A New Method To Identify Physically Stable Concentration of

Aug 8, 2017 - Accordingly, from the extrapolation of VFT fit to 100 s, we subsequently estimated Tg of FL as 271 K. The glass-transition temperature d...
0 downloads 6 Views 1MB Size
Subscriber access provided by University of Rochester | River Campus & Miner Libraries

Article

A new method to identify physically stable concentration of Amorphous Solid Disspersions (I): case of Flutamide + Kollidon VA64 K. Chmiel, J. Knapik-Kowalczuk, K. Jurkiewicz, W. Sawicki, R. Jachowicz, and M. Paluch Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b00382 • Publication Date (Web): 08 Aug 2017 Downloaded from http://pubs.acs.org on August 14, 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 27

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

A new method to identify physically stable concentration of Amorphous Solid Disspersions (I): case of Flutamide + Kollidon VA64 K. Chmiel 1 ,2 * , J. Knapik-Kowalczuk Jachowicz 4 and M. Paluch 1 ,2 1

1,2

, K. Jurkiewicz

1 ,2

, W. Sawicki 3 , R.

Institute of Physics, University of Silesia, ul. Uniwersytecka 4, 40-007 Katowice, Poland

2

Silesian Center for Education and Interdisciplinary Research, ul. 75 Pułku Piechoty 1a, 41500 Chorzow, Poland 3

Department of Physical Chemistry, Medical University of Gdansk, 84-416 Gdansk, Poland

4

Jagiellonian University, Faculty of Pharmacy, Department of Pharmaceutical Technology and Biopharmaceutics, Medyczna 9, 30-688 Kraków, Poland

*

corresponding author: [email protected]

TOC

Key words: amorphous solid dispersion, molecular dynamics, physical stability, dielectric spectroscopy, flutamide, kollidon VA64. 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

Abstract In this paper a novel approach to determine stable concentration in API-polymer systems is presented. As a model binary amorphous mixtures flutamide (FL) drug with a copolymer Kollidon VA64 (PVP/VA) has been used. It is worth to note that the finding an effective method to achieve this goal is a matter of great importance since physical stability of the amorphous pharmaceuticals is the key issue investigated worldwide. Due to the fact that molecular dynamics was found to be crucial factor affecting physical stability of disordered pharmaceuticals, we examined it for both neat FL and its PVP/VA mixtures by means of Broadband Dielectric Spectroscopy (BDS). Thorough investigation of the impact of polymeric additive on the molecular mobility of disordered FL, reviles unusual, previously un-reported behavior. Namely, simultaneously with the beginning of the recrystallization process, we observe some transformation from unstable supersaturated concentration of investigated mixture to the different, unknown concentration of FL-PVP/VA. Observed, during BDS experiment, transformation enables us to determine the limiting, highly physically stable concentration of FL in PVP/VA polymer (saturated solution), which is equivalent to FL + 41% wt. of PVP/VA. Described high physical stability of this unveiled system has been confirmed by means of long-term XRD measurements. According to our knowledge this is the first time when such a behavior has been observed by means of BDS.

2 ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27

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

Introduction According to a statistical report from the CDC (Centers for Disease Control and Prevention) the malignant neoplasms – cancers – are the second leading cause of death in the United States. Among all types of cancers, prostate cancer is the most frequently diagnosed solid organ malignancy and remains the second most common cause of cancer mortality amongst men worldwide. Due to the fact that its rate of occurrence is alarmingly high there is an urgent need to find new or to improve already existing pharmaceuticals for this disease1. A commonly applied practice in prostate cancer treatment is the antiandrogen therapy. The purpose of this therapy is to maximally block the action of the male hormones. The ideal antiandrogen should act exclusively as an androgen inhibitor and should manifest lack of any other hormonal and antihormonal activity2. Flutamide (FL), which is a synthetic non-steroidal drug, meets these criteria3 and similarly to bicalutamide is the most commonly used antiandrogen on the market. Despite of all of its advantages, the FL drug is characterized by low bioavailability which is associated with low aqueous solubility4. As a consequence of the low bioavailability, patients need to take a relatively high dose (750 mg/day) of this compound, which may result in frequently documented gastrointestinal side effects of FL. The undissolved residual mass of FL after oral administration might cause a nausea, vomiting, stomachache, diarrhea and even ulcers5. To overcome the gastrointestinal side effects of conventional FL, the novel formulations of this drug, with better bioavailability, should be developed. An effective approach to improve the drugs bioavailability is to change the physical form of these compounds by converting them from the crystalline to amorphous state6,7. In contrast to crystals, amorphous solids are characterized by higher internal energy and absence of long-range ordering (they present molecular arrangement similar to that of liquids). Change of the form of API usually results in improvement of its water solubility and accelerates its dissolution rate8,9. This ultimately leads to enhanced bioavailability of the transformed drug and allows a reduction in the API dose while preserving the same therapeutic effect. Consequently, it is possible to eliminate gastrointestinal side effects resulting from the undissolved mass of drug accumulating in the intestines. However, it should be noted that the converted crystalline drugs into amorphous form are thermodynamically unstable. It means that sooner or later these drugs will revert back to their initial – crystalline – form, simultaneously losing their superior features10,11. Tendency

3 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

toward recrystallization of drugs in the amorphous form is the most important limiting factor responsible for their restricted use in pharmaceutical technology12,13. The purpose of our studies was to obtain a physically stable amorphous form of FL drug. The physicochemical properties of both crystalline and amorphous form of examined pharmaceutical was investigated by means of X-ray diffraction (XRD) as well as Differential Scanning Calorimetry (DSC). Due to the fact that the molecular mobility of disordered materials is generally considered as the main factor governing their tendency toward recrystallization Broadband Dielectric Spectroscopy (BDS) has been applied to monitor the molecular dynamics and physical stability of studied antiandrogen. The measurements performed herein indicate that FL in its amorphous form will not be stable for a sufficiently long time required to introduce this drug into the market. Therefore, additional efforts have been made to find an effective way to stabilize it. Taking into account recent reports indicating that polymers with high glass transition temperature are able to improve physical stability of even extremely unstable pharmaceuticals due to the antiplasticization effect, we decided to stabilize this amorphous antiandrogen compound by means of high Tg polymer addition. Thus, we prepared drug-polymer binary mixtures, one of which proved to be resistant to the devitrification process at room temperature conditions. Furthermore, thorough dielectric examination of those systems led us to discover a method that potentially might be used to predict the lowest concentration of the polymers needed to effectively stabilize amorphous drugs.

