doi: Original scientific paper

This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version
of the manuscript. The manuscript will undergo, copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note
that during the production process errors may be discovered which could affect
the content, and all legal disclaimers that apply to the journal pertain.

1
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Accepted Manuscript
Title: Non-compendial vs compendial analytical tests - a
powerful tool for predicting in vitro similarity of highly
viscous oral suspension

Authors: Elena Kazandjievska1*, Iva Antova1, Slavica

Mitrevska1, Aleksandar Dimkovski1, Elena Dimov1, Maja
Hadzieva Gigovska 1, Packa Antovska1, Sonja Ugarkovic1,
Jasmina Tonic Ribarska2, Suzana Trajkovic-Jolevska2

1
Research & Development, ALKALOID AD Skopje, Blvd.
Aleksandar Makedonski 12, 1000 Skopje, R. North
Macedonia
2
Faculty of Pharmacy, Ss. Cyril and Methodius University,
Mother Theresa 47, 1000 Skopje, R. North Macedonia

DOI:
Received date: December 2018
Accepted date: February 2019
UDC: 615.451.22.032.311.015.4

Type of paper: Original scientific paper

Please cite this article

2
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Non-compendial vs compendial analytical tests - a powerful tool for predicting in vitro

similarity of highly viscous oral suspension

Elena Kazandjievska1*, Iva Antova1, Slavica Mitrevska1, Aleksandar Dimkovski1,

Elena Dimov1, Maja Hadzieva Gigovska 1, Packa Antovska1, Sonja Ugarkovic1,
Jasmina Tonic Ribarska2, Suzana Trajkovic-Jolevska2

1
Research & Development, ALKALOID AD Skopje, Blvd. Aleksandar Makedonski 12,
1000 Skopje, R. North Macedonia
2
Faculty of Pharmacy, Ss. Cyril and Methodius University, Mother Theresa 47,
1000 Skopje, R. North Macedonia

Abstract
In vitro dissolution profiles are increasingly used to evaluate drug release
characteristics of pharmaceutical products. The dissolution methods is expected to be an
appropriate tool for checking consistency of the pharmaceutical attributes by discriminating
similarities and dissimilarities between different drug formulations. Expansion in
development of novel “special” dosage forms, due to the manner in which these dosage
forms release the active pharmaceutical ingredient, usually requires applying non-
compendial dissolution strategy that differs from the traditional compendial
recommendations.
For demonstrating sameness in the dissolution profile, in vitro drug release
comparison between test and reference product of highly viscous oral suspension by
applying non-compendial peak vessel against conventional hemispheric vessel was
demonstrated in this study.
All reference batches exhibited high variability in dissolution data when using
hemispheric vessel due to forming mound compact mass at the bottom of the vessel.
Different strategies for samples manipulation, before and during dissolution period, were
performed in order to eliminate additional variabilities. Modifications of conventional USP

3
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

2 apparatus such as using peak vessel provided with more reproducible and reliable result
for distinguishing in vitro similarities between different formulations of oral suspensions.
Misinterpretation of dissolution data can lead to negative impact on product
development. Taking time to observe and evaluate what is happening to the product in the
vessel during dissolution is of curtail consideration for proper selection of the dissolution
strategy.

Keywords: oral suspensions; in-vitro release; hydrodynamic variability; USP apparatus 2/ Paddle
apparatus; peak vessel

Introduction

Pharmaceutical oral suspensions are liquid preparations consisting of solid particles,

usually medicinal agent, dispersed through liquid phase in which the particles are not soluble
(Brown et al., 2011). The external phase is an aqueous, organic, or oily lipid phase in which
the insoluble internal phase is uniformly dispersed (Brown et al., 2011).
Depending of the route of administration, oral administration remains the preferred
route with the highest degree of patient compliance and is the most user-friendly form of
drug delivery with majority of 84% of the most-sold pharmaceutical products in the European
market and US (Fotaki and Vertzoni, 2010). After oral administration of the product, the
extent of desired pharmacological effect depends of the amount of absorption of the active
pharmaceutical ingredient (API) in gastrointestinal tract (GIT). Determination of the
dissolution rate of drug product which is prerequisite for bioequivalence since the drug must
dissolve before it can be absorbed is of special analytical challenge.
Dissolution test plays essential part in the product development and could be adopted
as the surrogate basis for the decision as to whether two pharmaceutical products are same in
terms of bioequivalence. In vitro drug release is used to assure product sameness by profile
comparison between two products or pre-change and post change products using appropriate
in vitro test. Also, a specific value of in vitro dissolution is recognized in the application of
batch-to-batch quality control tests reflecting the levels of change in composition, site of

