A rapid and selective UPLC-DAD method was developed and validated for simultaneous analysis of the novel two-drug combination
Infection due to hepatitis C virus (HCV) is a leading cause for severe chronic liver disease, which can result in progressive liver damage such as cirrhosis and hepatocellular carcinoma. Thus, it is considered to be a great worldwide health problem specifically in Egypt, which has the greatest prevalence of the epidemic problem of HCV in the world in accordance with the reported Egyptian Demographic Health Survey [EDHS] that had reached 14.7%. So prevention of HCV becomes a national priority [
The available treatment options for HCV infection until 2011 were restricted to ribavirin with pegylated interferon combination. This drug regimen has limited efficacy, especially in genotype 1 infected patients, and was also accompanied with dangerous side effects [
In 2014, the directly acting antivirals were introduced in the market as a new anti-HCV generation. The main goal of these new drug therapies is decreasing the incidence of possible side effects for HCV patients. These powerful drugs encompass Nonnucleoside Inhibitors (NNIs) and Nucleoside Inhibitors (NIs) of HCV RNA polymerase (NS5A/5B) and Protease Inhibitors (PIs). As shown in Figure
Chemical structure of SF (a), DC (b), and IS (c).
Daclatasvir (DC) is the first discovered drug as HCV RNA polymerase NS5A replication complex inhibitor [
The increase in number of therapeutic options available for patients suffering HCV infections introduces a great challenge in the selection and management of HCV treatment. In regard to this, Therapeutic Drug Monitoring (TDM) is considered to be an important tool to assess the efficacy of drug regimen, help the clinicians to adjust the drug dosage, optimize the therapy or switch the treatment regimen, and overcome adverse events or therapeutic failure [
Due to the extensive use of the mentioned drugs in combination therapy, the therapeutic drug monitoring of concentrations of SF and DC concentrations in patients undergoing HCV therapy is very important and critical, especially in determining the treatment optimization and the potential of drug interaction.
Therefore, there were several reported methods for SF and/or DC determination in human plasma using HPLC-UV detection [
To the best of our knowledge, the adopted bioanalytical approach in this work will be the solely UPLC-PDA technique first developed for the simultaneous quantification of SF and DC in human plasma with a clinical application to their therapeutic drug monitoring after the oral intake of a coformulated tablet containing SF+DC (Darvoni® recently manufactured by Beacon Pharmaceuticals Limited) instead of the commonly two tablets dosage regimen of each one alone (Sovaldi® + Daklinza®). In accordance with US Food and Drug Administration (FDA) guidelines [
Drug standards for SF, DC, and Ledipasvir (IS) were kindly supplied by Memphis Co. for Pharmaceutical and Chemical Industries, Cairo, Egypt, with certified purity of 99.98 ± 0.421 for SF, 99.93 ± 0.231 for DC, and 99.87 ± 0.642 for Ledipasvir. Darvoni film-coated tablets (coformualted with 400 mg SF and 60 mg DC) were purchased from Beacon Pharmaceuticals Limited, Bangladesh. Drug-free human plasma was obtained from Kuwait Blood Bank, Al Jabriyah, Kuwait. HPLC grade acetonitrile and other used chemicals in the adopted method were of analytical grade and obtained from Sigma Aldrich, Dor-set, UK. “In house” HPLC grade water was prepared with a MilliQfilter purchased from Millipore, Watford, UK. Syringe membrane filters (13mm) were purchased from kinesis scientific expert, Cambridgeshire, UK. Nylon solvent filters (0.45 um) used for solvent filtration and Water 20-positions Extraction Manifold with SPE cartridges (Sep-Pak® Vac C18) used for sample preparation were purchased from Water Corporation, Milford, USA. SPE eluates were dried using DRI-BLOCK DB-3 evaporator which was purchased from Techne, Stone, UK.
Chromatographic separation was achieved using Waters® Acquity UPLC separation module with quaternary Solvent Manager (H-Class), online degasser with autosampler injector, and photodiode array detector coupled with Empower® software for data acquisition (Waters®, Milford, USA). Waters® Acquity UPLC BEH C18 column (2.1 mm × 50 mm, 1.7
Due to the possible degradation of DC in solution at high-intensity UV and visible light as previously reported by [
Quality control samples for SF and DC.