4 ACS Paragon Plus Environment

Page 4 of 27

Page 5 of 27

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

Experimental methods Materials. Flutamide drug of molecular mass MW = 276.21 g mol-1 was purchased from Sigma-Aldrich and used as received. Kollidon VA64 (PVP/VA) of molecular mass MW = 45 000 – 47 000 g mol-1 was purchased from BASF SE (Germany). The concentration of binary mixtures obtained for the purpose of this study was calculated as a molecular ratio of FL molecule to combined PVP and PVA mer. Preparation of binary system. The binary FL-PVP/VA systems were prepared at different molecular ratios which were assigned as 5:1 and 1:1. This corresponds to the following weight percentage concentrations of PVP/VA in each sample: 12.5% wt and 41.5% wt , respectively. To acquire homogeneous samples we mixed compounds with polymer at appropriate ratios in mortar for approximately 20-30 minutes. Prepared this way mixtures were then melted at T = 400 K and vitrified by fast transfer to a previously chilled copper plate. All measurements were performed immediately after preparation to avoid recrystallization. Differential Scanning Calorimetry (DSC). Thermodynamic properties of FL, Kollidon VA64 and their binary systems were examined using a Mettler–Toledo DSC 1 STARe System. The measuring device was equipped with a HSS8 ceramic sensor having 120 thermocouples. The instrument was calibrated for temperature and enthalpy using indium and zinc standards. Crystallization and melting points were determined as the onset of the peak, whereas the glass transition temperature as the midpoint of the heat capacity increment. The samples were measured in an aluminum crucible (40 µL). All measurements were carried out in range from 260.15 K to 450.15 K with 10 K/min heating rates. Broadband Dieletric Spectroscopy (BDS). The dielectric measurements of FL were carried out using Novo-Control GMBH Alpha dielectric spectrometer, in the frequency range from 10-1 Hz to 106 Hz at temperatures from 193.15 K to 473.15 K. The temperature was controlled by a Quattro temperature controller with temperature stability better than 0.1 K. Dielectric studies of FL and its binary systems were performed immediately after its vitrification by fast cooling of the melt in a parallel-plate cell made of stainless steel (diameter 10 mm, and a 0.1 mm gap with quartz spacers). Crystallization kinetics of FL at ambient pressure were carried out at 393.15 K. X-ray Diffraction (XRD) The X-ray diffraction experiments were performed at ambient temperature on a Rigaku-Denki D/MAX RAPID II-R diffractometer (Rigaku Corporation, Tokyo, Japan) with a rotating anode Ag KR tube (λ = 0.5608 Å), an incident

5 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

beam (002) graphite monochromator, and an image plate in the Debye−Scherrer geometry. The pixel size was 100 µm × 100 µm. Measurements were performed on sample-filled and empty capillaries, to allow subtraction of the background intensity. The beam width at the sample was 0.1mm. The two dimensional diffraction patterns were converted into one dimensional intensity data.

6 ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

Results and Discussion Physicochemical properties of amorphous FL In order to determine the melting point (Tm) of crystalline FL the DSC technique was applied. The thermogram obtained during heating of the sample is presented in Figure 1a. The heating rate applied in the calorimetric experiment was 10 K/min. As can be seen, the DSC curve of the measured sample exhibits one endothermic peak with an onset at Tm = 384 K. This endothermic event corresponds to the melting of the examined, crystalline API. Amorphous form of FL (chemical structure shown in Figure 2c) was prepared by the vitrificaton i.e. the rapid cooling of molten sample. Similarly to the case of crystalline FL, the thermal analysis of the vitrified drug was conducted via calorimetric technique. On the DSC thermogram obtained during heating step-like behavior, which is a manifestation of the glass transition, can be well visible at temperature equal to 274 K. Upon further heating of this sample non-isothermal cold crystallization also occurred. The onset of the exothermic peak corresponding to this phenomenon was registered at T=Tg+42K. This value can be compared with the results obtained for two different APIs: ezetimibe (Tg+81K) and nimesulide (Tg+54K). Taking into account that these pharmaceuticals were classified as materials relatively easily crystallizable14,15, one can consider FL as a drug manifesting similar features. Heating rate: 10 K/min crystalline Flutamide vitrified Flutamide Tm = 384 K

Heat Flow (endo up) [arb. units]

a) TC = 316 K

Tm = 383 K

b) Heat Flow (endo up) [arb. units]

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

260

280

300

320

0,30 0,27 0,24

Tg = 274 K

0,21 0,18

265 270 275 280 285 Temperature [K]

340

360

380

400

Temperature [K] Figure 1 DSC thermograms of crystalline (a) and vitrified (b) form of FL. Arrows indicates the crystallization onset and melting point. The inset shows the close up of glass transition process.

To ensure that the vitrified FL was fully amorphous and none of the residual crystalline fraction remained within the bulk, the X-ray diffraction measurements were 7 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

performed. Diffraction patterns for the crystalline (which was taken as a reference) and vitrified FL are presented in Figure 2a and b, respectively. As can be seen, the XRD pattern of quench cooled sample is characterized by a lack of the sharp Bragg peaks which appears only in case of crystalline material (compare on Figure 2a). This result indicates that the vitrified sample had no long-range three-dimensional molecular order. Thus, our sample had indeed an amorphous form by XRD.

Figure 2 X-ray diffraction patterns for crystalline FL (a) and FL obtained by vitrification method immediately after quenching (b). Chemical structure of FL (c).

Molecular dynamics and physical stability of amorphous FL Due to the fact that molecular dynamics was frequently found to be crucial factor affecting physical stability of disordered pharmaceuticals16,17,18,19,20, in this part of our studies we have investigated the molecular mobility of the amorphous form of FL via dielectric measurements in a wide frequency and temperature range. In Figure 3c we demonstrate the arbitrary chosen dielectric loss spectra which were obtained during heating of the amorphous sample. On the spectra collected below Tg, the loss characterized by a power law

ε " ( f ) = Bf − λ with λ < 0.2 can be well observed. This feature might be identified as the nearly constant loss (NCL), which is manifestation of the cage molecular dynamics. NCL can be well observed under certain conditions. The examined sample has to be characterized by αloss peak with narrow frequency dispersion. More importantly, although not a necessary condition is that a sample cannot exhibit any: β-relaxation, Johari-Goldstein β-relaxation separated from the α-process nor slow ɣ-relaxation process. This is because the intensity of any of these additional processes can overlap the NCL part of the dielectric spectra. An example of a pharmaceutical, in which NCL was also observed is prilocaine21. During additional heating one can observe, on the dielectric spectra of FL, a well-resolved peak 8 ACS Paragon Plus Environment

