You are on page 1of 9

Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis


journal homepage: www.elsevier.com/locate/jpba

Norelgestromin/ethinyl estradiol intravenous infusion formulation


optimization, stability and compatibility testing: A case study to
overcome polysorbate 80 interference in chromatographic analysis
Inas A. Abdallah a,b , Dana C. Hammell a , Hazem E. Hassan a,c, , Audra L. Stinchcomb a,
a
Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland, Baltimore, MD, United States
b
Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Misr International University, Cairo, Egypt
c
Department of Pharmaceutics and Industrial Pharmacy, Faculty of Pharmacy, Helwan University, Cairo, Egypt

a r t i c l e i n f o a b s t r a c t

Article history: Norelgestromin/ethinyl estradiol is a progestin/estrogen combination hormonal contraceptive indicated


Received 5 January 2016 for the prevention of pregnancy in women. The very poor solubility and wettability of these drugs,
Received in revised form 7 March 2016 along with their high potency (adsorption issues), give rise to difculties in designing intravenous (IV)
Accepted 10 March 2016
formulations to assess absolute bioavailability of products containing both drugs. The purpose of this
Available online 19 March 2016
study was to develop an IV formulation, evaluate its stability under different conditions and evaluate its
compatibility with IV sets for potential use in absolute bioavailability studies in humans. Also, a selec-
Keywords:
tive high-performance liquid chromatography (HPLC) method for quantication of ethinyl estradiol and
Norelgestromin
Ethinyl estradiol
norelgestromin in polysorbate 80 matrix was developed and validated.
Polysorbate 80 Norelgestromin/ethinyl estradiol IV solution was prepared using sterile water for injection with 2.5%
Intravenous solution ethanol and 2.5% polysorbate 80 as a cosolvent/surfactant system to obtain a nal drug solution of 25 g
ethinyl estradiol and 252 g norelgestromin from a concentrated stock drug solution. The stabilities of
the concentrated stock and IV solutions were assessed after storing them in the refrigerator (3.7 0.6 C)
and at room temperature (19.5 0.5 C), respectively. Additional studies were conducted to examine the
stability of the IV solution using an Alarias low sorbing IV administration set with and without an inline
lter. The solution was allowed to drip at 1 mL/min over a 60 min period. Samples were obtained at the
beginning, middle and end of the 60 min duration. The chemical stability was evaluated for up to 10
days. Norelgestromin and ethinyl estradiol concentration, purity, and degradant levels were determined
using the HPLC method. The norelgestromin/ethinyl estradiol IV formulation met the chemical stability
criteria when tested on day 1 through day 9 (216 h). Norelgestromin concentrations assayed in stock
and IV solutions were in the range of 90.098.5% and 90.998.8% after 9 days, respectively. As for ethinyl
estradiol, the assayed concentrations were in the range of 91.8100.9% and 92.7100.8% for the stock and
IV solutions, respectively. The administration set was found to be compatible with both drugs; the assayed
concentrations were in the range of 99.2100.3% for norelgestromin and 96.3102.7% for ethinyl estradiol,
but the inline lter showed some adsorption of ethinyl estradiol; where the assayed concentrations were
in the range of 98.199.8% for norelgestromin and 95.997.4% for ethinyl estradiol.
The present study provided evidence supporting the suitability of an intravenous formulation for
norelgestromin/ethinyl estradiol using ethanol/polysorbate 80 as a cosolvent/surfactant system. Both
IV and concentrated stock solutions when stored at room temperature and refrigeration, respectively,
were found to be chemically stable up to 9 days. These results indicated that this formulation is chem-
ically stable and can be used over the time period tested. This IV formulation can be used to evaluate
the absolute bioavailability of products containing norelgestromin and ethinyl estradiol provided that
microbial testing of the IV formulation is performed.
2016 Published by Elsevier B.V.

Corresponding author at: Department of Pharmaceutical Sciences, University of Maryland, School of Pharmacy, 20 N Pine Street, Room: N521, Baltimore, MD 21201,
United States.
Corresponding author at: Department of Pharmaceutical Sciences, University of Maryland, School of Pharmacy, 20 N Pine Street, Room: N525, Baltimore, MD 21201,
United States.
E-mail addresses: hhassan@rx.umaryland.edu (H.E. Hassan), astinchc@rx.umaryland.edu, audra@f6pharma.com (A.L. Stinchcomb).

http://dx.doi.org/10.1016/j.jpba.2016.03.024
0731-7085/ 2016 Published by Elsevier B.V.
146 I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153

