WO2017050825A1

WO2017050825A1 - A method for quantifying bevacizumab

Info

Publication number: WO2017050825A1
Application number: PCT/EP2016/072421
Authority: WO
Inventors: Rachel LEGERON; Fabien XUEREB; Dominique BREILH; Alain Pierre GADEAU; Jean-Michel Boiron; Jean-Marie Schmitter; Jean-William DUPUY; Stéphane CHAIGNEPAIN
Original assignee: Institut National De La Sante Et De La Recherche Medicale (Inserm); Université De Bordeaux; Chu De Bordeaux; Centre National De La Recherche Scientifique
Priority date: 2015-09-21
Filing date: 2016-09-21
Publication date: 2017-03-30

Abstract

The present invention relates to the use of an Internal Standard compound in a method for quantifying Bevacizumab in a sample by mass spectrometry, wherein the said Internal Standard compound is described in the specification.

Description

A METHOD FOR QUANTIFYING BEVACIZUMAB Field of the invention

This invention relates to the field of antibody quantification. It more precisely relates to the quantification of Bevacizumab in a biological sample.

Background of the invention

Monoclonal antibodies (mAbs) constitute a therapeutic class which knows the strongest current rate of development in the field of pharmaceutical biotechnology. There are to date more than 25 mAbs marketed in various fields such as oncology, immunology, ophthalmology and cardiology. Among them, Bevacizumab (Avastin^®) is an IgGl monoclonal antibody directed against the vascular endothelial growth factor (VEGF) and neutralizes the biological activity of VEGF, resulting in anti-angiogenic activity (Presta, L.G., et al., Cancer Res, 1997. 57(20): p. 4593-9.). Several clinical phase II and III studies have shown the interest of the combination of Bevacizumab with standard chemotherapy fluoropyrimidine-based compared to chemotherapy alone in patients with metastatic colorectal cancer (mCRC) (Hurwitz, H., et al., N Engl J Med, 2004. 350(23): p. 2335-42.; Saltz, L.B., et al., J Clin Oncol, 2008. 26(12): p. 2013-9.; Giantonio, B.J., et al., J Clin Oncol, 2007. 25(12): p. 1539-44.). These large randomized trials have demonstrated it benefit in terms of objective response rate (ORR), progression-free survival (PFS) and / or overall survival (OS) thus making Bevacizumab a very important drug in the therapeutic management of mCRC both in the first-line and in second-line treatment. Bevacizumab is also used in the first line treatment of metastatic breast cancer, advanced or metastatic non- small cell lung cancer and advanced or metastatic kidney cancer always in combination with chemotherapy. Although Bevacizumab significantly improved the therapeutic management of mCRC, it is characterized by inter-individual variability in clinical response (efficacy and safety) as well as other mAbs. However the causes of this inter- individual clinical variability remain poorly understood (Gao, B., et al., J Clin Oncol, 2012. 30(32): p. 4017-25.; Lu, J.F., et al., Cancer Chemother Pharmacol, 2008. 62(5): p. 779-86; Wu, J.Y., et al., Chin Med J (Engl), 2010. 123(7): p. 901-6). To better understand why some patients respond to treatment and others not, why some have more side effects than others, several factors were studied. It has already been demonstrated that pharmacokinetic (PK) variability that characterized monoclonal antibodies (Tabrizi, M.A., CM. Tseng, and L.K. Roskos, Drug Discov Today, 2006. 11(1-2): p. 81-8) could be a cause of response variability to treatment. Evaluation of mAb concentrations in biological fluids is an essential prerequisite for the determination of pharmacokinetic parameters to assess the relationship between response and drug exposure. In this context, it is necessary to develop quantification methods sufficiently sensitive and specific to ensure accurate and reliable measurement of expected therapeutic concentrations in plasma between 75 μg/ml and 270 μ^πύ.

Given the proteic nature of mAbs, their high degree of homology with the endogenous antibody and the low concentrations at which they are expected in the plasma environment, the determination of plasma concentrations of monoclonal antibodies is difficult. To date, reference techniques to identify and quantify mAbs and particularly Bevacizumab rely on enzyme-linked immunosorbent assay methods (ELISA) due to their high sensitivity and recover. However ELISA methods present important limits especially in terms of reliability. Indeed, given their high sensitivity, linearity of response does not cover the full range of expected therapeutic concentration in plasma which requires dilution of the sample and can lead to damage accuracy and precision of measurements concentrations (Tabrizi, M.A., CM. Tseng, and L.K. Roskos, Drug Discov Today, 2006. 11(1-2): p. 81-8; Yang, Z., et al.,. Anal Chem, 2007. 79(24): p. 9294-301). In addition, the specificity of ELISA techniques can be affected by cross-reactions with other molecules present in biological matrices and can generate false negatives or false positives (van den Broek, I., W.M. Niessen, and W.D. van Dongen,. J Chromatogr B Analyt Technol Biomed Life Sci, 2013. 929: p. 161-79). Moreover, Bevacizumab can be found in three different forms in the human plasma: free, bound and partially bound to its target, VEGF. The choice and the knowledge of the measured fraction are crucial for the interpretation of the plasma concentrations achieved and pharmacokinetic data resulting. The ELISA kit currently marketed for the Bevacizumab dosing plasma assesses only the free fraction of the drug. However, it seems that storage and analysis conditions of plasma samples alter the physiological equilibrium established in vivo between the different plasmatic forms of mAb, thus leading to overestimate or underestimate the measured concentrations when quantification is conducted exclusively on free fraction. An analytical method that does not disturb the physiological equilibrium would be ideal. Liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) was then viewed as an ideal candidate given its high specificity, repeatability and accuracy of the measurement. LC-MS/MS allows more reliability analysis than immunoassay technique while maintaining a sufficient sensitivity for antibody quantification in human plasma. LC-MS/MS which is already widely used for the quantification of small molecules is a growing analytical tool for analysis of therapeutic proteins and proteomic biomarkers. The specificity of LC-MS/MS detection is achieved through the analysis in multiple reaction monitoring (MRM) mode of intact protein or of a surrogate peptide of the protein of interest obtained after enzymatic proteolysis of samples (Duan, X., et al., J Chromatogr A, 2012. 1251: p. 63-73; Dubois, M., et al.,. Anal Chem, 2008. 80(5): p. 1737-45; Fernandez Ocana, M., et al., Anal Chem, 2012. 84(14): p. 5959-67; Heudi, O., et al., Anal Chem, 2008. 80(11): p. 4200-7; Lesur, A., E. Varesio, and G. Hopfgartner, J Chromatogr A, 2010. 1217(1): p. 57-64; Liu, H., et al.,. Anal Biochem, 2011. 414(1): p. 147-53).

Finally, given that the steps of pre-treatment of plasmatic sample and analysis by mass spectrometry can have several different sources of variability, internal calibration approaches are necessary and have been developed to correct theses bias and improve accuracy of the assay. All of the approaches are based on the addition in fixed and known amount of an internal standard (IS). Since MS detection of mAb involves peptides derived from proteolysis of the protein, stable isotope labeled (SIL) or structural analogue of either the intact protein or the signature peptide are generally used as IS (Yang, Z., et al., Anal Chem, 2007. 79(24): p. 9294-301; Dubois, M., et al., Anal Chem, 2008. 80(5): p. 1737-45, Fernandez Ocana, M., et al., Anal Chem, 2012. 84(14): p. 5959-67, Heudi, O., et al., Anal Chem, 2008. 80(11): p. 4200-7). One of the widely described phenomena could lead to variability in quantification especially in electrospray ionisation (ESI) mode is ion suppression source caused by matrix components (Stokvis, E., H. Rosing, and J.H. Beijnen, Rapid Commun Mass Spectrom, 2005. 19(3): p. 401-7). The degree of ion suppression may vary greatly from one matrix to another and depends largely on the chemical structure of the analyte. However this phenomenon is difficult to control and may vary during day to day analysis especially due to several parameters such as the temperature and pressure in the source.

SIL of signature peptides are often used to correct variability resulting from ions suppression phenomenon and also chemical degradation of peptide. The alternative to synthesize the stable isotope labeled of the intact protein is difficult to implement given the complexity of the chemical structure of protein, and recourse to production in a same way of the analyte is very expensive and time consuming.

There remains a need for a method aimed at precisely quantifying Bevacizumab, especially Bevacizumab contained in a biological fluid such as plasma, which method would be alternative or improved as compared to the Bevacizumab quantification methods that are already available in the art.

Considering that it is difficult to maintain the physiological equilibrium between the different forms of Bevacizumab in a sample containing a mixture of these forms, there is a need for a method for quantifying the total fraction of Bevacizumab. Such a quantification method would overcome the variability associated with ex vivo perturbation of the in vivo equilibrium making pharmacokinetic analysis of Bevacizumab more reliable.

Summary of the invention

The present invention relates to a method for quantifying Bevacizumab in a biological sample. This invention notably concerns a method for quantifying Bevacizumab in a sample originating from an individual that has been administered the Bevacizumab antibody.

The Bevacizumab quantification method of the invention is aimed notably at assessing the pharmacokinetic profile of this therapeutic antibody in individuals in need thereof.

The quantification method described throughout the present specification fully complies with the relevant FDA guidelines for bioanalytical methods validation and allows assessing plasmatic concentration of a total fraction of bevacizumab, which includes free bevacizumab and bevazicumab bound to plasma proteins, such as bevacizumab bound to its target VEGF.

More precisely, it is described herein the use of an Internal Standard compound in a method for quantifying Bevacizumab in a sample by mass spectrometry, wherein the said Internal Standard compound is selected in a group comprising :

- a labeled surrogate peptide of bevacizumab selected in a group comprising the peptides of SEQ ID NO 1 (FTFSLDTSK), SEQ ID NO 2 (STAYLQMNSLR), and SEQ ID NO 3 (VLIYFTSSLHSGVPSR), or - a monoclonal antibody which generates upon trypsin proteolysis one or more surrogate peptides selected in a group comprising the peptides of SEQ ID NO 4 (FTISADTSK), SEQ ID NO 5 (DTYIHWVR) and SEQ ID NO 6 (NTA YLQMNSLR) .

In some embodiments, the said Internal Standard compound is used in a quantification method of Bevacizumab in a sample to improve accuracy and precision of the method comprising the steps of :

a) subjecting to trypsin proteolysis a pre -proteolysis mixture consisting of the said sample in which is present a known and fixed amount of the said Internal Standard compound, whereby an assay mixture is provided, and wherein the said assay mixture comprises

- proteolysis peptides derived from Bevacizumab, which proteolysis peptides comprise surrogate proteolysis peptides derived from Bevacizumab, and

- the said labeled surrogate peptide of bevacizumab when it is used as the Internal Standard compound or proteolysis peptides derived from the said monoclonal antibody when the said monoclonal antibody is used as the Internal

Standard compound,

b) determining by mass spectrometric analysis the ratio of (i) a selected surrogate proteolysis peptide derived from Bevacizumab to (ii) a surrogate peptide of internal standard used selected in a group comprising (ii-a) the said labeled surrogate peptide of bevacizumab or (ii-b) a selected surrogate proteolysis peptide derived from the said monoclonal antibody, and

c) calculating from the ratio determined at step b) the amount of Bevacizumab in the said sample.

In some embodiments wherein the Internal Standard compound is a monoclonal antibody, the said antibody preferably consists of Trastuzumab.

In some embodiments of step b) of the quantification method above, (i) the said surrogate proteolysis peptide derived from Bevacizumab is a peptide of SEQ ID N° 1 and (ii-a) the surrogate peptide used as an Internal Standard is the labeled peptide of SEQ ID N° l .

