Process development and scale-up for the preparation of the 1-methyl-quinazoline-2,4-dione wnt inhibitor SEN461

Article abstract of DOI:10.1021/op400145w

A practical and scalable route to the Wnt inhibitor SEN461 1 is described herein. The optimized route consists of nine chemical steps. The intermediates are solids and were isolated by filtrations. Critical reactions steps in the medicinal chemistry route were modified for an initial scale-up process, and as a result, we developed a synthetic procedure for the preparation of multihundred gram quantities of the final product. A further process development for the phase 1 clinical batch campaign is reported.

Full text of DOI:10.1021/op400145w

Article
pubs.acs.org/OPRD
Process Development and Scale-Up for the Preparation of the
1‑Methyl-quinazoline-2,4-dione Wnt Inhibitor SEN461
Matteo Betti,*,† Eva Genesio,‡ Alessandro Panico,§ Salvatore Sanna Coccone,§ and Paul Wiedenau⊥
‡
†Process Chemistry Unit and Compound Management & Analysis Unit, Siena Biotech SpA, 53100 Siena, Italy
S
* Supporting Information
ABSTRACT: A practical and scalable route to the Wnt inhibitor SEN461 1 is described herein. The optimized route consists of
nine chemical steps. The intermediates are solids and were isolated by filtrations. Critical reactions steps in the medicinal
chemistry route were modified for an initial scale-up process, and as a result, we developed a synthetic procedure for the
preparation of multihundred gram quantities of the final product. A further process development for the phase 1 clinical batch
campaign is reported.
INTRODUCTION
RESULTS AND DISCUSSIONS
■
■
MedChem Synthesis. The first synthesis of SEN461
The Wnt signaling pathway is known to have a fundamental
role in tissue development, differentiation and homeostasis; this
is achieved by regulating a specific pool of genes that strictly
control temporal and spatial regulation of cell growth,
movement and cell survival. Chronic activation of the canonical
Wnt pathway promotes uncontrolled cell growth and survival,
and can consequently drive cancer formation in a range of
tissues.1−3 In the past decade a number of studies demonstrated
that up-regulation of the Wnt pathway is present in various
types of cancer (e.g., lung, breast, pancreatic, gastric, colorectal
cancers, medulloblastoma, glioblastoma, hepatocellular carcino-
ma).4 For this reason targeting the canonical Wnt pathway can
be a promising strategy in the identification and development
of new anticancer agents. To date at least three Wnt-mediated
intracellular signaling pathways have been identified:5,6 the
calcium-mediated and planar polarity pathways8 and the
canonical Wnt-β-catenin pathway. Siena Biotech has set up a
drug discovery research program to identify new inhibitors of
the β-catenin-dependent Wnt pathway for the treatment of
glioblastoma multiforme.7 After a screening campaign, a
primary hit series was discovered; medicinal chemistry
modifications and optimization of the original structures led
to the identification of an advanced lead compound, SEN461
shown in Figure 1.
(Scheme 1), carried out in our Medicinal Chemistry
laboratories, delivered gram quantities of SEN461 to support
in vitro and preliminary in vivo testing.
The synthesis, although long with nine steps, is convergent,
but it had some important major drawbacks which necessitated
its redesign: (a) the anthranilic acid 2 starting material is quite
expensive; (b) the esterification of 4 with sulfuric acid and
subsequent free basing gave 5 contaminated with 5−10% of its
dimer a (Figure 2), which was difficult to separate in this stage
and in the following stages; (c) the acylpiperazine 12 proved to
be unstable, giving piperazine hydrochloride as a decom-
position product and possibly contributing to the poor yield of
the coupling step to give amide 13; (d) the acid 8 and its
corresponding lithium salt have extremely poor solubilities and
consequently, were difficult to handle; (e) N-methylation in the
last step was relatively low-yielding and required chromato-
graphic purification.
Redesign of the Synthesis. The alternative sequence
shown in Scheme 2 was designed, preserving the ring-closure
step to intermediate 7 but starting from the significantly less
expensive nitrobenzoic acid 14.9 Although the sequence
features the same number of steps and is more linear, it
overcomes a number of problems of the original route, namely
the use of the toxic and corrosive triphosgene and the use of the
unstable acylpiperazine 12. Introducing the N-Me group early
on (7 to 16) in the synthesis avoids the relatively low-yielding
final step of the original procedure, as well as renders the
intermediates, especially the acid 17, more soluble and more
easy to handle.
Initial Scale-Up: 50 and 500 g Campaigns. Two
campaigns were carried out to synthesise SEN461 batches for
preclinical studies. The first campaign furnished 50 g of the
desired compound and included some route scouting and
reaction optimization work, which is described in detail in the
corresponding section. For the second campaign, which
Received: June 3, 2013
Published: July 23, 2013
Figure 1. Chemical structure of the Wnt inhibitor SEN461.
© 2013 American Chemical Society
1042
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
Scheme 1. MedChem Route to compound 1
to complete conversion to imidazolide, however the reaction
with the amino ester 5 was very slow and gave only 30%
conversion after 5 days in DCM at different temperatures (Tint
= 20 or 40 °C). Activation with thionyl chloride or oxalyl
chloride followed by addition of 5 and triethylamine afforded
the desired amide 15 in 18 h and higher yields (86−90%). For
the scale-up oxalyl chloride was chosen as activating agent
because the reaction profile was slightly cleaner. In contrast to
the 50 g campaign, the acid chloride solution was not
completely evaporated in the 500 g campaign, but was partially
concentrated and then added to the suspension of the
aminoester 5 hydrochloride salt and the triethylamine. The
desired product 15 was recovered in 77% yield (instead of 87%
as in the former campaign) and 98% HPLC purity after
crystallization from iPrOAc.
Stage 3. The method of choice for reducing the aromatic
nitro group of 15 is catalytic hydrogenation as there are no
incompatible functional groups present. As the hydrogenation
with H2 gas was not yet available at the time of initial scale-up,
we chose to examine transfer hydrogenation conditions
(anhydrous Pd−C 10 wt %, NH4COOH, EtOH as solvent).
In order to obtain a preliminary evaluation of the process
thermal profile, two reaction mixtures (respectively with 2.5
and 1.5 wt % of Pd−C) were warmed from 20 to 60 °C (Tint)
in 30 min. In the first experiment gas evolution (CO2 and NH3)
started from 25 °C internal temperature, while in the second
the onset was around 35 °C, possibly due to the lower catalyst
loading. In any case, the reaction mixtures were warmed until
Figure 2. Dimeric impurity a monitored in the MedChem synthesis of
5.
furnished 500 g of final product as well as several 100s of grams
of various intermediates, the synthetic route remained the same.
