Chapter-4

Analysis of Duloxetine hydrochloride and its related compounds in Pharmaceutical dosage forms and in-Vitro Dissolution studies by stability indicating UPLC

___________________________________________

4.1 Introduction of Duloxetine hydrochloride and survey of

analytical methods

Duloxetine hydrochloride is indicated for the treatment of major

depressive disorder (MDD). Its chemical designation is (+)-(S)-N-methyl-

β-(1-naphthyloxy)2-thiophenepropylamine hydrochloride. It is available

as 20 mg, 30 mg and 60 mg as capsules. It is a white to slightly

brownish white solid, which is slightly soluble in water. The molecular

formula of Duloxetine hydrochloride is C18H19NOS•HCl and molecular

weight of Duloxetine hydrochloride is 333.88

Fig: 4.1.F1: Chemical structure of Duloxetine hydrochloride

Molecular formula: C18H19NOS•HCl, Molecular weight : 333.88

Duloxetine is indicated for the acute and maintenance treatment of

major depressive disorder (MDD), as well as acute management of

generalized anxiety disorder. Also used for the management of

neuropathic pain associated with diabetic peripheral neuropathy, and

fibromyalgia. It has been used in the management of moderate to severe

stress urinary incontinence (SUI) in women. Duloxetine lacks affinity for

monoamine receptors within the central nervous system. It is a potent

inhibitor of neuronal serotonin and norepinephrine reuptake and a less

potent inhibitor of dopamine reuptake. The mechanism of action of

Duloxetine in SUI has not been determined, but is thought to be

associated with the potentiation of serotonin and norepinephrine activity

in the spinal cord, which increases urethral closure forces and thereby

reduces involuntary urine loss.

The aim of present study was to develop a new analytical method for

the determination of related substances and for quantitative estimation

of Duloxetine hydrochloride along with in vitro dissolution studies in

pharmaceutical dosage forms by liquid chromatography LC to establish

inherent stability of Duloxetine through stress studies under a variety of

ICH recommended test conditions. In the present study, the additional

benefit of in vitro dissolution studies is, Duloxetine degrades rapidly in

acidic pH conditions. As per regulatory requirement stability indicating

LC methods are required for acidic dissolution mediums. Hence

developed UPLC is useful for estimation of drug release in presence of

degradation impurities. Duloxetine has been determined in

pharmaceutical preparations by LC [1, 2], by spectrophotometric [3] and

by LCMS [4, 5, 6] methods. No liquid chromatographic method was

reported in major pharmacopeia’s like USP, EP, Japanese Pharmacopeia

and British Pharmacopeia. Literature survey reveals that no stability-

indicating liquid chromatographic method was reported for

determination of related substances and for quantitative estimation of

Duloxetine hydrochloride along with in vitro dissolution studies in

pharmaceutical dosage forms.

Hence a reproducible stability-indicating reverse phase UPLC method

was developed for the quantitative determination of Duloxetine

hydrochloride and its three impurities. The method was also applied for

study of in-vitro dissolution profiles in pharmaceutical dosage forms.

4.2. Development of a new analytical method for the determination

of related components and assay of Duloxetine hydrochloride in

pharmaceutical dosage forms by UPLC

4.2.1 Materials

Standards of Duloxetine Hydrochloride, Impurity-1, Impurity-2 and

Impurity-3, and Duloxetine 60 mg capsules were supplied by Dr. Reddy’s

laboratories limited, Hyderabad, India. The HPLC grade acetonitrile,

methanol, tertahydrofuran and analytical grade KH2PO4 and ortho

phosphoric acid were purchased from Merck, Darmstadt, Germany.

Water was prepared by using Millipore MilliQ Plus water purification

system.

4.2.2 Equipment

Acquity UPLCTM system (Waters, Milford, USA) we used consists of a

binary solvent manager, a sample manager and a photodiode array (PDA)

detector. The output signal was monitored and processed using

empower2 software. Cintex digital water bath was used for hydrolysis

studies. Photo stability studies were carried out in a photo stability

chamber (Sanyo, Leicestershire, UK). Thermal stability studies were

performed in a dry air oven (Cintex, Mumbai, India).

4.2.3 Chromatographic Conditions

LC was carried out on a Waters Aquity UPLC with photodiode array

detector. The output signal was monitored and processed using

empowers software. The chromatographic column used was Zorbax XDB

C-18, 50 mm x 4.6 mm i.d with 1.8 µm particles. The separation was

achieved on a gradient method. The mobile phase A contains a mixture of

0.01 M KH2PO4 (pH 4.0) buffer, tetrahydro furan and methanol in the

ratio 67: 23: 10 (v/v/v); respectively and the mobile phase B contains a

mixture of 0.01 M KH2PO4, (pH 4.0) buffer and acetonitrile in the ratio

60: 40 (v/v); respectively.

The flow rate of mobile phase was 0.6 mL min-1. The UPLC gradient

program was set as: time (min) / % solution B: 0/0, 6.0/0, 8.0/100,

13.0/100, 14.0/0 and 16.0/0. The column temperature was maintained at 40 ºC and the detection was monitored at a wavelength 236 nm. The

injection volume was 5 µL.

Fig: 4.2.F1: Chemical structures of Duloxetine hydrochloride

impurities

Impurity-1

Molecular formula: C18H19NOS•HCl

Molecular weight : 333.88

Impurity-2

Molecular formula: C18H19NOS•HCl

Molecular weight : 333.88

Impurity-3

Molecular formula: C10H8O

Molecular weight : 144.17

4.2.4 Sample preparation

A stock solution of Duloxetine hydrochloride (2.0 mg mL-1) was

prepared by dissolving an appropriate amount of drug in diluent (methanol: water 90:10 (v/v); respectively). Working solutions containing

200 and 20 μg mL-1 were prepared from this stock solution for

determination of related substances and for assay determination,

respectively. A mixed stock solution (0.5 mg mL-1) of the impurities (denoted Impurity-1 to Impurity-3) was also prepared in methanol.

