Subscriber access provided by GRIFFITH UNIVERSITY
Article
Drug–excipient interactions in the solid state: The role of different stress factors CORINNA GRESSL, Michael Brunsteiner, Adrian Davis, Margaret Landis, Klimentina Pencheva, Garry Scrivens, Gregory W. Sluggett, Geoffrey P. F. Wood, Heidrun Gruber-Woelfler, Johannes G. Khinast, and Amrit Paudel Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b00677 • Publication Date (Web): 23 Oct 2017 Downloaded from http://pubs.acs.org on October 25, 2017
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 17
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
Drug–excipient interactions in the solid state: The role of different stress factors Corinna Gressl†, Michael Brunsteiner†, Adrian Davis‡, Margaret Landis§, Klimentina Pencheva‡, Garry Scrivens‡, Gregory W. Sluggett§, Geoffrey P. F. Wood§, Heidrun Gruber-Woelfler†,∥, Johannes G. Khinast†,∥, Amrit Paudel*† †
Research Center Pharmaceutical Engineering, Graz, Austria Pfizer Worldwide R&D, Sandwich, Kent, UK § Pfizer Worldwide R&D, Groton, CT, USA ∥University of Technology, Institute of Process and Particle Engineering, Graz Austria ‡
KEYWORDS crystal surface structures, drug excipient interactions, molecular modelling, polymorphism, solid state chemistry, water activity ABSTRACT: Understanding properties and mechanisms that govern drug degradation in the solid state is of high importance to ensure drug stability and safety of solid dosage forms. In this study, we attempt to understand drug–excipient interactions in the solid state using both theoretical and experimental approaches. The model active pharmaceutical ingredients (APIs) under study are Carvedilol (CAR) and Codeine Phosphate (COP), which are known to undergo esterification with citric acid (CA) in the solid state. Starting from the crystal structures of two different polymorphs of each compound, we calculated the exposure and accessibility of reactive hydroxyl groups for a number of relevant crystal surfaces, as well as descriptors that could be associated with surface stabilities using molecular simulations. Accelerated degradation experiments at elevated temperature and controlled humidity were conducted to assess the propensity of different solid forms of the model APIs to undergo chemical reactions with anhydrous CA or CA monohydrate. In addition, for CAR, we studied the solid state degradation at varying humidity levels and also under mechano activation. Regarding the relative degradation propensities, we found that variations in the exposure and accessibility of molecules on the crystal surface play a minor role compared to the impact of molecular mobility due to different levels of moisture. We further studied drug–excipient interactions under mechano-activation (co-milling of API and CA) and found that the reaction proceeded even faster than in physical powder mixtures kept at accelerated storage conditions.
1 2
Graphical Abstract
1
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
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
Page 2 of 17
INTRODUCTION The chemical stability of drugs is one of their most critical quality attributes as the safety and the efficacy of the drug must be guaranteed throughout the shelf-life of the product. Forced degradation studies of APIs are often conducted in solution in order to identify potential chemical degradation pathways; however, the degradation pathways observed in the solid state can be fundamentally different, especially in the presence of excipients (e.g., in tablets). Thus, there is a need to better understand chemical stability in the solid state in the presence of excipients, in order to prevent potential quality issues and ensure product shelf-life. In general, the susceptibility of an API towards reactions with excipients is dependent on the existence of a possible chemical reaction pathway and upon the energy of the associated transition state, which acts as an energy barrier for the reaction. Depending on the electronic structure of the API molecule and the reaction mechanism, the height of this energy barrier can vary widely, and thus, influences how fast an API degrades. This is also true for reactions in the solid state. However, in the solid state, structural properties of the API crystals can have a prominent effect on the stability of the drug product. In Figure 1 some of these structural properties are illustrated. Defects and grain boundaries destabilize the API crystals, and thus, are likely to enhance reactions with the surrounding excipient. Also, the orientation of the API, exposing or hiding reactive groups of the API molecule, could possibly influence the reactivity. Furthermore, the molecular mobility of the API and the diffusion of reactants, such as O2, through the API crystal affect the reaction rate. In previous studies, it was shown that the crystal surface structures of different polymorphs of an API impact the degree of chemical degradation with ammonia gas.1,2 Other authors studied the reactivity of different polymorphs of tertbutylacetate towards photooxidation. They found that after desolvation of ethanol, the reactive polymorph possesses a void in the crystal allowing oxygen to penetrate the crystal, thus inducing the reaction.3 These studies demonstrate the importance of the crystal structure on chemical stability of APIs. In other studies, it was found that desolvated crystals of dialuric acid possess voids, but are still inert to oxidation under dry conditions.4 The authors hypothesized that solid–gas reactions actually might take place in a moisture layer on the solid material, thus highlighting the importance of humidity on reactivity. Losev et al.5 studied solid state reactions via co-grinding and found that trace amounts of water are needed for "solid state" reactions to take place. These studies provided important insights. However, studies on this topic still remain rare at this point.