Page 8 of 27

Page 9 of 27

corresponding to the structural (α) relaxation process. This peak is very visible at T > 279 K (i.e. above the liquid-glass transition temperature of FL) and moves toward higher frequencies with increasing temperature. In order to check if heating modifies the shape of α-relaxation peak, we horizontally shifted the dielectric loss spectra taken at temperatures from 213K to 291K to superimpose on the reference spectrum at 283 K. The masterplot constructed this way, presented in Figure 3b, indicates that the shape of the FL’s structural relaxation is temperature independent. To characterized the shape of the α-loss peak we fitted it by means of the one-sided Fourier transform of the Kohlrausch-Williams-Watts (KWW) function22,23. The value of the obtained βKWW parameter is equal to 0.86. It is worth recalling that the value of βKWW might vary within the range from 0 to 1. If this parameter is equal to 1 it denotes that the α-relaxation peak is narrow and symmetric, and obviously correspond to the Debye case. On the other hand, if the value of βKWW is approaching 0 the shape of structural relaxation is asymmetric and broad. Taking into account the recently established anticorrelation in the van der Waals glass formers narrow α-loss peak (large value of βKWW) should exhibit high value of the dielectric strength (∆εα)24. FL having βKWW = 0.86 and ∆εα = 85, follows this anticorrelation similarly to the prilocaine (βKWW = 0.75, ∆εα = 20)21, cimetidine (βKWW = 0.80,

10-1 100 101 102 103 104 105 106 107 108 109 101010111012 1013 2

10

1

Flutamide

101

βKWW = 0.845

100

NCL

Tg = 271

10-1

b) c) α - process

10-2 267 K

0

153 K

∆T = 2K ∆T = 10K

297 K 102 263 K 101

-1

100 TC = 295K Crystallization onset

α - process

-2

10-1

Dielectric loss ε"

2

Dielectric loss ε"

∆εα = 35)24 and bikalutamide (βKWW = 0.85, ∆εα = 60)25.

log τα

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

-2

10

a)

-3 3,4

3,6

slope = 0.13 -1

10

0

10

1

10

2

10

NCL 3

10

4

10

Frequency [Hz]

-3

5

10

10 106

Temperature [1000/T] Figure 3 Temperature dependence of τα in the supercooled liquid has been described by VFT equation (red solid line) (a). Horizontally shifted dielectric loss spectra taken at temperatures from 213 K to 291 K to superimpose on the reference spectrum, dashed line represents both βKWW function and NCL(b). Dielectric loss spectra of amorphous FL.

9 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 27

Black lines indicates the α process, red line indicates the NCL and the purple dashed lines corresponds to the crystallization process (c).

From the analysis of loss spectra collected above the liquid-glass transition we determined the temperature dependence of the α-relaxation times for FL (see Figure 3a). To obtain the values of τα at various temperatures, the experimental data were fitted by using the Havrilak-Negami (HN) function (Eq. 1):26

ε * (ω ) = ε ∞ +

∆ε

[1 + (iωτ ) ]

a b

HN

(1) where ε*(ω) is the complex dielectric permittivity relative to vacuum and ε*(ω)=ε’(ω)-iε”(ω), ε∞ is high frequency limit permittivity, ε0 is the permittivity of vacuum, ∆ε is dielectric strength, ω is equal to 2πf, τHN is the HN relaxation time, a and b represents symmetric and asymmetric broadening of relaxation peak. Based on the fitting parameters determined above the values of τα were calculated by means of the following formula: −

τα / β

1

1

  πa  a   πab  a = τ HN sin   sin     2 + 2b    2 + 2b  (2)

To parameterize the τα(T) dependence of FL in the supercooled liquid state we used the Vogel-Fulcher-Tamman (VFT) equation that is defined as follows:27,28,29



B    T − T0 

τ α (T ) = τ ∞ exp

(3) The fit parameters: log(τ∞), B, T0, that have been obtained from the fitting procedure, of the temperature evolution of FL’s structural relaxation times at T > Tg, are equal to -14,1; 1994,2 and 218,2 respectively. To obtain Tg values, we used commonly known definition Tg = T(τα = 100 s). Accordingly from the extrapolation of VFT fit to 100 s we subsequently estimated Tg of FL as 271 K. The glass transition temperature determined by this method is in well agreement with that previously obtained from calorimetric studies as well as with the literature data30. Referring back to Figure 3c, it should also be noted that above 293 K the intensity of structural relaxation peak starts to decrease with heating. Such a drop in the intensity of the αrelaxation loss peak reflects the decrease of ∆ε which is proportional to the number of units involved in structural relaxation31.

10 ACS Paragon Plus Environment

Page 11 of 27

∆ε = ε S − ε ∞ ≈ Nµ 2

(4) where εs is the static dielectric constant and ε∞ its high frequency analogue, N is the number of reorienting units (dipoles) and the µ is the average electric dipole moment of the sample. During the crystallization process the number of reorienting dipoles contributing to the primary α-relaxation rapidly decreases7. Therefore, one can conclude that at T = 293 K the recrystallization of amorphous form of FL begins. Due to the fact that T = 293 K is the temperature at which pharmaceuticals are usually stored, we performed at this T the time dependent isothermal BDS experiment. This measurement has been performed in order to characterize the physical stability of FL at standard storage conditions (T ~ room T and ambient p). The representative frequency dependence of the real (ε’) and imaginary (ε”) parts of complex dielectric permittivity obtained during this experiment are presented in Figure 4a and b, respectively. Frequency [Hz]

102

εs

101

102

103

Frequency [Hz]

104

105

10-1

ε S − ε ∞ = ∆ε =

t = 0s

100

2

π

101

102

103

104

105

106



∫ ε " (ω )d ln ω

10

2

0

t=

a)

101

0s

b) 100

∆t = 180 s

∆t = 180 s

101

10-1

ε∞ t = 11160 s

10

-2

T = 293 K

1,00 0,75

t = 11 160 s

Dielectric loss ε"

Dielectric permitivity ε'

100

Normalized dielectric permitivity ε'N

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

Crystallization onset

0,50 0,25

Crystallization endset

t1/2

0,00

0

2000

4000

6000

8000

10000

c) 12000

Time [s]

Figure 4 Upper panel presents dielectric spectra of the real and imaginary parts of the complex dieletrcic permittivity during an isothermal crystallization at 293 K. Lower panel shows normalized dielectric constants ε’N as a function of time from crystallization processes occurring at 293 K.

