Harish Dureja

The goal of this project was to use complexation to develop a dual-drug resinate system, to mask the bitter taste of levocetirizine dihydrochloride and montelukast sodium. The maximum drug loading onto the resin, polacrilin potassium, USP (Tulsion 339) was found to be 1:3 drug-to-resin ratio. Box-Behnken design methods were used to study the effect of processing parameters such as swelling time (X

), on cumulative percentage drug release. 

The cumulative percent drug release was found to be minimal at extreme pH values (X

).The maximum drug release was found at high values of stirring time, even at low values of swelling time. One-way analysis of variance (ANOVA) was applied to the cumulative percentage drug release to study the fitting and the significance of the model. 

Differential scanning calorimetry (DSC) and Fourier Transform Infrared (FTIR) analyses of the optimized formulation (F3) confirmed the drug complex formation, suggesting that this model can be used to optimize dual drug release. Finally, the dual-drug resinate was formulated into orally disintegrating tablets (ODT), whose quality was found to be within pharmacopoeial limits. Drug release rate studies of the tablets in simulated gastric fluid and in phosphate buffer (pH 6.8) showed better results than the marketed formulation. Results suggest that dual-drug resinate can be a cost effective and efficient method for masking unpleasant taste in dual drug formulations. 

	Organoleptic (i.e., sense-related) characteristics can have a significant impact on how closely patients adhere to any therapeutic regimen (1). Taste is one of the key factors that determine patient compliance and the commercial success of any oral pharmaceutical formulation. Thus, it is important for any oral dosage form to have a pleasing taste. When the formulation is bitter or tastes bad, taste masking technology is used to make it more palatable to patients.  This involves the development of a system that provides a physical barrier between the active substance and the taste buds, modifies the drug solubility, or alters human taste perception in some way (2).
	
	Unpleasant taste may be masked by various methods (3), including the use of flavors, sweeteners and amino acids, polymer coatings, effervescent systems, granulation, freeze-drying (4), adsorption on ion-exchange resin, solid dispersion, or chemical modification, for example, using insoluble prodrugs, multiple emulsions, or salt formation. Other techniques involve the use of liposomes, microencapsulation, complexation, and rheological modification. 
	
	Ion-exchange resins offer an inexpensive and effective way to mask unpleasant taste in pharmaceuticals (5). These resins are cross-linked polymers available as high molecular weight polyelectrolytes having extensive charged functional  sites. They are water insoluble and exchange their exchangeable ions with charge ions in the surrounding ionic medium (6, 7). The ion-exchange resin binds a drug to form a drug-resin complex, known as resinate. 

	Recently, an alternative form, a “dual-drug resinate” was introduced, in which two drugs are loaded onto the same resin, providing drug-release characteristics similar to those of single-drug resinates (8). The authors applied the dual-drug resinate approach, using ion-exchange resin to develop a drug-resin complex (DRC) to improve the palatability of a combination of levocetirizine dihydrochloride and montelukast sodium. This combination is often used to treat respiratory distress resulting from allergies. 

	The goal was to mask the taste and formulate orally disintegrating  tablets (ODTs), which are much easier for patients to take than conventional tablets, particularly when patients are children or elderly, or when they are bedridden, nauseous, or psychotic. ODTs use superdisintegrants, which allow tablets to disintegrate instantaneously after coming into contact with the tongue (9). 
	
	Box-Behnken design methods were used to optimize the dual-drug resinate complex to mask the taste of levocetirizine dihydrochloride and montelukast sodium, and to formulate the optimized complex into ODTs. Levocetirizine dihydrochloride is an active H1-receptor antagonist. It is the R enantiomer of cetirizine hydrochloride, a racemic compound with antihistaminic properties. Montelukast sodium is a selective and orally active leukotriene receptor antagonist that inhibits the cysteinyl leukotriene CysLT1 receptor (10). 

