Research Article

Microwave-Assisted Development of Orally Disintegrating Tablets by DirectCompression

Kishor V. Kande,1 Darsheen J. Kotak,1 Mariam S. Degani,1 Dmitry Kirsanov,2,3

Andrey Legin,2,3 and Padma V. Devarajan1,4

Received 12 October 2016; accepted 29 November 2016

ABSTRACT. Orally disintegrating tablets (ODTs) are challenged by the need for simpletechnology to ensure good mechanical strength coupled with rapid disintegration. Theobjective of this work was to evaluate microwave-assisted development of ODTs based onsimple direct compression tableting technology. Placebo ODTs comprising directly com-pressible mannitol and lactose as diluents, super disintegrants, and lubricants were preparedby direct compression followed by exposure to >97% relative humidity and then microwaveirradiation for 5 min at 490 W. Placebo ODTs with hardness (>5 kg/cm2) and disintegrationtime (<60 s) were optimized. Palatable ODTs of Lamotrigine (LMG), which exhibited rapiddissolution of LMG, were then developed. The stability of LMG to microwave irradiation(MWI) was confirmed. Solubilization was achieved by complexation with beta-cyclodextrin(β-CD). LMG ODTs with optimal hardness and disintegration time (DT) were optimized bya 23 factorial design using Design Expert software. Taste masking using sweeteners andflavors was confirmed using a potentiometric multisensor-based electronic tongue, coupledwith principal component analysis. Placebo ODTs with crospovidone as a superdisintegrantrevealed a significant increase in hardness from ∼3 to ∼5 kg/cm2 and a decrease indisintegration time (<60 s) following microwave irradiation. LMG ODTs had hardness >5 kg/cm2, DT < 30s, and rapid dissolution of LMG, and good stability was optimized by DOE andthe design space derived. While β-CD complexation enabled rapid dissolution and moderatetaste masking, palatability, which was achieved including flavors, was confirmed using anelectronic tongue. A simple step of humidification enabled MWI-facilitated development ofODTs by direct compression presenting a practical and scalable advancement in ODTtechnology.

KEY WORDS: Lamotrigine; microwave irradiation; orally disintegrating tablet; taste masking; β-cyclodextrin.

INTRODUCTION

Orally disintegrating tablets (ODTs) rapidly disintegratein the mouth to provide an in situ dispersion enabling ease ofadministration. ODTs have thereby created a revolution aspatient-friendly alternatives to the conventional tablets and

capsules, especially for geriatric patients and the dysphagic(1,2). Balancing two opposing requirements, namely, rapiddisintegration time (DT) and adequate hardness, coupledwith good palatability is the major challenge in ODTdevelopment. Lyophilization was among the first processesreported for ODT development, wherein freeze drying ofaqueous dispersions filled into blister alveoli cavities enabledthe formation of porous tablets (3–5). Vacuum dryingfollowed as an alternative to freeze drying (6). Nonetheless,while rapid disintegration was achieved, both processesresulted in porous fragile structures. An adapted cottoncandy process produced floss-like rapidly dissolving crystal-line structures, which enabled ODTs with rapid DT, but couldnot overcome the limitation of poor strength (7). Thisprocess, moreover, involved high temperatures, further limit-ing drug candidates that could be incorporated (8).

Wet molding technology, which involves moistening thepowder blend with a hydroalcoholic solvent followed by

1Department of Pharmaceutical Sciences and Technology, Institute ofChemical Technology, Deemed University, Elite Status and Centreof Excellence (Maharashtra), N.P. Marg, Matunga (E), Mumbai,400019, Maharashtra, India.

2 Institute of Chemistry, St. Petersburg State University,Universitetskaya nab. 7/9, Mendeleev Center, 199034, St. Peters-burg, Russia.

3 Laboratory of Artificial Sensory Systems, ITMO University,Kronverkskiy pr., 49, 197101, St. Petersburg, Russia.

4 To whom correspondence should be addressed. (e-mail:pvdevarajan@gmail.com)

AAPS PharmSciTech (# 2016)DOI: 10.1208/s12249-016-0683-z

1530-9932/16/0000-0001/0 # 2016 American Association of Pharmaceutical Scientists

forcing the wetted mass into mold plates, was also evaluatedfor the design of ODTs. While rapid DT was optimized, lowhardness resulted in high breakage during handling. Interest-ingly, Sano and Itai demonstrated that subjecting wet moldedtablets comprising mannitol and silicon dioxide to drying bymicrowave irradiation (MWI) enhanced their hardness,without compromising on the rapid disintegration (9). Theeffects of MWI on the increased porosity enabled by thewater vapor generated aided rapid disintegration. Enhancedhardness was ascribed to the dissolution of mannitol in thewater vapor, with subsequent solidification coupled with achange in the polymorphic form. In another study, Sano et al.demonstrated superior ODT properties with l-HPC as thedisintegrant (10). Wet molding dictates the need for waterduring processing, and being a slow process is not readilyadaptable for large-scale manufacture.

