International Journal of Polymeric Materials and

Polymeric Biomaterials

ISSN: (Print) (Online) Journal homepage: [Link]

Quality by design approach for formulation

development and evaluation of carboplatin loaded
ethylcellulose nanosponges

Muhammad Ahsan Hafiz, Nasir Abbas & Nadeem Irfan Bukhari

To cite this article: Muhammad Ahsan Hafiz, Nasir Abbas & Nadeem Irfan Bukhari (2021): Quality
by design approach for formulation development and evaluation of carboplatin loaded ethylcellulose
nanosponges, International Journal of Polymeric Materials and Polymeric Biomaterials, DOI:
10.1080/00914037.2021.1933978

To link to this article: [Link]

Published online: 05 Jul 2021.

Submit your article to this journal

Article views: 18

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

[Link]
INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS
[Link]

Quality by design approach for formulation development and evaluation of

carboplatin loaded ethylcellulose nanosponges
Muhammad Ahsan Hafiz , Nasir Abbas , and Nadeem Irfan Bukhari
Punjab University College of Pharmacy, University of the Punjab, Lahore, Pakistan

ABSTRACT ARTICLE HISTORY

Nanosponges, porous nanocarriers were explored for encapsulation of carboplatin by employing a Received 8 April 2021
quality by design approach. Nanosponges were prepared by double emulsion solvent evaporation Accepted 20 May 2021
using the central composite design. Ethylcellulose, 450 mg, polyvinyl alcohol, 2.3% (w/v) and phase
KEYWORDS
evaporation rate, 760 rpm yielded the optimized spherical particles of size 272.92 ± 12.31 nm, PDI
Carboplatin; central
0.390, with zeta charge of
22.3 mV. The entrapment efficiency of carboplatin-loaded nano- composite design;
sponges was 56.27 ± 2.52%. A sustained carboplatin release of 79.03% and 95.94% within 12 h at ethylcellulose nanosponges;
pH 4.5 and 6.8, respectively suggested that nanosponges could be promising carriers for sustained optimization; quality by
release of hydrophilic therapeutic candidates. design; risk assessment;

GRAPHICAL ABSTRACT

1. Introduction interest in this technology is ascribable to proposition of the

merged advantages of both, the nano- and micro- particulate
Carboplatin is a platinum-based antineoplastic drug and an
delivery systems. They are non-collapsible structures and
effective chemotherapeutic agent for testicular, ovarian, cer-
have a porous surface, thus are capable of encapsulating a
vical, bladder, and lung cancers as well as for the other
wide variety of molecules including the lipophilic and
newly diagnosed malignancies[1]. Recently, there has been
increasing interest in the development of nanoparticulate hydrophilic drugs within their colloidal particles.
systems of carboplatin in order to reduce its unfavorable Furthermore, they release the encapsulated drugs in a pre-
side effects and improve efficacy[1,2]. However, the aqueous dictable and controlled manner[7,8].
solubility of carboplatin poses encapsulation challenges dur- In spite of certain advantages, the NS-based products are not
ing formulation as polymeric nanoparticles[3]. A recently commercially available yet. NS development as drug cargo poses
studied nanosystem is the nanosponge (NS) delivery system challenges regarding their structural viability during various
which offers a predictable release of a drug compared to the processes of encapsulation[5]. For this reason, scalable and con-
nanoparticulate delivery system where a burst drug release is sistent methods of preparation are difficult to define for this for-
of the major concern. Therefore, the NS delivery system has mulation. In this sense, the quality by design (QbD) approach
gained popularity in pursuit of achieving better drug efficacy may support and accelerate the development of NSs for the opti-
at the site of the target[4]. NSs are a novel class of drug mized drug encapsulation. QbD is a systematic risk-based
delivery systems in nanoparticles that have emerged a prom- approach covering various aspects, like quality target product
ising role in the encapsulation of both pharmaceuticals and profile (QTPP), critical quality attributes (CQAs), critical process
the cosmeceuticals. They are sponge-like structures with an parameters (CPPs) and risk assessment. These concepts have
average particle size of less than 1 lm[5,6]. The enhanced been described in the International Council on Harmonization

CONTACT Nadeem Irfan Bukhari nadeem_irfan@[Link] Punjab University College of Pharmacy, University of the Punjab, Lahore-54000, Pakistan.
ß 2021 Taylor & Francis Group, LLC
2 M. AHSAN ET AL.

(ICH) guidelines: Pharmaceutical Development-Q8 (R2)[9], 2. Materials and methods

Quality Risk Management-Q9[10], and Pharmaceutical Quality
2.1. Materials
System-Q10[1]. The regulatory agencies around the world are
now encouraging the use of such approaches for formulation Carboplatin API was procured from M/s Onko Ilac San,
development and scale-up and post-approval changes[1,2,13]. The Istanbul, Turkey. Different grades of ethylcellulose
emulsion solvent evaporation method using appropriate poly- (EthocelTM, 10 cps, 20 cps and 100 cps), dichloromethane
mer is commonly used for the preparation of NSs[14]. (DCM), polyvinyl alcohol (PVA), were obtained from
Ethylcellulose has been used as the coating material for the for- Sigma-Aldrich, Pakistan. High-performance liquid chroma-
mulation of nano- and micro-sponges, with improved perform- tography (HPLC) grade methanol was obtained from Merck,
ance[15–17] for encapsulation of drugs due to its hydrophobic Germany. All the reagents used were of analytical grade.
character[8,18,19]. Generally, the solvent evaporation method is
used for the encapsulation of drugs within micro- and nano-
2.2. Methods
spheres. However, this method has limitations with the
hydrophilic drugs due to their tendency to diffuse out from the 2.2.1. Determination of QTPP and CQAs
polymer phase to the external aqueous phase[8,19]. According to ICH Q8, the QTPP provides a prospective
The problem of low drug encapsulation in NSs could be summary of the product attributes to achieve its desired
overcome by adopting the double emulsion (w/o/w) solvent quality concerning safety and efficacy. The QTPP classifica-
evaporation method. This method involves the preparation of tion helps in the identification of the CQAs of the drug
the primary and secondary emulsions using ultrasonication product. Based on the QTPP, the potential quality attributes
and high-speed homogenization, respectively followed by can be selected and with the help of risk the (or criticality)
removal of the organic solvent at a specific phase evaporation assessment, CQAs can be defined with justification. The
rate (stirring speed and duration)[20]. Several factors impact the CQAs extracted from the QTPP are used to monitor the
formation of NS particles by double emulsion (w/o/w) solvent process and product development[6,2,3]. Moreover, the CQAs
evaporation method including the polymer concentration, drug must be within an acceptable range and limits to ensure the
to polymer ratio, phase evaporation rate and pH of the outer performance of the product[24].
aqueous phase[21,2]. The purpose of the present study was,
therefore, to seek feasibility of loading of a model hydrophilic 2.2.2. Risk assessment and screening of factors
drug, carboplatin, in NSs using double emulsion solvent evap- Initial risk assessment was carried out to identify the factors
oration technique by exploring the QTPP, CPPs and CQAs. that may impact the quality of the product considerably.
As it is difficult to comprehend the inter-connections of Accordingly, the Ishikawa fishbone illustration (Figure 1) was
various factors in combination to establish the theoretical rela- developed to find out various material attributes (MAs) and
tionship with the properties which necessitates the use of math- process parameters (PPs) that might cause variability of CQAs
ematical models thus, the central composite design (CCD) was and may lead to product failure. Prioritization of the risk fac-
employed for optimization studies[2,3]. After determination of tors in the NS formulation process, based on the qualitative
the CQAs, a step of risk assessment, DoE studies and gener- analysis was performed using the risk estimation matrix
ation of the design space were followed under the QbD (REM)[2,3]. The study of the low, medium and high-risk fac-
approach. Thereon, the optimized formulations prepared from tors, based on the fishbone depiction led to the selection of
the developed design space were evaluated for compatibility, critical MAs (CMAs) and CPPs through REM. Then screening
morphology and drug release. of the factors was carried out for performing the optimization

