A versatile LC method for the simultaneous quantification of latanoprost, timolol and benzalkonium chloride and related substances in the presence of their degradation products in ophthalmic solution

Abstract

A challenging and multipurpose stability-indicating HPLC method is developed and validated for the quantification of latanoprost, timolol, benzalkonium chloride (BKC) and related substances in the presence of their degradation products in ophthalmic solution. Various stress conditions, e.g. acid, base, oxidation, heat, UV radiation and heat were employed to assess the stability-indicating nature of the method. A strategic experimental approach was implemented for the method development. The desired chromatographic separation was achieved on a reverse phase cyano column {Hypersil BDS CN (250 × 4.6 mm), 5 μm} under gradient elution conditions. A mixture of phosphate buffer of pH 3.2, acetonitrile and methanol was used as the mobile phase. The determination was carried out at 295 nm and 210 nm. Due to the very low content of latanoprost (0.005%) in the formulation, the injection volume was optimized as 80 μl in order to achieve an optimum response. In totality, peaks of latanoprost and its two known impurities, four BKC homologs, timolol and its known impurities were obtained with a minimum resolution of 3.0. This sensitive method was found to be precise and linear through out the range with a lowest quantification of 0.068 μg ml⁻¹ for 15-keto latanoprost, 0.030 μg ml⁻¹ for latanoprost acid, 0.018 μg ml⁻¹ for latanoprost (unknown impurity), 0.011 μg ml⁻¹ for timolol impurity I and 0.013 μg ml⁻¹ for timolol (unknown impurity). The method was validated according to ICH guidelines. This versatile method can be used as a combined assay and related substance method for said ophthalmic solution formulation.

---

Introduction

Impurity profiling of active pharmaceutical ingredients (API) in both bulk material and finalized formulations is one of the most challenging tasks for pharmaceutical analytical chemists in the industrial environment.^1,2 The presence of unwanted or in certain cases unknown chemicals, even in small amounts, may influence not only the therapeutic efficacy but also the safety of the pharmaceutical products.³ The quantification of such components becomes thornier, when the amount of APIs or drugs which are used in formulation is very low and in combination with one or more drugs.

The ophthalmic solution, for which the analytical method is developed and validated, contains latanoprost (0.005%), timolol maleate (0.5%) and BKC (preservative) (0.02%) (Fig. 1) in very low amounts. Latanoprost and timolol decrease elevated intraocular pressure (IOP) by different mechanisms of action. Latanoprost, a prostaglandin F2alpha analogue, is a prostanoid selective prostaglandin F2 (FP) receptor agonist that reduces the IOP by increasing the outflow of the aqueous humour. The main mechanism of action is increased uveoscleral outflow. Timolol is a non-selective β adrenergic receptor blocking agent and it lowers IOP by decreasing aqueous humour formation in the ciliary epithelium.⁴ Benzalkonium chloride (BKC) is a mixture of alkylbenzyldimethylammonium chlorides of various even numbered alkyl chain lengths. It has been considered one of the safest synthetic biocides known and has a long history of efficacious use. Its use as a preservative in cosmetics such as eye and nasal drops attests to its general safety.^5,6

Fig. 1 Structures of latanoprost, timolol, BKC and their related substances.

As per the drug master files of latanoprost⁷ and timolol⁸ there are two known impurities of latanoprost: 1. Latanoprost acid (PhXA 85, (5Z,9α,11α,15R)-9,11,15-trihydroxy-17-phenyl-18,19,20-trinor-prost-5-en-1-oic acid) (major degradant in acid conditions) and, 2. 15-keto latanoprost (9α,11α-dihydroxy-15-oxo-17-phenyl-18,19,20-trinor-prost-5Z-en-1-oicacid, isopropyl ester). There is one known impurity of timolol: timolol impurity I, which was quantified and identified along with unknown impurities and preservative in the reported method.

An extensive literature survey was carried out, which revealed that the said formulation is not official according to any pharmacopoeia, and only timolol maleate is officially a drug substance and individual formulation according to the Indian Pharmacopoeia,⁹ the British Pharmacopoeia¹⁰ and the US Pharmacopoeia.¹¹ A few reports have been previously published regarding the determination of latanoprost by HPLC,^12–14 the stability of latanoprost,^15,16 the determination of BKC by HPLC⁶ and the determination of timolol by HPLC and NMR, voltammetric and HPTLC techniques.^17–21 There is no method available for the simultaneous quantification of latanoprost, timolol and BKC in a formulation. The presented method was developed in such a way that, in a single analysis two drugs, one preservative, their known/unknown impurities and degradation products present in very low amounts can be determined with higher accuracy and precision.

