Bioanalysis of bevacizumab and infliximab by high-temperature reversed-phase liquid chromatography with fluorescence detection after immunoaffinity magnetic purification

Analytica Chimica Acta 916 (2016) 112e119

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Bioanalysis of bevacizumab and infliximab by high-temperature reversed-phase liquid chromatography with fluorescence detection after immunoaffinity magnetic purification Kenichiro Todoroki a, Tatsuki Nakano a, Yasuhiro Eda a, Kaname Ohyama b, Hideki Hayashi c, Daiki Tsuji d, Jun Zhe Min a, Koichi Inoue e, Naoki Iwamoto f, Atsushi Kawakami f, Yukitaka Ueki g, Kunihiko Itoh d, Toshimasa Toyo'oka a, * a

Laboratory of Analytical and Bio-Analytical Chemistry, School of Pharmaceutical Sciences, University of Shizuoka, Shizuoka, Japan Department of Pharmacy Practice, Graduate School of Biomedical Sciences, Nagasaki University, Japan c Laboratory of Pharmacy Practice and Social Science, Gifu Pharmaceutical University, Gifu, Japan d Laboratory of Clinical Pharmacology and Genetics, School of Pharmaceutical Sciences, University of Shizuoka, Shizuoka, Japan e Laboratory of Clinical and Analytical Chemistry, College of Pharmaceutical Sciences, Ritsumeikan University, Japan f Unit of Translational Medicine, Department of Immunology and Rheumatology, Graduate School of Biomedical Sciences, Nagasaki University, Japan g Sasebo Chuo Hospital, Japan b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Bioanalytical methods for bevacizumab and infliximab have been developed.
The drugs in plasma samples were purified using immunoaffinity magnetic beads.
The purified drugs were separated with high-temperature reversedphase LC.
Developed methods were successfully applied to bioanalysis of clinical samples.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 December 2015 Received in revised form 12 February 2016 Accepted 19 February 2016 Available online 23 February 2016

This study presents two simple and rapid methods for the quantification of therapeutic mAbs based on LC. Two mAbs (bevacizumab and infliximab) in plasma samples were purified using magnetic beads immobilized with a commercially-available idiotype antibody for each mAb. Purified mAbs were separated with HT-RPLC and detected with their native fluorescence. Using immunoaffinity beads, each mAb was selectively purified and detected as a single peak in the chromatogram. The HT-RPLC achieved good separation for the mAbs with sharp peaks within 20 min. The calibration curves of the two mAbs ranged from 1 to 20 mg mL1 (bevacizumab) and 1e10 mg mL1 (infliximab), and they had strong correlation coefficients (r2 > 0.998). The LOD of bevacizumab and infliximab was 0.07 and 0.15 mg mL1, and the LLOQ of bevacizumab and infliximab was 0.12 and 0.25 mg mL1, respectively. Thus, the sensitivities were sufficient for clinical analysis. Immunoaffinity purification with HT-RPLC produced a selective and

Keywords: Therapeutic antibodies High-temperature reversed-phase liquid chromatography Immunoaffinity purification Bevacizumab Infliximab

* Corresponding author. Department of Analytical and Bio-Analytical Chemistry, School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Suruga, Shizuoka 422-8526, Japan. E-mail address: [email protected] (T. Toyo'oka). http://dx.doi.org/10.1016/j.aca.2016.02.029 0003-2670/© 2016 Elsevier B.V. All rights reserved.