In some embodiments of step b) of the quantification method above, (i) the said surrogate proteolysis peptide derived from Bevacizumab is a peptide of SEQ ID N° 1 and (ii-b) the surrogate proteolysis peptide from Internal Standard monoclonal antibody is a peptide of SEQ ID N° 5.

In some embodiments of step a) of the quantification method above, trypsin is incubated during a time period ranging from 7 to 9 hours, and is preferably of about 8 hours.

Description of the figures

Figure 1. Fragmentation spectra of the doubly charged precursor ion of the two surrogate peptides of bevacizumab ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) (Figure 1A) and ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) (Figure IB) of the SIL surrogate peptide ⁶⁸FTFSLDTSK⁷⁶* (SEQ ID NO 1) (Figure 1C) and of the surrogate peptide of mAbIS ³¹DTYIHWVR³⁸ (SEQ ID NO 5) (Figure ID). All of the fragmentation spectra achieved after LC-ESI-QqQ analysis were compared against in silico spectra generated from Protein Prospector online bioinformatics tool.

Figure 2. LC-MS/MS chromatogram obtained after analysis of human serum sample spiked with 40( g/ml of bevacizumab (QC3) and with SIL peptide just before proteolysis (Figures 2A, 2B, 2C) or mAbIS before any step of treatment (Figure 2A, 2B, 2D). Analysis was conducted on the LC-ESI-QqQ system under chromatogram and mass spectrometry conditions described in experimental section.

Chromatographic peaks correponding to the sum of the most intense SRM transitions of the doubly charged surrogate peptide of bevacizumab ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1 - Figure 2A) and ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2 - Figure 2B) , the doubly charged SIL peptide ⁶⁸FTFS LDTS K⁷⁶ * (SEQ ID NO 1 - Figure 2C) and the doubly charged surrogate peptide of mAbIS DTYIHWVR (SEQ ID NO 5 - Figure 2D).

Figure 3. Assessment of bevacizumab and mAbIS trypsin proteolysis as function of trypsin/total protein ratio through measurement of mean peak area (AUC) corresponding to the two surrogate peptides of bevacizumab (Figure 3a), (Figure 3b) and the surrogate peptide of mAbIS (Figure 3c). QC1, QC2 and QC3 samples spiked with mAbIS were analyzed by LC-ESI-QqQ analysis under chromatographic separation and mass spectrometry detection described in experimental section. Results were expressed as mean AUC +sd (n=3). Figure 4. Assessment of bevacizumab and mAbIS trypsin proteolysis as function of time of trypsin incubation (4 hours, 6 hours and 8 hours and 16 hours) through formation of the two surrogate peptides of bevacizumab (-■- FTFSLDTSK (SEQ ID NO 1), -♦- STAYLQMNSLR (SEQ ID NO 2) and the surrogate peptide of mAbIS (- A- DTYIHWVR (SEQ ID NO 5)). Chromatographic separation and mass spectrometry detection were performed as described in experimental section. Results were expressed as mean AUC (n=6).

Figure 5. LC-MS/MS chromatograms issued from the MRM analysis of blank human plasma sample (Figure 5A), human plasma sample spiked with trastuzumab and SIL peptide (Figure 5B), human plasma sample spiked with bevacizumab and SIL peptide (Figure 5C), human plasma sample spiked with bevacizumab and trastuzumab (Figure 5D). All sample were treated according to the process described in the examples (protein A purification and concentration by ultra filtration) and submitted to tryptic proteolysis step. Detailed description of the invention

This invention provides for a quantification method of Bevacizumab in a sample. The invention's quantification method of Bevacizumab may be especially useful for determining the pharmacokinetic profile of Bevacizumab in individuals that are administered with this specific therapeutic antibody. Determination of pharmacokinetic profile of bevacizumab might be necessary in pharmacokinetic/pharmacodynamic studies exploring inter and intra-individual variability in clinical response, in comparative studies of administration way or also in pharmacokinetics studies of biosmiliars.

The invention thus relates to the use of an Internal Standard compound in a method for quantifying Bevacizumab in a sample by mass spectrometry, wherein the said Internal Standard compound is selected in a group comprising:

- a labeled peptide selected in a group comprising the peptides of SEQ ID NO 1, SEQ ID NO 2, and SEQ ID NO 3, and

- a monoclonal antibody which generates upon trypsin proteolysis surrogate peptides selected in a group comprising the peptides of SEQ ID NO 4, SEQ ID NO 5 and SEQ ID NO 6. The invention also relates to a method for quantifying Bevacizumab in a sample by mass spectrometry, said method comprising the use of an Internal Standard compound selected in a group comprising:

- a monoclonal antibody which generates upon trypsin proteolysis surrogate peptides selected in a group comprising the peptides of SEQ ID NO 4, SEQ ID NO 5 and SEQ ID NO 6.

In particular, the method for quantifying Bevacizumab in a sample comprises the steps of :

a) subjecting to trypsin proteolysis a pre-proteolysis mixture consisting of the said sample in which is present a fixed and known amount of the said Internal Standard compound , whereby an assay mixture is provided, and wherein the said assay mixture comprises

- proteolysis peptides derived from Bevacizumab, and

- the said labeled surrogate peptide of bevacizumab when it is used as the Internal Standard compound or proteolysis peptides derived from the said monoclonal antibody when the said monoclonal antibody is used as the Internal Standard compound, b) determining by mass spectrometric analysis the ratio of (i) a selected proteolysis surrogate peptide derived from Bevacizumab to (ii) a surrogate peptide of

Internal Standard selected in a group comprising (ii-a) the said labeled surrogate peptide of bevacizumab and (ii-b) a proteolysis surrogate peptide derived from the internal standard monoclonal antibody, and

In one embodiment of the invention, said Internal Standard compound is a Stable Isotope Labeled (SIL) peptide selected in a group comprising the peptides of SEQ ID N° 1, SEQ ID N°2, and SEQ ID N°3.

In another embodiment, said Internal Standard compound consists of a monoclonal antibody such as Trastuzumab. In one embodiment of the use described herein, at step b) of the method, the said surrogate proteolysis peptide derived from Bevacizumab is selected in a group comprising the peptides of SEQ ID N° 1, SEQ ID N°2, and SEQ ID N°3.

In particular, at step b) of the said method, the selected proteolysis surrogate peptide derived from Bevacizumab is a peptide of SEQ ID N° 1.

Still particularly, at step b) of the said method, (i) the selected proteolysis surrogate peptide derived from Bevacizumab is a peptide of SEQ ID N° 1 and (ii) the surrogate peptide of Internal Standard is a labeled peptide of SEQ ID N° l .

In one embodiment of the use described herein, at step a) of the method, trypsin is added at a trypsin / total protein molar ratio ranging from 1/100 to 1/10, advantageously from 1/75 to 1/25 and preferably froml/60 to 1/40.

In one embodiment of the use described herein, at step a) of the method, trypsin is incubated during a time period ranging from 7 to 9 hours, and is preferably of about 8 hours.

In one embodiment of the use described herein, the said sample is selected in a group comprising whole blood, plasma and serum, or a sample derived therefrom.

In one embodiment of the use described herein, the said sample consists of a sample derived from a human organism. The inventors have shown that a precise quantification of Bevacizumab in a sample, which may be also termed " test sample" herein, may be allowed through the design of a method wherein the amount of Bevacizumab, if present, in the said sample is determined by a mass spectrometry method after:

A) calculation of a ratio of

(i) the spectrometry signal generated by a selected surrogate proteolysis peptide from

Bevacizumab

to

(ii) a surrogate peptide selected in a group comprising :

(ii-a) a peptide analog consisting of a labeled form of the selected surrogate proteolysis peptide from Bevacizumab, which labeled peptide is used as an Internal

Standard compound, or (ii-b) a surrogate proteolysis peptide generated upon trypsin proteolysis of a monoclonal antibody which is used as an Internal Standard compound (e.g. Trastuzumab ), and

B) determination of the amount of Bevacizumab, if present, in the said sample by reporting the calculated ratio for the said test sample to a calibration curve of ratio values.

The kind of the Internal Standard compound(s) that is (are) used strongly contributes to the accuracy and precision of the Bevacizumab quantification method that has been designed by the inventors and that is described herein.

According to other aspects, the accuracy and precision of the Bevacizumab quantification method described herein may also be provided by a combination of various parameters, which include (i) the kind of method steps, (ii) the order in which the successive method steps are carried out and (ii) the specific reagents that are used, which encompasses the specific Internal Standard compounds used.

More precisely, the inventors have designed two kinds of highly specific

Internal Standard compounds that may be used for improving accuracy and precision of the absolute quantification of Bevacizumab in a sample by a method wherein quantification is performed by mass spectrometry, (i) a monoclonal antibody, especially Trastuzumab and (ii) labeled peptides of the same amino acid sequence as non-labeled surrogate peptides generated by trypsin proteolysis of Bevacizumab.

The present invention pertains to the use of an Internal Standard compound in a method for improving accuracy and precision of the absolute quantification of Bevacizumab in a sample by mass spectrometry, wherein the said Internal Standard compound is selected in a group comprising:

- a labeled surrogate peptide of bevacizumab selected in a group comprising the peptides of SEQ ID N° 1 (FTFSLDTSK), SEQ ID N°2 (STA YLQMNSLR) , and SEQ ID N°3 (VLIYFTSSLHSGVPSR), and

- a monoclonal antibody which generates upon trypsin proteolysis one or more surrogate peptides selected in a group comprising the peptides of SEQ ID N° 4 (FTISADTSK), SEQ ID N°5 (DTYIHWVR) and SEQ ID N° 6 (NTA YLQMNSLR).

In preferred embodiments, the said monoclonal antibody is Trastuzumab. Bevacizumab is a monoclonal antibody that has been widely described in the art. Notably, Bevacizumab is contained in the pharmaceutical composition marketed under the name of Avastin . Bevacizumab may be used for various therapeutic purposes, including for treating colorectal cancer, breast cancer, lung cancer, renal cancer, brain cancer, gynecological cancers (e.g. ovarian cancer) as well as eye diseases such as Age- related Macular Degeneration (AMD) and diabetic retinopathy. As used herein, Bevacizumab is a monoclonal antibody directed against VEGF comprising:

- two identical light chains comprising the amino acid sequence of the following SEQ ID NO 7 :

DIQMTQSPSSLSASVGDRVTITCSASQDISNYLNWYQQKPGKAPKVLIYFTSSLH SGVPS

RFSGSGSGTDFTLTISSLQPEDFATYYCQQYSTVPWTFGQGTKVEIKRTVAAPSV FIFPP

SDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDST YSLSSTLT

LSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC, and

- two identical heavy chains comprising the amino acid sequence of the following SEQ ID NO 8 :

EVQLVESGGGLVQPGGSLRLSCAASGYTFTNYGMNWVRQAPGKGLEWVGWI NTYTGEPTY

AADFKRRFTFSLDTSKSTAYLQMNSLRAEDTAVYYCAKYPHYYGSSHWYFDV WGQGTLVT

VSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVH TFPAVL QSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPP CPAPEL

LGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNA KTKPREE

QYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQ VYTLPPS

REEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYS KLTVDK SRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

As shown in the examples herein, this invention provides for specific Internal Standard compounds allowing an accurate and precise quantification of Bevacizumab in a sample by mass spectrometry.

In some embodiments, the one skilled in the art may perform any of the protein quantification methods by mass spectrometry which are known in the art, and especially any of the monoclonal antibody quantification methods which are known in the art, which include those which are referred to in the present specification, including in the description of the prior art methods.

Generally, the one or more Internal Standard compounds that are described herein are used in monoclonal antibody quantification methods by mass spectrometry where a known amount of one of the two kinds of Internal Standard compounds is added at a step preceding the step of quantification by mass spectrometry.

