EA039676B1 - Systems and methods for reducing lab-to-lab and/or instrument-to-instrument variability of multi-attribute method (mam) by run-time signal intensity calibrations - Google Patents

Abstract

Systems and methods are described for reducing lab-to-lab and/or instrument-to-instrument variability of Multi-Attribute Methods (MAM) analyses via run-time signal intensity calibration. In various aspects, multiple MAM-based instruments each have detectors and different instrument conditions defined by different instrument models or sets of settings. Each MAM-based instrument receives respective samples and a reference standard as a calibrant. Each MAM-based instrument detects, via its detector, sample isoforms of its respective sample and reference standard isoforms of the reference standard. The MAM-based instruments are associated with processor(s) that determine, via respective MAM iterations, correction factors and sample abundance values corresponding to the sample isoforms. The correction factors are based on the reference standard, and the sample abundance values are based on the correction factors. A variance value of the sample abundance values may be reduced based on correction factors of each of the MAM-based instruments.

Description

Link to related applications

This application claims priority of U.S. Provisional Application No. 62/763110 (filed June 8, 2018) and U.S. Provisional Application No. 62/746323 (filed October 16, 2018). The entire contents of each of the above provisional applications are incorporated herein by reference.

The field of technology to which the invention relates

The present invention generally relates to reducing multi-parameter method (MAM) variability between laboratories and between instruments through on-line signal intensity calibrations.

Background of the invention

The development of biotherapeutic agents typically involves the monitoring of certain attributes of therapeutic molecules, during which such attributes are identified as critical quality attributes (CQA) in order to measure and evaluate the safety and efficacy of the product. Assays designed to measure and evaluate quality attributes may use mass spectrometry (MS). MS is usually understood as an analytical technique based on the ionization of chemicals and the sorting of ions based on their mass/charge ratio. Thus, using the MS device, it is possible to measure the mass of a molecule in a sample. In the case of polypeptide attributes, the use of mass spectrometry allows a large number of quality attributes to be assessed using only a few assays.

MS can be used to monitor post-translational modifications (PTM), including glycosylation profiles, and/or excipients using both data obtained through ultraviolet (UV) detection and mass data through the implementation of multiple analyte analysis methods /attributes or so-called multi-parameter methods (MAM). MAM uses a combination of MS data and automated identification and relative quantification of attributes (Rogers, RS, et al. 2015. Development of a quantitative mass spectrometry multi-attribute method for characterization, quality control testing and disposition of biologics. mAbs 7:5 , 881-890). Because of the efficiency and quality control advantages of MAM, it is increasingly being used with MS, for example, to provide increased selectivity, sensitivity, and flexibility for quality attribute analysis. MAM refers to an analytical approach that can quantify multiple attributes of products and processes (eg quality attributes/CQA) within a single analysis. For example, MAM-based assays are generally aimed at tracking the isolation and purification of a target product, but are increasingly being used for quality control assays at the release of batches (eg samples).

For example, the MAM procedure based on proteolytic digestion followed by liquid chromatography/mass spectrometry (LC-MS) analysis of proteolytic peptides (peptides are fragments of larger polypeptides produced by proteolysis) can be used to quantify a set of quality attributes. therapeutic proteins. This procedure takes advantage of the resolution provided by a mass spectrometric detector, and the MS intensity of each isoform (including modified and unmodified forms) of a proteolytic peptide can be used for quantitation.

Combining mass spectrometry with MAM can be challenging because MS procedures typically require highly trained analytical chemists and complex laboratory infrastructure. In particular, the main difficulty in MAM analysis based on mass spectrometry is the observed high variability in the prepared samples used in the laboratory and the different instruments in these laboratories. For example, variability in sample preparation may occur due to different analytical chemists performing sample preparation differently, resulting in differences between prepared samples. For example, after receiving a sample, an analytical chemist typically performs a complex procedure (eg, proteolytic digestion) to prepare the sample for introduction into the analyzer. Due to the complexity of the sample preparation procedure, samples may differ to some extent even if the original sample is the same. The value of various indicators may change due to modifications due to the duration of the sample preparation procedure. These artificial modifications lead to inaccuracies and variations in MAM results. Varying digestion efficiencies during sample preparation by an analytical chemist also contribute to lab-to-lab variability. Variability can also arise from laboratory measuring instruments using different settings or using different operating models. Currently, in order to guarantee reproducible attribute measurements, not only should all analytical laboratories use a similar instrument model, but the instruments must also be brought into the same condition. However, limiting labs to a particular model can also deter labs from upgrading their equipment, preventing them from benefiting from cutting-edge advances.

The disadvantages of MS-MAM based analysis also arise from the assumptions and approaches used.

- 1 039676 used in traditional MAM procedures. For example, in a traditional MS-MAM procedure, each attribute score (eg, different state of modification of an amino acid residue in a peptide) is determined based on the MS responses (eg, peak areas) of the modified peptide and the unmodified peptide under the following assumptions: (1) unmodified and modified peptides are characterized by a reproducible recovery from laboratory to laboratory; (2) unmodified and modified peptides are characterized by the same MS-response coefficient and (3) artificially introduced changes in attributes are insignificant.

Because of these assumptions, the conventional MAM procedure depends on several necessary conditions, including the provisions that (1) the digestion efficiency is reproducible from laboratory to laboratory; (2) MS instrument states are exactly the same, and (3) sample preparation performed in different labs introduces minimal or constant artificial change. However, in reality, such required conditions are difficult to meet due to differences in sample preparation procedures, analytical chemist skills, instrumentation, reagent quality, and so on. The degree of extraction of the peptide can also be subject to fluctuations, which leads to additional variability.

Additional difficulties in achieving the required conditions include differences in MS instrument models or instrument settings. For example, the maintenance of one lab meter may be different from the maintenance of another lab meter. In addition, the response coefficient for peptides containing different variants may differ in different laboratories. Additional difficulties in achieving the required conditions include differences in sample preparation procedure, skill of analytical chemists, quality of devices and reagents, and instrument condition. The amount of artificially introduced modification may also vary.

As a result, traditional MS-based MAM methods lack robustness with variability between labs and/or instruments.

In addition, the main disadvantage of the multi-parameter method (MAM) based on mass spectrometry is the high variability of the results obtained between analytical chemists and instruments. Reproducible attribute measurements require not only the same instrument model for all analytical laboratories, but the instruments themselves must also be set up in the same way. This poses challenges in the long term, given the rapid development of new chromatographic and mass spectrometric technologies. In addition, differences in cleavage efficiency and artificial modifications (eg, oxidation, deamidation, isomerization of aspartic acid residues, and fragmentation) during sample preparation also contribute to lab-to-lab variability. These shortcomings need to be addressed in order to ensure the long-term success of MAM, for example under the current GMP regulations.

Short description

Therefore, there is a need for systems and methods to reduce MS-based multi-parameter method (MAM) variability between labs and/or instruments through on-line signal strength calibrations.