 Levocetirizine dihydrochloride and montelukast sodium samples were provided by Arion Health Care, Ltd. Samples of 

 (USP)-grade polacrilin potassium resin (Tulsion 339) were provided by Thermax, Ltd.Other chemicals used in this project were analytical grade; high-performance liquid chromatography (HPLC)-grade solvents were used for HPLC analysis, and HPLC-grade water was used throughout this study.
	
	

. Approximately 20 g of Tulsion 339 resin was consecutively washed three times with 100 mL of deionized water, 100 mL of 95% ethanol, 100 mL of 50% ethanol, and finally with 100 mL of deionized water. Supernatant was removed consecutively by sedimentation and decantation. The washed resin was dried overnight at 50 ºC in a hot air oven and kept in a closed vial.
	
	

 Different quantities of activated resin were transferred to 20 mL of deionized water and allowed to swell for 30 minutes. Levocetirizine dihydrochloride and montelukast sodium were added separately to each beaker at different  ratios of drug:resin (1:1, 1:2, 1:3, 1:4, 1:5 and 1:6), and stirred using a magnetic stirrer for three hours at room temperature. The mixtures were filtered, and residues were washed with 5 mL of deionized water. The unbound drug in the filterate was estimated by HPLC, using Agilent 1200 series on Reliasil ODS 250 × 4.6 mm, 5-µm column operated at 250ºC using methanol:water (78:22) as the mobile phase; flow rate was maintained at 1.0 mL/min, and detection carried out at 240 nm.
	
	

The DRC was prepared by batch process, keeping the quantity of drug constant, and using Box-Behnken design methods, as will be described subsequently. The resin was allowed to swell in 20 mL water under magnetic stirring for 15-60 min at room temperature. Levocetirizine dihydrochloride and montelukast sodium were added to the swollen slurry at the maximum drug-to-resin ratio, under magnetic stirring, and the resultant mixtures were stirred for one to six hours. The pH of the solution was adjusted to 1.2, 4.0, and 6.8. The DRC was separated by filtration, and residue was washed with 5 mL of deionized water to remove any uncomplexed drug, and dried at room temperature. The complex was then stored in an air-tight glass vial.
	
	

 Statistically designed experiments using Box-Behnken methods (Design-Expert 8, Version 8.0.7.1 software) were performed to study the effect of three factors-swelling time (X

	Box-Behnken design is an independent quadratic design in which the treatment combinations are at the midpoints of edges of the process space and at the center. It is a rotatable (or near rotatable) design, and requires three levels (low, medium and high) of each factor. The  study of three factors at three levels (-1, 0, +1) using Box-Behnken design leads to 15 complexes of drug and resin (F1-F15). These formulations (F1-F15) are listed in 

					Table I: Multiple linear regression for percent drug-resin complex (DRC) of levocetirizine dihydrochloride and montelukast sodium.

		Differential scanning colorimetry (DSC). The DRC was sealed in an aluminium pan. The reference was an empty sealed aluminium pan. Nitrogen flow rate was maintained at 60 mL/min, heating was done at a ramp rate of 10ºC/ min with equilibration at 30ºC. The sample was heated until the tempereature reached 350ºC. 

		FTIR Studies. FTIR analysis was performed using IR Affinity-1 FTIR, to identify the drugs. A total of 2% (w/w) of DRC was mixed  with dry potassium bromide (KBr). This powder mixture was then dried and further compressed into KBr discs under a hydraulic press at 10,000 psi. Each KBr disc was scanned over a wave number region of 400-4000 cm

. The characteristic peaks were recorded.

 dissolution rate study of  DRC in 0.1 N HCl and phosphate buffer (pH 6.8). Drug-release studies were performed in a USP XXII Type II tablet dissolution apparatus (LabIndia DS-8000), using 0.1 N HCl (pH 1.2) and phosphate buffer (pH 6.8), with 0.5% sodium lauryl sulphate as dissolution media. A DRC equivalent to 5 mg of levocetirizine dihydrochloride and 10 mg of montelukast sodium was taken in 900 mL of each dissolution media. The temperature and speed of rotation were maintained at 37 ± 0.5ºC and 50 rpm, respectively. Samples (10 mL) were withdrawn at suitable time intervals, filtered through a nylon filter, and analyzed, directly or after appropriate dilution, for drug concentration using HPLC at 240 nm. The cumulative percent drug release was calculated.

