12 was consistently present at no more than 0.2
% (HPLC area % at 210 nm). In the isolated intermediate 7, dimethyl sulfate was determined to
be < 1 ppm. This late phase development synthesis of intermediate 7 is outlined in scheme 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N
O
N+
OH
N
Cl
N
O
N+
N
NH2
3
Br
Br
POCl3 / NPr3
toluene
O-
O-
NH
N
O
N+
Br
O-
acetonitrile
y: 96% from 1
1
2
4
H2,Raney-Ni
THF
O
O
O
O
or:
I
S
N
N
O
O
N
N
O
O
N
N
N
O
O
or:
N
NH
N
N
N
Br
Br
NH
NH2
Br
KOtBu / THF
y: 78%
THF
N
N
y: 78% from 4
7
6
5
Scheme 2: Development synthesis of key intermediate 7
ACS Paragon Plus Environment

Page 11 of 40
Organic Process Research & Development
1
2
3
4
N
N
N
5
6
7
8
9
O
O
N
NH
N
N
O
Br
NH2
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
N
10
11
12
Scheme 3: Key impurities in the synthesis of dactolisib
Development of the key step, a Suzuki coupling reaction, to form intermediate crude 912
In the manufacturing of the first development batch, 6 mol % of PdCl2(PPh3)2 (the same catalyst
used in the research synthesis) was used. The reaction was performed in a DMF / H2O mixture
using K2CO3 as the base. At the completion of the reaction, the product was isolated after tedious
extractions with methylene chloride, followed by multiple precipitations / crystallizations. In the
first GMP development batch, KHCO3 was used as a base to avoid possible corrosion of the glass-
lined vessels in the plant by the more basic K2CO3 at elevated temperatures, as observed in the
corrosion testing lab. Hence, quinoline-3-boronic acid (8) was dissolved in a solution of DMF and
aqueous KHCO3 and added to compound 7 and the PdCl2(PPh3)2 catalyst (3 mol %) in DMF.
During the addition, the internal temperature of the reaction mixture had to be kept above 95 °C
(usually the IT (internal temperature) was in the range from 96 °C to 102 °C and the jacket
temperature was 110 °C – 115 °C) to keep the catalytic cycle active and to run the reaction in an
addition-controlled manner. If the internal temperature during the addition dropped to < 95 °C, a
significant amount of the desbromo impurity 11 formed. Under the ideal operating conditions
outlined above, impurity 11 usually formed at 0.5 % - 3 % area % (HPLC at 210 nm). An HPLC
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 40
1
2
3
4
5
6
7
8
9
check immediately after the complete addition of 8 showed < 5 % of compound 7. The reaction
mixture was then stirred at > 95 °C for additional 2 h. To ensure the complete removal of residual
compound 7 in the purification process of crude 9 → pure 9, the limit of residual 7 in the process
control (IPC) of the step crude 9 was set to < 2 % (area % HPLC). The addition time of compound
8 was set at 0.5 h – 2 h. The starting material 3-quinoline boronic acid 8 began to exothermally
decompose at 116 °C with a decomposition energy of 427 J / g. To ensure the thermal safety of
the process, an addition-controlled reaction to crude 9 and a high dilution of starting material 8 to
approximately 5 % in DMF / KHCO3 (aq) were necessary.
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 product crude 9 was isolated by filtration after precipitation by the addition of water. The
product crude 9 was isolated at a 90 – 97 % yield with a purity of > 90 % (HPLC, area %) and
contained 6000 – 7000 ppm Pd and 30 – 200 ppm Ni. However, this purity and the Pd and Ni
levels were not acceptable for API quality. To achieve the required level of purity of the drug
substance of > 99 % (HPLC area %) and residual Pd and Ni levels of < 2 ppm, further purification
steps were necessary. The low tolerable level of Pd (< 2 ppm) in API was due to the high projected
daily dose of dactolisib. Furthermore, starting materials 1, 3, and 8 and intermediates 4 and 5 had
positive Ames test results. The limits of the Ames-positive starting materials and intermediates
were checked based on the ICH guideline for product pure 9 or the corresponding tosylate salt of
pure 9 at a level of 1.9 ppm for each of these impurities. This level can be achieved for each
impurity and for the methylating agent dimethyl sulfate, which was used in the conversion of
intermediate 6 → 7. Potential byproducts or intermediates of the nitro group reduction
(hydroxylamine, etc.) were not considered in this genotoxic evaluation.