Experimental

Materials and reagents

All experiments were performed using ‘A class’ volummetric glassware. Analytical grade sodium dihydrogen orthophosphate dihydrate (S. D. Fine Chem., Ahmedabad, India), HPLC grade acetonitrile and orthophosphoric acid (Spectrochem, Ahmedabad, India), and methanol (Finar Chem., Ahmedabad, India) and HPLC grade Milli Q water (Millipore, Bedford, MA, USA) were used in the preparation of the mobile phase for gradient elution. The mobile phase was filtered through a 0.45 μm PVDF filter (Millipore, Barcelona) and degassed under vacuum, prior to use. To check the stability-indicating nature of the method, samples were subjected to various stresses, for which sodium peroxide, hydrochloric acid and sodium hydroxide (Finar Chem., Ahmedabad, India) were used. For the preparation of the reference solution and system suitability solution, pharmaceutical grade reference standards of latanoprost (Chirogate, Taiwan), timolol (FDC Ltd., India), BKC (Sigma-Aldrich, India), latanoprost acid (Chirogate, Taiwan), 15-keto latanoprost (Carbomer Inc.) and timolol impurity I (FDC Ltd, India) were used.

Analytical solutions

Stock solutions. Stock solutions of latanoprost, 15-keto latanoprost and latanoprost acid were suitably diluted by acetonitrile to get concentrations of 1000 μg ml⁻¹, 50 μg ml⁻¹ and 50 μg ml⁻¹ respectively. Simultaneously stock solutions of timolol (2000 μg ml⁻¹), timolol impurity I (250 μg ml⁻¹) and BKC (5000 μg ml⁻¹) were prepared with water.

Resolution solution. A solution containing 5.0 ml each of latanoprost and timolol stock, 4.0 ml of BKC stock, 1.0 ml each of 15-keto latanoprost and latanoprost acid stock and 200 μl of timolol impurity I stock diluted to 100 ml with water, was used as a resolution solution to confirm system suitability.

Reference solution. A solution containing 50 μg ml⁻¹ of latanoprost, 100 μg ml⁻¹ of timolol and 200 μg ml⁻¹ of BKC was used as a reference solution.

Sample solutions. For the quantification of latanoprost, BKC and their related substances, the formulation vial was directly used as sample solution and for the determination of timolol, the sample was diluted with water to get a concentration of timolol of 100 μg ml⁻¹.

Chromatographic conditions

The liquid chromatograph consisted of a Waters model with a quaternary pump, an automated injector with a 100 μl loop and a detector (PDA 2998 or dual wavelength). The empower software was used for data collection and processing.

A filtered and degassed solution of 0.05 M monobasic sodium phosphate dehydrate was prepared and the pH was adjusted to 3.20 using orthophosphoric acid, this solution was used as mobile phase A and a mixture of methanol and acetonitrile in equal volumes was used as mobile phase B. All the sample and standard solutions were prepared using purified water as the diluent. The separation was achieved on a reverse phase cyano column, Hypersil BDS CN (250 × 4.6 mm), 5 μm and gradient elution (Table 1: Gradient elution program) for mobile phase A and B, was performed for 55 min at a flow of 1.0 ml min⁻¹. The detection was performed at 210 nm (for latanoprost and BKC) and 295 nm (for timolol). An injection volume of 80 μl for both the standard and the sample was used. The eluted peaks were well separated and sharp under the above mentioned chromatographic conditions. The related substances were quantified by an area normalization method, while the potency of the drug substances was estimated by an external standard method.

Table 1 Gradient elution program

Time/min	Mobile phase A (%)	Mobile phase B (%)
0.00	64	36
25.0	64	36
43.0	15	85
43.1	64	36
55.0	64	36

---

Stress study: (forced degradation study)

Stress testing of drug products can aid in the identification of likely degradation products, which can reveal the intrinsic stability of the product. A specific stability-indicating analysis method has the ability to measure the responses of analytes and related impurities in the presence of potential degradation products.²² The interference of degradation products was investigated in both drug substances and drug products by analyzing the sample preparation with different stress conditions such as acid hydrolysis, base hydrolysis, oxidative hydrolysis, UV light stress and heat stress. The conditions for stress degradation are summarized in Table 2.