		Sensory evaluation for  bitterness of DRC. A bitterness evaluation  test was performed to compare the bitterness of the DRC to that of each of the drugs individually. Six healthy human volunteers evaluated the effectiveness of taste masking.  Each volunteer held a DRC equivalent to a dose of 5 mg of levocetirizine dihydrochloride and 10 mg of montelukast sodium in his or her mouth for 10 seconds, then spat out the mixture and rinsed his or her mouth with distilled water. Their perceived bitterness levels were recorded instantly and then after 10-30 seconds, using the following scoring values: 0 = tasteless, 1 = slight, 2 = moderate, 3 = strong, 3+ = very strong (11).

 Using direct compression, mouth-dissolving tablets of the DRC were made. The equivalent amount of both drugs was taken. The tablets were prepared using crospovidone and sodium starch glycolate as superdisintegrants (see 

). All the ingredients were accurately weighed and passed through mesh #60. The powder blend was evaluated for micromeritic properties such as angle of repose, bulk density, tapped density, powder flow properties and porosity. A mixed blend of DRC and excipients was then compressed to form tablets, each weighing 150 mg.

					Table II: Ingredients for orally disintegrating tablets of levocetirizine dihydrochloride and montelukast sodium.

 The prepared tablets were evaluated for various official and non-official specifications:
	
	

 Twenty tablets were randomly selected, and the average weight was calculated. Then, the individual tablets were weighed and the individual weight was compared with the average weight.
	
	

 Tablets were evaluated for hardness using the Monsanto hardness tester. Each tablet was placed in contact with the tester’s lower plunger, and a zero reading was taken. The upper plunger was then forced against a spring by turning a  threaded bolt until the tablet fractured. The force of fracture was recorded and the zero force reading was subtracted from it.
	
	

 Twenty tablets were weighed and placed in a Roche friabilator, and the apparatus was rotated at 25 rpm for 4 min. The tablets were then dusted and weighed.
	The friability is given by the formula:
	F= (1-W/W

 = weight of the tablets before test
	W= weight of the tablets after test
	Water absorption ratio. A piece of tissue paper folded twice was placed in a small petri dish containing 6 mL of water. A tablet was put on the tissue paper and allowed to wet completely. The wetted tablet was then weighed. Water absorption ratio, R was determined using following equation:
	R= 100 x (W

 = weight of tablet before water absorption
	W

 = weight of tablet after water absorption
	
	

A piece of tissue paper (10.75 × 12 mm), folded twice, was placed in a culture dish containing 6 mL of water. A tablet was put on the paper and the time for complete wetting was measured (12). 
	
	

Ten tablets were weighed and the average weight was calculated. The tablets were then powdered, and a quantity of powder equivalent to 5 mg of levocetirizine dihydrochloride and 10 mg of montelukast sodium was dissolved in 100 mL of 0.1 N hydrochloric acid. The amount of levocetirizine dihydrochloride and montelukast sodium was then determined by HPLC.
	
	In vitro 

Six tablets were placed in a disintegration apparatus containing distilled water maintained at 37°C ± 2°C. The time required for the tablet to disintegrate completely, without any palpable mass remaining in the apparatus, was recorded.
	
	In vitro 

 Drug release studies were performed using a USP XXII Type II tablet dissolution apparatus (LabIndia DS-8000). The tablet was taken in 900 mL of 0.1 N hydrochloric acid (pH 1.2) with 0.5% sodium lauryl sulfate (SLS). The temperature and speed of rotation were maintained at 37 ± 0.5ºC and 50 rpm, respectively. 10-mL samples were withdrawn at 15-min intervals, filtered in a nylon filter, and analyzed by HPLC at 240 nm directly or after appropriate dilution. The cumulative percent drug release was calculated.
	