29 30 31 32 33 34 35 36 37 38 39 40 41 42
Figure 1. Factors governing drug–excipient interactions in the solid state.
In the present study, we aim to improve our understanding of drug–excipient interactions in solid dosage forms. In particular, we want to investigate the role of the crystal structure versus the role of moisture and mechanical stress on reactivity and stability of drugs in the solid state. We selected Carvedilol (CAR) and Codeine phosphate (COP) for our study, as these APIs have been reported to readily esterify with citric acid (CA), a common excipient.6,7 Three different esters can be formed. Two esters are formed through the reaction with the terminal carboxylic acid group of CA, which are diastereomeres, and one ester is formed through the reaction with the central carboxylic acid group of CA (see Figure 2). In the solid state, chemical reactions between APIs and excipients can be expected to take place on the crystal surface, where the reactants come in contact with one another. The surface structure of a crystal is determined by the ordering of the crystal at a molecular level and can be predicted based on the unit cell and the crystal habit. Moreover, the stacking of molecules impacts the mobility on the crystal surface. Crystal defects are weak points of a crystal and are known to be sites were water can be adsorbed easily.8,9 This can trigger drug–excipient interactions as the adsorbed water can improve access of different conformations of the reactants.
2
ACS Paragon Plus Environment
Page 3 of 17
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
43 44 45 46 47
Molecular Pharmaceutics
For CAR and COP crystals, we used Molecular Dynamics (MD) simulations to model structural properties of their crystal surface. To quantify the surface properties that could influence reactivity, we established surface descriptors representing accessibility of the reactive site of molecules and the stability of the surface. For each API, we investigated two different polymorphs (or, in the case of COP, two different hydrates) as this allows us to eliminate molecular/electronic differences, and thus, to concentrate purely on effects of the crystal surface structure.
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
6,7
Figure 2. Esterification reactions of CAR (top panel) and COP (bottom panel) with CA.
Experimentally, we carried out degradation studies, where physical mixtures of API and CA were stored at controlled humidities and elevated temperatures. For one of the model compounds (CAR), we further studied the reaction kinetics under different storage temperatures and humidities. Furthermore, we investigated the drug–excipient interactions induced by mechano-activation (milling) and the reactivity of the model APIs with CA in aqueous solutions. METHODS Modelling of API crystal surfaces. All modelling work was carried out with classical molecular mechanics and dynamics simulations using the GROMACS 5.0.4 package.10 As a model potential, the Generalized Amber Force Field (GAFF) was utilized.11,12 The restricted electro-static potential (RESP) charges of molecules were obtained by using B3LYP exchange and correlation functionals together with the cppVTZ basis set to calculate the electrostatic potential and subsequent fitting of the RESP charges to this electrostatic potential.13 For hydrated crystals, the TIP3P water model14 was used. Crystal structure input files were obtained from the Cambridge Structural Database (CSD)15 and were processed with GDIS 0.9916 to create the 15 most likely crystal faces according to the Bravais-Friedel-Donnay-Harker (BFDH) habit which is entirely based on geometric properties of the crystal unit cell.17 Each of these crystal surface structures was then processed with open-babel and several in-house scripts to build slabs of approximately 40 × 40 Å and a thickness of > 80 Å. The obtained slabs were used as input files for the molecular mechanics and dynamics simulations. Attachment energies for each of the surface slabs were calculated to predict the attachment energy habit of the selected API crystals. The attachment energy model is more advanced than the BFDH model and takes into account the growth rate of crystal faces. The larger the attachment energy, the faster the growth of the face, hence the less morphologically important this face will be.18–20 For each of the surfaces of the resulting attachment energy habit, we calculated the following four descriptors: •
Exposed Reactive Surface Area (ERSA): For surface slabs of approximately 40 × 40 Å in size and a thickness of > 80 Å, we performed molecular dynamics (MD) simulations in a box with periodic boundary conditions. The height of the box perpendicular to the crystal surface was approximately 200 Å. The MD simulations were performed for 100 ps at a temperature of 300 K (velocity rescaling) and a pressure of 1 bar (Berendsen barostat). A Particle Mesh Ewald algorithm was used to account for electrostatic long-range interactions. From the trajectory, snapshots were taken at an interval of 1 ps. For calculating the ERSA, we used the GROMACS tool gmx_sasa.
•
Number of Close Encounters (nCE): For calculating the nCE, we performed a 2 ns MD simulation (T = 300 K and p = 1 bar) of each crystal face in contact with amorphous CA. We then used the GROMACS tool gmx_distance to calculate the frequency that the oxygen of the secondary alcohol of CAR or COP comes into close proximity to one of the three carboxyl carbons of CA. An encounter was considered "close" if the distance between the oxygen and the carbon was smaller than 3 Å.