The amount of Bevacizumab in the initial sample is finally determined against a calibration curve obtained by a method comprising the steps of :

(a) preparing a serial or a set of calibration samples (CS) comprising

(a-1) a fixed and known amount of an Internal Standard compound selected in a group comprising

- a labeled surrogate peptide of bevacizumab selected in a group comprising the peptides of SEQ ID NO 1 (FTFSLDTSK), SEQ ID NO 2 (STAYLQMNSLR), and SEQ ID NO 3 (VLIYFTSSLHSGVPSR), and

- a monoclonal antibody which generates upon trypsin proteolysis one or more surrogate peptides selected in a group comprising the peptides of SEQ ID NO 4 (FTISADTSK), SEQ ID NO 5 (DTYIHWVR) and SEQ ID NO 6

(NTAYLQMNSLR) in the embodiment wherein Trastuzumab is used as Internal Standard,

(a-2) a serial or a set of increasing amounts of Bevacizumab,

(b) subjecting each of the calibration sample prepared at step (a) to trypsin proteolysis, whereby a serial or a set of calibration assay samples (CAS) is provided,

(c) determining by mass spectrometric analysis, in each of the said calibration assay samples (CAS), the ratio of (i) the said selected surrogate proteolysis peptide derived from Bevacizumab to (ii) a surrogate peptide selected in a group comprising (ii-a) the said labeled surrogate peptide of bevacizumab when it is used as the Internal Standard compound or (ii-b) a proteolysis surrogate peptide derived from the said monoclonal antibody when the said monoclonal antibody is used as the Internal Standard compound, whereby a serial of ratio values is provided, and

(d) generating a calibration curve from the serial of ratio values that is provided at step (c).

As used herein, the term "test sample" or "sample tested" encompasses material that originates from a body fluid that was previously collected from an individual, especially from a human individual. As used herein, a "test sample" encompasses any material originating from an individual to which Bevacizumab has been administered one or more times.

In some embodiments, a test sample originates from a body fluid selected in a group comprising whole blood, plasma and serum.

In some embodiments, a test sample derives from a body fluid after dilution in an aqueous solution, e.g. a saline solution or a buffer solution.

However, in most preferred embodiments, a test sample is not subjected to any pretreatment, and particularly is used without being subjecting to a dilution step, for performing the Bevacizumab quantification method described in the present specification.

Internal Standard compounds for quantifying Bevacizumab

Two kinds of Internal Standard compounds are provided herein for improving accuracy and precision of the absolute quantification of Bevacizumab, labeled surrogate peptide of bevacizumab and monoclonal antibodies, respectively.

As shown in the examples herein, the high specificity against endogenous plasmatic proteins and the high physico-chemical homology between (i) these Internal Standard compounds or proteolysis peptides generated therefrom, and (ii) Bevacizumab or proteolysis peptides generated therefrom, allows a highly precise quantification of Bevacizumab in a sample.

Labelled peptides as Internal Standard compounds As shown in the examples herein, the peptides of SEQ ID NO 1, 2 and 3 consist of peptides that are generated after subjecting Bevacizumab to proteolysis by trypsin. These peptides have been shown herein to be unique and specific to Bevacizumab. Notably, it is shown herein that the peptides of any of SEQ ID NO 1, 2 or 3 are not found in a collection of tryptic peptides obtained after having subjected a composition of human polyvalent IgG antibodies to proteolysis by trypsin. Further, by performing an in silico analysis which consisted of an alignment of each of the amino acid sequences of SEQ ID NO 1, 2 or 3 against sequences of peptides generated by trypsin proteolysis of the known proteins of the human proteome, the inventors have also shown that there it was unlikely that there was homologous sequence to any of the peptides of SEQ ID NO 1, 2 or 3 (See Table 2 herein).

It is herein specified that trypsin proteolysis of Bevacizumab generates in silico a total of 54 distinct tryptic peptides without missed cleavage by trypsin. Among them 25 tryptic peptides without missed cleavage were identified after experimental analysis.

It is further specified that the selected peptides of SEQ ID NO 1, 2 and 3 are the Bevacizumab-derived surrogate peptides having no cystein in their sequence with the lowest Bit score from the in silico sequence alignment. The particularity of the SEQ ID NO

1 and 2 is the most sensitive mass spectrometry signal in multiple reaction monitoring (MRM) acquisition mode.

Still further, the uniqueness of each of the Bevacizumab tryptic peptides of SEQ ID NO 1,

2 and 3 has allowed their use, after their labelling, as Internal Standard compounds in a Bevacizumab quantification method by mass spectrometry.

The peptides of SEQ ID NO 1, 2 or 3 may be labelled according to various methods known in the art, provided that the labelled peptide is discriminated from its non- labelled counterpart in a mass spectrometry analysis method.

Preferably, the peptides of SEQ ID NO 1, 2 or 3 are labelled with one or more stable isotopes. Stable isotopes may be selected in a group comprising 2 H, 13 C, 15 N and

18 O. Preferably, stable isotopes are selected in a group comprising 1 ⁱ3^JC and 1¹5^JN.

A Stable Isotope Labelled peptide, due to a sufficient mass increment relative to the same but unlabeled peptide, is thus discriminated from the said unlabeled peptide by mass spectrometry analysis. Otherwise said, a Stable Isotope Labelled peptide selected in a group comprising the surrogate peptides of SEQ ID NO 1, 2 and 3 is discriminated by mass spectrometry analysis, from the non-labelled surrogate peptides of the same respective amino acid sequences that are generated upon trypsin treatment of Bevacizumab.

The Stable Isotope Labelled peptides of SEQ ID NO 1, 2 and 3 may also be termed "labelled surrogate peptides of bevacizumab" in the present specification.

In some embodiments, any of the surrogate peptides of SEQ ID NO 1, 2 and 3, when used as Internal Standard compounds in a Bevacizumab quantification method by mass spectrometry, may be labelled by both the 13 C and 15 N isotopes.

Stable Isotope Labelled (SIL) peptides may be synthesized according to techniques well known in the art. Illustratively, the SIL peptides may be obtained from JPT

Peptide Technologies GmbH (Berlin, Germany).

In some preferred embodiments of step b) of the method, the proteolysis surrogate peptide derived from Bevacizumab is a peptide of SEQ ID N° 1.

In some of these preferred embodiments of step b) of the method, (i) the selected proteolysis surrogate peptide derived from Bevacizumab is a peptide of SEQ ID

N° 1 and (ii) the surrogate peptide of Internal Standard is a labelled surrogate peptide of

SEQ ID N⁰1. Monoclonal antibodies as Internal Standard compounds

In other embodiments, the Internal Standard compound which is used may be a monoclonal antibody, and preferably a humanized IgG monoclonal antibody, which shall not be present in combination with Bevacizumab in the starting sample used for Bevacizumab quantification, e.g. which shall not be present initially in a plasma sample of a patient who received Bevacizumab chemotherapy.

Satisfying all these conditions, Trastuzumab was preferably selected as a monoclonal antibody Internal Standard compound. As shown in the examples herein, proteolysis by trypsin of the selected monoclonal antibody Trastuzumab notably generates the three peptides of SEQ ID NO 4, 5 and 6.

It is specified that the selected peptides of SEQ ID NO 4, 5 and 6 are the monoclonal antibody-derived peptides, and especially the Trastuzumab-derived surrogate peptides having no cystein in their sequence with the lowest Bit score from the in silico sequence alignment. The particularity of the SEQ ID NO 5 is the most sensitive mass spectrometry signal in MRM acquisition mode.

The peptides of SEQ ID NO 4, 5 and 6 may also be termed "surrogate peptides of trastuzumab" in the present specification.

In some embodiments, a specific surrogate peptide is selected in a group comprising the peptides of SEQ ID NO 4, 5 and 6.

By performing an in silico analysis which consisted of an alignment of sequence of SEQ ID NO 4, 5 or 6 against protein sequences of the human proteome subjected to trypsin proteolysis, the inventors have also shown that it was unlikely that there was sequence homologous to peptides of SEQ ID NO 4, 5 or 6 (See Table 2 herein).

Trastuzumab is a monoclonal antibody directed against HER-2/Neu and is widely known in the art. Trastuzumab is used as an active ingredient for treating breast and gastric cancers. Trastuzumab is contained in the pharmaceutical composition marketed under the name of Herceptin .

As used herein, Trastuzumab is a monoclonal antibody comprising : - a first light chain comprising the amino acid sequence of the following SEQ ID NO 9 : DIQMTQSPSSLSASVGDRVTITCRASQDVNTAVAWYQQKPGKAPKLLIYSASFL YSGVPS RFSGSRSGTDFTLTISSLQPEDFATYYCQQHYTTPPTFGQGTKVEIKRTVAAPSV FIFPP

SDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDST YSLSSTLT

LSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC,

- a first heavy chain comprising the amino acid sequence of the following SEQ ID NO 10 :

EVQLVESGGGLVQPGGSLRLSCAASGFNIKDTYIHWVRQAPGKGLEWVARIYP TNGYTRY

ADSVKGRFTISADTSKNTAYLQMNSLRAEDTAVYYCSRWGGDGFYAMDYWG QGTLVTVSS

ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFP AVLQSS GLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPPKSCDKTHTCPPCP APELLG

GPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKT KPREEQY NSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVY TLPPSRD

ELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKL TVDKSR

WQQGNVFSCSVMHEALHNHYTQKSLSLSPGK,

- a second light chain comprising the amino acid sequence of the following SEQ ID NO 11 :

DIQMTQSPSSLSASVGDRVTITCRASQDVNTAVAWYQQKPGKAPKLLIYSASFL YSGVPS

RFSGSRSGTDFTLTISSLQPEDFATYYCQQHYTTPPTFGQGTKVEIKRTVAAPSV FIFPP

SDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDST YSLSSTLT

LSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC, and

- a second heavy chain comprising the amino acid sequence of the following SEQ ID NO 12 :

EVQLVESGGGLVQPGGSLRLSCAASGFNIKDTYIHWVRQAPGKGLEWVARIYP TNGYTRY

ADSVKGRFTISADTSKNTAYLQMNSLRAEDTAVYYCSRWGGDGFYAMDYWG QGTLVTVSS ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFP AVLQSS

GLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPPKSCDKTHTCPPCP APELLG

GPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKT KPREEQY

NSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVY TLPPSRD ELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKL TVDKSR

WQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

As shown in the examples herein, the use of a monoclonal antibody as described above as an Internal Standard compound allows a highly accurate and precise quantification of Bevacizumab in a sample.

Generating a calibration curve

The precise quantification of Bevacizumab by mass spectrometric analysis is allowed by the use of an Internal Standard compound, the presence of which in combination with Bevacizumab in a sample permits the calculation of ratio values of (i) the spectrometry signal generated by a selected proteolysis surrogate peptide derived from

Bevacizumab to (ii) the spectrometry signal generated by a selected surrogate peptide selected in a group comprising (1) a Stable Isotope Labelled peptide chosen among the surrogate peptides of SEQ ID NO 1, 2 and 3 and (2) a surrogate peptide generated by trypsin treatment of a selected monoclonal antibody, especially the selected Trastuzumab antibody, chosen among the peptides of SEQ ID NO 4, 5 and 6.

As it will be further detailed in the present specification, the quantification of

Bevacizumab is performed by reporting the ratio value calculated for the sample tested, or test sample, to a calibration curve of ratio values generated with known amounts of

Bevacizumab and fixed and known amount of a selected Internal Standard compound.

For generating a calibration curve, a serial or set of calibration samples (CS) are prepared, wherein :

- each calibration sample contains a known amount of Bevacizumab,

- each calibration sample contains a fixed and known amount of a selected Internal Standard compound, and

- the serial or set of calibration samples are prepared so as to cover an amount range of Bevacizumab encompassing at least the amount range of Bevacizumab which is expected to be contained in a test sample.

