The characterization and dissolution performances of spray dried solid dispersion of ketoprofen in hydrophilic carriers

Siok-Yee Chan*,Yin-Ying Chung,Xin-Zi Cheah,ErynYen-Ling Tan, Joan Quah

School of Pharmaceutical Sciences,Universiti Sains Malaysia,11800 Pulau Pinang,Malaysia

The characterization and dissolution performances of spray dried solid dispersion of ketoprofen in hydrophilic carriers

Siok-Yee Chan*,Yin-Ying Chung,Xin-Zi Cheah,ErynYen-Ling Tan, Joan Quah

School of Pharmaceutical Sciences,Universiti Sains Malaysia,11800 Pulau Pinang,Malaysia

ARTICLEINFO

Article history:

Received 12 January 2015

Received in revised form 6 April 2015

Accepted 23 April 2015

Available online 14 May 2015

Solid dispersion Amorphous

Polymer

PVP

PVPVA

ketoprofen

Solid dispersion is one of the most promising strategies to improve oral bioavailability of poorly soluble API.However,there are inconsistent dissolution performances of solid dispersion reported which entails further investigation.In this study,solid dispersions of ketoprofen in three hydrophilic carriers,i.e.PVP K30,PVPVA 6:4 and PVA were prepared and characterized.Physical characterization of the physical mixture of ketoprofen and carriers shows certain extent of amorphization of the API.This result is coinciding to evaluation of drug-polymer interaction usingATR-FTIR whereby higher amorphization was seen in samples with higher drug-polymer interaction.XRPD scanning confirms that fully amorphous solid dispersion was obtained for SD KTP PVP K30 and PVPVA system whereas partially crystalline system was obtained for SD KTP PVA.Interestingly,dissolution profiles of the solid dispersion had shown that degree of amorphization of KTP was not directly proportional to the dissolution rate enhancement of the solid dispersion system.Thus,it is concluded that complete amorphization does not guarantee dissolution enhancement of an amorphous solid dispersion system.

©2015 The Authors.Production and hosting by Elsevier B.V.on behalf of Shenyang Pharmaceutical University.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Today,many newly identified active pharmaceutical ingredients(API)are classified as low solubility,which give rise to a low bioavailability when administered orally.These APIs are usually classified as BCS Class II or Class IV,where solubility is the limiting step for absorption.There are many strategies introduced to increase the solubility and bioavailability of BCS Class II or Class IV compounds.These include particle size reduction,formation of salt,formation of co-crystal,inclusion complex with cyclodextrin,amorphization and formation of solid dispersion[1,2].Solid dispersion is one of the most promising strategies to improve oral bioavailability of poorly solubleAPI[2,3].The preparation of solid dispersion involves formation of eutectic mixture of drugs with their carriers either via melting or fusion through solvent evaporation of their physical mixtures[4].

Conventional concept of a successful production of solid dispersion is related to the formation of amorphous solid dispersion(ASD).It is a metastable solid state that possesses higher solubility as compared to their corresponding crystalline structure.When amorphous solid dispersion is exposed to aqueous media such as distilled water or gastric fluid,the polymer which acts as a carrier will be dissolved,thus the API can be released as very fine,colloidal particles from the solid dispersion dosage form.The increase in surface area of the API as a result of molecularly dispersed or also known as amorphous dispersion is claimed to be one of the main reasons which results in an increased dissolution rate of drug,hence increasing the bioavailability of drug.Besides,polymeric system in the solid dispersion has been reported to facilitate the dissolution process.This phenomenon is detailed in a review by Craig Duncan whereby two types of dissolution process of solid dispersion were identified[5].On one hand,the carrier shall be able to completely dissolve the API before it could be fully released into the bulk.On the other hand,due to the low solubility of API in polymer,the hydrophilic polymer is wetted but the API would release intact in its highly soluble form(such as amorphous state)without dissolution into the carrier.In the former case,dissolution process is considered as carrier dependent whereas the latter case is termed as drug dependent dissolution process in which the physical properties of the drug is the determinant of the overall dissolution performance.In this theory,solubility of anAPI in a polymeric system is a crucial parameter which leads to the vast study of solid solubility of API in polymeric system in the recent years[6,7].In addition to solubilisation effect of the polymeric carrier,high molecular weight of the polymer carrier could also kinetically stabilize amorphous state of API through interaction with API.Examples of polymer in which the stabilization of API are proven include PVP,HPMC and HPMC phthalate[8-10].Furthermore, polymers also inhibit recrystallization of amorphous drug in solid dispersion,thus forming a stable ASD[1,2].Hence,it is theorized that the formation of fully amorphous solid dispersion would enhance the dissolution rate and solubility of the formulated API.

Among the vast publications under the umbrella of solid dispersion,appallingly,a few reports have revealed inconsistent dissolution performances of fully amorphous solid dispersion(SD)with the limited discussion emphasis on the said issue[11-15].Only one cited paper has highlighted the issue by naming the observed dissolution performance of its investigated SD as“anomalous dissolution behaviour”[14].Moreover, in the recent years,despite the high prevalence of publication on the subject of SD,inconsistence dissolution phenomenon of SD system is under reported,possibly due to the success led publication.Even though this issue seems minor as it was shown in only a few cited papers,however,the ignorant of this inconsistency together with the limited knowledge in tackling the underlying causes of this inconsistency would lead to production of an SD product that fail the quality assurance.This may further cause the drainage of effort and investment in the manufacturing of SD at large scale.Therefore,more investigations are urgently needed in this realm to fill in the knowledge gap in the thorough understanding of the real performances of ASD.