1. Introduction purchased from Sigma Aldrich (St. Louis, MO, USA). Norelgestromin
was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA,
Contraception is achieved mainly by inhibiting ovulation USA). Polysorbate 80 (NF) was purchased from Spectrum Chemi-
through the combined activity of two main components: cals MFG Corporation (New Brunswick, NJ, USA). Triethylamine was
estrogen and progestin. Norelgestromin/ethinyl estradiol is a pro- purchased from Fisher Scientic (Fair Lawn, NJ, USA). Sterile water
gestin/estrogen combination hormonal contraceptive indicated for for injection was purchased from Hospira (Lake Forest, IL, USA).
the prevention of pregnancy in women [1,2]. Ethinyl estradiol Intravia sterile empty medication delivery bags (250 mL) were
(EE), 19-Norpregna-1,3,5(10)-trien-20-yne-3,17-diol, (17 ), is a purchased from Baxter Health Corporation (Deereld, IL, USA).
synthetic derivative of natural estrogen estradiol, having high Alaris low sorbing set (107 in.) and IV extension (17 in., 0.2 m l-
estrogen receptor potency and used for treatment of vasomotor ter) (Taxol sterile) were purchased from CareFusion Corporation
symptoms associated with menopause, female hypogonadism, pro- (San Diego, CA, USA). Distilled water was puried by a MilliQ plus
static carcinomapalliative therapy, breast cancer, and oral and water system (Millipore; Billerica, MA, USA).
emergency contraception [3,4]. While, norelgestromin (NE), 18, 19-
dinorpregn-4-en-20-yn-3-one,13-ethyl-17-hydroxyl,3-oxime, (17
2.2. Instruments
) formerly known as 17-deacetyl-norgestimate is a major metabo-
lite of norgestimate and a third generation progestin with little
Liquid chromatography experiments were performed with a
androgenic effect, estrogenic or glucocorticoid activity [5,6].
Waters UPLC-H class system (Milford, MA, USA) equipped with a
Contraceptive transdermal systems containing NE and EE were
PDA detector (Milford, MA, USA) and a uorescence detector (Mil-
developed to reduce the number of unintended pregnancies. Accu-
ford, MA USA). The uorescence detection wavelengths were set at
rate determination of drug release and absorption rate is crucial to
ex/em = 280/310. The UV detection wavelengths were set at 242,
ensure safety of individuals using these transdermal delivery sys-
280 and 254 nm.
tems (TDS). By conducting clinical pharmacokinetic studies with
NE/EE TDS and intravenous administration, absolute bioavailability
2.3. Chromatographic conditions for quantication of
and pharmacokinetics can be calculated [1,7]. Determining abso-
norelgestromin and ethinyl estradiol
lute bioavailability of newly developed NE and EE products (e.g.,
TDS) is crucial to ensure therapeutic levels of both NE and EE are
2.3.1. Chromatographic method development
achieved after using these products.
A Symmetry C18 (4.6 mm 150 mm, 5 m) (Waters ; Milford,
As such, the primary task of this study was to develop an IV for-
MA, USA) HPLC column coupled with Phenomenex Luna secu-
mulation of NE and EE containing therapeutically effective doses
rity C18 guard column (4.0 mm 3.0 mm) (Phenomenex; Torrance,
of the active ingredients, while exhibiting the least possible toxi-
CA, USA) was used. The mobile phase composition was (A): ace-
city for use in absolute bioavailability studies in humans [8]. This
tonitrile and (B): water containing 0.1% v/v triethylamine adjusted
involves solubility testing of NE and EE in compatible vehicles for
to pH = 6.6 (A: 35, B: 65, v/v) used at ow rate of 1.5 mL/min. The
parenteral administration, stability testing under different stor-
relatively low organic phase composition was necessary to avoid
age conditions, compatibility testing in IV infusion sets with and
chromatographic interference induced by polysorbate 80. Column
without in-line lters, and the presence of reliable quantication
and autosampler temperatures were set at 60 C and 25 C, respec-
methods for both drugs to validate the results of these tests. The
tively. An injection volume of 50 L was used.
very poor solubility and wettability of NE and EE along with their
high potency give rise to difculties in designing pharmaceutical
IV formulations. In practice, different methods are often used for 2.3.2. Chromatographic method validation
improving the solubility of slightly soluble or practically insolu- The developed chromatographic method was validated for lin-
ble substances in water. The four most commonly used techniques earity, limit of detection and quantication, specicity, accuracy,
for solubilization of drugs for parenteral use are: adjustment of pH, precision and robustness as per International Conference on Har-
addition of a cosolvent, addition of a surfactant [911], or complex- monization (ICH) guidelines [22].
ation [1215]. In this study we evaluated different combinations of
solvents, cosolvents and surfactants that can be used parenterally 2.3.2.1. Linearity. The linearity was determined at six levels of NE
to solubilize both drugs. and EE over the range 0.150 g/mL. Absorbance of each calibra-
The stability of the nal composition is also of utmost impor- tor was measured and the graph mean absorbance (y-axis) versus
tance since degradation of active ingredients in liquid dosage forms concentration (x-axis) was plotted. Correlation coefcient (r), y-
is much more likely than in other dosage forms (e.g., solid dosage intercept and slope of regression line were estimated.
forms). As such, we have to assess the stability of the IV formulation
and the adequate storage conditions. 2.3.2.2. Specicity. Specicity of the method was ascertained by
There is no report in the primary scientic literature describing a evaluating the presence of interferences at the retention time of
validated HPLC method for simultaneous determination of norelge- NE and EE. The purity of the intact drug was checked using diode
stromin and ethinyl estradiol. Most of the published HPLC methods array detector.
were for the quantication of ethinyl estradiol either with other
estradiol degradation products [16] levonorgestrel [4,1719], des- 2.3.2.3. Accuracy. Accuracy of the method was determined by cal-
ogestrel [20], or drospirenone [2,3,21]. Also, none of the published culating recoveries of NE and EE. The amount of NE and EE was
methods discussed the difculty in analyzing these hormones in estimated by measuring the peak areas and by tting these values
the presence of polysorbate 80 with its signicant spectroscopic to the straight-line equation of calibration curve. Recovery exper-
and chromatographic interference. iments were conducted to determine accuracy of the method. The
accuracy of the assay was evaluated in triplicate at three concen-
2. Materials and methods
tration levels, i.e., 0.3, 3 and 30 g/mL for both compounds.
2.1. Materials
2.3.2.4. Precision.
Ethinyl estradiol USP reference standard, HPLC grade acetoni- 2.3.2.4.1. Repeatability. Standard solutions of both NE and EE
trile, HPLC grade methanol and 200 proof ethanol (USP) were (0.3, 3 and 30 g/mL) were analyzed six times within the same day
I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153 147