For the sake of clarity, each calibration sample comprises the same fixed and known amount of the selected Internal Standard compound. Illustratively, the amount range of Bevacizumab which is covered by the serial or set of calibration samples, when expressed as a final concentration in the calibration samples, may range from 6.25 μg/mL to 1,000 μg/mL. For example, a serial or set of calibration samples may comprise eight calibration samples comprising Bevacizumab at respective final concentrations of 6.25 μg/mL, 12.5 μg/mL, 25 μg/mL, 50 μg/mL, 100 μg/mL, 250 μg/mL, 500 μg/mL and 1000 μg/mL.

Illustratively, the given amount of the selected monoclonal antibody used as the Internal Standard compound, especially the given amount of Trastuzumab, is preferably an amount which generates mass spectrometry signal of the same order of magnitude as a mid-range calibration standard of bevacizumab in order to limit the difference in mass spectrometry signal intensity generated by the respective amounts (i) of surrogate peptides derived from trypsin proteolysis of the said monoclonal antibody used as the Internal Standard compound and (ii) of the tryptic peptides derived from Bevacizumab.

Illustratively, in the embodiments wherein a SIL-peptide is used as the Internal

Standard compound, the given amount of a SIL-peptide is preferably an amount which generates mass spectrometry signal of the same order of magnitude as a mid-range calibration standard of bevacizumab .

Indeed, the amount of bevacizumab that may be found in a test sample, especially in a test sample consisting of a human serum sample originating from a patient treated by Bevacizumab, may vary, depending of (i) the amount of Bevacizumab which has been administered to the said patient, (ii) the time period when the serum sample as been collected since the starting time period of the treatment, (ii) the time period of collection of the serum sample since the last administration of Bevacizumab, and (iv) physiological parameters which may be specific to the said patient, such as the rate of clearance of Bevacizumab from the blood.

Illustratively, the given amount of the selected monoclonal antibody used as the Internal Standard compound, especially of Trastuzumab, may be of a final concentration in a calibration sample of about 2 μg/mL, as shown in the examples herein.

Still illustratively, the given amount of the selected Stable Isotope labelled surrogate peptide used as the Internal Standard compound may be of a final concentration of about 0.2 μg/mL. In some embodiments, the serial or set of calibration samples may further comprise one or more control calibration samples which do not contain Bevacizumab.

Most preferably, a calibration sample is prepared starting from a body fluid sample initially exempt of Bevacizumab or of the selected Internal Standard compound, and preferably serum or plasma from a non-human mammal or from a human individual.

Then, each of the calibration sample is subjected to the same purification, concentration and trypsin treatment as that which is described for the test samples elsewhere in the present specification, so as to provide a serial or a set of calibration assay samples (CAS).

Then, each calibration assay sample is subjected to spectrometric analysis, in the same conditions as those described for the test samples elsewhere in the present specification and the values of the spectrometry signals generated by (i) a selected surrogate peptide generated by trypsin proteolysis of Bevacizumab and (ii) by the selected Stable Isotope Labelled surrogate peptide (also termed "labelled surrogate peptide of bevacizumab") used as the Internal Standard compound or by the selected peptide (also termed "surrogate peptide of monoclonal antibody") generated by trypsin proteolysis of the monoclonal antibody, especially by Trastuzumab, used as the Internal Standard compound, are then measured.

Then, for each of the calibration assay sample (CAS), a ratio of (i) the spectrometry signal value generated by the selected Bevacizumab surrogate peptide to (ii) the spectrometry signal value generated by the selected Internal Standard surrogate peptide.

As it will be further detailed in the present specification, a spectrometric signal value may consist of the peak area of specific SRM (Selected Reaction Monitoring), or more precisely of the sum of the peak areas of specific SRM, generated by a selected peptide of interest, typically by a selected surrogate tryptic peptide derived from Bevacizumab or by a surrogate peptide derived from Internal Standard described herein.

Thus, it is provided a serial or a set of ratio values, each ratio value being calculated from a calibration assay sample obtained from a starting calibration sample comprising known amounts, e;g. known final concentrations, of Bevacizumab and fixed and known amount of the Internal Standard compound. A calibration curve may then be generated by plotting the serial or set of calculated ratio values versus the corresponding theoretical Bevacizumab amounts, e;g. versus the corresponding known final concentrations of Bevacizumab.

As used herein, a "final" concentration of Bevacizumab is the concentration of Bevacizumab in an initial Calibration Sample (CS), which CS comprises a known added amount of Bevacizumab.

Detailed description of a method for quantifying Bevacizumab

The present invention concerns a method for quantifying Bevacizumab in a sample comprising the steps of:

a) subjecting to trypsin proteolysis a pre -proteolysis mixture consisting of the said sample in which is present a known amount of the said Internal Standard compound , whereby an assay mixture is provided, and wherein the said assay mixture comprises

- proteolysis peptides derived from Bevacizumab, which proteolysis peptides comprise surrogate peptides derived from Bevacizumab, and

- the said labeled surrogate peptide when it is used as the Internal Standard compound or proteolysis surrogate peptides derived from the said monoclonal antibody when the said monoclonal antibody is used as the Internal Standard compound,

b) determining by mass spectrometric analysis the ratio of (i) a selected proteolysis surrogate peptide derived from Bevacizumab to (ii) the said labeled surrogate peptide or a proteolysis surrogate peptide derived from the said monoclonal antibody, and

c) calculating from the ratio determined at step b) the quantity of Bevacizumab in the said sample.

As used herein, the expression "surrogate peptide" encompasses:

i) the specific peptide generated by trypsin proteolysis of the Bevacizumab discriminating it from other plasma endogenous proteins

ii) in the embodiments of the method wherein the Internal Standard compound which is used is a labelled peptide as described herein, the said labelled peptide is added just before proteolysis step without being cleaved by trypsin during the step of treatment with trypsin. Consequently, in these embodiments, a surrogate peptide is the labeled surrogate peptide of bevacizumab used as the Internal Standard compound itsef; iii) in the embodiments of the method wherein the Internal Standard compound which is used is a monoclonal antibody, a surrogate peptide is a specific peptide generated by trypsin proteolysis of the said internal standard monoclonal antibody, especially of Trastuzumab, discriminating it from other plasma endogenous proteins

In the embodiments of the quantification method wherein the Internal Standard compound which is used is a monoclonal antibody, then the said monoclonal antibody is cleaved in a plurality of tryptic peptides during the step of treatment with trypsin. Illustratively, when the internal standard monoclonal antibody is Trastuzumab, then the said monoclonal antibody is cleaved in a plurality of tryptic peptides, including each of the surrogate peptides of SEQ ID NO 4, 5 and 6.

The present invention also pertains to a specific embodiment of the quantification method above, which specific embodiment of the method for quantifying Bevacizumab in a sample, which may also be termed a test sample, comprises the steps of : i) enriching the said sample in IgG antibodies, whereby an IgG-enriched composition is provided,

ii) concentrating the IgG-enriched composition obtained at step i), so as to obtain a concentrated IgG-enriched composition,

wherein a known and fixed amount of an Internal Standard compound is added and wherein:

- the known amount of the Internal Standard compound is added before step i) in the embodiments wherein the Internal Standard compound is a monoclonal antibody, such as Trastuzumab, or

- the known amount of the Internal Standard compound is added before step iii) in the embodiments wherein the Internal Standard compound is a Stable Isotope

Labelled peptide,

iii) subjecting the concentrated IgG-enriched composition obtained at step ii) to trypsin proteolysis, whereby an assay mixture is provided, and wherein the said assay mixture comprises

- proteolysis peptides derived from Bevacizumab which proteolysis peptides comprise bevacizumab surrogate peptides, and - peptides derived from Internal Standard compound including surrogate peptides of Internal Standard selected in a group comprising (ii-a) the labeled selected surrogate peptide of bevacizumab and (ii-b) proteolysis surrogate peptides derived from the said monoclonal antibody, and

iv) determining by mass spectrometric analysis the ratio of (i) a selected surrogate proteolysis peptide derived from Bevacizumab selected in a group comprising the peptides of SEQ ID NO 1, 2 and 3, to (ii) a surrogate peptide of the Internal Standard selected in a group comprising (ii-a) the said labeled surrogate peptide of bevacizumab selected in a group comprising SEQ ID NO 1, 2 and 3 and (ii-b) a proteolysis surrogate peptide derived from the said monoclonal antibody selected in a group comprising the peptides of SEQ ID NO 4, 5 and 6 in the embodiment wherein Trastuzumab is used as Internal Standard, and

v) calculating from the ratio determined at step b) the quantity of Bevacizumab in the said sample.

In the Bevacizumab quantification method above, steps iii), iv) and v) are identical to steps a), b) and c) of the Bevacizumab quantification method described previously. However, in the method above, steps i) and ii) precede step a) of the method previously described. Indeed, in the previously described method, step a) may be preceded by steps identical to steps i) and ii) of the Bevacizumab quantification method above.

A quantification method of Bevacizumab according to the invention is described in more detail below, and specific embodiments thereof are disclosed in the examples herein.

Sample preparation

In some embodiments, the sample which is used in the quantification method originates from a whole blood sample that has been previously collected from an individual. The blood cells, and especially erythrocytes, are removed by centrifugation so as to obtain a plasma sample. In some other embodiments, coagulation of the whole blood sample is allowed to occur and a serum sample is obtained.

In further embodiments, the sample which is used in the quantification method may consist of other extracellular fluids such as lymphatic fluid (endolymph or perilymph) and interstitial fluid. Most preferably, at least for determining the pharmacokinetic profile of Bevacizumab in an individual, the said sample is a blood plasma sample or a blood serum sample, or a sample derived from blood plasma or blood serum.

In some embodiments, the initial sample is subjected to dilution, e;g. in an aqueous medium such as in a saline solution or in a buffer solution, before being used as the assay sample in the Bevacizumab quantification method according to the invention.

However, in the most preferred embodiments, the initial sample, such as a plasma sample or a serum sample, is used without any pre-treatment and in particular is used as such undiluted.

In some embodiments of the Bevacizumab quantification method wherein the

Internal Standard compound is a monoclonal antibody, especially Trastuzumab, the sample is added with a known amount of the selected Internal Standard compound at that step..

In these embodiments, there is thus provided a sample containing a known amount of the Internal Standard compound and an unknown amount of Bevacizumab.

In preferred embodiments, the said sample comprises only one Internal

Standard compound.

In these embodiments, where a monoclonal antibody is used as internal standard, the Internal Standard compound is subjected to each of the further steps of the Bevacizumab quantification method described herein.

Enriching the sample in IgG antibodies

In some embodiments of the Bevacizumab quantification method described herein, the sample, optionally comprising the Internal Standard compound, is enriched in IgG antibodies. It is reminded that the sample contains the Internal Standard Compound in embodiments wherein the said Internal Standard Compound consists of amonoclonal antibody

This step precedes step a) of the general Bevacizumab quantification method and is step i) of the specific embodiment thereof.

Various methods for enriching a sample in IgG antibodies are known in the art. In some embodiments, enrichment in IgG antibodies may be performed by ammonium sulfate precipitation, by using methods well known in the art, so as to obtain an IgG-enriched composition. In some other embodiments, enrichment in IgG antibodies may be performed by affinity chromatography, which includes the use of chromatography substrates onto which have been immobilized relevant ligands such as protein A, protein G or alternatively antibodies binding to the Fc portion of IgG antibodies, as well as nucleic acid or peptide aptamers that bind to the Fc portion of IgG antibodies.

The step of enrichment in IgG antibodies allows separating antibodies from other abundant plasma proteins and thus contributes to improve sensitivity and reproducibility of the Bevacizumab quantification method.

