key: cord-0873873-p0pe0ojh
authors: Habler, Katharina; Brügel, Mathias; Teupser, Daniel; Liebchen, Uwe; Scharf, Christina; Schönermarck, Ulf; Vogeser, Michael; Paal, Michael
title: Simultaneous quantification of seven repurposed COVID-19 drugs remdesivir (plus metabolite GS-441524), chloroquine, hydroxychloroquine, lopinavir, ritonavir, favipiravir and azithromycin by a two-dimensional isotope dilution LC-MS/MS method in human serum
date: 2021-01-28
journal: J Pharm Biomed Anal
DOI: 10.1016/j.jpba.2021.113935
sha: db8c5aa423147de943ca3f689e2b4b4e9a2f39f9
doc_id: 873873
cord_uid: p0pe0ojh

BACKGROUND: The present COVID-19 pandemic has prompted worldwide repurposing of drugs. The aim of the present work was to develop and validate a two-dimensional isotope-dilution liquid chromatrography tandem mass spectrometry (ID-LC-MS/MS) method for accurate quantification of remdesivir and its active metabolite GS-441524, chloroquine, hydroxychloroquine, lopinavir, ritonavir, favipiravir and azithromycin in serum; drugs that have gained attention for repurposing in the treatment of COVID-19. METHODS: Following protein precipitation, samples were separated with a two-dimensional ultra high performance liquid chromatography (2D-UHPLC) setup, consisting of an online solid phase extraction (SPE) coupled to an analytical column. For quantification, stable isotope-labelled analogues were used as internal standards for all analytes. The method was validated on the basis of the European Medicines Agency bioanalytical method validation protocol. RESULTS: Detuning of lopinavir and ritonavir allowed simultaneous quantification of all analytes with different concentration ranges and sensitivity with a uniform injection volume of 5 µL. The method provided robust validation results with inaccuracy and imprecision values of ≤ 9.59 % and ≤ 11.1 % for all quality controls. CONCLUSION: The presented method is suitable for accurate and simultaneous quantification of remdesivir, its metabolite GS-441525, chloroquine, hydroxychloroquine, lopinavir, ritonavir, favipiravir and azithromycin in human serum. The quantitative assay may be an efficient tool for the therapeutic drug monitoring of these potential drug candidates in COVID-19 patients in order to increase treatment efficacy and safety.

The present coronavirus disease 2019 (COVID- 19) pandemic is a global challenge for health care systems. Most people infected with SARS-CoV-2, the virus that causes COVID-19, develop mild to moderate symptoms and recover on their own. Some patients with COVID-19, however, develop severe symptoms and require hospitalization up to the point of intensive care therapy.

Corticosteroid therapy with dexamethasone proved to have some beneficial effect on severely ill COVID-19 patients [1] . So far, no other drugs have been shown to be effective for specific treatment of COVID-19 and optimized supportive care remains the mainstay of therapy. In view of the high urgency, considerable attention has been focused on the repurposing of drugs with in vitro antiviral activity against SARS-CoV-2. Several substances are considered as potential drug candidates and therefore subject to ongoing clinical research [2] .

Of particular note is investigational COVID-19 drug remdesivir, a promising broadspectrum antiviral agent originally developed to treat hepatitis C, where special approvals have been granted for the treatment of severe illness from COVID-19 [3] .

Other potential drug candidates are the antiviral substance favipiravir that is used to treat influenza [4] , and fixed-dose combination medication of the antiviral lopinavir and its pharmacological booster ritonavir (potent CYP3A4 inhibitor) for the prevention and treatment of HIV [5] . Azithromycin, a broad-spectrum antibiotic with antiviral and immunomodulatory properties that is primarily eliminated as an unchanged drug, also represents an interesting candidate in the search for COVID-19 drug therapy [6] .