Direct compression, on the other hand, is a simple andcost-effective technique for tablet manufacture (11) adaptableon high-speed machines, with the added advantage of being adry process. ODTs manufactured by direct compressionmethods at lower hardness exhibited rapid DT, but highfriability and even tablet rupture during opening of the blisterposed serious issues (12–14). Increasing the hardness resultedin prolonged DT (15), thereby limiting application of thedirect compression process for ODT manufacture.

In this paper, we present a new approach for the designof ODTs, adapting direct compression as the first step ofODT manufacture followed by MWI. Water, a criticalrequirement to enable enhanced tablet porosity, was providedthrough a simple yet innovative step of humidification of thedirectly compressed tablets prior to microwave irradiation.Using this approach, placebo ODTs of high hardness (>5 kg/cm2) and rapid disintegration (<60 s) were successfullydeveloped. The method was then successfully adapted foroptimization of palatable ODTs of Lamotrigine (LMG),

which exhibited rapid dissolution by design of experiment(DOE) approach.

MATERIAL AND METHODS

Materials

Ac-Di-Sol (FMC Biopolymers), crospovidone(Kollidon® CL-SF, BASF), directly compressible mannitol(Perlitol-200), beta-cyclodextrin (β-CD) and sodium stearylfumarate (Roquette Pharma), lactose DC (Meggle Pharma),sodium starch glycolate (Primogel, DFE Pharma), and pre-gelatinized starch (UNI-PURE WG 220, National StarchFood Innovation) were received as gifts. Potassium sulfatewas purchased from S. D. Fine Chemicals, Ltd. Lamotriginewas a gift sample (Cipla Pvt. Ltd., India).

Preparation of ODTs

Direct Compression

Placebo ODTs were prepared by mixing diluents, binder,and superdisintegrants, followed by mixing with the lubri-cants, sodium stearyl fumarate and magnesium stearate, in apolyethylene bag. The blend was compressed on a rotary

Table I. Placebo ODT Formulation Batches

Ingredients (mg/tablet) DC1 DC2 DC3 DC4 DC5 DC6

DC mannitol 147 147 147 155 153 152DC lactose 40 40 40 40 40 40Pre-gelatinized starch – – – – – –Ac-Di-Sol 10 – – – – –SSG – 10 – – – –Crospovidone – – 10 2 4 5PVP K-25 2 2 2 2 2 2Sod. stearyl fumarate 1 1 1 1 1 1Mg stearate – – – – – –Total wt. (mg) 200 200 200 200 200 200

RH exposure timeHardness (kg/cm2) Initial 3 3.2 2 3 3 3

After MWI 3.2 3.2 3.5 3.3 3.5 3.730 min +MWI – – – 5 5.2 5.22 h +MWI 4.4 4.2 5.2 – – –3 h +MWI 4.5 4.2 5.3 – – –

DT (s) Initial 175 177 57 154 130 122After MWI 160 162 26 148 123 11430 min +MWI – – – 90 67 482 h +MWI 135 130 14 – – –3 h +MWI 130 117 14 – – –

Table II. Independent Variables and Levels for DOE

Parameters Low level (−1) High level (+1)

A: Crospovidone (%) 2.5 7.5B: Microwave irradiation time (min) 2 5C: Humidity exposure time (min) 20 40

Kande et al.

tablet press using 8-mm flat punches to obtain tablets ofapproximately 200 mg.

Humidification of Tablets

Humidification of tablets was carried out by exposure oftablets for various time intervals in a humidity chambermaintained at 97% relative humidity (RH) using a saturatedsolution of potassium sulfate (16)

Microwave Irradiation

The placebo tablets were subjected to microwaveirradiation at 490 W in a microwave oven (Catalyst™ System,Cata 2R).

Effect of the following variables on tablet hardness andDT was evaluated:

(a) Microwave exposure time (5, 7, and 10 min)(b) Effect of initial tablet hardness (2 and 3 kg/cm2)(c) Effect of superdisintegrant type and concentration

(Ac-Di -Sol , sodium starch glyco la te , andcrospovidone)

(d) Effect of humidification time prior to microwaveirradiation (30 min, 2 h, and 3 h)

The various tablet batches evaluated are described inTable I.

Development of LMG ODTs

Effect of Microwave Irradiation on LMG Stability

LMG as a powder was exposed to microwave irradia-tion for 5 min at 490 W. A sample of about 10 mg of the

MWI sample was accurately weighed and transferred to a100 mL volumetric flask. Methanol (5 mL) and volume weremade up to 100 mL with double distilled water filteredthrough a 0.22-μm membrane filter (stock I, 100 μg/mL).One milliliter of stock I solution was further diluted to10 mL with mobile phase filtered through a 0.22-μmmembrane filter to obtain a solution of 10 μg/mL. Analysiswas performed by HPLC at room temperature (25°C) usinga Jasco Instrument (PU-980, Japan) equipped with a WatersSpherisorb® 250 × 4.6-mm column and a Jasco photodiodearray detector at 210 nm. The mobile phase comprised ofphosphate buffer pH 3/acetonitrile/methanol/THF(64:15:20:1) at a flow rate of 1 mL/min. The sample(100 μL) was injected into the system and the concentrationof LMG was extrapolated from a standard plot in theconcentration range 2–10 μg/mL prepared in a mannersimilar to the sample preparation.