Figure 1: Ishikawa fishbone illustration for carboplatin-loaded ethylcellulose nanosponges.

 INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS 3

studies. For this purpose, the failure modes and effects analysis experimental results by the following equation:
(FMEA) tool were used to screen the factors whereby each X X  X 
Y ¼ bo þ ðbi Xi Þ þ bij Xi Xj þ bii Xi 2 (1)
critical factor was quantitatively ranked in relation to the
occurrence (O), severity (S) and detectability (D) with a score where, Y represented measured response, bo was constant-
order between 1 and 10[25]. Finally, an overall risk was esti- coefficient, bi, bii and bij were the linear, quadratic and
mated by calculating the risk priority number (RPN) ¼ O  S interaction coefficients, respectively. Xi and Xij were the
 D[10,2,3,26]. FMEA-based rank-order scoring of MAs and variable parameters for the encapsulation process. The term
PPs, in respect to their effects on particle size, PDI and zeta Xi showed the main effect on a response when one variable
potential for the prepared blank NSs, was evaluated at two lev- was changed at a time. The Xij term showed the interaction
els of the following factors: the amount of ethylcellulose, the between Xi and Xj illustrated the behavior of the response
concentration of PVA, stirring speed, stirring time, phase for changing two factors simultaneously. The Xi2 represented
evaporation rate, pH of the outer aqueous phase and ultraso- polynomial term to estimate the non-linearity in the effect
nication time. However, the PDI and zeta potential were of the variable Xi. Thereafter, analysis of variance
employed only in the screening study and not carried forward (ANOVA), implemented in DXV analyzed the statistical sig-
R

for the evaluation in the optimization study.

nificance (p < .05) of the model coefficients[27].

2.2.3. Preparation of drug-loaded ethylcellulose

2.2.5. Characterization of nanosponges
nanosponges
[Link]. Entrapment efficiency and yield of nanosponges.
All the drug-loaded NSs were prepared by utilizing various
The amount of carboplatin entrapped in the NSs was esti-
quantities of ethylcellulose by double emulsion solvent evap-
mated by the reverse phase liquid chromatographic method,
oration technique[2]. Briefly, 5 mL aqueous solutions of car-
as reported previously[28]. The analytical method was carried
boplatin (10 mg/mL) was emulsified in 10 mL of DCM
out using an isocratic HPLC system, LC 2030C 3D Plus
containing ethylcellulose (330 mg–660 mg), by means of
model, with photodiode array (PDA) detector (Shimadzu,
probe sonication (Hielscher, Germany) at 100% amplitude
for 5 min to prepare a primary emulsion (w/o). This w/o Japan). A stability-indicating HPLC method was developed
emulsion was further size-reduced in 15 mL of PVA solution and validated on the Shim-pack C18-column (250 mm 
under the high-speed homogenizer (Heidolph Instruments, 4.6 mm; 5 mm) maintained at 25  C. Liquid chromatography
Schwalbach, Germany) at 10,000 rpm for 5 min to obtain a was performed with a PDA detector at 229 nm and a flow
double emulsion (w/o/w). The PVA was used as a stabilizer rate of 1 mL/min. The total run time for analytes to be sepa-
at different concentrations (1–2%, w/v) to prevent the separ- rated was 10 min. The mobile phase comprised the methanol
ation of phases. The final double emulsion was stirred at a as organic phase and a mixture of sulfuric acid, 0.001 N and
certain stirring rate (400–800 rpm) to enable the evaporation sodium sulfate, 0.02 M as an aqueous phase in the ratio of
of DCM. Following stirring of 4 h at room temperature, the 10:90. The method of analysis involved dispersion of the NS
reaction mixture was a vacuum-filtered through a 0.22 mm formulation equivalent to 10 mg of drug in 1 mL of DCM
filter to collect NSs, vacuum-dried at 40  C for 12 h and and sonication in a bath sonicator (Ultrasonic bath,
kept in a desiccator for further characterization. Ultrawave, UK) for 30 min. The resultant residue was redis-
persed in 1 mL of the mobile phase, vortexed for 5 min and
centrifuged at 8000 rpm for 15 min. The supernatant was
2.2.4. Optimization by the design of experiment
used for analysis. The NS’s yield and entrapment efficiency
After the screening studies, optimization of the significant
(%) were estimated by the following equations.
factors in relation to the major CQAs namely, particle size
(Y1) and entrapment efficiency (Y2) was attempted by apply- WeightofNSsðmgÞ
Yieldð%Þ ¼  100 (2)
ing a rotatable CCD experimental design. In this design, the Weightofcompoundsadded
selected critical factors were mathematically denoted at five
levels as –a, –1, 0, þ1, þa. In the case of the rotatable CCD DrugestimatedinNSsðmgÞ
Entrapmentefficiencyð%Þ ¼  100
approach, 23 factorial designs, 6-star points for the deter- Theoreticalamountofdrugadded
mination of curvature, and 6 center points (total 20 runs) (3)
are used for the estimation of experimental reproducibility.
All the experiments were conducted in random order and in
triplicate. The collected data were analyzed by Design-Expert [Link]. Particle size, polydispersity index (PDI) and zeta
(DXV) software, (Ver. 13, Stat Ease Inc. Minneapolis, MN, potentials. The mean particle size, PDI and zeta potential
R

USA) and the 3D response surface plots were drawn to investi- were determined by photon correlation spectroscopy (PCS)
gate the interactions between the critical factors and responses. using a particle size analyzer (Zetasizer, Malvern
Furthermore, a design space was constructed to achieve formu- Instruments, UK). All the measures were recorded at a
lation optimization by setting the CQAs range as per the wavelength of 633 nm and 90 scattering angle. In this
acceptance criteria. The validation of the design space was done method, the dispersions containing 2 mg of formulation
by preparing the checkpoint formulations. The polynomial were prepared in Millipore filtered water and immediately
models incorporating the interactive terms were fitted to the used to determine the particle diameter and PDI.
4 M. AHSAN ET AL.