General enrichment in IgG antibodies is preferred to a specific enrichment in Bevacizumab, e.g. by an immunocapture based on the recognition of the binding site of the antigen. The inventors have observed that such specific immunocapture method does not allow dosing the total fraction of Bevacizumab in the sample. Furthermore, development of such a specific immunocapture method is lengthy because it would require the production of an antibody with a capture system.

Further, general enrichment in IgG antibodies by using a method of precipitation of plasma proteins possesses substantial drawbacks. Such a method for general precipitation of plasma proteins, although it is simple, fast, inexpensive and allows access to the measurement of total protein fraction, the resulting plasma proteins-enriched mixture is not sufficiently enriched in IgG, which is detrimental to the repeatability of the subsequent step of trypsin proteolysis, and finally be detrimental to the accuracy of the Bevacizumab quantification method.

Still further, IgG enrichment by methods of protein depletion, e.g. methods of protein depletion by ion exchange or affinity chromatography, while removing albumin and optionally other abundant plasma proteins also causes a partial depletion in IgG antibodies which is a concern for reproducibility of a monoclonal antibody quantification method. The inventors practical experience is that these kind of enrichment methods do not allow removing the majority of plasma proteins and simultaneously do not allow preserving the content in IgG antibodies.

Preferably herein, enrichment in IgG antibodies by using protein A or protein G chromatography is preferred.

IgG enrichment by subjecting the sample to protein A or protein G chromatography allows depletion of almost the whole plasma proteins while retaining the whole IgG antibodies initially contained therein, which includes the whole Bevacizumab antibodies initially contained therein.

Most preferably, enrichment in IgG antibodies is performed by protein A chromatography.

As shown in the examples herein, when the IgG enrichment step is performed by protein A chromatography, recovery of Bevacizumab is of about at least 85 %.

In the embodiments wherein protein A chromatography is used, elution of the retained IgG antibodies is conventionally performed at an acidic pH, generally at a pH in the range of 2-3, preferably at a pH of 2.8, and the fraction containing the most part of the IgG antibodies is collected in a buffer at basic pH so as to provide an IgG-enriched composition having a pH value ranging from neutral pH to slightly basic pH, advantageously at a pH in the range of 8.0 to 9.0, preferably at a pH range of 8.5 to 9.5, and most preferably at a pH of about 8.0, so as to allow optimal trypsin proteolysis in a further step of the method.

In some embodiments of the Bevacizumab quantification method wherein the

Internal Standard compound is a monoclonal antibody, especially Trastuzumab and wherein the said Internal Standard compound has not already been added to the starting sample before IgG enrichment, then the IgG-enriched composition may be added with a known amount of the selected Internal Standard compound at this step.

In these embodiments, there is thus provided an IgG-enriched composition containing a known amount of the Internal Standard compound and an unknown amount of Bevacizumab.

In preferred embodiments, the said IgG-enriched composition comprises only one Internal Standard compound.

In the embodiments where a reference monoclonal antibody, especially Trastuzumab, is used as the Internal Standard compound, then the said Internal Standard compound is subjected to each of the further steps of the Bevacizumab quantification method described herein. Concentrating the IgG-enriched composition

In some embodiments, and especially in embodiments wherein the IgG- enriched composition is obtained by a step of chromatography wherein sample dilution is susceptible to occur, the said composition is then subjected to a concentration step, so as to provide a concentrated IgG-enriched composition.

This step precedes step a) of the general Bevacizumab quantification method and is step ii) of the specific embodiment thereof.

In these embodiments, the concentration step may be performed by any method known in the art, including dialysis and filtration, e.g. microfiltration or ultrafiltration.

In preferred embodiments, the concentration step is an ultrafiltration step wherein a filter membrane of a relevant cut-off value is used.

Contrary to what was expected initially when conceiving the Bevacizumab quantification method of the invention, it has been shown that the ultrafiltration step does not cause a significant loss in IgG antibodies. Consequently, in contrast to what was expected, the antibodies contained in the IgG-enriched composition do not significantly adsorb onto the ultrafiltration membrane.

Illustratively, the ultrafiltration step may be performed by using an ultrafiltration membrane having a cut-off value of about 100 kDa.

In the embodiments wherein the concentration step is an ultrafiltration step, a buffer exchange is performed during the ultrafiltration step so as to optimize the conditions of the further steps of the method are conducted. Notably, the buffer exchange that may be performed during the ultrafiltration step allows obtaining a concentrated IgG-enriched composition in which the subsequent step of proteolysis by trypsin will be optimally realized.

As shown in the examples herein, performing the concentration step by ultrafiltration allows a recovery of Bevacizumab of about 90 %. Pre-proteolysis mixture preparation

This precedes step a) of the general Bevacizumab quantification method and also precedes step iii) of the specific embodiment thereof.

In the embodiments of the method wherein the Internal Standard compound is a selected Stable Isotope Labelled peptide, then concentrated IgG-enriched composition obtained at the previous method step is added with a known and fixed amount of the selected Internal Standard compound. Indeed, according to these embodiments, no Internal Standard compound was added in any previous step of the method. There is thus provided a pre-proteolysis mixture containing a known amount of an Internal Standard compound and an unknown amount of Bevacizumab, irrespective of the step in which the said Internal Standard compound was added.

In preferred embodiments, the said pre-proteolysis mixture comprises only one Internal Standard compound.

In some most preferred embodiments, the said pre-proteolysis mixture comprises the labelled peptide of SEQ ID NO 1 or the labelled peptide of SEQ ID NO 2 as the Internal Standard compound. The labelled peptide of SEQ ID NO 1 and the labelled peptide of SEQ ID NO 2 may also be termed "surrogate peptides" herein.

In some other most preferred embodiments, the said pre-proteolysis mixture comprises Trastuzumab as the monoclonal antibody used as the Internal Standard compound.

When the selected Internal Standard compound is added at this step of the Bevacizumab quantification method it subjected proteolysis step and quantification step.

Proteolysis step

This step is step a) of the general Bevacizumab quantification method and is step iv) of the specific embodiment thereof.

At this step, trypsin is added to the pre-proteolysis mixture, so as to generate tryptic peptides from Bevacizumab and, in the embodiments wherein the Internal Standard compound is a monoclonal antibody, especially Trastuzumab, also tryptic peptides generated by trypsin proteolysis of the said internal standard monoclonal antibody. The specific tryptic peptides derived from the internal standard monoclonal antibody may also be termed "surrogate peptides" herein.

The inventors have determined that the proteolysis step shall be performed in conditions that are optimal for :

(i) generating all the expected surrogate tryptic peptides form Bevacizumab and, in the relevant embodiments, also from the internal standard monoclonal antibody, and thus avoiding missed cleavages, and

(ii) avoiding trypsin autolysis. Preferably, it is used a purified trypsin having a low ability to autolysis. Illustratively, it may be used a trypsin termed Trypsin Gold which is marketed by the company Promega (Madison, WI, United States).

It has been determined herein that optimal proteolysis conditions are reached by using a trypsin / total protein molar ratio ranging from 1/100 to 1/10, and advantageously a trypsin / total protein molar ratio ranging from 1/75 to 1/25.

In most preferred embodiments, the optimal proteolysis conditions are reached by using a trypsin / total protein molar ratio ranging from 1/60 to 1/40, e;g. a trypsin / total protein molar ratio of about 1/50.

A trypsin / total protein molar ratio of 1/50 or close thereto is particularly suitable for an optimal tryptic cleavage of both Bevacizumab and Trastuzumab.

At a reduced trypsin / total protein molar ratio; missed cleavage is expected. This means that the expected tryptic peptides from Bevacizumab, and that the expected tryptic peptides from the internal standard monoclonal antibody might not be generated or might be only partly generated, thus introducing a high variability in the method and leading to weakly reproducible Bevacizumab quantification values.

At an excessive trypsin / total protein molar ratio, an autolysis of trypsin occurs which lowers with time the amount of active trypsin for protein cleavage, which trypsin autolysis also causes a high variability and low reproducibility of the proteolysis step and thus also leading to weakly reproducible Bevacizumab quantification values.

In most preferred embodiments, the proteolysis step is performed in non- denaturing conditions, i.e. in conditions which do not cause protein denaturation. Notably, the proteolysis step is performed in the absence of a protein denaturation agent such as urea.

In some embodiments, proteolysis in the presence of trypsin is performed during a period of time that may be optimally adapted by the one skilled in the art.

For example, proteolysis is performed during a time period ranging from 7 to 9 hours, and is preferably of about 8 hours.

Advantageously, proteolysis is performed at 37 °C during a period of time ranging from 4 hours to 16 hours, preferably from 6 hours to 10 hours, and most preferably ranging from 7 hours to 9 hours, the time period being ideally around 8 hours. Illustratively, in the embodiments of the quantification method wherein the surrogate peptide of Bevacizumab is the peptide of SEQ ID NO 1, a 8 hour incubation time with trypsin at 37°C allows to reach the maximum concentration of the said tryptic peptide of SEQ ID NO 1 which is used as the signature peptide for Bevacizumab quantification.

In some embodiments of the proteolysis step, reduction of the pre-proteolysis mixture is performed before addition of trypsin, e.g. by adding a reducing agent such as dithiothreitol (DTT).

In some embodiments of the proteolysis step, prevention of oxidation of free cysteine residues and prevention of reformation of disulfide bridges may be performed by addition of an alkylation agent such as Iodoacetamide.

As shown in the examples herein, the overall Bevacizumab recovery of the quantification method, from the initial sample used to the composition obtained after trypsin proteolysis which is used for mass spectrometry quantification is of about 39.4 %. Quantification of Bevacizumab

This step is step d) of the general Bevacizumab quantification method and is step v) of the specific embodiment thereof.

The step of quantification of Bevacizumab is performed by mass spectrometry, according to techniques of protein quantification by mass spectrometry that are known in the art.

Preferably, the step of quantification is performed according to the method of Liquid Chromatography coupled to tandem Mass Spectrometry (LC-MS/MS), as it is shown in the examples herein.

Preferably, it is used a triple quadrupole (QqQ) mass spectrometer equipped with an ESI source operating in positive ion mode and using multiple reaction monitoring (MRM) mode for quantification.

In some embodiments, Liquid Chromatography is performed with a reverse phase chromatography substrate.

Then, in some embodiments, the most abundant state of charge of selected surrogate tryptic peptides derived from Bevacizumab, internal standard monoclonal antibody and labelled surrogate peptide observed preferably between 200 m/z and 2000 m/z in ESI ionization source and are selected and fragmented. At the quantification step by mass spectrometry, it is researched the Selected Reaction Monitoring (SRM) transitions specific of

(i) the selected surrogate tryptic peptide(s) of Bevacizumab and of

(ii) the internal standard surrogate peptide(s) selected, depending on the specific embodiment of the quantification method which is performed, in a group comprising

(i) the selected labelled surrogate peptide(s) or (ii) the tryptic surrogate peptide(s) derived from the internal standard monoclonal antibody.

SRM transitions of tryptic peptides from Bevacizumab, of tryptic peptides from the internal standard monoclonal antibody and of the labelled peptides are preferably established after comparing the fragmentation spectra obtained from pure solutions of each of these peptides, with in silico fragmentation spectra generated with a relevant available software tool, such as MS-product query available with the bioinformatics tool protein prospector.

Preferably, at this step, quantification of Bevacizumab is based on the ratio of the sum of the peak areas of specific SRM of Bevacizumab and the sum of the peak areas of the internal standard selected surrogate peptide.

More precisely, the amount of Bevacizumab in the sample tested, e.g. the concentration of Bevacizumab in the test sample, is determined by reporting the ratio value that is calculated for the said test sample to a calibration cure that was generated as previously described elsewhere in the present specification.

As shown in the examples, the quantification described herein allows linearity between peak area ratios and Bevacizumab sample concentration, and especially between peak area ratios and Bevacizumab plasma concentration.