Phase Solubility Study

A phase solubility study was carried out using themethod reported by Higuchi and Connors (17). Increasingconcentrations of β-CD of 1, 2, 4, 6, 8, and 10 mM wereprepared in distilled water and 3 mL filled in glass bottles.Excess LMG (50 mg) was added to these solutions and thebottles stoppered and agitated in a constant temperatureshaker water bath at 37 ± 2°C for 72 h. LMG without β-CDserved as the reference. Following equilibrium, the superna-tant was withdrawn and centrifuged at 10,000 rpm for 15 minand assayed for LMG content by UV spectrophotometry(UV1650PC, Shimadzu Corporation, USA) at λmax of276 nm. Experiments were performed in triplicate. A phasesolubility graph of drug concentration vs. β-CD concentrationwas plotted and the apparent stability constant (K1:1) was

Table III. Batches for Taste Masking

Ingredients (mg/tablet) DC7 DC8 DC9 DC10 DC11 DC12 DC13

Pineapple flavor – 2 2 2 – – –Vanilla flavor – – – – 2 2 2Sucralose – – 2 4 – 2 4

Fig. 1. HPLC chromatograms of LMG. a Standard LMG. b LMG after MWI indicating stability

MWI-Enabled Development of DC ODT

calculated from the initial straight line portion of the phasesolubility diagram using the following equation:

K1:1 ¼ SlopeIntercept 1−Slopeð Þ

LMG-β-CD Inclusion Complex

The LMG-β-CD inclusion complex was prepared by thewet kneading method. A mixture of LMG and β-CD in a 1:1molar ratio was kneaded in a mortar with ethanol–water (1:1)to obtain a paste-like consistency (18). The paste was dried inan oven at 50°C, pulverized, passed through a 60-mesh sieve,and stored in a desiccator until further use.

Characterization of LMG-β-CD Inclusion Complex

FTIR Spectrophotometry. Samples of LMG and theLMG-β-CD inclusion complex were prepared in the form of

KBr pellets and scanned from 4000 to 400 cm−1 on a FTIRspectrophotometer (Perkin-Elmer, Model Spectrum RX).

Differential Scanning Calorimetry. Differential scanningcalorimetry (DSC) thermograms were obtained on a differ-ential scanning calorimeter (Perkin-Elmer, Shelton, USA).Samples (5 mg) of LMG and LMG-β-CD inclusion com-plexes were sealed in an aluminum pan and heated from 30 to300°C at a heating rate of 10°C/min using an empty pan as areference under a purge of nitrogen (18 mL/min).

Powder XRD. The powder X-ray diffraction (XRD)spectra of LMG and LMG-β-CD complexes were recordedusing an X-ray diffractometer (Rigaku Miniflex, Japan) with acopper tube anode at a scanning rate of 5°/min and thediffraction angle (2θ) in the range 0–80°.

Design of Experiment Approach: LMG ODTs

A two-level full factorial design (23) with a center pointwas adopted using Design Expert® 7 software to analyze theeffect of critical material attributes (A: concentration ofcrospovidone) and critical process parameters (B:microwave irradiation time and C: humidity exposure time)on the desired critical quality attributes (Y1:disintegrationtime and Y2: hardness). The variables and levels areindicated in Table II.

LMG ODTs

For the preparation of LMG ODTs, the LMG-β-CDinclusion complex equivalent to 25 mg LMG was mixed withthe diluents, disintegrants, and lubricants and the tabletscompressed as described in BDirect Compression^

Fig. 2. Phase solubility study of LMG in β-CD solution

Fig. 3. FTIR spectra of LMG (a), β-CD (b), and the LMG-β-CD complex (c)

Kande et al.

In Vitro Evaluation of LMG ODTs

ODTs were evaluated for standard tablet properties includ-ing hardness, DT, weight variation, friability, and dimensions.Drug content was determined by UV spectrophotometry(UV1650PC, Shimadzu Corporation) at λmax of 276 nm.

Wetting Time

Wetting time was measured by placing the ODTs on atissue paper in a Petri dish (i.d. = 6.5 cm) containing 10 mL ofwater and monitoring the time for complete wetting. Threereplicates were performed.

In Vitro Dissolution Study

In vitro dissolution (n= 6) was performed in 900 mL of 0.1 NHCl maintained at 37 ± 0.5°C using USP type II dissolutionapparatus with sinkers (Electrolab, India) at a paddle speed of50 rpm. At predetermined time intervals, 10 mL sample waswithdrawn and replaced with fresh medium (37°C). The drug was

quantified by UV spectrophotometry (UV1650PC, ShimadzuCorporation) at λmax of 276 nm.