Table 1. QTPP of cisplatin-loaded ethylcellulose nanosponges.

Target product profile Target Justification
Drug delivery system Nanosponge Nanosponge formulation of optimal particle size is preferred for predictable drug
release compared to other nanosystems.
Dosage type Controlled release It is pre-decided in objectives to formulate Nanosponges with double emulsion
solvent evaporation method for controlled release.
Route of administration Intermediate for a localized delivery system. Nanosponge can be used as a promising intermediate for a localized delivery system.
Particle size analysis <300 nm Lack of particle size control may lead to uneven loading of drugs and fluctuation in
release rate.
Physical state Stability of drug substance in nanostructures. Interactions between drug substance and polymer may affect the physicochemical
properties of the formulation.
Entrapment efficiency Maximum possible achieved. Low entrapment efficiency and low yields may lead to loss of drug from the system
Percent yield which indicates process-related issues as well as affect efficacy of the formulation.
Drug release Controlled release compared to free drugs. A sustained release profile is necessary to produce prolonged effect hence it is a
critical attribute. It is mainly influenced by the porosity of the particles.

[Link]. Fourier-transform infrared (FTIR) spectroscopy. [Link]. Statistical analysis. The values of particle size, PDI
The samples were subjected to FTIR measurement using and zeta potential are reported as mean ± SD. Statistical ana-
FTIR spectrometer (PerkinElmer, USA) containing attenu- lysis and the goodness of fit statistical criteria (Predicted
and Adjusted R2, Adequate precision) implemented in DXV
R
ated total reflection (ATR) Ge crystal cell. In this method,
a small amount of sample was added to the plate of the were used, respectively for the selection of significant model
ATR instrument using a spatula such that the crystal was terms (factors) and the models. Significance was considered
covered by the sample. Then the metal tip is locked onto at p-value < .05.
the plate over the crystal. By this method, spectra of
carboplatin, ethylcellulose, physical mixture and the opti-
3. Results and discussion
mized formulation were then recorded over the range
of 400–4000 cm
1. 3.1. Elements of quality by design
3.1.1. Defining QTPP and product quality attributes
[Link]. Scanning electron microscopy (SEM). Morphological
The definition of QTPP is the primary step in the QbD-
analysis of optimized NSs (C1) was performed using a scan-
driven formulation development. The QTPP consists of a
ning electron microscope (JEOL Ltd, Tokyo, Japan). NSs of
summary of quality characteristics that primarily ensure the
the selected formulation were mounted onto metal stubs
safety and efficacy of the product. The QTPP summary of
using dual-sided adhesive tape. The stubs were coated with
the carboplatin-loaded ethylcellulose NSs is presented in
the gold using a sputter coater. The particles were then
Table 1. Selection of the CQAs is performed as the second
examined for morphological studies.
step of the QbD. CQAs are the essential product quality
attributes extracted from QTPP which affect the final prod-
[Link]. In-vitro release study. The optimized formulations
uct, due to which their effect has to be monitored and
of carboplatin-loaded NSs (C1 and C2) were subjected to
explored. The CQAs of ethylcellulose NSs are listed in
the in-vitro release studies by the dialysis membrane
Table 2. It is pertinent to mention that the prepared NSs
method. Briefly, the dried NSs equivalent to 10 mg of the
may act as intermediate carriers for a localized delivery sys-
drug was dispersed in 5 mL of Millipore water and inserted
tem. Therefore, all the CQAs and CPPs were explained with
in a dialysis bag (MWCO 12–14000 Daltons, Medicell
respect to the intermediate product.
International Ltd, 239 Liverpool Road, London) and sealed
to form a closed pouch. The dialysis bag was then sus-
pended separately in 50 mL of acetate buffer, pH 4.5 and 3.1.2. Determination of CMAs and CPPs
phosphate buffer, pH 6.8 media maintained at 37 ± 0.5  C The CMAs and CPPs affecting the quality of the product
under continuous stirring at 100 rpm. The samples (2 mL) were identified. The polymer and stabilizer concentrations
were withdrawn at predetermined time intervals up to 12 h were considered to be prominent MAs affecting the entrap-
and replenished with the same volume of fresh media main- ment efficiency and particle size. The stirring speed and
tained at the same conditions. The samples withdrawn were time of the secondary emulsification step were found as that
centrifuged at 5000 rpm for 2 min. The resultant supernatant process parameter that might have impacted the particle size
was filtered through 0.45 mm to be used for HPLC analysis. distribution and entrapment efficiency. The ultrasonication
The amount of drug in the given sample was quantified time and pH of outer the aqueous phase were considered to
from the calibration curve (R2 ¼ 0.998), established during be quality attributes. Moreover, the phase evaporation rate
the method validation studies. The similarity factor (f2) and was taken into consideration to study their impact on the
kinetics of drug release were studied using the DD Solver, size and entrapment efficiency of the nanodispersion.
an add-in program for comparison and modeling of the Different material attributes and process parameters were
drug release profiles[29]. presented in Ishikawa illustration (Figure 1). The Ishikawa
 INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS 5

Table 2. REM strategy linking MAs and PPs to the CQAs by qualitative analysis.
Material attributes (Mas) and process parameters (PPs)
Critical quality Homogenization Homogenization Phase pH of the outer
attributes (CQAs) Amount of polymer PVA concentration speed time evaporation rate aqueous phase Ultrasonication time
Particle size High High High Medium Medium Low Low
Polydispersity High Medium Medium Medium Low Low Medium
index
Entrapment High High Medium High High Medium Low
efficiency
Drug release High Medium Medium Medium High Low Medium