4
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

manufacturing, scale of manufacturing and process and equipment changes providing better
product understanding (Siewert et al, 2003).
As modern pharmaceutical development gains in recent years, complex formulation
became more and more prevalent in the pharmaceutical industry. Due to its “special” design
formulation which in turns lead to different physicochemical and release behavior, different
apparatus, procedures and technics for in vitro determination are employed on case by case
basis. Recent paper reviews classified high viscosity suspension as “special dosage form”.
As a result of high quantity of viscosity and sedimentation building agents in the matrix,
several steps should be considered while performing drug release study of highly viscose
suspension for oral use (Brown et al., 2011). In order to prevent variability in the dissolution
data: homogeneity of samples by applying proper sample preparation (mixing acceleration,
frequency, time, course of shaking); employing procedure for sample introduction;
treatment of samples (filtration and filter compatibility and stability of samples in
appropriate media), selecting of proper dissolution conditions (apparatus, agitation rates).
According to the Food and Drug Administration (FDA) Dissolution Database, USP apparatus
2 is the most common apparatus for determination of the rate of drug release from oral dosage forms.
It is recommended for approximately 70% of the dissolution methods and it’s considered as
apparatus of first choice for predicting in vitro similarity for oral suspensions (Shohin et al., 2016).
Lower agitation rates of 25-50 rpm are usually recommended for this dosage form (Siewert et al.,
2003). In some cases, when high viscosity persists, in order to prevent sedimentation and
accumulation at the bottom of the vessel, higher rates may be applied. However, not in all cases
apparatus 2 is the best apparatus to use for testing suspension (Parker and Gray, 2006). Recent
findings refer to its high variability when producing dissolution data which simultaneously leads to
misinterpretation and possible wrong outcomes (Quershi and Shabnam, 2001). Investigations have
indicated on existing of “dead zone” at the bottom of the vessel underneath the dissolution paddle
due to very low mixing hydrodynamics (Liu and Vivilecchia, 2005). Additionally, this leads to “cone
formation”, effect which mainly confined to dosage forms that are formulated with high amount of
insoluble excipients that form compact mass in which the active ingredient is trapped and leads to
non-reproducible results. For eliminating these variabilities, usually operating with higher rotation
of the paddle, displacing of the formation can be achieved but, in that case, the discriminatory power
of the method will be compromised.

5
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Literature surveys reveal several approaches for bridging the USP 2 variability by
changing the geometry of the vessel thus changing the hydrodynamics in the environment
referring to solid dosages forms only (Collins and Nair, 1998, Legace et al., 2004; Liu and

Vivilecchia, 2005; Quershi, 2006). Far from our knowledge, until now there are no published

studies for using non-compendial peak vessel in evaluating dissolution drug release from oral
suspension as dosage forms.
Therefore, the aim of this study is to challenge the dissolution behavior of different
formulation of highly viscose suspension by modifying the general recommended dissolution
conditions and use non-compendial approach for performing the dissolution test. The
analysis focus on the impact of sample preparation and sample introduction technique as well
as the effect of the vessel geometry (peak vessel) in providing more sensitive thus more
accurate and reproducible results.

Materials and methods

Samples used for in vitro study
Laboratory test batches (test product) for orally administered suspension with API
BCS (Biopharmaceutical Classification System) class II were manufactured in the Research
and Development department in Alkaloid AD, Skopje, North Macedonia. All together twenty
formulations were made and coded randomly from A1 to A20, but only two were selected
(Table 1), as they resulted in forming satisfactory suspensions and showing relevant results.
Commercially available batches (reference products) of the same dosage form and
dosage strength were purchased from two different markets, German and Ireland. The
samples were checked for their expiry dates before purchasing. The suspensions were
randomly coded as listed in Table 1.

Table 1

Aliquot of 5 mL suspension containing 200 mg of API, which reflects typical dose of

the product, were transferred in the dissolution media. For applying the suspension in the
dissolution vessel, 5 mL dose syringe, 10 mL plastic syringe and 10 mL plastic syringe
6
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

attached with cannula (Bent Cannula W/LUER 4.75 in form Agilent) were used. Sample
introduction was performed by applying the samples above and in the dissolution media, on
non-rotating paddle and during rotation of the paddles. Due to the high viscosity of the
samples, transfer of the aliquots was performed by weighing. The syringes were weighed
before and after adding the product and the weight difference was related to product density.
All suspension containers prior withdrawing of the appropriate dose of suspension
and transferring in the dissolution vessel were vigorously mixed for 5 minutes. Quantitation
of assay content, for evaluating homogeneity of the withdrawn aliquot, was performed on
high performance liquid chromatography (HPLC).

API solubility
The solubility of API was determined in four different solvents, such as: phosphate
buffer (pH 6.8 and pH 7.2), acetate buffer (pH 4.5) and hydrochloric acid buffer (pH 1.2).
An amount of API equivalent to the highest individual dose that can be administered was
added in 250 mL of each medium. After stirring for 1 hours, drug solubility in each medium
was determined.