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119

113

accurate bioanalysis without an LC-MS/MS instrument. Both methods could become general-purpose analytical methods and complement the results obtained with conventional LBA. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Therapeutic monoclonal antibodies (mAbs) and their related products are being increasingly developed and used for the treatment of a variety of diseases, such as cancer, rheumatoid arthritis (RA), autoimmunity diseases, and inflammation [1e3]. Therapeutic mAbs typically possess long drug efficacy and few side effects. However, the pharmacokinetics (PK) and pharmacodynamics (PD) of therapeutic mAbs are very complicated compared with lowmolecular-weight pharmaceuticals [4,5]. Because of the neonatal Fc receptor (which functions in mAbs for catabolism before blood transition, binding to target antigens, and transportation), most therapeutic mAbs are concentration dependent and show nonlinear pharmacokinetic behavior [5e8]. The PK and PD analyses of therapeutic mAbs have mainly utilized ligand binding assays (LBA) [9,10], such as the enzyme-linked immunosorbent assay (ELISA) and the chemiluminescent immunoassay. Although LBA permits a high sensitivity and high-throughput analysis, there is the potential for cross-reactivity of capture antibodies and low accuracy [11]. Recently, various alternative liquid chromatography e tandem mass spectrometry (LC-MS/MS) methods have been applied to analyze therapeutic mAbs in serum or plasma samples [12e19]. These methods enable a sensitive bioanalysis of therapeutic mAbs; however, they have limitations, such as time-consuming trypsin digestion and the manual purification process for tryptic peptides using solid-phase extraction (SPE) cartridges. Recently, several improved methods to speed up the trypsin digestion [20,21] and purification [22] processes have been developed, however, complicated procedures such as denature, reduction and alkylation before the digestion are still remaining. In 2009, Damen et al. reported a bioanalytical method for identifying trastuzumab by conventional HPLC [23]. In that study, trastuzumab in a serum sample was purified with an immunoaffinity column immobilized with an anti-idiotype mAb. The purified trastuzumab was separated by high-temperature reversed-phase LC (HT-RPLC) [24] and detected with its native fluorescence. Theoretically, this method could be sufficiently sensitive for the PK analysis of a mAb without an expensive LC-MS/MS instrument. However, the LLOQ of the method was 5 mg mL1 and the intra-day precision was less than or equal to 17.0%. This low sensitivity and insufficient precision can be ascribed to an inadequate immune-purification protocol. On the other hand, size-exclusion chromatographic (SEC) analysis of therapeutic mAb has already been reported [25]. In this method, infliximab was reacted with fluorescent labeled TNF-a and the resulting immune-complex was analyzed by SEC with fluorescence detection. However, the method requires fluorescently labeled ligands, and it did not afford sharp peaks in the chromatograms. The method developed here is a simple, sensitive, accurate and rapid quantification of therapeutic mAbs in plasma samples from cancer and RA patients using the combination of immunoaffinity magnetic purification and HT-RPLC with fluorescence detection. Target drugs in blood samples were purified with immunoaffinity magnetic beads immobilized with the commercial anti-idiotype mAb for each therapeutic mAb. The use of magnetic beads leads

to an accurate collection of the target and prevents excessive dilution during purification, which results in a better accuracy and sensitivity compared with previous methods [23]. The purified drug was further separated by HT-RPLC using a large-pore size coreeshell column. The separated drugs are then detected with their own fluorescence. We selected two therapeutic mAbs as target drugs, bevacizumab and infliximab. Bevacizumab is a humanized mAb against vascular endothelial growth factor A, and it is used alone or with other drugs to treat metastatic colon cancer, certain lung cancers, renal cancer, ovarian cancer, and glioblastoma. Since the patent of bevacizumab will expire in 2018, its biosimilars will be produced by many pharmaceutical companies. Infliximab is a chimeric mAb against tumor necrosis factor alpha, and it is indicated for RA, Crohn's disease, ulcerative colitis, etc. An infliximab biosimilar was first approved by the European Medicines Agency [26] in 2013, and further sales by many manufacturers can be expected. Therefore, characterization and bioanalysis methods for biosimilar and biobetter mAbs would be of the same importance as analysis of their original drugs [27]. Our method could also be applied in the evaluation of biological equivalencies in biosimilar development. The developed methods were validated for sensitivity, repeatability, linearity, accuracy, and precision. We successfully applied both methods to the bioanalyses of plasma samples obtained from cancer and RA patients who were administered each drug. To the best of our knowledge, the present report is the first to describe accurate and sensitive bioanalytical methods for bevacizumab and infliximab using immunoaffinity purification e HTRPLC. 2. Materials and methods 2.1. Reagents, solutions, and apparatus Deionized and distilled water, purified using the ELGA Purelab Flex system (ELGA, Marlow, UK), was used to prepare all aqueous solutions. LC grade acetonitrile and isopropanol were purchased from Kanto Chemicals (Tokyo, Japan). Tocilizumab (ACTEMRA® 80 mg for Intravenous Infusion), bevacizumab (Avastin® 400 mg/ 16 mL Intravenous Infusion), and trastuzumab (HERCEPTIN® Intravenous Infusion 150, 150 mg/7.2 mL) were produced by Chugai Pharmaceutical (Tokyo, Japan). Cetuximab (ERBITUX® Injection, 100 mg/20 mL) and infliximab (REMICADE for Intravenous Infusion100) were produced by Merck Serono (Tokyo, Japan) and Mitsubishi Tanabe Pharma (Osaka, Japan), respectively. Dynabeads M280 Tosyl-Activated (particle size: 2.8 mm) and the DynaMag Spin were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Both the anti-bevacizumab and the anti-infliximab idiotype antibodies were ELISA grade and obtained from Abnova Corporation (Taipei, Taiwan). Peptone from animal tissue was obtained from SigmaeAldrich (St. Louis, MO, USA). Control human plasma was obtained from volunteers. All other chemicals were of the highest purity available and were used as received. 2.2. Coupling of anti-bevacizumab or infliximab idiotype antibodies to magnetic beads Anti-bevacizumab or infliximab idiotype antibodies were