plot elaborates the effect of the expected formulation param- performance. Homo genization speed and time to prepare
eters of the double emulsion solvent evaporation method on secondary emulsion showed reasonable RPN scores, yet to
the quality attributes of the final product. simplify the optimization and reduce process variability,
their best conditions were opted. The homogenization speed
above 10,000 rpm did not reduce particle size significantly,
3.2. Risk assessment and screening of factors
thereby, this speed and 5 min homogenization time were
The initial risk assessment studies showed that the different considered suitable to obtain an optimized and stable emul-
factors such as the amount of polymer, PVA concentration, sion. Finally, the factors with an RPN score of 100 and
homogenization speed and time, pH of the outer aqueous above were identified as the critical parameters/attributes
phase, ultrasonication time and phase evaporation rate, might and carried forward for further optimization studies by
have a considerable effect on the PDI and particle size. After DoE. In this study, the critical factors included the amount
the qualitative evaluation, these factors were marked as with of polymer, PVA concentration and phase evaporation rate.
low, medium, and high risk based on their potential of risk
as per REM methodology (Table 2). The published reports
suggested the effect of amounts of ethylcellulose and PVA on 3.3. Analysis of optimization data
the NS size and entrapment efficiency. A study, using Different formulation parameters and their coded levels
Ishikawa figure revealed the effect of PVA and ethylcellulose (CMAs) selected for preparation of nanosponges as per CCD
concentration on NS size, PDI and entrapment efficiency by experimental design are shown in Table 4. The measured
Box Behnken design[30]. The effect of ethylcellulose on the responses of the formulations, evaluated using Design ExpertV
R

particle size of the econazole nitrate NSs was also software are presented in Table 5. The software suggested dif-
reported[18]. Another study demonstrated the effect of the ferent model sources for each response based on the results of
ethylcellulose:PVA ratio and stirring rate on the particle size Adjusted R2, Predicted R2, sequential p-value and predicted
and in-vitro release[21]. Moreover, process parameters related residual error sum of squares (PRESS), as shown in Tables 6a
to the double emulsion solvent evaporation method were and 6b. The quadratic models were suggested for both
reported, wherein the pH of the outer aqueous phase, ultraso- responses, particle size and the entrapment efficiency based on
nication time and solvent removal rate were found to be the
the considerable agreement between the adjusted R2 and
potential risk factors[28,31].
Predicted R2 values estimated for each of the linear, cubic,
Before the selection of the final critical parameters/attrib-
quadratic and two-factor interaction (2FI) models. The follow-
utes, trial batches were prepared. Using information gained
ing equations of quadratic models were obtained:
from the extensive literature survey and trial results, evalu-
ation of the risk assessment was performed following the Y1 ¼ 384:1 þ 71:87X1
63:66X2
78:51X3 (4a)
guidelines of FMEA, as shown in Table 3. In this method,
the rank order score by different parameters aided in segre- Y2 ¼ 45:14 þ 6:56X1 þ 4:59X2 þ 2:42X3 þ 1:07X2 2 þ 1:10X1 2
gation of the risk potential of the MAs and PPs, hence the
þ 0:345X3 2
0:1407X12
2:55X13 0:97X23
screening of the low-risk factors from the high-risk ones.
The literature and the present findings supported polymer (4b)
and PVA concentrations as being CMAs for the particle size In Eqs. 4a and 4b, Y1 represents the particle size and Y2
and entrapment efficiency and, thus were classified as the shows entrapment efficiency, whereas X1, X2, X3 represent
high-risk factors. Different viscosity grades of ethylcellulose the three CMAs, i.e., the amount of polymer, PVA concen-
caused the minimal variability in drug release, therefore eth- tration and phase evaporation rate, respectively. The positive
ylcellulose 10 cps was selected for the optimization studies. sign shows the synergistic effect of independent factors on
The phase evaporation rate of organic solvent was CPP and the response while the negative sign reflects the antagonistic
the high-risk factor, as it should not be beyond the permis- effect of the factor on the response.
sible limit. While pH of the outer aqueous phase was a low-
risk factor as it revealed almost pH-independent drug solu-
bility. The ultrasonication time and amplitude, along with 3.3.1. Effect of factors on particle size
other parameters of primary emulsification, were the low- The average diameter of particles varied from 216.4 nm to
risk factors as they did not ameliorate the process 625.9 nm, indicating the reliance of the response toward the
6 M. AHSAN ET AL.

Table 3. FMEA based rank-order scoring of MAs and PPs in respect to their effects on particle size and PDI for prepared
blank nanosponges, evaluated by two levels of each factor (number of formulations ¼ 14).
MAs & PPs Severity (S) Detectability (D) Occurrence (O) RPN score
Amount of polymer 7 6 6 252
PVA concentration 5 5 6 150
Stirring speed 5 4 5 100
Stirring time 4 3 4 48
Phase evaporation rate 7 6 5 210
pH of outer aqueous phase 2 1 2 4
Ultrasonication time 2 1 1 2
Abbreviations: PDI: polydispersity index; RPN: risk priority number.