As can be seen both the intensity of structural relaxation loss peak as well as the static dielectric permittivity of FL decrease with time of experiment reflecting the progress of cold crystallization. The increase in the crystallization degree can be easily analyzed calculating the normalized real permittivity (ε’N) by means of following formula:

11 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

ε ' N (t ) =

Page 12 of 27

ε ' (0 ) − ε ' (t ) ε ' (0 ) − ε ' (∞ ) (5)

where ε’(0) is the initial static dielectric permittivity, ε’(∞) is the long-time limiting value, and ε’(t) is the value at time t. The data, normalized in this fashion and plotted against time, are presented in Figure 4c. As can be seen this analysis clearly shows that the amorphous form of FL, stored at 293 K, begins to crystallize after ~20 minutes of preparation. The halflife crystallization time is equal to 67 minutes, while the full crystallinity of amorphous FL has been achieved after 3.4 hour. This result clearly shows that amorphous form of FL is extremely unstable and it has to be stabilized in order to fulfill the required shelf life time. Therefore our further efforts were concentrated on improvement of the physical stability of this drug. It has been frequently reported that polymers with high Tg value are able to effectively stabilize even extremely unstable in amorphous form pharmaceuticals. Since polymers are commonly used as excipients in pharmaceutical industry, we tried suppress the cold crystallization of FL by means of one of their representatives. From the broad spectra of polymers that potentially can be used for that purpose we choose Kollidon VA64 polymer (PVP/VA). This polymer is a block co-polymer made of polyvinylpyrrolidone and polyvinylacetate mers with a structure shown in Figure 5. Our choice was motivated by the recent report in which authors confirmed that Kollidon VA64 absorbed significantly less moisture than the other well known in pharmaceutical industry polymers. This analysis was conducted for neat polymers as well as amorphous solid dispersion systems (ASD)32. The fact of absorbing noticeably less water is particularly desirable in ASD formulation since water absorbed during storage tend to accelerates the recrystallization of the amorphous solid dispersions33.

Effect of Kollidon VA64 polymer on the physical stability of amorphous FL In order to evaluate the effect of Kollidon VA64 on the physical stability of amorphous form of FL, the binary mixture in molecular ratio Flutamid 5:1 Kollidon VA64 (FL 5:1 PVP/VA) has been prepared and examined by various experimental techniques. First, we applied differential scanning calorimetry to check the homogeneity of the prepared system. It has been many times demonstrated that if the phase separation exists in the mixture, the two separate thermal events (Tg) - first corresponding to the polymer-rich and second to the drug-rich - should be well visible on DSC thermogram34. As can be seen in Figure 5, where the DSC traces of pure FL, pure PVP/VA polymer as well as FL 5:1 PVP/VA are 12 ACS Paragon Plus Environment

Page 13 of 27

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

presented, the examined mixture is characterized by a single glass transition event just like in case of the pure components. This result might suggests that in the prepared binary system the phase separation does not occurs i.e. the investigated system is homogenous. In the inset of Figure 5 one can observe that the midpoint of the glass transition temperature of FL 5:1 PVP/VA is equal to 281 K. This value is slightly higher than that for pure FL (Tg = 274 K), which is connected with the antiplasticization effect exerted by the polymer. The value of Kollidon’s VA64 glass transition temperature is equal to 376 K. By increasing the temperature of glass transition point of FL its molecular mobility is slowed down. As a consequence of that FL’s onset of cold crystallization shifts toward higher temperatures (Tc = 350 K) indicating improvement of the drug physical stability respect to the room temperature. Comparing the neat and mixed FL samples stored at the same T conditions, the mixture, without a doubt, is characterized by higher physical stability.

Figure 5. Thermograms of amorphous: FL (a), FL 5:1 PVP/VA (b) and PVP/VA (c). Upper inset shows chemical structure of Kollidon VA64. The lower inset presents the close-up for the glass transition of the FL 5:1 PVP/VA mixture.

To thoroughly investigate the impact of polymeric additive on the molecular mobility of disordered FL the dielectric measurements in a wide temperature and frequency range have been performed. The representative dielectric loss spectra of FL 5:1 PVP/VA mixture are shown in the inset of Figure 6. The spectra collected above Tg exhibit well resolved loss peak, which can be identify with the α-relaxation of the investigated drug-polymer mixture. This mode moves toward higher frequencies with increasing temperature and its intensity remains 13 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

unchanged until T = 313 K. Above this temperature the intensity of α-relaxation begins to rapidly decreases reflecting the reduction in the amorphous fraction of FL. Moreover, simultaneously to the crystallization, additional relaxation process emerges on the low frequency side of ε”(f). This process becomes more evident with the progress of crystallization. Furthermore, at a certain temperature at which the devitrification ceased (T = 347 K), this additional relaxation process remains visible on the dielectric loss spectra. It is worth noting that during further increase of temperature this – additional – process moves towards higher frequencies.

Figure 6 Left panel shows relaxation map of binary mixtures of FL 5:1 PVP/VA. Temperature dependence of τα(black circles) and τadd(blue circles) in the supercooled liquid has been described by VFT equations (red solid lines).The right panel presents the dielectric loss spectra of amorphous binary mixture of FL 5:1 PVP/VA. Black lines indicates the α process, red lines corresponds to the crystallization process and the blue lines represents the additional process.

In order to identify the origin of the “additional” relaxation process, which appears during recrystallization of FL 5:1 PVP/VA mixture, we fitted dielectric loss spectra (marked in blue in the inset of Figure 6) by means of HN function and then determined the temperature dependences of its relaxation times. As can be seen in Figure 6, where we compared τadd(T) with the τα(T), the temperature evolution of the “additional” relaxation times exhibits nonlinear behavior, and follows VFT equation. It is reasonable to assume that this additional process is in fact the primary relaxation process from different than initial concentration of FL in the FL:PVP/VA mixture. To confirm our hypothesis we decided to exclude the possibility of associating this additional process with the pure PVP/VA polymer which remained amorphous after re-crystallization of FL from the mixture. Extrapolation of the VFT fit to 100 s, allowed us to determine the Tg value of this additional process. Taking into account that the 14 ACS Paragon Plus Environment