The system and methods described herein are disclosed with the aim of reducing multivariable method (MAM) variability between laboratories and/or instruments during operation. In various embodiments, it is described that these systems and methods can be used to determine a score of an attribute in a sample (e.g., score values of each quality attribute in a sample) using a measured or known attribute score in a reference standard (e.g., reference standard values quantitative indicator of each quality attribute) as a calibrant. This new technique improves inter-laboratory precision and/or inter-instrument accuracy and reduces variability across labs and/or instruments. For example, this new method requires no or minimal additional work by an analytical chemist or laboratory, since data on the reference standard for MAM has usually already been obtained. This is because, in a typical MAM procedure, the reference standard is analyzed in parallel with samples for other purposes, such as checking system suitability or authenticity. In addition, the use of a reference standard as a calibrant is also advantageous because in a reference standard most of the attribute scores remain constant over the lifetime of the standard, and for this reason the reference standard can be used as a calibrant in a particular way to correct for differences between instruments or sample preparation procedures.

In the various embodiments described herein, systems and methods are described for reducing the variability of MAM assays between laboratories or instruments through on-line signal intensity calibrations. For example, in some embodiments, such systems and methods may include a first MAM-enabled device containing trans

- 2 039676 output detector. The first MAM-enabled device may be characterized by the first device state defined by at least one of (1) the model of the first device or (2) the first set of settings. The first MAM-enabled instrument can be adapted to receive the first sample and reference standard. The first MAM-enabled instrument can be further adapted to detect, by means of the first detector, the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard.

The systems and methods may further include one or more processors associated with the first MAM-enabled device. One or more processors associated with the first MAM-enabled instrument may be adapted to determine, by a first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample. The first set of correction factors may be based on a reference standard. One or more processors associated with the first MAM-enabled instrument may be further adapted to determine a first set of sample score values corresponding to an isoform of the first sample, where the first set of sample score values is based on the first set of correction factors.

The systems and methods may further include a second MAM-enabled instrument comprising a second detector. The second device with MAM support may be characterized by the state of the second device, defined by at least one of: (1) the model of the second device or (2) the second set of settings.

In various embodiments, the state of the second device may be different from the state of the first device.

A second MAM-enabled instrument can be adapted to accept a second sample and reference standard. The second MAM-enabled instrument can be further adapted to detect, via the second detector, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard.

The systems and methods may further include one or more processors associated with a second MAM-enabled device. One or more processors associated with the second MAM-enabled instrument may be adapted to determine, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample. The second set of correction factors may be based on a reference standard. One or more processors associated with the second MAM-enabled instrument may be further adapted to determine a second set of sample score values corresponding to an isoform of the second sample, where the second set of sample score values is based on a second set of correction factors.

Given the correction factors defined for each instrument, the measurement variability between the first MAM-enabled instrument and the second MAM-enabled instrument can be reduced. For example, the variation value for the first set of sample score values and the second set of sample score values may be reduced by considering the first set of correction factors for the first MAM-enabled instrument and the second set of correction factors for the second MAM-enabled instrument.

In further embodiments, calibration systems and methods are disclosed for reducing the variability of an MAM-enabled instrument over multiple time periods by calibrating signal strengths during operation. In such embodiments, the MAM-enabled instrument may receive a first sample and a reference standard for a first period of time.

The MAM-enabled instrument can detect the first isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard through the detector for a first period of time.

One or more processors may be adapted to determine, by a first iteration of the MAM, during a first time period, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard.

One or more processors may also be adapted to determine, by a first iteration of the MAM, and within a first period of time, a first set of indicator values for a sample corresponding to the isoform of the first sample. The first set of sample score values may be based on the first set of correction factors. The device with MAM support during the first period of time can be characterized by the state of the first device, given by the first set of settings.

An instrument with MAM support for a second period of time can be adapted to accept a second sample and reference standard.

The MAM-enabled instrument can detect, by means of a detector, for a first period of time, the first sample isoform in the first sample and the first reference standard isoform in

- 3 039676 reference standard.

One or more processors may be configured, by means of a second iteration of the MAM over a second period of time, to determine a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard.

One or more processors may also be adapted to determine, through a second MAM iteration, and over a second time period, a second set of sample score values corresponding to the isoform of the second sample. The second set of score values in the sample may be based on the second set of correction factors. The device with MAM support during the second period of time can be characterized by the state of the second device, determined by the second set of settings.

The state of the second MAM-enabled device during the first time period may be different from the state of the second MAM-enabled device during the second time period.

Considering the correction factors determined for each time period, the measurement variability between MAM iterations of the MAM enabled instrument between the first time period and the second time period can be reduced. For example, the variation value for the first set of sample score values and the second set of sample score values may be reduced by considering the first set of correction factors for the first MAM-enabled instrument and the second set of correction factors for the second MAM-enabled instrument.

As described hereinafter, for example with reference to FIG. 4a, 4b, 5a, and 5b, analysis of various isoforms demonstrates that the use of the systems and methods disclosed herein reduces inter-laboratory and/or instrument variability (e.g., RSD intralaboratory precision) by a factor of two or three compared to existing procedures with MAM support. With the systems and methods disclosed, there is no longer a need for consistency between laboratories on instrument models, on instruments, or on MAM iterations performed over different time periods. At the same time, minor changes in the digestion procedure from laboratory to laboratory, as well as changes due to automation, do not significantly affect the results of analyzes obtained in different laboratories, with different instruments or MAM iterations performed at different times.

The new system and methods also provide additional options for other tools, such as a triple quadrupole instrument for monitoring a selected reaction, because due to the disclosed systems and methods, consistency in response factors between different peptide isoforms is no longer required.

In addition, the disclosed systems and methods have significant advantages over conventional methods because the new systems and methods significantly reduce inter-laboratory variability in results by calibrating signal strengths during operation. The new system and methods described herein essentially eliminate the requirement for MAM to use single equipment/instruments, which is a major problem in existing MAM workflows. Moreover, the new systems and methods are applicable to any MAM-enabled application where precise quantification of quality attributes is required. In addition, since reference standards are typically analyzed in parallel with samples in biopharmaceutical workflows, no additional work is required from the analytical chemist.

According to the above and the contents of this document, the present invention includes improvements in computer functionality or improvements in other technologies, at least for the reason that the claims state that, for example, MAM-enabled device(s) can be improved by reducing variability among MAM-enabled instruments performing MAM procedures with the same sample (e.g., with the same sample of proteolytic (proteolytic) peptide(s) and using a reference standard as a calibrant. That is, the present invention describes improvements operation of the computer itself, or any other technology or technical field, as the systems and methods reduce variability among MAM-enabled devices.The foregoing improves on the prior art, at least because traditional MAM methods require conditions that are difficult to meet. due to differences in sample preparation procedure, skill of analytical chemists, quality of instruments and reagents.

The present invention relates to improvements in other technologies or technical fields, at least because MAM-enabled instruments can be calibrated even if they are of different instrument models and have different sets of settings, even among different laboratories.

The present invention includes the use of unique improvements for a particular device or in the conditions of use of a particular device, such as MAM-enabled devices.