For subsequent development batches, the goal was to further reduce the amount of the Pd catalyst
to decrease the catalyst cost and the level of residual Pd contamination in the isolated product
ACS Paragon Plus Environment

Page 13 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
crude 9. Initially, we tested a range of Pd / C catalysts with or without the addition of PPh3.13
Reactions were performed in DMF / H2O or in NMP / H2O. In all cases, the target product crude
9 was formed as the major reaction product; however, a significant amount of impurity 11 (> 10
% area %; HPLC) was present in the reaction mixture. Therefore, the reaction to crude 9 with the
PdCl2(PPh3)2 catalyst was further investigated. We tried to reduce catalyst loading while
simultaneously adding 2 Pd equiv of PPh3 to the reaction mixture. The Pd catalyst levels could be
reduced to 0.3 mol %, and the reaction remained addition-controlled with regard to 8 (see Table
1). Other homogeneous catalysts than PdCl2(PPh3)2 were not considered due to the air stability of
this catalyst and cost considerations. In laboratory experiments, we were able to reduce the catalyst
amount to < 0.2 mol % by adding 2 Pd equiv of PPh3 and achieve an acceptable yield and purity
of crude 9. Without the addition of PPh3, the Suzuki coupling did not work at these low catalyst
loads. We hypothesized that adding 2 equiv PPh3 could help to stabilize the catalytically active
Pd(0) complex and that PPh3 might act as a reducing agent for the Pd(II) to Pd(0) conversion.
Table 1: Key experiments for the Suzuki coupling to product crude 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
Catalyst
Reaction Products by HPLC
Yield of
isolated
crude 9
Not
Pd / C 5 % (20 % m/m of 7)
type 39 paste (Johnson Matthey)
PdCl2(PPh3)2; 3 mol %
72.3 % (9); 1 % (7); 23.8 % (11)
isolated
95 %
94.7 % (9); 2.6 (7); 2.6 (11)
PdCl2(PPh3)2; 0.3 mol % + 0.6 mol %
PPh3
96.3 % (9); 1.6 % (7); 1.5 % (11)
97 %
The reaction was finally run in the following manner: starting material 7 and the Pd catalyst in
DMF were placed in a reactor, and a solution of 8 in DMF / KHCO3 (aq) was added at 90 – 95 °C.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 40
1
2
3
4
5
6
7
8
9
The reaction mixture was stirred for 2 h and then checked by HPLC. The product was precipitated
by the addition of H2O and isolated via filtration. According to this optimization, the amount of
Pd could be reduced by a factor of 10 to a level of < 700 ppm in the isolated product crude 9. The
yield of the crude product remained approximately same (90 – 95 %) with a purity of > 98 %
(HPLC area %). Starting material 8 was only soluble in this DMF / KHCO3 (aq) solution at > 60
°C. However, when this solution was aged for a prolonged period of time (> at 60 °C; > 5 h) prior
to the addition to compound 7 / catalyst mixture, a significantly higher amount of the desbromo
impurity 11 was formed, probably due to the hydrolysis of DMF under these conditions. The 1H-
NMR data indicated that stirring a solution of compound 8 in DMF /KHCO3 (aq; D2O) at a jacket
temperature of 70 °C resulted in the formation of 0.6 mol % potassium formate versus DMF after
4 h and that after 19 h, 2.6 mol % potassium formate formed. Based on this result, in another
laboratory experiment, we added 65 mol % of potassium formate (corresponding to 2.6 mol %
potassium formate versus DMF) to the solution of 8 in DMF / KHCO3 (aq). After reaction with
the starting material 7 under normal Pd-catalyzed conditions, the IPC of the reaction mixture
showed < 2 % of residual 7 and the formation of 3.1 % of impurity 11, which was depleted to <
1% (HPLC area %) in the isolated product crude 9. Potassium formate presumably serves as a
reducing agent for the reaction 7 → 11. This hypothesis was further supported by a control
experiment in which a mixture of compound 7 together with the catalyst in the presence of an
aqueous KHCO3 solution but in the absence of starting material 8 was stirred for several hours at
an internal temperature (IT) of > 95 °C. After 4 h at > 95 °C, 23 % of impurity 11 formed.
Furthermore, in another control experiment, DMF was replaced by N,N-dimethylacetamide as the
reaction solvent; after 4 h stirring at > 95 °C without the addition of starting material 8, only
approximately 1 % of desbromo 11 formed. Then, 3-quinoline boronic acid (8) was added, 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
ACS Paragon Plus Environment

Page 15 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
reaction to the product crude 9 proceeded smoothly providing product 9 at > 96 % area % and
desbromo 11 at approximately 1.5 % (HPLC area %) in the in-process control 2 h after the addition
of 8. Usually, the solution of starting material 8 was maintained for a maximum of 2 h before
addition to the reaction mixture of 7. In an experiment in which the solution of starting material 8
in DMF / KHCO3 was stirred for 7 h at a jacket temperature of 70 °C, 8 % of impurity 11 was
observed in the product crude 9. This amount of impurity 11 could be depleted in the purification
step to pure 9; however, a concomitant yield loss was observed. Excess 3-quinoline boronic acid
(8) was initially set at 1.2 equiv; however, in larger batches, it could be successfully reduced to 1.1
equiv and even to 1.05 equiv. However, the optimal amount of the starting material 8 appeared to
be 1.1 equiv versus 7.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the initial development batches, the product crude 9 was isolated after drying, which had an
impact on the throughput of the process since the water wet product crude 9 had to be dried,
resulting in long cycle times. However, the quality of the isolated product crude 9 did not meet
the quality requirements for the drug substance with regards to HPLC purity and residual Pd,
resulting in the need for a recrystallization / purification step. For the initial development batches,
the dried intermediate crude 9 was recrystallized in a formic acid /methanol system. Later, it was
found that purification could be performed in acetic acid / H2O, and the drying step at intermediate
crude 9 was omitted. These alterations yielded a satisfactory purity of the product pure 9 with
regard to HPLC purity and residual Pd.