Chloroquine und hydroxychloroquine that are used for the treatment of malaria and autoimmune disorders such as systemic lupus erythematosus were temporarily withdrawn for treatment of COVID-19 given that large-scale trials have failed to show J o u r n a l P r e -p r o o f any survival benefit [7] . However, due to the usage of unreliable data in some studies the WHO resumed the hydroxychloroquine arm of COVID-19 Solidarity trial.

Typically, the dosage of repurposed COVID-19 therapeutics is derived from in vitro generated half maximum effective concentration (EC50) values for SARS-CoV-2 and pharmacology-based pharmacokinetic models previously developed in other diseases and clinical conditions. However, corresponding dosage regimens will not necessarily translate into adequate drug exposure in critically ill COVID-19 patients due to pathophysiological alterations [8] . It is conceivable that dosage regimens of repurposed drugs may lead to subtherapeutic or toxic concentrations without clear clinical benefit. Accordingly, there is an urgent need to generate high-quality pharmacokinetic data for drug repurposing against COVID-19 [9] . In addition, polytherapy can result in unpredictable drug levels due to relevant drug-drug interactions. Consideration should therefore be given to an individualized optimal dosing strategy, where therapeutic drug monitoring (TDM) can further contribute to establish efficacy and safety. Different quantitative methods have been described in the literature for analysis of individual components or part of the above-mentioned repurposed COVID-19 drugs.

Recently, an UHPLC tandem mass spectrometry (MS/MS) assay was published for the simultaneous quantification of the prodrug remdesivir and its active metabolite GS-441524 [10] that exhibits a considerably longer elimination half-life (t½) of approximately 24 hours [11] . Several assays are described for the quantification of repurposed COVID19 drugs, such as HPLC with a diode-array detector (DAD) and LC-MS/MS for chloroquine [12, 13] , LC-MS/MS for hydroxychloroquine [14, 15] , HPLC coupled to UV detection and LC-MS/MS for lopinavir, ritonavir [16, 17] , favipiravir [18, 19] and LC-MS/MS for azithromycin [20, 21] . Especially, isotope dilution liquid J o u r n a l P r e -p r o o f chromatography tandem mass spectrometry (ID-LC-MS/MS) with application of stable isotope labelled analogues as internal standard is considered the gold standard in TDM [22] . However, to our knowledge no multi-analyte isotope dilution ID-LC-MS/MS method provides simultaneous quantification of remdesivir, and its metabolite GS-441524, chloroquine, hydroxychloroquine, lopinavir, ritonavir, favipiravir and azithromycin in a single analytic run with two different separation stages.

Two-dimensional liquid chromatography (2D-LC), where the eluate of a first column (the first dimension) is transferred to a second column (the second dimension), is becoming more widely adopted in the analysis of pharmaceutical compounds, especially when dealing with complex samples [23] . Therefore, the main objective of the present study was to establish a quantitative multi-analyte ID-LC-MS/MS method for these pharmaceuticals in human serum. To allow a high sample throughput with minimized matrix effects, we aimed to separate the analytes of interest after manual sample cleanup using a 2D-LC-setup, combining online solid phase extraction (SPE) with elution on an analytical column. Stable isotope-labelled drugs U-ring-remdesivir-13 C6, GS-441524-13 C5 and favipiravir-13 C1 15 N1 were obtained from Alsachim (Straßbourg, Grand Est, France). Chloroquine-D4 phosphate, hydroxychloroquine-D4 sulfate, ritonavir-D6, lopinavir-D8, and azithromycin-13 C1D3 were from Toronto research chemicals. UHPLC-grade water, acetonitrile, and methanol were from J.T. Baker (Jackson, TN, USA). UHPLC-grade formic acid was purchased from Biosolve (Dieuze, France). All used chemicals were of the highest purity available from the commercial suppliers. Table 1 . 

Sample preparation with manual protein precipitation was combined with analysis of the extracts by two-dimensional chromatography, including online solid phase 

Sample analysis was performed with a 2D Acquity Table 3 . Analytes were quantified with the TargetLynx V4.1 software (Waters) using the following settings: polynome type, linear; origin, excluded; weighting function, 1/x.