- 4 039676

The present invention includes specific features that differ from those that characterize the well-understood, conventional, traditional activities in the field of development or research of biotherapeutic agents, and/or non-standard steps are added that focus the focus of the present invention on specific applicable applications, for example, reducing the variability of the multivariate method (MAM) between laboratories and instruments through on-line signal intensity calibrations.

In addition, to overcome the shortcomings described herein, new calibration systems and methods are described for calculating an attribute score in a sample of a measured attribute used in a reference standard as a calibrant. In a reference standard, most quality attributes remain constant throughout the life of the standard and can therefore serve as a calibrant to correct for differences between instruments or sample preparation procedures. Since the reference standard data is usually obtained by the typical MAM method, no additional work is required by the analytical chemist. Test data with a large number of attributes demonstrated that this methodology significantly reduces inter-instrument variability. With this methodology, it is no longer necessary to match the instrument model and sample preparation procedure. As a result of the change in the splitting procedure, improvements in modern instrumentation will not have a significant impact on the result of the analysis. The new systems and methods also provide the ability to calibrate other instruments, such as a triple quadrupole selective response monitor, since consistent response factors between different peptide isoforms are no longer required.

As described herein, multi-parameter methods based on proteolytic digestion followed by LC-MS analysis of proteolytic peptides are designed to quantify a set of quality attributes of therapeutic proteins. These methods take advantage of the resolution provided by a mass spectrometry (MS) detector, with the MS intensity of each isoform (including modified and unmodified forms) of a proteolytic peptide used for quantitation. Due to the high specificity of these methods for each clinically significant quality attribute, such methods have received considerable attention in the biopharmaceutical industry.

The advantages will become more apparent to those skilled in the art from the following description of the preferred embodiments which have been shown and described by way of illustration. It will be understood that the present embodiments may be capable of other and different embodiments, and their details are open to modification in various aspects. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive.

Brief description of graphic materials

The figures described below represent various aspects of the system and methods disclosed herein. It should be understood that each figure represents an embodiment of a particular aspect of the disclosed systems and methods, and that each of the figures is intended to be consistent with possible embodiments thereof. In addition, where possible, the following description refers to the reference numerals found in the following figures, in which features shown in multiple figures are indicated by consistent reference numerals.

The drawings show the devices discussed herein, however, it should be understood that the embodiments of the present invention are not limited to the exact devices and tools that are shown, with FIG. 1 shows an exemplary MAM-enabled instrument comprising a mass spectrometer according to the various embodiments disclosed herein.

In FIG. 2 shows an exemplary flow diagram for reducing variability between a first MAM-enabled instrument and a second MAM-enabled instrument through in-service signal strength calibrations according to the various embodiments disclosed herein.

In FIG. 3 is an illustration of a method for reducing the variability of an MAM-enabled instrument over multiple time periods by calibrating signal strength during operation in accordance with the various embodiments disclosed herein.

In FIG. 4a is a graph depicting the reduction in inter-instrument/inter-laboratory score variation for a given isoform in six exemplary MAM iterations according to the various embodiments disclosed herein.

In FIG. 4b is a graph depicting the reduction in instrument/interlaboratory bias for six exemplary isoforms, including the isoform shown in FIG. 4a according to various embodiments disclosed herein.

In FIG. 5a is a diagram showing the decrease in the variation of quantitative values.

- 5 039676 inter-instrument/inter-laboratory measurements in twelve exemplary MAM iterations according to the various embodiments disclosed herein.

In FIG. 5b is a graph depicting the reduction in inter-instrument/inter-laboratory variation across 57 quality attributes according to the various embodiments disclosed herein.

In FIG. 6 is a graph of measured scores of two attributes in a sample at two different instrument setups and two different sample preparation procedures according to the various embodiments disclosed herein.

In FIG. 7 is a chart depicting an illustrative comparison of within-lab precision with and without response factor calibration according to the various embodiments disclosed herein.

In FIG. 8 is a diagram of an embodiment of artificial modification calibration for type 2 attributes showing similar results as for response factor a-calibration according to the various embodiments disclosed herein.

In FIG. 9 is a graph of the results of b-calibration of artificial modification for type 3 attributes compared to response coefficient a-calibration according to the various embodiments disclosed herein.

In FIG. 10 is an embodiment diagram showing the results of calibrating with two standards (a and b) for type 3 attributes compared with a single standard response factor calibration (a) according to the various embodiments disclosed herein.

In FIG. 11 is a diagram showing the improvement made to the consistency of two datasets acquired on two different LC-MS systems by calibrating the response factor according to the various embodiments disclosed herein.

In FIG. 12 is a graph showing the measured value of glycoform content in four sequences, in triplicate of each sequence, according to the various embodiments disclosed herein.

In FIG. 13 is a chart showing the relationship of attribute score and RSD within a sequence of three attributes according to the various embodiments disclosed herein.

The preferred embodiments in the figures are for illustrative purposes only. Alternative embodiments of the systems and methods presented herein can be applied without deviating from the principles of the present invention described in this document.

Detailed description

In FIG. 1 shows an exemplary MAM-enabled instrument 100 comprising a mass spectrometer (MS) 102 according to the various embodiments disclosed herein. This document describes that MAM can be used to simultaneously analyze multiple properties or attributes (eg, quality attributes) of a molecule (eg, protein or isoform). To measure properties or attributes of individual molecules, mass spectrometer 102 may receive a sample 110 of molecules or peptides, such as an isoform sample or isoform reference standard, as described herein. The mass spectrometer 102 can convert the received molecular sample into ions, whereby the ionized forms of the molecular sample can be filtered out and identified.

Typically, a mass spectrometer contains an ion source, a mass analyzer and a detector. For example, in the case of an ion source 114, a small sample (eg, sample 110) of molecules or peptides is ionized (151), usually into cations, by adding one or more protons. Mass analyzer 116 sorts and separates ions (eg, ions 152) according to their mass and charge. A detector (eg, detector 140) measures separated ions, and the results may be recorded and visually displayed by a computing device (eg, computing device 142) having one or more processors, such as with a graph or other report. One or more processors may be part of MAM-enabled device 100 or part of a separate computing device (eg, computing device 142).

Ions 152 can be detected electronically by detector 140, where ions 152 have different intensities, thereby generating different or varying ion intensities (i.e., signals) that detector 140 detects. Ions 152 detected by detector 140 can be read, stored and/or analyzed in the computing device 142, where, for example, detected ions can generate electronic information (eg, peak areas of registered ions, etc.). Thus, the MAM capable instrument 100 described herein can generate ion intensity information. The intensity of the ions may be displayed by means of a two-dimensional (2D) plot, plot or recording, where, for example, the y-axis of the mass spectrum of such a scheme may represent the intensity of the signal of the ions. Usually the value of the mass-to-charge ratio

- 6 039676 (m/z) is measured on the x-axis of a diagram, graph or record, where m is the molecular or atomic mass number, and z is the charge number of the ion.