Quantifying Bevacizumab with the quantification method described herein allows a high quantification precision, a high quantification repeatability, as well as Bevacizumab quantification over a wide range of amounts.

In the embodiments where only SEQ ID NO 1 (FTFSLDTSK) is used as a surroagte peptide of Bevacizumab for quantification and where stable isotope labeled SEQ ID NO 1 (FTFSLDTSK) is used as the Internal Standard compound, the accuracy of the Bevacizumab quantification method according to the invention ranges from 101.7 % to 110.6 , as shown in the examples. In this embodiment, the precision of the Bevacizumab quantification method according to the invention ranges from 7 % to 10 , as shown in the examples herein.

Further, the Bevacizumab quantification method according to the invention allows a linearity of the quantification measure from 12.5 μg/mL or less to 500 μg/mL or more.

According to FDA guidelines for bioanalytical method validation, it is thus shown herein that the Bevacizumab quantification method according to the invention is at the same time sufficiently sensitive and reproducible to quantify bevacizumab in human plasma samples.

The present invention is further illustrated, without being limited thereto.

Examples

A. Materials end Methods

A.l. Chemicals and reagents

Bevacizumab was obtained from an aqueous 25 mg/mL solution (Avastin^®, Roche, Basel, Switzerland) and the mAbIS trastuzumab was obtained from lyophilised powder (Herceptin^®, Roche, Basel, Switzerland) reconstituted at a 21 mg/mL solution in water for injection. The solution of human IgG pooled from several healthy donors used to confirm the uniqueness of the surrogate peptide was from LFB-biomedicaments (Tegeline^®, Courtaboeuf, France). Stable isotope labeled (SIL) of surrogate peptide of Bevacizumab used as IS in the second intern calibration approach was synthesized by JPT Peptide

Technologies GmbH (Volmerstrasse, Berlin, Germany) labeling lysine with 13 C and 15 N to obtain a sufficient mass increment relative to unlabeled peptide (8 Dalton). Trypsin Gold mass spectrometry was purchased from Promega (Madison, WI, USA). DL-Dithiotreitol (DTT) and iodoacetamide (IAA) were from Acros organics (Geel, Belgium). Synthetic surrogate peptide of Bevacizumab and mAbIS were purchased from Thermo Fisher Scientific GmbH (Ulm, Germany). Protein A high capacity kit were purchased from Pierce biotechnology (Nab Spin Kits, Rockford, IL, USA). Ultra centrifugal filters with a cut off of 100-kDa were obtained from Merck Millipore (Amicon-Ultra^® 100-kDa, Billerica, Massachusetts, USA). HPLC-MS grade acetonitrile was from Scharlau (Barcelona, Spain) and Ultrapure water for HPLC was produced using a filtration system (Aquadem^® EFP 310, Veolia water STI, Antony, France). Human plasma was used as a biological matrix for the development of the assay and the preparation of the calibration standards and the quality controls. Plasma was obtained from blood of several anonymous healthy volunteers by Etablissement Francais du Sang Aquitaine-Limousin (EFSAL), Bordeaux, France. A.2. Plasma sample preparation

Sets of calibration standards were obtained by diluting Bevacizumab solution for injection in pooled human plasma to 12.5, 25, 50, 100, 250 a 500 μg/mL to cover the expected concentration range in therapeutics. The 500 μg/mL calibration point was prepared by diluting in 720 μΙ_, of human plasma, 80 μΙ_, of 5 mg/mL Bevacizumab plasmatic solution obtained previously from the Bevacizumab aqueous stock solution for injection diluted to 1/5 in human plasma. The other calibration points were then obtained by serial dilutions in human plasma from the same plasmatic stock solution at 5mg/ml. Blank plasma samples without Bevacizumab were prepared. Sets of calibration points were used for calibration of quantitative analysis and linearity tests. Three quality controls (QCs) at 4, 200 and 400 μg/mL (low, medium and high calibration curve concentration) were used to assess accuracy and precision of quantification method during analytical validation step. They were prepared in the same way that calibration points by diluting Bevacizumab solution for injection in pooled human plasma. A.3. Internal standards preparation

In the first internal calibration approach using mAbIS, 6.5μί of 2^g/mL trastuzumab solution was added to 65 μΙ_, of each plasma sample (standard, quality control or plasma patient samples) before any treatment sample steps. In the second internal calibration approach using SIL surrogate peptide, 5μί SIL surrogate peptide solution was added to 65 μΙ_, of each standard, quality control or plasma patient samples just before proteolysis step as internal standard.

A.4. Purification on Protein A column

Purification of IgG was achieved by means of Protein A high capacity kit used according to the manufacturer's protocol. After conditioning the columns by loading two times 400 μL· of binding buffer, 55 μΐ_^ of each plasma sample (spiked or not with mAbIS) was loaded on the columns and the columns were incubated for 10 min. After three washes with 400 μL· of binding buffer, IgG were then eluted with 400 μΙ_, of IgG elution buffer. To determine which elution fraction contain the purified antibody, this step was repeat two times to obtain three elution fractions which were analysed by measuring the relative absorbance of each at 280 nm. Measurements were performed in triplicate by Tecan's Infinite M200pro micro plaque reader (data not shown). As the majority of the IgG (more than 94%) were eluted in the first elution fraction, exclusively this fraction was collected into collection tube containing 40 μΙ_, of neutralizing buffer (Tris buffer, pH 8.5).

A.5. Sample concentration and buffer exchange by ultrafiltration

Purified sample was introduced on Amicon Ultra^® 100-kDa and the device was centrifuged for 10 min at 14,000xg. After discarding the filtrate, 415 μΙ_, of 100 mM ammonium bicarbonate proteolysis buffer (pH 8) was introduced into the device and a second ultrafiltration by centrifugation was performed during 10 min at 14,000xg. To recover the concentrated IgG sample, the Amicon^® Ultra 100-kDa filter device was placed upside down in a clean microcentrifuge tube and centrifuged 1 min at l,000xg. This step allows concentrating (x 15) the purified sample in the proteolysis buffer.

A.6. Tryptic proteolysis

25 μΐ_^ of proteolysis buffer (100 mM ammonium bicarbonate buffer, pH 8) was added to each concentrated sample. A reduction step was performed by addition of 5.5 μΐ_^ of DTT 100 mM to each sample (final concentration of 10 mM) and incubation at 55 °C during 20 min. To prevent oxidation of the free cysteins and reformation of the disulfide bridges, an alkylation steps was performed by addition of 6 μΐ_^ IAA 100 mM to each sample (final concentration of 10 mM). After 30 min incubation at room temperature, protected from light, 5 μΐ_^ of trypsin 23.8 pmole^L in proteolysis buffer was added to each sample (ratio trypsin/total sample proteins 1/50 mole/mole) for 8 hours incubation at 37 °C. This amount of trypsin is the optimum amount providing a sufficient proteolysis of plasma samples, firstly avoiding missed cleavages and secondly avoiding trypsin autolysis, in order to limit variability. These data were determined and confirmed by LC-MS/MS analysis of purified bevacizumab plasma sample digested with an increasing ratio of trypsin. A.7. Experimental identification and selection of signature tryptic peptides of bevacizumab and mAbIS

In a first experimental study to confirm the uniqueness of the in silico selected signature peptides of Bevacizumab and mAbIS digest of each mAb was analysed by LC- MS/MS on a nano-LC-ESI-LTQ orbitrap system against a subset of the UniProt SwissProt Homo sapiens database incremented by sequence of Bevacizumab and mAb IS to establish the experimental peptide map of Bevacizumab and trastuzumab (mAbIS). This LC-MS/MS system used consisted of a U-3000 Ultimate nano LC system coupled to a nanospray LTQ- OrbiTrap XL mass spectrometer (ThermoFinnigan, San Jose, CA, USA) equipped with an ESI source. Ten microliters of each protein digest were separated on an analytical 75-μιη- inner diameter x 15-cm CI 8 PepMap™ column (LC Packings) with a 5-40% linear gradient of solvent B (water/acetonitril 20:80 (v/v) containing 0.1% formic acid (v/v)) for 70 min. The separation flow rate was set at 200 nL/min. Data were acquired in a data- dependent mode alternating a Fourier Transform Mass Spectrometry (FTMS) scan survey over the range m/z 300-1700 and six MS/MS scans in an exclusion dynamic mode. To establish the peptide signature of Bevacizumab and mAb IS, peptides were identified with SEQUEST through the Bioworks 3.3.1 interface (Thermo-Finnigan, Torrence, CA, USA) against a subset of the UniProt SwissProt Homo sapiens database incremented by sequence of Bevacizumab and mAbIS. Only b- and y-ions were considered for mass calculation and two missed trypsin cleavages were allowed. Tryptic peptides were validated using the following criteria: DeltaCN > 0.1, Xcorr > 1.5 (single charge), 2.0 (double charge), 2.5 (triple charge), 3.0 (Quadruple charge), and Peptide Probability < 0.001. Proteins were validated as soon as two different specific peptides were identified.

In a second experimental study a digest of human IgG polyvalent therapeutic solution (Tegeline^®) was also analysed in the same way to ensure that signature peptides of Bevacizumab and mAb IS were absent of human IgG polyvalent digest.

In a third experimental study, tryptic digest of Bevacizumab and mAbIS were analyzed by LC-ESI-QqQ system. Fragmentation of previously specific peptides identified and selected in silico and in the first experimental study was achieved in « product ion scan » mode. The selection criteria were the peptide ionization efficiency and its MS/MS fragmentation properties to ensure enough intensity of the SRM transitions allowing the better sensitivity of the method. A.8. LC-MS/MS identification and quantification of Bevacizumab

Bevacizumab quantification was performed on a LC-MS/MS system consisted of 1100 Series HPLC from Agilent (Santa Clara, CA, Etats-Unis) connected to a API3200 triple quadrupole (QqQ) mass spectrometer from AB Sciex (Foster City, CA, USA) equipped with an ESI source operating in positive ion mode and using multiple reaction monitoring (MRM) mode for quantification. The quantification LC-MS/MS system was controlled with Analyst 1.5.1 software. Twenty microliters of each proteolysed purified sample were separated on an analytical XB-C18 Aeris Peptide, 150 mm x 2.1 mm i.d., 3.6 μιη reverse phase column (Phenomenex, Torrance, CA, USA) with elution gradient of 2 solvents A and B for peptides separation. Solvent A was water with formic acid 0.1% (v/v) and solvent B was acetonitrile with formic acid 0.1% (v/v). The separation flow rate was set at 200 μΏ min during 35 min as follow (time in min/% solvent B): 0.0 min/10%; 1.0 min/10%; 10.0 min/60%; 20.0 min/10%; 30.0 min/10%. The most abundant state of charge of signature peptides of Bevacizumab and mAbIS detected between 200 and 2000 m/z in ESI ionisation source were selected and fragmented.

