Understanding API–Polymer Proximities in Amorphous Stabilized

Aug 22, 2014 - Anuji Abraham and George Crull. Drug Product Science and Technology, Material Science Division, Bristol-Myers Squibb, New Brunswick, ...
0 downloads 0 Views 1MB Size
Subscriber access provided by UNIV OF TASMANIA

Brief Article

Understanding API-Polymer Proximities in Amorphous Stabilized Composite Drug products using Fluorine-Carbon 2D HETCOR Solid-state NMR Anuji Abraham, and George Crull Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/mp400629j • Publication Date (Web): 22 Aug 2014 Downloaded from http://pubs.acs.org on September 2, 2014

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.

Molecular Pharmaceutics 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 20

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

Molecular Pharmaceutics

1

Understanding API-Polymer Proximities in Amorphous Stabilized Composite Drug products using Fluorine-Carbon 2D HETCOR Solid-state NMR

Anuji Abraham and George Crull

Bristol-Myers Squibb, Material Science Division, Drug Product Science and Technology, New Brunswick, New Jersey 08903

Abstract

A simple and robust method for obtaining fluorine-carbon proximities was established using a

19

F-13C

hetero-nuclear correlation (HETCOR) two-dimensional (2D) solid-state nuclear magnetic resonance (ssNMR) experiment under magic-angle spinning (MAS). The method was applied to study a crystalline active pharmaceutical ingredient (API), Avagacestat containing two types of fluorine atoms and its APIpolymer composite drug product. These results provide insight into the molecular structure, aid with assigning the carbon resonances, and probe API–polymer proximities in amorphous spray dried dispersions (SDD) This method has an advantage over the commonly used 1H-13C HETCOR because of the large chemical shift dispersion in the fluorine dimension. In the present study, fluorine-carbon distances up to 8 Å were probed, giving insight into the API structure, crystal packing and assignments. Most importantly, the study demonstrates a method for probing an intimate molecular level contact between an amorphous API and a polymer in an SDD, giving insights into molecular association and understanding of the role of the polymer in API stability (such as recrystallization, degradation etc) in such novel composite drug products.

Keywords

NMR, Solid-state NMR,

19

F-13C HETCOR, Fluorine-Carbon Proximities, SDD, API–polymer

interactions, ssNMR, spray dried dispersion, ASD, amorphous solid dispersions

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 2 of 20

2

Introduction Hetero-nuclear correlation (HETCOR)1-3 is a well established NMR method widely used in solution and solid-state nuclear magnetic resonance (ssNMR) for studying organic, inorganic and biological materials. The commonly used solid-state 1H-13C HETCOR3 is usually limited in resolution by strong proton-proton homonuclear interactions at moderate magic angle spinning (MAS) speeds and relies heavily on efficient homonuclear decoupling. 19F is an alternative nucleus with high sensitivity due to its high natural abundance and gyromagnetic ratio, allowing long range distances to be probed, but is typically in low enough concentrations so that homonuclear couplings are less severe. Fluorine is present in many active pharmaceutical ingredients (API), drug product intermediates and composite materials containing polymer and API such as spray dried dispersions (SDD) or hot melt extrudates.4-8 These amorphous solid dispersions (ASD) are emerging approaches in the drug development industry, and have become an important tool in enabling development of poorly soluble API. However, understanding of the API-polymer interaction and the role of the polymer in stabilizing the amorphous API in such ASD or hot melt extrudates are lacking. While some polymers tend to stabilize the amorphous API, others do not inhibit the crystallization of amorphous API. The amorphous nature of these materials limits the utility of X-ray powder diffraction and differential scanning calorimetric techniques, and makes structural insight from NMR measurements particularly invaluable. The high sensitivity and large chemical shift dispersion of 19F makes it well suited for studying the conformation of such compounds and ligand binding sites in proteins.9-11 In the past the use of 19F-NMR was limited by demanding hardware requirements imposed by the proximity of the