The ID-LC-MS/MS method was validated on the basis of the guideline of bioanalytical method validation from the European Medicines Agency (EMA), 21 July 2011 [24] . The J o u r n a l P r e -p r o o f multi-analyte assay was validated in terms of calibration curve, intra-and inter-assay accuracy and precision, carry-over, selectivity, matrix effect, recovery, dilution integrity, and stability.

All non-zero-calibrators were processed together with a zero calibrator (without analyte, but added internal standard) and a blank (without analyte, without internal standard). Non-zero calibrators should be within ± 15 % of the nominal (theoretical) value, except of the lowest calibrator (termed lower limit of quantification, LLOQ)

concentration where it should be within ± 20 %.

Imprecision was expressed with the coefficient of variation (CV), inaccuracy with 

Carry-over was tested by injection of blank serum samples from different donors after the highest calibrator. The peak area in blanks should not exceed 20 % of the LLOQ peak area and 5 % of the ISTD peak area.

To test for selectivity, leftover anonymized sera from intensive care patients (n=30), who were not treated with the drugs in this study, but typically with polypharmacy, were investigated. Absence of interfering substances is accepted when the response is ≤ 20 % of the LLOQ for the analytes and ≤ 5 % for the ISTD.

We also assessed the mean ion ratios for quantifiers and qualifiers 

Matrix effect testing was done by post-column infusion experiments according to Bonfiglio et al. [26] . Neat analyte and ISTD solutions were directly infused into the mass spectrometer with 0.10 ng/min while processed blank sera (n=2) were injected to the chromatographic system. Corresponding chromatograms were compared to chromatograms obtained by the injection of pure methanol. Deviations in the baseline or breakdowns of signal intensities would indicate relevant matrix effects.

Quantitative matrix effect and recovery were examined according the EMA guideline [24] , and Matuszweski et al [27] at two different concentrations (low/QC B, high/QC D level). Using the certified reference materials, stock solutions at 1 mg/mL in methanol were prepared for remdesivir, chloroquine, hydroxychloroquine, ritonavir, lopinavir, The internal standard normalized matrix factor (MF) was calculated as the ratio of the MF of the analyte and the MF of the corresponding internal standard and its CV should be ≤ 15 %.

To test dilution integrity, a serum sample was prepared with concentrations exceeding the highest calibrator 6 (termed upper limit of quantification, ULOQ) by a factor of x 

Stability was tested with QC B and QC D samples (refer to Table 1 for concentrations) that were stored up to 4 h at RT (benchtop stability) and -20°C for four weeks.

Autosampler stability of processed samples was also tested for 24 h at 8°C. Freeze and thaw stability was tested in 3 cycles at -20°C (freeze time > 12 h) and thawing at room temperature. Stability was tested for each storage condition by measuring QC samples in triplicate using freshly prepared calibration samples. Stability in all conditions was accepted with a deviation of ≤ 15 % from the nominal concentration.

The validated 2D-UHPLC-MS/MS method was applied to anonymized leftover serum blood samples from patients receiving hydroxychloroquine, ritonavir, azithromycin, or remdesivir (six samples per analyte) . Analysis was conducted in accordance with the

Committee (document number KB 20/029).

According to the summary of product characteristics and National institute of Health (NIH) COVID19 treatment guidelines [28] dosages are as follows: remdesivir, 200 mg on day 1 and then 100 mg daily for up to 9 days; hydroxychloroquine, 2x 200-400 mg daily; ritonavir, 2x 100 mg daily combined with 400 mg lopinavir; azithromycin, 1x 500 mg daily. Specimens were immediately transferred to the laboratory and sera obtained from collection tubes without gel additive by centrifugation at 2000 x g for 10 minutes at 20°C. Supernatants were stored up to two weeks at -80°C until analysis.