Recrystallization of the crude drug substance and removal of Pd
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 40
1
2
3
4
5
6
7
8
9
Crude 9 is almost completely insoluble in most organic solvents, making it very difficult to
design a feasible recrystallization system. Formic acid was identified as a good solvent for product
9, which probably forms a soluble formate salt. We therefore designed a recrystallization process
that involved the dissolution of crude 9 in formic acid and the addition of activated charcoal
(NORIT C EXTRA USP grade) to reduce Pd and Ni. Additionally, charcoal served as a
decolorizing agent. In addition to Smopex-23414, a Pd and Ni scavenger was added to further
reduce the levels of heavy metals. This mixture was stirred at room temperature; charcoal and
Smopex-234 were filtered off and MeOH was finally added at 60 °C. Under these conditions,
methyl formate formed; the purified product pure 9 precipitated and could be isolated by filtration.
To enhance the purity of the drug substance to the required level of > 99 % (HPLC area %) and
reduce the residual Pd and Ni levels to < 2 ppm, the preceding purification method had to be
repeated a second time. In the first cycle, approximately 80 % of the Pd initially present in crude
9 was removed by charcoal and Smopex-234; approximately 12 % of Pd ended up in the mother
liquor and 5 % of the initial Pd amount remained in the precipitated product. In the second cycle,
approximately 90 % of the remaining Pd was removed by charcoal and Smopex-234. Thus, the
purified drug substance was isolated at a > 90 % yield with an HPLC purity of > 99 % and Pd and
Ni levels of < 2 ppm. After the first recrystallization cycle, the Pd level was in the range from 20
to 100 ppm. When only charcoal was used, the once-precipitated product pure 9 still contained
approximately 200 ppm Pd, which was too high. Therefore, a combination of charcoal and
Smopex-234 was used as a Pd scavenger in the subsequent development. QuadraSil MP and 3-
mercaptopropyl ethyl sulphide silica were also investigated as Pd scavengers, and their efficacy
for Pd removal was similar to that of Smopex-234. Then, Smopex-234 was used as a scavenger
in combination with charcoal. Experiments using only Smopex-234 were not performed since 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
ACS Paragon Plus Environment

Page 17 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
addition of charcoal proved to be necessary to remove the color from product pure 9.
Recrystallization and an investigation of Pd removal without the addition of charcoal and Smopex-
234 were not performed. The problem of H2 and CO2 release due to Pd-catalyzed decomposition
of formic acid was addressed in the scale-up campaign, and safety measures were implemented.
The exhaust of the reactor was passed through a vessel filled with water and was then directly
released over the roof. The mother and washing liquor tanks were immediately sent to an
incineration plant without intermittent storage.
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
However, this process has the following four disadvantages:
-
Product crude 9 originates from a precipitation process from DMF / H2O containing salts
(K2CO3 and KHCO3). The purification procedure from intermediate crude 9 to product pure
9 is performed under anhydrous conditions (methanol, formic acid, and methyl formate), and
thus, residual salts from the previous Suzuki coupling are not removed, which may lead to a
potential risk of elevated sulfated ash levels in the drug substance.
-
-
Intermediate crude 9 has to be dried; otherwise, the formation of formic acid methyl ester
with concomitant precipitation of pure 9 is not complete due to the presence of water, which
will result in a large yield loss after isolation of pure 9.
The presence of a significant residual amount of Pd in crude 9 may result in H2 and CO2
formation due to the catalytic decomposition of formic acid. Indeed, in a pilot plant
campaign, pressure was formed in a drum of mother liquor, which can lead to a potentially
dangerous situation, especially since H2 is formed. Special safety measures had to be
implemented, and these measures are not acceptable for a commercial process.
-
The volume changes in this step are not favorable for the plant. Intermediate crude 9 is
dissolved in 5 volumes of formic acid and then concentrated to a low volume (2-fold) with
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 40
1
2
3
4
5
6
regard to the starting intermediate crude 9. Then, 8 volumes of methanol are added. Due to
these disadvantages, an alternative purification method had to be sought.
