Novel methodology to perform incurred sample reanalysis (ISR) blood spot (DBS) cards – Experimental data using darolutamide and filgotinib
Vinay Kiran, 1 Abhishek Dixit, 1 Bhavesh Babulal Gabani, 1 Nuggehally R. Srinivas2 and Ramesh Mullangi*1
Abstract
Different options on doing incurred sample reanalysis (ISR) on dried blood spot (DBS) cards were investigated using drugs belonging to various therapeutic areas: (a) darolutamide (to treat prostate cancer) and (b) filgotinib (to treat rheumatoid arthritis). The proposed novel methodology included the generation of half-DBS and quarter-DBS discs after initial blood collection using the full-DBS disc. Accordingly, blood collection via DBS was performed in male Sprague Dawley rats following intravenous and oral dosing of darolutamide; in male Balb/C mice following intravenous and oral dosing of filgotinib. The ISR data generated from the full-DBS disc, half-DBS disc and quarter-DBS disc were compared for the assessment of the proposed methodology. Quantification of darolutamide and filgotinib was accomplished using liquid chromatography-electro spray ionization/tandem mass spectrometry (LC-ESI-MS/MS) methods. Darolutamide and filgotinib ISR samples, that were collected and prepared using full-, half- and quarter-DBS discs, met the acceptance criteria for ISR analysis. In conclusion, this is the first report showing a viable tool for the performance of ISR on DBS cards. The use of quarter- or half-DBS discs would aid in not only ISR but also in long term storage experiments of analytes since it would avoid the need of additional blood sampling in subjects.
KEY WORDS: DBS; ISR; darolutamide; filgotinib; LC-MS/MS.
1. INTRODUCTION
Incurred sample reanalysis (ISR) is not only gaining importance but has become a mandatory requirement in bioanalytical method validation as this parameter provides a basis for early detection and/or identification with respect to the discrepancies between original and repeat analysis (Rocci et al., 2007; Srinivas, 2011; Rudzki et al., 2017 & 2019). The rigorous conduct of the ISR in early drug development process keeps a check on method reproducibility issues and it supports the validity of the pharmacokinetic data gathered in both preclinical and clinical studies. In recent times, even in the drug discovery stage, the concept of ISR has been introduced to gain the confidence on laboratory conditions and addressing reproducibility issues, which arise due to matrix effect, recovery, stability of analyte(s)etc (Giri et al., 2019). Another recent work examined ISR for ocular samples collected from various eye tissues namely aqueous humor, vitreous humor, retina-choroid and lens (Gabani et al., 2020).
As per the regulatory guideline, the samples for ISR analysis will be selected from the time points near Cmax (maximum concentration in plasma) and around the elimination phase. In case of plasma samples, one can withdraw the sample for ISR analysis from the same vial and apportion into various aliquots for ISR analysis. However, in the case of DBS, repeat analysis from the same DBS card is not possible in routine practice. However, in the case of DBS, repeat analysis from the same DBS card is not possible in routine practice and also, might not be appropriate. In practice it may be possible although laborious to make multiple punches (DBS) for initial analysis and ISR can be taken from a different spot. However, this may represent an indirect ISR measurement as the same spot is not repeated for analysis. Given the back-ground, the intent of the present work is to propose a new strategy and frame work to conduct the ISR using DBS discs that are into two or four equal parts along with full DBS card serving as a control for such ISR analysis. In order to demonstrate the applicability of this strategy, we used two drugs from different therapeutic areas namely darolutamide, which was recently approved for the treatment of non-metastatic castrate-resistance prostate cancer (Nubeqa) and filgotinib, which was recently filed in US and Japan to treat the patients suffering from mild-to-severe rheumatoid arthritis (Pharma Japan, 2019; (Gilead, 2019).