In order to unfold the possible underlying causes of the dissolution inconsistency of SD,in the current study,the aspect of carrier was focused on.Here,three hydrophilic polymers were used as a carrier in the preparation of spray dried solid dispersion,i.e.PVP K30,PVPVA 6:4 and PVA.PVP K30 and PVPVA 6:4 are both amorphous polymers.These polymers are widely used in the production of solid dispersion due to the good stabilizing effect on the production of ASD[8-10].Many have reported the dissolution enhancement of API by using both PVP and PVPVA as carrier systems,even though PVPVA has reported to be slightly more hydrophobic than PVP[16,17].On the other hand,PVA is a water soluble crystalline polymer,with a high value of viscosity at 4%,i.e.40 mPa s.These three polymers are chosen based on the ability in the production of different amorphicity SD system,in order to relate the amorphicity to dissolution performances of the SD of a poorly soluble drug,i.e.ketoprofen.

ketoprofen is a nonsteroidal anti-inflammatory drug of the propionic acid derivative group that is used for relief of pain and inflammation,treatment of rheumatoid arthritis,osteoarthritis and other muscle and joint conditions[18-21].It is classified as BCS class II compound with no previous report of polymorphic forms.It possesses a low theoretical glass transition temperature(Tg)at circa-5 to-6°C[22]which may present alteration of molecular mobility at biological temperature during dissolution of its amorphous form.Its poor solubility and low Tg characteristic renders it to be a suitable model of poorly soluble API in the current study.Here,ketoprofen is formulated into amorphous solid dispersion by utilizing spray drying method,accompanied with polymers of amorphous property such as PVPVA 6:4,PVP K30 and polymers of crystal property such as PVA.The resultant solid dispersions of ketoprofen were characterized and relate to their dissolution performances correspondingly.

2. Materials and methods

2.1. Materials

The raw ketoprofen(KTP)was purchased from AFINE Chemical LTD,BN:1102017.Three hydrophilic polymers were chosen as the carrier systems in the current study,i.e.Polyvinylpyrrolidone K30(PVPK 30),Vinylpyrrolidone-vinyl acetate copolymer Kollidon®VA 64(PVPVA)and Polyvinyl alcohol(PVA).Both Polyvinylpyrrolidone K30,Kollidon®30(average MW 40,000)and Vinylpyrrolidone-vinyl acetate copolymer Kollidon®VA 64 were generous gifts from BASF distributed via ELITE ORGANIC SDN.BHD.Polyvinyl alcohol was obtained from BDH Laboratory,England.

2.2. Preparation of physical mixture and spray dried solid dispersion

There are a few solution-based technologies to produce an amorphous solid dispersion,such as precipitation by addi-tion of antisolvents,mechanical activation,hot melt extrusion, freeze drying and spray drying[1].In this study,spray drying method was adopted due to the utilization of gentle temperatures and little exposure time as compared to other methods such as hot melt extrusion[23].To produce an amorphous solid dispersion,this method works by constantly dividing a liquid stream into very fine droplets through atomization process.The droplets then reach the glass compartment and contact with hot gas,thus get dried into fine particles,which are then separated from the drying gas and collected into a chamber.

Total of 5.0 g of KTP and polymer mixtures in 30%KTP loading were prepared using 50%v/v ethanol in water.The spray-dried solid dispersion products were prepared using a Buchi mini-spray dryer B-290,with operating parameters set at:inlet temperature 80°C;outlet temperature 60°C;feed rate 5 ml/min;aspirator 100°C.It is worth emphasizing that the inlet temperature used shall not be exceeding 80°C in order to control the outlet at circa 60°C so to avoid the transition of the resultant solid into its semisolid state in the collecting chamber.This is crucial because of the low theoretical Tg of the mixture which will undergo solid transition process if it was kept in a temperature higher than its Tg(theoretical Tg of the mixtures are calculated in section 3.3,Table 4).The yield obtained was approximately 40-50%.

On the other hand,physical mixture(PM)of KTP and polymers were prepared by trituration mixing of the weighed powders(30%KTP in polymer)with a mortar and pestle.

2.3. Characterization of the spray-dried KTP-polymer solid dispersions

青海省海西州德令哈市民族学校的索卫华对海西地区民族中学学生数学学习中导致数学基础较差的社会环境、家庭教育、教师的知识面及传授方法等智力和非智力因素进行了剖析．中央民族大学苏傲雪对全国31所内地新疆高中班学校的932名教师和1 873名学生进行了问卷调查和测试，对影响内地新疆班学生数学学业成绩的因素作了统计分析．云南曲靖师范学院孙雪梅综合应用调查测试、作业分析、口头报告和比较研究方法，从表征视角调查了七年级彝族学生数学学习现状，发现七年级彝族学生在解决数学问题上的数学表征水平发展不均衡，而七年级彝族学生与同年级汉族及其他少数民族学生没有显著差异．

Solid state characterization of the prepared solid dispersions and physical mixtures of KTP and polymer were carried out by scanning in X-ray powder diffraction(XRPD),differential scanning calorimetry(DSC)and Attenuated total reflectance-Fourier transform infrared(ATR-FTIR).

2.3.1. XRPD analysis

XRPD analysis of raw materials,PMs and solid dispersion were performed with an XRPD,Bruker D8 Advance equipped with a copper X-ray Tube(1.54060 Å).Samples were pressed into a sample holder to generate a flat and smooth plane surface.The samples were then exposed to an X-ray beam with voltage of 40 kV and a current 40 mA.All measurements were performed from 3°to 50°(2θ)coupled with scanning speed of 0.02°/ step and 1 s for every scan step to cover the characteristic peaks of the crystalline KTP.Area under the diffracted peaks was calculated by integrating the XRPD diffractograms using ORIGIN 8.0 software.

2.3.2. Water content measurement

Water content of the obtained samples was determined by heating approximately 10 mg of sample using a hotplate at 100°C for 10 min.The weight of the samples before and after heating was recorded.The percentage of weight loss after 10 min of heating at 100°C was taken as the function of water content in the sample.

2.3.3. DSC measurement

DSC measurements were performed with Perkin Elmer Pyris 6 DSC.Approximately 2-4 mg of samples was packed in crimped aluminum pan and heated under dry nitrogen purge.Samples were heated from 20°C to 200°C at 10°C/min,all samples were scan in duplicate.The results were analyzed using Pyris Data Analysis.