J o u r n a l P r e -p r o o f 3. Results

The calibration curve was generated using 6 calibrators with a linear regression model and weighting factor 1/x. The linearity of the method was shown for all calibrations with R 2 ≥ 0.993 for all analytes in this study. All calibrators met the EMA specifications with ± 15 % of nominal value, except the LLOQ where it was within ± 20 %. A representative analytical chromatogram is given in Figure 2 .

For all analytes intra-and inter-day inaccuracy was ≤ 9.59 %intra-and inter-day imprecision was ≤ 11.1 %, respectively. The respective analytical performance data is summarized in Table 4 .

Peak-areas in blank serum samples injected after the highest calibrator was ≤ 10. 

The post-column infusion experiment indicated no relevant matrix effects at the analytic retention time for the analytes and ISTDs in this study given that no noticeable differences were observed between the injection of processed blank sera and pure methanol. The results for qualitative matrix effect testing are shown in Figure 3 .

The CV of the internal standard normalized matrix factor ranged from 0.68 % to 6.22 % and recoveries were between 86.8 % and 107 %. Quantitative matrix effect and recovery results are summarized in Table 4 .

Dilution integrity was given for all analytes with inaccuracy and imprecision values of ≤ 9.02 % and ≤ 10.8 % for the 1:5 dilution.

Stability was given for all analytes with deviations ≤ 14.4 % for all conditions tested.

Benchtop stability (4 h, RT), autosampler stability (24 h, 8°C), freeze thaw testing (-J o u r n a l P r e -p r o o f

We developed and validated a two-dimensional isotope dilution UHPLC-MS/MS method for the simultaneous quantification of remdesivir and its metabolite GS-441524, chloroquine, hydroxychloroquine, ritonavir, lopinavir, favipiravir and azithromycin in human serum.

The crucial advantage of combining two-dimensional chromatography with isotope dilution standardization is that matrix effects can be minimized, which is particularly important in samples from critically ill patients with the risk of unforeseen analytical interference. Furthermore, the use of an online-extraction column contributes to the longevity of the analytical column and avoidance of mass spectrometer contamination by transferring remaining matrix components (in particular peptide/protein residues and salts) directly to the waste. In addition, one-dimensional liquid chromatography is not always capable of efficiently separating analytes of interest in complex samples.

However, multidimensional separation methodologies have higher technical requirements when compared to one-dimensional LC-systems.

To quantify all analytes with a uniform injection volume and adequate sensitivity, we performed analyte detuning for ritonavir and lopinavir by adjusting the collision energies. For the quality controls tested, both intra-and inter-day inaccuracy and imprecisions were ≤ 9.59 % and ≤ 11.1, complying with the EMA specifications [24] .

Carry-over was negligible for all analytes, except for chloroquine where it was ≈ 76 % of the lowest calibrator. Even though we tested several wash conditions, carry-over could not be reduced any further, probably due to the relatively low chloroquine concentration in the LLOQ and our autosampler instrumentation. To solve this issue, samples with different analytes can be injected in alternating order e.g. chloroquineremdesivir-chloroquine given that these two drugs are not administered simultaneously unextracted plasma, presumably due to the presence of endogenous esterases [10] .

In order to prevent the degradation of remdesivir in vitro, corresponding clinical samples should therefore be processed either immediately or after storage of up to four weeks at -20°C or up to 6 months at -80°C [10] .

We applied the method to anonymized serum samples from patients receiving hydroxychloroquine, ritonavir, azithromycin or remdesivir. Concentrations to be expected from known pharmacokinetic data were obtained within the calibration range of the assay, taking into account the average blood-plasma ratio of 7.2 (± 4.2) for hydroxychloroquine [29] . In agreement with a recently published pharmacokinetic study in two patients receiving treatment with remdesivir [30] , the concentration of the prodrug remdesivir (t½, ≈ 1h) was < LLOQ in our samples, while GS-441524 (t½, ≈ 24h) was consistently present in corresponding specimens, stressing the necessity for concomitant GS-441524 monitoring.