Page 14 of 27

Page 15 of 27

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

added polymer has a glass transition temperature at 376 K (see DSC thermogram presented in Figure 5), while this additional process is characterized by Tg = 317 K, one can conclude that the newly appeared process does not reflect the segmental relaxation of the residual Kollidon VA64. Based on this analysis we confirmed our assumption that the newly formed additional relaxation process observed during recrystallization of FL 5:1 PVP/VA mixture corresponds to the structural relaxation of the limiting concentration of FL:PVP/VA which is not able to further crystallize. Due to the fact that during the BDS experiment we observe some kind of transformation from unstable saturated concentration of investigated mixture to the different, unknown, and probably highly physically stable concentration of FL:PVP/VA, we performed additional DSC measurements. This experiment consisted of several steps. During the first step the freshly prepared amorphous mixture of FL 5:1 PVP/VA was heated up from 253 K to 319 K with heating rate of 10 K/min. Next, the sample was stored for 20 min at 319 K, and cooled down to 253 K with a cooling rate of 20 K/min. After this stage of the experiment we re-heated the mixture to 319 K, then stored it for the next 40 min, and one more time cooled down it to 253 K. This process described in the preceding sentence was repeated three times. The results from this experiment are presented in Figure 7a. As can be seen if the sample is fully amorphous on the DSC thermogram only one glass transition event can be observed (see green line in Figure 7). It should be highlighted that its value corresponds well to that obtained from the dielectric studies. During the progress of experiment time i.e. with the progress of FL crystallization, one can observe appearance of the second, significantly smaller, glass transition (marked with red arrow). This would indicate that some transformation occurred in the mixture. It is worth noting that this effect is more and more evident as the cold crystallization proceeds. When the second Tg becomes more apparent, the first Tg is vanishing up to a point of complete disappearance. From that moment, the sample ceases to crystallize and on its DSC thermogram only one Tg (associated with this newly established concentration) might be observed. Its value is equal to 311 K, which corresponds very well to that obtained from the BDS experiment. It should be noted that the heat capacity change at Tg (∆Cp) of this settled FL:PVP/VA concentration is significantly smaller than that for the initial material (FL 5:1 PVP/VA). This is because, after crystallization of the FL within the mixture, the remaining drug-polymer concentration is just a small fraction of the original sample mass i.e. a huge part of the sample is in the crystalline form. Presence of the crystalline fraction of FL amongst amorphous mixture does not necessarily imply that the two compounds are immiscible. As reported this 15 ACS Paragon Plus Environment

Molecular Pharmaceutics

phenomenon can be explained by the solubility limits. Considering polymers as “solvent” and their ability to dissolve a crystalline API one shall take into account the situation when obtained mixture is supersaturated and conditions for crystallization are favorable. In this case the excess amount of the API will recrystallize to the point of reaching its solubility limits within the polymer35. Time dependent data can be better visualized by presenting temperature dependence of derivative of Heat Flow with respect to temperature (see Figure 7b). The maximum of the

Tg = 281 K Tg = 281 K Tg = 282 K Tg = 283 K Tg = 284 K Tg = 297 K Tg = 299 K Tg = 302 K Tg = 311 K

a)

b)

after 180 min after 140 min after 100 min after 60 min after 20 min after amorphization

d Heat Flow / d T [arb. units]

peaks in Figure 7b represents the mid points of the glass transitions shown in Figure 7a. Heat Flow (endo up) [arb. units]

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 16 of 27

Heating rate 10 K/min

260

270

280 290 300 310 Temperature [K]

320

270

280 290 300 Temperature [K]

310

Figure 7 Left panel presents the thermograms in temperature dependence of heat flow of amorphous binary mixture FL 5:1 PVP/VA in a during isothermal crystallization in T = 319 K. Right panel shows thermograms presenting temperature dependence of first derivative of heat flow over temperature of amorphous binary mixture FL 5:1 PVP/VA during isothermal crystallization in T = 319 K. Black arrows indicates the Tg related to the α process and red arrows represents Tg related to the additional process.

The influence of PVP/VA content on the Tg value of the FL:PVP/VA binary mixture might be well described by the Gordon-Taylor/Kelley-Bueche equation defined as follows: Tg =

W1Tg1 + KW2Tg 2 W1 + KW2 (6)

where the Tg, Tg1, Tg2 are the glass transition temperatures of the drug-polymer mixture, the amorphous drug, and the polymer, respectively; W1 and W2 are the weight fraction of the drug and polymer and K is a parameter that can be calculated by means of the formula:

K≈

∆C p 2 ∆C p1 (7)

16 ACS Paragon Plus Environment

Page 17 of 27

where ∆Cp1 and ∆Cp2 are the changes in heat capacity at Tg of drug and polymer, respectively. Therefore, to determined this newly established concentration of FL:PVP/VA mixture we compare, Tg values from experiments, with those determined by G-T/K-B equation, concentration dependence of the glass transition temperature for the systems FL:PVP/VA. From this procedure we obtained the Tg value of the newly established concentration. The comparison described above is presented in Figure 8. Gordon-Taylor Kelley-Bueche equation experimental points point predicted from obtained data

380

360

340

Tg [K]

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

320

300

Tg value obtained after crystallization of FL 5:1 PVP/VA Concentration of newly formed mixture after crystallization of FL 5:1 PVP/VA predicted based on G-T/K-B eq.

FL 5:1 PVP/VA

280

260 0

20

40

60

80

100

% Kollidon VA64 Figure 8 Glass transition temperatures of FL - PVP/VA mixtures. The blue points corresponds to the experimentally determined Tg values. The dashed black line represents the Gordon-Taylor/Kelley-Bueche equation. The red circle is the point obtained by using Tg related to the τadd process (Tg obtain after “complete” crystallization).

As can be seen, accordingly to this procedure the concentration of FL:PVP/VA mixture which has been achieved after recrystallization of FL 5:1 PVP/VA should be equal to 1:1 in molecular ratio (i.e. FL + 41% wt. PVP/VA). One has to keep in mind that sometimes experimentally obtained Tg values differ from the theoretically predicted value. This might be due to the volume nonadditivity originated from the nonideal mixing of the drug and polymer35,36. In case any deviation from GT prediction authors suggest the preparation of a experimentally determined dependence of glass transition temperature vs concentration. Using this dependence as the base to determine newly formed concentration is far more reliable than GT equation due to a possibility of additional H-bonding interactions between API and polymer. However, taking into account that glass transition temperature of the previously prepared FL 5:1 PVP/VA mixture, as well as the rest of the experimentally determined points, is in good agreement with the G-T/K-B prediction (blue circles in Figure 8), the Tg value obtained for system established after crystallization should, with high accuracy, let us determine its concentration. 17 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

The phenomenon observed during crystallization of FL 5:1 PVP/VA is far from the commonly observed behavior in API-polymer systems. Additional efforts have been made to validate if this transformation occurs only in this particular concentration or if this is a universal feature for those components in binary mixture. As a consequence, the whole procedure was repeated for different initial concentrations (FL 10:1 PVP/VA) and obtained results led us to the exact same stable concentration. Analysis of the spectra collected above Tg ,of the FL 10:1 PVP/VA mixture, exhibit well resolved loss peak, which can be identify with the α-relaxation of the investigated drugpolymer mixture. Similarly to the previously presented concentration, this mode moves towards higher frequencies with increasing temperature and its intensity remains unchanged until T = 303 K. Above this temperature the crystallization process begins and the intensity of α-relaxation rapidly decreases. Moreover, similarly to the case of FL 5:1 PVP/VA, additional relaxation process emerges on the low frequency side of ε”(f). This process becomes more evident with the progress of crystallization, and at a certain temperature equal to 327 K (at which the devitrification ceased), this additional relaxation process remains visible on the dielectric loss spectra. It should be noted that that during further increase the temperature this – additional – relaxation process moves towards higher frequencies. Extrapolation of the VFT fits of the α- and α’-processes to 100 s, allowed us to determine the Tg values of examined system (right panel of Figure 9). Tg value (273K) determined on the basis the α-process is in very good agreement with the calorimetric data of FL 1:1 PVP/VA. Moreover, glass transition temperature obtained on the basis the additional (α’) process is equal to 316 K which was in a perfect agreement with results obtained for different initial concentration (i.e. FL 5:1 PVP/VA).