2.3.4. ATR-FTIR spectroscopy

AttenuatedTotal Reflectance-FourierTransform Infrared(ATRFTIR)spectra were recorded over a wavenumber range of 500 cm-1to 4000 cm-1with a resolution of 4 cm-1and 32 scans using Thermo Nicolet FTIR Nexus spectrometer coupled with ATR accessory.The spectra were analyzed using OMNIC software.

To quantify the API content in a medium sample,a Perkin-Elmer Lambda XLS UV/VIS spectrophotometer(USA)was used. λmaxspecified for KTP which was identified to be devoid of any interference from the added excipients at 259 nm was used. Calibration curves were constructed for known concentration of KTP in distilled water at 259 nm by using Beer Lambert plots.Each point in the calibration line was an average value of three measurements.

2.5. Dissolution studies

Dissolution tests were performed using paddle method in a calibrated Varian VK7000 Dissolution Apparatus.900 ml of distilled water was used as a dissolution medium.The dissolution medium was set at 37.0±0.5°C and the paddle speed of 50 rpm was used.Pure drug,polymer,physical mixtures and spray-dried products were sieved to a controlled particle size range of 100-106µm.Then,the samples equivalent to 20 mg of KTP(based on formulation composition)were added to the dissolution medium upon the start of dissolution experiment.10 ml of the samples was withdrawn at 2,5, 10,15,20,30,40,50,60 and 120 min.The volume of dissolution medium withdrawn was immediately replaced by introducing the same volume of fresh medium into the dissolution vessel.The samples were then filtered with mixed cellulose ester microfilter of 0.45µm pore size(MFS membrane filter,Lot no.41CLCA)and analyzed for content of KTP using UV-vies spectroscopy at 259 nm.To compare the dissolution performances between the PM and SD systems,similarity factor(f2),which could be expressed by Equation(1),was used[24].

where n is the number of time points,Rtis the percentage of drug release of a reference batch at the time t and Ttis the percentage of drug release of the comparison batch at time t. When f2is greater than 50(i.e.50-100),this indicates the sameness or equivalence of the both compared profiles.Conversely, when f2is less than 50 is an indication that the two profiles are different.In this study,dissolution profiles were compared up to a point after 80%drug release of the formulation. Similarity factor is an independent approach that measuresthe similarity in percentage between the 2 profiles of dissolution.This model was utilized as a tool to provide a gross idea on the rank order of the dissolution performance differences between the prepared amorphous SD and their PM systems.

3. Results and discussion

3.1. X-ray powder diffraction analysis

Amorphicity of a solid dispersion has previously been concluded to be pertinent in ensuring the final performances of solid dispersion.In light of this,the current study is trying to complement the existing knowledge by forming solid dispersion and relate the essentiality of amorphousness to the final performance of solid dispersion.To do this, XRPD was employed to check the crystallinity of all the prepared samples.Fig.8 shows the XRPD diffractograms of all the physical mixture and their corresponding spray dried samples.

The diffractograms showed the characteristic diffracted peak of ketoprofen in the region of 13°to 25°(Fig.1A).These peaks were also noted in all the physical mixtures which indicate the presence of crystalline KTP in those samples(Fig.1B-D).Crystallinity of each sample was quantitatively determined by calculating area under the diffracted peaks of the XRPD diffractogram.The results were tabulated in Table 1.It is interesting to note that the crystallinity of KTP has been reduced merely by trituration using mortar and pestle.There is a trend of in the crystallinity reduction whereby area under the diffracted peaks of PM KTP PVP K30 is lowest followed by

Halo patterns were shown in the diffractograms of spray dried 30%KTP in PVP K30 and PVPVA 6:4,respectively(Fig.1E and F respectively),which indicated the absence of crystalline material in these samples.In contrast,characteristic peaks of the ketoprofen in the region of 13°to 25°and PVA at 19.4°are detected in spray dried 30%KTP in PVA which correspond to 19.99%of crystallinity(Fig.1G and Table 1).This ascertains the presence of crystalline traces of ketoprofen in the SD KTP PVA sample.Here,the amorphous PVP carriers have shown to be able to produce fully ASD in comparison to the partially crystalline carrier PVA.
It is to bear in mind that,according to the theoretical assumption,the presence of crystalline material in the solid dispersion may deteriorate dissolution performance of the system as compare to the same system without crystalline traces[3].This result shall be compared to the dissolution performance which will be presented and discussed in the dissolution section of this paper.

Table 1-Crystallinity of PM and SD systems calculated from area under the peaks diffracted in XRPD diffractograms.

3.2. Water content of physical mixture and solid dispersion

Water content has been shown to be a critical parameter for dissolution performance and stability of solid dispersion system [26,27].Product with lower moisture content is preferred over that with higher moisture content.This parameter may determine physical stability and overall performances of the solid dispersion as it will affect crystallization tendency of amorphous solid dispersion.Hence it is critical to characterize the water content of solid dispersion.Fig.2 displays water content of PM and their corresponding spray dried products.Based on Fig.2,it is noted that all the spray dried products possess lower water content as compared with their corresponding physical mixture.The low moisture content of SD products was due to the water evaporation from the mixture in spray drying condition.The lower the water content of the mixture,the more stable the mixture.

SD 30 KTP and PVA shows the lowest water content when compared with the PVPVA 6:4 and PVP K30.The presence of high water content in SD KTP and PVPVA 6:4 and PVP K30 shows that the products are not completely dried after spray drying. Besides,it may also attribute to the hygroscopicity of the PVP polymers[28].

3.3. DSC analysis

The solid dispersion and their corresponding carriers system employed in the current study were characterized using DSC. Fig.3 presents DSC thermograms for the carrier system.DSC thermograms of PVP K30 and PVPVA 6:4 reveal glass transition temperatures at circa 165.03°C and 107.53°C,respectively, with the absence of any melting peak(Fig.3A and B).These results suggest amorphous characteristics of PVP polymers. Besides,a broad endotherm ranging from 20 to 100°C is observed in both the thermogram of pure PVP K30 and PVPVA 6:4 which was attributed to the water loss from the hygroscopic polymers upon heating.Conversely,no transition was seen in thermogram of PVA but a board endotherm was observed at temperature circa 190.03°C due to its crystalline property(Fig.3C)which has been confirmed in the XRPD data (Fig.1H).In order to detect theTg of PVA,it was reheated after complete melting in the first cycle,the Tg was then obtained from the second heating cycle which is shown in Fig.3D with the values detected at 73.61°C.