4.2.5 Preparation of tablets sample solution

Duloxetine hydrochloride is a capsule conatins pellets containing 60 mg of Duloxetine. Twenty capsules of Duloxetine (60 mg) were

individually weighed to get the average weight of the pellets. An amount

of pellets equivalent to 20 mg of Duloxetine was transferred to 100 mL volumetric flask, added 70 mL of methanol: water 90:10, v/v (diluent)

sonicated for 20 min and diluted to 100 mL (200 µg mL-1). The resulting

solution was centrifuged at 3,000 RPM for 15 min (supernatant solution was used for purity evaluation). 10 mL of supernatant solution was

transferred and diluted to 100 mL with diluent (20 µg mL-1). This

solution was filtered using 0.45 µm nylon 66-membrane filter and used

for the assay analysis.

4.2.6 Generation of stress samples

One lot of Duloxetine hydrochloride capsules (60 mg) were selected for

stress testing. From the ICH Stability guideline: “stress testing has to be

carried out on a single lot of material. Different kinds of forced

degradation conditions (i.e., heat, humidity, acid, base, oxidative and

light) were employed on one lot of Duloxetine capsules based on the

guidance available from ICH Stability Guideline (Q1AR2). The details of

the stress conditions used are as follows:

a) Acid hydrolysis:

Drug solution in 1N HCl was exposed at 60°C for 3 h.

b) Base hydrolysis:

Drug solution in 1N NaOH was exposed at 60°C for 3 h

c) Oxidative stress:

Drug solution in 6% H2O2 was exposed at 30°C temperature for 3 h.

d) Water hydrolysis:

Drug solution in water was exposed at 60°C for 3 h.

e) Thermal stress:

Pellets were subjected to dry heat at 60°C for 10 days.

f) Photolytic degradation:

The photo degradation was carried out by exposing the Duloxetine

hydrochloride capsules pellets in solid state to light providing an overall

illumination of not less than 1.2 million lux hours and an integrated near

ultraviolet energy of not less than 200 W h/m2, which took about 10

days period in photo stability chamber.

4.3 Method development and optimization of chromatographic

conditions

Forced degradation studies were performed to develop a stability

indicating UPLC method for the quantitative determination and purity

evaluation of Duloxetine hydrochloride capsules. Stressed samples

obtained during forced degradation studies including the samples of

impurity-1, impurity-2 and impurity-3 were utilized in the UPLC method

development.

4.3.1 Selection of wavelength

Impurities and Duloxetine hydrochloride solutions were prepared in

diluents and scanned in UPLC PDA detector; all the impurities and

Duloxetine are having UV maxima at about 236 nm (Fig: 4.3.F1). Hence

detection at 236 nm was selected for method development purpose.

Fig: 4.3.F1: Typical UV spectrums of Duloxetine and its impurities.

4.3.2 Column Selection

The main target of the chromatographic method was to get the

separation among impurities namely impurity-1, impurity-2 and

impurity-3 and degradation products (mainly at 0.87 RRT) from

Duloxetine peak and elution of Duloxetine symmetrical peak. For initial

screening trials the mobile phase used was a mixture of 0.01 M KH2PO4

(pH 4.0) buffer, tetrahydro furan and methanol in the ratio 67: 23: 10

(v/v/v); respectively, the flow rate was 0.6 mL min-1. Inertsil ODS-3,

50 mm × 2.1 mm, 2μm particles, Zorbax XDB C-18, 50 mm x 4.6 mm i.d

with 1.8 µm particles and Waters Aquity BEH C18, 50 x 2.1 mm; 1.7μm

columns were used with isocratic elution.

Ab

s

or

b

a

nc

e

Overlay spectra of Duluxetine

and its impurities

For Inertsil ODS-3, 50 mm × 2.1 mm, 2μm particles column,

impurity-2 was co-eluted with Duloxetine peak and also impurity-3 was

eluted as a broad peak. For Waters Aquity BEH C18, 50 x 2.1 mm; 1.7

μm column, impurity-2 was co-eluted with Duloxetine peak and the

resolution between Duloxetine and degradation product at 0.87 RRT was

poor. For Zorbax XDB C-18, 50 mm x 4.6 mm with 1.8 µm particles

column, impurity -1, degradation product at 0.87 RRT, impurity-2 and

impurity-3 were well separated from Duloxetine peak, but impurity-3

was strongly retained (RT was at about 29 min). Hence Zorbax XDB C-

18, 50 mm x 4.6 mm with 1.8 µm particles column was chosen for

further method development.