In addition, the response factor can be determined by analyzing the ion intensity/signals generated from the MS information. The response factor may be equal to the ratio of the ion intensity signal generated by the molecule or isoform (eg, as determined from 152 ions) and the amount of molecule or isoform that generates the signal. Response factors (eg, k) and ionic intensities (eg, Ii) are further described in this document, for example, in relation to table. one.

It should be understood that the ionization of molecules (eg, isoforms) by means of a mass spectrometer can be performed in various ways. Although the mass spectrometer embodiment of FIG. 1 is presented and described in one such way, it is to be understood that any mass spectrometer or method for performing mass spectrometry can be used for the systems and methods described herein. For example, the mass spectrometer 102 may comprise or be based on a mass spectrometric tool based on Orbitrap, TOF (time of flight), and/or one or three quadrupoles.

Computing device 142 may include one or more processors and/or one or more computer storage devices for reading, storing, or analyzing information about molecules, isoforms, reference standards, ions, or other information described herein. One or more processors and/or one or more computer storage devices of computing device 142 may also be used to implement any of the functions, methods, flow charts, or other elements described herein in relation to reducing the variability of MAM analysis between laboratories or instruments by signal intensity calibration during operation. In addition or alternatively, MAM-enabled device 100 may include one or more processors and/or one or more computer storage devices of computing device 142, which may also be used to implement any of the functions, methods, flow diagrams, or other elements described in this document in relation to reducing the variability of MAM analysis between laboratories or instruments by calibrating the signal intensity during operation, as described in this document. In FIG. 1 shows that computing device 142 may be causally connected to MAM-enabled device 100 directly (eg, fixedly connected by a cable such as a universal serial bus (USB) cable) or via a computer network (either private or public, such as the Internet). ) and may be communicatively coupled to any of the components of the mass spectrometer 102, including, for example, any of the ion sources 114, mass analyzer 116, and/or detector 140.

In particular, either MAM capable device 100 or computing device 142 may be a computing device that may include one or more processors as well as one or more computer storage devices. Memory devices may include one or more forms of volatile and/or non-volatile, fixed and/or removable memory devices such as read only memory (ROM), electronic programmable read only memory (EPROM), random access memory (RAM), erasable electronic programmable Read Only Memory (EEPROM) and/or other hard drives, flash memory, MicroSD memory cards, and others. Storage devices may store an operating system (OS) (eg, Microsoft Windows, Linux, Unix, etc.) capable of providing the functionality discussed in this document. The storage devices may also store machine-readable instructions, including any of one or more applications, one or more software components, and/or one or more application programming interfaces (APIs), which may be implemented to simplify or perform features, functions, or other information. described in this document, such as any methods, processes, elements or limitations that are illustrated, presented or described in various flowcharts, illustrations, diagrams, figures and/or other information in this document. For example, at least some of the applications, software components, or APIs may be, contain, or otherwise be part of a machine learning component and/or search engine optimization component, where each is adapted to provide their various functionality discussed in this document. . It should be understood that one or more other applications executed by the processor(s) of MAM-enabled device 100 or computing device 142 may be provided.

The processor(s) of MAM-enabled device 100 or computing device 142 may be coupled to the storage devices of MAM-enabled device 100 or computing device 142 via a computer bus responsible for transmitting electronic data, data packets, or other electronic signals to the processor(s) and to and from storage devices, to implement or execute machine-readable instructions, methods, processes, elements or restrictions,

- 7 039676 which is illustrated, presented or described in various block diagrams, illustrations, diagrams, figures and/or other contents of this document.

The processor(s) of MAM-enabled device 100 or computing device 142 may communicate with the storage device via a computer bus to launch and execute an operating system (OS). The processor(s) may also interact with the storage device via a computer bus to create, read, update, delete, or otherwise access or act on data stored in the storage devices of the MAM-enabled device 100 or computing device 142 and/or in a database of MAM-enabled instrument 100 or computing device 142 (eg, a relational database such as Oracle, DB2, MySQL, or a NoSQL-based database such as MongoDB). Data stored in storage devices and/or databases may include all or part of any data or information described herein, including, for example, one or more search queries, one or more transaction details, and user profile information.

The MAM-enabled device 100 or computing device 142 may further comprise a communications component adapted to transmit (e.g., send and receive) data through one or more external/network ports to one or more networks or local terminals, such as a computer network and/or computing device 142 described herein. In some embodiments, the communication component may include client/server platform technology such as ASP.NET, Java J2EE, Ruby on Rails, Node.js, a web service or online API responsible for receiving and responding to electronic requests. The processor(s) of MAM-enabled device 100 or computing device 142 may implement a communications component of MAM-enabled device 100 or computing device 142 that may cooperate to implement or execute machine-readable instructions, methods, processes, elements, or constraints, as illustrated. , is represented or described in the various flowcharts, illustrations, diagrams, figures and/or other contents of this document. In some embodiments, the communications component of MAM-enabled device 100 or computing device 142 may include or interact with one or more transceivers (e.g., WWAN, WLAN, and/or WPAN transceivers) that operate in accordance with IEEE standards, standards 3GPP or other standards and which can be used to receive and transmit data through the external/network ports of the MAM-enabled device 100 or computing device 142.

MAM capable device 100 or computing device 142 may further comprise or implement an operator interface adapted to present information to an administrator or operator and/or to receive input from an administrator or operator. The operator interface of MAM-enabled instrument 100 or computing device 142 may provide a display screen (eg, via computing device 142-106). Any of the MAM capable device 100 or computing device 142 may also provide input/output components (eg, ports, capacitive or touch input panels, keys, buttons, lamps, LEDs) that can be directly accessed through any of the device 100 s. capable of, or connected to, the MAM or computing device 142, or may be indirectly accessed through or connected to the terminal of the MAM-enabled device 100 or computing device 142. In some embodiments, an administrator or operator may access the server 102 through the operator interface of the MAM-enabled instrument 100 or computing device 142 and/or I/O components to view information, make changes, enter training data, and/or perform other functions.

In some embodiments, any of the MAM-enabled device 100 or computing device 142 may perform the functions described herein as part of a cloud network, or may otherwise communicate with other hardware or software components within the cloud to send, receive, or otherwise analyze the data or information described in this document.

In general, a computer program or computer-based product according to some embodiments may comprise a storage medium usable for computer use, or a tangible, non-volatile computer-readable medium (e.g., standard random access memory (RAM), an optical disk, a universal serial bus (USB) drive, or the like). the like), on which computer-readable program code or computer instructions are recorded, which computer-readable program code or computer instructions can be installed on the MAM-enabled device 100 or computing device 142 or otherwise adapted for execution by their processor (processors) (for example, operating in associated operating system on storage devices) to provide, implement, or execute machine-readable instructions, methods, processes, elements, or limitations illustrated, represented, or described in various flowcharts, illustrations, diagrams. grams, figures and/or other content of this document. In this regard, the program code may be implemented in any desired programming language and may be implemented as machine code, as

- 8 039676 sampler code, bitcode, interpreted source code, or the like (for example, in Golang, Python, C, C++, C#, Objective-C, Java, Scala, Actionscript, Javascript, HTML, CSS, XML, etc.) d.).