In order to understand the solid state changes of the solid dispersion system after spray drying,solid state of PM system is determined.Fig.4 shows the DSC thermograms of KTP and the physical mixtures prepared in this study.Based on the DSC curves,KTP shows a sharp melting endotherm at 96.06°C with enthalpy of fusion(ΔH)110.96±17.5 J/g(Fig.4A).A broad endotherm and a sharp peak are observed in the thermogram of physical mixture(PM)30%w/w KTP+PVP K30 and PM 30% w/w KTP+PVPVA 6:4(Fig.4B and C)which are corresponding to both the loss of residual water in the polymer and the melting endotherm of KTP,respectively.On the other hand,only a sharp melting peak was seen with PM KTP PVA due to the less hygroscopicity of PVA that doesn’t give rise to the broad water loss signal(Fig 4D).

Furthermore,the melting enthalpy,ΔHPM,of the PM in the DSC thermograms of physical mixture could be used to estimate the percentage of crystallinity of the KTP remain in the mixture.The crystallinity could be calculated using Eq.(2)and Eq.(3):

Where ΔHPureKTPis the enthalpy of pure ketoprofen, ΔHKTPcorrected in PMis enthalpy attributed by the KTP in the physical mixture,m is total weight of physical mixture and mKTPinPMis weight of KTP in the PM.The degrees of crystallinity calculated from DSC measurements are listed in Table 2.

Based on the values calculated,all the investigated systems reveal less than 100%of crystallinity despite the simple mixing process using a mortar and pestle.This result is in agreement to the deduction obtained from XPRD data(section 3.1, Table 1)whereby certain portion of the crystalline KTP has turned into amorphous through simple trituration with the carrier polymer.Similar observation has been reported before for ibuprofen and PVP whereby a spontaneous conversion of the physical mixtures into a stable glasslike form is observed due to the intense drug polymer interaction between the studied API and PVP[25].However,as far as DSC measurement is concerned,the possible interference from the dissolution of crystalline KTP into the carrier system upon heating in DSC shall not be excluded.

Fig.5 displays DSC thermograms of amorphous ketoprofen and the SD system prepared in this study.The absence of melting peak in SD30 KTP+PVP K30 and SD30 KTP+PVPVA 6:4(Fig.5B and C)indicates the absence of crystalline trace of KTP in the prepared spray dried system.This proposes amorphousness of the prepared products.This might be related to the intermolecular hydrogen bonding between KTP and PVP K30 and/or loss of drug mobility as the drug was entrapped in polymer after evaporation of solvent during the spray drying process[29].As predicted from the XRPD data,a small endotherm was detected at onset of 80.78°C in the thermogram of SD 30 KTP+PVA(Fig.5D)which is corresponding to the depressed melting point of ketoprofen.This phenomenon shows the presence crystalline trace in spray dried product of KTP in PVA.The calculated crystal traces in the spray dried product of KTP-PVA using Eq.(2)and Eq.(3)is approximated to be 5.22%. This value is in contrast to the crystallinity predicted from area under the diffracted peaks in XRPD diffractogram,i.e.approximately 19.99%.This may be due to the intimate contact between the KTP and polymer in SD sample that lead to the significant interference of KTP dissolution into the polymer bed upon heating in DSC.

Comparison between the experimental Tg and theoretical Tg calculated using Gordon-Taylor equation in Eq.(4)is carried out[30].

Table 2-Crystallinity of ketoprofen in physical mixture based on DSC data.

Table 3 lists all the numeric values used to calculate the theoretical Tg whereas Table 4 compares the experimental Tg and theoretical Tg of all the investigated solid dispersion systems.According to Table 4,the experimental and theoretical values differ significantly.This deviation occurs due to nonideal mixing of the spray-dried products[31].The lowering of Tg may be explained by incomplete drying of the product in the low inlet temperature during the spray drying process.This could be further evidenced by the detected water content of approximately 5-10%for all the SD systems(Fig.2).Besides, hydroscopicity of PVP polymer which absorbed a big amount of moisture in its structure during preparation of DSC scanning may also lead to the lowering of Tg via plasticization of water[32].Nevertheless,a single Tg detected for SD KTP PVP K30 and SD KTP PVPVA 6:4 systems devoid any other thermal event is a conventional indication of homogenous mixing ofsolid dispersion.Overall,Tg of the solid dispersion system of KTP-PVP K30 and KTP-PVPVA 6:4 has increased from the low Tg of the pure KTP(Fig.5A).The increase ofTg implies the lower molecular mobility in comparison to the pure KTP which possess a Tg at a low temperature of-3.78°C(Fig.5A).This might contribute to the better stability of the amorphous solid dispersion system.The enhanced of Tg is not seen in SD KTP PVA as the amorphousness of this system is too low to be detected.

Table 3-Values used to calculate theoretical Tg based on Gordon-Taylor equation.

Table 4-The comparison between theoretical Tg and experimental Tg of the investigated system.

3.4. Infrared spectra analysis

Drug-polymer interaction has been reported to be important for physical stability of a solid dispersion system[9].More recently,its importance was widely discussed for dissolution performance of a product whereby a high drug-polymer interaction may render a better dissolution performance for an immediate release dosage form[33,34].With that in mind,ATRFTIR was used in the current study to investigate the KTP interaction with the three investigated carriers system.Table 5 displays the chemical structure of the compounds employed in the current study.