(intra-day) and relative standard deviation (%RSD) was calculated 2.6. Physical and chemical stability assessment
for each concentration level.
2.3.2.4.2. Intermediate precision. Variation of results of three Stability studies were conducted by collecting samples up to
different concentrations (0.3, 3 and 30 g/mL) and between days 10 days from both stock solutions which were stored in the refrig-
(inter-day) was analyzed. Inter-day precision was determined by erator and IV solutions were stored at room temperature.
analyzing NE and EE for three times daily for three days.
2.6.1. Stock solution stability testing
Stock solutions (n = 3) were stored at refrigerated conditions. A
2.3.2.5. Limit of detection (LOD) and limit of quantication (LOQ).
2 mL sample was collected from each stock solution throughout the
LOD is dened as the lowest concentration of an analyte reliably
stability study.
differentiated from background levels. LOQ of an individual ana-
lytical procedure is the lowest amount of analyte quantitatively
2.6.2. Intravenous solution stability testing
determined with suitable precision and accuracy. LOD and LOQ
Intravenous solutions (n = 3) were stored at room temperature
were calculated using the following equation as per ICH guidelines.
in the IV Intravia medication delivery bag. A 2 mL sample was
LOD = 3.3 /S; LOQ = 10 /S; Where is the standard deviation
collected from each IV solution throughout the stability study.
of y-intercepts of regression lines and S is the slope of the calibra-
tion curve.
2.6.3. Intravenous medication delivery bags compatibility testing
Additional studies were conducted to examine the IV adminis-
2.3.2.6. Robustness. Robustness testing was done for both EE and tration set (Alaris low sorbing set) with and without the inline
NE by the use of design of experiment (DOE). Different factors were lter. The Intravia medication delivery bag had an IV line and
changed including: column temperature, ow rate, acetonitrile, extension (with and without the in-line lter) attached to it. The
water containing 0.1% v/v triethylamine, and excitation and emis- solution was allowed to drip at 1 mL/min over a 60 min period. Sam-
sion wavelengths. All were changed through a small range using the ples (20 mL) were obtained at the beginning, middle and end of the
Placket-Burman design [23]. A linear design with 8 experiments, 60 min duration before they were analyzed using the HPLC method.
and three center points was reported. All experiments were per-
formed in randomized order to minimize the effects of uncontrolled 3. Results and discussion
factors that may introduce a bias on the response. The coefcients
of the model were estimated by the least squares regression. If an intravenous formulation can be developed, a crossover
absolute bioavailability study in healthy female volunteers using
2.4. Solubility studies either clinically relevant doses or doses that provide quantiable
concentrations enable adequate characterization of the pharma-
Solubility studies were conducted to optimize the formulation cokinetic prole. In clinical practice, an aqueous solution of NE
of NE/EE intravenous solution to determine absolute bioavailabil- and EE is intended for parenteral administration to assess absolute
ity of these drugs. Solubility measurements were determined in bioavailability of these drugs after administering non-parenteral
various solvents/cosolvent systems including normal saline (with products (e.g., solid dosage forms and TDS). Solubilization of poorly
010% ethanol) and sterile water for injection (with 010% ethanol soluble drugs (e.g., NE and EE) has been a very important issue in
and combination of 2.5% ethanol with different percentages of screening studies of new chemical entities as well as in formula-
polysorbate 80 (0.625, 1.25, 2.5 and 3.75%). Amounts of EE and NE tion research. Often, in a certain volume of solvent, an adequate
were weighed in glass vials then 60 mL of each solvent were added. concentration of a drug cannot be achieved due to poor solubility
The samples (n = 3) were shaken at 25 2 C for 24 h, and then l- of the drug which makes development of a IV dosage formulation
tered through 0.2 m lter. The concentrations of the dissolved EE challenging [15]. In practice, different methods are often used for
and NE were analyzed using the developed HPLC method. improving the solubility of slightly soluble or practically insoluble
substances in water [12]. For example, often selected cosolvents
for formulating poorly soluble drugs are ethanol, glycerol, poly
2.5. Intravenous infusion preparation (ethylene glycol) and propylene glycol because of their low toxi-
city. However, because solubilization capacity of these cosolvents
Three IV solutions were individually prepared as follows: in a is not sufcient at non-toxic levels in many cases, many attempts
sterile glass serum bottle, 0.25 mL of EE stock solution (1 mg/mL) have been made to combine their addition with other solubiliza-
(dissolved in ethanol) and 2.52 mL of NE stock solution (1 mg/mL) tion techniques, such as pH adjustment, use of surfactant and use
(dissolved in ethanol) were mixed together. Then 12.23 mL of of cyclodextrin [13]. In this study we evaluated the solubility of
200 proof ethanol were added to the bottle (total volume of NE and EE in normal saline, normal saline with 10% ethanol as a
ethanol = 15 mL) followed by 15 mL of polysorbate 80. The solu- cosolvent, sterile water for injection with varying percentages of
tion was mixed until components were dissolved. 90 mL of sterile ethanol, and sterile water for injection with different percentages
water for injection were added to the bottle to bring the total vol- of ethanol and polysorbate 80.
ume to 120 mL (5 concentrated solution) and then mixed. Twelve
milliliters were transferred into a new sterile glass serum bottle 3.1. Chromatographic method development
using a sterile needle xed with a low retention lter (Acrodisc
Supor 25 mm, 0.2 m) (Pall life Sciences; Ann Arbor, MI USA), then To optimize the chromatographic conditions, Symmetry
48 mL of sterile water for injection, USP were added using another C18 column was used and provided good peak shape and
sterile syringe xed with a low retention lter to have a total vol- retention characteristics for EE and NE. The chromato-
ume of 60 mL. After mixing, the 60 mL were transferred to the sterile graphic conditions initially developed with a mobile phase of
IV medication delivery bag using a sterile syringe. The nal solution [methanol:acetonitrile:water] [50:30:20, v/v] and ow rate of
(5 diluted) contained 252 g of NE, 25 g of EE, 2.5% ethanol and 0.5 mL/min resulted in retention times of 5.8 and 7.5 min for EE
2.5% polysorbate 80. The nal concentration was 0.42 g/mL of EE and NE, respectively. The total run time was 12 min as depicted in
and 4.2 g/mL of NE. Intravenous solutions were stored at room Fig. 1. The 3D-plot chromatogram shown in Fig. 2 for both EE and
temperature. NE shows the irregularities in the baseline due to polysorbate 80
148 I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153