The SRM transitions specific of Bevacizumab and mAbIS were researched and monitored for the protein identification. SRM transitions were determined after analysis of fragmentation spectra obtained for each solutions of synthesized signature peptide of Bevacizumab (included SIL surrogate peptide) and internal standard mAbIS relative to the in silico fragmentation spectra generated from MS-product query of Protein Prospector online bioinformatics tool v 5.12 .2 (University of California, San Francisco, USA). In order to increase the signal-to-noise ratio and the accuracy, the most intense SRM transitions with the less variability were selected and summed for protein quantification. Moreover, it was checked after six consecutive analyzes conducted on the same sample that the ratio of the most intense SRM transitions was constant in each run with a coefficient of variation not exceeding 5% for each transitions. Bevacizumab quantification was based on the ratio of the sum of the peak areas of specific SRM of Bevacizumab and the sum of the peak areas of internal standard (surrogate peptide of mAbIS or SIL peptide). The different ionization and fragmentation parameters were optimized using the automatic tool of Analyst software (Analyst 1.5.1 software, AB Sciex) from each solution of synthesized surrogate peptide of Bevacizumab, mAbIS and SIL internal standard. Dependent parameters of the precursor ion (DP, EP and CEP CE) have been optimized after entering their respective m/z ratio (523.3 for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) first Bevacizumab surrogate peptide, 642.3 for ⁷⁷ STA YLQMNS LR⁸⁷ (SEQ ID NO 2) second Bevacizumab surrogate peptide, 527, 3 for ⁶⁸FTFSLDTSK*⁷⁶ (SEQ ID NO 1) SIL peptide and 545.3 for ³¹DTYIHWVR³⁸ (SEQ ID NO 5) mAbIS surrogate peptide). CXP which depends on ion fragment was optimized for each SRM transitions. The optimized values of these parameters are summarized in Table 1. Other parameters were optimized by ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) infusion. IonSpray voltage (IS) was set at 5000 eV and drying temperature at 300°C.

A.9. Analytical Validation

The analytical validation is a prerequisite to Bevacizumab quantification. The assay was validated according to the Food and Drug Administration (FDA) guidelines for bioanalytical method validation by determining linearity, accuracy, precision, specificity, sensibility and recovery. Calibration curve was constructed from the peak area ratios (Bevacizumab surrogate peptide/internal standard SRM sum) obtained after LC-MS/MS analysis of the six Bevacizumab calibrations standard of known concentration. Linearity between peak area ratios and Bevacizumab plasmatic concentration was assessed by determining equation and coefficient of correlation of calibration curve. Precision and accuracy were determined after intraday and interday variability study by the analysis of six replicates of each of the three QC samples each day, and for four days. Precision was assessed using the coefficient of variation (CV %) and accuracy using ratio between concentration measured on expected concentration. Specificity of the assay was assessed by analysing blank plasma samples versus Bevacizumab plasma samples spiked and not spiked with mAbIS or SIL internal standard. Recovery of purification step by affinity chromatography on protein A was assessed by comparing the means peak area values obtained between two conditions: blank plasma sample spiked with Bevacizumab before affinity chromatography, and blank plasma sample spiked with Bevacizumab immediately after affinity chromatography. In the same way recovery of ultrafiltration step was assessed by comparing mean peak area values obtained between two conditions: blank plasma sample spiked with Bevacizumab before ultrafiltration step, and blank plasma sample spiked with Bevacizumab immediately after ultrafiltration step. The proteolysis recovery was achieved by comparing means peak area values obtained between two conditions: blank plasma samples spiked with Bevacizumab just before proteolysis, and blank plasma samples spiked with synthetized specific peptide of Bevacizumab after proteolysis at concentration corresponding to a recovery of 100%. To determine recovery of each step of the assay, all of these analyzes were conducted in triplicates on the lower and upper QCs (QC1 and QC3).

The detection limit corresponds to the concentration that gives an area equivalent to three times the background noise of the chromatographic signal (background noise is measured at both sides of the peak with a width equivalent to 20 times the peak width at half -height).

The quantification limit corresponds to the minimum concentration giving a coefficient of variation less than 20% obtained after the analysis of six samples per concentration.

The stability of Bevacizumab specific peptides during LC-MS/MS analysis was conducted in triplicate for each of the three QCs during twelve hours at room temperature and was assessed by comparing the mean peak area measured at different times of the analysis (0, 2, 4, 6, 8 and 12 hours) with time zero peak area.

Example 1 : LC-MS/MS identification and quantification of Bevacizumab

SRM transitions of surrogate peptides of Bevacizumab, of mAbIS and SIL peptide were established after comparing the fragmentation spectra obtained from pure solutions of each of the previously selected peptides synthetized ( ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1), ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2), ³¹DTYIHWVR³⁸ (SEQ ID NO 5) and ⁶⁸FTFSLDTSK⁷⁶* (SEQ ID NO 1)), with in silico fragmentation spectra generated with MS-product query (figure 1). Both the transitions from the doubly charged ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) peptide (m/z 523.3) to the six single charge y-ions, and the transitions from the doubly charged ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) peptide (m/z 642.3) to the six single y-ions were used for the identification of Bevacizumab (transitions 523.3→898.4 (y₈ ²⁺), 523.3→797.3 (y₇ ²⁺), 523.3→650.3 (y₆ ²⁺), 523.3→563.2 (y₅ ²⁺), 523.3→450.2 (y₄ ²⁺), 523.3→335.2 (y₃ ²⁺) for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) peptide and transitions 642.3→1024.5 (y₈ ²⁺), 642.3→861.3 (y₇ ²⁺), 642.3→748.3 (y₆ ²⁺), 642.3→620.3 (y₅ ²⁺), 642.3→489.3 (y₄ ²⁺), 642.3→375.2 (y₃ ²⁺) for ⁷⁷ ST A YLQMNS LR⁸⁷ (SEQ ID NO 2) peptide). The four most intense SRM transitions of FTFSLDTSK (SEQ ID NO 1) peptide after analysis in MRM mode were summed for Bevacizumab quantification to increase signal-to-noise ratio with the less variability (transitions 523.3→797.3 (y₇ ²⁺), 523.3→650.3 (y₆ ²⁺), 523.3→450.2 (y₄ ²⁺), 523.3→335.2 (y₃ ²⁺)). The four following SRM transitions 527.3→805,3 (y₇ ²⁺), 527.3→658.3 (y₆ ²⁺), 527.3→458,2 (y₄ ²⁺), 527.3→343,2 (y₃ ²⁺) for peptide ⁶⁸FTFSLDTSK⁷⁶* (SEQ ID NO 1) were summed to monitored SIL peptide used as internal standard, and the two following SRM transitions 545.3→710.4 (y₅ ²⁺), 545.3→597.3 (y₄ ²⁺) for peptide ³¹DTYIHWVR³⁸ (SEQ ID NO 5) to monitor mAblS. Under chromatographic conditions described in experimental section, the retention time of the two Bevacizumab peptides were 9.8 min for the peptide ⁷⁷STAYLQMNSLR⁸⁷"(SEQ ID NO 2), and 10.8 min for the peptide "⁶⁸FTFSLDTSK⁷⁶"(SEQ ID NO 1). The SIL internal standard peptide ⁶⁸FTFS LDTS K⁷⁶ * (SEQ ID NO 1) was co-eluted with the non-labeled peptide at 10.8 min and mAblS surrogate peptide ³¹DTYIHWVR³⁸ (SEQ ID NO 5) was eluted in 9.6 min (figure 2).

Example 2 : Optimisation of proteolysis conditions

In order to achieve Bevacizumab therapeutic concentrations found in plasma and also to ensure reproducibility of the assay, trypsin concentration and cleavage time were optimized. Trypsin ratio relative to the amount of substrate should be sufficient to ensure complete proteolysis without be introduced in excess which could lead to trypsin autolysis. Promega trypsin was chosen for its limited self-digestion and an increase ratio of trypsin was tested to find the optimal conditions for a sufficient proteolysis with a good repeatability. Each of the three QCs samples of Bevacizumab in plasma spiked with mAblS at the same concentration were proteolysed at three different ratios trypsin/total proteins 1/40, 1/50 and 1/60 (mole/mole) under the conditions described in experimental section (figure 3). This analysis was conducted in triplicate to assess variability of the assay. Mean peak area corresponding to ⁶⁸FTFSTLDTSK⁷⁶ (SEQ ID NO 1) peptide were slightly lower when samples were proteolysed to the 1/40. However for the three trypsin/total proteins ratios tested, CV% was always less than 15% for ⁶⁸FTFSTLDTSK⁷⁶ (SEQ ID NO 1) peptide. Although mean peak area corresponding to ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) peptide were higher when samples were proteolysed to the ratio 1/60, variability of the results was more important with CV% more than 15% for QC2 and QC3 samples. CV% more than 15% were also observed with the ratio 1/40. However only samples proteolysed to the ratio 1/50 exhibited a CV% less than 15% for the three peptides monitored. Therefore we fixed the ratio trypsin/total proteins at 1/50 (mole/mole).

To improve the efficiency of the proteolysis we studied the need of protein denaturation by urea. Denaturation has the property to increase the accessibility of trypsin to its cleavage sites breaking tertiary structure of monoclonal antibody. Denaturation with urea at 6 mole/L was tested in triplicate with blank plasma samples spiked with Bevacizumab and mAbIS (QC3 samples). Mean peak area corresponding to signature peptides of Bevacizumab and mAbIS were analyzed and compared with samples without urea denaturation. Both experiments have been subjected to the same dilution factor. Although mean peak area were increased for the samples subjected to urea denaturation, dilution factor necessary to this step was higher than this increase which finally decrease the sensitivity of the assay. Therefore denaturation urea was not used during proteolysis step (data not shown).

To limit the time of proteolysis step and to increase repeatability of the assay, the formation of signature peptides of Bevacizumab and mAbIS after 4 hours, 6 hours, 8 hours and 16 hours of trypsin incubation was studied by measuring mean AUC of each surrogate peptide. This experiment was conducted in six replicates and variability was also assessed by coefficient of variation (CV%) (Figure 4). 6 hours of trypsin incubation were sufficient for 77 STAYLQMNSLR 87 to reach its maximum concentration that decrease after 6 hours, while 8 hours of trypsin incubation were required for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1).

Formation of 3 ^J 1¹DTYIHWVR 3^J8^O (SEQ ID NO 5) peptide was slower and increase slowly at least until 16 hours of trypsin incubation (Figure 4). Variability of the results was more important after 6 h of proteolysis rather than 8 h for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) with a CV of 10.2% at time 6 h and 4.6% at time 8 h while it is substantially the same for ⁷⁷STAYLQMNSLR⁸⁷ at 6h (10.7%) or 8h (10.8%) of trypsin incubation. However variability of the result (CV%) was more important at time 6 h (27.0%) and 8h (29.5%) for ³¹DTYIHWVR³⁸ (SEQ ID NO 5) rather than at time 16h (21.3%). Although no condition is optimal for both of the three peptides, an 8h incubation time with trypsin seems to be the best compromise since it allows to reach the maximum concentration of ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) which is the signature peptide used for Bevacizumab quantification (having the most intense signal with less variability in the measurements). This condition allow a sufficient precision (CV less than 15%) for the two Bevacizumab signature peptides while preserving sufficient amount of ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2).

Example 3: Analytical validation and comparison of the two internal calibration approaches

Selectivity was first checked for the two signature peptides of Bevacizumab, for the SIL peptide and for the signature peptide of mAbIS compared to pooled blank human plasma from different healthy donors (Figure 5a). The absence of signal in blank human plasma, at the retention time of each of the four signature peptides, and the absence of SRM transitions of the 4 peptides attested to their specificity for Bevacizumab, SIL and mAbIS respectively when compared to endogenous protein that could be present in blank human plasma treated with the same process. The low intensity of the background noise and the absence of other signal in blank human plasma demonstrate also the ability of the treatment process to provide clean samples which can ensure an easy interpretation of chromato grams and a good sensitivity.

Moreover, Selectivity was in a second time checked for the two signature peptides of Bevacizumab, for the signature peptide of mAbIS and SIL peptide compared to respectively pooled blank human plasma spiked with mAbIS and SIL peptide (Figure 5b), pooled blank human plasma spiked with Bevacizumab and SIL peptide (Figure 5c) and pooled blank human plasma spiked with mAbIS and Bevacizumab (Figure 5d). For these three experiences, the absence of signal at the retention time of Bevacizumab signature peptides, of mAbIS and of SIL peptide respectively, and the absence of SRM transitions of these peptides attested to their specificity (Figure 5b, c, d).