Fig.6 shows ATR-FTIR spectrum of crystalline KTP,amorphous KTP PM of KTP-PVP K30 and its SD systems.ATR-FTIR spectrum of crystalline KTP indicates two main peaks at region of 1695 cm-1and 1653 cm-1which corresponds to the C=O stretching of carbonyl group(Fig.6B).These peaks are broadening in melt quench of ketoprofen with an extra peak at 1738 cm-1(Fig.6C).This broadening was attributed to the alteration of solid state of KTP from crystalline to amorphous form after melt quenching.Besides,the fingerprint region shows a triplet in region of 704 cm-1for crystalline KTP and only doublet peak for melt quench sample(Fig.6C).These characteristic peaks could be used to probe the solid state of KTP with the triplet peaks represent crystalline and doublet peaks represents complete amorphousness of KTP(compare Fig.6B and C).

Examining the ATR-FTIR spectra of PM KTP PVP K30(Fig.6D) in comparison to crystalline KTP(Fig.6B),there is no apparent downshift of the PM spectra.However,in comparison to the pure PVP K30 spectra(Fig.6A),the characteristic peaks of C=O stretching of the pyrole group of PVP K30 wasdownshift after physical mixed with KTP.This indicates certain degree of drug polymer interaction via hydrogen bonding of OH KTP to the C=O group of PVP K30.After spray drying of PVP K30 with KTP,the characteristic peak at the fingerprint region of 704 cm-1is absent in the IR spectra of SD system (Fig.6E),which indicates the amorphousness of the SD system. Furthermore,the OH peaks of KTP in IR spectra of SD were further broadened,showing a less defined peak at 2976 cm-1. An apparent downshift of the wavenumbers of amorphous KTP from 1738 cm-1to 1722 cm-1and 1659 cm-1to 1655 cm-1were seen in the IR spectra of SD KTP PVP K30(Fig.6E).The significant down-shift of the wavenumber in the SD system signifies a higher degree of KTP-PVP K30 interaction as compared to its PM system.

Table 5-Chemical structure of the materials used in this study.

Similar trend is observed in ATR-FTIR spectra of PVPVA system whereby SD KTP PVPVA reveals doublet peaks at fingerprint region of 600-750 cm-1with the absent of peak 704 cm-1which shows the complete amorphousness of the SD system (Fig.7C).Besides,the C=O stretching of the carbonyl moiety of amorphous KTP and pyrole moiety of PVPVA 6:4 were both downshifted from 1659 cm-1and 1674 cm-1to 1655 cm-1in IR spectra of SD system(compare IR spectrum of Fig.7C,A and E).However,it was found that the C=O stretching(1732 cm-1) of the vinyl acetate moiety in both PM and SD preparations of KTP-PVPVA 6:4 did not shift.Therefore it was believed that the main interactions between KTP and PVPVA 6:4 occurs preferentially at the C=O group of the pyrole group rather than the C=O in vinyl acetate group.Apart from that,the OH stretching of carboxylic acid observed in amorphous KTP at 3061 cm-1is also shifted down to 3057 cm-1.The apparent shift of both the C=O and OH implies the participation of these moieties in KTP-PVPVA 6:4 hydrogen bond interactions of the SD system.

Unlike PVP and PVPVA carrier system,IR spectrum of PVA SD system reveals a different trend(Fig.8).PVA indicates a broad peak at O-H stretching region,i.e.3335 cm-1which corresponds to the O-H stretching of alkane.The bands at 1712 cm-1and 1716 cm-1of the carbonyl group are attributed to the absorption residual acetate groups due to the manufacture of PVA from hydrolysis of polyvinyl acetate(Fig.8A)[35].After the spray drying process of PVA with KTP,the triplet in the region of 704 cm-1is remained in the IR spectrum of the SD system (Fig.8E).This indicates the presence of crystal traces in the SD system which is in agreement to the result obtained for both the XRPD as well as DSC measurement.Since there is no proton acceptor moiety in PVA,thus no drug-polymer interaction is expected in KTP-PVA system.This prediction is consistent to the IR spectra of SD which shown merely a summation IR spectra of crystalline KTP,amorphous KTP and PVA.

To summarize the infrared results,SD of KTP PVP K30 and PVPVA 6:4 are presented with fully amorphous characteristic with certain extent of drug-polymer interaction.Unlike the SD KTP PVP K30,the presence of shoulder 1705 cm-1in SD KTP PVPVA 6:4 and non-shifting of C=O from acetate group of PVPVA 6:4 inferred limited interaction involving this group moiety of carbonyl in PVPVA 6:4 molecule.Hence,it is concluded that KTP PVP K30 reveals a higher drug-polymer interaction in comparison to KTP PVPVA 6:4 system.On the other hand,SD KTP PVA is suggested to be partially crystalline devoid any drug-polymer interaction.Overall,the intensity of drug-polymer interaction is in the trend of KTP PVPK30>KTP PVPVA 6:4>KTP PVA.Interestingly this trend is in accordance to the degree of amorphousness concluded from XRPD of the PM systems.Similar trend is seen in SD system,whereby the both the PVP and PVPVA carrier system revealed complete amorphousness follow by the partial crystalline behavior of PVA carrier SD system.Thus,it is hypothesized that degree of drug polymer interaction does interfere with the degree of amorphousness in solid dispersion during the manufacturing process.

3.5. Dissolution performances of the PM and solid dispersions system

3.5.1. The comparison of the dissolution profiles of PM systems

The ability of amorphous SD in producing formulations with enhanced dissolution rate and bioavailability was widely reported[1,5,12,36-39].This is due to the ability of amorphous system in exhibiting high level of supersaturation and thus higher apparent solubility than its crystalline counterpart[3]. In this study,the dissolution performances were assessed by comparing the dissolution profiles of pure KTP,PM and their corresponding SD systems using similarity factor as tabulated in Table 6.

3.5.2. The comparison between dissolution profiles of PMs and SDs systems

Fig.9 overlays the dissolution profiles of all the PM systems and ketoprofen alone.All the PM systems reveal higher initial rate of dissolution in comparison to the dissolution of ketoprofen alone.The increase the dissolution rate of the PM in comparison to the KTP may be due to the well-known wetting effect of the hydrophilic carrier.This effect is obvious when we compare the initial dissolution rate between PM systems and drug alone.However,the wettability effect is vague when these systems were formulated into SD system.The reverse trend was seen whereby early dissolution rate of PM KTP PVP K30 system is the slowest followed by

Table 6-Comparison of f2value among physical mixture products(PM KTP PVA,PM KTP PVPVA 6:4,PM PVP K30)and pure KTP.