0.04 =242
AU

Norelgestromin
0.02

0.00

0.008

0.006 =280 nm
0.004
AU

Ethinyl estradiol
0.002

0.000

0.030
=254 nm

0.020 Norelgestromin
AU

0.010

0.000

2000.00

1500.00 ex =280 nm, em = 310 nm


EU

1000.00
Ethinyl Estradiol
500.00

0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 10.50 11.00 11.50 12.00

Minutes

Fig. 1. Chromatogram of a 5 g/mL mixture of EE (ex/em = 280/310 and = 280 nm) and NE = 242 and 254 nm. The mobile phase was: methanol (A), acetonitrile (B) and
water (C). Chromatographic method: isocratic elution: 50% A, 20% B and 30% C. Flow rate: 0.5 mL/min and column temperature: 60 C.

0.054
0.052
NE 0.050

EE 0.048
0.046
0.044
0.042
0.040
0.038
0.036
0.034
0.032
0.030
0.028
0.026
0.024
AU

0.022
0.020
0.018
0.016
0.014
0.012
0.010
0.008
0.006
0.004
0.002
0.000
-0.002
-0.004
-0.006
-0.008

250.00

300.00

350.00 nm
400.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00

Minutes

Fig. 2. 3D plot chromatogram of a 5 g/mL mixture of EE and NE solubilized in 2.5% polysorbate 80 (PDA spectrum (210400 nm). The mobile phase was: methanol (A),
acetonitrile (B) and water (C). Chromatographic method: isocratic elution: 50% A, 20% B and 30% C. Flow rate: 0.5 mL/min and column temperature: 60 C.
I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153 149

0.030
=242 nm

0.020 Norelgestromin

0.010
AU
0.000

-0.010

1000.00 ex =280 nm, em = 310 nm

Ethinyl Estradiol
800.00

600.00
EU

400.00

200.00

0.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00
Minutes

Fig. 3. Chromatogram of a 5 g/mL mixture of EE (ex/em = 280/310) and NE ( = 242 nm). The mobile phase was: acetonitrile (A) and water containing 0.1% v/v triethylamine
adjusted to pH = 6.6 (B). Chromatographic method: isocratic elution: 35% A and 65% B. Flow rate: 1.5 mL/min and column temperature: 60 C.