Most of the reactions were carried out in a 10 L jacketed glass
reactor, and some adjustments to reduce solvent volumes and
to facilitate working with the reactor were made.
Stage 1. Studies directed toward the multihundred gram
synthesis of 5 began by exploring the feasibility of employing
thionyl chloride in methanol as solvent. During the 50 g
campaign, the aminoester 5 hydrochloride salt was prepared in
high yield (>98%) by adding 2.0 equiv of SOCl2 to a
suspension of amino acid 4 in MeOH (10 vol), followed by
evaporation of the solvent. In the 500 g campaign, the reaction
was run in the same solvent but more concentrated than
previously (3 vol), and the product was isolated by
precipitation with MTBE rather than evaporation to dryness.
The reaction was carried out in two main batches in 5 L round-
bottom flasks affording 5 in 98% yield.
Stage 2. Next, the coupling of acid 14 with amino ester 5
was examined.10 Even if the activation of acid 14 with CDI led
1043
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
Scheme 2. Redesigned synthesis of compound 1
no gas evolved anymore. In the workup we included an
aqueous wash to completely remove residual NH4COOH: it
allowed, during the 50 g campaign, to isolate a product with a
superior NMR purity (95% instead of 90%) in 94% yield. In the
500 g campaign the transfer hydrogenation was replaced with
hydrogenation using H2 gas, thus eliminating the need to
remove residual salts by aqueous washes. Having only a 2 L
hydrogenation jacketed reactor available, the reaction had to be
performed in several batches. Although the starting material
was poorly soluble in EtOH, the reaction proceeded rapidly
within 1 h to the product 6 which was highly soluble in EtOH.
This reduction is exothermal,11 and with no external cooling
applied, a temperature increase from 20 to 55 °C over 30 min
was detected using a thermometric probe. The desired
intermediate 6 was isolated in 91% yield and 99% HPLC purity.
Stage 4. Both CDI as well as triphosgene can be used to
close the ring as demonstrated on similar systems in the
Medicinal Chemistry group (Scheme 1). The former was
favored for scale-up due to its low toxicity and cost. We
observed that the absence of ammonium formate in the starting
material 6 was crucial for a clean ring-closure step: among the
side products, two major impurities b and c (Figure 3) were
identified (3−5% a/a by LC−MS).
Figure 3. Major impurities b and c identified in the presence of
ammonium formate.
The reaction initiates already at 15 °C, as shown by the
product precipitating as a white solid and causing a moderate
exotherm from 15 to 25 °C after complete CDI addition. For
isolation on multihundred grams scale, the white precipitate is
simply filtered and dried under vacuum giving 7 in high purity
(HPLC > 95%) and 88% yield. For practicality (hygroscopicity
of CDI) as well as to better control the exothermic reaction, the
order of addition was changed on kilogram scale. A solution of
the amide 6 was added to the CDI suspension, rather than
adding the solid CDI to the amide solution. The desired 1-H-
quinazoline-2,4-dione 7 was isolated with a slightly improved
yield (92%).
Stage 5. Together with introducing the N-methyl group
early on in the synthesis, the hazardous combination of NaH/
DMF12 was replaced with K2CO3/DMF,13 affording the
1044
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
intermediate 16 in 94% yield and excellent purity (>99% by
HPLC). In the kg-scale run, another 0.2 equiv of MeI (1.5
equiv instead of 1.3 equiv) had to be added to ensure complete
conversion. In previous trials, we observed complete conversion
with 1.3 equiv of the alkylating agent using the same batches of
reactants. We hypothesize that some MeI might have been lost
when charging the MeI/DMF solution to the addition funnel of
the pilot plant under reduced pressure.
Table 1. Optimisation of the TFA concentration and process
campaign experiments: stage 8
conversion/reaction
time at 50 °C (Tbath)
yield
(%)
entry
TFA
a
1
0.7 M
1.7 M
0.3%/1.5 h (20 °C)
1.5%/1 h (50 °C)
56%/1 h
−
a
2
−
72%/2 h
Stage 6. The hydrolysis of methyl ester 16 was performed by
addition of lithium hydroxide using a mixture THF/H2O as
solvent and heating at 60 °C (Tint). The rate of conversion of
the starting material was monitored by HPLC, and reaction was
completed in 2 h. In order to obtain the product as free acid the
organic solvent was evaporated, and the acid 17 was
precipitated by addition of HCl 1 N. Following this workup
procedure, the filtration of the product was very slow. Instead
acidifying the reaction mixture at 45 °C (rather than room
temperature) and adding 1 volume of acetone gave a precipitate
that was much easier to filter. The residual water could not be
completely removed (KF: 6−8% w/w) from the product by
drying in a vacuum oven.
Stage 7. CDI and oxalyl chloride were investigated as
activating agents for the coupling with the Boc-piperazine. The
activation with oxalyl chloride was quite slow and did not go to
completion in 18 h. CDI furnished complete activation of the
acid in 1 h (the presence of residual water in acid 17 required
an overcharge of the coupling agent to achieve full conversion
to intermediate imidazolide), while the subsequent coupling
with Boc-piperazine was slightly slower and was left to react
overnight for convenience. When the reaction mixture was
cooled to room temperature during the 50 g campaign, a
precipitate started to form and was filtered. The solid obtained
was free of imidazole and Boc-piperazine, but contained 5% (by
HPLC) of the acid 17, which was removed by basic aqueous
washes. In the 500 g campaign, with no residual acid left, the
basic washes could be eliminated and the workup was
significantly simplified.
Stage 8. The Boc deprotection reaction was initially
performed using HCl 4 N in dioxane14 in analogy to the
reaction in Scheme 1. This procedure had some issues: (a) the
reaction rate was very slow, requiring further additions of HCl
as well as heating; (b) the suspension that formed after some
hours was difficult to stir. We then turned our attention to
investigating the use of TFA for the Boc-deprotection.15 Some
trials were done by varying the percentage of TFA in DCM
(Table 1, entries 1−4): the best conditions ensuring a complete
and rapid deprotection were found when TFA 3.2 M in DCM
at 50 °C (Tbath) was used. On multihundred-gram scale (Table
1, entry 5) it was possible to reduce the concentration of TFA
by ca. 25% with respect to earlier work, maintaining a fast and
complete reaction.
Stage 9. The final step of the intermediate 19 with 2-
methoxy acetyl chloride was carried out in DCM using triethyl
amine as base. The reaction was fast, showing complete
conversion (by HPLC) already after 30 min. The workup was
done by simply washing the reaction mixture with H2O and
NaHCO3 solution, and the final product 1 was purified by
crystallisation from EtOAc, obtaining SEN461 in high yield
(90%) and purity (HPLC purity >99%).
