Unexpected Single Crystal Growth Induced by a Wire and New Crystalline Structures of Lapatinib

Article abstract of DOI:10.1021/acs.cgd.6b01271

Single crystal growth of lapatinib free base was induced by immersion of a copper wire into a supersaturated methanolic aqueous solution yielding monoclinic anhydrous plates (space group P2₁/c, Form 1) and needles of a previously unknown channel hydrate (in P4₂2₁2). Also, a new method has been developed herein to obtain anhydrous Form 1 via acid-base reaction of lapatinib ditosylate and sodium methoxide, avoiding the usage of an aqueous solution and hydrate formation. Anhydrous Form 2 as well as new solvates were produced via solution mediated transformation experiments, including a dichloromethane solvate with a powder X-ray diffraction pattern similar to that of anhydrous Form 2. Differential scanning calorimetry and solution equilibrium experiments helped to elucidate the interconversion pathways between Form 1, Form 2, and the solvates.

Full text of DOI:10.1021/acs.cgd.6b01271

Subscriber access provided by United Arab Emirates University | Libraries Deanship
Article
Unexpected single crystal growth induced by a
wire and new crystalline structures of lapatinib
Gabriel Lima Barros de Araujo, Matthias Zeller, Daniel Smith, Haichen Nie, and Stephen R. Byrn
Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.6b01271 • Publication Date (Web): 06 Sep 2016
Downloaded from http://pubs.acs.org on September 7, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Crystal Growth & Design 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 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Unexpected single crystal growth induced by a wire
and new crystalline structures of lapatinib
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
*
† ‡
§
‡
‡
‡
Gabriel L. B. de Araujo
, Matthias Zeller , Daniel Smith , Haichen Nie , Stephen R. Byrn
†
Faculty of Pharmaceutical Sciences, Department of Pharmacy, University of Sao Paulo, SP,
Brazil
§
Department of Chemistry, Purdue University, West Lafayette, IN, United States
‡
Department of Industrial and Physical Pharmacy, Purdue University, West Lafayette, IN,
KEYWORDS. crystal growth; crystal structure; crystal polymorphism; hydrates/solvates; phase
transition;
ABSTRACT. Single crystal growth of lapatinib free base was induced by immersion of a copper
wire into a supersaturated methanolic aqueous solution yielding monoclinic anhydrous plates
(
space group P2 /c, Form 1) and needles of a previously unknown channel hydrate (in P4 2 2).
1
2
1
Also, a new method has been developed herein to obtain anhydrous Form 1 via acidꢀbase
reaction of lapatinib ditosylate and sodium methoxide, avoiding the usage of an aqueous solution
and hydrate formation. Anhydrous Form 2 as well as new solvates were produced via solution
ACS Paragon Plus Environment
1

Crystal Growth & Design
Page 2 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mediated transformation experiments; including a dichloromethane solvate with a powder Xꢀray
diffraction (XRPD) pattern similar to that of anhydrous Form 2. Differential scanning
calorimetry (DSC) and solution equilibrium experiments helped to elucidate the interconversion
pathways between Form 1, Form 2 and the solvates.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
In spite of the innumerous advances in many fields of the crystallization science, the success of
unexpected approaches reveals why single crystal growth and the finding of new polymorphs of
1–3
organic entities are uncertain and sometimes considered more as a mixture of science and art
1
,3–5
6
7
8
9
with a pinch of creativity and luck.
Vaporꢀdiffusion , gels , highꢀpressure , laser beam ,
1
0
11
electric and magnetic fields are examples of interesting and creative methods to promote
crystal growth. Also, the presence of seeds that allow a primary surface to the first nucleation are
recognized to play a primordial role and can vary from highly engineered approaches as the use
12
13,14
of gold islands with selfꢀassembled monolayers surfaces or polymers
, unintentional
3
15,16
impurities like airborne particles and byproducts or degradation products
to the simplest
2
classical procedure of scratching the glassware. Braga et al. stated that crystal makers need
“sagacity and readiness of mind”, so they “will be ready to pick new avenues as the crystal
experiments will yield something unplanned for”. One curious episode was reported by Bucar et
3
al. After two years trying to repeat the growth of a specific polymorph of a dimethyl analogue of
the dichlorobenzylideneaniline, they succeed only after moving to a new laboratory operated by
a new student hired by phone, using new reagents and new glassware to avoid the presence of
3
unknown seeds from other polymorphs. For a crystal engineer faced with this wideꢀrange of
ACS Paragon Plus Environment
2