Table 4.3.T1: Experimental results on different stationary phases

Column Parameter Imp-

1 DUL*

Imp-

2

Imp-

3 Remarks

Inertsil ODS-3,

50 mm × 2.1 mm,

2μm particles

Retention

time 2.53 4.03 - 10.93

Imp-2 and

unknown (0.87

RRT) peeks were co-

eluting with

Duloxetine peak

USP Tailing 1.5 1.7 - 0.9

Zorbax XDB C-18,

50 mm × 4.6 mm,

containing 1.8μm particles

Retention time

2.56 3.83 5.00 28.6 Imp-1,

unknown (0.87

RRT), Imp-2

and Duloxetine peaks are well

separated from

each other. But Imp-3

USP Tailing 1.1 1.7 1.3 1.1

retention time

was about 29

min

Waters Aquity BEH C18, 50 x

2.1 mm; 1.7μm

column

Retention

time 1.42 2.26 - 9.39

Imp-2 peak was

co-eluting with

Duloxetine and unknown peak

at RRT 0.87 is

partially merged

with Duloxetine peak

USP Tailing 0.9 1.8 - 1.6

*Duloxetine

4.3.3 Effect of pH

To get better separation of impurities namely impurity-1, impurity-2,

impurity-3 and degradation products (mainly the unknown impurity at

0.87 RRT) from Duloxetine and to elute the impurities and Duloxetine as symmetrical peaks, different pH buffers from pH range 4.0 to 7.0 (pH 4.0

and 7.0) was studied. For pH 4.0 study, the mobile phase was “0.01 M

KH2PO4 buffer, tetra hydro furan and methanol in the ratio 67: 23: 10 (v/v/v), pH adjusted to 4.0 with ortho phosphoric acid”. For pH 7.0

study, the mobile phase was “0.01 M ammonium acetate buffer, tetra

hydro furan and methanol in the ratio 67: 23: 10 (v/v/v), pH adjusted to

7.0 with aqueous ammonia solution”. For above all pH study column used was Zorbax XDB C-18, 50 mm × 4.6 mm, 1.8 μm and flow rate was

0.6 mL min-1

At pH 7.0, Peak distortion of Duloxetine peak was noticed and co-

elution of impurity-2 with Duloxetine was observed and also impurity-3

was strogly retained (RT was about 19 min) in the column. At pH 4.0,

impurity-1, unknown (0.87 RRT), impurity-2 and Duloxetine peaks are

well separated from each other. But impurity-3 was strongly retained (RT

was about 29 min). Since impurity-3 was strongly retained when by pH

decreased from 7.0 to 4.0, no further reduction in the pH was studied.

Hence pH 4.0 was selected for method pH. To elute impurity-3 faster,

elution mode was switched to gradient elution mode from isocratic mode.

4.3.4 Effect of Column Temperature

The effect of column oven temperature was studied from 25° C to

40°C, the symmetry of Duloxetine peak was 1.8 at 25°C. When

temperature was increased to 40°C, the symmetry of Duloxetine peak

was improved to 1.4. Hence column temperature was chosen at 40°C.

4.3.5 Effect of Organic solvent

When “mixture 0.01 M KH2PO4 buffer, tetra hydro furan and

methanol in the ratio 67: 23: 10 (v/v/v), pH adjusted to 4.0 with ortho

phosphoric acid” was used as mobile phase (solvent A) in isocratic

elution, the compound impurity-3 was strongly retained in the column.

To elute impurity-3 faster, elution mode was changed to gradient mode

with solvent B (mixture of 0.01 M KH2PO4, (pH 4.0) buffer and

acetonitrile in the ratio 60: 40 (v/v); respectively).

When acetonitrile was used as a solvent in solvent A (buffer: tetra

hydro furan: acetonitrile: 67: 23: 10, v/v/v) instead of methanol, the

resolution between Duloxetine and impurity-2 and the resolution

between unknown impurity at RRT 0.87 and Duloxetine was poor.

Ratio’s of methanol and tetrahydrofuran in solvent A were optimized to

achieve good resolution between Duloxetine and impurity-2, and between

unknown impurity (at RRT 0.87) and Duloxetine peaks.

When methanol was used as a solvent in solvent B (buffer: methanol:

60:40, v/v) instead of acetonitrile, the retention time of impurity-3 was

shifted towards higher side (13 min) due to high interactions of impurity-

3 with column (lipophilicity). To improve the retention time of impurity-3,

the strength of the solvent was increased by replacing methanol with

acetonitrile. Different mixtures of buffer and acetonitrile were applied to

improve the retention time and symmetry of impurity-3. When 40%

acetonitrile was used as a solvent in solvent B retention time of impurity-

3 was improved (9.8 min) and symmetry of impurity-3 was also

improved.

Fig. 4.3.F1: Typical chromatograms from method development trials

4.3.6 Optimized chromatographic conditions for the determination

of related substances and assay of Duloxetine

Column : Zorbax XDB C-18, 50 mm × 4.6 mm,

containing 1.8 μm particles.

Elution : Gradient.

Mobile phase-A : Buffer, tetra hydro furan and methanol in the

ratio 67: 23: 10 (v/v/v).

Mobile phase-B : Buffer and acetonitrile in the ratio 60: 40 (v/v)

Flow rate : 0.6 mL min-1

Column temperature : 40°C

Wavelength of detection : 236 nm

Injection volume : 5 L

Run time : 16 min

Diluent : methanol: water 90:10 (v/v)

Gradient Program : Time (min)/ % solution B: 0.01/0, 6.0/0,

8.0/100, 13.0 /100, 14.0/0 and 16.0/0

Buffer : 0.01 M KH2PO4 buffer pH adjusted to 4.0 with

orthophosporic acid Retention time : Duloxetine about 4.15 min

Relative Retention Time : Impurity-1 - about 0.65

Impurity-2 - about 1.31

Impurity-3 - about 2.36

The interference from excipients gelatin, hypromellose, hydroxypropyl

methylcellulose acetate succinate, sodium lauryl sulfate, sucrose, sugar

spheres, talc, titanium dioxide, and triethyl citrate was also checked by injecting sample solutions of excipients. There was no interference found

at retention times of impurities (impurity-1, impurity-2 and impurity-3)

and Duloxetine peak.