Table 2 Stress study conditions

Acid stress conditions
Concentration of acid	5 N hydrochloric acid
Time	6 h – LTP, 12 h – TIM
Temperature	40 °C
Alkali stress conditions
Concentration of base	5 N sodium hydroxide
Time	4 min – LTP, 30 min – TIM
Temperature	25 °C, 40 °C
Oxidative stress conditions
Concentration of hydrogen peroxide	30% w/w H2O2
Time	2 h – LTP, 1 h – TIM
Temperature	40 °C
UV Radiation conditions
Exposure to UV Radiation	200 watt hours m^−2
Heat Stress conditions
Time	48 h
Temperature	40 °C

---

The experimental conditions that were followed for the stress study are as follows.

Acid hydrolysis. 0.2 ml and 2 ml of 5N hydrochloric acid were added into solutions of drug substances and drug products of latanoprost and timolol respectively and the solutions were kept at 40 °C for 6 h and 12 h.

Base hydrolysis. 0.2 ml and 2 ml of 5N sodium hydroxide solution were added into solutions of drug substances and drug products of latanoprost and timolol respectively and the solutions were kept at 25 °C and 40 °C for 4 min and 30 min.

Oxidative hydrolysis. 0.2 ml and 2 ml of 30.0% hydrogen peroxide solution were added into solutions of drug substances and drug products of latanoprost and timolol respectively and the solutions were kept at 40 °C for 2 h and 1 h.

UV radiation stress. The solutions of latanoprost and timolol drug substances and drug products were exposed to UV radiation of 200 watt·hours m⁻² in a photo stability chamber.

Heat stress. The solutions of latanoprost and timolol drug substances and drug products were exposed to 40 °C heat for 48 h in a water bath.

Validation study

Specificity. The specificity of the method was established by checking the interference of diluent, placebo, known impurities and degradation products with the analyte peaks. The specificity was evaluated by observing the chromatograms of blank samples and samples spiked with APIs and all known impurities. The peak purity of the peaks of interest was checked by comparing purity angle and purity threshold.

Precision.
Instrument precision: (Suitability of the system). The suitability of the system was checked by a single injection of the resolution solution and five replicate injections of the reference solution. The %RSD, theoretical plates, tailing factor and resolution were optimized as the system suitability parameters. The acceptance criteria were set as mentioned in Table 3.

Table 3 System suitability Data

For resolution solution
Parameter	Observed value	Acceptance criteria
At wavelength of detection 210 nm
Resolution between latanoprost acid and latanoprost	17.0	NLT 10.0
Resolution between latanoprost and 15-keto latanoprost	7.4	NLT 3.0
At wavelength of detection 295 nm
Resolution between timolol and timolol impurity I	5.5	NLT 3.0

---

For reference solution: (at 210 nm)
	Observed value
	Latanoprost			BKC homolog C-12
	Peak area	Theoretical plates	Tailing factor	Peak area	Theoretical plates	Tailing factor
Average	4655727	9686	1.2	12356241	23560	2.6
RSD	0.1%	N/A	N/A	0.1%	N/A	N/A
Acceptance criteria	RSD NMT 2.0%	NLT 2500	NMT 2.0	RSD NMT 2.0%	NLT 10000	NMT 4.0

---

For reference solution: (at 295 nm)
		Observed value (for timolol)
		Peak area			Theoretical plates				Tailing factor
Average		12529118			4104				1.8
RSD		0.1%			NA				NA
Acceptance criteria		RSD NMT 2.0%			NLT 2500				NMT 2.5

---

Method precision: (Repeatability). Six sets of sample preparations were prepared by spiking known amounts of impurities and were analyzed for estimation of impurities, while six sets of sample preparations were directly used for assay. Latanoprost, timolol, total BKC homologs, known impurities, unknown individual impurity and total impurities were calculated as mentioned in the test method for each set of test preparations.
Intermediate precision: (Reproducibility). The reproducibility was conducted on different chromatographic systems by different analysts and on different days. The analysis was conducted in the same manner as the method precision and the %RSD for all six sets of sample preparations and spiked sample preparations was determined. The difference in the % average results of latanoprost, timolol and BKC content along with known and unknown impurities was determined.