2. MATERIALS AND METHODS
2.1. Chemicals and Reagents
Darolutamide (purity: >97%; Supplementary Fig. 1) was obtained from Angene International Limited, China. Apalutamide-d3 (purity: 99.5 %; Supplementary Fig. 1) used as an internal standard for the quantitation of darolutamide was synthesized by the Medicinal chemistry group, Jubilant Biosys (Bangalore, India) using literature information (Pang, Wang & Chen,
2017) and characterized using chromatographic (HPLC, LC-MS/MS) and spectral techniques (IR, UV, Mass, 1H and 13C-NMR) by the Analytical research group, Jubilant Biosys. Filgotinib (purity: >95%; Supplementary Fig. 1) was purchased from Angene International Limited, Tsuen Wan, Hong Kong. Tofacitinib (IS; purity: 98%; Supplementary Fig. 1) used as an internal standard for the quantitation of filgotinib was purchased from Sigma-Aldrich (St. Louis, USA). Whatmann FTA DMPK card C DBS cards were purchased from GE, Bangalore. Silica gel sachets and sealable plastic bags for the storage of DBS cards were purchased from local market. The mouse and rat hematocrit value was measured using Mindray BC-5000Vet (Shenzhen, China). The hematocrit values ranged between 43-48% (mice) and 38-46% (rats), similar to the reported data (Mara et al., 2011; Santos et al., 2016).
2.2. Preparations of stock solutions for darolutamide, filgotinib and internal standards
Two separate primary stock solutions of darolutamide (1.10 mg/mL) and filgotinib (1.00 mg/mL) were prepared in DMSO. Appropriate secondary and working stocks of darolutamide and filgotinib were prepared from the respective primary stock solution by serial dilution with methanol:water (80:20, v/v). These solutions were used to prepare the calibration curve (CC) and quality controls (QCs).
Similarly, the ISs (tofacitinib and apalutamide-d3) primary stock solution (1.00 mg/mL) was made in DMSO. Each IS primary stock solution was diluted appropriately with methanol and subsequently used as IS working stock solution. The primary stock solutions of darolutamide, filgotinib and the ISs were stored at -20°C, which were found to be stable for 75 days. Working stock solutions were stored at 4°C for 25 days.
2.3. Preparation of quality control and calibration curve samples
Darolutamide: Samples for the determination of precision and accuracy were prepared by spiking control mouse blood with darolutamide (using secondary and working stocks from first primary stock solution) at appropriate concentrations 3.36 ng/mL (low quality control, LQC), 1108 ng/mL (medium quality control, MQC) and 1612 ng/mL (high quality control, HQC) (n=6 at each QC level/run). Using secondary and working stocks from second primary stock solution, calibration samples were prepared on each validation day. Peak area ratio of darolutamide to that of the IS was used for all calculations. A least squares linear regression (1/X2 weighting factor) of eight non-zero samples (1.12, 2.24, 11.2, 44.8, 560.0, 1008, 1456 and 2016 ng/mL) was used to define the calibration curve.
Filgotinib: In a similar way, QCs were prepared for filgotinib in rat blood from independent stock solution at nominal concentration of 4.01 (LQC), 1068 (MQC) and 1736 (HQC) ng/mL, respectively (n=6 at each QC level/run). The calibration curve standards were 1.30, 2.60, 13.0, 26.0, 312, 624, 1248 and 1924 ng/mL. A least squares linear regression (1/X2 weighting factor) of eight non-zero samples was used to define the calibration curve.
Unknown and QC samples were back-calculated using standard curve for each drug. For both the drugs, inter- and intra-day precisions were determined by calculating percent relative standard deviation (%RSD) that should be ±15% for all the QCs except for LLOQ QC where it should be ±20%. The inter- and intra-day accuracy expressed as percent relative error (%RE) was calculated by comparing the measured concentration with the nominal value and deviation was limited within ±15% except for LLOQ QC where it should be ±20%.