19

F and 1H resonance

frequencies. However, with the modern spectrometers these experiments are accessible. These experiments were performed on spectrometers configured for at least triple resonance, allowing concurrent high power decoupling of both protons and fluorine atoms. The utility of polarization (CP) MAS experiments has been demonstrated12, 13 including a selective

19

19

F-13C cross

F-13C CPMAS

experiment for compounds containing multiple inequivalent fluorine atoms.14 Spiess et al. demonstrated the application of F-C HETCOR for structural assignments in polymers.15 This work was extended by Schmidt-Rohr with studies of Nafion.16 Recently Dybowski and coworkers have applied F-C HETCOR to help assign a polymorph of Atorvastatin.17

ACS Paragon Plus Environment

Page 3 of 20

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

Molecular Pharmaceutics

3

Characterization of an API in the solid state is challenging and the peak assignments are usually tentatively made by using routine 1H-13C/15N CPMAS, 1H-13C cross polarization - polarization inversion (CPPI) and

13

C high power decoupling (HPDEC) NMR. Definitive assignments require selective

labeling, 2D 13C incredible natural abundance double quantum transition experiment (INADEQUATE)18 or 1H-13C HETCOR type experiments.

13

C INADEQUATE at natural abundance is found to be time

consuming (usually many days to obtain a spectrum) while 1H-13C HETCOR relies heavily on efficient homonuclear decoupling or ultra high MAS speeds. A 2D

19

F-13C HETCOR NMR experiment offers

significant advantages in the ease of set up, time saving and sensitivity. The only limitation of this experiment which has been observed was while studying compounds with long

19

F relaxation times,

which can be overcome by implementing a double CP (1H-19F-13C) experiment. The present study demonstrates the use of

19

F-13C HETCOR ssNMR under MAS to provide insight into the structure,

crystal packing and carbon assignments of a crystalline API. The study also demonstrates a method for probing API–polymer proximities in an amorphous SDD drug product providing evidence for a molecular level intimate contact between the API and the polymer. This evidence of molecular association gives insight into the role of polymer in retaining the amorphous API without allowing recrystallization.

Experimental The procedure of the preparation of Avagacestat, the API has been described by Gillman et al..19 The SDD was prepared in the course of experimental drug development in collaboration with Bend Research, Bend, OR. The SDD is comprised of 25% amorphous API and 75% amorphous Hydroxylpropyl methylcellulose- acetate succinate (HPMC-AS) M grade (from Ashland). Amorphous API control sample was prepared using a Bend mini spray drier from an acetone solution. XRPD data was recorded to confirm the API polymorph consistent with the single-crystal structure and to confirm the amorphous nature of the amorphous API control and SDD samples prepared.

Most of the solid-state NMR experiments were conducted on a Bruker AV III instrument operating at a proton frequency of 500.01 MHz (11.7 T) using a 4 mm quad resonance (CHNF) solid-state probe-head. Data was collected using pulse sequences from the Bruker library (Bruker-Biospin, Billerica, MA). Approximately 80 mg samples were spun at 13 kHz in a 4 mm ZrO2 for 1H-13C CPMAS and 1H-13C CPPI MAS20 experiments. Both the experiments used 4 ms cross polarization contact time, ramped from

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 4 of 20

4

50 to 100% on proton and a relaxation delay of 20 s. 100 µs polarization inversion time was used for CPPI experiment. Proton decoupling was applied using a TPPM sequence21 with an 8 µs pulse (62.5 kHz nominal band width) for both CPMAS and CPPI experiments.

Samples were spun at 10 kHz for

19

F HPDEC,

19

F-13C CPMAS and

19

F-13C HETCOR experiments.