Stability of analytical solutions. The aim of this study was to prove the stability of the sample solution and reference solution at room temperature. For the study, duplicate sets of spiked sample preparations and sample preparations as per the test method were prepared and kept on a bench top (25 °C ± 2 °C) and analyzed initially (0 days), after 1 day and after 2 days by a single injection of each set of sample preparations and spiked sample preparations into a liquid chromatography column, and chromatograms were recorded. The difference in the assay of latanoprost, timolol and BKC content was determined at each time interval. Known impurities, unknown individual impurities and total impurities were determined at each time interval against the respective initial results, and the difference in the results was determined.

Limit of detection (LOD) and limit of quantification (LOQ). The limit of detection and limit of quantification concentrations for latanoprost, timolol, 15-keto latanoprost, latanoprost acid and timolol impurity I were determined based on a linear regression method (standard deviation of response and slope method) as per ICH guideline Q2. Linearity was performed with five different solutions with concentrations 0.040 μg ml⁻¹ to 0.060 μg ml⁻¹ of latanoprost, 0.010 μg ml⁻¹ to 0.050 μg ml⁻¹ of timolol maleate equivalent to timolol, 0.025 μg ml⁻¹ to 0.125 μg ml⁻¹ of 15-keto latanoprost and latanoprost acid and 0.025 μg ml⁻¹ to 0.125 μg ml⁻¹ of timolol impurity I. Linearity graphs of concentration in μg ml⁻¹ (X-axis) versus average area (Y-axis) were plotted. The slope of regression and the residual standard deviation were calculated. LOD and LOQ concentrations were determined on the basis of the equation given below.

LOD = (3.3 × σ)/S and LOQ = (10 × σ)/S

Where, σ = residual standard deviation of regression line, S = slope of calibration curve.

Six replicate injections of these derived LOD and LOQ concentrations were performed and it was concluded that this ensured that the peaks were detected and the responses were reproducible.

Linearity study. The linearity study was performed in a range of LOQ to 150.0% (LOQ, 50%, 75%, 100%, 125% and 150%) concentration of known and unknown impurities, considering their limits. As the calculation was done by an area normalization method, an additional linearity up to 150.0% concentration for latanoprost and timolol in sample preparation was performed to meet the requirements of the calculation. Duplicate injections of each linearity solution were introduced into a liquid chromatography column and chromatograms were recorded. Linearity graphs of concentration in μg ml⁻¹ (X-axis) versus average area (Y-axis) were plotted. Different parameters for the linearity level were calculated.

Accuracy (by recovery). The accuracy of an analytical method should be established across its range. Accuracy was performed in the range of LOQ to 150.0% (LOQ, 50.0%, 100.0% and 150.0%) of the target concentration of latanoprost, timolol and their known and unknown impurities. Triplicate sets of accuracy sample preparations at each accuracy level were prepared and injected into a liquid chromatography column and chromatograms were recorded.

Robustness. The robustness of the method was demonstrated by performing the system suitability test as per the test method under normal conditions and under the altered conditions mentioned below.

1. Changing the temperature of the column (±5 °C, i.e. 20 °C and 30 °C).

2. Changing the wavelength of the detector (±2 nm, i.e. 208 nm and 212 nm, and 293 nm and 297 nm).

3. Changing the flow rate of the mobile phase (±10%, i.e. 0.9 ml min⁻¹ and 1.1 ml min⁻¹).

4. Changing the organic solvent ratio (±5% relative).

5. Changing the pH of the aqueous phase of the mobile phase (±0.2, i.e. 3.00 and 3.40).

Results and discussion

Optimization of chromatographic conditions

The quality is closely related not only with the quality of API itself, but also with the very small amounts of impurities in the preparation. Moreover, impurity levels of degradation products are a good indicator of active substance stability, and consequently the formulation stability. Therefore it is very important to give great consideration to these detrimental degradation products. In addition to chemical stability, impurities in pharmaceutical products can cause problems associated with toxicity or the performance of pharmaceutical products. Therefore, determination of related substances and degradation products in formulations is necessary.

HPLC along with UV-PDA detection was chosen as a simple, reliable and effective separation method for the determination of latanoprost, timolol, BKC and their related substances in the formulation. In the selection of optimal absorption wavelengths for the detection of the compounds, particular attention was paid to the absorption spectra of the degradation products of latanoprost and timolol to avoid unnecessary merging of peaks and also to give exact amount of impurities at their wavelengths, so wavelengths of 210 nm for latanoprost and 295 nm for timolol were selected.