	

A bitterness evaluation test was performed to compare the bitterness of the tablet to that of each of the pure drugs, (i.e., levocetirizine dihydrochloride and montelukast sodium). 
	
	

 The drug resin ratio of 1:3 was found to show the maximum binding of drug with the resin. Therefore, the ratio 1:3 was selected for further study of various process parameters. The results are tabulated in 

					Table III: Optimization of drug-resin ratio.

 Fifteen DRC formulations (F1- F15) of DRC were prepared and tested. Percent drug release was found to be minimum at extreme pH values (i.e., pH of 1.2 and pH 6.8) and at high level of swelling time (X

). The maximum drug release was found at pH 4.0, at high levels of stirring time, even at low levels of swelling time. These results suggest that high levels of stirring are required for maximum binding of the drug with the resin. The release of the drug also depends upon the pH of the medium. The multiple linear regression (MLR) for % DRC of levocetirizine dihydrochloride and montelukast sodium is given in 

. The model, developed from MLR to estimate effect (Y

 = % cumulative drug release of levocetirizine dihydrochloride
	Y

 = % cumulative drug release of montelukast sodium
	X

 shows the quadratic relationship term. 
	
	Batch F3 was selected as optimum batch for further formulation of ODTs.  ANOVA was applied on cumulative percentages of drug released, in order to study the fitting and significance of model. The F-test of equality for two variances was carried out to compare the regression mean square with the residual mean square (see 

) for levocetirizine dihydrochloride and montelukast sodium. The ratio F = 8.08 in the case of levocetirizine dihydrochloride and F = 4.52 in the case of montelukast sodium showed regression to be significant.

					Table IV: Analysis of variance of the regression (percent drug-resin complex).

	The estimated model, therefore, may be used as response surface for the % CDR, as shown by contour plots in Figures 1 and 2 and 3-D surface in Figures 3 and 4. 

					Figure 1: Contour plot for levocetirizine dihydrochloride.
					All figures courtesy of the authors.

					Figure 2: 3-D response surface for levocetirizine dihydrochloride.

Characterization of optimized drug-resin complex. 

 Results from DSC analyses of the pure drugs, levocetirizine dihydrochloride and montelukast sodium are shown in Figures 5 and 6. DSC analyses of the dual-drug resinate showed significant decrease in melting point of the pure drugs. DSC confirmed the interaction of formulation constituents with levocetirizine dihydrochloride and montelukast sodium and, therefore,decrease in the melting point. 

					Figure 3: Contour plot for montelukast sodium.

					Figure 4: 3-D response surface for montelukast sodium.

FTIR spectra of pure levocetirizine dihydrochloride and montelukast sodium showed characteristics peaks. A minor shifting of functional peaks was observed in the physical mixture of these drugs (see 

). The appearance of new peaks and major shifts of the functional peaks in the FTIR spectrum of batch F3 confirmed the formation of complex. 

					Table V: Fourier Transform Infrared (FTIR) data (peak at cm -1) of Levocetrizine hydrochloride, montelukast sodium, physical mixture, and optimized formulation (F3).

 dissolution rate study of DRC in 0.1 N HCl and in phosphate buffer pH 6.8.

 release studies of the optimized formulation were carried out in 0.1 N HCl (pH 1.2) and in phosphate buffer (pH 6.8), both  with 0.5 % SLS. Levocetirizine dihydrochloride showed 99.19% and 95.87%, while montelukast sodium showed 65.84% and 84.49% release in 0.1 N HCl and phosphate buffer (pH 6.8), respectively. The result is shown graphically in Figure 7.

					Figure 5: Differential Scanning Calorimetry Thermogram of levocetirizine dihydrochloride.

					Figure 6: DSC Thermogram of montelukast sodium.

Sensory evaluation for bitterness of DRC. 

The sensory evaluation  of the formulation showed that the resinate’s taste was more acceptable to the test volunteers than that of the pure drugs. 
	