Fluorine-carbon CP contact times of 4, 6 and 8 ms (ramped from 50 to 100% on fluorine channel) were used for the 2D HETCOR experiments. Proton decoupling was applied using a TPPM sequence with an 8 µs pulse (62.5 kHz nominal band width) for all the experiments. Proton decoupling using a TPPM sequence was also applied in the indirect fluorine dimension of the 2D HETCOR spectrum. Fluorine decoupling was applied using a Pidec sequence22 with a 4 µs pulse (62.5 kHz nominal band width) for both HETCOR and CPMAS experiments. 2D HETCOR spectra resulted from averaging 1024 transients for each of 128 t1 increments with a relaxation delay of 1 s between consecutive scans. The T1 relaxation times of proton and fluorine atoms were measured using a saturation recovery experiment, and were analyzed using Topsin 2.1 (Bruker-Biospin, Billerica, MA). Proton T1ρ experiment was done at a radio frequency field of 40 W. 2D HETCOR and variable temperature 19F HPDEC experiments on the amorphous API control sample were conducted on a Bruker AV III instrument operating at a proton frequency of 400.13 MHz (9.4 T) using a 4 mm triple resonance CFH solid-state probe-head. Samples were spun at 13.02 kHz. Proton decoupling was applied using a TPPM sequence with a 6.8 µs pulse (73.5 kHz nominal band width) for all the experiments. Proton decoupling using a TPPM sequence was also applied in the indirect fluorine dimension of the 2D HETCOR spectrum. Fluorine decoupling was applied using a Pidec sequence22 with a 4.4 µs pulse (56.8 kHz nominal band width) for the HETCOR experiment. Fluorine-carbon CP contact times of 2ms (ramped from 50 to 100% on fluorine channel) was used for the 2D HETCOR experiment. 2D HETCOR spectra obtained from averaging 1024 transients for each of 128 t1 increments with a relaxation delay of 1 s between consecutive scans.

The

13

C spectra were referenced indirectly to TMS using 3-methylglutaric acid23 and

19

F spectra were

referenced to external poly tetrafluoroethylene24 (PTFE) at -123.2 ppm. The temperature was maintained at a nominal 280 K (nominal) using a cooled stream of nitrogen gas for all the experiments except the 19

F variable temperature experiments, where the temperature was varied from 260K to 390K (nominal).

ACS Paragon Plus Environment

Page 5 of 20

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

Molecular Pharmaceutics

5

Results and Discussion Figure 1 shows the 1H-13C CPMAS with fluorine decoupling (top), 1H-13C CPMAS without fluorine decoupling (middle) and 1H-13C CPPI (bottom) MAS NMR spectra of the API molecule, Avagacestat. The molecular structure of the API is also shown in Figure 1 (inset). 1H-13C CPMAS spectrum with fluorine decoupling shows the tentative assignment of the molecule. In the CPMAS spectra the splitting (~400 Hz) of the C26 resonance observed is due to the residual dipolar splitting (second-order quadrupolar-dipolar cross term) from the adjacent quadrupolar chlorine-35 atom which was confirmed by doing an experiment at a lower field (9.4 T). The C8 resonance shows C-F J-coupling of about 245 Hz in the middle spectrum, which disappears in the top spectrum due to the fluorine decoupling applied during the experiment. 1H-13C CPPI MAS spectrum distinguishes CH3, CH2, CH and tertiary carbon groups in the structure. In the CPPI spectrum, CH2 group gives inverted peaks while CH gives no peak and C and CH3 groups give positive peaks. Figure 2 (top) shows the 19F HPDEC MAS spectrum of the crystalline API molecule. The spectrum shows the two types of fluorine atoms consistent with the structure. The resonance frequency for 19F was set in between CF3 and CF resonances. The peak at -65.7 ppm is assigned to the CF3 group and the broad peak near -117 ppm to CF. A splitting of the CF peak at -117 ppm is either due to a dynamic disorder in the aryl ring or due to a slight conformational change in aryl fluorine atom. Variable temperature experiments were conducted to a maximum of 390 K and the aryl fluorine resonances moved towards each other leading to a near coalescence, at the limit of the probe and molecular degradation. Figure 2 (middle) shows the spectrum at higher temperature (390 K) with minimal splitting. T1 relaxation times were measured as 310 and 350 ms for CF3 and CF groups, respectively. The rapid relaxation of CF could be due to the dipolar coupling of CF to the fast rotating CF3 in the molecule. Figure 2 (bottom) shows the 19F HPDEC MAS spectrum of the amorphous API. Figure 3 shows the 19F-13C HETCOR 2D NMR spectra of the API molecule under investigation at two fluorine-carbon CP contact times 6 ms (left) and 8 ms (right). The resonance frequency for 19F (indirect dimension) was set in the middle of CF3 and CF resonances. The spectra show a spatial correlation from two types of fluorine atoms to various carbon atoms in the crystal structure. Figure 4 shows the 3D crystal structure of the API with distances to various carbon atoms labeled from a radial coordinate centered on CF3. Experiments with shorter CP contact time probe the closest distances from CF3. The