For the purpose of separation, several RP columns were compared among which Hypersil BDS CN (250 × 4.6 mm), 5 μm was found to be suitable due to its high selectivity towards polar and hydrophilic compounds. A higher amount of buffer in the mobile phase yielded a better separation of the latanoprost peak, but it was not required for BKC and its homologs, hence to minimize the analysis time, gradient elution was chosen. In the first interval timolol and its related substances eluted, in the second portion latanoprost and its related substances eluted, and as the organic portion of mobile phase became higher BKC and its homolog eluted. As per scanned spectrums of latanoprost, timolol and their known impurities, the wavelengths were selected. The dual wavelength method that was used is more advantageous for quantifying related substances in a combinational product. Due to the very low label content of all the APIs, especially of latanoprost (0.005%), the injection volume was optimized at 80 μl, which is higher than the regular injection volume of 10–50 μl. In order to ensure easy identification in gradient noise and to obtain an optimum response throughout the concentration range for all related peaks, the injection volume was optimized at 80 μl.

In order to develop a good method, some important parameters were considered, e.g. a minimum resolution of 3.0 between all impurity and analyte peaks, good peak shapes and group-wise elution of peaks. The gradient elution program of the mobile phase was set in such a way that the timolol group (timolol, timolol impurity I and its degradation product) elute in the first portion of 15 min, after that the latanoprost group elutes in the next 15 min and after that all BKC homologs (BKC X, BKC C12, BKC C14 and BKC C16) elute with baseline separation in the next 20 min. To avoid any kind of carry over of degradation products and to extend the post run equilibrium, the runtime was optimized as 55 min. To further ensure the system suitability for the desired separation and accurate analysis, the peak shape and column efficiency parameters were optimized. On the basis of the above observations, a versatile, simple and sensitive method was established, through which simultaneous analysis of latanoprost, timolol, BKC and related substances was done accurately.

Method validation study

Stress study (forced degradation study). On exposing the samples to various stress conditions it was observed that latanoprost was highly sensitive and degradable towards acidic conditions, and timolol was towards basic conditions. Latanoprost acid, the known process impurity of latanoprost, was observed as a major degradant in acidic conditions, while in basic conditions unknown degradation was observed. The process impurity of timolol, timolol impurity I, was not observed as a degradant under any conditions. The detailed data of the forced degradation study are summarized in Table 4.

Table 4 Data of forced degradation study

Peak name	Acid	Base	Oxidation	UV	Heat
For latanoprost
Latanoprost acid	4.01%	0.11%	ND	ND	0.27%
15-Keto latanoprost	ND	ND	ND	0.42%	0.53%
Single maximum unknown impurity	0.08%	1.48%	ND	0.16%	ND
Total impurities	4.15%	1.59%	ND	0.58%	0.80%
For timolol maleate
Timolol impurity I	ND	ND	ND	ND	ND
Single maximum unknown impurity	0.03%	15.52%	0.99%	ND	ND
Total impurities	0.03%	15.52%	1.94%	ND	ND

---

The stability-indicating nature of the method was confirmed by the observation that the peaks of the degradation products were well separated from those of the analytes and related impurities. The purity of the analyte peaks was assessed by verifying the purity angle and purity threshold of analyte peaks under each stress condition. In all the studies the purity angle value was found to be less than the purity threshold which proved that the method produces interference-free peaks.

Specificity. The aim of the specificity study is to assess unequivocally the analyte in the presence of components that may be expected to be present. The retention times of the peaks of interest were noted, also the interferences of the placebo, the known impurities and the degradation products were assessed. It showed that the method was interference-free and all the peaks were pure and resolved from each other, which complied with the acceptance criteria.

Precision.
Instrument precision: (Suitability of system). The aim of confirming the system suitability is to ensure adequate analysis. %RSD, theoretical plates, tailing factor and resolution were calculated as system suitability parameters for replicate injections of the reference solution and a single injection of the resolution solution (Fig. 2,3). The results obtained are shown in Table 3. The parameters all complied with the acceptance criteria and system suitability was established.

Fig. 2 Chromatogram of the resolution solution.

Fig. 3 Chromatogram of the reference solution.