18 ACS Paragon Plus Environment

Page 18 of 27

Page 19 of 27

Flutamide 10:1 Kollidon VA64 ∆T = 2K

2 TC = 305 K Crystallization onset

Tg = 316 K

1

Tg = 273 K roc ess

0

α' -p

100

ess

-1

proc

-2

α-

10

-1

α - process

log τα

101

Dielectric loss ε"

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

α' - process

-4

10-2 10-1

100

101

102

103

104

105

2,8

3,0

3,2

3,4

3,6

-5

Temperature (1000/T)

Frequency [Hz]

Figure 9 Left panel presents the dielectric loss spectra of amorphous binary mixture of FL 10:1 PVP/VA. Black lines indicates the α process, red lines corresponds to the crystallization process and the blue lines represents the additional process. The right panel shows relaxation map of binary mixtures of FL 10:1 PVP/VA. Temperature dependence of τα(black circles) and τα’(blue circles) in the supercooled liquid has been described by VFT equations (red solid lines).

Based on this fact, one should consider this transformation as a typical behavior in case of supersaturated FL-PVP/VA systems.

Thermal properties, molecular dynamics and physical stability of amorphous mixture of flutamide and Kollidon VA64 in the molar ration 1:1 In the previous section it has been predicted that FL should be physically stable in the binary mixture with Kollidon VA64 in the molar ratio 1:1. In order to validate this prediction the binary system containing FL 1:1 PVP/VA has been prepared and measured by DSC, BDS as well as XRD techniques. In Figure 10 the DSC thermogram of FL 1:1 PVP/VA mixture has been compared with thermograms of pure FL, pure Kollidon VA64 as well as mixture of FL 5:1 PVP/VA. All of these DSC traces have been obtained during heating of the samples with a heating rate of 10 K/min. As can be seen the drug-polymer mixture in molar ratio 1:1 has glass transition temperature of 313 K.

19 ACS Paragon Plus Environment

Page 20 of 27

Flutamide FL 5:1 PVP/VA FL 1:1 PVP/VA Kollidon VA64 TC = 316 K

TC = 350 K

Heat Flow (endo up) [arb. units]

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

Heat Flow (endo up) [arb. units]

Molecular Pharmaceutics

Tg = 274 K Tg = 281 K Tg = 313 K Tg = 376 K

280

Heating rate: 10 K/min 270

285

300

315

330

345

FL 1:1 PVP/VA 300 320 340 Temperature [K] 360

375

390

Temperature [K] Figure 10Thermograms of amorphous: FL (a), FL 5:1 PVP/VA (b), FL 1:1 PVP/VA (c) and PVP/VA (d). Inset shows the close up of glass transition process of amorphous FL 1:1 PVP/VA

It is worth to note that the Tg value of the prepared FL 1:1 PVP/VA binary mixture is in good agreement with that obtained from the G-T/K-B prediction for the transformed, from FL 5:1 PVP/VA, sample. This result clearly indicates that crystallization in case of the FL 5:1 PVP/VA mixture leads to the establishment of new drug-polymer concentration with 1:1 molar ratio. To investigate the molecular dynamics of FL 1:1 PVP/VA binary mixture we employed BDS. The results of performed dielectric experiment are presented in the upper panel of Figure 11. As can be seen, the dielectric loss spectra collected below the glass transition temperature (red spectra) have not revealed the presence of any secondary relaxation processes – only NCL is visible. On the other hand, the α-relaxation process is very visible on the spectra registered at T>Tg.

20 ACS Paragon Plus Environment

Page 21 of 27

Frequency [Hz] 2

b)

299 K

103

∆T = 6 K

104

105

101

100

a) α-p roce ss α' proc from e ss FL 5 :1 PV P/VA

10-1

FL 1:1 PVP/VA NCL

-3

-4

-5

2,6

106 2 10

383 K

α - process

-2

-6

102

Tg = 316 K

0

-1

101

2,8

3,0

10-2

c)

10-4

T = 349 K

1

ε"/εmax

1

100

FL 1:1 PVP/VA additional process occured in FL 5:1 PVP/VA after crystallization 10-3

10-2

Temperature [1000/K]

Dielectric loss ε"

10-1

log τα

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

10-1

100

101

102

0,1 103

f/fmax

Figure 11 Temperature dependence of τα for FL 1:1 PVP/VA mixture in the supercooled liquid has been described by VFT equation (red solid line). Dielectric loss spectra of amorphous binary mixture of FL 1:1 PVP/VA (b). Black lines indicates the α process and the red lines corresponds to the NCL. Masterplot of two dielectric spectra obtained for both FL 1:1 PVP/VA and FL 5:1 PVP/VA after crystallization in temperature T = 349 K. Circles indicates primary relaxation for FL 1:1 PVP/VA and FL 5:1 PVP/VA, black and blue respectively.

Analysis of the dielectric loss spectra, which has been collected above the glass transition temperature of FL 1:1 PVP/VA, allows us to determine the temperature dependence of structural relaxation times. The τα(T) dependence is presented in Figure 11. Due to the fact that in the supercooled liquid region, the temperature evolution of τα usually shows nonArrhenius behavior, we parameterized these data over the entire T range by means of the VFT equation (τ∞ = -13,9; T0 = 245,8; B = 2550,6). From extrapolation the VFT fit to τα = 100 s, the kinetic glass transition temperature of the investigated mixture has been estimated as 316K (Tg = T(τα = 100 s)). This was in good agreement with the results acquired from DSC measurements as well as with the Tg value obtained after crystallization of FL 5:1 PVP/VA mixture. At this point, the thermal properties of glass transition itself lead us to the conclusion that binary mixture in ratio 1:1 is in good agreement with the mixture predicted from G-T equation. Moreover, we decided to compare the shape of the primary relaxation process in both the FL 1:1 PVP/VA and this additional process emerging during crystallization of FL 5:1 PVP/VA mixture (shown in Figure 11c). As can be seen in Figure 11c dielectric spectra are in perfect agreement which one can take as conclusive evidence identifying an additional process formed during crystallization of binary mixture in molecular ratio 5:1. Moreover, this 21 ACS Paragon Plus Environment

