Article
A Robust, Streamlined Approach to Bosutinib Monohydrate
Gregory J. Withbroe, Chris Seadeek, Kevin P. Girard, Steven M. Guinness, Brian C. Vanderplas,
and Rajappa Vaidyanathan*,†
Chemical Research and Development, Pfizer Inc., Eastern Point Road, Groton, Connecticut 06340, United States
ABSTRACT: This article describes a systematic approach used to streamline the process for the isolation of bosutinib
monohydrate, a promiscuous solvate former. A thorough understanding of the complex solid form landscape was garnered, and
this knowledge was used to develop a process that routinely delivered the correct solid form and excellent purity at the end of the
last bond-formation step, without the need for additional recrystallization and/or solid form conversion steps.
form at the end of the last synthetic step via a controlled
crystallization. This paper describes our efforts that culminated
in a robust, streamlined isolation process for bosutinib
monohydrate from 7.
INTRODUCTION
■
Bosutinib monohydrate (1·H2O) is a potent competitive dual
inhibitor of Src and Abl kinases that inhibits cell growth,
metastasis, and osteoclast activity and has been developed as a
treatment for chronic myelogenous leukemia (CML). Several
synthetic approaches to this compound were evaluated,1−4 and
the synthesis depicted in Scheme 1 was selected for the
manufacture of late-stage clinical supplies.5
RESULTS AND DISCUSSION
■
While the desired solid form, the monohydrate (1·H2O), exists
as a single polymorph, bosutinib is known to be a promiscuous
solvate former, and exhibits a propensity to form solvate/
hydrates, solvates, and higher-order hydrates under a variety of
conditions.6 Therefore, it was important to gain a full
understanding of the complex solid form landscape in order
to develop a process that would routinely deliver bosutinib
monohydrate (of acceptable purity) at the end of the last bond-
forming step.
Solid form screening was initiated on bosutinib using a
variety of single organic solvents, and binary systems with
water. In this screen, approximately 20 mg of 1·H2O was
weighed into 1.5 mL HPLC vials, and diluted with ∼300 μL of
the appropriate solvent or binary solvent/water mixture. More
1·H2O was added to ensure saturation, and the samples were
allowed to stir for 10 days at room temperature. The resulting
slurries were filtered using a 0.45 μm nylon centrifuge filter, and
the filtrates were gravimetrically analyzed for solubility. The
isolated solids were analyzed via PXRD, DSC, and TGA to
determine the solid form recovered, and the key findings are
summarized in Table 1.
A large number of screened solvent systems generated
undesirable solvated solid forms (column 3, Table 1). The
esters and ketones screened were a notable exception, where
the monohydrate was recovered. Similarly, hydrocarbon
solvents such as toluene, n-heptane, and xylene provided the
monohydrate, although the solubility of 1·H2O was too low to
warrant further investigation of these solvents (Table 2). Most
of the alcohol solvents resulted in the corresponding solvates,
and binary systems containing water and alcohols provided
solvate/hydrates.
Alkyl chloride 2 was treated with N-methylpiperazine 3 to
furnish adduct 4, which was reduced to provide aniline 5. A
three-component coupling reaction of 5 with cyanoacetamide 6
and triethyl orthoformate led to 7 as a mixture of cis/trans
isomers. Cyclization of 7 using POCl3 in sulfolane provided the
core structure of bosutinib (1). Basification of the reaction
mixture with aqueous KOH led to precipitation of the
hexahydrate, 1·6H2O. Recrystallization of the hexahydrate
from aqueous iso-propyl alcohol upgraded purity by removing
organic impurities and furnished the dihydrate-iso-propyl
alcohol solvate (1·2H2O·i-PrOH). This compound was
converted to the desired monohydrate solid form (1·H2O)
by treatment with hot water.
The sequence of synthetic transformations works very well
and produces acceptable quality active pharmaceutical ingre-
dient (API). The two additional steps after the final bond-
formation step were incorporated in order to achieve the
desired purity and solid form. Not surprisingly, an evaluation of
processing costs indicated that these steps contributed
significantly to the overall cost of API manufacture. In addition,
the solid form conversion step (step 6) was carried out as a
slurry-to-slurry transformation in water and produced the
product as a wide distribution of fine particles prone to
agglomeration, suggesting that filtration on large-scale would
likely be problematic, depending on the choice of equipment.
For these reasons, there was a huge impetus to eliminate the
last two steps, and strive to attain the desired purity and solid
Special Issue: Polymorphism and Crystallization 2013
Received: April 2, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op300087r | Org. Process Res. Dev. XXXX, XXX, XXX−XXX