4.4 Degradation behavior

UPLC studies on Duloxetine under different stress conditions

suggested the following degradation behavior.

4.4.1 Degradation in Acidic solution

Duloxetine underwent degradation in 1N HCl on heating at 60°C for 3

h. Major degradation was observed with formation of impurity-1,

impurity-3 and unknown impurity at RRT 0.87 (3.392 RT) (Fig: 4.4.F1(a) to Fig: 4.4.F1(b)).

Fig: 4.4.F1 (a): Typical UPLC chromatogram of acid hydrolysis

Fig: 4.4.F1 (b): Peak purity plot of Duloxetine under Acid

hydrolysis

Purity Angle Purity

Threshold Purity Flag Peak Purity

0.072 0.271 No Pass

4.4.2 Degradation in Basic solution:

When Duloxetine HCl pellets were exposed to 1 N NaOH at 60°C for 3 h, minor degradation was observed with formation of impurity-3 (Fig:

4.4.F2 (a) to Fig: 4.4.F2 (b)).

Fig: 4.4.F2 (a): Typical UPLC chromatogram of alkali hydrolysis

Fig: 4.4.F2 (b): Peak purity plot of Duloxetine under alkali

hydrolysis

Purity Angle Purity

Threshold Purity Flag Peak Purity

0.182 0.322 No Pass

4.4.3 Oxidative conditions

Duloxetine HCl pellets were exposed to 6% Hydrogen peroxide at 30

ºC for 3 h. Duloxetine has shown no significant sensitivity towards the

treatment of Hydrogen peroxide (Fig: 4.4.F3 (a) to Fig: 4.4.F3 (b)).

Fig: 4.4.F3 (a): Typical UPLC chromatogram of oxidative degradation

Fig: 4.4.F3 (b): Peak purity plot of Duloxetine under oxidative

degradation

Purity Angle Purity

Threshold Purity Flag Peak purity

0.160 0.341 No Pass

4.4.4 Degradation in Neutral (Water) solution

When Duloxetine HCl pellets were exposed to water at 60°C for 3 h,

major degradation products were observed with the formation of

impurity-1, impurity-3 and unknown impurity at RRT 0.87 (3.177 RT)

(Fig: 4.4.F4 (a) to Fig: 4.4.F4 (b)).

Fig: 4.4.F4 (a): Typical UPLC chromatogram of Water hydrolysis

Fig: 4.4.F4 (b): Peak purity plot of Duloxetine under water hydrolysis

Purity Angle Purity

Threshold Purity Flag Peak purity

0.198 0.274 No Pass

4.4.5 Photolytic conditions

When the drug pellets was exposed to light for an overall illumination

of 1.2 million lux hours and an integrated near ultraviolet energy of 200-

watt hours/square meter (w/mhr), no degradation of the drug was

observed (Fig: 4.4.F5 (a) to Fig: 4.4.F5 (b)).

Fig: 4.4.F5 (a): Typical UPLC chromatogram of Photo degradation

Fig: 4.4.F5 (b): Peak purity plot of Duloxetine under photo

degradation

Purity Angle Purity

Threshold Purity Flag Peak purity

0.112 0.251 No Pass

4.4.6 Thermal Degradation

Duloxetine HCl is stable for heat when drug pellets were exposed to

dry heat at 60°C for 10 days (Fig: 4.4.F6 (a) to Fig: 4.4.F6 (b)).

Fig: 4.4.F6 (a): Typical UPLC chromatogram of thermal degradation

Fig: 4.4.F6 (b): Peak purity of Duloxetine under thermal degradation

The unknown impurity (0.87 RRT) which was formed in acid

hydrolysis and aqueous hydrolysis was investigated with LCMS study.

The mass of unknown impurity was 298.4 and the mass of Imp-2 (4-(3-

methylamino)-1-(thiophen-2-yl) propyl) naphthalin-1-ol) was 298 and the UV spectra of Impurity-2 was similar to that of unknown impurity.

Hence, based LCMS data and UV spectra, the possible chemical name of

the unknown impurity could be ‘2-(3-methylamino)-1-(thiophen-2-yl) propyl) naphthalin-1-ol. The proposed structure of unknown impurity

was positional isomer of Impurity-2.

Fig: 4.4.F7: Typical UV spectrum of unknown impurity at RRT 0.87

Purity Angle Purity

Threshold Purity Flag Peal purity

0.112 0.250 No Pass

Fig: 4.4.F8: LC-Mass spectrum of unknown impurity at RRT 0.87

Peak purity test performed for the Duloxetine peak using photodiode

array (PDA) detector, data confirmed the purity of the peak for all the

stressed samples. Assay of all stressed samples were calculated using

qualified standard of Duloxetine Hydrochloride. Considering the purities

from the respective chromatograms of stressed samples, mass balance

(%assay + % degradents + % impurities) was calculated for each stress

sample. The mass balance of stressed samples was more than 99.1%.

This clearly demonstrates that the developed UPLC method was found to

be specific for Duloxetine in presence of its impurities (impurity-1,

impurity-2 and impurity-3) and degradation products.

4.5 Analytical method Validation, results and discussion

The developed and optimized UPLC method was taken up for

validation. The analytical method validation was carried out in

accordance with ICH guidelines [7, 8].

4.5.1 System Suitability Test

Duloxetine standard, impurity-1, impurity-2 and impurity-3 were

spiked at 0.30% level with respect to the concentration of Duloxetine in

diluent and injected for five times into UPLC system. Resolution between

impurities and Duloxetine, tailing factor for impurities and Duloxetine,

RSD% for the areas of impurities and Duloxetine was calculated. Good

resolution was obtained between impurities and Duloxetine (Fig: 4.5.F1).