Page 3 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
2,17
possibilities only two things are certain: for while there is no universal method and it is a
6
compound specific problem.
Herein, we report the investigation of new crystalline forms of lapatinib free base (LP) and the
unplanned discovery of an unusual approach to promote single crystal growth of two of these
forms. Lapatinib (Figure 1) is an effective and selective oral tyrosine kinase inhibitor first
approved by the FDA in 2007 for the treatment of metastatic HER2ꢀpositive breast cancer.
Recently, Lapatinib has been reported to shrink a metastatic brain lesion resulting from HERꢀ2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
8
positive breast cancer. In the case of lapatinib, the marketed salt lapatinib ditosylate (Tykerb®,
GlaxoSmithkline, here abbreviated as LPDS) has more than twenty reported polymorphs that can
be obtained from a wide range of conditions under thermodynamic and kinetic experimental
control (Table 1). Lapatinib free base is of interest because it can form salts with acidic polymers
such as HPMCP and HPMCAS and these amorphous salts are rather stable in the amorphous
1
9
state. However, lapatinib free base has been reported to form only four crystalline polymorphs
named Form 1, Form 2, Form X and Form Y with very limited thermodynamic or structural data
20,21
available.
To the best of our knowledge, no single crystal structure or interconversion studies
have been reported for any of the polymorphs that have been described so far. After months of
unsuccessful trials using classical procedures for growth of single crystals, in a moment of
inspiration (or despair since we were running out of material), we decided to offer a different
surface to promote a site for nucleation. Since at that time, we had only an old piece of copper
wire available on hand, it was introduced in the drug solution and left to slowly evaporate
yielding the new structures here described in the first attempt. Since that time we have repeated
the experiment with the same results.
ACS Paragon Plus Environment
3

Crystal Growth & Design
Page 4 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structural formula of lapatinib.
Table 1. Some of the crystal forms related to lapatinib free base and its salts.
Salts and Polymorphs of Lapatinib
Lapatinib anhydrate Form 1, Form 2 and solvates (1,4 dioxane, pyridine and Nꢀmethylꢀ2ꢀ
2
0
pyrrolidinone solvates)
Polymorphs of lapatinib ditosylate (Forms I, II, III, IV, V, VI, VII, VIII, IX, X, XI, XII,
2
2
XIII, XIV, XV, XVI, XVII, XVIII, XIX)
Lapatinib monotosylate (Form M1) and other salt forms of lapatinib in crystalline and
amorphous states (maleate, tartrate, fumarate, succinate, sulfate, diꢀhydrochloride, diꢀ
2
3
hydrobromide, phosphate)
2
1
Amorphous and crystalline lapatinib free base polymorphs (Forms X and Y)
Lapatinib ditosylate anhydrous and hydrate (Form A and Form B) and amorphous
24
ditosylate form
ACS Paragon Plus Environment
4

Page 5 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Table 1 continued
Salts and Polymorphs of Lapatinib
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
2
5
Lapatinib ditosylate forms APOꢀI and APOꢀII (isopropanol solvate)
2
6
Lapatinib ditosylate Nꢀmethylpyrrolidinone solvate
2
7
Coꢀcrystals of lapatinib monoacid salts
ACS Paragon Plus Environment
5