114

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119

coupled to tosyl-activated magnetic beads as follows. Briefly, a suspension of magnetic beads (33 mL, 1 mg) was placed in a 1.5-mL polypropylene tube. After removal of the solvent, the idiotype antibody (30 mg), dissolved in 100 mL of 100 mM sodium phosphate buffer (pH 7.4) containing 1.5 M ammonium sulfate, was added. This mixture was vortex-mixed at room temperature for 20 h with a microtube mixer (MT-360, TOMY SEIKO Corporation, Tokyo, Japan). After the addition of 50 mL of 1 M triseHCl buffer (pH 7.4) containing 1% peptone, the mixture was further vortex-mixed at room temperature for 20 h. After removal of the solution, the remaining beads were washed with 100 mM sodium phosphate buffer (pH 7.4) containing 0.1% Tween 20 (washing buffer). After washing, beads were dispersed in 100 mL of 100 mM sodium phosphate buffer (pH 7.4) and stored as a suspension at 4  C. 2.3. Isolation of therapeutic mAbs from plasma sample using immunoaffinity magnetic beads The affinity purification was executed using 1 mg immunoaffinity magnetic beads per sample. After removal of the solvent, the immunoaffinity beads were added to 100 mL of a plasma sample diluted 2e50 times with 100 mM phosphate buffer (pH 7.4). The mixtures were incubated with vortex-mixing at room temperature for 1 h. After incubation, the beads were washed twice with 100 mL of the washing buffer. Target mAbs were then eluted once with 100 mL of 100 mM citrate buffer (pH 3.1). Aliquots of 2 mL were injected onto the LC-fluorescence system. After elution, the resulting beads were reused after equilibration with 100 mL of 100 mM sodium phosphate buffer (pH 7.4). 2.4. LC system and conditions We used the Nexera ultrahigh-performance liquid chromatograph system (Shimadzu), which consisted of a CBM-20A system controller, an SIL-30AC auto sampler, two LC-30AD pumps, a DGU20A online degasser, a CTO-30A column oven, an SPD-M20A PDA detector and an RF-20A fluorescence spectrometer equipped with a 12-mL flow cell. The fluorescence intensity was monitored at excitation and emission wavelengths of 278 and 343 nm, respectively. The collected data were analyzed using a Lab Solutions LC (v. 1.21; Shimadzu); the peak areas and heights were estimated using the baseline-to-baseline method. 2.4.1. HT-RPLC The Aeris Widepore XB-C8 column, which is a core shell-type analytical column, packed with 3.6-mm coreeshell particles (150  2.1 mm I.D., Phenomenex, Torrance, CA, USA) was used. Mobile phase A was water containing 0.1% trifluoroacetic acid (TFA), while solvent B was 70% isopropanol: 20% acetonitrile: 9.9% water: 0.1% TFA. Gradient profiling involved an isocratic elution with A/B (90: 10) for 1 min, a linear gradient elution from A/B (90: 10) to A/B (75: 25) for 1 min, a linear gradient elution from A/B (75: 25) to A/B (50: 50) for 13 min, an isocratic elution with A/B (0: 100) for 5 min, and an isocratic elution with A/B (90: 10) for 8 min. The flow rates of the mobile phase and the column temperature were set at 0.4 mL min1 and 75  C, respectively. 2.5. Preparation of stock solutions, calibration standards, and quality control samples Avastin and Remicade injections were stored at approximately 4  C and were found to be stable for at least 6 months. These stock solutions were serially diluted with drug-free human plasma to obtain calibration standards at concentrations of 1, 2, 5, 10, and 20 mg mL1 bevacizumab or infliximab. For the preparation of