System suitability results were tabulated in Table 4.5.T1.

Fig: 4.5.F1: Typical UPLC chromatogram of system suitability

Table 4.5.T1: System Suitability results

Compound

(n=5)

Resolution (Rs) USP

Tailing factor (T)

RSD%

for area

Impurity-1 -- 1.1 0.4

Duloxetine 3.7 1.1 0.3

Impurity-2 2.5 1.3 0.8

Impurity-3 22.3 1.1 0.5

n = Number of determinations

4.5.2 Precision

Assay method precision (repeatability) study was evaluated by

carrying out six independent assays of Duloxetine hydrochloride test

sample against qualified reference standard and RSD of six consecutive

assays was 0.15% (Table 4.5.T2). Results showed insignificant variation

in measured response which demonstrated that the method was

repeatable with RSDs below 0.2%.

Table 4.5.T2: Precision results of the assay method

Preparation % Assay

1 99.9

2 99.8

3 100.1

4 99.7

5 99.9

6 99.7

Mean 99.9

SD 0.1516

% RSD 0.15

95% Confidence interval of mean 99.8 to 100.0

The precision of the related substance method was evaluated by

injecting six individual preparations of Duloxetine (0.2 mg mL-1) spiked with 0.30% of impurity-1, impurity-2 and impurity-3 with respect to

Duloxetine analyte concentration. The % RSD for area of impurity-1,

impurity-2 and impurity-3 for six consecutive determinations was below

2.0% (Table 4.5.T3).

Results showed insignificant variation in measured response which

demonstrated that the related substances method was repeatable with

RSDs below 2.0%.

Table 4.5.T3: Precision results of the RS method

Preparation Impurity-1 Impurity-2 Impurity-3

1 22350 16070 46126

2 22377 15886 46126

3 22772 15902 45210

4 22925 15932 45415

5 22715 16048 45525

6 22151 15652 43827

Mean 22548 15915 45340

SD 298.7 149.8 814.4

% RSD 1.3 0.9 1.8

Intermediate precision for assay and related substances method was

performed by carrying out six independent assays of Duloxetine

hydrochloride test sample against qualified reference standard and

calculated %RSD of six consecutive assays. Related substances method

was performed by injecting six individual preparations of Duloxetine (0.2

mg mL-1) spiked with 0.30% impurity-1, impurity-2 and impurity-3 with

respect to Duloxetine analyte concentration over different days, different

instruments and with different analysts (Table 4.5.T4).

Table 4.5.T4: Results of repeatability and Intermediate precision

S.No

Parameter

Variation

%RSD

for Assay

%RSD

for Related Substances

1

Repeatability

(a) Analyst-1

(b) Waters Acquity

UPLC system with PDA detector.

(c) Day-1

<0.50%

< 2.0%

2

Intermediate

precision

(a) Analyst-2

(b) Waters Acquity

UPLC system with TUV detector.

(c) day-2

<0.50%

< 2.0%

4.5.3 Limit of quantification (LOQ) and limit of detection (LOD)

LOQ and LOD were established for Duloxetine and for impurity-1,

impurity-2 and impurity-3 based on signal to noise ratio (S/N) method.

4.5.3.1 Limit of quantification (LOQ)

The quantification limits were established by S/N methodology and

established concentration values are reported in Table 4.5.T5.

Table 4.5.T5: LOQ values for Duloxetine and its impurities

S.No Impurity

name Concentration

Signal to noise

ratio

1 Duloxetine 0.080 µg mL-1 9.6

2 Impurity-1 0.026 µg mL-1 10.4

3 Impurity-2 0.054 µg mL-1 9.5

4 Impurity-3 0.028 µg mL-1 9.8

4.5.3.2 Limit of detection (LOD)

The detection limits were established by S/N methodology and

established concentration values are reported in Table 4.5.T6.

Table 4.5.T6: LOD values for Duloxetine and its impurities

S.No Impurity name Concentration Signal to noise

ratio

1 Duloxetine 0.020 µg mL-1 3.1

2 Impurity-1 0.010 µg mL-1 3.3

3 Impurity-2 0.020 µg mL-1 2.5

4 Impurity-3 0.010 µg mL-1 3.4

4.5.4 Precision at Limit of quantification

The precision of the related substance method was also checked by

injecting six individual preparations of impurity-1, impurity-2 and

impurity-3 at their LOQ level with respect to Duloxetine analyte concentration. Results showed insignificant variation in measured

responses which demonstrate that the method was repeatable at LOQ

level with RSD below 2.3 % (Table 4.5.T7).

Table 4.5.T7: Precision results of the RS method at LOQ level

Preparation Impurity-1 Impurity-2 Impurity-3

1 1035 1496 2143

2 1073 1558 2155

3 1032 1489 2171

4 1089 1466 2206

5 1051 1531 2185

6 1033 1492 2133

Mean 1052.1 1503.6 2165.5

SD 23.92 31.87 27.28

% RSD 2.3 2.1 1.3

4.5.5 Accuracy at LOQ level

The recovery studies for impurity-1, impurity-2 and impurity-3 were

carried out in triplicate at LOQ level of the Duloxetine target analyte

concentration (200 µg mL-1). The percentage recovery of impurity-1, impurity-2 and impurity-3 was calculated (Table 4.5.T8). The method

showed consistent and high absolute recoveries at LOQ level with mean

absolute recovery ranging from 94.8 % to 98.5%.