interference, which led to inaccurate, non-repeatable analyses of All experiments were performed in randomized order to mini-
both compounds in a polysorbate 80 containing matrix. Therefore mize the effects of uncontrolled factors that may introduce a bias
further optimization of mobile phase composition was required to on the response. The coefcients of the model were estimated by
minimize polysorbate 80 interference of the EE and NE drug peaks. least squares regression. The equation model for Y was as follows:
Successful chromatographic method development included the
use of a low organic mobile phase followed by high organic wash y = a + bx1 + cx2 + dx3 + ex4 + fx5 [NE]
cycle using 80% acetonitrile throughout the analysis sequence to
ensure no signicant surfactant build-up [24]. This method enabled y = a + bx1 + cx2 + dx3 + ex4 + fx5 + gx6 [EE]
accurate, repeatable analysis of samples containing polysorbate 80.
where, y is either capacity factor or number of theoretical plates to
Low organic mobile phase was optimized for both compounds;
g are the coefcients, and x1 x6 are the six previous factors exam-
isocratic elution was used with a mobile phase composed of ace-
ined in robustness testing. This DOE was completed for robustness
tonitrile:water containing 0.1% v/v triethylamine (pH = 6.6) [35:65;
testing for both EE and NE.
v/v]. Retention times of EE and NE were 9.8 and 13.8 min, respec-
All the factors were found to be non-signicant (p > 0.05) as
tively with a total run time of 20 min as shown in Fig. 3. The 3D-plot
shown on the coefcient plot presented in Fig. 5.
chromatogram shown in Fig. 4 of both EE and NE shows a regular
baseline with no interference that accounts for the accurate and
repeatable results obtained from this low organic mobile phase 3.3. Optimizing concentrations of the excipients
method.
The aim of the solubility testing reported herein was to solu-
bilize both drugs using the lowest concentration of excipients to
3.2. Chromatographic method validation minimize excipient-related toxicity [2528]. NE and EE are poorly
soluble drugs in aqueous solutions. As a result, we tested a num-
Chromatographic method validation was performed as sug- ber of parenterally compatible solvent/cosolvent systems with and
gested by International Conference on Harmonization (ICH) without surfactants to solubilize both drugs. Higher concentra-
guidelines [22]. Various analytical parameters were evaluated; tions of surfactants (e.g., polysorbate 80) are associated with toxic
including linearity, specicity, accuracy, precision, intermediate effects. These toxic effects are mainly due to their interaction with
precision, limit of detection, and quantication, as shown in Table 1. biological membranes and exertion of hemolytic effects. Based
on these reports and after assessing the benet/risk ratio of each
excipient, it was concluded that the rst excipient to be reduced
3.2.1. Robustness should be polysorbate 80, followed by the reduction of the per-
The ranges examined were small deviations from the method centage of ethanol. Indeed, a solvent/cosolvent system composed
settings and corresponding responses as capacity factor or number of 2.5% ethanol and 2.5% polysorbate 80 in sterile water for injec-
of theoretical plates considered (Y) were observed. A linear design tion (WFI) was deemed optimum to prepare an IV solution of NE
with eight experiments, with three center points, was reported and EE.
in Table 2. PlackettBurman design mandates performing eight Fig. 6 depicts the solubility of EE and NE in different solvent
experimental designs under different conditions (column tempera- systems comprised of normal saline (NS), NS with 10% ethanol as
ture, ow rate, mobile phase conditions and wavelength) followed cosolvent, WFI with varying percentages of ethanol (010%) and
by performing three experiments under optimal conditions (indi- sterile WFI with 2.5% ethanol with 0.625%, 1.25%, 2.5% and 3.75%
cated in bold in Table 2). polysorbate 80. The results showed that NS alone and NS with 10%
150 I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153

0.120
0.110
0.100
0.090
0.080
0.070
0.060
0.050

AU
0.040

EE NE 0.030
0.020
0.010
0.000
-0.010
-0.020
-0.030
-0.040

220.00
240.00
260.00
280.00
300.00
320.00 nm
340.00
360.00
380.00
400.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00
Minutes

Fig. 4. 3D plot chromatogram of a 5 g/mL mixture of EE and NE solubilized in 2.5% polysorbate 80 (PDA spectrum (210400 nm). The mobile phase was: acetonitrile
(A), and water containing 0.1% v/v triethylamine adjusted to pH = 6.6 (B). Chromatographic method: isocratic elution: 35% A and 65% B. Flow rate: 1.5 mL/min and column
temperature: 60 C.

Table 1
Summary of validation and regression equation parameters of the chromatographic methods used for determination of ethinyl estradiol and norelgestromin.

EE NE

Linearity:
Regression equation Y = 28,330,000 x 16,186.3 Y = 115,413 x 819.8
Correlation coefcient (r) 0.9998 0.9995
Range 0.150 g/mL 0.150 g/mL
LOD 0.015 g/mL 0.007 g/mL
LOQ 0.046 g/mL 0.022 g/mL
Precision:
Repeatability (Intraday)
(%RSD)*
QCL (0.3 g/mL) 2.05% 2.53%
QCM (3 g/mL) 1.42% 0.74%
QCH (30 g/mL) 1.99% 3.11%
Intermediate precision (Interday)
(% RSD)*
QCL (0.3 g/mL) 3.88% 0.77%
QCM (3 g/mL) 1.04% 0.73%
QCH (30 g/mL) 0.72% 2.29%
Accuracy:
(Mean S.D)*
QCL (0.3 g/mL) 102.63 2.10 101.59 2.57
QCM (3 g/mL) 101.48 1.44 101.69 0.75
QCH (30 g/mL) 98.29 1.96 101.67 3.16
Specicity Specicity of the method was adequate with no interfering peaks
observed at the retention time of either EE or NE

RSD: relative standard deviation.


*
Expressed mean of three replicates.

ethanol as a cosolvent, solubilized 5.9% and 86.6% of EE and 17.1% ethanol were not enough for complete solubilization of EE and
and 67.2% of NE, respectively. While solvent systems composed of NE. Higher percentages of polysorbate 80 (2.5% and 3.75%) were
WFI with varying percentages of ethanol as a cosolvent (0%10%) sufcient for complete solubilization; however, the lower concen-
solubilized 53.8105.7% of EE and 18.3109.9% of NE, respectively. tration of the polysorbate 80 was selected because of potential toxic
Low percentage of polysorbate 80 (0.625% and 1.25%) with 2.5% effects associated with higher polysorbate 80 percentages.
I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153 151

Table 2
Design of experiment for robustness testing of ethinyl estradiol and norelgestromin.

NE EE

Exp. No Column temperature Flow rate Acetonitrile Water Wavelength Excitation wavelength Emission wavelength

1 58 1.3 30 70 243 281 309


2 62 1.7 30 70 241 279 311
3 58 1.3 40 60 241 279 309
4 62 1.7 40 60 243 281 309
5 58 1.3 30 70 243 281 311
6 62 1.7 30 70 241 279 311
7 58 1.3 40 60 241 279 311
8 62 1.7 40 60 243 281 309
911* 60 1.5 35 65 242 280 310
*
Optimal parameters repeated in triplicate.