In FIG. 2 shows an exemplary flow diagram for reducing variability between a first MAM-enabled instrument 210 and a second MAM-enabled instrument 260 through on-line signal strength calibrations according to the various embodiments disclosed herein. In some embodiments, the first MAM-enabled instrument 210 and the second MAM-enabled instrument 260 may be located in the same laboratory. In other embodiments, the first MAM-enabled instrument 210 and the second MAM-enabled instrument 260 may be located in different laboratories. For example, the first MAM-enabled instrument 210 may be located in a first laboratory at a first geographic location, and the second MAM-enabled instrument 260 may be located in a second laboratory at a second geographic location. In some embodiments, the first MAM device 210 may be communicatively coupled to one or more processors associated with the second MS device over a computer network, as described in connection with FIG. one.

Each of the first MAM-enabled device 210 and the second MAM-enabled device 260 may be configured the same or similar to that described for the MAM-enabled device 100 shown in FIG. 1. Accordingly, the description with respect to FIG. 1 equally or similarly applies to any of the first MAM-enabled device 210 and/or the second MAM-enabled device 260. In addition, each of the first MAM-enabled device 210 and the second MAM-enabled device 260 may comprise a computing device (eg, computing device 142) as described in connection with FIG. 1. In various embodiments, the first MAM-enabled instrument 210 and the second MAM-enabled instrument 260 may be or comprise the mass spectrometry (MS) instruments described in connection with FIG. 1. In other embodiments, the first MAM-enabled tool 210 and the second MAM-enabled tool 260 may be or comprise tri-quadrupole tools.

In the embodiment shown in FIG. 2, the first MAM-enabled tool 210 includes a first detector (eg, detector 140 described herein in connection with FIG. 1). In addition, the first MAM-enabled device 210 is characterized by the state of the first device as determined by (1) the model of the first device or (2) the first set of settings. For the MAM-enabled instruments described herein, the instrument model may define how the MAM-enabled instrument analyzes, reads, or otherwise outputs molecular, isoform, ion, or other related information described in this document. For example, an exemplary embodiment of a device model or configuration for ionizing molecules is described herein for MAM capable device 100 shown in FIG. 1. Similarly, MAM-enabled devices may have a set of settings for a MAM-enabled device that can affect the operation of a MAM-enabled device. For example, settings can change how the MAM-enabled instrument performs ionization and/or changes the sensitivity of how ions are read or detected or otherwise reported.

Differences in any instrument model or instrument settings may cause the MAM-enabled instrument to function differently, which may cause the MAM-enabled instrument to detect and/or read molecules, isoforms, ions in a different way. Thus, MAM-enabled instruments with different instrument models or instrument settings may have different states and therefore may operate differently from instrument to instrument and/or lab to lab that may host such different instruments.

In the embodiment shown in FIG. 2, the first MAM-enabled instrument 210 is adapted to receive a first sample 204 (eg, a proteolytic peptide sample) and a reference standard 202 (eg, a proteolytic peptide). In the embodiment shown in FIG. 2, the concentration of an attribute (eg, a quality attribute) in the first sample 204 is unknown. However, in the reference standard 202, the concentration of the attribute is known and can therefore be used as a calibrant. For example, a reference standard may be a chemical sample that has a certain known chemical composition, for example, a sample with a certain lot number, or a lot can have a first chemical or attribute of 80%, a second chemical or attribute of 10%, and various minor percentages of other residual chemicals or attributes. The reference standard may have the same chemical composition or attributes and/or the same signal signature (which, for example, is detected by a mass spectrometer as described in connection with FIG. 1) as other reference standard samples of the same chemical or composition. . Thus, a reference standard can be used for quality control purposes where a test sample (eg, first sample 204 and/or second sample 254) is compared to a reference standard (eg, reference standard 202 and/or reference standard 252) to determine quality. , quantity, agreement, variation and/or discrepancy between a test item and a reference standard or pairs

- 9 039676 Tia. For example, for epoetin-alpha (recombinant erythropoietin) protein (eg, Epogen® from Amgen), a reference standard lot can be provided that can be used to compare batches of epoetin test samples for quality control measurements.

As shown in FIG. 2, by means of the first MAM instrument 210, the reference standard 202 is analyzed in parallel with the sample 204. Similarly, for the second MAM instrument 260, the reference standard 252 is analyzed in parallel with the second sample 254. In the case of the reference standards 202 and 252, most of the quality attributes remain constant during the lifetime of each reference standard and therefore serve as a common calibrant to correct for difference(s) between instruments or sample preparation procedures, e.g., difference(s) in conditions between a first MAM-enabled instrument 210 and a second MAM-enabled instrument 260 and/or preparation procedures samples performed by analytical chemists operating the first assisted instrument 210 and/or the second assisted instrument 260.

This document describes that the benefit of using a reference standard as a calibrant in each run (for example, where reference standards 202 and 252 may be samples of the same reference standard used as calibrant) is that you can eliminate most of the requirements for traditional MAM in terms of reproducibility between labs and instruments. For example, assumptions about reproducibility between labs and instruments can be simplified and revised as follows: (1) unmodified and modified peptides have reproducible recovery in the same LC/MS sequence (e.g. when compared from lab to lab); (2) unmodified and modified peptides have a reproducible MS response factor in the same LC/MS sequence (eg, when compared to the same response factor); (3) artificial performance changes are negligible. The above requirements are easier to meet than those of the conventional method since most of the interlaboratory reproducibility requirements are relaxed to reproducibility within a single LC/MS run (e.g. same lab, same analytical chemist and same day).

As shown in the embodiment in FIG. 2, the first MAM-enabled instrument 210 is adapted to detect with its first detector the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard. In the context of proteolytic peptides, a protein isoform may be a protein variant that is representative of a highly similar set of proteins that perform the same or similar biological roles. In some embodiments, the implementation of the isoform of the first sample may be a quality attribute used for quality control, as described in this document. In other embodiments, the quality attribute may be a protein, or an identified contaminant, or other measure of the quality of a sample or lot. For example, in various embodiments, a quality attribute may be defined by fragmentation, oxidation, glycation, hydroxylation, sequence variants, isomerization, deamination, C-terminal lysine, Associated glycans, and/or N-terminal glycans.

The first MAM capable device 210 may be associated with one or more processors. One or more processors may be contained in the first MAM capable device 210, or may be part of a computing device (eg, computing device 142) communicatively coupled to the first MAM capable device 210, as described, for example, with respect to FIG. 1. In the embodiment shown in FIG. 2, one or more processors associated with the first MAM capable instrument 210 are adapted to determine, by a first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample. It should be understood that the determination of the set of correction factors may be performed before, during, or after the first iteration of the MAM, such that the determination of the set of correction factors is associated with the first iteration of the MAM. It is also to be understood that the set of correction factors described herein may comprise a single correction factor or a combination of correction factors. In the embodiment shown in FIG. 2, the first set of correction factors is based on the reference standard 202. This relationship can be expressed by the following equation (1) for n+1 isoforms associated with a particular amino acid residue, where a _i represents the first set of correction factors and I and A represent intensity values ions and attribute scores for reference standard 202.