7
8
9
An alternative solution using an acetic acid / H2O mixture was found. To further reduce the
levels of Pd in crude 9, cysteine was added to the H2O used for precipitation of the product
crude 9. The Pd levels in isolated and dried product crude 9 were consistently < 500 ppm using
a final loading of 0.3 mol % PdCl2(PPh3)2. However, on the plant scale, the product crude 9 was
not dried and was dissolved as a water-wet filter cake in acetic acid; Smopex-234 and charcoal
were added, and the mixture was stirred at 90 °C for 1 h. Then, charcoal and Smopex-234 were
removed via filtration, and charcoal and Smopex-234 were added again to the filtrate; the
mixture was stirred for 1 h at 90 °C, and the solids were filtered off. The filtrate was evaporated
to a volume of approximately 5 – 5.5-fold versus the theoretical yield. The purified product was
precipitated by the addition of 16.4 volumes of deionized H2O. The yield of these two-telescoped
steps was approximately 85 % - 90 % and the purity was usually > 99.5 % HPLC (area %). The
product pure 9 was dried at 80 °C until AcOH was < 1 % (m/m). The final step of the late phase
synthesis of dactolisib is outlined in 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
N
N
N
PdCl2(PPh3)2 / PPh3
KHCO3 / DMF / H2O
recrystallisation
HOAc / H2O
O
N
O
N
O
N
N
N
N
N
N
Br
SMOPEX234
charcoal
y: 85 - 90%
N
N
N
N
OH
B
OH
7
8
crude 9
pure 9
Scheme 4: Synthesis of dactolisib from the key intermediate 7
ACS Paragon Plus Environment

Page 19 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The isolated product crude 9 contained a variable amount of H2O. In the lab (filtration over a
nutsch filter), the water levels were usually up to approximately 30 – 50 % of the theoretical
amount of the product crude 9, and in the plant (centrifugation), the water levels were 20 – 50 %
of the theoretical amount of the product crude 9. The solubility of the product pure 9 in the AcOH
/ H2O system reaches a maximum at a ratio of about 4:1 (AcOH : H2O) either at RT as well as 90
°C, see 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
SOLUBILITY OF PURE 9 IN ACETIC ACID / WATER SYSTEM
5°C
90 °C
160
140
120
100
80
147
62.7
62.3
60
38.8
40
14.8
2.7
11.7
0%
20
8.5
0
10%
20%
30%
40%
50%
60%
70%
80%
weight % of water in solvent mix
Figure 1: Solubility of pure 9 in AcOH / H2O system
Some residual H2O is beneficial to dissolve the crude 9 in AcOH. The impact of various amounts
of H2O brought into the recrystallization process from crude 9 → pure 9 was investigated. The
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 40
1
2
3
4
5
6
results are summarized in the following Table 2. In all experiments listed in table 2, charcoal and
Smopex-234 were used for the Pd removal.
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
H2O content in
crude 9
Yield
(%)
Residual Pd in pure 9 *
(ppm)
HPLC purity
(area %)**
(g/g crude 9)
10
20
92.9 %
92.9
< 1
< 1
< 1
< 1
< 1
99.9
99.9
99.7
99.8
99.4
50
93.6
100
93.6
200
95.7
* The starting material crude 9 contained 342 ppm Pd.
** The starting material crude 9 had an HPLC purity of 98.9 % (area %).
Table 2: Impact of the H2O content in crude 9 on the quality and yield of isolated pure 9
The data indicate that the H2O content in wet crude 9 is independent of the Pd removal efficiency
in this process. However, if > 100 % m/m H2O is present in crude 9 (calculated on dried basis),
the efficiency of the depletion of byproducts is decreased. Higher levels of residual H2O in wet
crude 9 slightly increase the yield, as expected.
With this final recrystallization process of dactolisib from acetic acid / H2O, it is possible to
easily remove up to 10 % of impurity 11 into the mother liquor; however, only approximately 2 %
of the remaining starting material 7 is purged. Hence, the limit of residual starting 7 in the in-
ACS Paragon Plus Environment

Page 21 of 40
Organic Process Research & Development
1
2
3
4
5
6
process control of the reaction 7 + 8 → crude 9 was set to < 2 %. The Pd levels in the isolated
purified product pure 9 were usually below 2 ppm.
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
CONCLUSION
We successfully developed a robust production process for the synthesis of dactolisib. The
amount of Pd catalyst PdCl2(PPh3)2 can be reduced to 0.3 mol % in the Suzuki coupling by the
addition of triphenylphosphine as a supplementary ligand. The intermediate crude 9 after the
Suzuki coupling was not dried in the final process, thus improving the process throughput. Pd can
be removed with charcoal / Smopex-234, and the final purified free base pure 9 of dactolisib was
isolated at an 85 – 90 % yield (Suzuki coupling and purification) with a purity of > 99.5 % (HPLC
area %) and Pd and Ni < 2 ppm. The number of isolated products can be reduced from 6 in the
first generation synthesis (scheme 1) to 4 in the final process (scheme 2 and 4).