99%/18 h
a
3
2.1 M
95%/1 h
−
98%/1.5 h
99% 18 h
b
4
3.2 M (50 g campaign)
2.4 M (500 g campaign)
100%/1 h
98
80
c
5
100%/3 h
a
b
Scale of experiment: 2 g of 18. Scale of experiment: 100 g of 18.
c
Scale of experiment: 0.7 kg of 18
(1) the process employed thionyl chloride, a hazardous
reagent;
(2) the presence of impurity d (Figure 6) was monitored in
the amide 15 formation;
(3) the N-methylation appeared not to be reproducible, even
using an excess of methyl iodide;
(4) stage 6 displayed complications in the purification and
isolation of 17 as free acid;
(5) the contamination by inorganics and water of acid 17
severely lowered the yield of the Boc-piperazine 18
formation step;
(6) the use of TFA in the Boc-deprotection stage raised an
issue with plant pipes compatibility.
Thionyl Chloride: Occasion of Concern. The excess of
thionyl chloride employed in the 500 g campaign was greater
than effectively required. The use of this reactant is always
occasion of concern especially in pilot-plant campaigns: its
intrinsic acidity leads to corrosion of pilot-plant pipes.
Moreover, the addition of this reagent is associated to a strong
exothermal effect, evolution of toxic gases such as sulfur dioxide
and potentially, in the case of SOCl2/MeOH system, methyl
chloride. It was observed that the reaction mass tends to get
very dense and even up to a jelly consistency at about half of
the addition of thionyl chloride, compromising the mixing
efficiency, and consequently, the heat transfer capacity. Dilution
with 4 vol of solvent together with a minor charge of thionyl
chloride (1.3 equiv vs 2 equiv) solved the problem, affording in
the manufacturing campaign the product with 97% yield and
desired purity.
A calorimetric measurement of the reaction was performed17
(Figures 4 and 5) and confirmed the following: release of heat
and gas was controlled by the rate of addition of thionyl
chloride; the peak of heat evolution was observed during the
first 25% of thionyl chloride charge; average gas evolution (3L/
min·kg) was constant throughout the addition. Consequently,
during the manufacturing campaign on 8.9 kg of acid 4 in
MeOH (4 vol), SOCl2 (5 L) was charged in 50 min,
maintaining the temperature below 35 °C.
Stage 2: An Oxalyl Chloride Byproduct. Changes versus the
initial scale-up procedure concerned mainly the adjustment of
equivalents of nitrobenzoic acid 14 in order to minimize the
formation of a new impurity d identified by NMR and HPLC−
MS analyses (Figure 6) throughout the process development
optimisation.
Process Development. Although the aforementioned
enabling methodology yielded excellent quality product in
very good yield, the following issues were identified by the
process development investigation:16
1045
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
amount of K2CO3 was increased from 1.3 to 2.0 equiv, and
their suspension in DMF was stirred 1 h at 50 °C before the
addition of MeI as a solution in the same solvent (after cooling
to 30 °C). The procedure was successfully applied to the demo
batch experiment, although a high inorganic content (up to 19
wt %, potassium salt byproducts) was measured in the product
isolated by precipitation from DMF/water. To overcome the
issue in the plant campaign, it was decided to increase from 5 to
10 vol the amount of water added during the precipitation,
obtaining the desired intermediate 16 with low inorganics
content (<0.5 wt %) in 95% yield.
Improving Isolation and Purification of Acid 17.
Considerable issues following the initial workup procedure
resulted from the very slow filtration of product 17 since the
collected solid had a particle size down to 1 μm. Some
difficulties were also encountered in removing water (KF: 4 wt
%) and inorganics (sulfated ashes: 18 wt %). Fortunately, the
acid 17 particle size was significantly improved if precipitation
occurred from THF/water, avoiding the solvent distillation.
Moreover, acetone washes (instead of water) on filter allowed
an easier drying in the oven (KF: 0.5−1.0 wt %).19 The
manufacturing batch of acid 17 was isolated with 96% yield,
99% HPLC purity, and <0.1 wt % inorganics content.
Stage 7. The coupling reaction failed to reach completion
when the starting material 17 was contaminated by inorganics
and water (70% final conversion), even if a clean product 18
was recovered after filtration (0.1% a/a of residual acid 17 was
observed by HPLC) in 67% yield. Employing the inorganics-
free batch of starting acid, the reaction went as expected.
Moreover, thanks to the low water content, the amount of
charged CDI could be lowered from 2.0 to 1.5 equiv,
minimizing the homocoupling of N-Boc-piperazine and
affording intermediate 18 in 91% yield, high purity (HPLC
99%) and suitable characteristics to be chosen as starting
material for the GMP stages (stages 8 and 9) of the
manufacturing campaign.
Figure 4. Specific heat of reaction (green line) as function of SOCl2
addition (violet line).
Figure 5. Gas evolution (blue line) as function of SOCl2 addition
(black line).
TFA and Plant Pipes Compatibility. A procedure with an
excess of TFA (almost 10 equiv) was deemed problematic
because of serious compatibility problems with the plant
apparatus and a complex workup procedure. Deprotection with
HCl 5 N in IPA and the subsequent hydrochloride salt
precipitation was favored on the basis of previous experience
with analogous compounds. DCM was selected as cosolvent in
order to ensure an efficient stirring during the salt formation.
Small-scale trials and demo batch production were conducted
without problems. However, during the cGMP campaign,
filtration rate of 19 as hydrochloride salt was unexpectedly slow
(see Plant Manufacturing Campaign section). A further lab
experiment demonstrated that the filtration rate was associated
with the unfavorable ratio between wet cake height and filter
diameter.
Figure 6. Impurity d.
When an excess of oxalyl chloride (1.3 equiv) was used to
activate the starting acid, 8% of d was detected by HPLC.
Instead, using the same excess of 14 (1.2−1.3 equiv) ensured a
complete consumption of oxalyl chloride without the impurity
generation.18 Moreover, the residual nitrobenzoic acid 14 could
be eliminated by crystallization with i-PrOAc. However,
employing the latter conditions, we experienced difficulties in
the demo batch experiment: the incomplete conversion (90%)
of the acid 14 was observed, also using 1.5 equiv of oxalyl
chloride. The cause was identified in the unexpected high water
content (5 wt % by KF analysis) of the batch of nitrobenzoic
acid supplied for the plant campaign. Due to the urgency to
deliver final product 1 and since the demo batch experiment
produced the required ester 5 with the expected yield (89%)
and purity (99% by HPLC), the excess of oxalyl chloride was
used also in the manufacturing campaign without a further
investigation.
Plant Manufacturing Campaign. The reported synthetic
route (Scheme 3) was followed to synthesize SEN461 in a
nine-stage procedure.
Non-GMP Stages: Pilot-Plant Scale-Up. Stages 1−7 were
carried out in a single run using 100−200 L Hastelloy vessels.