Crystal Growth & Design
Page 6 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EXPERIMENTAL SECTION
Materials
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Highꢀpurity (>99%) lapatinib free base and lapatinib ditosylate were obtained from LC
Laboratories (Woburn, MA) and used as received. Sodium carbonate and sodium methoxide
(25% by wt. in methanol) were procured from SigmaꢀAldrich (SigmaꢀAldrich, St. Louis, MO) as
were high purity solvents (HPLC grade): acetonitrile, 1,4ꢀdioxane, toluene, ethyl acetate,
tetrahydrofuran, dichloromethane, 2ꢀpropanol, acetone, methanol, heptane, methylꢀtert butyl
ether (MTBE).
Copper wire induced single crystal growth from methanol
LP free base was dissolved in methanol (0.5 mg/mL) under stirring and filtered through a 0.22
µm PVDF filter (Grace Davidso, Dearfield, IL). A 6 mL volume of this solution was transferred
to a 20 mL glass vial; then RO water (ca. 3 mL) was added dropwise until a slightly blue
turbidity was observed (Tyndall effect). An additional 3 mL of the methanolic LP solution was
added to the colloidal dispersion to yield a transparent solution. The vial was left partially
covered and a copper wire was immersed in the solution to promote crystallization. The solution
was allowed to slowly evaporate under ambient condition for one week. Another sample,
prepared the same way, but without the wire, and was kept under the same conditions. The
crystals obtained from the vial with the copper wire were collected by filtration and kept in a
sealed glass vial at 2 – 8 °C. No crystals were observed in the vial without the copper wire.
ACS Paragon Plus Environment
6

Page 7 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Preparation of crystalline forms
Polymorphism screening was performed using three approaches: acid base reactions of LPDS
with either sodium carbonate or sodium methoxide and slurries of LP free base. For all
approaches the drug was suspended in 10 solvents selected preferentially from different groups
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
2
8
according to their properties as categorized by C.ꢀH. Gu et al.. The acid base reaction was
conducted using ditosylate salt and a selected base (sodium carbonate or sodium methoxide).
LPDS was suspended in the selected solvent and stirred for 30 minutes at room temperature.
Either aqueous sodium carbonate or methanolic sodium methoxide was added dropwise to the
suspension and the resulting mixture stirred for 3 hours in a sealed vial at room temperature. The
resulting suspension was filtered and the powder was dried for 48 h under vacuum. Slurry
experiments were conducted by suspending LP free base in selected solvents and stirring for one
week in a sealed glass scintillation vial. The solids were collected by filtration and the powder
was dried for 48 h under vacuum. Additional solvent mediated transformation studies were
performed according to conditions described in Table 2.
Single crystal Xꢀray diffraction
Single crystal Xꢀray measurements were conducted on a Rigaku Rapid II curved image plate
diffractometer with a CuꢀKα Xꢀray microsource (λ = 1.54178 Å) with a laterally graded
multilayer (Goebel) mirror for monochromatization. Single crystals were mounted on Mitegen
microloop mounts using a trace of mineral oil and cooled inꢀsitu to 100(2) K for data collection.
2
9
Data were collected using the dtrek option of CrystalClearꢀSM Expert 2.1 b32. Data sets were
3
0
processed using HKL3000 and data were corrected for absorption and scaled using
31
Scalepack. The space groups were assigned and the structures were solved by direct methods
ACS Paragon Plus Environment
7

Crystal Growth & Design
Page 8 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
31,32
using XPREP within the SHELXTL suite of programs
and refined by full matrix least
2
33
34
squares against F with all reflections using Shelxl2014 with the graphical interface Shelxle.
H atoms attached to carbon and nitrogen atoms were positioned geometrically and constrained to
ride on their parent atoms, with carbon hydrogen bond distances of 0.95 Å for aromatic and
alkene H atoms, 0.99 for CH , and 0.98 for CH moieties, respectively. Methyl H atoms were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
allowed to rotate but not to tip to best fit the experimental electron density. NꢀH distances in the
plate (monoclinic) polymorph were freely refined and NꢀH distances restrained to 0.86(2) Å. In
the tetragonal hydrate polymorph, the amine H atom attached to N3 was set to 0.88 Å. That at
N4 is 1:1 disordered (see below) and the NꢀH distances were set to 0.91 Å. Water H atoms were
freely refined and their positions restrained based on hydrogen bonding considerations. U (H)
iso
+
values were set to a multiple of U (C/N) with 1.5 for OH and CH3, and 1.2 for NH , CꢀH and
eq
2
CH units, respectively.
2
For the needle shaped crystals (tetragonal polymorph) several types of disorder are observed.
The SO Me group is disordered over two orientations with the two moieties being in general
2
positions. Its disorder induces disorder of the water molecule of O8, and is correlated with
presence or absence of the water molecule of O11. Parallel to this general disorder there is
systematic disorder across symmetry elements and due to incompatible hydrogen bonds across
symmetry elements, causing 1:1 disorder of large sections of the solvate regions and for the
amine hydrogens of N4.
Systematic disorder: Hydrogen bonded to the amine atom N3 is the water molecule of O6,
which is in turn hydrogen bonded to its own counterpart across a twofold axis, causing 1:1
disorder of the H atoms of water molecule of O6. In its alternative position the second H atom of
O6 is Hꢀbonded to the lone pair of the amine of N4. When the H atom is engaged in a hydrogen
ACS Paragon Plus Environment
8