Table 4. Levels of critical factors investigated for CCD experimental domain. Table 5. CCD-based experimental runs along with the measured responses.
Coded levels Coded levels of variables Responses
Critical factors ̶ 1.68 ̶1 0 þ1 þ1.68 Trial X1 X2 X3 Particle size (nm) Entrapment efficiency (%)
X1 – Amount of polymer (mg) 214.5 330 495 660 775.5 01 ̶1 ̶1 ̶1 464.8 ± 8.5 37.7 ± 2.5
X2 – Percent of PVA (%, w/v) 0.65 1 1.5 2 2.35 02 1 ̶1 ̶1 625.9 ± 9.3 46.6 ± 3.0
X3 – Phase evaporation rate (rpm) 260 400 600 800 940 03 ̶1 1 ̶1 342.8 ± 4.1 44.1 ± 0.8
04 1 1 ̶1 465.1 ± 8.3 57.2 ± 1.3
05 ̶1 ̶1 1 300.7 ± 3.9 39.7 ± 2.9
06 1 ̶1 1 463.0 ± 10.6 52.9 ± 1.1
factors under study (Table 5). The particle size distribution 07 ̶1 1 1 216.4 ± 6.2 47.5 ± 1.6
and zeta potential are shown in Figures 2a and 2b, respect- 08 1 1 1 320.3 ± 4.3 64.3 ± 3.2
09 ̶ 1.68 0 0 279.4 ± 12.3 34.4 ± 2.7
ively. The smallest particle size could be obtained with the 10 1.68 0 0 538.8 ± 5.2 56.5 ± 3.2
low levels of X1 and the higher levels of both, X2 and X3. 11 0 ̶ 1.68 0 542.1 ± 9.8 44.72 ± 2.0
The estimated model and p-value showed that the input 12 0 1.68 0 326.0 ± 5.8 60.2 ± 3.1
13 0 0 ̶ 1.68 575.7 ± 6.4 43.8 ± 1.2
variables at individual levels were found to impact signifi- 14 0 0 1.68 291.3 ± 6.2 51.9 ± 2.8
cantly on the response. In contrast, the effects of factor inter- 15 0 0 0 381.0 45.1
action were little, suggesting the influence of one variable at 16 0 0 0 379.5 44.2
17 0 0 0 380.2 46.0
a time on the particle size. A test for ranking significance in 18 0 0 0 388.5 42.1
the regression model was performed based on p-value and 19 0 0 0 382.0 45.4
F-value. The smaller probability value (Prob > F) and the 20 0 0 0 386.6 45.2
larger F-value showed that the model coefficient was statistic-
ally significant. For particle size, linear model terms were
statistically significant terms as shown in Eq. 4a. The particles[3,4]. The size of the prepared nano-dispersion
adequacy of the model was determined by the adjusted R2. decreased remarkably by the increasing phase evaporation
In this case, the value of adjusted R2 was greater than 90%, rates (Figures 3b and 3c). Higher stirring rates lead to the
showing a considerable correlation between factors and faster evaporation of organic solvent from the final product,
responses[32]. thus the smaller size particles are formed[35].
The coefficient values in Eq. 4a depicted a synergistic
effect of the polymer concentration on the particle size. In
contrast, a reduction in the particle size was observed by an 3.3.2. Effect of factors on entrapment efficiency
increase in the PVA concentration and phase evaporation Relationships between the percent entrapment efficiency and
rates. The phase evaporation rate affected the particle size to the factors under study have been given in Table 5. The
a greater extent compared to the amount of the polymer amount of the carboplatin entrapped in the NSs was ranged
and PVA concentration alone. The 3D response surface from 34.4% to 64.3%, showing sensitivity of entrapment effi-
plots for the particle size with respect to CMAs are shown ciency toward the critical factors. The yield of the prepared
in Figure 3. A moderate increase in NS size occurred by NSs was fairly consistent withing range 61.2–64.5%, so this
increasing the level of polymer (Figures 3a and 3b), ascribed response was not included in the factor or optimization
to the higher viscosity of the organic phase due to the study. Information of the optimum levels of formulation var-
higher polymer content, resisting breaking down of droplets iables is required for obtaining the highest entrapment effi-
during emulsification, thereby facilitating the formation of ciency[3]. The highest drug entrapment was achieved when
larger particles[2]. However, the increasing amounts of poly- levels of all factors under study corresponded to 1. The posi-
mer led to the instability of primary emulsion due to an tive values of the coefficients of X1, X2 or X3 in the Eq. 4b
increased viscosity of the internal phase, causing coalescence indicated an increase in drug entrapment efficiency with the
of droplets[3]. increase in levels of X1, X2 or X3. The amount of polymer
The increasing concentration of the PVA led to a had a greater impact on the drug encapsulation compared to
decrease in the particle size (Figures 3a and 3c). This was the other two variables. The PVA concentration had a mod-
attributed to the higher concentration of stabilizer needed at erate influence whereas the phase evaporation rate had a
the oil-water interface to reduce the free energy which minimal effect. The influence of the polymer concentration
resisted the formation of floccules and aggregates of the on the drug entrapment is shown in Figures 4a and 4b.
 INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS 7

Table 6a. Statistical values of responses for different model sources, along with remarks suggested by Design-Expert.
Response Source Sequential p-value R2 Predicted R2 Adjusted R2 Press Remarks
Particle size Linear <.0001 0.9623 0.9379 0.9551 13,679 –
Quadratic .0405 0.9862 0.8949 0.9732 23139.05 Suggested
2FI .4092 0.9696 0.9425 0.9554 12672.42 –
Cubic .8666 0.9885 – 0.9637 – Aliased
Entrapment efficiency Linear <.0001 0.8852 0.8145 0.8637 201.87 –
Quadratic <.0001 1.0000 0.9997 0.9999 0.4483 Suggested
2FI .5057 0.9035 0.7876 0.8589 230.83 –
Cubic .2111 1.0000 0.9950 0.9999 – Aliased
Cases with leverages of 1 are not defined and left blank.

Table 6b. Goodness of fit statistical criteria authenticating the selection of model for each response (the p-values < .05 indicates sig-
nificant model terms).
p-value
Response Model F-value Adequate precision Standard deviation Prob > F
Particle size Quadratic 79.61 34.769 17.41 .0007
Entrapment efficiency Quadratic 20508.02 561.58 0.0768 <.0001

Figure 2. (a) Particle size distribution of carboplatin loaded ethylcellulose NSs (C1), (b) Zeta potential of C1.

Figure 3. 3D response surface plots showing average particle size with respect to two CMAs while keeping the third factor at a constant level.

Due to its hydrophilic nature, the drug might get leached The increasing PVA concentration led to a moderate
out to the external aqueous layers due to the concentration increase in drug entrapment (Figures 4a and 4c). This was
gradient. Therefore, an increase in the polymer concentration due to the increased viscosity of the PVA solution, posing
resulted in the higher viscosity of the organic phase which greater difficulty in the drug partitioning across the inter-
led to the decreased carboplatin diffusion into the aqueous face, causing the drug retention in the NS[31]. However,
layer. Moreover, an increased polymer content provided the higher stabilizer content could increase the drug adsorption
larger volume available for enclosing the drug into the par- onto the particle surface, thus decreasing the amount of
ticle surface, thus improving the drug encapsulation[36]. drug retained in the NSs. These opposing effects could
8 M. AHSAN ET AL.

Figure 4. 3D response surface plots showing average particle size with respect to two CMAs while keeping the third factor at a constant level.

Figure 5. Optimum zone indicated in yellow color by graphical optimization using DXV after defining the constraints of CQAs for nanosponge size between 250
R

and 300 nm and percent entrapment efficiency between 55 and 60%. (a) The phase evaporation rate is kept constant at 800 rpm and in (b) PVA concentration is
kept constant at 2% w/v.

influence the retention of carboplatin in the NSs. Thus, drug 55 and 60% as shown in Figure 5. Definitely, the control
encapsulation did not show a linear relationship with the strategy for CQAs of design space was proposed such that
PVA concentration. Figures 4b and 4c indicated that the main objective of optimizing the NS formulation and
the entrapped amount of the drug was increased with the process could not be hindered by the unrealistic targets.
increase in phase evaporation rate. The barrier formed Through navigation of design space, two checkpoint formu-
arround NSs, as a result of stirring might decrease the possi- lations (C1 and C2) were formulated as per the composition
bility of carboplatin movement to the external phase, given in Table 7. After characterization, the results of both
thereby improving the encapsulation[3,4]. formulations were tallied and the difference between the
above formulations was assessed. Satisfactory results were
observed within limits of the desired quality attributes.
3.4. Design space and validation of QbD output
Formulation C1 was predicted with the particle diameter of
An optimization and validation process was carried out by 251.31 nm with 59.37% entrapment efficiency. The predicted
means of DXV software to probe the optimized formula for
R
and observed characteristics of the checkpoint formulation
the preparation of carboplatin-loaded NSs[31]. Therefore, a C1 were in good agreement, thus were considered reprodu-
design space was constructed by overlaying the contour cible. Similarly, the predicted values of the particle size and
plots after setting the constraints of CQAs – the particle size entrapment efficiency for C2 were found to be 283.01 nm
between 250 and 300 nm and entrapment efficiency between and 59.37%, respectively. The findings (Table 7) depicted
 INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS 9

Entrapment efficiency (%)a

that the values were within the 95% confidence interval
range of both formulations.

