A Simple RP-HPLC Method to Simultaneously Assay the Contents of Lamivudine, Tenofovir, and Nevirapine in Fixed Dose Combined Oral Antiviral Medicines

An accurate and rapid reverse HPLC method has been developed and validated for the simultaneous quantification of lamivudine, nevirapine, and tenofovir disoproxil fumarate. Suitable separation was achieved on Phenomenex Synergi C18 (250× 4.6mm, 4μm) using mobile phase, methanol (50%): ammonium acetate buffer (adjusted to pH 2.80) (40%): acetonitrile (10%) in an isocratic mode. &e drugs were detected at 270 nm with a flow rate of 1.0ml/min, and the retention times were found to be 3.26, 5.42, and 7.55 minutes for lamivudine, nevirapine, and tenofovir disoproxil fumarate, respectively. &e developed method was validated per ICH guidelines. Good linearity was obtained within the concentration ranges of 10–59 μg/ml, 7–42 μg/ml, and 15–90 μg/ml with a correlation coefficient of not less than 0.990.&e%RSD values for precision (intraday and interday) and accuracy studies were found to be less than 2%.&e results obtained from quantitative analysis conform to USP content requirements for marketed tablet dosage forms, RICOVIR-LN, and tenofovir disoproxil fumarate/lamivudine tablets. &e method is therefore useful for routine quality control of antiretroviral tablet dosage forms containing tenofovir disoproxil fumarate, lamivudine, and nevirapine.


Introduction
HIV/AIDS is a major public health issue and as such forms a significant part of the Sustainable Development Goals, with the aim of ensuring healthy lives and promoting well-being for all at all ages [1]. e introduction of highly active antiretroviral therapy (HAART), a treatment regimen comprising of the combination of three or more antiretroviral drugs [2], has revolutionized the management of the disease condition, with the resultant dramatic reduction in mortality rates and the incidences of opportunistic infections [3]. In Ghana and other parts of the world, three of the widely used antiretrovirals (ARVs) in HAART include lamivudine (3TC), nevirapine (NVP), and tenofovir disoproxil fumarate (TDF). It may be argued that a lasting impact on morbidity and mortality on people living with HIV/AIDS (PLWHA) could be assured if the quality of ARVs was adequately monitored. With recent reports of substandard medicines available in Ghanaian health facilities [4], it has become particularly important to increase surveillance on the quality of all medicines procured for public health facilities, especially for antiretrovirals, used for a significant public health intervention programme. is therefore calls for the development and validation of reliable analytical methods to achieve such quality control purposes.
In the past, most of the methods developed had focused on the detection and quantitation of ARVs in biological samples [5,6]; just a handful had concentrated on methods for the oral dosage formulations [7][8][9]. In as much as these methods have provided useful tools to assess the qualities of some ARVs, it is necessary that very efficient alternative methods with the target scope of ARVs in the HAART programme be developed to achieve similar quality control outputs.
e aim of the current study therefore was to develop a simple, affordable, and reliable method for the quality assessment of ARVs used in the HAART programme in Ghana and other parts of the world, where applicable. us, the method when developed would be used for the analyses of 3TC, NVP, and TDF in monotherapies and fixeddose combination products as well.

Chemicals and Reagents.
e reagents and solvents used to prepare the samples and carry out the analyses were of analytical and HPLC grades, respectively. Solvents including acetonitrile and methanol were procured from Fisher Scientific, United Kingdom. Ammonium acetate and glacial acetic acid purchased from VWR International Limited and Merck House, respectively, were also used for the analysis. Purified water was freshly produced in-house, terminally sterilized with ultraviolet radiation, and filtered through a 0.45 µm membrane filter before being used to prepare all solutions and buffers.

Equipment and Instrument.
e HPLC system used comprised of Shimadzu prominence UFLC series system, consisting of LC-20A quaternary pump (part G1311 A), DGU-20A 5 in-line vacuum degasser (part no. G1322A), and SPD-20A ultraviolet detector. Data acquisition was performed by LC solutions software (version A. 10.02 Build 1757). e chromatographic separation was carried out using on a C18 Phenomenex Synergi column (250 × 4.6 mm; 4 µm). An electronic analytical balance (Mettler Toledo, AB204-S/FACT), a digital pH meter (Mettler Toledo Seven Compact pH/Ion S220), and a sonicator were also used.

Buffer Preparation.
A 500 ml of 0.02 M ammonium acetate buffer was prepared by weighing a determined quantity of ammonium acetate powder, dissolving with some amount of distilled water, and transferred into a 500 ml volumetric flask. e pH was adjusted to 2.8 with acetic acid, and with continuous stirring, the solution was topped up with distilled water to the required volume. It was then filtered with a 0.45 µm membrane filter.