EXPERIMENTAL SECTION
General:
All solvents and reagents used were of bulk quality.
The NMR spectra were recorded on a Bruker Avance 400 (1H NMR: 400 MHz; 13C NMR: 100
MHz) spectrometer in the indicated solvent. Coupling constants (J) are reported in hertz (Hz). The
accuracy of the measurement of the coupling constants was within +/- 0.2 Hz.
HPLC was performed on an Agilent 1100 or 1200 machine. HR-LC/MS were measured on a
Waters Acquity UPLC/Synapt G2 Q-TOF MS instrument and were performed by using
electrospray ionization in the positive ion mode after separation by liquid chromatography (Nexera
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 40
1
2
3
4
5
6
7
8
9
from Shimadzu). The elemental composition was derived from the mass spectra acquired at a high
resolution of approximately 30,000 on an LTQ Orbitrap XL mass spectrometer (Thermo
Scientific). A high mass accuracy below 1 ppm was obtained by using a lock mass.
All reactions were carried out under inert atmosphere (N2).
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
Synthesis of 2[4-(6-bromo-3-nitro-quinolin-4-yl amino)-phenyl]-2-methyl-propionitrile
(4). 6-Bromo-4-hydroxy-3-nitroquinoline (78.5 kg; 292 mol) was placed in a reactor together with
toluene (463 kg) at RT. Then, tripropylamine (62.7 kg; 438 mol; 1.5 equiv) was added. The
suspension was then heated to 80 °C, and POCl3 (89.4 kg; 584 mol; 2 equiv) was added within
approximately 1 h. The reaction mixture was heated to 90 °C and stirred at 90 °C – 95 °C for 2 h,
and the turnover of the reaction was checked by HPLC (requirement: < 1 % starting material). The
reaction mixture was cooled to 65 – 70 °C, and 345 kg toluene was distilled off. Then, four times
toluene (262 kg) was added, and the same amount of toluene was distilled off. The solvent was
exchanged to acetonitrile. Four times acetonitrile (235 kg) was added, and the same amount of the
solvent was distilled off. Finally, acetonitrile (628 kg) was added. The internal temperature was
decreased to 25 °C, and a solution of 2-(4-amino-phenyl)-2-methyl propionitrile (3) (49 kg; 306
mol; 1.05 equiv) in 78.5 kg acetonitrile was added within 1.5 h. During the addition of starting
material 3, product 4 precipitates from the reaction mixture. The reaction mixture was stirred at
RT for 18 h. The reaction was checked with HPLC (requirement: < 1.0 % area % intermediate 2).
Finally, H2O (3,000 kg) was added, and 1,268 kg of the solvent was distilled off under vacuum at
70 °C. After the distillation of the first 700 l of the solvent, additional H2O (900 kg) was added.
When the distillation was stopped, the pH was adjusted to 5 – 6 by adding 30 % NaOH (93 kg).
The reaction mixture was then cooled to 0 ° C within 7 h and stirred at that temperature for 2 h;
ACS Paragon Plus Environment

Page 23 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
product 4 was finally isolated via filtration. The filter cake was washed with acetonitrile : H2O =
1:10 (240 kg). The wet product was dried at 60 °C in vacuum to give 115.4 kg of product 4, which
corresponded to a 96 % yield. The purity of the production batches were in the range from 90 to
98 % (HPLC area %). This product was used in the next step without further purification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1H NMR (400 MHz, DMSO-d6) δ 10.15 (s, 1H), 9.07 (s, 1H), 8.71 (d, J = 1.9 Hz, 1H), 7.98 (dd,
J = 8.9, 1.9 Hz, 1H), 7.91 (d, J = 8.9 Hz, 1H), 7.47 (d, J = 8.6 Hz, 2H), 7.14 (d, J = 8.6 Hz, 2H),
1.68 (s, 6H).
13C NMR (101 MHz, DMSO-d6) δ 148.17, 147.72, 141.40, 140.59, 137.92, 135.58, 132.08,
130.19, 127.23, 126.49, 125.12, 123.02, 120.48, 120.42, 36.71, 28.77.
HR-LC MS: m/z [M + H]+ calcd for C19H16N4O2Br: 411.0457; found: 411.0350.
Synthesis of 2-[4-(8-bromo-2-oxo-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-yl)-phenyl]-2-
methyl-propionitrile (6).
Starting product 4 (113.9 kg, 277 mol) and Na2CO3 (5.7 kg) were combined with THF (1,481
kg) into an autoclave. The Raney nickel catalyst (21 kg) was added, and the mixture was
hydrogenated under normal pressure of molecular hydrogen up to the uptake of 3 equiv of H2
(usually, approximately 8 – 10 h). At this point, no significant consumption of H2 was observed.