Isolation of solid intermediates was performed by pressure/
vacuum filtration using a 60 cm Hastelloy filter. Drying was
carried out in tray dryer oven. No relevant deviations from the
planned procedure were observed, and intermediate 18 was
isolated in 65% overall yield starting from nitrobenzoic acid 14.
GMP Stages: Kilolab Scale-Up. Stage 8 was performed in
multiple batches, employing a 20 L glass reactor and a
A Not Reproducible N-Methylation. After a few familiar-
ization trials, the methylation of 7 appeared not to be
reproducible. An incomplete conversion was monitored (87−
93% a/a by HPLC) even using an excess of MeI (from 1.3 to
1.7 equiv). To drive the conversion of starting material 7, the
1046
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
Scheme 3. Pilot-plant campaign synthetic route
Hastelloy filter dryer. During the first run of stage 8 a very slow
filtration rate was observed. Trying to mitigate the impact of
the slow filtration, in view of the following runs, the input
weight was decreased, and the solvent composition was slightly
modified to decrease solvent viscosity.20 The first batch of
SEN461 1 (1.798 kg) was produced with 42% overall yield
starting from acid 14; the second batch (3.581 kg) was
produced with 38% overall yield in the GMP synthesis.
contribution of dichloromethane used in the last step of the
process.21
CONCLUSIONS
■
Initial scale-up and process optimization to the Wnt inhibitor
SEN461 1 is described herein. The optimized route consists of
nine chemical steps. The current nine-step linear route is
Isolation and Identification of SEN461 Process Impurity.
An unexpected impurity was detected at low level (0.2% a/a by
HPLC analysis) in the manufacturing API batch of SEN461. In
a preliminary investigation by LC−MS the APCI mass
spectrum contained m/z 913.6 as main peak. Analyzing the
MS spectra of the impurity and by a comparison with the
fragmentation pattern of SEN461, it seemed that the two
molecules were structurally related. The unknown compound
was isolated by preparative LC, and an NMR study was
performed with the aim to identify its molecular structure. The
inspection of mono- and bidimensional NMR spectra allowed
excluding the high-molecular weight impurity as some type of
polymeric contamination of the batch. Thanks to the
comparison with the proton spectrum of 1, the structure of
the impurity was determined to be a sort of dimer of SEN461
(compound e, Figure 7), probably formed with the
Figure 7. Elucidated structure of the SEN461 process impurity e.
1047
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
characterized by high yields (overall yield in the 50 g campaign:
44%; in the 500 g campaign: 38%); solid intermediates
throughout the synthesis, easy purification of intermediates and
final product and absence of chromatography; no highly
hazardous reagents; robust reactions with low thermal hazards;
no specialized equipment required and temperature range
reactions between 0 and 80 °C; availability of starting material
14 from various suppliers. Critical reaction steps in the
medicinal chemistry route were modified for an initial scale-
up process, and as a result, a synthetic procedure for
preparation of multihundred gram quantities of the final
product was developed. A further process development and the
clinical batch manufacturing campaign are also described: a
linear and efficient access to the key GMP-intermediate 18 was
developed, significantly simplifying the synthesis, reducing cost,
and maintaining high yields (overall yield in plant campaign:
42%).
HPLC: tR= n.d., not UV responsive intermediate.
1H NMR (400 MHz DMSO-d6): δ 7.83 (s, broad, 3H), 3.57
(s, 3H), 2.61 (d, J = 6.9 Hz, 2H), 2.24 (m, 1H), 1.90 (m, 2H),
1.79 (m, 2H), 1.50 (m, 1H), 1.27 (m, 2H), 0.97 (m, 2H).
13C NMR (100 MHz DMSO-d6): δ 175.9, 52.2, 44.7, 42.6,
35.5, 29.4, 28.5.
Synthesis of 4-[(5-Methoxy-2-nitro-benzoylamino)-
methyl]-cyclohexanecarboxylic Acid Methyl Ester (15).
5-Methoxy-2-nitro-benzoic acid 14 (0.4 kg, 2.0 mol, 1 equiv)
was charged to a 5 L four-necked round-bottom flask under a
blanket of nitrogen, with 1 N aq NaOH trap for scrubbing the
off-gases. Dichloromethane (3.2 L) and DMF (4.3 mL) were
respectively added, cooling at 15 °C internal temperature with
an ice bath. The mixture was stirred for 10 min.
Oxalyl chloride (186 mL, 2.2 mol, 1.1 equiv) was added
dropwise over 45 min. An endotherm was observed during the
addition (ΔT = −5 °C, from 15 to 10 °C). Gas evolution was
completed in 3.5 h, and then an HPLC check confirmed the
complete activation.
Part of the dichloromethane was removed by distillation
under reduced pressure (1.0 L at p = 200 mbar, Tint = 30 °C).
The resulting mixture was kept standing at room temperature
under nitrogen atmosphere for 18 h and then was added
dropwise in 2 h to a suspension of the 4-aminomethyl-
cyclohexanecarboxylic acid methyl ester 5 (414 g, 2.0 mol, 1
equiv) and TEA (830 mL, 6.0 mol, 3 equiv) in dichloro-
methane (1.6 L) at 10 °C. An exotherm was observed during
the addition (ΔT = +10 °C, from 10 to 20 °C). A sample was
checked by HPLC 15 min after the addition and a complete
conversion was observed.
EXPERIMENTAL SECTION
■
The reported yields are corrected for purity and water/solvent
content of the products. Generally, the reactions were
monitored by HPLC, and purities/conversions quoted refer
to HPLC area % at 215 nm. HPLC method: Waters Symmetry
C18 3.5 μm 4.6 mm × 75 mm column, flow rate 1.0 mL/min,
mobile phase A: 10 mM aq K2HPO4 buffer or 0.1% aq formic
acid; mobile phase B: acetonitrile or acetonitrile/0.1% formic
acid; gradient (13 min): 95:5 A/B to 10:90 A/B over 10 min, 2
min at 10:90 A/B, then 10:90 A/B to 95:5 A/B over 1 min.
UPLC-MS analyses were run using a Acquity Waters UPLC
equipped with a Waters SQD (ES ionization) and Waters
Acquity PDA detector. UPLC method: BEH C18 1,7 μm, 2,1 ×
50 mm column, flow rate 0.6 mL/min, mobile phase A: 0.1% aq
formic acid; mobile phase B: 0.1% formic acid acetonitrile;
gradient (3 min): 95:5 A/B to 100:0 A/B over 2 min, 40 s at
100:0 A/B, 100:0 A/B to 95:5 A/B over 10 s, then 10 s at 95:5
A/B. Retention times were expressed in minutes. Temperature:
40 °C. UV Detection at 215 and 254 nm. ESI+ detection in the
80−1000 m/z range.