Dissolution conditions
Dissolution of the test and reference products was carried out on Agilent 708- DS
standard compendial configuration. 1000 mL round bottom (hemispheric) vessels and non-
compendial 1000 mL peak vessels (protruded bottom) from Agilent were used.
Dissolution media 900 mL pH 1.2 (hydrochloric acid), pH 4.5 (acetate buffer), pH
6.8 (phosphate buffer) and pH 7.2 (phosphate buffer) were prepared as described in the
European Pharmacopoeia (Ph Eur., 5.17.1). All dissolution media were degassed prior
introducing in the apparatus and used at a temperature of 37±0.5 ᵒC.
A minimum of 6 vessels were sampled for each analysis. 5 mL aliquots of suspension
containing 200 mg of API were introduced in each vessel.
Paddle speed conditions at 50, 60, 65, 75, 85, 100 rpm were evaluated for choosing
the best agitation rate. Samples were taken at predetermined intervals of 5, 10, 15, 20, 30, 45
and 60 minutes. Aliquots of 10 mL were withdrawn at each time point through bent cannula
7
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

with stopper at which end 35-micron ultra-high molecular weight polyethylene (UHMW PE)
full flow filters were placed to ensure that no large undissolved particles are withdrawn. Due
to the high quantity of viscosity building agents in the suspension matrix which could result
with clogging the system while quantitation, samples were again filtered through 0.20 μm
regenerated cellulose (RC) membrane syringe filters.
All filters were purchased from Agilent Technologies (USA).
Sample solution in concentration of 0.2222 mg/mL in pH 4.5 and pH 1.2 were
evaluated in time interval of 7 h after the end of the dissolution period due to instability,
while samples prepared in pH 7.2 and 6.8 were more stable and evaluation was performed in
time interval of 48 h.

Reagents
Analytical grade methanol (CH3OH), sodium hydroxide (NaOH), hydrochloric acid
(HCl), 85% o-phosphoric acid (H3PO4), sodium acetate (CH3COONa), sodium chloride
(NaCl), glacial acid (CH3COOH) and potassium dihydrogen phosphate (KH2PO4),
acetonitrile (CH3OH) were purchased from Merck (Darmstadt, Germany).
Water was purified by a Werner water purification system, obtained in-house at
Alkaloid AD Skopje, Skopje, R. North Macedonia.

Quantitation methodology
Quantitation of collected samples was performed on Agilent Technologies 1290
Infinity Liquid Chromatographic System (Agilent Technologies, USA) equipped with a
Quaternary Pump VL, a column compartment, auto sampler and photo-diode array detector.
Instrument control, data acquisition and processing were done by using OpenLab
Chemstation chromatography software (version A.02.02/1.3.4). The separation was
performed on Zorbax XDB C18 (Agilent Technologies, USA), 150 x 3.0 mm, 5 μm using
solution of o-H3PO4 and CH3OH as a mobile phase in ratio 30:70 (v/v). The column
temperature was 35 ºC. Flow rate was 1.5 mL/min. Injection volume was 5 μL. UV detection
was performed at 221 nm.

8
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Results and discussion

Choosing the right media and dissolution conditions, for providing relevant results,
depends on the required release characteristics of the intended product, solubility and stability
of the analyte in the test medium (Brown et al. 2004). Pharmacokinetic properties of the API
indicate small intestines as site of absorption of the drug where pH of the environment
exhibits pH 6.5-7.5. From the results presented in Table 2 it can be concluded that API is
highly soluble in pH 7.2 and pH 6.8 thus showing sink conditions when 5 mL individual dose
of the oral suspension (equivalent to 200 mg of API) is applied in 900 mL of these buffer
media. This is not the case in pH 4.5 and pH 1.2 were solubility is respectively decreasing,
which is to be expected since the API is week acid with pKa value around 4, those showing
no satisfactory sink conditions. The stability of sample solution prepared in concentration of
0.22 mg/mL, indicated buffer pH 7.2 as most appropriate media with satisfactory data for
more than 2 days stability. Taking into consideration all above elaborated, media pH 7.2 was
chosen as most appropriate medium for initial screening of the dissolution behavior of the
suspension for establishing in vitro release strategy.

Table 2

Sample introduction
Vigorously mixing the suspension container for 5 min before withdrawing samples,
assured accurate and reproducible quantity of API between 98% and 101% assay with every
5 mL withdrawn aliquots of suspension.
Starting dissolution conditions were set as per recommendations for oral suspension
with agitation of 50 rpm (Siewert et al., 2003), in hemispheric vessel, with media replacement
during sampling intervals and by applying the suspension samples on non-rotating paddles.
Initially the dissolution study was performed by evaluating the dissolution behavior of the
reference product. Applying the suspension above the media provided with unsatisfactory
data which were visually apparent even with higher agitation of 75 rpm (Table 3). This could
be result of the media surface tension which enables the suspension sample to fall down in