Scanning Electron Microscopy Analysis

ODTs were mounted on metal stubs using double-sidedadhesive tapes and scanned on a scanning electron micro-scope (JSM-6510, Jeol, Japan). ODTs were evaluated beforeand after microwave irradiation.

Taste-Masked LMG ODT Using an Electronic Tongue

ODT batches with flavors (pineapple and vanilla) andsweeteners are reported in Table III. Taste masking was optimizedusing an electronic tongue. Placebo ODTs (labeled as DC7P–DC13P, respectively) and the plain drug, LMG, served as reference.

Multisensor System

The potentiometric multisensor system employed in thisstudy contained 18 potentiometric membrane sensors and

Fig. 4. XRD patterns for Lamotrigine (a) and the LMG-β-CD complex (b)

Fig. 5. DSC thermograms of LMG and the LMG-β-CD complex

MWI-Enabled Development of DC ODT

standard pH glass electrode (ZIP, Gomel, Belorussia). Tensensors were PVC-plasticized anion-sensitive electrodes andsix sensors were PVC-plasticized anion-sensitive electrodesbased on various ion exchangers and similar to thoseemployed earlier (19). Two sensors were based on

chalcogenide glass membranes with pronounced RedOxsensitivity (a sensor sensitive to the reduction/oxidationprocess on the electrode surface). All sensors were producedin the Laboratory of Chemical Sensors of Saint PetersburgState University. Sensor potentials were measured against a

Table IV. Experimental Design Model

Run order Parameters Response 1: DT(s)

Response 2: hardness(kg/cm2)

A: Crospovidone(%)

B: Microwave drying(min)

C: Humidity exposure(min)

1 −1 −1 −1 67.333 ± 1.52 3.066 ± 0.112 −1 +1 +1 50.666 ± 1.15 5.533 ± 0.113 +1 +1 −1 59.666 ± 0.57 6.533 ± 0.054 +1 −1 −1 15 ± 0.1 2.566 ± 0.115 +1 +1 +1 20.666 ± 1.15 5.933 ± 0.116 +1 −1 +1 24.333 ± 0.2 2.6 ± 0.17 −1 −1 +1 80.333 ± 0.57 2.533 ± 0.058 −1 +1 −1 110.33 ± 1.52 6.1 ± 0.459 0 0 0 20.666 ± 1.52 5.533 ± 0.05

* n = 3; 0 indicates center point

Fig. 6. Response surface plots of hardness as a function of the concentration of crospovidone (a), microwave irradiationtime (b), and humidity exposure time (c)

Kande et al.

Ag/AgCl reference electrode (Izmeritelnaya Tehnika, LLCMoscow, Russia) with 0.1 mV precision using a high-impedance multichannel digital mV-meter HAN-32 (SensorSystems, LLC, St. Petersburg, Russia). The mV-meter wasconnected to a PC for data acquisition and processing.

Data Processing

Principal component analysis (PCA) is the method ofdata dimensionality reduction and visualization of the hiddendata structure. Nowadays, it is widely employed in differentstudies, and detailed descriptions of the mathematical calcu-lations behind the PCA are available (20). Briefly, the PCAalgorithm looks for orthogonal directions of maximal vari-ance in the initial multidimensional space and projects thepoints on this new coordinate axis (principal components).The main outcomes of PCA are so-called score and loadingsplots, visualizing similarity of the studied samples and theinformation contained in the employed variables (sensorresponses in our case).

Stability Evaluation

LMG ODTs were packed in sealed HDPE bottles andsubjected to 40 ± 2°C/75% RH± 5% and 30 ± 2°C/65 ± 5% asper the ICH guidelines for 3 months. Samples were with-drawn at 1, 2, and 3 months and evaluated for appearance,hardness, DT, and drug content.

Fig. 7. Response surface plots of disintegration time as a function of the concentration of crospovidone (a), microwaveirradiation time (b), and humidity exposure time (c)

Table V. Regression Analysis

Term Tablet hardness (kg/cm2) Disintegration time (s)

Coefficient p value Coefficient p value

A 0.050 0.1976 −21.79 <0.0001B 1.68 <0.0001 4.96 0.0159C −0.20 0.0330 −11.37 <0.0001AB – – 1.62 0.3925BC – – 0.29 <0.0001AC – – −13.29 0.8767ABC – – 4.88 0.0175Constant 4.46 32.46R2 = 0.9353 R2 = 0.9345

MWI-Enabled Development of DC ODT

Statistical Analysis

All values are expressed as the mean ± SD of at least threeindependent experiments. Statistical analysis was performedusing one-wayANOVAwithDennett’s test and Student’s t tests.P < 0.05 was the criterion for statistical significance.