The reaction mixture was combined with the batch of the
second run (0.4 kg scale) in a 10 L jacketed reactor and washed
with: water (2 × 3.0 L); HCl 2 M (1 × 3.0 L); water (1 × 3.0
i
L). PrOAc (6.3 L) was added dropwise in 15 min to the
separated organic phase, and the formation of a pale-brown
precipitate was observed. The suspension was heated to Tint
=
75 °C for 40 min and then cooled to Tint = 0 °C for 1h. After
filtration on a Buchner funnel and drying until constant weight
in oven (p = 20 mbar, Toven = 20 °C, 8 h), 1.085 kg of 4-[(5-
methoxy-2-nitro-benzoylamino)-methyl]-cyclohexanecarboxylic
acid methyl ester 15 as an off-white solid was recovered. Yield:
77%.
Synthesis of 4-Aminomethyl-cyclohexanecarboxylic
Acid Methyl Ester (5). 4-Aminomethyl-cyclohexanecarboxylic
acid 4 (0.6 kg, 3.8 mol, 1 equiv) was charged to a 5 L four-
necked round-bottom flask under a blanket of nitrogen, with 1
N aq NaOH trap for scrubbing the off-gases. Methanol (2.0 L)
was added, cooling at 10 °C internal temperature with an ice
bath. The mixture was stirred for 10 min.
Thionyl chloride (559 mL, 7.6 mol, 2 equiv) was added
dropwise over 2.5 h, maintaining the temperature between 10
and 20 °C. After complete addition, the white suspension was
warmed to 25 °C in 1 h, and HPLC indicated complete
conversion.
Then the methanol was partially removed by distillation
under reduced pressure (1.1 L at p = 320 mbar, Tint = 42 °C).
The resulting mixture was cooled to 30 °C, and MTBE (2.0 L)
was added over 15 min, giving the precipitation of a white solid.
The suspension was cooled to −5 °C under mechanical stirring
for 1.2 h and then filtered on a Buchner funnel, followed by
washing with MTBE (500 mL).
UPLC−MS: tR = 1.23 min, m/z = 351 [M + 1]+.
HRMS calcd for C17H23N2O6 [M + 1]+ 351.15565, found
351.15566.
HPLC: tR = 6.48 min, purity 98.7%.
1H NMR (400 MHz CDCl3): δ 8.06 (d, J = 9.1 Hz, 1H),
6.91 (dd, J = 9.1 Hz, 2.8 Hz, 1H), 6.84 (d, J = 2.8 Hz, 1H), 5.73
(m, 1H), 3.84 (s, 3H), 3.60 (s, 3H), 3.26 (t, J = 6.5 Hz, 2H),
2.20 (m, 1H), 1.96 (m, 2H), 1.86 (m, 2H), 1.56 (m, 1H), 1.48
(m, 2H), 0.99 (m, 2H).
13C NMR (100 MHz DMSO-dd): δ 176.2, 166.2, 163.7,
140.0, 136.6, 127.5, 115.3, 114.9, 57.1, 52.0, 45.9, 43.0, 37.3,
30.0, 29.0.
Synthesis of 4-[(2-Amino-5-methoxy-benzoylamino)-
methyl]-cyclohexanecarboxylic Acid Methyl Ester (6). A
2 L glass pressure reaction vessel (Buchi Ecoclave) was charged
at 20 °C under a nitrogen blanket with a suspension of Pd/C
10 wt % (anhydrous, 5.0 g) in ethanol (10 mL) followed by a
suspension of 4-[(5-Methoxy-2-nitro-benzoylamino)-methyl]-
cyclohexanecarboxylic acid methyl ester 15 (200 g, 0.57 mol, 1
equiv) in ethanol (1.0 L). The resulting mixture was vigorously
stirred under hydrogen pressure (initial H2 pressure: 75 psi)
The final product 5 hydrochloride salt was dried in the
drying oven (p = 20 mbar, T = 20 °C, 16 h) to give 772 g of a
white solid. Yield: 98%.
UPLC−MS: tR= 0.53 min, m/z = 172 [M + 1]+.
HRMS calcd for C9H18NO2 [M + 1]+ 172.13376, found
172.13378.
1048
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
until complete reduction of the starting nitro derivative 15 was
observed by HPLC (theoretical absorption of H2 achieved after
1 h). The reaction mixture was then cooled to 20 °C under
nitrogen atmosphere, combined with the other 5 identical
reaction batches and filtered on a Buchner funnel through a pad
of powered cellulose to remove the catalyst, washing with
ethanol (2 × 250 mL). Mother liquors were transferred to the
10 L jacketed reactor and concentrated under reduced pressure
(from 9 to 4.5 L; p = 200 mbar, Tint = 45 °C); after cooling to 5
°C, water (4.5 L) was added dropwise in 25 min. The resulting
suspension was stirred at 8−10 °C for 1 h and filtered on a
Buchner funnel. The solid obtained was washed with EtOH/
H2O 1:1 (3 × 500 mL) and dried in a vacuum oven (Toven = 50
°C, p = 10 mbar, 18 h). Finally 1.01 kg of 6 as a pale-orange
solid was obtained. Yield: 91%.
over 15 min. A mild exotherm was observed during the addition
(ΔT = +8 °C; from 20 to 28 °C). After 3 h another portion of
MeI was added (35 mL, 0.6 mol, 0.2 equiv), driving the
reaction to completion (by HPLC) after overnight stirring at
room temperature.
The suspension was cooled to +10 °C, and water (5.0 L) was
added over 40 min. The mixture was stirred at this temperature
for 30 min and then filtered on a Buchner funnel, followed by
washing with water (3 × 1.0 L). The product was dried in an
oven (Toven = 50 °C, p = 20 mbar) until constant weight,
affording 903 g of 16 as a white solid. Yield: 94%.
UPLC−MS: tR = 1.46 min, m/z = 361 [M + 1]+. HPLC: tR =
7.54 min, purity 99.7%.
HRMS calcd for C19H25N2O5 [M + 1]+ 361.17368, found
361.17368.
UPLC−MS: tR= 1.01 min, m/z = 321 [M + 1]+.
1H NMR (400 MHz DMSO-d6): δ 7.45 (d, J = 2.6 Hz, 1H),
7.36 (m, 2H), 3.79 (m, 5H), 3.54 (s, 3H), 3.48 (s, 3H), 2.23
(m, 1H), 1.85 (m, 2H), 1.71−1.61 (m, 3H), 1.20 (m, 2H), 1.03
(m, 2H).
HRMS calcd for C17H25N2O4 [M + 1]+ 321.18146, found
321.18144.
HPLC: tR = 4.44 min, purity 99.3%.