9
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

the vessel and let it partially float on the top of the media (Fig. 1a). When pulling down the
shaft of the dissolution apparatus, these floating particles would adhere on the paddles and
therefore poor mixing of the suspension occurred (Fig. 1b, Table 3). By transferring the
suspension in the media at the side or at the bottom of the vessel, no floating or sticking
remains were evident. Nevertheless, lower dissolution rate for 60 minutes (below 85%) were
observed when using 5 mL dose syringe for sample introduction, at stirring speed of 75 rpm.
Improvements of the released rate in period of 45 minutes (above 90%) were observed when
using 10 mL plastic syringe as transferring device.
However, variability in the results persists to occur as evident by the high values of
relative standard deviation (RSD). In addition to this, when visually observed, the reference
product tends to form compact mound mass at the bottom of the vessel, with no consistent
behavior during the dissolution period (Fig. 2).

Fig. 1

In one set of six individual portion of reference suspension, some would remain
adhered to the vessel bottom while other tend to show displacement more rapidly which
correlates directly with the dissolution results (lower and higher released % respectively).
As suggested for some viscous suspension (Brown et al., 2011) although not typical
for this dosage forms, higher agitation speeds of 100 rpm were applied in order to prevent
accumulation of the suspension. The use of higher agitation speed had no positive impact in
decreasing the variability of the dissolution data (Table 3).

Fig. 2

Further investigation required modification of the compendial conditions, by

applying the suspension in the apparatus 2 while paddles were rotating, in attempt to prevent
sticking of the suspension at the bottom of the vessel. It was assumed that this occurrence
happens due to non-existing hydrodynamic motion in the media prior starting the dissolution
paddles, which allows the high viscosity and suspending agents to settle and adhere to the
10
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

bottom. Application of the suspension on rotating paddles required prolonged duration of

dissolution test as a compensation of the time needed to administer six portions of suspension
individually, in order to retain correct sampling intervals. Additionally, to prevent
disturbance of the hydrodynamic, the withdrawn volumes during sampling period were not
replaced with fresh dissolution media as the sink conditions in pH 7.2 remain preserved
(Table 4).

Table 3
Table 4

No significant difference was observed with 75 rpm agitation speed as for the
compact mass persists to occur in the vessels. Significant difference with acceptable
variability was detected only on very high mixing speed of 100 rpm when applying the
suspension with 10ml plastic syringe attached to cannula during rotation of the paddles.
There was more than twofold increase of the dissolution rate providing sharper profile than
75 rpm (Table 4, Fig. 4a & 4b). The compact mass was quickly displaced which reflected
with approximately 100% drug released evident at the end of 10 minutes compared to 45
minutes with 75 rpm.

Fig. 3

Regardless to low reproducibility between samples of the reference batches, this was
not the case with the suspensions of the test product - no significant variability between
individual suspensions samples in hemispheric vessel were demonstrated (Fig. 5b). This
observation was related with no compact accumulation of the test product under the rotating
paddle and eventually due to different formulation design (Fig. 3).

Fig. 4

Effects of different geometry of dissolution vessel

11
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Even though with 100 rpm no accumulation of the suspension from the reference
product was accomplished, yet the produced results should be considered with high
suspicion. It is well known that higher agitation rates compromises the ability of the method
to discriminate whether similarity or dissimilarity between products occurs due to
formulation or manufacturing attributes or due to experimental conditions (Quershi, 2006).
From the results presented in Fig. 4, all dissolution profiles reveal great similarity with no
significant difference between reference (B-IR and C-IR) and test products (A16 and A18).
All four batches exhibit dissolution rate above 85% for period of 15 minutes.

Fig. 5

Lower agitation rates of 75 rpm in hemispheric vessel point out difference in the
dissolution profile of test and reference batches. The percentage of drug released differed
significantly during the initial 10 minutes (Fig. 5a). Around 45% were released form all
reference batches and test batch A16, compared to 95 % released form A18. As dissolution
proceeded, there appeared to be no difference in the release profiles indicating similarity
between reference products and A16 test batch.
However, when applying non-compendial peak vessel at the same agitation speed of
75 rpm, the difference in dissolution behavior became more distinctive and totally different
from the previously presented. As seen in Fig. 5c, all commercial batches reviled similar
dissolution profile with test batch A18, approximately 100% released drug within 10 minutes
compared to 55% released from the test batch A16. In addition, the dense compact formation,
which occurs with the reference products, was immediately eliminated appearing with great
reproducibility in the dissolution data. The variability ranged from 0.4-6% in peak vessel
compared to 5-24% relative standard deviation (RSD) in hemispheric vessel (Fig. 5b & 5d).
The presence of a protrusion at the bottom of the vessel pronounced existing of better mixing
environment underneath the paddle resulting with less variability in the dissolution (released)
data.