Table 4.5.T8: Recovery at LOQ level

S.No

Impurity name

Mean recovery

(%)(n = 3 )

%RSD

1 Impurity-1 98.5 2.7

2 Impurity-2 94.8 2.2

3 Impurity-3 98.4 2.7

4.5.6 Linearity

4.5.6.1 Linearity of the assay method

The linearity of the assay method was established by injecting test

sample at 10%, 50%, 75%, 100%, 125% and 150% of Duloxetine assay

concentration (i.e.100 µg mL-1). Each solution was injected into UPLC

system (Table 4.5.T9). Calibration curve obtained by least square

regression analysis between peak area and the concentration showed

(Fig: 4.5.F2 and Table 4.5.T10) linear relationship with a regression

coefficient of 0.999. The best fit linear equation obtained was y =

30204.95 x con + 2701.18.

Table 4.5.T9: Linearity results of the assay method

Concentration

(µg mL-1)

Peak area

0.2 6062

10 303123

15 462023

20 607522

25 760123

30 903890

Correlation

Coefficient(r) 0.99992

Slope 30204.95

Intercept 2701.18

Fig: 4.5.F2: Linearity plot for Assay Method

Table 4.5.T10: Residual summary of Linearity results of Duloxetine

% of impurity

[w.r.t 200

µg/ml]

Mean area

Response chieved

Response calculated

thru Trend

line equation

Residual

(Response

practical -Response

theoretical)

Residual

square

0.2 6062 6684.55 623 387572.7

10 303123 302693.06 -430 184850.8

15 462023 453717.80 -8305 68976280.7

20 607522 604742.55 -2779 7725337.7

25 760123 755767.30 -4356 18972142.9

Residual sum of squares

192492369.67

Trend line equation y = 30204.95 x con + 2701.18

4.5.6.2 Linearity of the related substances method

Linearity experiments were carried out by preparing the Duloxetine sample solutions containing impurity-1, impurity-2 and impurity-3 from

LOQ to 150% (i.e. LOQ, 50, 100, 125 and 150%) with respect to their

specification limit 0.30% (Table 4.5.T11).

Calibration curve was drawn by plotting average area of the impurity (impurity-1, impurity-2 and impurity-3) on the Y-axis and concentration

on the X-axis (Fig: 4.5.F3 to Fig: 4.5.F5) which showed linear relation

ship with a regression coefficient of greater than 0.998 for all impurities.

Analysis of residuals indicated (Table 4.5.T12 to Table 4.5.T14) that the residuals were normally distributed around the mean with uniform

variance across all concentrations suggesting the homoscedastic nature

of data.

Table 4.5.T11: Linearity results of the RS method

S.No

% of impurity

w.r.t 200

µg/ml

Impurity-1

(Peak area)

Impurity-2

(Peak area)

Impurity-3

(Peak area)

1 LOQ 1037 1468 2155

2 0.15 12724 8523 21822

3 0.30 22620 15247 44727

4 0.375 27642 19151 56901

5 0.45 32773 22000 64105

Correlation

coefficient (r) 0.998439 0.998326 0.998558

Intercept 897.8831 656.335 374.1549

Slope 71666.6 48469.53 145724.8

Fig: 4.5.F3: Linearity plot for impurity-1

Linearity graph of Impurity-1

y = 71667x + 897.88

R2 = 0.9969

0

5000

10000

15000

20000

25000

30000

35000

0 0.1 0.2 0.3 0.4 0.5

% impurity w .r.t test concentration

Mean

are

a

Fig: 4.5.F4: Linearity plot for impurity-2

Linearity graph of Impurity-2

y = 48470x + 656.33

R2 = 0.9967

0

5000

10000

15000

20000

25000

0 0.1 0.2 0.3 0.4 0.5

% impurity w .r.t test concentrationM

ean

are

a

Fig: 4.5.F5: Linearity plot for impurity-3

Linearity graph of Impurity-3

y = 145725x + 374.15

R2 = 0.9971

0

10000

20000

30000

40000

50000

60000

70000

0 0.1 0.2 0.3 0.4 0.5

% impurity w .r.t test concentration

Mea

n a

rea

Table 4.5.T12: Residual summary of Linearity results of impurity-1

% of

impurity [w.r.t 200

µg/ml]

Mean

area Response

chieved

Response calculated

thru Trend

line equation

Residual (Response

practical -

Response theoretical)

Residual square

0.013 1037 1829.5 793 628133.9

0.15 12724 11647.9 -1076 1158048.1

0.30 22620 22397.9 -222 49344.4

0.375 27642 27772.9 131 17124.1

0.45 32773 33147.9 375 140515.8

Residual sum of squares

1993166.36

Trend line equation y = 71667x + 897.88

Table 4.5.T13: Residual summary of Linearity results of impurity-2

% of

impurity

[w.r.t 200

µg/ml]

Mean area

Response achieved

Response

calculated

thru Trend line

equation

Residual

(Response

practical -Response

theoretical)

Residual

square

0.027 1468 1965.0 497 247021.2

0.15 8523 7926.8 -596 355497.2

0.30 15247 15197.2 -50 2480.7

0.375 19151 18832.4 -319 101500.9

0.45 22000 22467.6 468 218670.8

Residual sum of squares

925170.77

Trend line equation y = 145725x + 374.15

Table 4.5.T14: Residual summary of Linearity results of impurity-3

% of

impurity [w.r.t

200

µg/ml]

Mean area Response

achieved

Response

calculated thru

Trend line

equation

Residual

(Response practical -

Response

theoretical)

Residual

square

0.027 2155 2414.3 259 67237.3

0.15 21822 22232.9 411 168814.2

0.30 44727 44091.6 -635 403752.0

0.375 56901 55020.9 -1880 3534615.1

0.45 64105 65950.3 1845 3405133.5

Residual sum of squares

7579552.06

Trend line equation y = 145725x + 374.15

4.5.7 Accuracy

4.5.7.1 Accuracy of the assay method

Accuracy of the assay method was established by injecting three preparations of test sample using Duloxetine hydrochloride capsules 60

mg at 50%, 100% and 150% of analyte concentration (20 µg mL-1). Each

solution was injected trice (n=3) into UPLC system and the mean peak

area of Duloxetine peak was calculated. Assay of test solution was

determined against three injections (n=3) of qualified Duloxetine standard. The method showed consistent and high absolute recoveries at

all three concentration (50,100 and 150%) levels with mean absolute

recovery ranging from 99.6 % to 101.2% (Table 4.5.T15). The obtained absolute recoveries were normally distributed around the mean with

uniform RSD values.