Method precision: (Repeatability). The method precision study shows the repeatability of the results obtained by the testing method. %RSD was determined for all the six sets of sample preparations, which should not be more than 2.0% for latanoprost, timolol and BKC content, and not more than 5.0% and 10.0% for impurities. The results showed confidence in the method by a very low %RSD of 0.1–0.9% for the assay method and 0.5%–2.0% for impurities, which showed that the method is precise for the intended purpose.
Intermediate precision: (Reproducibility). The purpose of this study is to demonstrate the reliability of the test results with variations. The maximum difference observed in precision and intermediate precision was 0.8% for the assay of drug substances and 0.03% for known and unknown impurities, which was very close to the method precision study results. For the assay the allowable % variation was 2.0% and all the results were within the acceptance criteria, so this study proved the method to be rugged enough for day to day use.

Stability of analytical solutions. The difference in both drug substances and preservative was compared with an initial sample at each respective time interval of 24 h and 48 h on the bench top (25 °C ± 5 °C), which showed a maximum difference of 0.8% in the results of the assay. The results complied with the acceptance criteria of 2.0% maximum difference. For individual impurities the maximum difference observed was 0.04%; for total impurities the difference in result after 48 h was not more than 0.06%, which showed that the sample solution is stable for 48 h on the bench top.

Limit of detection (LOD) and limit of quantification (LOQ). The derived LOD and LOQ for latanoprost, timolol, latanoprost acid, 15-keto latanoprost and timolol impurity I were of very low level as mentioned in Table 5. The reproducibility of LOD and LOQ concentrations was checked by performing replicate injections of these solutions, in which the %RSD was determined to be less than 10.0% for LOQ concentrations, and at the LOD concentration peaks were detected in replicate injections.

Table 5 LOD and LOQ concentrations

Name	Limit of detection		Limit of quantification
	Concentration/μg ml^−1	% with respect to test concentration	Concentration/μg ml^−1	% with respect to test concentration
15-Keto latanoprost	0.022	0.04	0.068	0.14
Latanoprost acid	0.010	0.02	0.030	0.06
Latanoprost (unknown impurity)	0.006	0.01	0.018	0.04
Timolol impurity I	0.004	0.004	0.011	0.01
Timolol maleate equivalent to timolol (unknown impurity)	0.004	0.004	0.013	0.01

---

Linearity study. The linearity of an analytical method is its ability to elicit test results that are directly, or by a well-defined mathematical transformation, proportional to the concentration of analyte in samples within a given range. The acceptance criteria for the correlation coefficient, set for the assay of drug substances, was 0.9990 for drug substances and 0.9800 for known and unknown impurities. The correlation coefficient observed for latanoprost, timolol and BKC was 0.9998, 0.9990 and 1.0000 respectively, and for latanoprost acid, 15-keto latanoprost and timolol impurity I it was 0.9997, 0.9975 and 1.0000 respectively. Also, the square of the correlation coefficient, slope of regression, RSD of the response factor, Y-intercept and bias at 100% level were calculated for each analyte and the impurities, which were within the acceptance criteria. To show the linearity of the concentration range the precision and accuracy were additionally performed at lower concentrations and higher concentrations. On the basis of the above observations, the method was found to be linear over the defined concentration range.

Accuracy (by recovery). The accuracy of an analytical method is the closeness of test results obtained by that method compared with the true values. The accuracy of an analytical method should be established across its range. Accuracy was performed in the range of LOQ to 150.0% (LOQ, 50.0%, 100.0% and 150.0%) of the target concentration of latanoprost, timolol and their known and unknown impurities. Triplicate sets of accuracy sample preparations at each accuracy level were prepared and injected into a liquid chromatography column and chromatograms were recorded. The amount recovered for latanoprost, timolol and BKC was nearly 100.0% and for impurities it was well within the set acceptance criteria. This proved that the method is accurate for the determination of assay and related substances of latanoprost, timolol, BKC and impurities, and analytes which are added are practically recovered with minimum difference. The data of the accuracy study are shown in Table 6.

Table 6 Data of accuracy study

Name	Single maximum impurity (latanoprost)	15-Keto latanoprost	Latanoprost acid	Single maximum impurity (timolol)	Timolol impurity I
Average recovery – LOQ	108.4%	106.4%	105.5%	121.4%	122.2%
Average Recovery – 150.0%	101.4%	106.0%	101.6%	109.9%	97.5%
Acceptance criteria	[<0.05%; 70.0–130.0%], [0.051–0.100; 75.0–125.0%]
	[0.101–0.500; 80.0–120.0%], [0.501–0.100; 85.0–115.0%]
	[>1.000; 90.0–110.0%]

---

Name	Latanoprost	Timolol	BKC
Average recovery – 50.0%	99.2%	100.1%	100.2%
Average recovery – 150.0%	101.5%	101.4%	99.1%
Acceptance criteria	Recovery should be within 98.0%–102.0%

---

Robustness. The robustness of an analytical procedure is a measure of its capacity to remain unaffected by small, but deliberate changes in the method parameters and provides an indication of its reliability during normal usage. As mentioned, all the small changes, the system suitability criteria of resolution, tailing factor, theoretical plates and %RSD were achieved within the acceptance criteria, which showed the method's robustness.