RESULTS

Preparation of ODTs

Placebo ODTs

The various placebo ODT batches are reported inTable I. It is evident from the table that exposure to humidityprior to MWI reflected a significant change in both hardnessand DT. Although humidity exposure time did not influencehardness, DT was significantly affected and inversely relatedto the exposure time. The superdisintegrant, however, had asignificant role and influenced both hardness and DT. Whilean increase in hardness was seen with all threesuperdisintegrants, this increase was substantial withcrospovidone as the disintegrant. Interestingly, crospovidonealso reflected a very low DT of 14 s (DC3), while the DTsseen with Ac-Di-Sol (DC1) and SSG (DC2) were significantlygreater than 60 s. Nevertheless, tablets with 10%crospovidone (DC3) revealed a rough surface. A decrease

in crospovidone concentration resulted in an acceptableappearance with a significant increase in hardness and a DTof less than 1 min. Moreover, this was achieved with just30 min of exposure to high humidity (D6).

ODT of LMG

The HPLC chromatogram following microwave irradia-tion revealed no additional peaks, confirming the stability ofLMG to MWI, as shown in Fig. 1.

Phase Solubility Study of LMG-β-CD

The phase solubility plot revealed a linear increase in theaqueous solubility of LMG as a function of β-CD concentra-tion up to 10 mM (Fig. 2). The linear host–guest correlationcoefficient was r2 > 0.97 and the slope < 1, with a stabilityconstant K of 524.72/M.

Characterization of the LMG-β-CD Inclusion Complex

FTIR Interaction of β-CD with LMG in Solid State. TheFTIR spectra of LMG and the LMG-β-CD complex are shownin Fig. 3. The infrared spectrum of LMG (Fig. 3a) ischaracterized by vibration peaks at 3450 cm−1 (N–H aromatic),3317 and 3213 cm−1 (C–H aromatic), 1630 cm−1 (C=N), and1556 cm−1 (C=C). The FTIR spectra of β-CD (Fig. 3b) showspeaks at 3377 cm−1 (O–H), 2926 cm−1 (C–H), 1157 cm−1 (C–H),and 1080 cm−1 (C–O). The FTIR spectra of the LMG-β-CDinclusion complex (Fig. 3c) shows broadening and reduction ofpeak intensities of aromatic N–H, C–H, and C=N groups,indicating interaction between LMG and β-CD.

Powder X-Ray Diffraction Analysis. The X-ray diffrac-tion profile of LMG revealed a crystalline nature (Fig. 4a).The XRD pattern of the LMG-β-CD (Fig. 4b) complexexhibited peaks with a decrease in the intensity of peaks.

DSC Study. The DSC thermogram of LMG revealed asharp endothermic peak at 221°C which corresponds to the

Fig. 8. Design space overlay plot

Table VI. Optimization and Statistical Validation

Predictedvalue

Observedvalue

%Deviation

T a b l e thardness(kg/cm2)

5.05 5.23 0.127

Disintegrationtime (s)

23.44 22 1.018

Parameters: A = 5.5%, B = 3.5 min, and C = 30 min

Kande et al.

melting point of LMG (Fig. 5). This endotherm was notexhibited by the LMG-β-CD complex.

DOE Approach

The design of experiment approach was applied to arriveat the optimum LMG ODT formulation with rapid disinte-gration and good hardness as the response parameters.Crospovidone concentration, microwave irradiation time,and humidity exposure time were selected as the variables.The hardness and DT of the LMG ODTs are depicted inTable IV. The main and interaction effects of the concentra-tion of crospovidone, microwave irradiation time, and hu-midity exposure time on the selected responses wereevaluated. Regression analysis was carried out and a p valueless than 0.05 was considered statistically significant. Stan-dardized regression coefficients represent the positive ornegative effects of each parameter and their interactions oneach of the tablet properties. The coefficient of determination(r2), which was doubly adjusted with degrees of freedom, wasemployed as an indicator of the model fit. The contour plots

revealed the effect of the variables on the hardness (Fig. 6)and DT (Fig. 7) of the LMG ODTs. Values of 0.9353 and0.9345 for the correlation coefficient R2 suggested goodcorrelation between the observed and model-predicted valuesof hardness and DT, respectively (Table V). The overlay plot(Fig. 8) depicts the design space representing optimal valuesof the three variables to arrive at LMG ODTs with desirableproperties of DT < 30 s and hardness > 5 kg/cm2.

We considered the following optimized conditions::A(concentration of crospovidone) = 5.5%; B (microwave irradia-tion time) = 3.5 min; and C (humidity exposure time) = 30 min,as shown in Table VI. Our results confirm that both responseswere consistent with the predicted values, validating theappropriateness of the experimental design. Optimized LMGODTs revealed that crospovidone exerted no effect on hard-ness. A linear increase in hardness with an increase inmicrowave irradiation time and a decrease with an increase inhumidity exposure time were observed. Concentration ofcrospovidone and humidity exposure time revealed a negativeeffect on DT, whereas microwave irradiation time exhibited apositive effect. A negative interaction effect was seen betweencrospovidone (A) and humidity exposure time (C).