quality control (QC) samples, a similar procedure was followed. Another stock solution was serially diluted in a different batch of drug-free human plasma to prepare the calibration samples to obtain QC samples containing bevacizumab or infliximab at concentrations of 1, 2, 5, 10, and 20 mg mL1. The validation sample at 100 mg mL1 was prepared to assess the accuracy and precision after dilution in drug-free human plasma. 2.6. Sample collection Plasma samples of cancer and RA patients were collected from Seirei Hamamatsu General Hospital and Nagasaki University Hospital, respectively. Colon cancer patients with bevacizumab (n ¼ 3; age 50e75 years) were recruited from Seirei Hamamatsu General Hospital in Hamamatsu, Japan. This study was approved by the Ethics Committees of Seirei Hamamatsu General Hospital and the University of Shizuoka, Shizuoka, Japan. Patients received periodic doses of Avastin as injections (every month at 8 mg/kg for at least 6 months). Serum samples of RA patients (n ¼ 9; age 34e70 years) with infliximab who fulfilled the American College of Rheumatology criteria were collected at Sasebo Chuo Hospital. The disease activity score for C-reactive protein and erythrocyte sedimentation rate in RA patients were 2.5 and 3.1, respectively. This study was approved by the Ethics Committee of Nagasaki University Hospital. Patients received periodic injections of infliximab every two or three months at 3 mg kg1. In both experiments, written informed consents were obtained from either the patients or their legal guardians after the purpose of these studies was explained to them. Two milliliters of blood was collected from the treated subjects at times when they underwent biochemical examination of blood, and the times of the administrations of each drug were recorded. 2.7. Method validation Proposed analytical method was partially (intra- and inter-day precisions, accuracy, recovery) followed FDA Bioanalytical method validation [28]. To obtain the validation parameters, peak areas were estimated by LabSolution, LC, and the baseline-to-baseline method was used for the quantification. 2.7.1. Precision The precision of the assays was determined by the repeated measurement of five (bevacizumab; 1, 2, 5, 10, and 20 mg mL1; n ¼ 3) or four (infliximab; 1, 2, 5, and 10 mg mL1; n ¼ 3) spiked samples. For intra-day precision, these levels were analyzed three times each day, whereas for inter-day precision, specimens of the spiked samples at the same concentrations were analyzed three times per day for three days (n ¼ 9). The spiked samples were then analyzed. The minimum acceptable precisions were <20% at 1 mg mL1 and <15% at other concentrations, respectively. 2.7.2. Accuracy The accuracy was determined by the repeated measurement of three levels (1, 5 and 10 mg mL1; n ¼ 3) spiked samples. The minimum acceptable bias were <20% at 1 mg mL1 and 15% at other concentrations, respectively. 2.7.3. Calibration curve For the quantitative analysis, calibration standard solutions (n ¼ 5) with concentrations ranging from 1 to 10 mg mL1L (1, 2, 5, 10, 20 mg mL1) were prepared by diluting the stock solutions. The equations of the calibration curves were determined using least squares linear prediction. The limit of detection (LOD) and the

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119

115

lower limit of quantification (LLOQ) were determined from signalto-noise ratios of 3 and 5, respectively. Concentrations of target mAbs were calculated by the standard addition method.

RPLC analysis of the remaining antibody in the resulting solution, more than 98% of anti-idiotype antibodies were bound to the beads (data not shown).

2.7.4. Recovery The extraction recoveries for both drugs were performed by comparing the analytical results for extracted samples at low (1 mg mL1), medium (5 mg mL1), and high (10 mg mL1) with unextracted standards that represent 100% recovery.

3.3. Evaluation of immunoaffinity purification

3. Results and discussion 3.1. LC separations of antibody preparations HT-RPLC, first reported by Dillon et al. from Amgen, is extensively used for structural characterization and investigation of the heterogeneity of mAbs [24,29,30]. According to the general retention model of proteins in reversed-phase LC [31], proteins are initially adsorbed, then solvated, and then desorbed. In contrast, in HT-RPLC analyses, a widepore Zorbax-300SB C8 column at high temperatures (>70  C) with a mobile phase containing solvents of high eluotropic strength, such as isopropanol and acetonitrile, are utilized. Thus, the adsorption of IgGs onto the stationary phase, refolding of denatured IgGs, and aggregation of IgGs could be greatly reduced. This leads to a good separation of IgGs with sharp peaks. Pore sizes of column packing materials can also affect protein unfolding and subsequent chromatographic properties [32]. Recently, coreeshell packing materials of various diameters and shell thicknesses have become commercially available. The thickness of the porous layer plays a major role in governing particle porosity [33]. More recently, a new 3.6-mm coreeshell wide-pore material (0.2-mm shell thickness) was marketed under the name Aeris Widepore. This column is promising for use in peptide and protein separation methods [34]. The mobile phases used in mAbs separation were the same as previous ones [23]; however, since the separation column was different, the flow rates and gradient program were optimized. Fig. 1 shows chromatograms of the four mAb preparations subjected to HT-RPLC. The preparations showed very similar molecular weights (ca. 150 kDa), and a good separation for the mAbs was achieved within 20 min. Their elution order was trastuzumab (11.6 min), infliximab (11.9 min), tocilizumab (12.3 min), and bevacizumab (12.8 min). These results showed that HT-RPLC could achieve good separation with sharper peaks for the mAb preparations with a short analysis time. 3.2. Preparation of the immunoaffinity magnetic beads

Fluorescence intensity

Immunoaffinity beads were prepared by incubating tosylactivated magnetic beads and each antibody solution (containing 30 mg of anti-idiotype antibody) with shaking for 20 h. From an HT-

0

5

10

15

20

Fig. 1. HT-RPLC chromatograms of four mAb preparations. 1, trastuzumab; 2, infliximab; 3, tocilizumab; and 4, bevacizumab. Each peak corresponds to 30 mg mL1.