Fig. 5. Regression coefcient plots obtained from robustness study following Plackett-Burman design using MODDE pro 11 (Umetrics, Sweden) (a) EE and (b) NE.

Better solubility was achieved for both EE and NE with a mean 3.4. Physical and chemical stability assessment
recovery of 99.9% for EE and 102.9% for NE when 2.5% polysorbate
80 was included as a surfactant to decrease the amount of ethanol. Stability studies improve quality in health care effectiveness.
As a result, the optimal composition of cosolvent/surfactant system The chemical integrity and activity of active ingredients in the
for solubilizing EE and NE was deemed to be 2.5% polysorbate 80 preparation must be conrmed. These types of studies are also
and 2.5% ethanol in WFI. used to investigate the formation of precipitates or toxic metabo-
152 I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153

Table 3
Stability of EE and NE during storage in glass bottles (stock solutions) and in medication delivery bags (Intravia , Baxter) (IV solutions).

Time (h) Concentrated Stock solution (EE) IV solution (EE) Concentrated stock solution (NE) IV solution (NE)

(Mean S.D)*

0 100.90 1.64 100.84 2.45 98.49 1.23 98.79 1.92


0.5 100.04 1.94 100.49 1.59 98.54 1.99 98.25 1.88
1 100.21 1.27 100.45 1.85 98.22 1.79 98.21 2.56
2 99.96 1.89 99.69 2.05 98.14 0.49 98.29 2.11
4 99.82 1.76 98.62 2.17 98.29 0.53 98.76 3.69
6 99.64 2.46 98.65 1.46 98.98 2.75 98.58 3.24
8 98.76 1.09 98.81 1.73 98.88 0.79 98.85 2.65
10 98.70 2.01 97.64 2.25 98.59 1.44 97.14 2.29
12 98.30 2.86 95.07 0.50 98.92 2.30 97.68 1.55
24 96.16 2.51 94.97 3.86 98.25 1.74 97.68 1.55
48 94.12 2.73 94.17 4.79 98.07 2.28 97.31 5.41
72 93.49 1.51 94.45 2.65 97.84 1.27 96.68 2.34
96 92.28 1.31 94.32 2.79 94.87 1.09 95.84 3.29
120 92.41 1.61 93.35 1.15 92.25 1.01 95.79 1.20
144 91.44 2.21 93.24 2.64 92.25 3.00 93.83 4.76
168 91.98 1.82 93.14 2.72 92.13 2.98 93.79 2.50
192 91.68 1.43 91.33 4.74 91.63 2.45 91.60 3.72
216 91.81 2.96 92.72 2.61 89.95 2.39 90.99 2.51
240 84.24 1.43 86.84 2.18 83.41 4.63 90.01 2.89
*
Expressed as percentage of original concentration remaining (n = 3) (Mean SD).

(a) So, all mixtures (n = 3) were physically compatible during the time
WFI + 2.5% ETOH + 3.75% PS - 80 of storage.
WFI + 2.5% ETOH + 2.5% PS - 80 During optimization of the chromatographic method, high vari-
WFI + 2.5% ETOH + 1.25% PS - 80 ability and error percentage were noticed in the recovery of both
WFI + 2.5% ETOH + 0.625% PS - 80
WFI + 10% ETOH
drugs. This variability was minimized by using siliconized Eppen-
WFI + 5% ETOH dorf microcentrifuge tubes and silanized glass vials for HPLC and
WFI + 4% ETOH low retention pipette tips. This proves both EE and NE are highly
WFI + 3% ETOH
WFI + 2% ETOH
adsorbed into plastic vials and Eppendorf microcentrifuge tubes.
WFI + 1% ETOH Therefore the selection of medication delivery bag and admin-
WFI + 0% ETOH istration set was very crucial due to the adsorption and binding
NS + 10% ETOH
problems encountered during optimization of sample preparation
NS
and chromatographic analysis. The binding issues necessitated the
0 20 40 60 80 100 120 140
use of the Intravia medication delivery bag for the EE and NE
Ethinyl Estradiol
% Recovery intravenous solution, in order to make sure the entire dose was in
(b) solution and none was adsorbed into the delivery bag for accurate
WFI + 2.5% ETOH + 3.75% PS - 80 dose delivery.
WFI + 2.5% ETOH + 2.5% PS - 80 Table 3 represents the stability data of EE and NE during stor-
WFI + 2.5% ETOH + 1.25% PS - 80 age in glass bottles (concentrated stock solutions) and in sterile
WFI + 2.5% ETOH + 0.625% PS - 80
WFI + 10% ETOH
medication delivery bags (Intravia , Baxter) (IV solution). As indi-
WFI + 5% ETOH cated in Table 3 the remaining concentrations of NE and EE were
WFI + 4% ETOH 90% after 9 days (216 h) of storage for either the IV solution in
WFI + 3% ETOH the Intravia medication delivery bag or the concentrated stock
WFI + 2% ETOH
solutions in sterile glass vials.
WFI + 1% ETOH
WFI + 0% ETOH Additional studies were conducted to examine the IV admin-
NS + 10% ETOH istration set with and without the inline lter. The solution was
NS allowed to drip at 1 mL/min over a 60 min period in the pres-
0 20 40 60 80 100 120 ence and absence of the inline lter. Samples were collected and
Norelgestromin analyzed in both cases from the beginning (020 min), middle
% Recovery (2040 min) and end (4060 min) of the 60 min period. The admin-
istration set was found to be compatible with both drugs, but the
Fig. 6. Solubility of EE and NE in different solvent systems. Data represents
mean S.D (n = 3). Normal saline (NS), sterile water for injection (WFI), ethanol
inline lter showed some adsorption of EE, as shown in Table 4.
(ETOH) and polysorbate 80 (PS80).
4. Conclusion