Furthermore, friability of <1% was an indication of thegood mechanical resistance of the tablet. The thickness was in

Fig. 9. Tablet wetting time images at various stages

Fig. 10. Scanning electron microscope images of the internal surface of untreated (a) andmicrowave-treated (b) tablets (arrows indicate porosity)

MWI-Enabled Development of DC ODT

the range of 2.24 ± 0.07 mm, and the drug content was99.13%, which was within acceptable limits. Tablets exhibitedcomplete wetting in 16 s (Fig. 9). SEM images of the tabletsurface before and after microwave irradiation are depictedin Fig. 10. A porous surface was clearly evident in themicrowave-irradiated tablets (Fig. 10b).

In Vitro Drug Release

The dissolution profiles of LMG and the optimized LMGODTs are shown in Fig. 11. While the optimized LMG ODTsreleased nearly 100% of the drug within 5 min, LMGexhibited a release of barely 50% at 30 min.

Taste Masking

Figure 12 shows the PCA score plot obtained for allanalyzed samples. The score plot provides a clear separation ofLMG and LMG ODTs from the placebo ODTs along the PC1axis. All placebo samples are located on the left part of the plotand have negative score values onPC1; all sampleswithLMGareon the right side of the plot. The observed separation can beattributed to the sensitivity of the sensor array towards thestudied drug. Separate PCA analysis of all active pharmaceuticalingredient (API) samples reveals that certain separation offormulations with different taste masking is also possible using apotentiometric multisensor system. The PCA score plot shown inFig. 13 suggests that formulations DC7, DC9, DC10, and DC12

are more similar to LMG in terms of sensor responses indicatingbitterness compared to formulations DC8, DC11, andDC13. Thedata also suggested that good taste masking was achieved simplyby the addition of vanilla flavor without the need for sucralose(DC 11), confirming the role of β-CD in taste masking.

Stability Study

Optimized ODT formulation was evaluated for stability.No significant difference in appearance, disintegration time,hardness, weight variation, and drug content was observed,confirming the stability of LMG ODTs (Table VII).

DISCUSSION

Development of ODTs presents a delicate balance ofhigh mechanical strength with low disintegration time. Directcompression as a process for the manufacture of ODTs hasmanifold advantages, including scalability, high capacity, andbeing a solvent-free process. The excipients that imparthardness usually increase the disintegration time. Hence,MWI was evaluated as a strategy to increase hardness with acorresponding decrease in DT. Mannitol is an excipient ofchoice for ODTs due to its negative heat of solution andpleasant taste. It is available as α-, β-, and δ-crystallinepolymorphs (21). Furthermore, the effect of microwaveirradiation on mannitol results in the conversion of the δ-form to the stable β-form, with a corresponding increase inthe hardness of the tablets, and no compromise on DT isdemonstrated (10). We therefore selected mannitol as thediluent in our study. As tablets with mannitol alone exhibitedhigh friability, we arrived at a diluent combination ofmannitol and lactose for the development of ODTs.

The effects of MWI on tablet hardness and porosity arerelated entirely to the water vapor generated. Microwaveirradiation causes vibrations in water molecules at highvelocities, resulting in partial conversion to water vapor.While the dissolution of mannitol in this water vapor andsubsequent solidification enabled the formation of solidbridges to increase the hardness, the water vapor-inducedexpansion of the tablet mass facilitated enhanced porosityand, hence, low DT. It is therefore evident that water is acrucial requirement in ODT development using MWI. Toadapt the advantage of MWI to direct compression, a dry

Fig. 11. Dissolution profiles of LMG ODT and for LMG drug

Fig. 12. PCA score plot for analyzed samples (API-containingsamples are marked with filled points, while placebo samples aremarked with empty points)

Fig. 13. PCA score plot for API samples

Kande et al.

process, we proposed one simple additional step, that ofexposure of ODTs to controlled humidity (>95%RH) forpredetermined time periods prior to microwave irradiation.The increased hardness and the decrease in DT observedconfirmed the effects of MWI on the ODTs.

The significant decrease in DT seen with crospovidoneis attributed to the high capillarity and the resulting rapidwater absorption (22). On the other hand, the swellingdisintegrants Ac-Di-Sol and SSG exhibited core formation,which hampered disintegration (23–25). A decrease incrospovidone concentration favored good appearance withdesired DT. The successful exploitation of MWI in thedevelopment of placebo ODTs triggered us to design LMGODTs.

LMG is a low-molecular-weight (256 g/mol) BCS class IIdrug which exhibits poor solubility and also has a bitter taste.β-CD is a good solubilizer, stabilizer, and a known tastemasking agent specifically for low-molecular-weight drugswhich can readily be entrapped in the β-CD cavity. A stableinclusion complex of LMG/β-CD at a 1:1 ratio, whichcorrelated with the Higuchi and Connors linear modelrelationship, was obtained, as confirmed by the high K value(17). Interaction of LMG with β-CD, as seen in the FTIRspectra, is indicative of solubility enhancement, while inter-actions specifically involving N groups also suggest tastemasking (26). The DSC thermogram and XRD spectra, whichindicated a significant decrease in crystallinity with possiblepartial amorphization, also proposed an enhanced dissolutionrate (27).