2.4. Animals dosing and blood collection on DBS cards
Twenty-four male Balb/C mice and eight Sprague Dawley rats were purchased form Vivo Biotech, Hyderabad, India. Rodents were housed in Jubilant Biosys animal house under controlled temperature (22 ± 2 °C) and humidity (30-70%) conditions with 12:12 h light:dark cycle and 15 air changes per hour. During the acclimatization rodents had free access to food and water. Protocols used in the study were approved by the Institutional Animal Ethics
Committee (IAEC) approved the protocols (mice: IAEC/JDC/2019/188R; rats: IAEC/JDC/2019/189R). After fasting (4 h fasting for mice and 12 h fasting for rats and during fasting period rodents had free access to water) mice (26-33 g) were divided into two groups (Group A and B) comprising twelve mice in each group, similarly rats (215-216 g) were divided into two groups (Group 1 and 2) having four rats in each group. Group-A and B mice received darolutamide by oral (10 mg/kg) and intravenous (1.0 mg/kg) routes, respectively. For oral administration a solution formulation was prepared using 50% PEG400, 30% propylene glycol and 20% of 5% glucose at 1.0 mg/mL strength. For intravenous dosing a solution formulation comprising 5% DMSO:5% Solutol:absolute alcohol (1:1, v/v) and 90% saline was prepared at strength: 0.1 mg/mL. The dose volume was 10 mL/kg for both the routes. Group-1 and 2 rats received filgotinib by oral (10 mg/kg) and intravenous (2.0 mg/kg) routes, respectively. For oral administration a solution formulation was prepared using Tween-80 and 0.5% methyl cellulose at 1.0 mg/mL strength. The dose volume was 10 mL/kg. For intravenous dosing a solution formulation comprising 5% DMSO:5% Solutol:absolute alcohol (1:1, v/v) and 90% saline was prepared at 0.5 mg/mL strength. The dose volume was 4.0 mL/kg.
Post-dosing, blood samples (80 µL) were collected at 0.083, 0.25, 1, 2, 6, 8 and 24 h (for intravenous route) and 0.25, 1, 2, 4, 8, 12 and 24 h (for oral route) from mice (sparse sampling technique was used) received darolutamide and at 0.083, 0.5, 1, 4, 8 and 24 h (for intravenous route) and 0.5, 1, 2, 4 and 8 10 and 24 h (for oral route) from rats received filgotinib through retro-orbital plexus into the vials containing K2.EDTA as an anticoagulant. Feed was provided to rodents post 2 h of dosing. For each drug at each time point 25 µL of blood was spotted on DBS card (in triplicate) using a calibrated pipette and left to dry at room temperature at least for 3 h. Post drying DBS cards were preserved in sealed plastic bags (having desiccant) at room temperature until bioanalysis.
2.5. DBS samples processing, original and ISR bioanalysis
At each time point from the triplicate DBS cards, the first set of DBS cards were used to punch out 6 mm diameter discs (Harris-Micro-Punch®) and used to obtain regular pharmacokinetic data. From the second set of DBS cards, 6 mm diameter discs were punched out and used for full-DBS disc ISR analysis. From the third set, each disc was again cut into two equal halves. One-half was used for half-DBS ISR analysis and another half was divided further into two-halves to get quarter-DBS disc and subsequently used for ISR analysis (Fig. 1). All the DBS-discs (full, half and quarter) were processed in similar lines as mentioned below for the quantitation of respective drugs (Saini, Sulochana, Zainuddin & Mullangi, 2018 for darolutamide and Dixit, Kiran, Gabani & Mullangi, 2020 for filgotinib). ISR of all DBS-discs (full, half and quarter) was done on day-50.
Darolutamide: Briefly, to each DBS card disc (full/half/quarter) in a glass tube, 200 μL of water was added vortex mixed for 10 min and sonicated for 10 min at room temperature. Post sonication 200 μL of acetonitrile enriched with the IS (25 ng/mL) was added vortex mixed for 10 min followed by centrifugation 10 min. Clear supernatant (200 μL) was separated and transferred into an HPLC vial for injection and 5.0 μL was injected onto LC-MS/MS system for analysis.
Chromatographic resolution of darolutamide was achieved on an Atlantis C18 column (50 × 4.6 mm; 3 µm) maintained at room temperature. An isocratic mobile phase, a mixture of 0.2% formic acid:acetonitrile (30:70, v/v) at a flow-rate of 0.8 mL/min was used for the resolution of darolutamide and the IS. The mass spectrometer was operated under positive ion mode. The transitions (Q1Q3) used for quantitation of darolutamide and the IS were m/z 399178 and 481453, respectively. Irrespective of the DBS disc type, the linearity range was 1.12-2016 ng/mL. Along with study samples quality control (QC) samples supplemented at concentrations of 3.36 (LQC), 1008 (MQC) and 1904 (HQC) ng/mL were analyzed.