- 10 039676

Thus, a reference standard (eg, represented by I and A ) is used as the calibrant, where the first set of correction factors (eg, ai) is determined from such values. As shown for equation (1), the first set of correction factors (eg, ai) is based on the ion intensities (eg, I00 and Ii ⁰ ) of the first isoform(s) of the reference standard and the score values in the first reference standard (eg, A00 and Ai ⁰ ) of the first isoform(s) of the reference standard. In table. 1 below describes the various notations used for equation (1) and/or as used elsewhere in this document:

Table 1

Designation Description Superscript 0 Values with superscript 0 represent reference standard values (e.g. samples 202 or 252 of the reference standard) A ⁰ Represents a known isoform score in the reference standard (for example, samples 202 or 252 of the reference standard) 1° Represents the measured ion intensity of the reference standard (e.g. samples 202 or 252 of the reference standard) Subscripts The subscript 0, 1, 2, ...ϊ,.,.η represents n+1 isoforms according to 0,1, 2, ...ϊ,.,.η given residue (including the unmodified form). The subscript 0 represents the most common form in the reference standard (e.g., samples 202 or 252 of the reference standard), which is usually the unmodified form To Represents the response coefficient of isoform i a; Represents a correction factor (for example, correction factor for response factor) for isoform i ao The value of ao is defined as 1 BUT li Represents the score of isoform i Represents the ion intensity (peak area) of isoform i

Equation (1) can be used to perform an intensity calibration (eg, ion intensity calibration) from a known attribute score in the reference standard (Ai ⁰ ).

In addition, the first set of correction factors (for example, a,) can be used to calibrate the response factor (k) associated with the first set of quantitative values in the sample (for example, A), to determine the value (values) of the ion intensity (Ii) isoforms (isoforms) of the first sample. This is shown in equation (2) below. In general, ion intensity values may represent peak areas as detected by MS detector 140 as described in connection with FIG. 1. It is shown below that the correction factor ai can be used to calibrate the response factor k which, when applied to the score value (A), provides a calibration of the ion intensity factor Ii. This is shown in equation (2) below, from which equation (1) is derived. ):

UkA° =l,i = 0,na _i kA _i - CA - 0,p

-« ₀ -1 (2)

I

Σ+ ⁼¹ y=o

In equation (2), the response factor (k) for each isoform (i) is modified by a correction factor (a,). The formula aik ⁰ represents the adjusted response factor for the reference standard (eg, 202), and a _i k represents the adjusted response factor for the sample (eg, the first sample 204). The response factor for the reference standard (k ⁰ ) may differ

- 11 039676 differ from the response factor of the first sample 204 (k) due to small differences in sample preparation, as well as differences in instrument sensitivity between injections (eg differences within the laboratory). Similarly, differences may also be introduced or created between the first sample 204 and the second sample 254 (eg, between laboratories). Equation (2) shows that in some embodiments, the correction factor a ₀ is set to 1 (a ₀ =1) to solve equation (2) to derive equation (1). However, it should be understood that it does not matter which isoform is set to a constant value (a ₀ =1); however, the most common isoform is usually established, which is usually the unmodified isoform.

The score value (Ai ⁰ ) of each attribute in the reference standard can be set either using the traditional MAM or the orthogonal method with better accuracy. If an absolute quantification method is used, all subsequent MAM analyzes of the same attribute become an absolute measurement after calibration.

Equations (1) and (2) can be used to calculate the score value (Ai) of each isoform i in a sample (eg, the first sample 204). For example, for equation (1), it is shown that the score value (Ai) of isoform i is determined from values including a correction factor (eg, ai) and signal intensity of each isoform (eg, isoform Ii).

In the embodiment shown in FIG. 2, one or more processors associated with the first MAM-enabled instrument 210 may further be adapted to determine a first set of sample score values (e.g., Ai as shown for equation (1)) corresponding to the isoform(s) of the first sample. where the first set of score values in the sample (eg, Ai as shown for equation (1)) is based on the first set of correction factors (eg, ai in equation (1)). As described herein, score values may be presented as percentages. As shown for equation (1), the first set of score values in the sample (eg, Ai) is also based on the ion intensity value(s) of the isoform(s) of the first sample (eg, Ii in equation (1)).

In FIG. 2 also provides an illustration of the calibration of the second MAM-enabled instrument 260. The second MAM-enabled instrument 260 is calibrated (270) using the same reference standard (or another sample of the same reference standard) that is used to calibrate (220) the first MAM-enabled instrument 210. Thus, both the second MAM-enabled instrument 260 and the first MAM-enabled instrument 210 give consistent results, as can be shown in report 280. It should be understood that the second MAM-enabled instrument 260 is adjusted and calibrated in the same or similar manner as as described for the first MAM-enabled instrument 210, including using the same reference standard (where, for example, reference standard 202 and reference standard 252 are examples of the same reference standard), so the description in this calibration document (for example, in number through equations (1) and (2)) is equally applicable to both the second MAM-enabled device 260 as it is to the first MAM-enabled device 210.

The second MAM-enabled tool 260 includes a second detector (eg, detector 140 described herein in connection with FIG. 1). In the embodiment shown in FIG. 2, the second MAM-enabled device 260 is characterized by the state of the second device as determined by at least one of (1) the model of the second device or (2) the second set of settings. As described herein, the second device state for the second MAM-enabled device 260 may be different from the first device state for the first MAM-enabled device 210. In some embodiments, the state of the second device may be different from the state of the first device because the second MAM-enabled device 260 may have a second device model that is different from the first device model for MAM-enabled device 210. In other embodiments, the state of the second device may differ from the state of the first device because the second MAM-enabled device 260 may have a second set of settings that is different from the first set of settings for the first MAM-enabled device 210.

The second MAM-enabled instrument 260 is adapted to receive a second sample 254 (eg, a proteolytic peptide sample) and a reference standard 252 (eg, a proteolytic peptide). The reference standard 252 may be another sample of the same reference standard as used for the reference standard 202. The second MAM-enabled instrument 260 is further adapted to detect by a second detector (eg, detector 140 described herein in connection with FIG. 1) an isoform of the second sample (i) in the second sample and the second isoform of the reference standard in the reference standard. It is to be understood that in some embodiments, each of the first sample, the second sample, and the reference sample can be associated with a common proteolytic peptide, so the first sample is the target proteolytic peptide, the second sample is the proteolytic peptide, and the reference standard is the proteolytic peptide.

- 12 039676

The second MAM-enabled device 260 is associated with one or more processors. In some embodiments, one or more processors are part of a single computing device (e.g., computing device 142), so the one or more processors associated with the first MAM-enabled device 210 are the same one or more processors that are associated with the second device 260 with MAM support. In other embodiments, one or more processors are part of different computing devices.