Optimisation of agitation speed and in vitro evaluation

12
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

In attempt to challenge the robustness of the dissolution condition, deliberate

manipulation of the hydrodynamic effects in peak vessel were performed by applying lower
agitation rate to a point of desired discrimination of the reproducibility of the reference drug
product (Liu and Vivilecchia, 2005).

Fig. 6

From the results presented in Fig. 6b, at 50 rpm significant variability occurs during
whole dissolution period with RSD ranging 10 - 45%, displaying extensive mounding.
Average released percent was less than 80% for a period of 45 minutes which excludes 50
rpm as relevant operating speed. Nevertheless, operating speed of 60 and 65 rpm provided
relevant dissolution results of about 95% release within 10 minutes and satisfactory
variability of less than 10% RSD during dissolution test. Based on visual observation and
robustness evaluation of hydrodynamic sensitivity, the dissolution procedure of 65 rpm
reflects more rugged and reliable dissolution profile of the reference product than 75 rpm.
Therefore, the rotational speed of 65 rpm was chosen for performing in vitro release testing.
Analyzing the graphics presented in Fig. 7a, when operated on modified USP 2
apparatus with peak vessel, on 65 rpm agitation speed, without media replacement during
sampling period and by applying the suspension samples in the media during rotation of the
paddles, similarity of test batch A18 and apparent dissimilarity of test batch A16 in
comparison to reference batches (B-DE, B-IR) in pH 7.2 can be deduced. In analogy to
previous conclusion, consistency of the dissolution behavior was verified by performing the
dissolution study in pH 1.2, pH 4.5 and pH 6.8 as recommended for simulating physiological
environment. The performed analysis, in relation to reference batch B-DE, excluded test
batch A16 as formulation with similar in vitro behavior and affirmed sameness in the released
rate with test batch A18 (Fig. 7b, 7c, 7d).
Additionally to close up the in vitro outcome between different formulations
(generic A18 vs brand formulation batch B-DE), model independent approach of f2 similarity
was applied by performing the dissolution test on additional 6 doses (total of 12 individual
doses of the products) in above mentioned different media (Table 5). In pH 7.2 and pH 6.8
13
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

both products release more than 85% of the drug within 15 minutes, which according to the
Guideline on the investigation of bioequivalence (CPMP/EWP/QWP/1401/98 Rev.1) the
profiles are accepted as similar without further mathematical evaluation. High values for f2
above 70 in pH 4.5 and pH 1.2, indicates very similar drug release behavior between test and
reference batch which in terms of similarity they exhibit in vitro equivalency. As expected,
the chosen dissolution strategy reflected minimal variability in comparison to compendial
hemispheric vessel providing more reliable results for differentiation of dissolution profiles
as a reflection of formulation change.

Fig. 7
Table 5

Conclusions

Understanding the physicochemical properties of the drug product is crucial for

determining the most effective methodology for predicting in vitro similarity between two
products. Suspensions with high viscosity require more work before a method can be
recommended due to its complexity and data variability. For highly viscose suspension,
special attention should be paid on sample homogeneity prior introducing in the vessel and
using standardized sample introduction procedure to ensure accurate and repeatable results.
As presented in this investigation, traditional compendial apparatus 2, although
recommended as first choice for dissolution methodology for oral suspension, is often not
the best approach for some highly viscose oral suspension due to its complex matrix
composition. Taking in consideration that when applying higher agitation rates of 100 rpm,
as recommended for preventing mounding at the bottom of the vessel, the discriminative
power of the method will be decreased therefore unable to distinct more realistic dissolution
behavior between different formulations which could result with misinterpretation of the
dissolution data. The presented non-compendial strategy: replacing hemispheric with peak
vessels; introducing suspension sample in vessel during rotating paddles; no replacement of
the withdrawn aliquots of samples with equal volume of fresh media during multipoint

14
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

dissolution testing, provides better hydrodynamics by removing the mounding and

preventing adhesion of the suspension in the round bottom vessel. Using this non-compendial
test helped overcoming the above-mentioned disadvantages of apparatus 2 thereby more
discriminative technique with relevant reproducible data for predicting in vitro equivalency
between different drug products (formulation) was provided.