Prepared QC samples (ng.mL−1) | SF | DC |
---|---|---|
LLOQ | 25 | 50 |
QL | 50 | 100 |
QM | 400 | 1600 |
QH | 3200 | 6400 |
ULOQ | 6400 | 12800 |
Sample clean-up was conducted using the following SPE procedure: The cartridges were first preconditioned with acetonitrile (1 mL) then equilibrated by water (1 mL), in a positive pressure manifold. A volume of 450
The assay validation was carried out in accordance with guidance for bioanalytical method validation recommended by the FDA [
The potential interferences from endogenous matrix components were investigated by evaluating ten different lots of human plasma as blank and at the LLOQ level of the spiked SF and DC. Drug-free plasma samples chromatograms were compared with those of the spiked plasma to ensure the absence of analytical interferences from endogenous substances present in plasma samples.
Eight-point calibration standard curves of SF and DC in plasma, ranging from 25 to 6400 ng.mL−1 and 50 to 12800 ng.mL−1, respectively, were prepared in triplicate for each run. The LLOQ is the lowest concentration in the standard calibration curve that back-calculates with good precision that does not exceed 20% of the CV and satisfactory accuracy which does not exceed 20% of the nominal concentration.
Accuracy, intraday precision, and interday precision values were determined by the analysis of spiked human plasma samples with five different concentrations for each of SF and DC, corresponding to the LLOQ, low, medium, high QC samples, and ULOQ three times on the same day and on three separate days. Accuracy was expressed as the % of deviation between the nominal and measured concentration (% error). Precision was calculated as coefficient of variation % (CV).
The overall recovery of SF and DC from spiked human plasma was determined at the LLOQ, low, medium, high QC samples, and ULOQ. The ratio of the peak area response of extracted QC samples for SF or DC to that of the IS was compared to that of unextracted standards obtained by injecting the corresponding concentration of DC and IS in the mobile phase and analyzed in triplicate. The extraction recovery was computed as previously published [
Stability tests were carried out under various conditions simulating those that a real human plasma samples may be subjected to during routine analysis. Stability studies of SF and DC in human plasma included the following:
(a) Freeze and thaw stability after three freeze-thaw cycles of stored plasma samples; (b) bench-top stability of plasma samples after storage at RT for 48 h; (c) long-term stability of plasma samples after storage at−80
For each of the previously mentioned conditions, three series of LLOQ, LQ, MQ, HQ, and ULOQ spiked plasma samples were analyzed. The SF and DC concentrations in the analyzed plasma samples were compared to freshly made QC samples. For all of the previously mentioned stability studies, SF and DC in plasma samples were considered as stable if the stability sample results were within 15% of nominal concentrations according to the FDA guidelines for bioanalytical method validation [
The chromatographic separation of SF and DC was carried out using different HPLC columns and various mobile phases. The proper chromatographic separation was achieved on Waters® Acquity UPLC BEH C18 column (2.1 × 50 mm, 1.7
Typical UPLC chromatograms of (a) drug-free human plasma; (b) blank plasma spiked with SF and DC at LLOQ, (c) blank plasma spiked with SF and DC at ULOQ, and (d) plasma sample from a volunteer 0.5 h after administration of Darvoni® (400 mg of SF/60 mg of DC).