One or more processors associated with the second assisted instrument 260 are adapted to determine, by a second MAM iteration, a second set of correction factors (eg, ai) corresponding to the isoform(s) of the second sample (i). It should be understood that the determination of the set of correction factors (eg, ai) may be performed before, during, or after the second iteration of the MAM such that the determination of the set of correction factors is associated with the second iteration of the MAM. In the embodiment shown in FIG. 2, the second set of correction factors (eg, ai) is based on the reference standard 252, which is the same reference standard as the reference standard 202. One or more processors associated with the second MAM-enabled instrument 260 are further adapted to determine the second set sample score values (eg, Ai) corresponding to the isoform(s) of the second sample, where the second set of sample score values (eg, Ai) is based on the second set of correction factors (eg, ai).

Taking into account the correction factors determined for each of the first MAM-enabled instrument 210 and the second MAM-enabled instrument 260, which are calibrated by common reference standards 202 and 252, as described herein, between the first MAM-based instrument 210 and the second instrument 260 based on MAM, measurement variability can be reduced when the same sample is analyzed by both instruments. Reduced variability can be demonstrated by consistent MAM results in the 280 report. be reduced by using a first set of correction factors (eg, ai) for the first MAM-enabled device 210 and a second set of correction factors (eg, ai) for the second MAM-enabled device 260, as described herein. As described herein, in some cases, the variation value for the first set of score values in the sample and the second set of score values in the sample can be reduced by at least 25 percent.

Thus, as shown with respect to FIG. 2, samples 204 and 254 and a common reference standard (eg, 202 and 252) can be analyzed in parallel to calibrate (220 and 270) the first MAM-enabled instrument 210 and the second MAM-enabled instrument 260, respectively. Thus, a report 280 reporting the concentration of each quality attribute in a sample can be generated using a known concentration in a reference standard (e.g., 202 and 252) to achieve consistent results even with different tools, e.g. tools having different conditions described in this document. For example, each of the first MAM-enabled instrument 210 and the first MAM-enabled instrument 216 may be adapted to generate a report (eg, report 280) containing or otherwise describing or reporting isoforms and/or quality attribute(s). For example, report 280 may contain information and data presented in a diagram and may include data and information such as response factors, peak areas, or other information, such as as described with respect to FIG. 1. In various embodiments, report 280 may be generated after calibration (220 or 270) using a reference standard 202 or 252, as described herein.

Fig. 3 is an illustration of a method 300 for reducing the variability of an MAM-enabled instrument (eg, MAM-enabled instrument 100) over multiple time periods by calibrating signal strength during operation in accordance with the various embodiments disclosed herein. It should be understood that the MAM-enabled device described in FIG. 3 is adjusted and calibrated in the same or similar manner as described for the first MAM capable instrument 210 shown in FIG. 2, including the use of a reference standard as disclosed herein for calibration (eg, including through Equations (1) and (2)) is equally applicable to the MAM-enabled instrument described in FIG. 3. However, with regard to FIG. 3, the same MAM-enabled instrument is used at different time periods (eg, different cycles, different MAM iterations, and/or different days, hours, etc.). The MAM-enabled device described in FIG. 3 may be adapted as described for the MAM-enabled instrument 100 shown in FIG. one.

Method 300 begins (302) at block 304, where the MAM-capable device described in FIG. 3, for a first period of time, may receive a first sample (e.g., sample 204, which may be a proteolytic peptide) and a reference standard (e.g., a reference

- 13 039676 standard 202, which may be a proteolytic peptide).

At block 306, the MAM-enabled instrument via a detector (eg, detector 140) can determine the isoform of the first sample (i) in the first sample and the first isoform of the reference standard in the reference standard within a first period of time.

At block 308, one or more processors (e.g., one or more of the processors described in connection with FIG. 1) may be adapted by a first iteration of the MAM over a first period of time to determine a first set of correction factors (e.g., ai) corresponding to the isoform(s). ) of the first sample (i), where the first set of correction factors (eg ai) is based on the reference standard.

At block 310, one or more processors may also be adapted to determine, through a first iteration of the MAM, and within a first period of time, a first set of score values in a sample (eg, Ai) corresponding to the isoform(s) of the first sample (i). The first set of score values in the sample may be based on the first set of correction factors (eg, a;). The device with MAM support during the first period of time can be characterized by the state of the first device, given by the first set of settings. The state of the first instrument may be the same or similar to that described with respect to FIG. 2.

At block 312, the instrument with support for the second time period may receive a second sample (eg, which may be a proteolytic peptide) and a reference standard (eg, which may be a proteolytic peptide).

At block 314, the MAM-enabled instrument may detect, by means of a detector (eg, detector 140), for a second period of time, the isoform of the second sample (i) in the second sample and the second isoform of the reference standard in the reference standard.

At block 316, one or more processors may be adapted to determine, by a second MAM iteration, over a second time period, a second set of correction factors corresponding to the isoform(s) of the second sample, where the second set of correction factors is based on a reference standard.

At block 318, one or more processors may also be adapted to determine, through a second MAM iteration, and over a second time period, a second set of sample score values corresponding to the isoform(s) of the second sample. The second set of score values in the sample may be based on the second set of correction factors. The device with MAM support during the second period of time can be characterized by the state of the second device, determined by the second set of settings. The state of the second MAM-enabled device during the first time period may be different from the state of the second MAM-enabled device during the second time period.

Considering the correction factors determined for each time period, the measurement variability between MAM iterations during the first time period and the second time period can be reduced. For example, the variation value(s) for the first set of sample score values and the second set of sample values may be reduced by considering the first set of correction factors for the first MAM-enabled instrument and the second set of correction factors for the second MAM-enabled instrument.

The embodiments shown in FIG. 4a and 4b described herein illustrate the analysis of P1 protein using variation reduction and calibration as described herein. P1 contains two O-linked glycosylation sites in one part of the molecule with 6 different glycoforms (isoforms), each of which is shown in FIG. 4b in this document. Diagrams 400 and 450 shown in FIG. 4a and 4b, respectively, were generated by analyzing sample P1 along with its reference standard by different analytical chemists using different LC/MS instruments, including the Q Exactive™ BioPharma Platform instrument and two Exactive™ Plus Orbitrap Mass Spectrometer instruments, each manufactured by the company ThermoFisher Scientific (San Jose, California). All data were processed with Chromeleon™ software provided by ThermoFisher Scientific to determine the peak areas of each peptide of interest. Score values (eg Ai) presented for each glycoform in FIG. 4a and 4b were quantified using iterations (402) of the traditional MAM as well as iterations (404) of the new calibrated MAM. The procedures (404) of the new calibrated MAM are based on calibration techniques to reduce the variability of multi-parameter methods (MAM) between laboratories or instruments by calibrating the intensity of signals during operation, as described in this document. To obtain score values (e.g., A, ⁰ ) of each attribute in the reference standard P1, the reference standard P1 was analyzed through iterations (402) of the traditional MAM six times, and the average of these six measurements was used as the known score in the reference standard.