4.5.7.2 Accuracy of the RS method

Accuracy of the related substances method established at 50%, 100% and 150% of the impurities specification limit (0.20%). Test solutions

were prepared in triplicate (n=3) with impurities (impurity-1, impurity-2

and impurity-3) at 0.15%, 0.30% and 0.45% level w.r.t. analyte concentration (0.2 mg mL-1). Mean % recovery of impurities calculated in

the test solution using the area of impurities standard at 0.30% level

with respect to analyte (Table 4.5.T15).

Table 4.5.T15: Recovery results

Amount

spiked

% Recovery

Mean ± % RSD

Duloxetine Impurity-1 Impurity -2 Impurity -3

50%

101.2±0.12 100.5±1.4 93.3±1.7 99.7±1.3

100 %

99.6±0.11 101.6±1.8 96.8±2.1 98.6±2.1

150 %

100.4±0.15 104.5±1.2 96.2±0.6 98.6±2.6

The related substances method showed consistent and high absolute

recoveries of all five impurities at all three concentration (50,100 and 150%) levels with mean absolute recovery ranging from 93.3 % to 104.5%

in drug product.

4.5.8 Solution state stability

The solution state stability of Duloxetine in diluent in the assay method was carried out by leaving the test solutions of sample in tightly

capped volumetric flasks at room temperature for two days. The same

sample solution were assayed for every 24 hours interval up to the study period, each time freshly prepared reference standard was used to

estimate the assay of test sample. The % RSD of assay of Duloxetine

during solution stability experiments was with in 1.0%.

The solution stability of Duloxetine in the related substance method

was carried out by leaving sample solution in tightly capped volumetric flask at room temperature for two days. Content of impurity-1, impurity-

2, and impurity-3 were checked for every 24 hours interval up to the

study period. No significant change was observed in the impurity content during solution stability experiments from the initial values up to the

study period. Hence Duloxetine sample solutions are stable for at least

48 hours in the developed method.

Mobile phase stability of the developed method was established by injecting fresh samples for 48 hours. Fresh sample solution was assayed

for every 24 hours interval up to the study period, each time freshly

prepared reference standard was used to estimate the assay of sample. The % RSD of assay of Duloxetine during solution stability experiments

was with in 1.0.

Table 4.5.T16: Solution stability results of the assay method

S.No Interval % Assay

Solution stability

% Assay

Mobile phase stability

1 0 h 99.7 99.5

2 24 h 99.2 99.0

3 48 h 98.7 99.4

% RSD 0.50 0.26

4.5.9 Robustness

To determine the robustness of the developed method experimental

conditions were purposely altered and the resolution between impurity-1

and Duloxetine and between Duloxetine and impurity-2 were evaluated.

In each of the deliberately altered chromatographic condition (flow rate

0.5 mL min-1 and O.7 mL min-1, varying column temperature from 35°C

to 450C), varying pH of mobile phase by −1 to +1 units), the resolution

between impurity-1, Duloxetine, impurity-2 and impurity-3 was greater

than 2.4 and tailing factor for Duloxetine and its impurities was less

than 1.3, illustrating the good robustness of the method.

4.5.10 Mass balance

The stressed samples of Duloxetine hydrochloride drug product were

assayed against the qualified reference standard and the results of mass

balance obtained were very close to 99%. The results of mass balance obtained in each condition are presented below (Table 4.5.T17).

Table 4.5.T17: Mass balance of the assay method

Stress condition Time

~%

Degradation

% Assay of

active substance

Mass balance

(%Assay +

%impurities

+% Degradation

products)

Acid hydrolysis

(1 N HCl,60°C ) 3 h 40.22 % 58.9 99.1

Base hydrolysis

(1 N NaOH, 60°C )

3 h 2.04 % 97.6 99.6

Oxidation (6% H202) 3 h 0.21% 99.6 99.8

Water hydrolysis

(60°C) 3 h 14.22% 84.6 99.3

Thermal (60° C) 10 days 0.13% 99.5 99.6

Light (photolytic

degradation) 10 days 0.21% 99.9 100.1

4.6 In vitro Dissolution Studies by developed stability indication

UPLC method

Dissolution tests are one of the tests most used in the

characterization of drugs and in the quality control of dosage forms. It

was introduced in the 1960’s and it is evolved into a test that

pharmaceutical manufacturers hope will better predict the in vivo

performance of drug products. In addition, the reliability and

discriminatory capabilities of dissolution methods for delayed release

products has gained much attention in recent years. The most widely

used dissolution tests for delayed release products use 900 or 1000 mL

of an acid medium followed by 900 or 1000 mL of aqueous buffer

medium with USP apparatus I (basket) or apparatus II (paddle) at

rotation speeds of 100 or 50 RPM, respectively. It is desirable to have an

in vitro dissolution method that is sensitive to formulation factors that

affect the dissolution process and in consequence bioavailability. In our

study, release characteristics of Duloxetine hydrochloride capsules were

studied according to the FDA dissolution method (Basket) at 37 C ± 0.5

using 1000 mL of 0.1N HCl (2 h) followed by 1000 mL of 6.8 phosphate

buffer (pH6.8) at 100 RPM rotation speed [9]. In vitro dissolution was

conducted on Duloxetine hydrochloride 60 mg test product and on

Duloxetine hydrochloride 60 mg innovator (Cymbalta) product. At

scheduled time intervals, 1 mL of sample was withdrawn and replaced

with fresh medium. The final solution was filtered through 0.45 µm nylon

membrane filter and then analyzed by the proposed UPLC method.