An IV solution of NE and EE was developed using 2.5% ethanol


lites/degradants, and prevent their formation and hence ensure and 2.5% polysorbate 80 as a cosolvent/surfactant system. The sta-
patient safety [2934]. bility studies indicated that at room temperature the IV solution
During the stock solution and intravenous solution stability was chemically stable up to 9 days (216 h) when stored in Intravia
studies, room and refrigeration temperatures were 19.5 0.5 C medication delivery bags. In addition, stability studies indicated
and 3.7 0.6 C, respectively. At the end of both studies, the that both drugs were compatible with the Alarias low sorbing IV
solutions were visually inspected; none of the solutions showed administration set (i.e., stable with minimal drug adsorption).
changes in color, NE or EE precipitation, or measurable losses of A new chromatographic method was developed and vali-
volume caused by evaporation at the different storage conditions. dated for the analysis of both compounds in the presence of
I.A. Abdallah et al. / Journal of Pharmaceutical and Biomedical Analysis 125 (2016) 145153 153

Table 4
Drug adsorbance testing of the intravenous administration set.

No inline lter Inline lter

EE NE EE NE

(Mean S.D) *

Beginning 102.74 4.70 100.33 6.71 96.99 2.94 99.07 0.002


Middle 100.03 4.15 100.94 4.04 95.94 0.83 99.82 0.004
End 96.34 3.26 99.20 5.28 97.39 1.69 98.09 4.02

Beginning = 020 min; Middle = 2040 min; End = 4060 min.


*
Expressed as percentage of original concentration remaining (n = 3) (Mean SD).