References

Brown, C.K., Chokshi, H.P., Nickerson, B., Reed, R.A., Rohs, B.R., Shan, P.A., 2004.
Acceptable Analytical Practices for Dissolution Testing of Poorly Soluble
Compounds. Dissolution Technologies 28(12), 56. Available at:
https://doi.org/10.14227/DT120405P6.
Brown, C.K., Friedel, H.D., Barker, A.R., Buhse, L.F., Keitel, S., Cecil, T.L., Kraemer, J.,
Morris, J.M., Reppas, C., Stickelmeyer, M.P., Yomota, C., Shah, V.P., 2011.
FIP/AAPS Joint Workshop Report: Dissolution/In Vitro Release Testing of
Novel/Special Dosage Forms. Indian Journal of Pharmaceutical Sciences 73(3), 338–
353.
Collins, C.C. and Nair, R.R., 1998. Comparative evaluation of mixing dynamics in USP
apparatus 2 using standard USP vessels and PEAKTM vessel. Dissolution
Technologies 5(2), 17–21. Available at: https://doi.org/10.14227/DT050298P17.
CPMP/EWP/QWP/1401/98 Rev.1, E.M. Agency, 2010. Guideline on the investigation of
bioequivalence. Available at: https://www.ema.europa.eu/en/documents/scientific-
guideline/guideline-investigation-bioequivalence-rev1_en.pdf.
Fotaki, N. and Vertzoni, M., 2010. Biorelevant Dissolution Methods and Their Applications
in In Vitro-In Vivo Correlations for Oral Formulations. The Open Drug Delivery
Journal 4(2), 2–13. Available at: https://doi.org/10.2174/1874126601004020002.
Lagace, M., Gravelle, L., di Maso, M., McClintock, S., 2004. Developing a discriminating
dissolution procedure for a dual active pharmaceutical product with unique solubility
characteristics. Dissolution Technologies 11(1), 13–17. Available at:
https://doi.org/10.14227/DT110104P13.
15
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Mirza, T., Joshi, Y., Liu, Q. J., Vivileccia, R., 2005. Evaluation of dissolution hydrodynamics
in the USP, Peak TM and Flat-Bottom Vessels. Drugs, 11–16. Available at:
https://dx.doi.org/10.14227/DT120105P11.
Parker, J. and Gray, V., 2006. Highlights of the AAPS workshop on dissolution testing for
the 21st century. Dissolution Technologies 13(3), 26–31. Available at:
https://doi.org/10.14227/DT130306P26.
Qureshi, S.A. and Shabnam, J., 2001. Cause of high variability in drug dissolution testing
and its impact on setting tolerances. European Journal of Pharmaceutical Sciences
12(3), 271-276.
Qureshi, S.A., 2006. Developing discriminatory drug dissolution test and profiles: Some
thoughts for consideration on the concept and its interpretation. Health (San
Francisco), 18–23.
Shohin, I.E., Grebenkin, D.Y., Malashenko, E.A., Stanishevskii, Y.M. and Ramenskaya,
G.V., 2016. A brief review of the FDA dissolution methods database. Dissolution
Technologies 23(3), 6–10. Available at: https://doi.org/10.14227/DT230316P6.
Siewert, M., Dressman, J., Brown, C., Shah, V., Robinson, J., Shah, V., Uppoor, R. and
Williams, R., 2003. FIP/AAPS Guidelines for Dissolution/In Vitro Release Testing
of Novel/Special Dosage Forms.
9th Commission, E.P., 2017, 5.17.1 Recommendations on Dissolution Testing., Ph. Eur.
(07/2010:51701), 761-763.

16
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Резиме

Некомпендијали vs компендијални аналитички тестови – моќна алатка за

предвидување на in vitro сличност помеѓу орални суспензии со голема
вискозност

Елена Казанџиевска1*, Ива Антова1, Славица Митревска1, Александар Димковски1,

Елена Димов1, Маја Хаџиева Гиговска1, Пацка Антовска1, Соња Угарковиќ1,
Јасмина Тониќ Рибарска2, Сузана Трајковиќ Јолевска2

1
Истражување и Развој, Aлкалоид AД Скопје, Бул. Александар Македонски 12, 1000
Скопје, Република Северна Македонија

2
Фармацевтски факултет, Универзитет „Св. Кирил и Методиј”, Мајка Тереза 47,
1000 Скопје, Република Северна Македонија

Клучни зборови: орални суспензии, in vitro ослободување, хидродинамична

варијабилност, USP апарат 2/ Апарат со примена на весла, peak чаши

In vitro тестовите за испитување степен на растворливост покажуваат

значителен пораст на искористување при евалуација и карактеризација на
фармацевтските производи. Очекувањата од методот за дисолуција се во насока на
соодветна проверка на конзистентноста на фармацевтските атрибути преку
дискриминирање на сличности и различности помеѓу различни дозажни формулации.
Зголемениот развој на „специјални” дозажни форми, во однос на начинот на
ослободување на активната супстанцијa, изискуваат примена на некомпендијални
стратегии за растворливост кои што се разликуваат од традиционалните
фармакопејски препораки.
За демонстрирање на сличност во профилот на ослободување помеѓу генерички
и оригинаторски високо вискозни суспензии за орална употреба, во овој труд е
17
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

прикажана компаративна in vitro студија на растворливост со употреба на

некомпендијални чаши со испакнато дно наспрема фармакопејски хемисферни чаши.
При употреба на хемисферни чаши, поради формирање на компактна маса на
дното од чашите, сите евалуирани оригинаторски серии покажаа значителна
варијабилност во резултатите. Со цел намалување варијабилност во добиените
резултати, различни стратегии на манипулација со примерокот, пред и за време на
изведба на тестот за растворливост, беа применети. Употребата на чаши со испакнато
дно, резултираше со голема репродуцибилност, а со тоа и поголема сигурност во
проценката на in vitro сличност помеѓу различни формулации на орални суспензии.
Несоодветна интерпретација на резултатите од тестот на растворливост може
да доведе до негативно влијание врз целокупниот исход од развојот на фармацевтскиот
препарат. За таа цел значително важна е визуелната обсервација и проценка на
однесувањето на препаратот за време на тестот за растворливост при донесување
одлука за правилен избор на стратегија за изведување на тест на растворливост.