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 6 of 20

6

red circle (Figure 4) depicts probe range of CF3 for the shorter contact times which is approximately 5.5Å. The black circle shows the CF3 probe range for longer contact time of 8 ms (~8Å). At 6 ms F-C cross polarization contact time, the fluorine atoms of the CF3 group at -65.7 ppm show a correlation to CF3 ([C20], 127.5 ppm) and to the adjacent carbon chain (CH2-CH2-CH) (~ 3.8 Å) at 30.5 [C19], 20 [C18] and 59.7 [C17] ppm, respectively. The rest of the molecule within the 5.5 Å probe distance doesn’t have carbon atoms, or in the case of the carbonyl at C29, is beyond 5.5 Å and so is not observed. Correlation to the CH2 ([C13], 42 ppm) inside of 5.5 Å is also observed but its peak intensity is weak, which indicates some mobility (leading to weaker dipolar couplings) due to its flexible nature. Interestingly, the CF3 group at -65.7 ppm also correlates to CF [C8] and a CH [C7], which is adjacent to the CF [C8] in the aromatic ring and not to the CH [C11] which is opposite to the CF [C8] in the aromatic ring, which suggests that it is a correlation to the neighboring molecule and not within the molecule (Figure 4). The CF3-CH ([C11], opposite of CF in the aromatic ring) distance is 8.6 Å while CF3-CH ([C7], adjacent to CF in the aromatic ring) distance is 8.9 Å. CF3 also correlates to the carbon atoms in the chlorinated aromatic ring of both the same molecule and the neighboring molecule (both are within 5.5 Å distance from CF3). The fluorine atom of the CF group at -117 ppm correlates only to CF [C8] and to the adjacent two carbon atoms ([C7] and [C9], ~ 1.4 Å), though correlations to longer distances appeared with increased signal to noise ratio. C8 resonance shows no C-F J-coupling, unlike observed in 1H-13C CPMAS spectrum, due to the fluorine decoupling applied during the acquisition period. The multiple fluorine atoms in the CF3 group, compared to the single environment in CF, allow the longer distances to be probed with same number of scans. Fluorine-19 T1 relaxation times for CF3 and CF groups were found to be 310 and 350 ms, respectively. A complete spatial correlation from two types of fluorine atoms to various carbon atoms in the crystal structure is shown, providing insight into structural properties, crystal packing and aiding in assignment. Figure 5 shows the 19F-13C HETCOR 2D NMR spectrum of the SDD sample under investigation at a CP contact time of 2 ms. Only CF3 group at -65.7 ppm is observed in the 19F indirect dimension as the drug loading is only around 25% of the total weight. Significantly more scans would likely be required to observe the aromatic CF fluorine. Figure 6 shows a comparison of 1H-13C CPMAS spectrum (top) of the polymer HPMC-AS with the CPMAS spectra of SDD, crystalline API and amorphous API. The 2D HETCOR spectrum shows a correlation of the fluorine atoms of the CF3 group of the API to the carbon atoms of CF3 group of API ([C20], 127.5 ppm) and also to the carbonyl groups of the polymer HPMCAS (167 ppm). A correlation of the fluorine atom of CF3 to the adjacent CH2 group ([C19], 30.5 ppm) is