Mobile Phase/Diluent Preparation.
e solvent system used was a mixture of buffer (pH � 2.8), methanol, and acetonitrile in the ratio 40 : 50 : 10 (v/v), respectively. is solution was also used as the diluent to prepare solutions of the standards and samples.

Preparation of Standard Solutions.
A stock standard solution containing 75 mg of 3TC, 35 mg of NVP, and 50 mg of TDF was prepared in a 100 ml volumetric flask. e content in the flask was sonicated at 37°C for 10 minutes before topping up to the required volume with the diluent, after cooling. e resulting solution was filtered with a 0.45 µm membrane filter. 3 ml of the stock solution was pipetted to prepare 50 ml of working standard solution containing 45 µg/ml of 3TC, 21 µg/ml of NVP, and 30 µg/ml of TDF.

Preparation of Sample Solutions.
Solutions of powdered product samples A (containing TDF and 3TC) and B (containing two formulations, with B 1 containing TDF and 3TC and B 2 containing NVP) were prepared for analysis. For samples A and B 1 , a weight of each containing an equivalent of 75 mg of 3TC was weighed and transferred into a 100 ml volumetric flask. About 50 ml of methanol was added, sonicated for 10 minutes at 37°C, and made up to volume with the diluent. e resulting solution was then filtered through a Whatman No. 1 filter paper, discarding the first 5 ml of the filtrate. 3 ml of the resulting filtrate was pipetted and transferred into a 50 ml volumetric flask and made up to volume using the diluent. For sample B 2 , a weight containing an equivalent of 35 mg of NVP was transferred into a 100 ml volumetric flask and a similar procedure as described above was employed to prepare the sample B solution.

Development and Validation of Method.
e method for detection, separation, and quantitation of the three antiretroviral drug substances in the product was developed empirically (S1), and validated in accordance with the International Council for Harmonisation (ICH) Q2(R) guidelines [12]. e validation parameters investigated included linearity and its range, limits of detection and quantitation, specificity, stability of test solution, robustness, accuracy, and precision.
2.5.1. Specificity/Selectivity. Specificity and selectivity were assessed by comparing the chromatograms from a matrix without expected analytes (blank sample) with that of matrices containing the expected analytes (that is, 3TC, NVP, and TDF) [13,14]. e selectivity/specificity was confirmed by comparing mean ± SD of the retention times for the analytes using one-way ANOVA followed by Tukey's post hoc test [15].

Precision.
Precision was demonstrated by determining repeatability (intraday precision) and intermediate precision of the method. Intraday precision was confirmed Practically insoluble in water, sparingly soluble or slightly soluble in methylene chloride, slightly soluble in methanol from the peak areas of the analytes, from triplicate injections of three dilute concentrations of the standard solution within the same day (that is, 15 µg/ml, 45 µg/ml, and 90 µg/ ml for 3TC; 7 µg/ml, 21 µg/ml, and 42 µg/ml for NVP; and 10 µg/ml, 30 µg/ml, and 59 µg/ml for TDF). Intermediate precision, on the other hand, involved studying the variation in response at 100% concentration of the working standard on three different days. e results obtained were statistically analysed by determining the relative standard deviations and carrying out ANOVA, where applicable [12].

Linearity and Range.
e linearity of the developed method was investigated by injecting six concentrations prepared from the stock standard solution, to obtain concentrations for 3TC, NVP, and TDF within the ranges of 15 µg/ml-90 µg/ml, 7 µg/ml-42 µg/ml, and 10 µg/ml-59 µg/ ml, respectively [12]. Triplicate determinations were carried out for each test concentration and the peak areas, reported as mean ± SD were plotted against test concentrations. Statistical analysis was performed by the least-squares method [13]. Linearity was predicted by estimating the regression coefficient (R 2 ) and the linear regression y-intercept of the response versus concentration plot. e regression model was also tested for fitness by determining the level of significance of the F-value of the model in an ANOVA at 5% risk level [13]. Additionally, the residual plots for the sets of test data were generated.

Limit of Detection (LOD) and Limit of Quantitation (LOQ).
e limits of detection and quantification were determined from the intercept on the y-axis slope from the linear regression model derived from the linearity test [12]. e formulae below were used for calculating the LOD and LOQ. e results are shown in Table 2. LOD � 3.3 × standard deviation of the response slope of the calibration curve , LOQ � 10 × standard deviation of the response slope of the calibration curve . (1) 2.5.6. Robustness. e robustness of the developed method was tested by monitoring the effects of deliberate changes in the flow rate (±0.1 ml/min) and the wavelength of detection (±2 nm) on the general performance of the developed method [13,18].

Stability of Test Solution.
e stability of the working standard solution was assessed over a 6-hour period at room temperature by monitoring the peak areas of the drug substances with time [13].

Analysis of Commercial Products.
e contents of 3TC, NVP, and TDF in samples A, B 1 , and B 2 as prepared above were assessed from peak areas from triplicate injections of the samples and applying the linear regression models obtained to the recorded peak areas.