1H NMR (400 MHz DMSO-d6): δ 8.20 (t, J = 5.4 Hz, 1H),
7.03 (d, J = 2.8 Hz, 1H), 6.80 (dd, J = 8.8 Hz, 2.8 Hz, 1H), 6.62
(d, J = 8.8 Hz, 1H), 5.89 (s, 2H), 3.66 (s, 3H), 3.55 (s, 3H),
3.03 (t, J = 6.2 Hz, 2H), 2.23 (m, 1H), 1.89 (m, 2H), 1.75 (m,
2H), 1.47 (m, 1H), 1.26 (m, 2H), 0.96 (m, 2H).
13C NMR (100 MHz DMSO-d6): δ 176.0, 161.7, 155.4,
150.8, 135.2, 124.0, 117.0, 116.0, 109.0, 56.2, 51.9, 47.2, 42.8,
36.2, 31.3, 29.9, 28.8.
Synthesis of 4-(6-Methoxy-1-methyl-2,4-dioxo-quina-
zolin-3-ylmethyl)-cyclohexanecarboxylic Acid (17). 4-(6-
Methoxy-1-methyl-2,4-dioxo-quinazolin-3-ylmethyl)-cyclohexa-
necarboxylic acid methyl ester 16 (1.0 kg, 2.8 mol, 1 equiv) was
suspended in THF (4.0 L) in a 10 L jacketed reactor, and a
solution of LiOH (76 g, 3.2 mol, 1.1 equiv) in water (1.0 L)
was added at 20 °C over 15 min. An endotherm was observed
during the addition (ΔT = −2 °C; from 19 to 17 °C). The
mixture was heated to 60 °C (Tint) for 30 min, then another
portion of water (1.0 L) was added to enable more efficient
stirring. After 2.5 h the conversion was complete by HPLC.
The organic solvent was distilled under reduced pressure
(Tint = 45−40 °C, p = 700−500 mbar; 3.5 L recovered in 1.5
h); water (3.0 L) was added and the mixture acidified to pH = 1
with HCl 3 M (1.1 L). Then acetone (1.0 L) was added and the
suspension cooled to +5 °C for 30 min. The mixture was filter
on a Buchner funnel, washing with water (2 × 1.0 L). The wet
solid was dried in oven (p = 20 mbar; 2 d at Toven = 50 °C, 8 h
at Toven = 80 °C), giving 930 g of 17 as a white solid. Yield:
90%.
13C NMR (100 MHz DMSO-d6): δ 176.2, 169.2, 150.0,
144.3, 119.7, 118.2, 116.1, 112.9, 56.3, 52.0, 44.9, 43.0, 37.6,
30.2, 29.0.
Synthesis of 4-(6-Methoxy-2,4-dioxo-1H-quinazolin-
3-ylmethyl)-cyclohexanecarboxylic Acid Methyl Ester
(7). To a suspension of 1,1-carbonyldiimidazole (859 g, 5.3
mol, 1.5 equiv) in acetonitrile (2.5 L) in a 10 L jacketed reactor,
a solution of 4-[(2-amino-5-methoxy-benzoyl-amino)-methyl]-
cyclohexanecarboxylic acid methyl ester 6 (1.13 kg, 3.5 mol, 1
equiv) in acetonitrile (7.5 L) was added over 1 h at room
temperature and nitrogen atmosphere. An exotherm was
observed (ΔT = +16 °C; from 11 to 27 °C). The reaction
suspension was heated to 55 °C (Tint), and after 2 h the
conversion was completed by HPLC analysis.
The suspension was cooled and kept at +5 °C for 40 min and
then filtered on a Buchner funnel, washing with acetonitrile (1
× 1.0 L). The product was dried by suction on the filter to give
1.12 kg of 7 as a white solid. Yield: 92%.
UPLC−MS: ,tR= 1.28 min, m/z = 347 [M + 1]+.
UPLC−MS: tR = 1.19 min, m/z = 347 [M + 1]+.
HRMS calcd for C18H23N2O5 [M + 1]+ 347.16073, found
347.16074.
HPLC: tR= 6.93 min, purity 99.1%.
HRMS calcd for C18H23N2O5 [M + 1]+ 347.16073, found
347.16072.
HPLC: tR = 6.33 min, purity 99.3%.
1H NMR (400 MHz DMSO-d6): δ 7.34 (d, J = 2.5 Hz, 1H),
7.28 (dd, J = 8.8 Hz, 2.5 Hz, 1H), 7.11 (d, J = 8.8 Hz, 1H), 3.76
(m, 5H), 3.55 (s, 3H), 2.24 (m, 1H), 1.86 (m, 2H), 1.65 (m,
3H), 1.22 (m, 2H), 1.04 (m, 2H).
1H NMR (400 MHz DMSO-d6): δ 7.48 (d, J = 2.6, 1H), 7.38
(m, 2H), 3.81 (m, 5H), 3.48 (s, 3H), 2.10 (m, 1H), 1.85 (m,
2H), 1.64 (m, 3H), 1.18 (m, J = 12.5 Hz, 3.4 Hz, 2H), 1.02 (m,
J = 12.5 Hz, 3.4 Hz, 2H).
13C NMR (100 MHz DMSO-d6): δ 176.0, 162.7, 155.3,
150.7, 134.2, 124.6, 117.4, 114.8, 108.9, 56.2, 52.0, 46.1, 42.8,
36.1, 30.0, 28.8.
13C NMR (100 MHz DMSO-d6): δ 177.3, 161.7, 155.4,
150.8, 135.2, 124.0, 116.9, 115.9, 109.9, 56.2, 47.3, 43.0, 36.3,
31.3, 30.1, 28.9.
Synthesis of 4-(6-Methoxy-1-methyl-2,4-dioxo-quina-
zolin-3-ylmethyl)-cyclohexanecarboxylic Acid Methyl
Ester (16). 4-(6-Methoxy-2,4-dioxo-1H-quinazolin-3-ylmeth-
yl)-cyclohexanecarboxylic acid methyl ester 7 (925 g, 2.7 mol, 1
equiv), potassium carbonate (480 g, 3.5 mol, 1.3 equiv), and
DMF (1.0 L) were respectively charged to a 10 L jacketed
reactor at room temperature under nitrogen flux. The
suspension was stirred for 1 h at 20 °C. No exotherm and
no gas evolution were observed. Then a solution of methyl
iodide (200 mL, 3.2 mol, 1.2 equiv) in DMF (1.0 L) was added
KF: 6.8 wt % water.