The turnover of the reaction was checked by HPLC (< 1 % of intermediate 4). The catalyst was
then filtered off and washed four times with THF (122 kg). THF was distilled off under vacuum
(1,400 kg). Then, twice THF (473 kg) was added and distilled off. More THF (1,416 kg) was
added, and the H2O content was measured (limit: 1000 ppm; KF-titration). Finally,
carbonyldiimidazole (86 kg, 531 mol; 1.9 equiv) was added. The reaction mixture was heated to
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 40
1
2
3
4
5
6
7
8
9
60 °C within 1.5 h and was stirred at this temperature for 3 h. The reaction was checked by HPLC
for completion (< 1 % of intermediate 5; HPLC area %), and the suspension was cooled to 0 °C –
5 °C within 3 h and stirred for another 2 h; then, the crude product was filtered. The filter cake was
washed with THF (471 kg). The crude THF wet product was charged into a reactor together with
H2O (439 kg) and suspended at IT 40 – 50 °C for 1 h. Then, the product was filtered, washed with
H2O and dried at 90 °C until the H2O content (KF) was < 0.4 % to obtain 88 kg of intermediate 6
(78 % yield). The purity of the isolated product was in the range from 92 % to 97 % (HPLC area
%) of the production batches.
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
1H NMR (400 MHz, DMSO-d6) δ 11.82 (s, 1H), 8.80 (s, 1H), 7.94 (d, J = 9.1 Hz, 1H), 7.89 –
7.80 (m, 2H), 7.75 – 7.67 (m, 2H), 7.65 (dd, J = 9.1, 2.1 Hz, 1H), 6.97 (d, J = 2.1 Hz, 1H), 1.82 (s,
6H).
13C NMR (101 MHz, DMSO-d6) δ 153.88, 143.76, 143.08, 135.16, 134.90, 132.86, 129.94,
129.91, 129.12, 127.25, 124.83, 122.80, 122.60, 119.30, 116.43, 37.32, 28.83.
HR-LC MS: m/z [M + H]+ calcd for C20H16N4OBr: 407.0507; found 407.0530.
Intermediate 5 could be isolated after hydrogenation by evaporation of THF and crystallization
from n-heptane. Analytical results of intermediate 5:
1H NMR (500 MHz, DMSO-d6; 100 °C) δ 8.67 (s, 1H), 7.95 – 7.85 (m, 1H), 7.79 (d, J = 8.8
Hz, 1H), 7.64 (s, 1H), 7.49 – 7.36 (m, 1H), 7.31 – 7.22 (m, 2H), 6.68 – 6.53 (m, 2H), 5.15 (br. s,
2H), 1.65 (s, 6H).
13C NMR (126 MHz, DMSO-d6; 100 °C) δ 144.18, 143.97, 140.56, 137.30, 130.70, 130.65,
127.75, 126.43, 125.17, 124.26, 123.09, 121.48, 119.02, 113.83, 35.20, 28.00.
HR-LC MS: m/z [M + H]+ calcd for C19H18N4Br: 381.0715; found 381.0746.
ACS Paragon Plus Environment

Page 25 of 40
Organic Process Research & Development
1
2
3
4
5
6
7
8
Synthesis of 2-(4-((3-aminoquinolin-4-yl)amino)phenyl)-2-methylpropanenitrile (10)
Intermediate 5 (5 g; 13.1 mmol) was dissolved in 100 ml 2-Me-THF together with 2 g (19.8
mmol; 1.5 equiv) triethylamine. Then, 0.5 g of Pd/C (10 %) catalyst was added, and the mixture
was hydrogenated in an autoclave at RT with 1 bar overpressure for a total of 8 h. A LC/MS check
of the reaction mixture showed complete consumption of starting material 5. The catalyst was
filtered off and washed with 2-Me-THF, and the 2-Me-THF product solution was washed with 50
ml 0.05 M citric acid, followed by two washes with 50 ml deionized H2O. The 2-Me-THF solution
of compound 10 was evaporated to dryness to obtain 3.5 g product 10 as a brown evaporation
residue (88 % yield).
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
Analytical results of compound 10
1H NMR (400 MHz, DMSO-d6) δ 8.63 (s, 1H), 7.96 – 7.79 (m, 2H), 7.73 – 7.62 (m, 1H), 7.43
– 7.30 (m, 2H), 7.30 – 7.19 (m, 2H), 6.62 – 6.50 (m, 2H), 1.61 (s, 6H).
13C NMR (101 MHz, DMSO-d6) δ 144.91, 143.99, 142.20, 137.31, 130.33, 129.04, 126.54,
126.10, 125.77, 125.11, 124.13, 122.68, 121.68, 114.00, 35.64, 28.41.
HR-LC MS: m/z [M + H]+ calcd for C19H19N4: 303.16042, found: 303.16068.
Synthesis of 2[4-(8-bromo-3-methyl-2-oxo-2,3-dihydro-1H-imidazo[4,5-c]quinolin-1-yl)-
phenyl]2-methyl-propionitrile (7).