Page 9 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
bond to the counterpart of O6 across the two fold axis, the lone pair of O6 acts as an acceptor for
the amine H atom of N4, thus causing 1:1 disorder of the H atoms on amine N4. This disorder in
turn causes disorder of H atoms at the water molecule of O8, which is H bonded to N4. To refine
this disorder, O6 and its H atoms were split into two half occupied moieties nearly related by the
two fold axis. O6 and O6B were constrained to have identical anisotropic displacement
parameters, ADPs.
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 disordered SO2Me group moieties were restrained to have similar geometries and Uij
components of ADPs of their atoms were restrained to be similar if closer than 1.7 Å. The
SO2Me disorder is correlated with disorder of the water molecule of O8 and with presence or
absence of the water molecule of O11. O8 was split into two positions with the same occupancy
ratios as the SO2Me group. The distance of O8 and O8B to N4 was restrained to be similar. Both
the major moiety of O8 as well as the minor moiety of O8B show additional disorder of their
hydrogen atoms, induced by the H atom disorder at N4, see above. The water molecule of O11 is
incompatible with O8B, with no indication of disorder as for O8 and the SO Me group visible.
2
O11 was thus refined as partially occupied and absent when O8B is present. Both O11 as well as
neighboring O7 are 1:1 disordered across two fold axes. Hydrogen atoms of O8, O11 and O7
were placed in positions compatible with the expected hydrogen bonding interactions. One of the
H atoms of O7 was additionally split due to hydrogen bonding across another two fold axis to a
symmetry created copy of O7. H atoms were placed in positions compatible with the expected
hydrogen bonding interactions and were refined with a damping factor and distance restraints
based on Hꢀbonding considerations. In the final refinement cycles the damping factor was
removed and H atoms were set to ride on their carrier oxygen atoms. A mild antiꢀbumping
restraint was applied during H atom refinement and was retained in the final refinement steps.
ACS Paragon Plus Environment
9

Crystal Growth & Design
Page 10 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Subject to these conditions, the occupancy ratio of the SO Me moieties and O8 refined to
2
0.810(7) to 0.190(7).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Xꢀray powder diffraction
XꢀRay powder diffraction patterns were measured on a Siemens/Bruker D5000 diffractometer,
using Ni filtered Cu Kα radiation (λ = 1.5418 Å) with an acceleration voltage of 40 kV and a
tube current of 40 mA, step size of 0.02, step time 5 s with an angular range of 4° < 2Θ < 40°.
Thermal analysis
Differential Scanning Calorimetry (DSC) curves were obtained using a DSC Q10 cell (TA
Instrument, USA). About 2 mg of sample were weighed into aluminum crucibles and
hermetically sealed. The samples were scanned at a heating rate of 10 °C/min, under a dynamic
N atmosphere (100 mL/min), in a temperature range of 25 °C to 200 °C.
2
RESULTS AND DISCUSSION
Single crystal growth experiments
Conventional approaches such as room temperature evaporation from solution, boiling and
cooling of a supersaturated solution and vapor diffusion were not successful in inducing the
growth of single crystals of lapatinib free base. However, introduction of a copper wire into a
supersaturated methanol/water solution of LP induced the nucleation and growth of single
ACS Paragon Plus Environment
10

Page 11 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
crystals with two different morphologies, plates and thin needles; both of which adhered to the
wire (Figure 2ꢀa and 2ꢀb). No crystal growth was observed using the same solution and
conditions in the absence of the wire. Crystals with regular shape were selected for singleꢀcrystal
Xꢀray diffraction analysis. Structures with atomic labels and crystal packing are shown in
Figures 3 and 4. Crystallographic data are reported Table 2. Calculated Xꢀray powder patterns
are presented in Figures 8ꢀd and Fig.9ꢀf.
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
Figure 2. Single crystal growth induced by copper wire: (a) lapatinib free base crystal growth in
a scintillation vial using a copper wire; (b) closeup view of needles attached in the copper wire;
ACS Paragon Plus Environment
11