LMG ODTs were successfully optimized by DOE toobtain LMG ODTs with hardness of 5.23 kg/cm2 and DT of22 s. The positive interaction effect seen between theconcentration of crospovidone, microwave irradiation time,and humidity exposure time on DT proposed that all threevariables were critical. The enhanced dissolution ratesconfirmed the role of β-CD as a solubilizer for LMG, withimprovement in LMG wettability and formation of readilysoluble complexes in the dissolution medium enablingenhanced dissolution (28).

The added difficulty with LMG was its bitter taste.Various strategies like ion exchange resin (29,30), complexa-tion with cyclodextrin (31,32), microencapsulation (33,34),film coating (35), flavors and sweeteners (36,37), andrheological modification (38) have been used to mask thebitter taste of drug. The potentiometric multisensor systemcoupled with principal component analysis confirmed that theLMG-β-CD inclusion complex in combination with vanillaflavor enabled successful taste masking of LMG ODTs.

CONCLUSION

We present an innovative yet simple and green approachfor the preparation of ODTs by direct compression coupledwith MWI. This technology is versatile, scalable, and adapt-able to a range of drugs. More importantly, this approach wassuccessfully extrapolated for the design of LMG ODTs whichalso exhibited good palatability with rapid dissolution of thedrug.

ACKNOWLEDGEMENTS

The authors are thankful to the University GrantsCommission, Government of India, Department of Science& Technology (DST), Government of India and RussianFoundation for Basic Research (grant INT/RUS/RFBR/P-195and RFBR no. 15-53-45105), and DST Prime MinistersFellowship for financial support. Dmitry Kirsanov andAndrey Legin acknowledge partial financial support fromGovernment of Russian Federation (grant 074-U01).

REFERENCES

1. Goel H et al. Orally disintegrating systems: innovations informulation and technology. Recent Patents Drug DelivFormul. 2008;2(3):258–74.

2. Sreenivas S et al. Orodispersible tablets: new-fangled drugdelivery system—a review. Indian J Pharmaceut Educ.2005;39(4):177.

3. Seager H. Drug‐delivery products and the Zydis fast‐dissolvingdosage form*. J Pharm Pharmacol. 1998;50(4):375–82.

4. Hori H et al. Olanzapine orally disintegrating tablets (ZyprexaZydisR) rapidly improve excitement components in the acutephase of first-episode schizophrenic patients: an open-labelprospective study. World J Biol Psychiatr. 2009;10(4-3):741–5.

5. Lafon L. Galenic form for oral administration and its method ofpreparation by lyophilization of an oil-in-water emulsion. 1986,Google Patents.

6. Gohel M et al. Formulation design and optimization of mouthdissolve tablets of nimesulide using vacuum drying technique.AAPS PharmSciTech. 2004;5(3):10–5.

7. Misra TK et al. Fast-dissolving comestible units formed underhigh-speed/high-pressure conditions. 2000, Google Patents.

8. Myers GL, Battist GE, Fuisz RC. Process and apparatus formaking rapidly dissolving dosage units and product therefrom.1998, Google Patents.

9. Sano S et al. Impact of active ingredients on the swellingproperties of orally disintegrating tablets prepared by micro-wave treatment. Int J Pharm. 2014;468(1):234–42.

Table VII. Stability Study (Mean ± SD; n = 3)

Stability conditions Sampling interval Disintegration time (s) Hardness (kg/cm2) Weight variation (mg) Drug content (%)

Initial 20.66 ± 1.73 5.1 ± 0.12 202.62 ± 1.22 99.37 ± 0.2330°C/65% RH 1 month 21.33 ± 1.52 5.3 ± 0.05 203.37 ± 2.43 99.45 ± 0.24

2 months 19.66 ± 1.52 5.2 ± 0.16 200.35 ± 0.53 99.63 ± 0.463 month 20.63 ± 1.54 5.3 ± 0.2 199.55 ± 0.53 99.63 ± 0.46

40°C/75% RH 1 month 23.01 ± 1.16 5.3 ± 0.15 201.71 ± 2.22 99.37 ± 0.192 months 21.66 ± 0.57 5.2 ± 0.2 199.17 ± 1.12 99.22 ± 0.433 months 23.56 ± 0.37 5.3 ± 0.13 200.22 ± 1.23 98.72 ± 0.43

MWI-Enabled Development of DC ODT

10. Sano S et al. Preparation and evaluation of swelling induced-orally disintegrating tablets by microwave irradiation. Int JPharm. 2011;416(1):252–9.

11. Sano S et al. Design and evaluation of microwave-treated orallydisintegrating tablets containing polymeric disintegrant andmannitol. Int J Pharm. 2013;448(1):132–41.