According to the EMA-guideline on bioanalytical method validation [24] , the calibration range of an analytical assay is defined by the calibrators. Samples with concentrations exceeding the highest calibrator, (termed upper limit of quantification, ULOQ) can be diluted into the calibration range, while concentrations below the lowest calibrator (termed lowest limit of quantification, LLOQ) are reported with "< LLOQ". Depending on the dosage regimens to be subject to TDM and the scientific objective (e.g.

pharmacokinetic sampling), the measuring range may be extended with additional calibrators. However, this would typically require a separate partial validation.

In conclusion, we report a robust two-dimensional ID 

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. 

Role of corticosteroid in the management of COVID-19: A systemic review and a Clinician's perspective

Ongoing Clinical Trials for the Management of the COVID-19 Pandemic

Compassionate Use of Remdesivir for Patients with Severe Covid-19

Experimental Treatment with Favipiravir for COVID-19: An Open-Label Control Study. Engineering (Beijing)

Role of Lopinavir/Ritonavir in the Treatment of Covid-19: A Review of Current Evidence, Guideline Recommendations, and Perspectives

Azithromycin for COVID-19: More Than Just an Antimicrobial?

No benefit of hydroxychloroquine in COVID-19: Results of Systematic Review and Meta-Analysis of Randomized Controlled Trials

COVID-19 pathophysiology: A review

Concerns about pharmacokinetic (PK) and pharmacokinetic-pharmacodynamic (PK-PD) studies in the new therapeutic area of COVID-19 infection

Development and validation of a UHPLC-MS/MS method for quantification of the prodrug remdesivir and its metabolite GS-441524: a tool for clinical pharmacokinetics of SARS-CoV-2/COVID-19 and Ebola virus disease

Remdesivir Gilead

Validation of a method for the simultaneous quantification of chloroquine, desethylchloroquine and primaquine in plasma by HPLC-DAD

High sensitivity methods to quantify chloroquine and its metabolite in human blood samples using LC-MS/MS

Population Pharmacokinetics of Hydroxychloroquine in Japanese Patients With Cutaneous or Systemic Lupus Erythematosus

Simultaneous quantitation of hydroxychloroquine and its metabolites in mouse blood and tissues using LC-ESI-MS/MS: An application for pharmacokinetic studies

Simple and simultaneous determination of the hiv-protease inhibitors amprenavir, atazanavir, indinavir, lopinavir, nelfinavir, ritonavir and saquinavir plus M8 nelfinavir metabolite and the nonnucleoside reverse transcriptase inhibitors efavirenz and nevirapine in human plasma by reversed-phase liquid chromatography

Simultaneous determination of HIV protease inhibitors amprenavir, atazanavir, indinavir, lopinavir, nelfinavir, ritonavir and saquinavir in human plasma by high-performance liquid chromatography-tandem mass spectrometry

Favipiravir Pharmacokinetics in Nonhuman Primates and Insights for Future Efficacy Studies of Hemorrhagic Fever Viruses

Pharmacokinetics of Favipiravir in Critically Ill Patients With COVID-19

Quantitative determination of azithromycin in human plasma by liquid chromatography-mass spectrometry and its application in a bioequivalence study

Simplified LC-MS/MS method enabling the determination of azithromycin in human plasma after a low 100mg dose administration

After another decade: LC-MS/MS became routine in clinical diagnostics

Two-dimensional liquid chromatography in pharmaceutical analysis. Instrumental aspects, trends and applications

Clinical and Laboratory Standards Institute. Liquid chromatography-mass spectrometry methods

The effects of sample preparation methods on the variability of the electrospray ionization response for model drug compounds

Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS

COVID-19) Treatment Guidelines. National Institutes of Health

A dose-ranging study of the pharmacokinetics of hydroxy-chloroquine following intravenous administration to healthy volunteers

Pharmacokinetics of remdesivir and GS-441524 in two critically ill patients who recovered from COVID-19

J o u r n a l P r e -p r o o f