Crystal Growth & Design
Page 12 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
c) optical microscopy photomicrographs (10x) of plates under polarized light and (d) a needle
under nonpolarized light.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Crystallographic data of lapatinib plates and needles obtained from recrystallization in
methanol/water seeded with a copper wire.
Crystalline Modification
Anhydrous (plates)
Form 1
3.4 Hydrate (needles)
Solvate A (3.4 hydrate)
.
Formula
C H ClFN O S
C H ClFN O S 3.4H O
29 26
4
4
29 26
4
4
2
Temperature
Space group
Cell parameters
100 K
100 K
P2 /c
P4 2 2
1
2 1
a = 38.752(3) Ǻ
b = 5.3361(4) Ǻ
c = 12.5021(8) Ǻ
β = 90.447 °
2585.16
a/b = 26.2160 (4) Ǻ
c = 8.6021 (2) Ǻ
3
Cell volume (Ǻ )
5912.04
8
Z
4
2
2
Rꢀfactor ([F > 2σ(F )], %)
7.91
5.58
ACS Paragon Plus Environment
12

Page 13 of 45
Crystal Growth & Design
1
2
3
4
5
6
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
Figure 3. Crystal structures from single crystal Xꢀray diffraction (ORTEP view, thermal
ellipsoids at the 50% probability level) showing atoms labeling: a) LP anhydrous (Form 1); b)
LP 3.4 hydrate.
ACS Paragon Plus Environment
13

Crystal Growth & Design
Page 14 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. (a) Intermolecular Hꢀbonding interactions and (b) crystal packing view along the b
axis in anhydrous LP obtained by recrystallization from methanol/water with copper wire.
ACS Paragon Plus Environment
14

Page 15 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
The plate like crystals obtained are anhydrous and the Xꢀray powder pattern calculated from
the single crystal corresponds very well to the anhydrous Forms 1 and X claimed by different
patents (Support Information Table S1 and Figure 6ꢀf). Form 1 is reported to be an anhydrate
obtained from the reaction between lapatinib ditosylate suspended in ethyl acetate and sodium
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
2
0
carbonate in aqueous solution. Form X is described to be formed when the salt is suspended in
acetonitrile and equilibrated at 40 °C, followed by reaction with sodium carbonate in an aqueous
2
1
solution yielding LP free base. Normalized Xꢀray powder diffraction patterns of Form 1 and
Form X are practically the same as can be observed in Tablet S1 and shown in Figure 5.
2
0,21
Figure 5. Comparison of Form X and Form 1 Xꢀray powder data extracted from patents.
The LP needle shaped crystals, on the other hand, have a characteristic channel hydrate
structure, solvate A, with several types of disorders (vide infra), with ca. 3.4 water molecules for
every molecule of lapatinib free base. Figure 6 shows a comparison of the conformation of
lapatinib in the plates (Form1) and needles (solvate A). The two forms can be regarded as
conformational polymorphs, if the solvate water molecules of the hydrate are ignored. In Form
ACS Paragon Plus Environment
15

Crystal Growth & Design
Page 16 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
, the quinzaolinamine ring and the appended furan ring as well as the aminomethylsulfonyl side
chain are nearly coplanar. The phenyl rings bearing the chlorine and the fluorine atoms, on the
other hand, are not coplanar with the quinzaolinamine ring. In the solvate A, the quinzaolinamine
ring, the appended furan ring and the two phenyl rings bearing the Cl and F atoms are nearly
coplanar. Here, the aminomethylsulfonyl side chain is not coplanar with the larger ring system
with the CꢀN bond being oriented nearly perpendicular to the ring system (Figures 3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Overlay of the conformations of lapatinib motif:(blue) in hydrate structure and (red)
anhydrous.
The four molecules in the unit cell in Form 1 are held together in pairs by one NꢀHꢀꢀꢀO
hydrogen bond (2.21 Å) and one NꢀHꢀꢀꢀN hydrogen bond (2.16 Å). In the solvate, the nearly
planar ring systems of the eight molecules in the unit cell are arranged in two types of stacks
along the z axis. The stacks form the channels (Figure 7) that accommodate the solvent water
molecules. The stacks of planar rings are held together by van der Waals contacts as well as one
ACS Paragon Plus Environment
16