	

 The micromeritic properties of the powder blend were evaluated and the results are tabulated in 

					Table VI: Evaluation of micromeritic properties of the powder blend.

	
	
	After  the micromeritic studies, ODT were evaluated for various parameters (e.g., weight variation, hardness, friability, water absorption ratio, wetting time, drug content, disintegration time and in-vitro drug release). The results are tabulated in 

					Figure 7: Cumulative percent drug release of F3 in simulated gastric fluid (left) and phosphate buffer pH 6.8 (right).

	The ODTs were found to pass the weight variation test as per I.P. (150 ± 7.5). The hardness was found to be 4.1 kg/cm3, which is sufficient to withstand mechanical shocks of handling in manufacture, packaging, and transportation. The friability was 0.76% and disintegration time was found to be 47 sec, which is acceptable. The water absorption ratio was found to be 75.90. The wetting time was found to be 35 seconds, and drug content was 98% for levocetirizine and 95% for montelukast.

	In-vitro dissolution rate study of ODTs. 

 drug release studies were performed in two different dissolution media, simulated gastric fluid and phosphate buffer of pH 6.8, and results were compared with those for the commercial formulation (tablet) in the same media. In simulated gastric fluid, ODT showed 99.64% and 54.85% release for levocetirizine dihydrochloride and montelukast sodium, respectively, whereas the release values for the marketed formulation were 93.86% for levocetirizine dihydrochloride and 48.17% for montelukast sodium. In phosphate buffer of pH 6.8, the ODT showed 90.28% release for levocetirizine dihydrochloride and 75.57% release for montelukast sodium. Comparable values in the commercial formulation were 75.12% for levocetirizine dihydrochloride and 75.55% for montelukast sodium. The results are shown graphically in Figures 8 and 9.

					Figure 8: Comparative percent release of levocetirizine dihydrochloride in simulated gastric fluid and in phosphate buffer pH 6.8.

					Figure 9: Comparative percent release of montelukast sodium in simulated gastric fluid and in phosphate buffer pH 6.8.


	Using complexation with ion-exchange resin, and the Box-Behnken model for optimizing the complexes, the taste of levocetirizine dihydrochloride and montelukast sodium was effectively masked. Taste masking was optimized based on such parameters as stirring time, pH effect and swelling time of resin. The maximum drug release was found at high values of stirring time, even at low values of swelling time. DSC and FTIR analyses confirmed the drug complex formation. The taste of the resulting complex was found to be palatable. 
	
	Results suggest that the mathematical model developed in the present study can be applied to formulate a taste-masked drug resin complex for different drugs. Retrospectively, a dual-drug resinate of desired release characteristics can also be developed. ODTs formulated using crosspovidone and sodium starch glycolate showed faster disintegration and enhanced drug release. The ODTs showed better release when compared with the commercial formulation. These formulations can be considered for scaleup.
	
	

 380 (1) 36-45 (2004).
	2. T.H.T. Hoang, et al., 

434 (1-2) 235-42 (2012).
	3. V. Sayeed et al,  “Solid Dosages and Excipients 2015” eBook supplement to 

 p. 12. (2015). 
	4. H. Sohi, Y. Sultana, and R.K. Khar, 

30 (5) 429-48 (2004).
	5. A. Helmy et al., 

7 (12) 831-44 (2011).
	6. V. Anand et al., 

6 (17) 905-14 (2001).
	7. I. Singh et al., 

32 (2) 91-100 (2007).
	8. H. X. Zeng et al., 

 37 (2) 201-7 (2001).
	9. R. Sheshala et al., 

. Accessed on 14/10/12.
	11. D. Shukla et al., 

57 (4) 337-45 (2009).
	12. H. Sunada, et al., 

 When referring to this article, please cite it as J. Mundlia, R.K. Marwaha, and H. Dureja et al., “Using a Dual-Drug Resinate Complex for Taste Masking,” 

Articles

Box-Behnken modeling was used to optimize a resinate complex, to mask the taste of levocetirizine dihydrochloride and montelukast sodium in orally disintegrating tablets.