ACS Paragon Plus Environment

Page 7 of 20

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

Molecular Pharmaceutics

7

also observed in Figure 5. The correlation observed to the carbonyl groups of the polymer HPMC-AS suggests that the polymer may be hydrogen bonded to the NH2 group of the API which is spatially close to the CF3 group. In this way the carbonyl groups of the HPMC-AS can be spatially close to the CF3 group of the API. The fact that this inter-molecular API-polymer correlation is seen at a shorter CP contact time suggests that it is a correlation to the carbonyl of the polymer and not to the carbonyl of the API itself, as the crystalline pure API did not show such a correlation within its own carbonyl group (intra-molecular) at a short CP contact time. Although one could argue that the amorphous API molecule can have a different spatial arrangement when compared to the crystalline API, bringing the carbonyl group closer to the CF3 group leading to a stronger intra-molecular correlation, which can be observed at shorter contact times. This can be reasoned as this API-polymer correlation was not seen (spectrum not shown) when HPMC-AS polymer was changed within M grade with low succinoyl and acetyl content. Figure 7 shows the 19F-13C HETCOR 2D NMR spectrum of the amorphous API. The fluorine atoms of the CF3 group at -65.7 ppm show a correlation only to C20 and not to the carbonyl region, and the fluorine atom of the CF group at ~ -117 ppm shows a correlation to C8. This observation confirms the correlation of the fluorine atoms of the CF3 group of the API to the carbonyl groups of the polymer HPMC-AS (167 ppm) and not to the carbonyl group of the API, in the SDD sample under investigation (Figure 5). The formation of a hydrogen bond between the carbonyl group of the polymer and the NH2 group of the API might hold the API with the polymer, restricting it from ordering by itself and forming a crystal lattice at long shelf lives. Therefore, this evidence of having a molecular level intimate contact between the API and the polymer gives insight into molecular association and therefore, the role of polymer in retaining the amorphous API without leading to its recrystallization. These data also support the SDD being a solid solution rather than a mixture. Proton T1 and T1ρ experiments were also done on the SDD sample and resulted in single T1 (3.6 s) and T1ρ (7.9 ms) values for both which again is consistent with a homogeneous drug-polymer solid solution.

Conclusions

A simple and robust

19

F-13C HETCOR experiment in solid-state was demonstrated for measuring

fluorine-carbon distances, and therefore probing API-polymer proximities. The method has been applied to a crystalline API Avagacestat probing fluorine-carbon distances up to 8 Å and giving insight into the API structure, crystal packing and assignments. Clear evidence for a molecular level intimate contact between API and the polymer was observed in amorphous spray dried dispersion, giving insights into

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 8 of 20

8

molecular association and therefore, understanding the role of polymer in API stability. The APIpolymer distance was estimated to be less than 5 Å. The method shows a great advantage over the usual 1

H-13C HETCOR, having large chemical shift dispersion in the fluorine-19 dimension.

Acknowledgements:

Authors would like to thank A. Patel for providing the samples and for the helpful discussions, D. Wu for the initial crystal structure determination and J. Struppe for the helpful discussions on implementing 1D

19

F-13C CPMAS experiments. Authors would also like to thank V. Rao and M. A Galella for

insightful discussions, and D. McNamara for helping to prepare the amorphous API control sample and for helpful discussions.

ACS Paragon Plus Environment

Page 9 of 20

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

Molecular Pharmaceutics

9

Figure 1: 1H-13C CPMAS with and without fluorine decoupling and 1H-13C CPPI MAS NMR spectra of the API molecule, Avagacesta. The 1H-13C CPMAS spectrum with fluorine decoupling shows the tentative assignments. 1H-13C CPPI MAS spectrum gives the assignment of CH2 (inverted), CH (no signal) and C and CH3 groups (positive). Inset: The molecular structure of the API with atom numbering.