Page 17 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
hydrogen bond connecting a nitrogen atom of the quinzaolinamine ring to a water molecule
which in turn is hydrogen bonded to an adjacent quinzaolinamine ring. The estimated solvent
3
accessible volume is 570 Å (9.6% of the unit cell volume), with a probe radius of 1.4 Å,
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
(
calculated using the contact surface approach of the Mercury software). Substantial disorder is
observed in the solvate: The SO Me group in solvate A is disordered over two orientations in
2
general positions. Its disorder induces disorder of one of the water molecules (O8), and is
correlated with presence or absence of another water molecule (O11). Parallel to this general
disorder there is systematic disorder across symmetry elements due to incompatible hydrogen
bonds across symmetry elements, causing 1:1 disorder of large sections of the solvate regions
and for the amine hydrogens (N4).
Figure 7. Water channels in LP 3.4 hydrate structure.
ACS Paragon Plus Environment
17

Crystal Growth & Design
Page 18 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Crystal space of lapatinib
Although complex, demanding and often not possible to achieve, the single crystal growth and
structure determination are still fundamental steps to map the crystal space, obtain a standard for
powder diffraction and understand the structural aspects of polymorphism phenomena like
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
,35
solvatation and interconversions.
In order to explore the crystal space of lapatinib free base
polymorphism screening and interconversion studies were carried out and results were
summarized in Table 3 and Scheme 1.
ACS Paragon Plus Environment
18

Page 19 of 45
Crystal Growth & Design
1
2
3
4
5
6
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
Scheme 1. Overview of polymorphic conversions in LP crystal forms.
ACS Paragon Plus Environment
19

Crystal Growth & Design
Page 20 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Summary of results of LP polymorphism screening.
Solvent
Acidꢀbase reaction with Acidꢀbase reaction with Slurry for one week
sodium carbonate
sodium methoxide
at room temperature
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Acetonitrile
Solvate A (3.4 hydrate)
***
***
***
a
b
(condition 1 )
Acetonitrile
(condition 2 )
Form 2
***
1,4ꢀDioxane
***
***
***
***
Form 2 unchanged
Form 2 unchanged
***
Toluene
***
Ethyl acetate
Tetrahydrofuran
Dichloromethane
***
***
***
***
***
***
Solvate B
Solvate C
Form 1
1ꢀOctanol
Solvate C
***
Acetone
Methanol
Heptane
Solvate A (3.4 hydrate)
***
***
***
***
Form 1
Form 1
***
Form 1
***
MTBE (methylꢀtert
butyl ether)
***
a
condition 1: reaction was carried out in suspension at 25 °C, under stirring for 1 hour. Then,
crystals were filtered and dried under vacuum for 48 hours at room temperature.
b
condition 2: reaction was carried out in suspension at 40 °C, under stirring for 1 hour. The
resulting yellow suspension was stirred at 5 °C for 2 hours. The product was filtered and dried
under vacuum for 48 hours at room temperature.
*** incomplete reaction or impurity profile inadequate.
ACS Paragon Plus Environment
20

Page 21 of 45
Crystal Growth & Design
1
2
3
4
5
6
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
Figure 8. Normalized Xꢀray powder diffractogram of LP solvates: a) solvate obtained from
slurry of Form 2 in octanol (one week at RT); b) solvate from acidꢀbase reaction of LPDS with
sodium carbonate in aqueous and acetone; c) solvate from acidꢀbase reaction of LPDS with
sodium carbonate in aqueous and acetonitrile; d) Xꢀray powder pattern calculated from LP 3.4
hydrate.
ACS Paragon Plus Environment
21

Crystal Growth & Design
Page 22 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 9. Normalized Xꢀray powder diffractograms of nonꢀsolvate LP crystals obtained
experimentally: a) LP crystals obtained from acidꢀbase reaction of LP with sodium carbonate in
water and acetonitrile (at 40 °C); b) LP crystals from slurry in methanol (one week at RT); c) LP
crystals from slurry in acetone (one week at RT); d) LP crystals obtained from acidꢀbase reaction
of LPDS with sodium methoxide in methanol; e) LP crystals obtained from acidꢀbase reaction of
LPDS with sodium methoxide in heptane; f) Xꢀray powder pattern calculated from anhydrous LP
Form 1 (100 K data, additional low angle peak at 4.612°).
ACS Paragon Plus Environment
22