Intermediate 6 (166 kg; 408 mol) was placed together with THF (889 kg) in a reactor. Then, a
solution of 20 % m/m KOtBu (343 kg; 611 mol; 1.5 equiv) in THF was added within 2 – 3 h under
cooling at 20 °C – 25 °C. The addition tank was rinsed with THF (45 kg). The mixture was stirred
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 40
1
2
3
4
5
6
7
8
9
at 20 °C – 25 °C for 0.5 h, and dimethyl sulfate (77 kg; 610 mol; 1.5 equiv) was added within 2.5
h. The reaction mixture was stirred for 2 h at 25 ° - 30 °C. Then, the reaction was checked with
HPLC (requirement < 2 % of 6). A solution of 5 % NH3 in H2O (316.5 kg) was added, and the
mixture was stirred for 0.5 h at 25 °C – 30 °C. Product 7 was filtered and washed with THF (182
kg) followed by a mixture of methanol (93 kg) and H2O (93 kg). The crude wet product was
charged in a reactor together with DMF (1717 kg) and activated charcoal (8.6 kg). The mixture
was heated at 120 °C within 2.5 h and stirred at that temperature for 0.5 h. Then, the charcoal was
filtered off and rinsed with DMF (345 kg). The filtrate was cooled within 5 h to 5 °C and stirred
at that temperature for 1.5 h; then, the product was filtered and washed with TBME (452 kg). The
product was dried in vacuum at 45 °C to obtain 133.8 kg product 7 (78 % yield). The purity level
of the 9 production batches: > 99.0 % (area % HPLC); impurity 12 < 0.2 % (area % HPLC).
1H NMR (400 MHz, acetic acid-d4) δ 9.20 (s, 1H), 8.22 (d, J = 9.2 Hz, 1H), 8.02 – 7.88 (m, 2H),
7.81 (dd, J = 9.2, 2.0 Hz, 1H), 7.73 (d, J = 8.6 Hz, 2H), 7.23 (d, J = 2.0 Hz, 1H), 3.76 (s, 3H), 1.90
(s, 6H).
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
13C NMR (101 MHz, acetic acid-d4) δ 154.10, 144.79, 140.04, 133.67, 132.19, 130.97, 130.94,
129.25, 128.17, 127.10, 123.87, 123.67, 123.10, 120.87, 115.63, 37.24, 28.27, 27.71.
HR-LC MS: m/z [M + H]+ calcd for C21H18N4OBr: 421.0664; found 421.0636.
Analytical data of compound 12
1H NMR (400 MHz, CDCl3) δ 7.82 – 7.71 (m, 2H), 7.53 – 7.47 (m, 3H), 7.31 (d, J = 9.1 Hz,
1H), 6.88 (d, J = 2.2 Hz, 1H), 3.86 (s, 3H), 3.79 (s, 3H), 1.85 (s, 6H).
13C NMR (101 MHz, CDCl3) δ 153.76, 153.71, 143.58, 136.00, 134.34, 130.99, 129.14, 126.96,
126.03, 124.18, 123.86, 116.97, 115.43, 115.12, 112.92, 37.36, 29.76, 29.49, 29.29.
ACS Paragon Plus Environment

Page 27 of 40
Organic Process Research & Development
1
2
3
4
HR-LC MS: m/z [M + H]+ calcd for C22H20N4O2Br: 451.07642; found 451.07642.
5
6
7
8
9
Synthesis of 2-methyl-2-[4-(3-methyl-2-oxo-8-quinolin-3-yl-2,3-dihydro-1H-imidazo[4,5-
c]quinolin-1-yl)-phenyl]propionitrile (crude 9; pure 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
Intermediate 7 (155 kg; 368 mol) was placed in a reactor together with DMF (961 kg) at RT
under N2. The mixture was degassed by evacuating and flushing several times with N2. Then,
bis(triphenylphosphine) palladium dichloride (0.78 kg; 1.1 mol; 0.3 mol %) and
triphenylphosphine (0.62 kg; 2.4 mol; 0.64 mol %) were added. In a second reactor, 3-quinoline
boronic acid 8 (70 kg; 405 mol; 1.1 equiv) was dissolved in DMF (721 kg) at 60 °C. A solution of
KHCO3 (207 kg) in H2O (1033 kg) was added at 60 °C, and the solution of 8 was degassed.
Meanwhile, the first reactor was heated to 95 °C – 100 °C, and the hot solution of 8 was added
within 1.5 h at > 95 °C to the starting 7. The reaction mixture was stirred for 2– 6 h at this
temperature and was checked by HPLC (requirement: < 2 % 7; area %). A solution of L-cysteine
(13.7 kg) in demineralized H2O (115 kg) was added within 20 minutes. The mixture was stirred at
> 95 °C for 2 h. Then, demineralized H2O (712 kg) was added. The mixture was cooled within 9
h to 0 °C and was stirred at that temperature for 4 h. The product crude 9 was filtered and washed
with H2O (2,296 kg),to obtain 209 kg wet crude 9 (theoretical 100 % yield: 172 kg), which was
directly converted to the product pure 9.