56.27 ± 2.52
61.23 ± 3.14
The results exhibited that different levels of CMAs,
selected from within the design space for formulating nano-
dispersions could yield similar results as predicted by soft-

Experimental results
ware. Therefore, it was demonstrated that the design space
formed could reduce the risk of product failure by control-
ling variations within the permissible limits.

3.5. FTIR analysis

Particle size (nm)a
272.92 ± 12.31
305.21 ± 10.04

The FTIR spectra of pure drug, polymer, physical mixture

and the carboplatin-loaded NSs are shown in Figure 6. The
FTIR spectrum of carboplatin showed an absorption band at
3259.70 cm–1 corresponding to the N–H stretching of amino
groups. The absorption band at 1631.78 cm–1 corresponded
to the presence of the carbonyl functional group. The pres-
ence of aliphatic C-H stretching vibrations was evident from
the characteristic peak observed at 3136.26 cm–1. The func-
Entrapment efficiency (%)

tional groups of ethyl cellulose were characterized by an

absorption peak at 1053.13 cm–1 which was attributed to the
C–O–C stretching vibrations. The absorption peaks at
59.37
55.80

2870.08 cm–1 and 2974.23 cm–1 corresponded to the C–H

stretching frequency, as shown in Figure 6b. The absorption
Predicted values

peaks for the carboplatin due to N–H stretching along with

the specific peaks of ethylcellulose were observed in the both
physical mixture (Figure 6c) and the drug-loaded NSs
(Figure 6d). The FTIR scan of the physical mixture did not
Table 7. Optimized (checkpoint) formulations of drug-loaded nanosponges prepared by navigation of design space.

display any significant interaction between the substances.

Particle size (nm)

The spectrum of the carboplatin-loaded polymeric NSs

revealed that the drug sustained its chemical structure after
251.31
283.04

the formation of NSs.

3.6. Morphological characteristics

The surface characteristics of the NSs are important to
determine the release kinetics[18]. When the particles were
X3 (rpm)

scanned under the electron microscope, the NSs were spher-

760
770

ical in shape with a porous outer peel (Figure 7). The por-
ous surface might be due to the phase evaporation during
the hardening of the particle. This surface porosity of
sponges is an important feature in the delivery system for
the controlled release characteristics of the entrapped drug
(% w/v)

particles[36]. The inner side of NSs was formed by void

2.3
2.0
X2

structure with spaces on the rigid shell that may be fabri-

cated from both polymer and the drug and as a result of
which higher entrapment efficiency was observed in the
The results are expressed as mean ± SD.

NSs. However, a fraction of irregular-shaped porous ethyl-

cellulose particles was also observed probably due to second-
X1 (mg)
450
500

ary emulsification in the external aqueous phase.

3.7. In-vitro release study

The in-vitro carboplatin release profile from the optimized
Formulation ID

NS formulations C1 and C2 showed a biphasic type sus-

tained-release pattern (Figure 8). Both the formulations
exhibited an initial burst release of the drug followed by the
C1
C2

slow and continual drug release pattern. The initial trend

a
10 M. AHSAN ET AL.

Figure 6. FTIR spectra of (a) carboplatin; (b) ethylcellulose; (c) physical mixture of drug and polymer; (d) optimized NS formulation.

was due to the presence of the weakly entrapped drug on NSs in the body and at the site of action may be improved.
the surface of NSs. The later sustained-release trend was It is therefore proposed that the prepared NSs may be
attributed to the slow diffusion and release of drug from the loaded into a hydrogel for further prolonging the release
polymer matrix. Higher burst drug release from the formu- and for achieving localization at the site of action.
lation was observed for the formulation C1 with a slightly Factors influencing drug release from NSs might
lower polymer amount compared to the formulation C2 con- include drug-polymer ratio, stirring speed and surfactant
taining relatively the higher polymer level which might be concentration for the secondary emulsification. The results
due to the formation of the less porous and thick poly- of the kinetic modeling with First order, Higuchi and
mer matrix. Korsmeyer Peppas showed linearity with the Higuchi model,
The burst release of formulation C1 was about 45% in 3 h as its R2 value ranging from 0.9872–0.9996 was greater
in acetate buffer, pH 4.5. On the contrary, the formulation than 0.98. It supported that the release of the encapsulated
C2 showed an initial release of 36% in 3 h with the subse- carboplatin from the NS pores followed the diffusion pro-
quent drug release of 74% in 12 h. The rate and extent of cess[38]. In this case, as the rate of phase evaporation
drug release in acetate buffer, pH 4.5 was lesser compared increases, the number of pores formed on the particle sur-
to the drug release in phosphate buffer, pH 6.8. Both formu- face increases[39,40]. Further exploration of the release mech-
lations, regardless of the pH of the media showed a consid- anism by the Peppas model showed that the release of
erable similarity in their release profiles, as indicated by the carboplatin was also by non-Fickian type since the value of
comparable values of similarity factor (f2) of 61.32 and 65.70 Peppas-derived parameter, n ranged from 0.522–0.529.
in pH 4.5 and pH 6.8 media, respectively. These results Therefore, it could be inferred that both diffusion and ero-
demonstrated that the drug release could be prolonged by sion mechanisms were involved in the drug release from the
more than 12 h compared to the pure drug (which released NSs. The above findings were suggested that the quality by
drugs within less than 3 h). Previous studies reported similar design and the risk assessment approach were supportive for
findings with regards to the ethylcellulose-based drug car- the factor analysis and optimization of the carboplatin-
riers[37]. By controlling drug release, the residence time of loaded nanosponges, in line with the previous report[41].
 INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS 11

Figure 7. Scanning electron microscopy images of optimized batch (C1) showing porous structures. Legends show the image magnification for visualizing the particles.

Figure 8. In-vitro drug release profile of nanosponge formulations studied at (a) acetate buffer pH 4.5, and (b) phosphate buffer pH 6.8 according to dialysis bag
membrane method (n ¼ 3).

4. Conclusion formulation parameters of nanosponges for optimization of

The optimized carboplatin-loaded ethylcellulose nano- particle size and drug entrapment efficiency. The optimized
sponges were successfully developed using QbD and risk nanosponges exhibited a porous structure wherein the car-
analysis, performed with FMEA. The risk potential of for- boplatin sustained its intact structure. The nanosponges
mulation and process-related parameters were identified and released carboplatin in a sustained release pattern in both
the central composite design systematically revealed the media for more than 12 h which followed the Higuchi and
12 M. AHSAN ET AL.