Conclusion

A challenging, versatile HPLC method was developed for the simultaneous determination of latanoprost, timolol, BKC and related substances in an ophthalmic solution. The method was very simple and effective, and based on the validation data it can be concluded that within an analysis time of 55 min, two drugs, one preservative, three known impurities and other unknown related substances were determined accurately and precisely. The method was able to detect very low concentrations of analyte and impurities, which is very useful for doing stability analysis of formulations. The method was also found to be specific and free from the interference of diluent, impurities, placebo and other degradation products. The method was also found to be linear and accurate over concentrations ranging from the LOQ to 150.0% of target concentrations. Furthermore the method was unaffected by variations in day to day analysis; hence it is rugged and robust. On the basis of the above inferences, it can be concluded that the method can be implemented for quality control purposes as well as the stability analysis of the said formulation.

Acknowledgements

All raw data from the validation work is archived at Cadila Pharmaceutical Ltd. All the validation work was performed at Analytical Research Laboratory, Cadila Pharmaceutical Ltd, Dholka, Gujarat, India. The author is thankful to M/s Cadila Pharmaceuticals Ltd, Dholka, India for providing facilities and infrastructure for the study.

References

1. ICH guideline Q3A (R2), Impurities in New Drug Substances, October 25, 2006.
2. ICH guideline Q7A, Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients, November 2005.
3. A. Van de Water, R. Xhonneux, R. Reneman and P. Janssen, Eur. J. Pharmacol., 1988, 156, 95–103.
4. http://www.cipladoc.com/therapeutic/pdf_cipla/latim_eye_drops.pdf .
5. http://en.wikipedia.org/wiki/Quaternary_ammonium_compound .
6. J. Mehta, K. Patidar and N. Vyas, E-J. Chem., 2010, 7(1), 11–20.
7. Drug master file (applicant's part), Chirogate international inc., 2008, pp. 27–28.
8. Drug master file (applicant's part), FDC limited., 2007, pp. 1–14.
9. Indian Pharmacopoeia, part 3, 2007, pp. 1808–1810.
10. US Pharmacopoeia, USP-32 NF-27, 2009, pp. 3748–3749.
11. British Pharmacopoeia, online version, 2010, Monograph No.: 0572.
12. M. Ashfaq, I. U. Khan and M. N. Asghar, Anal. Lett., 2006, 39(11), 2235–2242.
13. HPLC methods for recently approved pharmaceuticals, George Lunn, John Wiley & Sons Inc., New Jersey, 2005, Page 339–340.
14. P. Widomski, P. Baran, P. Gołębiewski, I. Surowiec and A. Bielejewska, Acta Chromatogr., 2008, 20(2), 157–164.
15. R. Varma, J. Winarko, T. Kiat-Winarko and B. Winarko, Invest. Ophthalmol. Vis. Sci., 2006, 47, 222–225.
16. P. V. Morgan, S. Proniuk, J. Blanchard and R. J. Noecker, J. Glaucoma, 2001, 10(5), 401–405.
17. P. M. Lacroix, B. A. Dawson, R. W. Sears and D. B. Black, Chirality, 1994, 6, 484–491.
18. N. Erk, Pharmazie, 2003, 58(7), 491–493.
19. M. H. Türkdemir, G. Erdögdu, T. Aydemir, A. A. Karagözler and A. E. Karagözler, J. Anal. Chem., 2001, 56(11), 1047–1050.
20. S. P. Kulkarni and P. D. Amin, J. Pharm. Biomed. Anal., 2000, 23(6), 983–987.
21. HPLC methods for recently approved Pharmaceuticals,George Lunn, 2005, Page 171 and 339.
22. US Food and Drug Administration ( 2001) Guidance for industry: analytical procedures and methods validation: chemistry, manufacturing, and controls documentation. Draft Guidance. US Food and Drug Administration.

---