12. Bi Y et al. Preparation and evaluation of a compressed tabletrapidly disintegrating in the oral cavity. Chem Pharm Bull.1996;44(11):2121–7.

13. Bi Y, Yonezawa Y, Sunada H. Rapidly disintegrating tabletsprepared by the wet compression method: mechanism andoptimization. J Pharm Sci. 1999;88(10):1004–10.

14. Gupta A. Recent trends of fast dissolving tablet-an overview offormulation technology. Int J Pharmaceut Biol Arch.2010;1(1):1–10.

15. Iveson SM et al. Nucleation, growth and breakage phenomenain agitated wet granulation processes: a review. Powder Technol.2001;117(1):3–39.

16. Rockland LB. Saturated salt solutions for static control ofrelative humidity between 5° and 40°C. Anal Chem.1960;32(10):1375–6.

17. Higuchi T, Connors A. Phase-solubility techniques. Adv ChemInstrum. 1965;4:212–217.

18. Patel H et al. Preparation and characterization of etoricoxib-β-cyclodextrin complexes prepared by the kneading method. ActaPharma. 2007;57(3):351–9.

19. Rudnitskaya A et al. Assessment of bitter taste of pharmaceu-ticals with multisensor system employing 3 way PLS regression.Anal Chim Acta. 2013;770:45–52.

2 0 . E s b e n s e n KH . P r i n c i p a l c ompon e n t a n a l y s i s(PCA)—introduction. In: Esbensen KH, editor. Multivariatedata analysis in practice—an introduction to multivariate dataanalysis and experimental design. 5. Oslo: Camo Software AS;2001:19–74.

21. Walter-Levy L. Cristallochimie-sur les variétés cristallines du D-mannitol. CR Acad Sc Paris Ser C. 1968;267:1779–82.

22. Augsburger LL et al. Superdisintegrants: characterization andfunction. Encyclop Pharmaceut Technol. 2007;20:269–90.

23. Zhao N, Augsburger LL. The influence of swelling capacity ofsuperdisintegrants in different pH media on the dissolution ofhydrochlorothiazide from directly compressed tablets. AAPSPharmscitech. 2005;6(1):E120–6.

24. Gohel MC et al. Preparation and assessment of novelcoprocessed superdisintegrant consisting of crospovidone and

sodium starch glycolate: a technical note. AAPS PharmSciTech.2007;8(1):E63–9.

25. Gryczke A et al. Development and evaluation of orallydisintegrating tablets (ODTs) containing ibuprofen granulesprepared by hot melt extrusion. Colloids Surf B: Biointerfaces.2011;86(2):275–84.

26. Singh J, Garg R, Gupta GD. Enhancement of solubility ofLamotrigine by solid dispersion and development of orallydisintegrating tablets using 32 full factorial design. J Pharm(Cairo). 2015;5:828453.

27. Sharma M, Garg R, Gupta G. Formulation and evaluation ofsolid dispersion of atorvastatin calcium. J Pharmaceut Sci Innov.2013;2(4):73–81.

28. Shinde VR et al. Enhanced solubility and dissolution rate ofLamotrigine by inclusion complexation and solid dispersiontechnique. J Pharm Pharmacol. 2008;60(9):1121–9.

29. Yewale CP et al. Formulation and development of taste maskedfast-disintegrating tablets (FDTs) of chlorpheniramine maleateusing ion-exchange resins. PharmDev Technol. 2013;18(2):367–76.

30. Bhise K, Shaikh S, Bora D. Taste mask, design and evaluationof an oral formulation using ion exchange resin as drug carrier.AAPS PharmSciTech. 2008;9(2):557–62.

31. Stojanov M, Wimmer R, Larsen KL. Study of the inclusioncomplexes formed between cetirizine and α‐, β‐, and γ‐cyclodextrin and evaluation on their taste‐masking properties.J Pharm Sci. 2011;100(8):3177–85.

32. Goudanavar P, Shah SH, Hiremath D. Development andcharacterization of lamotrigine orodispersible tablets: inclusioncomplex with hydroxypropyl B cyclodextrin. Int J Pharm PharmSci. 2011;3(3):208–14.

33. Bora D, Borude P, Bhise K. Taste masking by spray-dryingtechnique. AAPS PharmSciTech. 2008;9(4):1159–64.

34. Hu X et al. Preparation and evaluation of orally disintegratingtablets containing taste-masked microcapsules of berberinehydrochloride. AAPS PharmSciTech. 2013;14(1):29–37.

35. Chen J-C, Bunick FJ, McNally G. Fast dissolving/disintegratingcoating compositions. 2013, Google Patents.

36. Somoza V et al. Method for the identification of bitter tastingcompounds and bitter taste modulating compounds. 2015. USPatent 20,150,362,481.

37. Santi PAD, Nelson DG. Taste masking of phenolics using citrusflavors. 2001, Google Patents.

38. Skrabanja ATP, Tully RE. Oral liquid antidepressant solution.2000, Google Patents.

Kande et al.