Peppas release mechanism. Thus, the ethylcellulose nano- development Q8(R2). [Link]
sponges could be used as promising intermediate carriers Q8_R2_Guideline.pdf
[11] ICH. International Conference on Harmonisation of Technical
for a localized delivery system. Requirements for Registration of Pharmaceuticals for Human
Use. ICH harmonised tripartite guideline: pharmaceutical quality
systems (Q10). [Link]
Acknowledgement tific-guideline/international-conference-harmonisation-technical-
requirements-registration-pharmaceuticals-human_en.pdf
The authors acknowledge Punjab University College of Pharmacy,
[12] Mishra, V.; Thakur, S.; Patil, A.; Shukla, A. Quality by Design
University of Punjab, Lahore for the provision of the required facilities
(QbD) Approaches in Current Pharmaceutical Set-Up. Expert.
for the preparation and characterization of nanosponges. We are also
Opin. Drug. Del. 2018, 15, 737–758. DOI: 10.1080/17425247.
thankful to NUST University, Islamabad for extending the Lab facilities
2018.1504768.
for performing scanning electron microscopy studies.
[13] Mehta, R.; Teckoe, J.; Schoener, C.; Workentine, S.; Ferrizzi, D.;
Rajabi-Siahboomi, A. Investigation into the Effect of
Ethylcellulose Viscosity Variation on the Drug Release of
Disclosure statement Metoprolol Tartrate and Acetaminophen Extended Release
No potential conflict of interest was reported by the author(s). Multiparticulates-Part I. AAPS PharmSciTech. 2016, 17,
1366–1375. DOI: 10.1208/s12249-015-0465-z.
[14] El-Habashy, S. E.; Allam, A. N.; El-Kamel, A. H. Ethyl Cellulose
Nanoparticles as a Platform to Decrease Ulcerogenic Potential
ORCID
of Piroxicam: formulation and In Vitro/In Vivo Evaluation. Int.
Muhammad Ahsan Hafiz [Link] J. Nanomedicine. 2016, 11, 2369–2380. DOI: 10.2147/IJN.
Nasir Abbas [Link] S93354.
Nadeem Irfan Bukhari [Link] [15] Maiti, S.; Kaity, S.; Ray, S.; Sa, B. Development and Evaluation
of Xanthan Gum-Facilitated Ethyl Cellulose Microsponges for
Controlled Percutaneous Delivery of Diclofenac Sodium. Acta
References Pharm. 2011, 61, 257–270. DOI: 10.2478/v10007-011-0022-6.
[16] Zhang, J.; Wang, X.; Liu, T.; Liu, S.; Jing, X. Antitumor Activity
[1] Jose, S.; Juna, B.; Cinu, T.; Jyoti, H.; Aleykutty, N. Carboplatin of Electrospun Polylactide Nanofibers Loaded with 5-
Loaded Surface Modified PLGA Nanoparticles: Optimization, Fluorouracil and Oxaliplatin against Colorectal Cancer. Drug.
Characterization, and In Vivo Brain Targeting Studies. Colloids Deliv. 2016, 23, 784–790. DOI: 10.3109/10717544.2014.916768.
Surf B Biointerfaces. 2016, 142, 307–314. DOI: 10.1016/[Link]- [17] Amrutiya, N.; Bajaj, A.; Madan, M. Development of
surfb.2016.02.026. Microsponges for Topical Delivery of Mupirocin. AAPS
[2] Sadhukha, T.; Prabha, S. Encapsulation in Nanoparticles PharmSciTech. 2009, 10, 402–409. DOI: 10.1208/s12249-009-
Improves Anti-Cancer Efficacy of Carboplatin. AAPS 9220-7.
PharmSciTech. 2014, 15, 1029–1038. DOI: 10.1208/s12249-014- [18] Sharma, R.; Pathak, K. Polymeric Nanosponges as an
0139-2. Alternative Carrier for Improved Retention of Econazole
[3] Desai, N. Challenges in Development of Nanoparticle-Based Nitrate onto the Skin through Topical Hydrogel Formulation.
Therapeutics. AAPS J. 2012, 14, 282–295. DOI: 10.1208/s12248- Pharm. Dev. Technol. 2011, 16, 367–376. DOI: 10.3109/
012-9339-4. 10837451003739289.
[4] Wang, X.; Wang, J.; Wu, W.; Li, H. Vaginal Delivery of [19] Selvamuthukumar, S.; Anandam, S.; Krishnamoorthy, K.;
Carboplatin-Loaded Thermosensitive Hydrogel to Prevent Local Rajappan, M. Nanosponges: A Novel Class of Drug Delivery
Cervical Cancer Recurrence in Mice. Drug Deliv. 2016, 23, System-Review. J. Pharm. Pharm. Sci. 2012, 15, 103–111. DOI:
3544–3551. DOI: 10.1080/10717544.2016.1205158. 10.18433/j3k308.
[5] Ahmed, R.; Patil, G.; Zaheer, Z. Nanosponges–a Completely [20] Iqbal, M.; Zafar, N.; Fessi, H.; Elaissari, A. Double Emulsion
New Nano-Horizon: Pharmaceutical Applications and Recent Solvent Evaporation Techniques Used for Drug Encapsulation.
Advances. Drug. Dev. Ind. Pharm. 2013, 39, 1263–1272. DOI: Int. J. Pharm. 2015, 496, 173–190. DOI: 10.1016/[Link].2015.
10.057.
10.3109/03639045.2012.694610.
[21] Aldawsari, H. M.; Badr-Eldin, S. M.; Labib, G. S.; El-Kamel,
[6] Abdul, A. H.; Bala, A. G.; Chintaginjala, H.; Manchikanti, S. P.;
A. H. Design and Formulation of a Topical Hydrogel
Kamsali, A. K.; Dasari, R. R. D. Equator Assessment of
Integrating Lemongrass-Loaded Nanosponges with an
Nanoparticles Using the Design-Expert Software. IJPSN 2019,
Enhanced Antifungal Effect: In Vitro/In Vivo Evaluation. Int. J.
13, 4766–4772. DOI: 10.37285/ijpsn.2020.13.1.5.
Nanomedicine. 2015, 10, 893–902. DOI: 10.2147/IJN.S74771.
[7] Bolmal, U. B.; Manvi, F.; Rajkumar, K.; Palla, S. S.; Paladugu,
[22] Kumar, P. M.; Ghosh, A. Development and Evaluation of Silver
A.; Reddy, K. R. Recent Advances in Nanosponges as Drug
Sulfadiazine Loaded Microsponge Based Gel for Partial
Delivery System. PCI-Approved-IJPSN. 2013, 6, 1934–1944. Thickness (Second Degree) Burn Wounds. Eur. J. Pharm. Sci.
DOI: 10.37285/ijpsn.2013.6.1.3. 2017, 96, 243–254. DOI: 10.1016/[Link].2016.09.038.
[8] Kapileshwari, G. R.; Barve, A. R.; Kumar, L.; Bhide, P. J.; Joshi, [23] Rapalli, V. K.; Khosa, A.; Singhvi, G.; Girdhar, V.; Jain, R.;
M.; Shirodkar, R. K. Novel Drug Delivery System of lulicona- Dubey, S. K. Application of QbD Principles in Nanocarrier-
zole-Formulation and Characterisation. J. Drug. Deliv. Sci. Tec. Based Drug Delivery Systems. In: Pharmaceutical quality by
2020, 55, 101302. DOI: 10.1016/[Link].2019.101302. design. Academic Press: London, 2019; pp. 255–296.
[9] ICH. International Conference on Harmonisation of Technical [24] Lan, Z.; Shirui, M. Application of Quality by Design in the
Requirements for Registration of Pharmaceuticals for Human Current Drug Development. Asian. J. Pharm. Sci. 2017, 12, 1–8.
Use. ICH harmonised tripartite guideline: quality risk manage- [25] Chang, R.-K.; Raw, A.; Lionberger, R.; Yu, L. Generic
ment Q9. [Link] Development of Topical Dermatologic Products, Part II:
guideline/international-conference-harmonisation-technical-req Quality by Design for Topical Semisolid Products. AAPS J.
uirements-registration-pharmaceuticals-human-use_en-[Link] 2013, 15, 674–683. DOI: 10.1208/s12248-013-9472-8.
[10] ICH. International Conference on Harmonisation of Technical [26] Nayak, A. K.; Ahmed, S. A.; Beg, S.; Tabish, M.; Hasnain, M. S.
Requirements for Registration of Pharmaceuticals for Human Application of Quality by Design for the Development of
Use. ICH harmonised tripartite guideline: pharmaceutical Biopharmaceuticals. In: Pharmaceutical quality by design.
 INTERNATIONAL JOURNAL OF POLYMERIC MATERIALS AND POLYMERIC BIOMATERIALS 13