Page 23 of 45
Crystal Growth & Design
1
2
3
4
5
6
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
Figure 10. Normalized Xꢀray powder diffractogram LP crystals obtained experimentally: a) LP
crystals obtained from acidꢀbase reaction of LP with sodium carbonate in water and acetonitrile
(at 40 °C); b) LP Form 2 from commercial source (arrows indicate presence of Form 1); c)
solvate obtained from slurry of Form 2 in dichloromethane (one week at 25 °C).
ACS Paragon Plus Environment
23

Crystal Growth & Design
Page 24 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 11. DSC curves obtained in dynamic N atmosphere (100 mL/min) at a heating rate of 10
2
°C/min: a) hydrate obtained from acidꢀbase reaction of LPDS with sodium carbonate in water
and acetonitrile (condition 1); b) hydrate from ethanol/water single crystal experiments; c)
solvate obtained from slurry of Form 2 in octanol (one week at RT); d) solvate obtained from
slurry of Form 2 in dichloromethane (one week at RT); e) Form 2 obtained from acidꢀbase
reaction of LP with sodium carbonate in aqueous and acetonitrile (at 40°C); f) LP Form 2 from
commercial source; g) Form 1 obtained from acidꢀbase reaction of LPDS with sodium methoxide
in methanol.
ACS Paragon Plus Environment
24

Page 25 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Xꢀray powder patterns of LP solvates and its anhydrous forms obtained in the screening are
presented in Figures 8 and 9, respectively. Acidꢀbase reactions of LPDS with sodium carbonate
in acetronitrile/water or ethyl acetate/water had been described to produce LP Form X or Form 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
20,21
respectively.
By dissolving LP in heated dichloromethane and allowing the solution to cool
2
0
gradually it is reported to crystallize into Form 2. The same procedure when conducted in 1,4ꢀ
2
1
dioxane is claimed to lead to Form Y. In our screening, however, reactions carried out in an
acetonitrile/water environment were resulted in formation of a solvate and Form 2, depending on
experimental conditions. No complete reaction was accomplished in ethyl acetate as solvent. To
avoid the hydrate formation a reaction in absence of water was proposed with sodium methoxide.
XRPD patterns of crystals from acidꢀbase reaction of LPDS with sodium methoxide in methanol
(Fig.9ꢀd) and heptane (Fig. 9ꢀf) and those obtained from slurry conversion of commercial Form 2
in methanol (Fig.9ꢀb) and acetone (Fig. 9ꢀc) showed a good match with the diffractogram
calculated from the single crystal structure of the anhydrous plates (Fig.9ꢀf), and with the
previously reported Forms X and 1 (Table S1). Their DSC curves (Fig. 11ꢀg) indicate that this
crystal form has a wellꢀdefined melting endortherm at = 137 °C (Tonset) with an enthalpy of
fusion of 76.2 J/g (Tpeak=144 °C).
It is known that solvent mediated anhydrate/hydrate phase transformations are complex
processes that depend on competition of kinetic and thermodynamic factors related to water
3
6
activity in the surrounding solvent and temperature. As the acetonitrile/water ratio was kept
the same between both procedures, the hydrate/anhydrate transition temperature in this system
seems to be an important factor to be explored in the crystallization methodologies described in
2
0,21
literature.
XRPD patterns reveal that the solvates obtained are in good agreement to the
channel hydrate structure obtained from the single crystal growth experiments (Fig.8ꢀb, 8ꢀc and
ACS Paragon Plus Environment
25