To confirm the specificity of the prepared immunoaffinity magnetic beads, plasma samples containing different drugs were spiked at 10 mg mL1 were purified with these beads and their eluate were analyzed by HT-RPLC (Fig. 2). Different drugs and other human IgG could not be captured on the beads and their peaks were not observed on the chromatograms. The prepared beads selectively captured the target drugs (Fig. 3). The immunoaffinity beads were reused at least five times after equilibration with 100 mM sodium phosphate buffer (pH 7.4). In this treatment, a carry-over of target drugs was not observed from the HT-RPLC analysis of the blank plasma and drug-spiked (10 mg mL1) samples. 3.4. Method validation Different concentrations ranging from 1 to 20 mg mL1 (bevacizumab) and 1e10 mg mL1 (infliximab) with five replicates were chosen to draw each calibration curve. The calibration curves for both mAbs were observed between the peak area and analyte concentration with good correlation coefficients (r2 > 0.998). The LOD of bevacizumab and infliximab was 0.07 and 0.15 mg mL1, respectively, and the LLOQ of bevacizumab and infliximab was 0.12 and 0.25 mg mL1, respectively. In a previous study [16], the LLOQ for trastuzumab was 5 mg mL1, despite the large volume of sample injection (13 mL), and had insufficient sensitivity for drug monitoring. The ranges of concentrations tested were chosen based on the drug package inserts; the effective blood concentrations of the drugs ranged from 50 to 500 mg mL1 (bevacizumab) and 10e100 mg mL1 (infliximab). The results of the intra- and inter-day precisions, accuracy and recovery of assays are listed in Table 1. The intra-day assay precisions of bevacizumab and infliximab ranged from 0.4% to 2.6% and 0.8%e4.7%, respectively; and the inter-assay precisions ranged from 2.7% to 7.4% and 4.9%e7.1%, respectively. The bias were 1.4%e1.1% and 1.2e3.1% for bevacizumab and infliximab, respectively. The inter- and the intra-assay precisions of CV < 8% and bias of CV < 4% were within the acceptable limits in both assays. Recoveries for bevacizumab and infliximab ranged from 98.6% to 101.1% and 98.8%e103.1%, respectively. Each drug was captured at almost 100% recoveries with excess amount of antiidiotype antibodies, followed directly analyzed by LC with their native fluorescence. Thus, relationship between drug concentrations and their fluorescence intensities was direct proportional. 3.5. Determination of bevacizumab in plasma samples from cancer patients Fig. 4 shows typical chromatograms of plasma samples obtained from three colon cancer patients who were administered bevacizumab. When the analyzed concentration exceeded the calibration range, plasma samples were diluted appropriately and reanalyzed. Unlike the drug-spiked plasma analysis, several peaks were observed in the chromatograms from the cancer patient samples. However, the peak for bevacizumab was not obscured by other components, and was detected as single, sharp peak. Concentrations of bevacizumab in the three plasma samples were 363.0, 64.5, and 167.7 mg mL1. For comparison, same samples were analyzed at UV 260 nm detection (Supplemental data). Bevacizumab concentrations calculated by standard calibration curve were almost identical to those given by fluorescence detection (concrete data was added into the text). As a result, we confirmed

116

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119

(b)

Fluorescence intensity

(a)

2 1

0

5

(min)

10

15 0

5

(min)

10

15

Fig. 2. Chromatograms of plasma samples spiked with 10 mg mL1 of (a) infliximab and (b) bevacizumab after immuno-affinity purification using each inverted anti-idiotype antibody. Peaks: 1, 10 mg mL1 bevacizumab standard solution; 2, 10 mg mL1 infliximab standard solution.

(b)

Fluorescence intensity

(a)

0

5

(min)

10

15 0

5

(min)

10

15

Fig. 3. Chromatograms of plasma samples spiked with 10 mg mL1 of bevacizumab (a) with and (b) without immuno-affinity purification using anti-idiotype antibody.