This crude product contained 151 ppm Pd and < 1 ppm Ni (after drying in the lab). The HPLC
purity of this intermediate crude 9 was 98.8 % (area % HPLC).
Acetic acid (1634 kg) was placed in a reactor together with the above crude product (209 kg).
Then, Smopex-234 (17.5 kg) was added. The mixture was stirred at RT for 2 h; charcoal (17.5 kg)
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 40
1
2
3
4
5
6
7
8
9
was added, followed by AcOH (204 kg). The mixture was heated to 90 °C within 1.5 h and stirred
at that temperature for 1 h. Then, charcoal and Smopex-234 were filtered off and washed with
AcOH (369 kg). To the filtrate, Smopex-234 (17.5 kg) followed by charcoal (17.5 kg) and AcOH
(89 kg) were added. The mixture was stirred for 1 h at 90 °C. Then, Smopex-234 and charcoal
were removed by filtration. The filter cake was washed with AcOH (553 kg). AcOH was distilled
off under vacuum at 70 °C to a residual volume of approximately 950 l in the vessel (measured by
radar). Then, demineralized H2O (2,888 kg) was added at 80 – 90 °C. The mixture was seeded (0.5
kg of pure 9). Stirring was continued for 0.5 h; the mixture was cooled within 7 – 10 h to 0 °C and
stirred at that temperature for 4 h; then, the product pure 9 was filtered and washed with a mixture
of demineralized H2O (368 kg) and AcOH (128 kg) followed by demineralized H2O (2,943 kg).
The wet product (253 kg) was dried in a paddle dryer at 80 °C under full vacuum (< 10 mbar) to
obtain 152.2 kg purified product pure 9 (88 % yield; based on intermediate 7). The product
contained < 1 ppm Pd and < 1 ppm Ni. HPLC purity 99.8 % area %. Assay AcOH < 1 % m/m.
1H NMR (400 MHz, acetic acid-d4) δ 9.14 (s, 1H), 8.82 (d, J = 2.3 Hz, 1H), 8.39 (d, J = 9.0 Hz,
1H), 8.23 (d, J = 2.0 Hz, 1H), 8.14 – 8.04 (m, 2H), 7.93 – 7.97 (m, 3H), 7.89 – 7.82 (m, 2H), 7.79
(ddd, J = 8.4, 7.0, 1.3 Hz, 1H), 7.72 – 7.62 (m, 1H), 7.36 (d, J = 1.8 Hz, 1H), 3.79 (s, 3H), 1.81 (s,
6H).
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
13C NMR (101 MHz, acetic acid-d4) δ 154.12, 147.17, 144.83, 144.78, 139.98, 135.59, 135.40,
133.84, 132.65, 131.96, 131.19, 130.04, 129.38, 128.54, 128.39, 128.03, 127.77, 127.17, 126.62,
126.40, 123.74, 123.71, 118.99, 114.57, 37.32, 28.31, 27.81.
Anal. calcd for C30H23N5O: C, 76.74; H, 4.94; N, 14.92. Found: C, 76.34; H, 4.92; N, 14.82.
HR-LC MS: m/z [M + H]+ calcd for C30H24N5O: 470.1981; found 470.2009.
ACS Paragon Plus Environment

Page 29 of 40
Organic Process Research & Development
1
2
3
4
Analytical data of compound 11:
5
6
7
8
1H NMR (400 MHz, acetic acid-d4) δ 9.32 (s, 1H), 8.42 (d, J = 8.7 Hz, 1H), 7.96 – 7.89 (m, 2H),
7.86 (ddd, J = 8.5, 7.0, 1.1 Hz, 1H), 7.81 – 7.71 (m, 2H), 7.57 – 7.48 (m, 1H), 7.36 (d, J = 8.4 Hz,
1H), 3.78 (s, 3H), 1.89 (s, 6H).
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
13C NMR (101 MHz, acetic acid-d4) δ 154.34, 144.62, 138.79, 134.50, 133.73, 130.77, 129.12,
128.27, 127.88, 127.16, 123.94, 123.73, 123.45, 121.09, 114.24, 37.16, 28.13, 27.86.
HR-LC MS: m/z [M + H]+ calcd for C21H19N4O: 343.15534; found: 343.15552.
ASSOCIATED CONTENT
NMR data of intermediates of the synthesis are provided.
The following files are available free of charge.
NMR data of all intermediates and listed impurities (PDF)
AUTHOR INFORMATION
Corresponding Author
Markus Baenziger; Novartis Pharmaceuticals, GDD/TRD/Chemical & Analytical Development
(CHAD), 4056 Basel, Switzerland
markus.baenziger@novartis.com
Present Addresses
ACS Paragon Plus Environment