To compare the 2 internal standard approaches, monoclonal antibody (mAbIS) and stable isotope labeled peptide (SIL peptide), we assessed for each approaches linearity, accuracy and precision by intraday repeatability assays as described in experimental section. The peak area of each signature peptide of Bevacizumab was normalized either by the peak area of the mAbIS or by the peak area of the SIL peptide. For these two internal standards, linear regression was used to establish calibration curve from results of the LC- ESI-QqQ analysis of the six standards. The r² values were 0.9904 for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and 0.9660 for ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) when mAbIS was used. The r² values were 0.9994 for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and 0.9987 for STAYLQMNSLR (SEQ ID NO 2) when SIL peptide was used. The intraday precision of the QCs was between 8.7% and 11.7% for the ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and between 8.07 % and 23.9% for ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) when mAbIS was used. The intraday precision of the QCs was between 5.49% and 8.9% for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and between 5.39% and 10.09 % for ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) when SIL peptide was used. Considering these preliminary results from intraday repeatability assay, the internal calibration approach using SIL peptide provide less variability than mAbIS. Interday repeatability study was therefore only assessed for SIL peptide approach (see results in table 3). The interday precision of the QCs was between 6.98% and 9.85% for ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and between 11.3 % and 19.2 % for ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2). Accuracy of the QCs was between 101.7% and 110.6% for the ⁶⁸FTFSLDSTK⁷⁶ (SEQ ID NO 1) and between 104.6% and 110.4% for ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2). Given the results obtained after repeatability assay, we opted to quantify bevacizumab with the approach using SIL peptide FTFSLDTSK* as internal standard basing bevacizumab quantification on the ratio of the peak area of ⁶⁸FTFSLDTSK⁷⁶ bevacizumab surrogate peptide and the peak area of FTFSLDTSK* SIL peptide. Indeed this approach ensures to quantify bevacizumab with sufficient precision and accuracy according to the FDA guidelines for bioanalytical method validation. Identification of bevacizumab is provided by the presence of the two surrogate peptides for which we have shown the proportionality between the generated signal and the amount of bevacizumab.

Since ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) present less variability in the measurement than ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) recovery of affinity chromatography on protein A and ultrafiltration step was determined for bevacizumab through mean peak area of ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1). Recovery of affinity chromatography was assessed comparing mean peak area of plasma spiked with Bevacizumab before affinity chromatography step, with blank plasma spiked with Bevacizumab just before ultrafiltration step. Recovery obtained for this step was 84+4.6% (CV=5.5%). Recovery of ultrafiltration step was assessed comparing mean peak area of ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) derived from pre-treated and proteolysed plasma sample spiked with Bevacizumab just before ultrafiltration, with serum spiked with Bevacizumab just before proteolysis step. Recovery obtained for this step was 92+9.2% (CV=10.0%). Overall recovery of sample pre-treatment before proteolysis was assessed by comparing mean peak area of ⁶⁸FTFSLDTSK⁷⁶ derived from pre-treated and proteolysed plasma sample spiked with Bevacizumab before any step of the treatment, with plasma spiked with Bevacizumab just before proteolysis. This overall recovery of sample pre-treatment was 74.38+10.3% (CV%=13.8) with generally better recovery at the highest concentration of Bevacizumab. Recovery of proteolysis step was assessed for the two surrogate peptide of bevacizumab by comparing mean peak area of ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) derived from treated and proteolysed plasma sample spiked with Bevacizumab before proteolysis, with treated and proetolysed plasma spiked with the synthetized ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) and ⁷⁷ ST A YLQMNS LR⁸⁷ (SEQ ID NO 2) after proteolysis. Recovery of proteolysis was 47.7+0.6% (CV%=1.25 %) and 47.3+0.4% (CV%=0.8 %) for ⁶⁸FTFSLDTSK⁷⁶ and ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) respectively.

Table 1. Ionization and fragmentation parameters of each identification and quantification SRM transitions from the two doubly charged surrogate peptides of bevacizumab, the doubly charged surrogate peptide of mAbIS and the doubly charged SIL peptide.

Cell

Declustering Entrance Collison Cell exit entrance

Peptide potential, potential, Energy, transition potential, potential,

DP (eV) EP (eV) CE (eV) CXP (eV)

CEP (eV)

523,3→898,4 8,5

523,3→797,3 8,5

523,3→650,3 6,5

⁶⁸FTFSLDTSK⁷⁶ 40 6,5 25 30

523,3→563,2 8,5

523,3→450,2 4,5

523,3→335,2 6,5

642,3→1024,5 10

642,3→861,3 35

642,3→748,3 7

⁷⁷STAYLQMNSLR⁸⁷ 50 3,5 20 35

642,3→620,3 5

642,3→489,3 5

642,3→375,2 5

545,3→597,3 4

³¹DTYIHWVR³⁸ 70 10 26 38

545,3→710,4 4 Cell

Declustering Entrance Collison Cell exit entrance

Peptide potential, potential, Energy, transition potential, potential,

DP (eV) EP (eV) CE (eV) CXP (eV)

CEP (eV)

527,3→805,3 34

527,3→658,3 8

FTFSLDTSK* 40 6,5 26 30

527,3→458,2 6

527,3→343,2 6

Table 2: Specific tryptic peptides of bevacizumab and mAblS identified in silico with their respective highest BLAST score obtained against UniProtKB_HUMAN database.

Subject

Bit

Position Sequence E-Value identities

Score

(%) ^a

Bevacizumab

68-76 (heavy

chain) FTFSLDTSK (SEQ ID NO 1) No hits found

77-87 (heavy 100% chain STAYLQMNSLR (SEQ ID NO 2) 32,9 0,007 (9/9→l l)

46-61 (light 1,00E- 87%(13/15→ chain) VLIYFTSSLHSGVPSR (SEQ ID NO 3) 39,2 04 16)

88-98 (heavy 5,00E- 100% chain) AEDTAVYYCAK (SEQ ID NO 13) 39,2 05 (11/11→11)

44-65 (heavy GLEWVGWINTYTGEPTYAADFK (SEQ ID 1,00E- 86% chain) NO 14) 53,7 10 (19/22→22)

20-38 (heavy LSCAASGYTFTNYGMNWVR (SEQ ID NO 8,00E- 84% chain) 15) 57,5 11 (16/19→19)

19-42 (light VTITCSASQDISNYLNWYQQK (SEQ ID NO 9,00E- 95% chain) 16) 68,9 15 (20/21→24)

62-103 (light FSGSGSGTDFTLTISSLQPEDFATYYCQQYS 5,00E- 88% chain) TVPWTFGQGTK (SEQ ID NO 17) 78,6 18 (37/42→42) niAMS

68-76 (heavy

chain) FTISADTSK (SEQ ID NO 4) 25,7 1,3 89% (8/9→9)

31-38 (heavy

chain) DTYIHWVR (SEQ ID NO 5) 29, 1 0,1 88% (7/8→8)

77-87 (heavy 5,00E- 100% chain) NTAYLQMNSLR (SEQ ID NO 6) 36,3 04 (10/10→11)

25-42 (light 1,00E- 75%(12/16→ chain) ASODVNTAVAWYQQKP (SEQ ID NO 18) 38,8 04 18) ^a: Subject identity corresponds to the number of common amino acid residue between "subject sequence" / "query sequence". The number after the arrow corresponds to the amino acid length of the sequence of interest. Table 3. Results of intra- and inter-day validation on pooled human plasma sample spiked with bevacizumab (QCl, QC2, QC3) and analysed by LC-QqQ MRM method under ratio of the ⁶⁸FTFSLDTSK⁷⁶ (SEQ ID NO 1) or ⁷⁷STAYLQMNSLR⁸⁷ (SEQ ID NO 2) bevacizumab surrogate peptide to SIL peptide ⁶⁸FTFS LDTS K* ⁷⁶ (SEQ ID NO 1). All QC samples were submitted to Protein A enrichissement, concentration by ultrafilitration and proteolysis during 8 hours at 1/50 (mole/mole) trypsin ratio.

Theoretical Observed

Accurarcy Precision

concentration (mean + n

(%) (CV%)

(Mg/mL) S.D.)^a ⁶⁸FTFSLDTSK⁷⁶

40 41,77 + 2,30 104,42 5,49 6

200 211,5 + 12,82 105,75 8,06 6

Intra-day

430,83 +

400 38,75 107,71 8,99 6

40 40,68 + 4,007 101,69 9,85 24 (over 4 days)

219,36 +

Inter-day 200 17,72 109,68 8,08 24 (over 4 days)

442,46 +

400 30,91 110,61 6,98 24 (over 4 days)

⁷⁷STAYLQMNSLR⁸⁷

40 38,45 + 2,072 96, 13 5,39 6

175,17 +

Intra-day 200 16,57 87,53 9,46 6

417,5 +

400 42,13 104,38 10,09 6

40 41,84 + 4,72 104,59 11,28 24 (over 4 days)

Inter-day 217,86 +

200 41,89 108,93 19,23 24 (over 4 days)

Table 4 : Sequences

Claims

1. The use of an Internal Standard compound in a method for quantifying Bevacizumab in a sample by mass spectrometry, wherein the said Internal Standard compound is selected in a group comprising :

2. The use according to claim 1, wherein the method for quantifying Bevacizumab in a sample comprises the steps of :

a) subjecting to trypsin proteolysis a pre -proteolysis mixture consisting of the said sample in which is present a fixed and known amount of the said Internal Standard compound , whereby an assay mixture is provided, and wherein the said assay mixture comprises - proteolysis peptides derived from Bevacizumab, and

- the said labeled surrogate peptide of bevacizumab when it is used as the Internal Standard compound or proteolysis peptides derived from the said monoclonal antibody when the said monoclonal antibody is used as the Internal Standard compound,

b) determining by mass spectrometric analysis the ratio of (i) a selected proteolysis surrogate peptide derived from Bevacizumab to (ii) a surrogate peptide of Internal Standard selected in a group comprising (ii-a) the said labeled surrogate peptide of bevacizumab and (ii-b) a proteolysis surrogate peptide derived from the internal standard monoclonal antibody, and

3. The use according to any one of claims 1 and 2, wherein the said Internal Standard compound is a Stable Isotope Labeled (SIL) peptide selected in a group comprising the peptides of SEQ ID N° 1, SEQ ID N°2, and SEQ ID N°3.

4. The use according to any one of claims 1 and 2, wherein the said Internal Standard compound consists of a monoclonal antibody such as Trastuzumab.

5. The use according to any one of claims 2 to 4, wherein, at step b) of the said method, the said surrogate proteolysis peptide derived from Bevacizumab is selected in a group comprising the peptides of SEQ ID N° 1, SEQ ID N°2, and SEQ ID N°3.

6. The use according to any one of claims 2 to 4, wherein, at step b) of the said method, the selected proteolysis surrogate peptide derived from Bevacizumab is a peptide of SEQ ID N° 1.

7. The use according to any one of claims 2 to 4; wherein, at step b) of the said method, (i) the selected proteolysis surrogate peptide derived from Bevacizumab is a peptide of SEQ ID N° 1 and (ii) the surrogate peptide of Internal Standard is a labeled peptide of SEQ ID N°l.

8. The use according to any one of claims 2 to 7, wherein, at step a) of the method, trypsin is added at a trypsin / total protein molar ratio ranging from 1/100 to 1/10, advantageously from 1/75 to 1/25 and preferably froml/60 to 1/40.

9. The use according to any one of claims 2 to 8, wherein, at step a) of the method, trypsin is incubated during a time period ranging from 7 to 9 hours, and is preferably of about 8 hours.

10. The use according to any one of claims 1 to 9, wherein the said sample is selected in a group comprising whole blood, plasma and serum, or a sample derived therefrom.

11. The use according to any one of claims 1 to 10, wherein the said sample consists of a sample derived from a human organism.