Synthesis of 4-[4-(6-Methoxy-1-methyl-2,4-dioxo-qui-
nazolin-3-ylmethyl)-cyclohexanecarbonyl]-piperazine-
1-carboxylic Acid tert-Butyl Ester (18). 1,1-Carbonyldiimi-
dazole (723 g, 4.5 mol, 2.1 equiv) and acetonitrile (4.5 L) were
respectively charged to a 10 L jacketed reactor under nitrogen
flux, and the suspension was stirred at 20 °C for 15 min. 4-(6-
Methoxy-1-methyl-2,4-dioxo-quinazolin-3-ylmethyl)-cyclohexa-
necarboxylic acid 17 (750 g, 2.1 mol corrected for water
content, 1 equiv) was added portionwise (6 × 120 g + 1 × 30
1049
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
g) over 40 min. Gas evolution (CO2) and exotherm (ΔT= +10
°C; from 17 to 27 °C) were observed. The mixture was heated
to +50 °C (Tint) for 1 h and checked by HPLC (activation
complete). Then a solution of N-Boc-piperazine (524 g, 2.8
mol, 1.3 equiv) in acetonitrile (1 L) was added over 10 min. An
endotherm was observed during the addition (ΔT= −1 °C;
from 53 to 52 °C). The reaction was heated to reflux (Tint= 80
°C) for 2 h (99.5% a/a HPLC conversion) and at 45 °C (Tint)
overnight. The obtained white suspension was cooled to 10 °C
for 30 min and filtered on a Buchner funnel, followed by
washing with acetonitrile (500 mL).
13C NMR (100 MHz DMSO-d6): δ 173.7, 161.8, 155.4,
150.8, 135.3, 124.0, 117.0, 116.0, 109.9, 56.2, 47.4, 46.8, 46.2,
42.7, 36.4, 31.4, 30.1, 29.3.
Synthesis of 6-Methoxy-3-{4-[4-(2-methoxy-acetyl)-
piperazine-1-carbonyl]-cyclohexylmethyl}-1-methyl-
quinazoline-2,4-dione (1). To a solution of 6-Methoxy-1-
methyl-3-[4-(piperazine-1-carbonyl)-cyclohexyl-methyl]-quina-
zoline-2,4-dione 19 (430 g, 1.0 mol, 1 equiv) in dichloro-
methane (2.0 L) mechanically stirred in a 10 L jacketed reactor,
triethylamine (172 mL, 1.2 mol, 1.2 equiv) was added all at
once at 18 °C under nitrogen flux. A slight exotherm was
abserved after the addition (ΔT= +1 °C; from 18 to 19 °C).
Then a solution of 2-methoxyacetyl chloride (120 g, 1.1 mol,
1.1 equiv) in dichloromethane (0.8 L) was added over 15 min,
observing an exotherm with ΔT= +10 °C (from 18 to 28 °C).
The mixture was stirred at 20 °C for 1 h when HPLC indicated
complete conversion.
The product was dried on filter by suction until constant
weight to give 960 g of 18 as a white solid. Yield: 91%.
UPLC−MS: tR = 1.51 min, m/z = 515 [M + 1]+.
HRMS calcd for C27H39N2O4 [M + 1]+ 515.28701, found
515.28702.
HPLC: tR = 7.54 min, purity 99.8%.
1H NMR (400 MHz DMSO-d6): δ 7.46 (d, J = 2.6 Hz, 1H),
7.41−7.35 (m, 2H), 3.83 (m, 5H), 3.48 (s, 3H), 3.41 (m, 4H),
3.26 (m, 4H), 2.51 (m, 1H), 1.71−1.61 (m, 5H),1.49 (s, 9H),
1.28−1.05 (m, 4H).
The reaction mixture was washed with water (1 × 1.5 L),
aqueous NaHCO3 2.5 wt % (1 × 2.0 L) and water (2 × 1.5 L).
Part of the organic solvent (2.0 L) was distilled under reduced
pressure (Tint= 20 °C, Tjacket = 70 °C, p = 400 mbar), and
EtOAc (2.0 L) was added, resulting in a precipitation of a white
solid. The suspension was stirred at 0 °C for 18 h, filtered on a
Buchner funnel, and then washed with EtOAc (1 × 1.0 L). The
isolated product was dried in the vacuum oven (Toven= 25 °C, p
= 20 mbar) until constant weight: 460 g of SEN461 1 as an off-
white solid was obtained. Yield: 90%.
13C NMR (100 MHz DMSO-d6): δ 174.1, 161.8, 155.4,
154.4, 150.9, 135.5, 124.0, 117.0, 116.0, 109.9, 79.8, 56.2, 47.4,
45.2, 41.4, 40.6, 36.4, 31.4, 30.1, 29.2, 28.7.
Synthesis of 6-Methoxy-1-methyl-3-[4-(piperazine-1-
carbonyl)-cyclohexylmethyl]-quinazoline-2,4-dione
(19). 4-[4-(6-Methoxy-1-methyl-2,4-dioxo-quinazolin-3-yl-
methyl)-cyclohexanecarbonyl]-piperazine-1-carboxylic acid
tert-butyl ester 18 (700 g, 1.4 mol, 1 equiv) was dissolved in
dichlomethane (3.5 L) at 20 °C in a 10 L jacketed reactor. A
solution of TFA (911 mL, 11.8 mol, 8.4 equiv) in dichlo-
methane (0.7 L) was added over 10 min. An exotherm was
observed during the addition (ΔT= +3 °C; from 20 to 23 °C).
The solution was heated to 38 °C (Tint) for 2.5 h (99.8%
HPLC conversion) and then cooled to 5 °C under nitrogen flux
and stirred overnight at this temperature.
Dichloromethane (3.0 L) was distilled under reduced
pressure (p = 400−500 mbar, Tint = 20−30 °C, Tjacket = 45−
80 °C). The organic mixture was basified with aqueous NaOH
15 wt % until pH = 13, maintaining the temperature at 10−20
°C during the addition of the base. The organic layer was
washed with water (3 × 3.0 L). The organic solvent was
UPLC−MS: tR= 1.10 min, m/z = 415 [M + 1]+.
HRMS calcd for C25H35N4O6 [M + 1]+ 487.25569, found
487.25570.
HPLC: tR= 5.34 min, purity 99.7%.
1H NMR (400 MHz CDCl3): δ 7.66 (d, J = 3.0, 1H), 7.27
(dd, J = 9.1 Hz, 3.0 Hz, 1H), 7.15 (d, J = 9.1, 1H), 4.11 (s, 2H),
3.93 (d, J = 7.2 Hz, 2H), 3.89 (s, 3H), 3.60−3.42 (m, 11H),
3.41 (s, 3H), 2.43 (m, 1H), 1.92 (m, 1H), 1.82−1.76 (m, 4H),
1.53 (m, 2H), 1.17 (m, 2H).