18
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Table 1. List of samples used for in vitro dissolution study

Viscosity Country of
Batch code Exp. date Status
(cP) origin
A-16 ~ 800 NA Macedonia Test product
A-18 ~ 1300 NA Macedonia Test product
B-IR ~ 1400 04.2017 Ireland Reference product
C-IR ~ 1400 05.2017 Ireland Reference product
B-DE ~ 1400 10.2018 Germany Reference product

19
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Table 2. API solubility in four different buffer media

Medium Solubility (mg/mL)
Buffer pH 1.2 0.029
Buffer pH 4.5 0.086
Buffer pH 6.8 2.60
Buffer pH 7.2 4.35

20
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Table 3. In vitro drug release of reference product when applied with compendial conditions on USP 2 in pH 7.2 buffer
position of
75 rpm/ non-rotating paddles/ with media replacement 100 rpm/ non-rotating paddles/ with media replacement
device for transfer application in the
(n=6) 5 min 10 min 15 min 20 min 30 min 45 min 60 min 5 min 10 min 15 min 20 min 30 min 45 min 60 min dissolution vessel
mean (%) 18.6 28.7 36.0 41.8 51.2 71.3 80.0
/ 5 mL dose syringe above the medium
RSD (%) 19.4 18.2 17.7 14.1 10.7 25.0 17.4
mean (%) 16.8 23.9 31.9 38.1 46.8 59.3 70.8 28.5 54.6 80.0 84.4 96.5 95.9 / in the medium at the
5 mL dose syringe
RSD (%) 20.3 10.4 5.5 5.4 5.1 3.7 3.3 17.6 17.3 18.0 20.0 2.2 3.5 / side of the vessel
mean (%) 35.2 55.0 71.0 78.1 84.1 89.6 93.2 in the medium at the
/ 10ml plastic syringe
RSD (%) 16.5 19.0 26.3 21.3 16.9 9.7 4.6 side of the vessel
mean (%) 21.5 28.8 35.5 42.1 60.0 85.8 102.3 10ml plastic syringe at the bottom of the
/
RSD (%) 30.3 25.3 19.6 13.3 13.3 14.9 1.1 attached with cannula vessel

21
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Table 4. In vitro drug release of reference product when applied with non-compendial conditions on USP 2 in pH 7.2 buffer
position of
75 rpm/ rotating paddles/ without media replacement 100 rpm/ rotating paddles/ without media replacement
device for transfer application in the
(n=6) 5 min 10 min 15 min 20 min 30 min 45 min 60 min 5 min 10 min 15 min 20 min 30 min 45 min 60 min dissolution vessel
mean (%) 13.7 26.8 35.2 42.8 65.0 83.1 / in the medium at the
/ 5 mL dose syringe
RSD (%) 25.5 19.7 13.4 11.3 25.7 22.2 / side of the vessel
mean (%) 31.1 47.1 60.8 75.9 90.7 98.3 100.2 73.14 101.01 101.73 101.74 101.71 / / 10ml plastic syringe in the medium at the
RSD (%) 9.9 18.6 21.0 20.1 15.7 9.0 4.8 5.91 1.23 0.23 0.16 0.16 / / attached with cannula side of the vessel
mean (%) 16.4 28.4 34.6 41.7 63.5 92.5 101.0 10ml plastic syringe at the bottom of the
/
RSD (%) 18.2 16.0 13.1 11.7 24.2 14.6 1.0 attached with cannula vessel

22
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

Table 5. f2 similarity factor in pH 7.2, pH 6.8, pH 4.5 and pH 1.2 performed in peak vessels
between test product (A18) and reference product (B-DE), on 65rpm/ application on rotating
paddles/ without media replacement
A18 (peak v.)
(n=12) pH 7.2 pH 6.8 pH 4.5 pH 1.2
B-DE (peak v.) NA* NA* 73 80
*NA-not applicable (According to the Guideline on the investigation of bioequivalence, Doc. Ref. CPMP/EWP/QWP/1401/98
Rev. 1, where more than 85.0% of the drug is dissolved within 15 minutes, dissolution profiles may be accepted as similar
without further mathematical evaluation)

23
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

a) b)
Fig. 1. Dissolution behavior of reference suspension when applied above the medium -
particles of suspension floating on the surface of the medium (a), suspension adhering on the
paddle (b).