A satisfactory drug recovery in sample extraction is crucial for the simultaneous estimation of SF and DC in human plasma at low concentration levels. Various extraction procedures were tried such as liquid–liquid extraction with different solvents and protein precipitation techniques, but they gave very low drug recovery which lead to significant interference from plasma peaks background. Deproteinization using acetonitrile was carried out due to the fact that SF and DC are ~ 61-65% [
Plasma sample extraction and chromatographic separation procedures were carried out to obtain a selective simultaneous determination for SF and DC. Various lots of blank human plasma from different sources were carefully evaluated for interference from endogenous matrix components. A typical chromatogram of blank human plasma is illustrated in Figure
Two calibration standard curves of SF and DC in plasma, ranging from 25 to 6400 ng.mL−1 and 50 to 12800 ng.mL−1, respectively, were prepared in triplicate for each run. The calibration plots showed linearity over the previously mentioned concentration ranges and the determination coefficient (r2) was not less than 0.998. The LLOQ is the lowest concentration in the standard calibration curve that back-calculates with good precision that does not exceed 20% of the CV and good accuracy within 20% of the nominal concentration. The LLOQ of SF and DC was 25 and 50 ng.mL−1, respectively. A typical chromatogram for a spiked plasma sample containing the LLOQ for both SF and DC is presented in Figure
Data results for accuracy (expressed as % bias = [measured concentration – nominal concentration] / nominal concentration × 100) and precision (expressed as % CV) presented in Table
Accuracy, intraday, and interday precision and % extraction recovery for SF and DC in their QC samples in human plasma (n=6)
Nominal Concentration | | ||||
---|---|---|---|---|---|
Average of the measured concentration | Accuracy | Intra-day precision | Inter-day precision | % Extraction recovery ± %CV | |
25 | 24 | -4.0 | 9.6 | 8.6 | 96.0 ± 1.2 |
50 | 45 | -10.0 | 4.6 | 5.1 | 90.0 ± 3.5 |
400 | 425 | 6.3 | 5.2 | 9.3 | 106.3 ± 1.3 |
3200 | 3430 | 7.2 | 3.9 | 7.4 | 107.2 ± 2.8 |
6400 | 6123 | -4.3 | 3.8 | 6.1 | 95.7 ± 1.5 |
| |||||
Nominal Concentration | | ||||
Average of the measured concentration | Accuracy | Intra-day precision | Inter-day precision | % Extraction recovery ± %CV | |
| |||||
50 | 54 | 8.0 | 6.5 | 3.7 | 108.0 ± 2.3 |
100 | 95 | -5.0 | 9.2 | 9.1 | 95.0 ± 1.7 |
1600 | 1490 | -6.9 | 4.7 | 7.5 | 93.1 ± 1.1 |
6400 | 6690 | 4.5 | 5.4 | 6.4 | 104.5 ± 2.0 |
12800 | 12357 | -3.5 | 2.8 | 8.2 | 96.5 ± 3.6 |
Accuracy, intraday, and interday precision and % extraction recovery for SF and DC in their QC samples in human plasma (n=6)
Nominal Concentration | | ||||
---|---|---|---|---|---|
Average of the measured concentration | Accuracy | Intra-day precision | Inter-day precision | % Extraction recovery ± %CV | |
25 | 24 | -4.0 | 9.6 | 8.6 | 96.0 ± 1.2 |
50 | 45 | -10.0 | 4.6 | 5.1 | 90.0 ± 3.5 |
400 | 425 | 6.3 | 5.2 | 9.3 | 106.3 ± 1.3 |
3200 | 3430 | 7.2 | 3.9 | 7.4 | 107.2 ± 2.8 |
6400 | 6123 | -4.3 | 3.8 | 6.1 | 95.7 ± 1.5 |
| |||||
Nominal Concentration | | ||||
Average of the measured concentration | Accuracy | Intra-day precision | Inter-day precision | % Extraction recovery ± %CV | |
| |||||
50 | 54 | 8.0 | 6.5 | 3.7 | 108.0 ± 2.3 |
100 | 95 | -5.0 | 9.2 | 9.1 | 95.0 ± 1.7 |
1600 | 1490 | -6.9 | 4.7 | 7.5 | 93.1 ± 1.1 |
6400 | 6690 | 4.5 | 5.4 | 6.4 | 104.5 ± 2.0 |
12800 | 12357 | -3.5 | 2.8 | 8.2 | 96.5 ± 3.6 |
The average of % extraction recovery for SF and DC in their respective concentration ranges varied from 90.0 to 107.2 % and 93.1 to 108.0 % for SF and DC, respectively, with % CV values ranging from 1.2 to 3.5 % and 1.1 to 3.6 % for SF and DC, respectively, as presented in Table
The stability of SF and DC in human plasma samples at five levels (LLOQ, LQ, MQ, HQ, and ULOQ) was investigated under the following conditions that the common and routine clinical samples are usually subjected to [
Stability of SF and DC in plasma samples under several storage conditions.