Table 1 Intra- and inter-day precisions, accuracy and recovery for (a) bevacizumab and (b) infliximab. Spiked concentration (mg mL1)

(a) 1 2 5 10 20 (b) 1 2 5 10

Precision (%)

Accuracy (Bias, %)

Recovery (%)

7.4 6.1 6.1 4.1 2.7

1.4 e 1.3 þ1.1 e

98.6 98.7 98.7 101.1 99.7

5.8 7.1 4.9 5.3

1.2 e þ3.1 þ1.8

98.8 100.4 103.1 101.8

Intra-day

Inter-day

0.4 0.8 0.8 1.9 2.6 3.3 4.7 4.1 0.8

that fluorescent peak corresponding to bevacizumab was not at least influenced by co-eluting fluorescent contaminants. Obtained values were in good agreement with average Cmax values (323 mg mL1) and trough concentrations (ca. 100 mg mL1) obtained in a clinical analysis of bevacizumab at same injected dose (every month at 8 mg kg1) [10].

3.6. Determination of infliximab in plasma samples from RA patients To investigate the applicability of the proposed method, it was used for the quantification of infliximab in plasma obtained from nine RA patients who were chronically-administered infliximab (for more than six months). Fig. 5 shows typical chromatograms of plasma samples obtained from RA patients. When the analyzed concentration exceeded the calibration range, plasma samples were diluted appropriately and re-analyzed. The concentrations of

Fluorescence intensity

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119

117

1 2

5

(min)

Fluorescence intensity

0

10

15 0

10

15

5

(min)

10

15

3

0

5

(min)

Fig. 4. Chromatograms of plasma samples obtained from colon cancer patients who were administered bevacizumab. Bevacizumab concentrations: 1, 363.0 mg mL1 (50-fold dilution); 2, 64.5 mg mL1 (10-fold dilution); and 3, 167.7 mg mL1 (20-fold dilution).

2

Fluorescence intensity

1

5

Fluorescence intensity

0

(min)

10

15 0

10

15

5

(min)

10

15

3

0

5

(min)

Fig. 5. Typical chromatograms of plasma samples obtained from RA patients who were chronically-administered infliximab. Infliximab concentrations: 1, 7.74 mg mL1; 2, 15.5 mg mL1; and 3, 10.44 mg mL1.

infliximab in nine plasma samples ranged from 0.48 to 10.44 mg mL1 (Table 2), and these values were within the range of average concentrations at 22.3 (15.3e29.4), 14.6 (7.3e22), and 2.8

(0.6e6.8) mg mL1 at 2, 4, and 8 weeks after 3 mg kg1 injections, respectively [35].

118

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119

Table 2 Infliximab concentrations in plasma samples obtained from 9 RA patients. No. A B C D E F G H I

Sex

Age

Disease duration

Infliximab concentration (mgL1)

M F F F F F F F M

57 34 54 55 49 70 67 63 64

8y2m 11 y 8 m 9m 20 y 10 m 4 y 10 m 7 y 10 m 7y9m 24 y 9 m 1y5m

3.31 2.57 0.48 7.74 3.03 3.62 8.75 9.13 10.44

Dosages of Remicade for all patients were 3 mg kg1.

[11]

[12]

[13]

[14]

4. Conclusions In this study, we developed a simple, sensitive, accurate, and rapid analytical method for detecting bevacizumab and infliximab in plasma samples using immunoaffinity purification e HT-RPLC with fluorescence detection. These mAbs were selectively purified from plasma using magnetic beads immobilized with a commercial anti-idiotype antibody for each mAb. Using HT-RPLC with a widepore coreeshell column, the purified mAbs were successfully separated as sharp peaks within 20 min. The sensitivities, precisions, and accuracies of both methods were sufficient for the bioanalysis of plasma from cancer and RA patients. We successfully applied both methods to the bioanalyses of plasma samples obtained from cancer and RA patients who were administered each drug. The methods described here are applicable for planning optimal therapeutic programs, and also for the evaluation of biological equivalencies in the development of biosimilars.

[15]

[16]

[17]

[18]

[19]

[20]

Acknowledgments

[21]

This work was supported by the Research Foundation for the Electrotechnology of Chubu and JSPS KAKENHI Grant Number 25460040.

[22]

[23]

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.aca.2016.02.029.

[24]

References

[25]