The dissolution characteristics of Duloxetine hydrochloride are given

Table 4.6.T1 and Fig: 4.6.F1. Analyzing the in-vitro dissolution profile of

test and innovator, it could be concluded that test showed the faster

release rate than that of innovator. Initially at 15 min and at 30 min,

dissolution rate of test product slightly faster than the innovator product,

but from 45 min to 90 min both test and innovator products were

releasing in a similar way. The release of the total amount of Duloxetine

was achieved at 90 min for innovator product; test product was released

completely at 45 min in the medium. Based these observations from test

and innovator products, it could be concluded that initial time points up

to 30 min were critical for in- vitro dissolution mapping. This result was

in agreement with the literature and the dissolution method mentioned

above would be appropriate to simulate the in vitro release of Duloxetine

hydrochloride for innovator and test products.

Table 4.6.T1: Results of In-vitro release profile of Duloxetine

hydrochloride 60 mg capsules for test and Innovator

Time in

minutes

% Dissolution

Test product Innovator

15 54 44

30 91 72

45 93 89

60 93 93

90 93 93

Fig: 4.6.F1: In-vitro release profile of Duloxetine hydrochloride 60

mg capsules for test and Innovator

4.7 Summary and Conclusions

A new reverse phase ultra performance liquid chromatographic

method was developed for assay and impurities estimation Duloxetine

hydrochloride with stability-indicating power for pharmaceutical dosage forms. The proposed developed method is capable of separating all

Duloxetine hydrochloride impurities in presence degradation impurities

formed under forced degradation studies and also proposed method can be employed for assay determination of Duloxetine hydrochloride in

pharmaceutical dosage forms. The column used for development of this

method was Zorbax XDB C-18, 50 mm × 4.6 mm, 1.8 µm column. The detection wavelength was to monitor eluted components was 236 nm.

Duloxetine hydrochloride capsules pellets were exposed to degradation

conditions of acid, oxidative, base, hydrolytic, photolytic and thermal

degradation. The degradation impurities were well separated from Duloxetine peak and its related compounds, showed the stability-

indicating nature and power of the developed method. The developed

UPLC method was also applied for in-vitro dissolution studies for test and innovator products. The developed method was validated as per ICH

guidelines for precision, accuracy, linearity, specificity, limit of detection,

limit of quantification, system suitability and robustness and found to be precise, accurate, linear rugged, robust and specific. Thus the developed

and validated liquid chromatographic method can be used for stability

analysis and production samples analysis of Duloxetine hydrochloride in pharmaceutical dosage forms.

Table 4.7.T1: Summary of analytical method validation

Test Parameter

Related Substance method

Assay

method

Imprity-1 Impurity-2 Impurity-3

Precision (RSD)

1.3 0.9 1.8 0.15

LOD

(µg mL-1) 0.010 0.020 0.010 0.020

LOQ

(µg mL-1) 0.026 0.054 0.028 0.080

Precision

at LOQ(%RSD) 2.3 2.1 1.3 -

Linearity

(Corre coefficient) 0.9984 0.9983 0.9985 0.9999

Accuracy % at

100%

(Drug product)

101.6 96.8 98.6 99.6

Robustness the resolution between Impurity-1, Duloxetine, Impurity-2 and Impurity-3 was greater than 2.4

Solution

Stability

Standard and test solutions were stable for 48 hours at room temperature

Mobile phase

Stability

Mobile phase was stable for 48 hours at room temperature

Specificity

For all stress conditions, purity angle is less than

the purity threshold and there is no purity flag observed for purity results

References:

[1] P.Srinivasulu, K.S. Srinivas, R.S. Reddy, Mukkanti K and R.

Bucchireddy, Pharmazie. 64 (2009) 10-3.

[2] J. Yang, X. Lu, Y. Bi, F. Qin and F. Li, Chromatogr. 66 (2007)

389-393.

[3] M.M. Kamila, N. Mondal and L.K. Ghosh, Pharmazie. 62 (2007)

414-415.

[4] P.S. Selvan, V. Gowda, U. Mandal, W.D. Sam Solomon and T.K.

Pal, J. Chromatogr. B. 858 (2007) 269-275.

[5] D.K. Satonin, J.D. McCulloch, F. Kuo and M.P. Knadler, J.

Chromatogr. B. 852 (2007) 582-589.

[6] P.J. Jansen, P.L. Oren, C.A. Kemp, S.R. Maple and S.W.

Baertschi, J. Pharm. Sci. 87 (1998) 81-5.

[7] International Conference on Harmonization (1995) ICH guidelines

on validation of analytical procedures: text and methodology Q2

(R1): FDA. Federal Register 60:11260.

[8 International Conference on Harmonization (1996) ICH guidelines

on stability testing: Photo stability testing of New drug substances

and products (Q1B).

[9] www.accessdata.fda.gov/scripts/cder/dissolution/index.cfm.