Similar phenomena have been observed by Terifie and coworkers[40]where hydrophilicity of the interacted polymer is altered after it was incorporated into the solid dispersion. According to the authors,this alteration is ascribed to the strong hydrogen bond interaction between the carboxylic group of the poorly soluble drug and oxygen moiety of the hydrophilic polymer which made the polymer difficult to be solubilized[40]. Similarly,in the current study,the deterioration in wettability might be ascribed to the interaction between the carboxylic and oxygen moieties of KTP and PVP or PVPVA 6:4 after spray drying which lead to the lower initial dissolution rate as compared to their corresponding PM.

Fig.10 presents the comparison of dissolution profile of KTP, PM and SD for the three systems investigated in this study.According to Fig.10,dissolution rate of SD for KTP PVPVA 6:4 and KTP PVA system are higher than their corresponding PM system (Fig.10C and D).Interestingly,the most hydrophilic polymer of PVP K30 shows slowest dissolution rate among the SD (Fig.10D).This results is consistent to the conclusion drawn from a report by Mauludin and co-worker whereby dissolution of SD KTP PVA system dissolve faster than SD KTP PVP K30 system[41].However,that published dissolution result(by Mauludin and co-worker)revealed a higher release rate of SD KTP PVP K30 system than its corresponding PM and KTP alone which is not in agreement to the result of the current study [41].Based on Fig.10A,SD KTP PVP K30 not only shows a slower release profile among the tested SD,furthermore,its releaserate is even slower than its corresponding PM system and Pure KTP alone with the significant different deduced from the similarity factor calculated in Table 7.

Table 7-Comparison of f2Value among Spray driedproducts(SD KTP PVA,SD KTP PVPVA 6:4,SD PVP K30) and their corresponding physical mixtures.

3.5.3. Crystalline traces of KTP is not a hurdle in dissolution of KTP

Table 8 summarizes the dissolution performances of all the investigated samples in this study.Based on Table 8,it shows that dissolution rate of pure KTP was significantly improved by the physical addition of polymer PVA,PVPVA and PVPK30. On the other hand,it is interesting to note that dissolution rate of spray dried SD KTP-PVA,PVP K30 and PVPVA 6:4 do not consistently show the highest rate when compared with PM and pure drug.SD 30 KTP PVP K30 shows the lowest dissolution rate when compare with pure KTP and PM.The slight basic property of PVP polymer[42]might fasten dissolution of the acidic molecule of KTP.However this is not observed as SD KTP PVP system present the slowest dissolution rate among all the tested systems.Thus,pH of the dissolved polymer is unlikely to be the main factor for the different dissolution profiles observed for the case of a pH-dependent API like ketoprofen.

It is also intriguing that the partially crystalline dispersion,SD 30 KTP PVA,despite its crystallinity,reveals the greatest enhancement in dissolution profile of KTP over the fully amorphous SD systems(Fig.10D).This indicates that the crystallinity of the solid dispersion does not present as a hurdle for dissolution process of the KTP.Amorphous solids generally have higher solubility as compared to their crystalline counterpart.However they may undergo solution mediated phase transformation to their corresponding less soluble metastable or crystalline form in the dynamic of the dissolution medium[43].After conversion,the less soluble solid will then give rise to the slower dissolution rate.In this study,dissolution rate of the partially crystalline SD of 30 KTP PVA not only did not slow down the dissolution performance of KTP but further increase its dissolution rate despite the present of crystal traces.Thus the present of crystalline material as a result of the mediation of the amorphous solid to their crystalline counterpart is not a sole contributor for the limited enhancement of dissolution.

3.5.4. Ability of the carrier in sustaining high surface area of the solute in dissolution process

Based on the types of dissolution mechanism proposed by Craig Duncan[5],dissolution shown in the current study might be categorized as carrier dependent process as the dissolution rate of the same API is highly depends on its carrier,i.e.the KTP domain would need to first dissolve in the wetted polymer bed of its carrier before releasing mechanism.In this respect,KTP solubility in PVA shall be higher than in PVP K30 as PVA reveal a higher dissolution rate than KTP alone whereas PVP K30 revealed a slower dissolution rate than KTP alone.However,the presence of crystalline trace in SD KTP PVA and limited interaction of PVA and KTP might suggest the reverse.Generally, drug-polymer interaction plays an important role in enhancing dissolution rate of API in solid dispersion due to the better solubilisation effect of the API exerted by the polymer carrier [33,34].However,this study shows an exception,higher drugpolymer interaction which may impart higher solubility of KTP in the polymer,did not give rise to a faster dissolution rate but rather slows down the dissolution rate.

Dissolution process of drug is a kinetic process,even though an API may be highly soluble in a polymer carrier fluid;it may need a long duration to be completely solubilized in the polymer bed or medium.Thus,solely considering API’s solubility in the polymer might not be sufficient to predict the optimum performances of a poorly soluble API.Furthermore there is no cutting point presented in numerical term of which extend of the solubility of an API would lead to polymer dependent or drug dependent dissolution process.

Moreover,due to the dynamic movement of the dissolution medium during dissolution process,the hydrophobic domain,i.e.drug rich domain,may recrystallize or agglomerate during the release process.These negative effects may worsen the dissolution rate when the carrier dissolved away too rapidly and doesn’t play a role in sustaining the high effective surface area of solute at vicinity of the dissolving layer during dissolution process.The phenomena agglomeration is clearly seen in the current study as evident by the visible particles in the midst of the dissolution vessel while dissolution process(Fig.11A)despite the control of particle size(100-160µm)before the dissolution process.The growing of particles was noted fast in the extremely hydrophilic PVP that release the KTP rapidly due to the rapid dissolution of PVP polymer. At the same time,the protecting effect of the PVP polymer as amorphous stabilizer and steric hindrance of crystal growth is vanishing and hence the value of the original amorphous solid dispersion would have been lost(Fig.11B).Also,the growing of particle size reduces the effective surface area for dissolving during the dissolution process.This is the mainfactors for the unpredictable dissolution of solid dispersion, particularly the fast dissolving carrier,i.e.PVP homopolymer.