polysorbate 80, which causes signicant interference during chro- [12] P. Chaudhari, P. Sharma, N. Barhate, P. Kulkarni, C. Mistry, Solubility
matographic analysis. This IV formulation can be used for future enhancement of hydrophobic drugs using synergistically interacting
cyclodextrins and cosolvent, Curr. Sci. 92 (2007) 15861591.
clinical bioavailability studies to determine absolute bioavailabil- [13] K. Kawakami, N. Oda, K. Miyoshi, T. Funaki, Y. Ida, Solubilization behavior of a
ity of other pharmaceutical formulations containing NE and EE (e.g., poorly soluble drug under combined use of surfactants and cosolvents, Eur. J.
contraceptive transdermal systems). Pharm. Sci. 28 (2006) 714.
[14] D. Guo, J.P. Cain, S. OConnell, E.R. Gardner, S. Pisle, W.D. Figg, et al.,
Preformulation study of NSC-726796, AAPS PharmSciTech 13 (2012) 661673.
Acknowledgements [15] R.G. Strickley, Solubilizing excipients in oral and injectable formulations,
Pharm. Res. 21 (2004) 201230.
[16] L. Havlikora, L. Novakova, L. Matysora, J. Sicha, P. Solich, Determination of
Funding for this project was made possible, in part, by the estradiol and its degradation products by liquid chromatography, J.
Food and Drug Administration through grant U01FD004955. Views Chromatogr. A. 1119 (2006) 216223.
expressed in this manuscript do not reect ofcial policies of the [17] A. Ravindra, P. Hima, K.N. Swamy, K.V. Kumar, Validated RP-HPLC method for
simultaneous estimation of levonorgestrel and ethinylestradiol in combined
Department of Health and Human Services; nor does any men-
dosage form, J. Sci. Inn. Res. 2 (2013) 642650.
tion of trade names, commercial practices, or organization imply [18] M. Ines, R.M. Santoro, N.M. Kassab, M. Hasegawa, E.R. Kedor-Hackmann,
endorsement by the United States Government. Determination of steriod hormones in oral contraceptives by high
performance liquid chromatography, Drug Dev. Ind. Pharm. 28 (2002)
The authors would like to thank Dr. Sam G. Raney and Dr.
741747.
Priyanka Ghosh for their valuable discussions. [19] A.M. Bond, I.D. Heritage, M.H. Briggs, Reversed-phase high performance liquid
chromatography for the determination of steriod hormones in oral
contraceptives, J. Chromatogr. A. 315 (1984) 313320.
References
[20] M. Sarat, C. Rambabu, A validated simultaneous RP-HPLC method for
determination of desogestrel and ethinyl estradiol tablets, Int. J. Pharm.
[1] L.S. Abrams, D.M. Skee, J. Natarajan, F.A. Wong, G.D. Anderson, Pharm. Sci. 4 (2012) 115119.
Pharmacokinetics of a contraceptive patch (Ortho Evra) containing [21] R.C. Patel, D.K. Rathod, K.S. Rajesh, V.S. Patel, RP-HPLC method development
norelgestromin and ethinyloestradiol at four application sites, Br. J. Clin. and validation for estimation of drospirenone and ethinyl estradiol in bulk
Pharmacol. 53 (2002) 141146. and combined dosage form, Pharmagene 1 (2013) 1520.
[2] V.B. Silva, A. Arturo, G. Galdos, C. Maria, A. Mothe, B. Pallastrelli, et al., [22] ICH Validation of analytical procedures test and methodology Q2 (R1),
Simultaneous determination of ethinyl estradiol and drospirenone in oral International Conference on Harmonization, IFPMA, Geneve, 2005.
contraceptive by high performance liquid chromatography, Braz. J. Pharm. Sci [23] J. Burman, R.L. Plackett, The design of optimum multifactorial experiments,
49 (2013) 521528. Biometrika 33 (1946) 305325.
[3] S. Warkad, B. Santhakumari, A.V. Chandewar, Development and validation of [24] W.P. Wuelng, K. Kosuda, A.C. Templeton, A. Harman, M.D. Mowery, R.A.
a simple and sensitive RP-HPLC method for simultaneous estimation of Reed, Polysorbate 80 UV/vis spectral and chromatographic
drospirenone and ethinylestradiol in combined tablet dosage form, Int. J. characteristicsdening boundary conditions for use of the surfactant in
Pharm. Pharm. Sci. 4 (2012) 452457. dissolution analysis, J. Pharm. Biomed Anal. 41 (2006) 774782.
[4] F. Ahmadi, R. Ghordoie-milan, Determination of levonorgestrel and ethinyl [25] G.H.W. Ard, Formulation-related problems associated with intravenous drug
estradiol in pharmaceutical formulations by H-point standard addition delivery, J. Pharm. Sci. 87 (1998) 787796.
method in nonaqueous solvent using simultaneous addition of both analytes, [26] M.J. Taylor, S. Tanna, T. Sahota, In vivo study of a polymeric glucose-sensitive
J. Rep. Pharm. Sci. 1 (2012) 718. insulin delivery system using a rat model, J. Pharm. Sci. 99 (2010) 42154227.
[5] X. Liu, C. Ding, Q. Ge, X. Zhi, Z. Zhou, Determination of norelgestromin in [27] J.F. Krzyzaniak, F.A. Alvarez Nnez, D.M. Raymond, S.H. Yalkowsky, Lysis of
rabbit plasma by LCMS/MS and its application to the pharmacokinetic study human red blood cells. Comparison of in vitro and in vivo hemolysis data, J.
of ORTHO EVRA, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 877 (2009) Pharm. Sci. 86 (1997) 12151217.
311315. [28] J.F. Krzyzaniak, S.H. Yalkowsky, Lysis of human red blood cells. Effect of
[6] F.A. Wong, R.W. Edom, M. Duda, J.P. Tischio, M. Huang, S. Juzwin, et al., contact time on surfactant-induced hemolysis, Int. J. Pharm. Sci. Technol. 52
Determination of norgestimate and its metabolites in human serum using (1997) 6669.
high-performance liquid chromatography with tandem mass spectrometric [29] B. Benaj, T. Dine, M. Luyckx, B. Gressier, C. Brunet, F. Goudaliez, Stability and
detection, J. Chromatogr. B Biomed. Sci. Appl. 734 (1999) 247256. compatibility studies of zorubicin in intravenous uids and PVC infusion bags,
[7] L.S. Abrams, D. Skee, J. Natarajan, F.A. Wong, Pharmacokinetic overview of J. Pharm. Biomed. Anal. 14 (1996) 695705.
Ortho Evra/Evra, Fertil. Steril. 77 (2002) S3S12. [30] R.R. Carr, D. Decarie, M.H.H. Ensom, Stability of epinephrine at standard
[8] K. Kovcs, I. Antal, G. Stampf, I. Klebovich, K. Ludnyi, Composition concentrations, Can. J. Hosp. Pharm. 67 (2014) 197202.
optimization and stability testing of a parenteral antifungal solution based on [31] J. Kaiser, I. Kramer, Long-term stability study of clofarabine injection
a ternary solvent system, AAPS PharmSciTech 11 (2010) 285293. concentrate and diluted clofarabine infusion solutions, J. Oncol. Pharm. Pract.
[9] B. Liu, W.P. Gordon, W. Richmond, T. Groessl, T. Tuntland, Use of solubilizers 18 (2012) 213221.
in preclinical formulations: effect of Cremophor EL on the pharmacokinetic [32] P. Martel, I. Petit, F. Pinguet, S. Poujol, C. Astre, M. Fabbro, Long-term stability
properties on early discovery compounds, Eur. J. Pharm. Sci. 15 (2015) of 5-uorouracil stored in PVC bags and in ambulatory pump reservoirs, J.
928987. Pharm. Biomed. Anal. 14 (1996) 395399.
[10] P.V. Hoogevest, X. Liu, A. Fahr, Drug delivery strategies for poorly [33] S.S. Murthy, H.G. Brittain, Stability of revex nalmefene hydrochloride
water-soluble drugs: the industrial perspective, Expert Opin. Drug Deliv. 8 injection, in injectable solutions, J. Pharm. Biomed. Anal. 15 (1996) 221226.
(2011) 14811500. [34] J.F. Carpenter, M.A. Mcnulty, L.J. Dusci, K.F. Ilet, Stability of omeprazole
[11] P.P. Constantinides, A. Tustian, D.R. Kessler, Tocol emulsions for drug sodium and pantoprazole sodium diluted for intravenous infusion, J. Pharm.
solubilization and parenteral delivery, Adv. Drug Deliv. Rev. 56 (2004) Technol. 22 (2006) 9598.
12431255.