Filgotinib: Briefly, to each DBS card disc (full/half/quarter as the case may be) in a glass tube, 200 µL of 0.2% formic acid enriched with 100 ng/mL of IS was added, thereafter the contents were vortex mixed for 5 min (Thermomixer®, Eppendorf) and sonicated (Elmasonic S300 H) for 15 min at room ambient temperature. After sonication, to the same microcentrifuge, 1.0 mL of ethyl acetate was added and the mixtures were vortexed further for 3 min. The samples were centrifuged at 14,000 rpm for 3 min. The clear organic layer (800 µL) was pipetted out after centrifugation and dried under gentle stream of nitrogen (Turbovap®, Zymark®,Kopkinton, MA, USA) at 50C. The residue was reconstituted with 200 µL of 0.2% formic acid in acetonitrile and 150 µL clear supernatant was aliquoted into an HPLC vial and 5.0 µL was injected onto the column for LC-MS/MS analysis.
Filgotinib and the IS were separated on a Gemini C18 column (100 × 4.6 mm; 3 µm) maintained at room temperature using a mixture of 0.2% formic acid:acetonitrile (20:80, v/v) at a flow-rate of 0.9 mL/min. The mass spectrometer was operated under positive ion mode. The transitions (Q1Q3) used for quantitation of filgotinib and the IS were m/z 426.3291.3 and 313.2149.2, respectively. Irrespective of the DBS disc type, the linearity range was 1.30-1924 ng/mL. Along with study samples quality control (QC) samples supplemented at concentrations of 3.90 (LQC), 936 (MQC) and 1352 (HQC) ng/mL were analyzed.
2.6. Incurred samples reanalysis (ISR)
Recent guideline has emphasized on the necessity of ensuring ISR reproducibility. As per the guidance, 10% of the samples should be reanalyzed if total samples size is less than 1000. As per the guidance, the difference in concentrations between the initial value and the ISR should be less than ± 20% of their means for at least 67% of the repeats (US DHHS et al., 2018). For ISR analysis, from the oral and intravenous arms for darolutamide a total of 18 samples were selected for each disc (full/half/quarter). From the intravenous route 0.083, 1 and 6 h and for oral route 1, 4 and 8 h time point samples were selected. In a similar way, for filgotinib DBS samples at 0.083, 1, 4 and 8 h (for intravenous route) and 1, 4 and 8 h (for oral route) were selected for ISR analysis (total of 28 samples) for each disc (full/half/quarter).
3. RESULTS
For darolutamide the initial time points (in both routes) blood samples showed high concentration above ULOQ (upper limit of quantitation) were diluted appropriately to bring the concentration within linearity range. The accuracy observed for the mean of backcalculated concentrations for all the discs (full/half/quarter) in calibration curves for darolutamide was within 92.1-109% and 89.3-107%, for pharmacokinetic and ISR samples, respectively; while the precision (CV) values ranged from 2.32-10.9 and 1.48-6.90%, respectively in the calibration curve samples with a regression (r2) of 0.993. In the QC samples, the accuracy and precision (%CV) ranged between 92.2-108% and 5.66-7.31%, respectively in all the discs belonging to the pharmacokinetic samples analysis of darolutamide. In ISR samples analysis the accuracy ranged between 95.8-110% and the precision values were <5.61% in all the analyzed discs.
Similarly, for filgotinib, the accuracy values across the DBS discs were within 87.3-114% and 90.2-110%, for pharmacokinetic and ISR samples, respectively; while the precision (CV) values ranged from 1.51-8.11 and 3.94-9.46%, respectively in the calibration curve samples (r2: 0.993). In the QC samples, the accuracy and precision (%CV) ranged between 88.1-111% and 4.51-9.01%, respectively in all the discs belonging to the pharmacokinetic samples analysis of filgotinib. In ISR samples analysis the accuracy ranged between 94.3-111% and the precision values were in the range of 4.11-8.87% in all the analyzed discs.