Molecular Pharmaceutics

might be taken as verification of the method where G-T equation can be used to determine limiting, stable concentration appearing during crystallization process from supersaturated mixture. There is still one thing we would like to point out, DSC as well as BDS experiments indicate that the investigated FL 1:1 PVP/VA mixture do not reveal any tendency toward recrystallization. On the DSC thermogram we cannot observe any exothermic peak as well as on the dielectric loss spectra the α-relaxation peak does not change its intensity even up to a melting temperature of pure FL. It is opposite to previously examined mixture with higher drug concentration. In order to confirm high physical stability of FL 1:1 PVP/VA system we decided to performed additional long-term XRD measurements. These studies were performed on the sample-filled capillaries stored at standard storage conditions (T = room T, ambient p and RH = 25%) and obtained results are shown in Figure 12. As can be seen on Xray diffraction patterns of FL 1:1 PVP/VA mixture exhibit an absence of sharp Bragg peaks. These results indicate that the examined system does not revealed any tendency towards cold crystallization for more than 270 days (up to date), which correlates with high physical stability. One might consider this as a confirmation of our hypothesis that based on the transformation observed during BDS experiment we indeed are able to appoint the solubility limits in the FL – PVP/VA systems. This is the crucial step in order to determine the concentration of binary mixture resistant to the devitrification process.

FL 1:1 PVP/VA after 200 days of storage at room temperature

Intensity [a.u.]

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 22 of 27

FL 1:1 PVP/VA immediately after vitrification

0

10

20

30

40

50

60

Scattering angle 2 Theta [°] Figure 12 X-ray diffraction patterns obtained for FL 1:1 PVP/VA mixtures in three different time periods.

22 ACS Paragon Plus Environment

Page 23 of 27

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

Conclusions In this paper, we evaluated the physical stability of amorphous drug flutamide by means of various experimental techniques such as DSC, BDS and XRD. All performed studies clearly indicate that this compound is highly physically unstable. As it has been proven the neat amorphous FL fully re-crystallizes after only 45 minutes of storage at temperature 293 K. Taking into account that disordered APIs have to be physically stable at standard storage conditions for a minimum of three years, vitrified FL requires stabilization. In order to suppress re-crystallization tendency of FL we mixed it with block co-polymer of PVP and PVA named Kollidon VA64 in molar ration FL 5:1 PVP/VA. This binary amorphous system is characterized by higher, equal to 281 K glass transition temperature and consequently also lower tendency towards devitrification than pure drug. Unfortunately, this amount of Kollidon VA64 polymer is not sufficient to fully stabilize the investigated drug – during both DSC and BDS experiments the amorphous FL undergoes the cold crystallization. It is worth to highlight that during the dielectric studies we were able to observe unusual, previously un-reported behavior. Namely, simultaneously with the beginning of the drug devitrification an additional loss peak appears on the low frequency side of ε”(f). This process was became more and more evident with the progress of crystallization. At a certain temperature, equal to 347 K, at which a decrease in intensity of the α-relaxation peak ceases, only this additional peak remains visible on dielectric loss spectra. Our analysis supported by additional DSC measurements clearly indicates that this additional loss peak reflects structural relaxation of different – highly physically stable – concentration of FL in binary mixture with Kollidon VA64. During the time of experiment, one can observe recrystallization of the excess of FL (under API supersaturated conditions) unveiling the stable drug-polymer system. Observed, during BDS experiment, transformation enables us to determine the limiting physically stable concentration of FL in PVP/VA polymer (saturated solution), which is equivalent to molar ratio 1:1. Described high physical stability of this unveiled system has been confirmed by means of long-term XRD measurements. According to our knowledge this is the first time when such a behavior has been observed by means of BDS. We realize that utilizing broadband dielectric spectroscopy in this particular measurements requires additional investigation, however we strongly believe that this method might become irreplaceable tool in predicting limiting polymer concentration needed to stabilize amorphous APIs. In order to test if this phenomenon is commonly occurring in API-polymer systems the authors strongly encourage further exploration of this kind of behavior.

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

ACKNOWLEDGMENT The authors are grateful for the financial support received within the Project No. 2015/16/W/NZ7/00404 (SYMFONIA 3) from the National Science Centre, Poland.

24 ACS Paragon Plus Environment

Page 24 of 27

Page 25 of 27

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

References Kenneth D. Kochanek, M.A., Sherry L. Murphy, B.S., JiaquanXu, M.D., and Betzaida

(1)

Tejada-Vera. Deaths: Final Data for 2014. NVSR Volume 65, Number 4.122 pp. (PHS) 2016-1120. Labrie, F., Mechanism of action and pure antiandrogenic properties of flutamide.

(2)

Cancer 1993, 72: 3816–3827. Neri R, Florance K, Koziol P, Van Cleave S. A biological profile of a non-steroidal

(3)

antiandrogen, SCH 13521 (4’-nitro-3’-trifluoromethyl-isobutyranilide). Endocrinology 1972; 91:427-37. Madhusudhan B., Rambhau D., Apte S. S., Gopinath D. Oral Bioavailability of

(4)

Flutamide from 1‐O‐Alkylglycerol Stabilized o/w Nanoemulsions. J. Dispers. Sci. Technol. 2007, 28, 1254—1261. Patil, B. S., H. S. Mahajan, and S. J. Surana. Preparation and evaluation of

(5)

flutamidemucoadhesivemicroparticles

loaded poloxamer

suppositories.

Journal

of

Pharmaceutical Investigation. 2016, 46.2: 139-146. Kaminski, K.; Adrjanowicz, K.; Wojnarowska, Z.; Grzybowska, K.; Hawelek, L.;

(6)

Paluch, M.; Zakowiecki, D.; Mazgalski, J. Molecular dynamics of the cryomilled base and hydrochloride ziprasidones by means of dielectric spectroscopy. J. Pharm. Sci. 2011 , 10 , 22479. Grzybowska, K., Chmiel, K., Knapik-Kowalczuk, J., Grzybowski, A., Jurkiewicz, K.,

(7)

& Paluch, M. Molecular Factors Governing the Liquid and Glassy States Recrystallization of Celecoxib in Binary Mixtures with Excipients of Different Molecular Weights. Mol. Pharmaceutics, 2017, 14(4), 1154-1168. Murdande, B. S.; Pikal, M. J.; Shanker, R. M.; Bogner, R. H. Solubility Advantage of

(8)

Amorphous Pharmaceuticals: I. A Thermodynamic Analysis. J. Pharm. Sci. 2010, 99(3), 1254-1264. Murdande, B. S.; Pikal, M. J.; Shanker, R. M.; Bogner, R. H. Solubility Advantage of

(9)

Amorphous Pharmaceuticals, Part 3: Is Maximum Solubility Advantage Experimentally Attainable and Sustainable? J. Pharm. Sci. 2011, 100, 4349−4356. (10)

Kothari, K.; Ragoonanan, V.; Suryanarayanan, R. Influence of Molecular Mobility on

the Physical Stability of Amorphous Pharmaceuticals in the Supercooled and Glassy States. Mol. Pharmaceutics 2014, 11, 3048−3055.