13C NMR (100 MHz DMSO-d6): δ 174.2, 168.0, 161.8,
155.4, 150.8, 135.3, 124.0, 117.0, 116.0, 109.9, 71.2, 59.0, 56.3,
47.4, 45.3, 44.7, 42.2, 36.4, 31.4, 30.1, 29.2.
M.p. (DSC): 164 °C.
Water content (KF): 1.7 wt %.
Pd (ICP-MS): 5 ppm.
removed under reduced pressure (p = 400−500 mbar, Tint
=
20−30 °C, Tjacket = 45−80 °C), and the residual was rinsed
with MTBE (2.5 L). The suspension was cooled to 10 °C for 1
h and filtered on a Buchner funnel. The isolated product was
dried in the vacuum oven (p = 20 mbar, Toven = 25 °C) until
constant weight in order to obtain 320 g of 19. The mother
liquors were concentrated under reduced pressure (p = 300−
400 mbar, Tint = 20−45 °C, Tjacket = 45−70 °C; 1.0 L of
distillate recovered), affording a suspension which was stirred at
10 °C for 1 h and then filtered on a Buchner funnel to give a
second crop. The isolated product was dried in the vacuum
oven (p = 20 mbar, Toven = 25 °C) until constant weight in
order to obtain 135 g of 19. Yield (combined): 81%.
UPLC−MS: tR = 0.92 min, m/z = 415 [M + 1]+.
ASSOCIATED CONTENT
■
S
* Supporting Information
Copies of 1H NMR and 13C NMR spectra. Experimental
procedures for the preparation of most compounds in plant
scale. Copies of analytical data of SEN461 process impurity
identification. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: mbetti@sienabiotech.it
HRMS calcd for C22H31N4O4 [M + 1]+ 415.23456, found
415.23455.
Present Addresses
⊥Aptuit (Verona), 37135 Verona, Italy.
§Novartis Vaccines, 53100 Siena, Italy.
HPLC: tR = 3.41 min, purity 98.6%.
1H NMR (400 MHz DMSO-d6): δ 7.47 (d, J = 2.6 Hz, 1H),
7.41−7.35 (m, 2H), 3.82 (m, 5H), 3.48 (s, 3H), 3.34 (m, 4H),
2.62−2.45 (m, 5H), 1.75−1.60 (m, 5H), 1.28−1.04 (m, 4H).
Notes
The authors declare no competing financial interest.
1050
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051

Organic Process Research & Development
Article
(21) We explain the dimer formation as a consequence of a double
nucleophilic attack of the starting amine 19 and the methylene bridge
position of the 2-methylacetyl moiety of 1 to DCM in basic conditions.
For further details see the Supporting Information.
ACKNOWLEDGMENTS
■
We thank the many operators and support personnel at Aptuit
Verona. We especially recognize Arianna Ribecai, Angelo
Manzo, Federico Scaravelli for their support in process
development; Simone Battaggia, Simone Guelfi, Francesco
Tinazzi for their work in the pilot-plant campaign; Yves Saint-
Denis for managing multiple aspects of the research timeline
with the business timelines; Stefano Provera for his NMR
studies and interpretation. We also gratefully acknowledge
Arianna Nencini and Maurizio Varrone for the early medicinal
chemistry work in Siena Biotech.
REFERENCES
■
(1) Nusse, R. Cell Res. 2005, 15, 28−32. Barker, N.; Clevers, H. Nat.
Rev. Drug Discovery 2006, 5, 997−1014.
(2) Clevers, H. Cell 2006, 127, 469−480.
(3) Barker, N.; Clevers, H. Nat. Rev. Drug Discovery 2006, 5, 997−
1014.
(4) Rey, J. P.; Ellies, D. L. Dev. Dyn. 2010, 239, 102−114.
(5) Moon, R. T.; Shah, K. Nature 2002, 417, 239−240.
(6) Nelson, W. J.; Nusse, R. Science 2004, 303, 1483−1487. Seto, E.
S.; Bellen, H. J. Trends Cell Biol. 2004, 14, 45−53.
(7) De Robertis, A.; Valensin, S.; Rossi, M.; Tunici, P.; Verani, M.;
De Rosa, A.; Giordano, C.; Varrone, M.; Nencini, A.; Pratelli, C.;
Benicchi, T.; Bakker, A.; Hill, J.; Sangthongpitag, K.; Pendharkar, V.;
Liu, B.; Mee Ng, F; Wen Then, S.; Jing Tai, S; Cheong, S. M.; He, X.;
Caricasole, A.; Salerno, M. Mol. Cancer Ther. 2013, 12, 1180−1189.
(8) Veeman, M. T.; Axelrod, J. D.; Moon, R. T. Dev. Cell 2003, 5,
367−377.
(9) Aldrich; 25 g at €154/g; 500 g at €2.90/g.
(10) Direct coupling of benzoic acid 14 with amino acid 4 would cut
the current synthesis by two steps and was therefore investigated.
While activation of 14 as acyl chloride was complete, subsequent
addition of amino acid 4 gave only very poor conversion.
(11) Gygax, R. Thermal Process Safety, Data Assessment, Criteria,
Measures; Expert Commission for Safety in the Swiss Chemical
Industry (ESCIS): Lucerne, 1993; Vol. 8.
(12) Urben, P. G., Ed. Bretherick’s Handbook of Reactive Chemical
Hazards; Oxford: New York, 2006; Vol. 1.
(13) Somanathan, R.; Smith, K. M. J. Heterocycl. Chem. 1981, 18,
1077.
(14) Green, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis; Wiley-Interscience: New York, 1999; pp 518−525, and
736−739.
(15) Shendage, D. M.; Froehlich, R.; Haufe, G. Org. Lett. 2004, 6,
3675−3678. Englund, E. A.; Gopi, H. N.; Appella, D. H. Org. Lett.
2004, 6, 213−215.
(16) SEN461 was nominated for clinical evaluation, and the
manufacturing of a GMP batch was outsourced. The process
development studies and the manufacturing campaign were carried
out within Aptuit API Development Manufacturing facilities based in
Verona, Italy.
(17) The experiments were run in a RC1 Reaction Calorimeter,
Mettler Toledo.
(18) The activation step was monitored by HPLC (at 220 nm)
previously adding benzylamine to the reaction sample and having an
indication of the residual oxalyl chloride by observing the UV−visible
product obtained.
(19) The possibility of introducing a purification step at the
beginning of stage 7 was considered with the aim of avoiding
contamination by inorganics and potential risks in the pilot-plant
campaign. Even if the rework procedure was put in place and tested, it
was not used since a wash from inorganics was not required.
(20) The permeation rate during washing was improved using a
DCM/MTBE mixture instead of pure MTBE. For further details see
the Supporting Information.
1051
dx.doi.org/10.1021/op400145w | Org. Process Res. Dev. 2013, 17, 1042−1051