Comparison of ISR values vs. original values in full-, half- and quarter-DBS discs using Bland-Altman plots for darolutamide by oral are presented in Fig. 2A to 2C and all the disc values are collaged in Fig. 2D. Similarly, in Fig. 3A, 3B and 3C the ISR vs. original values for darolutamide intravenous samples were presented for full, half and quarter-DBS discs, respectively. In Fig. 3D, the ISR vs. original values for all the discs are shown. Both in oral and intravenous drug dosing, the ISR values for the analyzed samples for darolutamide were within ±20% when compared with original values.
In similar lines, the ISR values vs. original values for filgotinib full, half and quarter-DBS discs in oral pharmacokinetic study samples in Fig. 4A, 4B and 4C, respectively. In Fig. 4D the collective data from all the discs (full, half and quarter discs) was shown. In a similar fashion, Fig. 5A to 5C depicts the ISR vs. original values for filgotinib intravenous pharmacokinetic study samples obtained from full, half and quarter-DBS discs, respectively. In Fig. 5D the ISR vs. original values for all the discs are shown. Both in oral and intravenous routes, for all the analyzed samples ISR values for filgotinib were within ±20% when compared with original values.
Supplementary Table 1 and Supplementary Table 2 enlist the percentage difference between ISR value vs. original value for each drug in full-, half- and quarter-DBS disc for darolutamide and filgotinib, respectively.
4. DISCUSSION
Whole blood sample bioanalysis is cumbersome as compared to plasma/serum bioanalysis of samples and therefore, it is less preferred to be used as a matrix to generate relevant pharmacokinetic data in animals and humans. However, there are many drugs like cyclosporin (Fahr, 1993), chloroquine (Kaewkhao et al, 2019) tacrolimus (Venkataramanan et al., 1995) and desidustat (ZYAN1) (Kansagra et al., 2018) etc. whose pharmacokinetics have been derived using the whole blood matrix. One of the reasons for choosing whole blood as a matrix for the above mentioned drugs is due to the higher partitioning of the drug into red blood cells relative to plasma and therefore, in such instances, blood as opposed to plasma would likely represent as a key surrogate to define the pharmacokinetics of the drug. Also, due to higher levels in blood relative to plasma in such instances, it would help to achieve reasonable sensitivity for the quantitation of drug(s) in blood. Similar to analysis of plasma samples using LC-MS/MS, the procedures to establish matrix effect would be a key for analysis based on blood samples (Jemal & Xia, 2006; Srinivas, 2009). In addition, the establishment of ISR would become another aspect that need to be considered for bioanalytical methods that use blood as the matrix.
Because DBS is now being increasingly recognized as a great option to collect blood samples (Sulochana, Daram, Srinivas & Mullangi, 2019; Dittakavi, Jat & Mullangi, 2019; Linder et al., 2019; Kim et al., 2019; Kumar Saini et al., 2019), we were interested to develop and implement a novel strategy for carrying out ISR analysis using quarter-, half- and full- DBS discs. As depicted in the Fig. 1, this framework of dividing DBS cards (into two equal pieces or four equal pieces, as the case may be) would render itself as a useful approach for preparing aliquots of study samples collected on DBS for ISR analysis. Also, the same strategy could be applied for the experimental designs to facilitate collection of the long term stability data using the spiked samples on DBS cards. Furthermore, this approach would maximize the utility of DBS approach in pharmacokinetic studies especially those where multiple analytes need to be analyzed with different methods (extraction/chromatography and detection). In drug development, the need of analysis of multiple analytes is essential and a strategy should be in place for such analyses (Srinivas, 2004 & 2006). In such cases, quarter or half portion of DBS cards can be considered for different assays to facilitate analysis of multiple analytes.