24
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

1 2 3

Fig. 2. Dissolution behavior of reference product during sampling intervals in hemispheric

vessel, (1, 3- adhered mound; 2 - displaced mound).

25
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

a)

b)

Fig. 3. Dissolution behavior of test product in hemispheric vessel applied on rotating paddles,
observed at the beginning (a) and at middle of dissolution period (b).

26
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

100rpm_pH 7.2 100rpm_pH 7.2

110 10
100
90
80
% released

RSD (%)
70
60
5
50
40
30
20
10
0 0
0 5 10 15 20 25 30 35 40 45 50 55 60 0 5 10 15 20 25 30 35 40 45 50 55 60
Time (minutes) Time (minutes)
Batch C-IR Batch B-IR Batch C-IR Batch B-IR
Batch A18 Batch A16 Batch A16 Batch A18

a) b)
Fig. 4. In vitro release profile of reference products B-IR, C-IR, and test products A16, A18
performed on 100 rpm in pH 7.2 hemispheric vessel/application on rotating paddles/without
media replacement (a - % released in relation of time; b - RSD (%) in relation of time).

27
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

hemispheric v._75rpm_pH 7.2 hemispheric v._75rpm_pH 7.2

110 30
100
90
80
20
% released

RSD (%)
70
60
50
40 10
30
20
10
0 0
0 5 10 15 20 25 30 35 40 45 50 55 60 0 5 10 15 20 25 30 35 40 45 50 55 60
Time (minutes) Time (minutes)
Batch B-DE Batch A16 Batch B-DE Batch A16
Batch A18 Batch B-IR Batch A18 Batch B-IR
Batch C-IR Batch C-IR

a) b)

peak v._75rpm_pH 7.2 peak v._75rpm_pH 7.2

110 10
100
90
80
70
RSD (%)
% released

60
5
50
40
30
20
10
0 0
0 5 10 15 20 25 30 35 40 45 50 55 60 0 5 10 15 20 25 30 35 40 45 50 55 60
Time (minutes) Time (minutes)

Batch A16 Batch A18 Batch A16 Batch A18

Batch B-IR Batch C-IR Batch B-IR Batch C-IR

c) d)
Fig. 5. In vitro release profile of reference products B-IR, C-IR, B-DE and test products A16,
A18 performed on 75 rpm in pH 7.2 hemispheric vessel (a - % released in relation of time; b
– RSD (%) in relation of time) and peak vessel (c - % released in relation of time; d – RSD
(%) in relation of time) application on rotating paddles/without media replacement.

28
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

peak v._pH 7.2_Batch B-DE peak v._pH 7.2_ Batch B-DE

60
120
50
100
40

rsd (%)
% released

80
30
60
20
40
10
20
0
0 0 5 10 15 20 25 30 35 40 45 50 55 60
0 5 10 15 20 25 30 35 40 45 50 55 60
Time (minutes)
Time (minutes)

50rpm 60rpm 65rpm 50rpm 60rpm 65rpm

a) b)
Fig. 6. In vitro release profile of reference products B-IR, B-DE performed on 50, 60, 65 rpm
in peak vessel in pH 7.2/application on rotating paddles/without media replacement (a, c - %
released in relation of time; b, d - rsd (%) in relation of time).

29
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

pH 7.2 pH 6.8
110 110
100 100
90 90
80 80
% released

% released
70 70
60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 5 10 15 20 25 30 35 40 45 50 55 60 0 5 10 15 20 25 30 35 40 45 50 55 60
Time (minutes) Time (minutes)
Batch B-DE_ (peak v.) Batch B-DE_ (peak v.)
Batch B-IR_ (peak v.) Batch B-IR_ (peak v.)
Batch A16_ (peak v.) Batch A16_ (peak v.)
Batch A18_ (peak v.) Batch A18_ (peak v.)

a) b)

pH 4.5 pH 1.2
70 30
60
50
% released

% released

20
40
30
10
20
10
0 0
0 5 10 15 20 25 30 35 40 45 50 55 60 0 5 10 15 20 25 30 35 40 45 50 55 60
Time (minutes) Time (minutes)
Batch B-DE_ (peak v.) Batch B-DE_ (peak v.)
Batch B-IR_ (peak v.) Batch B-IR_ (peak v.)
Batch A16_ (peak v.) Batch A16_ (peak v.)
Batch A18_ (peak v.) Batch A18_ (peak v.)

c) d)
Fig. 7. In vitro release profile of test (A16, A18) and reference products (B-DE, B-IR),
performed on 65 rpm peak vessel in pH 7.2 (a), pH 6.8 (b), pH 4.5 (c) and pH 1.2 (d),
application on rotating paddles/without media replacement.

30
*Corresponding author email: ekazandzievska@alkaloid.com.mk
doi: Original scientific paper

31
*Corresponding author email: ekazandzievska@alkaloid.com.mk