Nominal Concentration (ng.ml−1) | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
| | ||||||||||
25 | 50 | 400 | 3200 | 6400 | 50 | 100 | 1600 | 6400 | 12800 | ||
(a) After three freeze/thaw cycles | % Bias | -2.9 | 5.2 | -5.4 | 4.2 | 6.1 | 8.5 | -3.5 | -5.8 | -3.1 | -9.4 |
| |||||||||||
(b) short-term storage (48 h) at RT | % Bias | 4.1 | 4.1 | 7.3 | -5.7 | -4.8 | 5.1 | 6.4 | 9.2 | 8.4 | -5.0 |
| |||||||||||
(c) long-term storage at −80°C for 6 months | % Bias | 3.5 | -3.7 | 3.4 | 5.1 | 7.3 | 8.4 | 9.4 | -2.0 | -5.7 | -6.3 |
| |||||||||||
(d) Standard solutions kept in refrigerator up to 10 days | % Bias | -2.5 | 6.5 | 3.8 | 2.8 | 4.1 | 7.3 | -5.7 | -3.9 | -2.8 | 1.7 |
| |||||||||||
(e) Solutions kept on bench are stable for 5 days | % Bias | 3.9 | 6.3 | 8.2 | 9.0 | 1.8 | 4.7 | 3.9 | 2.8 | -3.7 | -4.5 |
| |||||||||||
(f) Frozen plasma samples which undergo heating process (60°C for 60 min) | % Bias | -6.7 | -5.7 | 3.7 | 4.1 | 5.9 | 8.6 | 4.7 | 8.1 | -9.4 | -7.2 |
| |||||||||||
(g) Dried extracts (after SPE procedure) kept at −20°C for 6 days | % Bias | 9.5 | 4.7 | 5.8 | -9.3 | -6.3 | -7.8 | 9.5 | -5.2 | -6.0 | -3.7 |
| |||||||||||
(h) Reconstituted extracts in the mobile phase kept at 4°C for 4 days in the auto sampler | % Bias | -5.7 | -4.0 | -5.2 | 4.9 | 8.1 | 7.2 | 7.6 | 6.1 | 8.5 | 5.3 |
Concentrations of SF and DC were estimated using the adopted UPLC-PDA method in the plasma samples obtained from three healthy male volunteers after the oral intake of one Darvoni® tablet coformulated with 400 mg of SF and 60 mg of DC. To confirm the clinical application of the suggested method, a typical UPLC chromatogram of a plasma sample obtained from one of the volunteers after 0.5 h from the oral intake of one Darvoni® tablet is presented in Figure
A rapid, accurate, precise, sensitive, and selective UPLC-PDA method was developed for simultaneous quantification of SF and DC in plasma samples has been adopted in this work for the first time. The developed method had been validated according to FDA guidelines for bioanalytical method validation. The method applicability was confirmed by the analysis of plasma samples of three healthy volunteers after the oral intake of coformulated Darvoni® tablet. So far, the previously published analytical methods developed for pharmacokinetic investigations of SF and DC were all based on costly LC-MS/MS equipment after the oral intake of two tablets for each drug, which decreased the feasibility of the routine analysis of SF and DC. The adopted UPLC-UV method is considered to be greatly applicable for the routine TDM of SF and DC in plasma at conventional clinical laboratories where LC-MS/MS equipment is not present. The suggested method will be suitable for standard clinical laboratories that do not possess LC-MS/MS equipment. With respect to large-scale pharmacokinetic analysis, the proposed method will be a satisfactory, simple, cheap, fast, and easier to set up alternative UPLC chromatographic method coupled with DAD. Stability studies of SF and DC under various common conditions to which both drugs may be subjected to during sample handling and analysis through routine TDM process illustrated that both drug concentrations remained approximately unchanged in plasma, in their stock solutions, and in processed plasma samples under different storage conditions. All in all, the adopted UPLC-DAD technique and the data results of the stability studies analyses could be valuable for dosing both drugs and appropriately dealing with their plasma samples both in routine clinical application and in TDM. This clinical application gave the opportunity to the optimization of the drug treatment which can improve the life quality and increase the therapy efficacy itself. It can lead also to a cost saving outcome, reducing side effects, and consequently a better clinical cost for patient’s care.
The data used to support the findings of this study are included within the article.
The authors receive no specific grant from funding agencies in the public, commercial, or not-for-profit sectors to develop this study.
The authors declare that there are no conflicts of interest regarding the publication of this paper.