[1] A.C. Chan, P.J. Carter, Therapeutic antibodies for autoimmunity and inflammation, Nat. Rev. Immunol. 10 (2010) 301e316. [2] L.M. Weiner, R. Surana, S. Wang, Monoclonal antibodies: versatile platforms for cancer immunotherapy, Nat. Rev. Immunol. 10 (2010) 317e327. [3] M.X. Sliwkowski, I. Mellman, Antibody therapeutics in cancer, Science 341 (2013) 1192e1198. [4] E.D. Lobo, R.J. Hansen, J.P. Balthasar, Antibody pharmacokinetics and pharmacodynamics, J. Pharm. Sci. 93 (2004) 2645e2668. [5] W. Wang, E.Q. Wang, J.P. Balthasar, Monoclonal antibody pharmacokinetics and pharmacodynamics, Clin. Pharmacol. Ther. 84 (2008) 548e558. [6] T. Suzuki, A. Ishii-Watabe, M. Tada, T. Kobayashi, T. Kanayasu-Toyoda, T. Kawanishi, T. Yamaguchi, Importance of neonatal FcR in regulating the serum half-life of therapeutic proteins containing the Fc domain of human IgG1: a comparative study of the affinity of monoclonal antibodies and Fcfusion proteins to human neonatal FcR, J. Immunol. 184 (2010) 1968e1976. [7] C.M. Ng, E. Stefanich, B.S. Anand, P.J. Fielder, L. Vaickus, Pharmacokinetics/ pharmacodynamics of nondepleting anti-CD4 monoclonal antibody (TRX1) in healthy human volunteers, Pharm. Res. 23 (2006) 95e103. [8] N. Hayashi, Y. Tsukamoto, W.M. Sallas, P.J. Lowe, A mechanism-based binding model for the population pharmacokinetics and pharmacodynamics of omalizumab, Br. J. Clin. Pharmacol. 63 (2007) 548e561. [9] C.W.N. Damen, E.R. de Groot, M. Heij, D.S. Boss, J.H.M. Schellens, H. Rosing, J.H. Beijnen, L.A. Aarden, Development and validation of an enzyme-linked immunosorbent assay for the quantification of trastuzumab in human serum and plasma, Anal. Biochem. 391 (2009) 114e120. [10] J.L.G. Bender, P.C. Adamson, J.M. Reid, L. Xu, S. Baruchel, Y. Shaked, R.S. Kerbel,

[26] [27]

[28] [29]

[30]

[31] [32]

[33]

[34]