Academic Press: London, 2019; pp. 399–411. DOI: 10.1016/ Entrapment Efficiency Using a w/o/w Double Emulsion
B978-0-12-815799-2.00019-8. Technique. Colloid. Surface. B. 2004, 33, 177–183. DOI: 10.
[27] Bolton, S.; Bon, C. Pharmaceutical Statistics: practical and 1016/[Link].2003.10.003.
Clinical Applications. CRC Press: New York, 2009. [35] Sharma, N.; Madan, P.; Lin, S. Effect of Process and
[28] Alex, A. T.; Joseph, A.; Shavi, G.; Rao, J. V.; Udupa, N. Formulation Variables on the Preparation of Parenteral
Development and Evaluation of Carboplatin-Loaded PCL Paclitaxel-Loaded Biodegradable Polymeric Nanoparticles: A
Nanoparticles for Intranasal Delivery. Drug Deliv. 2016, 23,
Co-Surfactant Study. Asian. J. Pharm. Sci. 2016, 11, 404–416.
2144–2153. DOI: 10.3109/10717544.2014.948643.
DOI: 10.1016/[Link].2015.09.004.
[29] Zhang, Y.; Huo, M.; Zhou, J.; Zou, A.; Li, W.; Yao, C.; Xie, S.
[36] Tiwari, A.; Mishra, M. K.; Shukla, A.; Yadav, S. K.
DDSolver: An Add-in Program for Modeling and Comparison
of Drug Dissolution Profiles. AAPS J. 2010, 12, 263–271. DOI: Microsponge: An Augmented Drug Delivery System. Am. J.
10.1208/s12248-010-9185-1. Pharm. Tech. Res. 2016, 6, 79–95.
[30] Ghose, A.; Nabi, B.; Rehman, S.; Md, S.; Alhakamy, N. A.; [37] Pandav, S.; Naik, J. Preparation and In Vitro Evaluation of
Ahmad, O. A.; Baboota, S.; Ali, J. Development and Evaluation Ethylcellulose and Polymethacrylate Resins Loaded
of Polymeric Nanosponge Hydrogel for Terbinafine Microparticles Containing Hydrophilic Drug. J. Pharm. 2014,
Hydrochloride: Statistical Optimization. In Vitro and In Vivo 2014, 1–5. DOI: 10.1155/2014/904036.
Studies. Polymers. 2020, 12, 2903. [38] Higuchi, T. Mechanism of Sustained-Action Medication.
[31] Zhou, Y. Z.; Alany, R. G.; Chuang, V.; Wen, J. Optimization of Theoretical Analysis of Rate of Release of Solid Drugs
PLGA Nanoparticles Formulation Containing L-DOPA by Dispersed in Solid Matrices. J. Pharm. Sci. 1963, 52, 1145–1149.
Applying the Central Composite Design. Drug. Dev. Ind. DOI: 10.1002/jps.2600521210.
Pharm. 2013, 39, 321–330. DOI: 10.3109/03639045.2012.681054. [39] Costa, P.; Lobo, J. M. S. Modeling and Comparison of
[32] Kollipara, S.; Bende, G.; Movva, S.; Saha, R. Application of Dissolution Profiles. Eur. J. Pharm. Sci. 2001, 13, 123–133.
Rotatable Central Composite Design in the Preparation and DOI: 10.1016/S0928-0987(01)00095-1.
Optimization of Poly(Lactic-co-glycolic acid) Nanoparticles for [40] Zhou, M.; Shen, L.; Lin, X.; Hong, Y.; Feng, Y. Design and
Controlled Delivery of Paclitaxel . Drug Dev Ind Pharm. 2010,
Pharmaceutical Applications of Porous Particles. RSC Adv.
36, 1377–1387. DOI: 10.3109/03639045.2010.487263.
2017, 7, 39490–39501. DOI: 10.1039/C7RA06829H.
[33] Galindo-Rodriguez, S.; Allemann, E.; Fessi, H.; Doelker, E.
[41] Kanwal, U.; Mukhtar, S.; Waheed, M.; Mahreen, A.; Abbas, N.;
Physicochemical Parameters Associated with Nanoparticle
Formation in the Salting-out, Emulsification-Diffusion, and Shamim, R.; Hussain, K.; Rasool, F.; Hussain, A.; Bukhari, N. I.
Nanoprecipitation Methods. Pharm. Res. 2004, 21, 1428–1439. Fixed Dose Single Tablet Formulation with Differential Release
DOI: 10.1023/B:[Link]. of Amlodipine Besylate and Simvastatin and Its
[34] Meng, F.; Ma, G.; Liu, Y.; Qiu, W.; Su, Z. Microencapsulation Pharmacokinetic Profile: QbD and Risk Assessment Approach.
of Bovine Hemoglobin with High Bio-Activity and High Drug Des. Dev. Ther. 2021, 15, 2193–2210.