25 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

(11)

Karmwar, P.; Graeser, K.; Gordon, K. C.; Strachan, C. J.; Rades, T. Investigation of

properties and recrystallization behavior of amorphous indomethacin samples prepared by different methods. Int. J. Pharm. 2011, 417, 94−100. (12)

Dengale, S. J.; Grohganz, H.; Rades, T.; Lö bmann, K. Recent advances in co-

amorphous drug formulations. Adv. Drug Delivery Rev. 2016, 100, 116−125. (13)

Kothari, K.; Ragoonanan, V.; Suryanarayanan, R. The Role of Drug − Polymer

Hydrogen Bonding Interactions on the Molecular Mobility and Physical Stability of Nifedipine Solid Dispersions. Mol. Pharmaceutics 2015, 12, 162−170. (14)

Knapik J, Wojnarowska Z, Grzybowska K, et al. Physical stability of the amorphous

anticholesterol agent (ezetimibe): the role of molecular mobility. Mol Pharmaceutics. 2014;11:4280-4290. (15)

Knapik J, Wojnarowska Z, Grzybowska K, Tajber L,Mesallati H,Paluch KJ,Paluch

M. Molecular Dynamics and Physical Stability of Amorphous Nimesulide Drug and Its Binary Drug-Polymer Systems. Mol Pharmaceutics. 2016;13(6):1937-46 Bhugra, C.; Shmeis, R.; Krill, S. L.; Pikal, M. J. Prediction of Onset of Crystallization from Experimental Relaxation Times. II. Comparison between Predicted and Experimental Onset Times. J. Pharm. Sci. 2008, 97, 455−472. (17) Andronis, V.; Zografi, G. Molecular mobility of supercooled amorphous indomethacin, determined by dynamic mechanical analysis. Pharm. Res. 1997, 14, 410−414. (18) Zhou, D.; Zhang, G. G. Z.; Law, D.; Grant, D. J. W.; Schmitt, E. A. Thermodynamics, molecular mobility and crystallization kinetics of amorphous griseofulvin. Mol. Pharmaceutics 2008, 5, 927−936. (19) Bhardwaj, S. P.; Arora, K. K.; Kwong, E.; Templeton, A.; Clas, S. D.; Suryanarayanan, R. Correlation between molecular mobility and physical stability of amorphous itraconazole. Mol. Pharmaceutics 2013, 10, 694−700. (20) Bhardwaj, S. P.; Suryanarayanan, R. Molecular Mobility as an Effective Predictor of the Physical Stability of Amorphous Trehalose. Mol. Pharmaceutics 2012, 9 (11), 3209−3217. (21) Wojnarowska, Z., Rams-Baron, M., Knapik, J., Ngai, K. L., Kruk, D., & Paluch, M.. (16)

Dynamic Properties of Glass-Formers Governed by the Frequency Dispersion of the Structural α-Relaxation: Examples from Prilocaine. The Journal of Physical Chemistry B, 2015;119(39), 12699-12707. (22) (23) (24)

Kohlrausch, R. Pogg. Ann. Phys.1847, 12, 393 Williams, G.; Watts, D. C. Trans. Faraday Soc. 1970, 66, 80 Paluch, M., Knapik, J., Wojnarowska, Z., Grzybowski, A., & Ngai, K. L. (2016).

Universal Behavior of Dielectric Responses of Glass Formers: Role of Dipole-Dipole Interactions. Physical review letters, 116(2), 025702 26 ACS Paragon Plus Environment

Page 26 of 27

Page 27 of 27

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

(25)

Szczurek, J., Rams-Baron, M., Knapik-Kowalczuk, J., Antosik, A., Szafraniec, J.,

Jamroz, W., Dulski M., Jachowicz R., Paluch, M.. Molecular dynamics, recrystallization behavior and water solubility of amorphous anticancer agent bicalutamide and its polyvinylpyrrolidone mixtures. Mol. Pharmaceutics. 2017 14(4), 1071-1081. (26)

Kremer, F., Schönhals, A., Eds. Broadband Dielectric Spectroscopy; Springer: Berlin,

2003 (27)

Vogel H.; Das Temperaturabhangigkeitgesetz der Viskosität von Flüssigkeiten. J.

Phys. Z. 1921, 22, 645−646. (28)

Fulcher G. S.; Analysis of Recent Measurements of the Viscosity of Glasses. J. Am.

Ceram. Soc. 1925, 8, 339−355. (29)

Tammann, G.; Hesse, W. Die Abhängigkeit der Viscosität von der Temperatur bie

unterkühlten Flüssigkeiten. Z. Anorg. Allg. Chem. 1926, 156, 245−257. (30)

N.S. Trasi, L.S. Taylor Effect of additives on crystal growth and nucleation of

amorphous flutamide. Cryst. Growth Des. 2012, 12, pp. 3221–3230 (31)

M. Rams-Baron, Z. Wojnarowska, K. Grzybowska, M. Dulski, J. Knapik, K.

Jurkiewicz, W. Smolka, W. Sawicki, A. Ratuszna, and M. Paluch; Towards a better understanding of the physical stability of amorphous anti-inflammatory agents: the role of molecular mobility and molecular interaction patterns. Mol Pharmaceutics. 2015 Oct 5;12(10):3628-38. (32)

Xie, Tian; Taylor, Lynne S. Effect of Temperature and Moisture on the Physical

Stability of Binary and Ternary Amorphous Solid Dispersions of Celecoxib. J. Pharm. Sci. 2017, 106 (2017) 100-110 . (33)

Andronis, V., Yoshioka, M., and Zografi, G. Effects of sorbed water on the

crystallization of indomethacin from the amorphous state. J. Pharm. Sci. 1997; 86: 346– 351 (34)

Baird, J. A.;Taylor,L. S.Evaluation of amorphous solid dispersion properties using

thermal analysis techniques Adv. Drug Delivery Rev.2012,64(5)396–421 (35)

Paudal A, Van Humbeeck J, Van den Mooter G. Theoretical and experimental

investigation on the solid solubility and miscibility of naproxen in poly(vinylpyrrolidone). Mol Pharmaceutics. 2010;7:1133-11148. (36)

Couchman P, Karasz F. A classical thermodynamic discussion of the effect of

composition on glass transition temperatures. Macromolecules. 1978;11: 117-119.

27 ACS Paragon Plus Environment