E.M. Cooney-Qualter, D. Stempak, H.X. Chen, M.D. Nelson, M.D. Krailo, A.M. Ingle, S.M. Blaney, J.J. Kandel, D.J. Yamashiro, Phase I trial and pharmacokinetic study of bevacizumab in pediatric patients with refractory solid tumors: a children's oncology group study, J. Clin. Oncol. 26 (2008) 399e405. A.N. Hoofnagle, M.H. Wener, The fundamental flaws of immunoassays and potential solutions using tandem mass spectrometry, J. Immunol. Methods 347 (2009) 3e11. O. Heudi, S. Barteau, D. Zimmer, J. Schmidt, K. Bill, N. Lehmann, C. Bauer, O. Kretz, Towards absolute quantification of therapeutic monoclonal antibody in serum by LC-MS/MS using isotope-labeled antibody standard and protein cleavage isotope dilution mass spectrometry, Anal. Chem. 80 (2008) 4200e4207. M.F. Ocana, I.T. James, M. Kabir, C. Grace, G. Yuan, S.W. Martin, H. Neubert, Clinical pharmacokinetic assessment of an anti-MAdCAM monoclonal antibody therapeutic by LC-MS/MS, Anal. Chem. 84 (2012) 5959e5967. H. Li, R. Ortiz, L. Tran, M. Hall, C. Spahr, K. Walker, J. Laudernann, S. Miller, H. Salimi-Moosavi, J.W. Lee, General LC-MS/MS method approach to quantify therapeutic monoclonal antibodies using a common whole antibody internal standard with application to preclinical Studies, Anal. Chem. 84 (2012) 1267e1273. M. Dubois, F. Fenaille, G. Clement, M. Lechmann, J.-C. Tabet, E. Ezan, F. Becher, Immunopurification and mass spectrometric quantification of the active form of a chimeric therapeutic antibody in human serum, Anal. Chem. 80 (2008) 1737e1745. Y. Cong, Z. Zhang, S. Zhang, L. Hu, J. Gu, Quantitative MS analysis of therapeutic mAbs and their glycosylation for pharmacokinetics study, Proteomics Clin. Appl. (2015), http://dx.doi.org/10.1002/prca.201500098. W.S. Law, J.-C. Genin, C. Miess, G. Treton, A.P. Warren, P. Lloyd, S. Dudal, C. Krantz, Use of generic LCeMS/MS assays to characterize atypical PK profile of a biotherapeutic monoclonal antibody, Bioanalysis 6 (2014) 3225e3235. D. Lebert, G. Picard, C. Beau-Larvor, L. Troncy, C. Lacheny, B. Maynadier, W. Low, N. Mouz, V. Brun, C. Klinguer-Hamour, M. Jaquinod, A. Beck, Absolute and multiplex quantification of antibodies in serum using PSAQ™ standards and LC-MS/MS, Bioanalysis 7 (2015) 1237e1251. K. Peng, K. Xu, L. Liu, R. Hendricks, R. Delarosa, R. Erickson, N. Budha, M. Leabman, A. Song, S. Kaur, S.K. Fischer, Critical role of bioanalytical strategies in investigation of clinical PK observations, a Phase I case study, mAbs 6 (2014) 1500e1508. K.J. Bronsema, R. Bischoff, W.W.M. Pim Pijnappel, A.T. van der Ploeg, N.C. van de Merbel, Absolute quantification of the total and antidrug antibody-bound concentrations of recombinant human a-glucosidase in human plasma using protein G extraction and LC-MS/MS, Anal. Chem. 87 (2015) 4394e4401. B. An, M. Zhang, R.W. Johnson, J. Qu, Absolute quantification of the total and antidrug antibody-bound concentrations of recombinant human a-glucosidase in human plasma using protein G extraction and LC-MS/MS, Anal. Chem. 87 (2015) 4023e4029. Y. Shen, G. Zhang, J. Yang, Y. Qiu, T. McCauley, L. Pan, J. Wu, Online 2D-LC-MS/ MS assay to quantify therapeutic protein in human serum in the presence of pre-existing antidrug antibodies, Anal. Chem. 87 (2015) 8555e8563. C.W.N. Damen, E.J.B. Derissen, J.H.M. Schellens, H. Rosing, J.H. Beijnen, The bioanalysis of the monoclonal antibody trastuzumab by high-performance liquid chromatography with fluorescence detection after immuno-affinity purification from human serum, J. Pharm. Biomed. Anal. 50 (2009) 861e866. T.M. Dillon, P.V. Bondarenko, M.S. Ricci, Development of an analytical reversed-phase high-performance liquid chromatography-electro spray ionization mass spectrometry method for characterization of recombinant antibodies, J. Chromatogr. A 1053 (2004) 299e305. S.-L. Wang, L. Ohrmund, S. Hauenstein, J. Salbato, R. Reddy, P. Monk, S. Lockton, N. Ling, S. Singh, Development and validation of a homogeneous mobility shift assay for the measurement of infliximab and antibodies-toinfliximab levels in patient serum, J. Immunol. Methods 382 (2012) 177e188. A. Beck, J.M. Reichert, Approval of the first biosimilar antibodies in Europe A major landmark for the biopharmaceutical industry, Mabs 5 (2013) 621e623. A. Beck, F. Debaene, H. Diemer, E. Wagner-Rousset, O. Colas, A.V. Dorsselaerb, ranib, Cutting-edge mass spectrometry characterization of originator, S. Cianfe biosimilar and biobetter antibodies, J. Mass Spectrom. 50 (2015) 285e297. FDA, CDER, Guidance for Industry: Bioanalytical Method Validation, US Department of Health and Human Services, 2001. T.M. Dillon, P.V. Bondarenko, D.S. Rehder, G.D. Pipes, G.R. Kleemann, M.S. Ricci, Optimization of a reversed-phase high-performance liquid chromatography/mass spectrometry method for characterizing recombinant antibody heterogeneity and stability, J. Chromatogr. A 1120 (2006) 112e120. H. Liu, G. Gaza-Bulseco, E. Lundell, Assessment of antibody fragmentation by reversed-phase liquid chromatography and mass spectrometry, J. Chromatogr. B-Analytical Technol. Biomed. Life Sci. 876 (2008) 13e23. X.D. Geng, F.E. Regnier, Rention model for proteins in reversed-phase liquid chromatography, J. Chromatogr. 296 (1984) 15e30. J.L.M. McNay, E.J. Fernandez, Protein unfolding during reversed-phase chromatography: I. Effect of surface properties and duration of adsorption, Biotechnol. Bioeng. 76 (2001) 224e232. F. Gritti, I. Leonardis, J. Abia, G. Guiochon, Physical properties and structure of fine core-shell particles used as packing materials for chromatography relationships between particle characteristics and column performance, J. Chromatogr. A 1217 (2010) 3819e3843. S. Fekete, R. Berky, J. Fekete, J.-L. Veuthey, D. Guillarme, Evaluation of a new

K. Todoroki et al. / Analytica Chimica Acta 916 (2016) 112e119 wide pore core-shell material (Aeris WIDEPORE) and comparison with other existing stationary phases for the analysis of intact proteins, J. Chromatogr. A 1236 (2012) 177e188. [35] G.J. Wolbink, A.E. Voskuyl, W.F. Lems, E. de Groot, M.T. Nurmohamed, P.P. Tak,

119

B.A.C. Dijkmans, L. Aarden, Relationship between serum trough infliximab levels, pretreatment C reactive protein levels, and clinical response to infliximab treatment in patients with rheumatoid arthritis, Ann. Rheumatic Dis. 64 (2005) 704e707.