•
Root Mean Square Fluctuation (RMSF): The RMSF gives the fluctuation of surface molecules around their average position. Before calculating the RMSF, the central regions of the crystal slabs were aligned. The RMSF
3
ACS Paragon Plus Environment
Molecular Pharmaceutics
84 1 85 2 86 3 87 4 5 88 6 89 7 90 8 91 9 92 10 93 11 94 12 95 13 96 14 97 15 98 16 99 17 100 18 101 19 102 20103 21104 22105 23106 24107 25108 26109 27 28110 29111 30112 31113 32114 33115 34116 35117 36118 37119 38120 39121 40122 41123 42124 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 17
was then calculated between the initial configuration corresponding to the perfect and unrelaxed crystal structure and snapshots of the 2 ns MD trajectory. •
Binding energy (Ebind): The Ebind was calculated by comparing the energy of a crystal slab before and after one molecule was removed from the crystal lattice at the surface.
The surface descriptors ERSA, nCE, RMSF, and Ebind were calculated for each individual crystal face and the numbers were then weighted according to the relative surface area of the face on the crystal habit. Solubility predictions. The aqueous pH-solubility profiles of CAR and COP and their esters with CA at 25 °C were estimated by Advanced Chemistry Development, Inc. (ACD/Labs) v12.0.21 Materials. Citric acid anhydrous (ACS reagent > 99.5 %), citric acid monohydrate (ACS reagent ≥ 99.0 %), toluene (ACS reagent ≥ 99.5 %), methanol (ACS reagent ≥ 99.8 %), N,N-dimethylformamide (ACS reagent > 99.8 %), and acetone (ACS reagent > 99.5 %) were purchased from Sigma Aldrich. CAR raw material was purchased from Indoco Remedies, India. COP raw material was purchased from a pharmacy (Neutor Apotheke, Graz). Preparation of solid forms of CAR and COP. CAR Form II and Form III were recrystallized as described in the literature.22 For obtaining CAR Form II, 1.7 g CAR were dissolved in 200 ml toluene by constant stirring at 60 °C. The solution was then kept at room temperature and crystals were harvested via filtration after two days. CAR Form III was produced by dissolving 4 g CAR in 100 ml methanol by stirring at 50 °C. The solution was then left at room temperature for crystallization and crystals were harvested via filtration after one day. Amorphous CAR was produced by hot-melt extrusion (HME) and subsequent cryo-milling. HME was carried out using a bench top 9 mm-twin screw extruder (Three-Tec, Seon, Switzerland) at extrusion temperature of 115 °C and screw speed of 100 rpm. Bulk amorphous CAR was prepared via HME to have a consistent thermal history of the samples used for all the experiments and chemical purity following HME was confirmed by UHPLC analysis (method described below). Alternatively, amorphous CAR was produced by cryomilling at 25 Hz for 60 minutes. The amorphous material was confirmed as such based on short and wide angle X-ray scattering (SWAXS). The chemical purity of amorphous CAR was confirmed by ultra high performance liquid chromatography (UHPLC). The polymorphic form of CAR crystals was characterized by SWAXS patterns and characteristic peaks in the Raman fingerprint region. The polymorphic purity was confirmed by differential scanning calorimetry (DSC) via analysis of the melting peaks. The two hydrates of COP were recrystallized according to the methods described in the literature.23 The hemi-hydrate of COP (COP-HH) was recrystallized by dissolving 2.0 g of COP-HH (raw material) in 125 ml of N,N-dimethylformamide in an 500 ml-Erlenmeyer-flask under stirring and heating to 90 °C. The solution was then kept at ambient conditions for 4 days. The COP-HH crystals were then filtered from the solution and dried at ambient conditions for 1 hour. For obtaining the sesqui-hydrate (COP-SH), 3.0 g COP raw material was dissolved in 3.5 ml water at room temperature. COP-SH crystals were then precipitated from the solution by slowly adding 10.0 ml acetone. The crystals were harvested by filtration and dried at ambient conditions. Recrystallization of COP-HH and COP-SH was confirmed by comparing Raman spectra of the hydrates to the literature.24,25 Drug–excipient interactions in the solid state. In order to assess the drug–excipient interactions of CAR and COP with CA experimentally, samples containing physical mixtures of API and CA were stored at elevated temperatures and controlled humidity. The drug load was kept constant (1 part API + 10 parts CA w/w) in all samples. Furthermore, we consistently used API crystals of identical particle size ranges (as obtained by sieving) when comparing different crystal forms. An overview over the degradation studies is shown in Table 1. Table 1. Summary of the storage conditions, APIs, and particle sizes that were used in the different degradation studies. Humidity [% RH]
T [°C]
API solid forms
Particle size [µm]
~ 100
70
CAR Form II
90-180
CAR Form III
90-180
CAR Form II
200-355
CAR Form III
200-355
CAR Form II