Crystal Growth & Design
Page 26 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8
ꢀd); it is also supported by DSC curves (Figure 11ꢀa and 11ꢀb) which show a similar
dehydratation event in the range of 40ꢀ80 °C. This unusual water loss behavior at low
temperatures with a broad endothermal peak is characteristic of channel solvates and has been
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
36–39
observed for sodium ibruprofen dihydrate and carbamazepine dehydrate.
The LP crystals
obtained from acidꢀbase reaction of LPDS with sodium carbonate in water and acetonitrile at
0 °C with subsequent cooling to 5 °C yielded crystals with XRPD patterns in good agreement
4
2
0
with Form 2 described in patent WO 2009/079547. Nevertheless, DSC curves reveal a
different thermal behavior from experimental Form 2 and the commercial sample. Both samples
present a first endothermal event assigned to melting of Form 2 in the range of 85ꢀ115 °C, 25 °C
lower than for Form 1. Nevertheless, in the commercial sample there is evidence of a Form 1
impurity and/or a phase transition with an additional endothermic peak in agreement with
melting of Form 1. This can be explained by the presence of small amounts Form 1, as indicated
by small peaks in the XRPD pattern (Fig. 7b – indicated by arrows). The small amounts of Form
1
might also act as crystallization seeds from the melt of Form 2. The DSC results in association
with stability studies involving solvent equilibration of mixtures of both polymorphs (Table 4)
indicate that Form 1 is the stable form from 25 up to 150 °C.
ACS Paragon Plus Environment
26

Page 27 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Table 4. Stability of polymorphs formed by slurry methods in different solvents.
Initial Composition Condition
Final
Composition
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
Mixture 1:1 (w/w) Slurry in methanol at 25°C for 1 Form 1
(Form 1 and Form 2) week.
Mixture 1:1 (w/w) Suspension in toluene at 70 °C Form 1
Form 1 and Form 2
for 48 hours.
Mixture 1:1 (w/w) Slurry in dichloromethane at 25 Solvate B
Form 1 and Form 2
Form 1
°C for 1 week.
Suspension heated to reflux in Solvate B
dichloromethane
(39.6
°C)
according to procedure described
in patent (14) to obtain Form 2.
Form 1
Slurry in dichloromethane at 25 Solvate B
°C for 24 hours.
The dichloromethane solvate also has a structure similar to that of Form 2 as evidenced by its
XRPD and DSC curve (Figs 7 and 8d). After loss of solvent this structure changes into Form 2.
An 1ꢀoctanol solvate was obtained via a slurry method and presents a XRPD pattern different
from any of the other forms. DSC curves indicate three events in the range of 45ꢀ150 °C: the first
transition, a endothermic event, is assigned to solvent loss in the range of 45ꢀ114 °C, followed by
an exothermic event assigned to a crystallization of Form 1 at 113 °C with subsequent melting
between 123ꢀ140 °C. It is noteworthy that Form Y could not be obtained in any of our screening
2
1
experiments, or even by reproducing the literature procedure.
ACS Paragon Plus Environment
27

Crystal Growth & Design
Page 28 of 45
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSION
The immersion of a copper wire into a methanol/water solution of lapatinib free base was able
to induce the nucleation and growth of single crystals with two different morphologies, plates
and thin needles. Structural refinement indicates that the needles are a 3.4 hydrate with a channel
structure and the plates correspond to an anhydrous form that corresponds to previously reported
forms 1 and X. A new acidꢀbase reaction between lapatinib ditosylate and sodium methoxide in
methanol or heptane can successfully be used to produce this last form avoiding the presence of
the hydrate. Form 2 can selectively be formed by desolvation of the hydrate, or from several
other solvates with structures similar to the hydrate. The combination of thermal analysis,
solvent equilibrium studies and Xꢀray diffraction allowed the identification of new solvates and
the establishment of Form 1 as the most stable form in a wide range of temperature.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
28

Page 29 of 45
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
ASSOCIATED CONTENT
Supporting Information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table S1. Comparison of Xꢀray diffraction data of lapatinib free base anhydrous polymorphs
1
8,19
reported in patents
and calculated pattern from single crystal data (anhydrous plates). This
material is available free of charge via the Internet at http://pubs.acs.org.
Accession Codes
Complete crystallographic data, in CIF format, have been deposited with the Cambridge
Crystallographic Data Centre. CCDC 1501183ꢀ1501184 contains the supplementary
crystallographic data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
AUTHOR INFORMATION
Corresponding Author
*Email:gabriel.lbaraujo@gmail.com
Funding Sources
This project was funded by the São Paulo Research Foundation (FAPESP) [grant#2015/15456ꢀ5
and grant#2015/05685ꢀ7, SP, Brazil].
Notes
The authors declare no competing financial interest.
ACS Paragon Plus Environment
29