We selected two drugs namely darolutamide and filgotinib for this ISR analysis using the proposed DBS methodology since it would be relevant in blood sampling procedures in cancer and rheumatoid arthritis patients, respectively. Moreover, we used both oral and intravenous routes of drug administration to gather pharmacokinetic samples for both drugs for the use in the structured ISR analysis. The ISR data supports the utility of our proposed methodology for either of the two drugs. Although for experimental demonstration, we used mouse/rat DBS apportioned to quarter-and half- DBS discs with full DBS disc serving as a control, the same strategy and proposal can be applied for ISR in human DBS for either of the two drugs or for other drugs.
We believe this innovative strategy and proposal has a significant benefit in delineating ISR in DBS, which otherwise would have been difficult or cumbersome due to the need of collection of multiple DBS discs for ISR analysis. Furthermore, the strategy of quarter or half-DBS discs could also be potentially employed to perform long-term storage experiments, which in turn would avoid use of additional blood volume.
5. CONCLUSION
In this paper we report the novel strategy with well-planned experimental frame work to perform ISR analysis with traditional DBS cards using darolutamide and filgotinib as reference drugs. The proposed strategy was evaluated using quarter- and half-DBS discs in relation to full-DBS disc. The ISR data suggested that quarter- and half-DBS discs, in addition to full-DBS discs can be used for original and ISR analysis to establish the reproducibility and ruggedness of the bioanalysis.
References
Dittakavi, S., Jat, R.K., & Mullangi, R. (2019). Quantitation of ivosidenib, a novel mutant IDH1 inhibitor on mice DBS: Application to a pharmacokinetic study. Drug Research, 69, 505-511.
Dixit, A., Kiran, V., Gabani, B.B., & Mullangi, R. (2020). Validated LC-MS/MS method for quantitation of a selective JAK1 inhibitor, filgotinib in rat plasma and its application to a pharmacokinetics study in rats. ADMET & DMPK, (in press).
Fahr, A. (1993). Cyclosporin clinical pharmacokinetics. Clinical Pharmacokinetics, 24, 472-495.
Gabani, B.B., Kiran, V., Praharaj, S., Srinivas, N.R., & Mullangi, R. (2020). Incurred sample reanalysis (ISR) of cefuroxime in rabbit ocular tissues - A case study. Biomedical Chromatography, 34, e4737.
Gilead. (2019). https://www.gilead.com/news-and-press/press-room/press-releases
Giri, P., Patel, N., Joshi, V., Giri, S., & Srinivas, N.R. (2019). Incurred sample reanalysis in drug discovery bioanalysis. Biomedical Chromatography, 33, e4430.
Jemal, M., & Xia, Y. Q. (2006). LC-MS Development strategies GLPG0634 for quantitative bioanalysis. Current Drug Metabolism, 7, 491-502.
Kaewkhao, K., Chotivanich, K., Winterberg, M., Day, N.P., Tarning, J., & Blessborn, D. (2019). High sensitivity methods to quantify chloroquine and its metabolite in human blood samples using LC-MS/MS. Bioanalysis, 11, 333-347.
Kansagra, K.A., Parmar, D., Jani, R.H., … Barot K. (2018). Phase I clinical study of ZYAN1, a novel prolyl-hydroxylase (PHD) inhibitor to evaluate the safety, tolerability, and pharmacokinetics following oral administration in healthy volunteers. Clinical Pharmacokinetics, 57, 87-102.
Kim H.M., Park, J.H., Long, N.P., Kim, D.D., & Kwon, S.W. (2019). Simultaneous determination of cardiovascular drugs in dried blood spot by liquid chromatographytandem mass spectrometry. Journal of Food and Drug Analysis, 27, 906-914.
Kumar Saini, N., Sulochana, S.P., Kiran, V., Mallurwar, S.R., Richter, W., Srinivas, N.R., & Mullangi, R. (2019). Simultaneous determination of bendamustine and hydroxybendamustine in mice dried blood spots and its application in a mice pharmacokinetic study. Journal of Pharmaceutical and Biomedical Analysis, 174, 168174.
Linder, C., Neideman, M., Wide, K., von Euler, M., Gustafsson, L.L., Pohanka, A. (2019). Dried blood spot self-sampling by guardians of children with epilepsy is feasible: comparison with plasma for multiple antiepileptic drugs. Therapeutic Drug Monitoring,