Figure 2:

19

F HPDEC MAS NMR spectra of the crystalline API (top), crystalline API at higher

temperature (390 K) (middle) and amorphous API (bottom) samples under investigation. The spectrum shows the two types of fluorine atoms in the structure. The resonances at -65.7 and -117 ppm are assigned to the CF3 and CF groups, respectively and the rest of the peaks are the spinning side bands. A coalescence of the split is observed at higher temperature. T1 relaxation times were determined to be 310 and 350 ms for CF3 and CF groups, respectively. Figure 3: 19F-13C HETCOR 2D NMR spectra of the API molecule under investigation at two fluorinecarbon CP contact times 6 ms (left) and 8 ms (right). The 19F (indirect dimension) peaks at -65.7 and 117 ppm are assigned to the CF3 and CF groups, respectively and the rest of the peaks are the spinning side bands. Rotor synchronization was not possible due to hardware limitation. The spectra show a spatial correlation from two types of fluorine atoms to various carbon atoms in the structure (shown in the boxes). Each of the 2D spectra resulted from averaging 1024 transients for each of 128 t1 increments. Figure 4: 3D crystal structure of the API (8 molecules) with distances to various carbon atoms labeled from a radial coordinate centered on CF3. Experiments with shorter CP contact times probe the closest distances from CF3 (~ 5.5 Å), shown with red circle, while longer contact times probe longer distances (~ 8 Å), shown with black circle. yellow: fluorine, green: chlorine, red: oxygen, blue: nitrogen and gold: sulfur. Figure 5: 19F-13C HETCOR 2D NMR spectrum of the amorphous SDD sample under investigation at a fluorine-carbon CP contact time of 2 ms. The 19F (indirect dimension) peak at -65.7 ppm is assigned to the CF3 group, and the rest of the peaks are the spinning side bands (SSB). The spectrum shows a spatial correlation of the fluorine atoms of the CF3 group of the API to the carbon of the CF3 group of API and

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 10 of 20

10

also to the carbonyl groups of the polymer, HPMC-AS. The 2D spectrum resulted from averaging 2048 transients for each of 128 t1 increments. Figure 6: 1H-13C CPMAS NMR spectrum of the polymer HPMC-AS in comparison to the spectra of SDD, crystalline API and amorphous API. The spectrum highlights the carbonyl peak which is interacting with the API in the amorphous SDD sample.

Figure 7:

19

F-13C HETCOR 2D NMR spectrum of the amorphous API control sample at a fluorine-

carbon CP contact time of 2 ms. The 19F (indirect dimension) peaks at -65.7 and -117 ppm are assigned to the CF3 and CF groups, respectively and the rest of the peaks are the spinning side bands. The spectrum shows a correlation of the fluorine atoms of the CF3 group at -65.7 ppm to C20 and the fluorine atom of the CF group at -117 ppm to C8. The 2D spectrum resulted from averaging 2048 transients for each of 128 t1 increments.