Table 8-Summary of dissolution performances among KTP with PVA,PVPVA 6:4,PVP K30.

In contrast,even though PVA is a partially crystalline polymer with limited interaction with KTP,its high viscosity in solution,i.e.40 mPa s at 4%might present as a barrier for the solution mediated transformation of the drug domain during dissolution(as compared to the low viscosity of PVP and PVPVA 6:4 in solution i.e.,less than 5-8 mPa s at same concentration).Nonetheless,the release mechanism of PVA which was detailed in a report by Mallapragada and co-wrokers,showed the possibility of this polymer to be released in a continuing lamellar unfolding manner might further assist the release of drug devoid the extensive agglomeration of the hydrophobic drug domain[44].Therefore,ability of the carrier in sustaining high surface area of the solute during the dissolution process is crucial.

4. Conclusion

ketoprofen readily forms amorphous solid dispersion by spray drying with PVP and PVPVA polymers.However,only partially amorphous system obtained when spray drying with PVA polymer.This may be attributed by the weaker drug-polymer interaction in PVA system that lead to the lower degree of amorphization.

The amorphous property,which is deemed to be an important factor for dissolution enhancement of solid dispersion, is shown to be insignificant for dissolution enhancement of solid dispersion preparation in the current study.Complete amorphization of the SD 30 KTP PVP K30,which may be formed due to the high degree of drug-polymer interaction,does not guarantee absolute dissolution enhancement of the highly active amorphous form of ketoprofen.In this study,the solution mediated solid transformation process which gives rise to the hydrophobic crystalline traces of KTP is just a bridge to the real hurdle of the dissolution of API.Competing factor between the bigger size of recrystallized drug and re-dissolution of these particles in its recrystallized form may be the main hurdle in slowing down dissolution of the API.

Hence,the true advantage of amorphous solid dispersion is the competing balanced between the amorphicity of the drug, its recrystallization tendency and more importantly the hydrophobic agglomeration.More investigation should be carried out to ensure the key properties that determine the beneficial side of solid dispersion.

AcAcknowledgements

The authors acknowledge the financial support received from the short term grant 304/PFARMASI/6313055,Universiti Sains Malaysia in carrying out this work.

REFERENCES

[1]Patel JR,Carlton RA,Yuniatine F,et al.Preparation and structural characterization of amorphous spray-dried dispersions of tenoxicam with enhanced dissolution.J Pharm Sci 2012;101:641-663.

[2]Song Y,Wang L,Yang P,et al.Physicochemical characterization of felodipine-kollidon VA64 amorphous solid dispersions prepared by hot-melt extrusion.J Pharm Sci 2013;102:1915-1923.

[3]Newman A,Knipp G,Zografi G..Assessing the performance of amorphous solid dispersions.J Pharm Sci 2012;101:1355-1377.

[4]Serajuddin ATM..Solid dispersion of poorly water-soluble drugs:early promises,subsequent problems,and recent breakthroughs.J Pharm Sci 1999;88:1058-1066.

[5]Craig DQM..The mechanisms of drug release from solid dispersions in water-soluble polymers.Int J Pharm 2002;231:131-144.

[6]Marsac P,Shamblin S,Taylor L..Theoretical and practical approaches for prediction of Drug-Polymer miscibility and solubility.Pharm Res 2006;23:2417-2426.

[7]Paudel A,Van Humbeeck J,Van den Mooter G..Theoretical and experimental investigation on the solid solubility and miscibility of naproxen in Poly(vinylpyrrolidone).Mol Pharm 2010;7:1133-1148.

[8]Ilevbare GA,Liu H,Edgar KJ,et al.Understanding polymer properties important for Crystal growth inhibition-impact of chemically diverse polymers on solution Crystal growth of Ritonavir.Cryst Growth Des 2012;12:3133-3143.

[9]Matsumoto T,Zografi G..Physical properties of solid molecular dispersions of indomethacin with Poly(vinylpyrrolidone)and Poly(vinylpyrrolidone-co-vinylacetate)in relation to indomethacin Crystallization.Pharm Res 1999;16:1722-1728.

[10]Taylor LS,Zografi G..Spectroscopic characterization of interactions between PVP and indomethacin in amorphous molecular dispersions.Pharm Res 1997;14:1691-1698.

[11]Verheyen S,Blaton N,Kinget R,et al.Mechanism of increased dissolution of diazepam and temazepam from polyethylene glycol 6000 solid dispersions.Int J Pharm 2002;249:45-58.

[12]Tajarobi F,Abrahmsén-Alami S,Larsson A..Dissolution rate enhancement of parabens in PEG solid dispersions and its influence on the release from hydrophilic matrix tablets.J Pharm Sci 2011;100:275-283.

[13]Moneghini M,Carcano A,Zingone G,et al.Studies in dissolution enhancement of atenolol.Part I.Int J Pharm 1998;175:177-183.

[14]van Drooge DJ,Hinrichs WLJ,Frijlink HW..Anomalous dissolution behaviour of tablets prepared from sugar glass-based solid dispersions.J Control Release 2004;97:441-452.

[15]Saers ESk,Craig DQM..An investigation into the mechanisms of dissolution of alkyl p-aminobenzoates from polyethylene glycol solid dispersions.Int J Pharm 1992;83:211-219.

[16]Ashland.Plasdone™K povidones and Plasdone™S-630: properties for Spray dried and melt-extruded solid dispersions.2012.

[17]Bühler V..Soluble polyvinylpyrrolidone(Povidone). Polyvinylpyrrolidone excipients for pharmaceuticals.2005. p.5-124.