Method Development.
e chromatographic method was developed empirically, guided by the physicochemical properties like acid dissociation constant (pKa) and partition coefficient (cLogP) ( Table 1). e conditions comprising suitable column, mobile phase composition, wavelength for analyte detection, flow rate, column temperature, and injection volume were determined empirically by monitoring the resolution, peak symmetry, and run time for the analytes (S1). e investigations resulted in the choice of a mobile phase system consisting of methanol (50%), ammonium acetate buffer pH of 2.8 (40%) and acetonitrile (10%), and a flow rate of 1 ml/min, among others (Table 3), which were then used to develop the chromatographic method ( Figure 1).

Method Validation.
e results from the validation are summarized in Table 2. In establishing the specificity and selectivity of the developed method, the chromatogram generated from the blank sample showed only noises, with no apparent peak observed within 0-20 minutes (S2). Upon injection of the working standard solutions (containing 3TC, NVP, and TDF), three resolved peaks were observed. It was further shown that the retention times for each of the analytes were different from each other (p < 0.0001; Table 2). e outcome showed that the method was capable of independently detecting the three analytes and distinguishing one from the other, thus specific and selective. In order to evaluate the quantifying power of the method, its accuracy was also determined. It was shown that the percentage recoveries obtained, complied with the acceptance criteria of 98%-102% [13,14] (with minimum deviations) over the concentration range, 80%-120% of test concentration for the three analytes (Tables 2 and S3). e responses for 3TC, NVP, and TDF were showed to be linear within the ranges 15 µg/ml-90 µg/ml, 7 µg/ml-42 µg/ml, and 10 µg/ml-59 µg/ml, respectively (Table 2 and Figure 2). e regression coefficient of correlation (R 2 ) obtained for the abovementioned analytes was 0.9959, 0.9948, and 0.9971, respectively. F-values from the linear regression models were also shown to be significant, further demonstrating strong correlations between the concentration of the analytes and peak area responses (Table 2). e method was shown to detect at least 5.5032 µg/ml, 3.1496 µg/ml, and 3.9267 µg/ml of 3TC, NVP, and TDF, respectively. However, quantitation could only be carried out with the method when their concentrations were at least 16.6762 µg/ml, 9.5443 µg/ml, and 10.0433 µg/ml, respectively (Table 2).
Method precision was demonstrated with repeatability and intermediate precision (Tables 2 and S4). It was observed in both evaluations that the RSDs for the method responses (peak areas) were less than 2%, proving their consistency and precision [12,13]. In addition, it was  Deliberate changes in the wavelength detection did not show significant differences in peak areas (RSD <2%; % deviation < 5%). In addition, proportionate detection was maintained (p > 0.005). Method is fairly robust. * Minimal deviations of results from acceptance criteria.    Journal of Chemistry observed from the intermediate precision determinations that the peak areas recorded on different days for the different concentrations adopted were not significantly different from each other (F (4,18) � 0.08900, p � 0.9847) ( Tables 2 and S4).

Journal of Chemistry
In testing for the robustness of the method, it was observed that deliberate changes in the flow rate did not significantly affect the resolution of the peaks, as they remained well separated from each other (F (2,6) � 60322; p < 0.0001 and F (2,6) � 1949; p < 0.0001). Upon changes effected to the wavelength of detection, the deviations observed were also not significant (that is, % deviations < ±5%) (Tables 2 and S5). us, the method was considered robust. e analytes in the test solution were also found to be stable within the period of analysis, which in the current study was estimated to be 6 hours ( Figure 3).

Assay of Sampled Products.
e validated method demonstrated its usability in the analysis of commercial products.
e outcome of such investigation showed that the products complied with content assay specifications as indicated in the United States Pharmacopeia [19] (Table 4 and Figure 4).

Conclusion
An accurate, precise, and selective reverse HPLC method for the simultaneous estimation of lamivudine, nevirapine, and tenofovir disoproxil fumarate has been developed and validated. e method was validated per the ICH guidelines and passed all tests for the various validation parameters. e developed method is specific and selective as well as robust, accurate, and precise. e HPLC method was successfully applied to quantitatively estimate the active pharmaceutical ingredients in commercial products A 1 (containing TDF 300 mg/ 3TC 300 mg tablets + NVP 200 mg) and A 2 (containing TDF 300 mg/ 3TC 300 mg).

Data Availability
Processed data used to support the findings of this study are included within the article. Some of the data related to the validation of the developed method are also included within the article while others are included as Supplementary Data (S1-S5).

Conflicts of Interest
e authors declare that they have no competing interests. Acknowledgments e authors are grateful to Danadams Pharmaceutical Industry Limited, Spintex, Accra, for the donation of the pure antiretroviral APIs used in this research. e authors are also thankful to Mr. Brown of Ghana Standards Authority for providing HPLC instrument and all reagents used.