ACS Paragon Plus Environment

Page 11 of 20

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

Molecular Pharmaceutics

11

References

1. Holland, G. P.; Alam, T. M. Multi-dimensional 1H-13C HETCOR and FSLG-HETCOR NMR study of sphingomyelin bilayers containing cholesterol in the gel and liquid crystalline states. J Magn Reson 2006, 181, (2), 316-326. 2. Abraham, A.; Ilott, A. J.; Miller, J.; Gullion, T. (1)H MAS NMR study of cysteine-coated gold nanoparticles. J Phys Chem B 2012, 116, (27), 7771-7775. 3. Vogt, F. G.; Katrincic, L. M.; Long, S. T.; Mueller, R. L.; Carlton, R. A.; Sun, Y. T.; Johnson, M. N.; Copley, R. C.; Light, M. E. Enantiotropically-related polymorphs of {4-(4-chloro-3-fluorophenyl)-2-[4-(methyloxy)phenyl]1,3-thiazol-5-yl} acetic acid: crystal structures and multinuclear solid-state NMR. J Pharm Sci 2008, 97, (11), 4756-4782. 4. Yoshioka, M.; Hancock, B. C.; Zografi, G. Inhibition of indomethacin crystallization in poly(vinylpyrrolidone) coprecipitates. J Pharm Sci 1995, 84, (8), 983-986. 5. Van den Mooter, G.; Wuyts, M.; Blaton, N.; Busson, R.; Grobet, P.; Augustijns, P.; Kinget, R. Physical stabilisation of amorphous ketoconazole in solid dispersions with polyvinylpyrrolidone K25. European Journal of Pharmaceutical Sciences 2001, 12, (3), 261-269. 6. Aso, Y.; Yoshioka, S.; Kojima, S. Molecular mobility-based estimation of the crystallization rates of amorphous nifedipine and phenobarbital in poly(vinylpyrrolidone) solid dispersions. J Pharm Sci 2004, 93, (2), 384-391. 7. Graeser, K. A.; Patterson, J. E.; Zeitler, J. A.; Gordon, K. C.; Rades, T. Correlating thermodynamic and kinetic parameters with amorphous stability. European Journal of Pharmaceutical Sciences 2009, 37, (3–4), 492498. 8. Hancock, B. C.; Shamblin, S. L.; Zografi, G. Molecular mobility of amorphous pharmaceutical solids below their glass transition temperatures. Pharm Res 1995, 12, (6), 799-806. 9. Goetz, J. M.; Poliks, B.; Studelska, D. R.; Fischer, M.; Kegelbrey, K.; Bacher, A.; Cushman, M.; Schaefer, J. Investigation of the binding of fluorolumazines to the 1-MDa capsid of lumazine synthase by 15N{19F} REDOR NMR. J. Am. Chem. Soc. 1999, 121, 7500–7508. 10. Metha, A. K.; Studelska, D. R.; Fischer, M.; Giessauf, A.; Kemter, K.; Bacher, A.; Cushman, M. Investigation of the binding of epimer A of the covalent hydrate of 6,7-bis(trifluoromethyl)-8-D-ribityllumazine to recombinant F22W Bacillus subtilis lumazine synthase mutant by 15N{19F} REDOR NMR. J. Org. Chem. 2002, 67, 2087–2092. 11. McDowell, L. M.; McCarrick, M. A.; Studelska, D. R.; Guilford, W. J.; Arnaiz, D.; Dallas, J. L.; Light, D. R.; Whitlow, M.; Schaefer, J. Conformations of Trypsin-Bound Amidine Inhibitors of Blood Coagulant Factor Xa by Double REDOR NMR and MD Simulations. J. Med. Chem. 1999, 42, 3910-3918. 12. Hagaman, E. W.; Burns, J. H. The determination of local structure in organofluorides using fluorine-19 carbon-13 dipolar coupling. Fuel 1993, 72, (8), 1239-1243. 13. Cherry, B. 19F Solid State NMR Workshop. 48th Rocky Mountain Conference, Breckenridge, CO 2006. 14. Crull, G. B.; Grosso, J.; Scaringe, R. Selective F-19 to C-13 Cross Polarization, a Novel Tool for Structure Elucidation in Pharmaceuticals. 51st Experimental NMR Conference, Daytona Beach, FL 2010. 15. Lee, Y. J.; Clark, C. G.; Graf, R.; Wagner, M.; Müllen, K.; Spiess, H. W. Solid-State Organization of Semifluorinated Alkanes Probed by 19F MAS NMR Spectroscopy. The Journal of Physical Chemistry B 2009, 113, (5), 1360-1366. 16. Chen, Q.; Schmidt-Rohr, K. 19F and 13C NMR Signal Assignment and Analysis in a Perfluorinated Ionomer (Nafion) by Two-Dimensional Solid-State NMR. Macromolecules 2004, 37, (16), 5995-6003. 17. Wang, W. D.; Gao, X.; Strohmeier, M.; Wang, W.; Bai, S.; Dybowski, C. Solid-state NMR studies of form I of atorvastatin calcium. J Phys Chem B 2012, 116, (11), 3641-3649. 18. Buddrus, J.; Lambert, J. Connectivities in molecules by INADEQUATE: recent developments. Magnetic Resonance in Chemistry 2002, 40, (1), 3-23. ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 12 of 20