[18]Kulkarni PK,Dixit M,Kumar YS,et al.Preparation and evaluation of ketoprofen beads by melt solidification technique.Der Pharm Sin 2010;1:31-43.

[19]Jachowicz R,rnberg EN,Pieszczek B,et al.Solid dispersion of ketoprofen in pellets.Int J Pharm 2000;206:13-21.

[20]Patil S,Sherikar A,Sujit Patil AP..Improvement of physicochemical characteristics and dissolution profile of poorly water soluble drug:ketoprofen by solid dispersion technique.Int J Res Pharm Sci 2010;1:450-453.

[21]Yang R,Wang Y,Zheng X,et al.Preparation and evaluation of ketoprofen hot-melt extruded enteric and sustainedrelease tablets.Drug Dev Ind Pharm 2008;34:83-89.

[22]Malaj L,Censi R,Mozzicafreddo M,et al.Influence of relative humidity on the interaction between different aryl propionic acid derivatives and poly(vinylpyrrolydone)K30:evaluation of the effect on drug bioavailability.Int J Pharm 2010;398:61-72.

[23]Bhavesh B,Patel JKP,Chakraborty Subhashis,Shukla Dali.. Review:revealing facts behind spray dried solid dispersion technology used for solubility enhancement.Saudi Pharm J 2014;14.

[24]FDA.Guidance for Industry.Research CfDEa editor.. Dissolution testing for immediate release solid oral dosage forms.Rockville:U.S.Department of Health and Human Services,Food and Drug Administration;1997. p.8-10.

[25]Bogdanova S,Pajeva I,Nikolova P,et al.Interactions of Poly(vinylpyrrolidone)with ibuprofen and naproxen: experimental and modeling studies.Pharm Res 2005;22:806-815.

[26]Jørgensen AC,Torstenson AS..Humid storage conditions increase the dissolution rate of diazepam from solid dispersions prepared by melt agglomeration.Pharm Dev Technol 2008;13:187-195.

[27]Kanaujia P,Lau G,Ng WK,et al.Investigating the effect of moisture protection on solid-state stability and dissolution of fenofibrate and ketoconazole solid dispersions using PXRD,HSDSC and Raman microscopy.Drug Dev Ind Pharm 2011;37:1026-1035.

[28]Callahan JC,Cleary GW,Elefant M,et al.Equilibrium moisture content of pharmaceutical excipients.Drug Dev Ind Pharm 1982;8:355-369.

[29]Abd Alaziz DMSO,Elshamy AE,Neseem DI..Formulation and evaluation of binary and ternary solid dispersions of domperidone by solvent evaporation method.Afr J Pharm Pharmacol 2014;8:5.

[30]Gordon M,Taylor JS..Ideal copolymers and the second-order transitions of synthetic rubbers.i.non-crystalline copolymers.J Appl Chem 1952;2:493-500.

[31]Pinal R..Entropy of mixing and the glass transition of amorphous mixtures.Entropy 2008;10:207-223.

[32]Verreck G,Six K,Van den Mooter G,et al.Characterization of solid dispersions of itraconazole and hydroxypropylmethylcellulose prepared by melt extrusionpart I.Int J Pharm 2003;251:165-174.

[33]Li Y,Pang H,Guo Z,et al.Interactions between drugs and polymers influencing hot melt extrusion.J Pharm Pharm 2014;66:148-166.

[34]Harsh C,Hg Chong,Eman ATEF..Correlating the behavior of polymers in solution as precipitation inhibitors to tis amorphous stablization ability in solid dispersion.J Pharm Sci 2013;102:1924-1935.

[35]Shehap AM..Thermal and spectroscopic studies of polyvinyl alcohol/sodium carboxy methyl cellulose blends. Egypt J Solids 2008;31:75-91.

[36]Chan SY,Qi S,Craig DQM..Enhanced dissolution of naproxen using hot melt extruded polyvinylpyrrolidone vinyl acetate formulations.APS Pharm Sci 2012,12-14th September 2012;East Midlands Conference Centre, University of Nottingham:APS PharmSci 2012;2012.

[37]Li B,Konecke S,Wegiel LA,et al.Both solubility and chemical stability of curcumin are enhanced by solid dispersion in cellulose derivative matrices.Carbohyd Polym 2013;98:1108-1116.

[38]Yan Y-D,Sung JH,Kim KK,et al.Novel valsartan-loaded solid dispersion with enhanced bioavailability and no crystalline changes.Int J Pharm 2012;422:202-210.

[39]Alleso M,Chieng N,Rehder S,et al.Enhanced dissolution rate and synchronized release of drugs in binary systems through formulation:amorphous naproxen-cimetidine mixtures prepared by mechanical activation.J Control Release 2009;136:45-53.

[40]Terife G,Wang P,Faridi N,et al.Hot melt mixing and foaming of soluplus®and indomethacin.Polym Eng Sci 2012;52:1629-1639.

[41]Mauludin R,Pamudji JS,Ruyanti D..Dissolution imporvement of ketoprofen tablets by solid dispersion method.Asian J Pharm Clin Res 2011;4:119-124.

[42]Li B,Liu H,Amin M,et al.Enhancement of naringenin solution concentration by solid dispersion in cellulose derivative matrices.Cellulose 2013;20: 2137-2149.

[43]Liu T,Wang B,Dong W,et al.Solution-mediated phase transformation of a hydrate to its anhydrous form of donepezil hydrochloride.Chem Eng Technol 2013;36:1327-1334.

[44]Raina SA..Phase behavior of supersaturated solutions of poorly soluble small molecules[Ph.D.].Ann Arbor:Purdue University;2014.

*Corresponding author.Universiti Sains Malaysia,11800 Pulau Pinang,Malaysia.Tel.:+60 4 6532233. E-mail address:sychan@usm.my(S.-Y.Chan). Peer review under responsibility of Shenyang Pharmaceutical University.

http://dx.doi.org/10.1016/j.ajps.2015.04.003

1818-0876/©2015 The Authors.Production and hosting by Elsevier B.V.on behalf of Shenyang Pharmaceutical University.This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).