12 19. Gillman, K. W.; Starrett, J. E.; Parker, M. F.; Xie, K.; Bronson, J. J.; Marcin, L. R.; McElhone, K. E.; Bergstrom, C. P.; Mate, R. A.; Williams, R.; Meredith, J. E.; Burton, C. R.; Barten, D. M.; Toyn, J. H.; Roberts, S. B.; Lentz, K. A.; Houston, J. G.; Zaczek, R.; Albright, C. F.; Decicco, C. P.; Macor, J. E.; Olson, R. E. Discovery and Evaluation of BMS-708163, a Potent, Selective and Orally Bioavailable γ-Secretase Inhibitor. ACS Medicinal Chemistry Letters 2010, 1, (3), 120-124. 20. Wu, X. L.; Zilm, K. W. Complete Spectral Editing in CPMAS NMR. Journal of Magnetic Resonance, Series A 1993, 102, (2), 205-213. 21. Bennett, A. E.; Rienstra, C. M.; Auger, M.; Lakshmi, K. V.; Griffin, R. G. Heteronuclear decoupling in rotating solids. The Journal of Chemical Physics 1995, 103, (16), 6951-6958. 22. Liu, S. F.; Schmidt-Rohr, K. High-Resolution Solid-State 13C NMR of Fluoropolymers. Macromolecules 2001, 34, (24), 8416-8418. 23. Barich, D. H.; Gorman, E. M.; Zell, M. T.; Munson, E. J. 3-Methylglutaric acid as a 13C solid-state NMR standard. Solid state nuclear magnetic resonance 2006, 30, (3-4), 125-129. 24. Katoh, E.; Sugimoto, H.; Kita, Y.; Ando, I. Structures of polytetrafluoroethylene oligomers as studied by high-resolution solid-state 19F NMR and their properties. Journal of MOLECULAR STRUCTURE 1995, 355, 21-26.

ACS Paragon Plus Environment

Page 13 of 20

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

Molecular Pharmaceutics

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

Page 14 of 20

CF3

CF

-40

-60

-80

-100

ACS Paragon Plus Environment

-120

ppm

Page 15 of 20

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

Molecular Pharmaceutics

CF

CF3

ACS Paragon Plus Environment

Molecular Pharmaceutics

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 ACS Paragon Plus Environment

Page 16 of 20

Page 17 of 20

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

Molecular Pharmaceutics

HPMC-AS CO-

CF3

SSB CF3

ACS Paragon Plus Environment

Molecular Pharmaceutics

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

ACS Paragon Plus Environment

Page 18 of 20

Page 19 of 20

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

Molecular Pharmaceutics

CF3 CF

ppm -140 -120

CF

-100 -80 CF3

-60 -40 150

100

50

ACS Paragon Plus Environment

ppm

Molecular Pharmaceutics

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

OH

Page 20 of 20

HPMC-AS CH3

HO

O

HO

HO

O

O

OH

R

F

F

O

O

O

CH3

HO

API

F

Cl

~3A

CH3 F F

N

O

F

~3A

Cl

CH3

O

O OH F

O

O

O

O

CH3

API

F

HO

O O

O

O H2N

R O

O

O O

OH

O O

O

HO

OH

API F Cl

O

S

H2N

O

O

N

O

S H2N

O

N

O

S O

O

~3A

O O

R

N

F N

CH3 O

O

O

HO

N

F H3C O

H3C O

O O

O O

O HO HO

HO

O CH3

N

F O

H3C OH

HO

OH O

N

O

O

O

O

O

OH

N

HO

O

O

O O HO

O OH

HPMC-AS

ACS Paragon Plus Environment

O

O CH3

R