Fig. 4a is a chart 400 describing the decrease in the variation of the values of the

- 14 039676 metrics (eg Ai) for a given instrument/interlaboratory isoform within six exemplary MAM iterations. For example, the y-axis 410y of chart 400 shows score values (eg, Ai) as a percentage. The x-axis 410x of chart 400 shows six exemplary MAM iterations 1-6 in which traditional sample score values determined by iterations (402) of the traditional MAM are compared with calibrated score values determined using iterations (404) of the new calibrated MAM. , as described for the various embodiments disclosed herein. For example, MAM iterations 1-6 may be MAM iterations conducted on different MAM devices (e.g., the first MAM-enabled device 210 and the second MAM-enabled device 260 described in connection with FIG. 2) or MAM iterations 1-6 may represent MAM iterations conducted at different times on the same MAM-enabled instrument (eg, as described for FIG. 3).

In particular, in the embodiment shown in FIG. 4a, chart 400 shows the measured values of quantitative indicators (eg, Ai) for the glycoform (isoform), which are determined using both iterations (402) of the traditional MAM and iterations (404) of the new calibrated MAM. In particular, the glycoform in diagram 400 represents the 2nd nucleus sialic acid (SA). Iterations 1-3, shown on the x-axis 410x, were measured with the first MAM-enabled instrument, i. the first Exactive™ Plus Orbitrap Mass Spectrometer. Iterations 4-6, shown on the x-axis 410x, were measured with a different, second MAM-enabled instrument, i.e. another Exactive™ Plus Orbitrap Mass Spectrometer. Chart 400 shows that the inter-instrument variability in score values (eg, Ai) within each iteration 1-6 using procedures (404) of the new calibrated MAM is significantly reduced compared to the variability in score values within each iteration 1-6 at using the procedures (402) of the traditional MAM. In other words, the consistency of score values (e.g., Ai) between instruments within each iteration 1-6 using procedures (404) of the new calibrated MAM is significantly improved compared to the variability in score values within each iteration 1-6 using procedures ( 402) traditional MAM.

In FIG. 4b is a graph showing the reduction in instrument/interlaboratory bias for six exemplary isoforms, including the isoform shown in FIG. 4a according to various embodiments disclosed herein. The y-axis 460y of chart 450 shows the relative standard deviation (RSD) of intralaboratory precision as a percentage. RSD correlates with score values (eg Ai). In particular, a decrease in RSD indicates a decrease in the volatility of score values.

The x-axis 460x of plot 400 shows six exemplary isoforms (1HexNAc, 1Core, 1SA, 2SA, 2nd core SA, and non-glycosylated isoform). Analysis for the SA 2nd isoform is shown in FIG. 4a as described herein. In FIG. 4b shows the percentage reduction in Relative Standard Deviation (RSD) (eg, glycoforms) for each of the six isoforms, as evidenced by comparing iterations (402) of the traditional MAM with iterations (404) of the new calibrated MAM. Each of the six isoforms can be used as quality attributes to check the quality control of batches of a given sample (eg, P1). Therefore consistency (i.e. reduced variability/reduced deviation) when measuring the same sample is important. Through diagram 450 in FIG. 4b shows that iterations (404) of the new calibrated MAM significantly reduce the RSD (eg, two to three times), from a maximum of 25 to a maximum of 5%, even when P1 glycoforms are measured on different instruments.

The embodiments shown in FIG. 5a and 5b described herein illustrate the attribute analysis of the second P2 protein. In particular, with respect to diagrams 500 and 550 shown in FIG. 5a and 5b, reference standard P2 was incubated at 40° C. for 4 weeks. Then 20% of this stressed sample was added to the reference standard P2 to obtain a test sample. This test sample, together with both the P2 reference standard and the stressed sample, was digested with trypsin following two different protocols, and each digest product was analyzed in triplicate on two different LC/MS instruments with different columns, mobile phase, and gradient. The Thermo Q Exactive™ BioPharma Platform was used as the first MAM-enabled instrument, and the Orbitrap Fusion™ Lumos™ Tribrid™ Mass Spectrometer (ThermoFisher Scientific) was used as the second MAM-enabled instrument. Reference standard P2 was analyzed on the first MAM-enabled instrument six times, and the average measured score value of each attribute was used as the standard score value. All data and information obtained was analyzed to obtain the peak area corresponding to each peptide isoform and associated score values.

- 15 039676

In addition, with respect to Figures 5a and 5b, 57 quality attributes for P2 were analyzed. For example, the 57 quality attributes include fragmentation, oxidation, glycation, hydroxylation, sequence variants, isomerization, deamination, C-terminal lysine, O-linked glycans, and/or N-linked glycans. These quality attributes covered a wide range of quantitative values from 0.005 to 5%.

In FIG. 5a is a graph 500 describing the decrease in variation in score values (eg, Ai) between instruments/between laboratories over twelve exemplary MAM iterations. For example, the y-axis 510y of chart 500 shows score values (eg, Ai) as a percentage. The x-axis 510x of chart 500 shows twelve exemplary MAM iterations 1-12 in which standard sample scores determined by iterations (402) of the traditional MAM are compared with calibrated score values determined using iterations (404) of the new calibrated MAM. as described for the various embodiments disclosed herein. For example, MAM iterations 1-12 may be MAM iterations performed using different instruments (e.g., the first MAM instrument 210 and the second MAM-enabled instrument 260 as described for FIG. 2), or MAM iterations 1-12 may be MAMs performed using different time periods on the same MAM instrument (as described for example in FIG. 3).

Namely, in the embodiment shown in FIG. 5a, chart 500 shows measured quantitation values (eg, Ai) for K117 hydroxylation as determined using both traditional MAM iterations (402) and new calibrated MAM iterations (404). In particular, chart 500 illustrates the K117 hydroxylation quantitation measured using two cleavage protocols and two LC/MS instruments, in triplicate each. MAM iterations 1-3 and 7-9 refer to split protocol 1, and assay runs 4-6 and 10-12 refer to split protocol 2. MAM iterations 1-6 were performed on a Thermo Q Exactive™ BioPharma Platform instrument and MAM iterations 7-12 were performed on an Orbitrap Fusion™ Lumos™ Tribrid™ Mass Spectrometer instrument. Plot 500 shows that the inter-instrument variability of scores (eg, Ai) across all MAM iterations 1-12 using procedures (404) of the new calibrated MAM is significantly reduced compared to the score variability across MAM iterations 1-12 using procedures (402) traditional MAM. In other words, the consistency of score values (e.g., Ai) between instruments at each iteration of 1-12 MAM using procedures (404) of the new calibrated MAM is significantly improved compared to the variability of score values at each iteration of 1-12 MAM using procedures ( 402) traditional MAM. Thus, in FIG. 5a illustrates the measured K117 hydroxylation quantitations for two different instrument models. Through the conventional MAM procedure (402), different score variations were obtained on two different MAM capable instruments. However, procedures (404) of the new calibrated MAM provided consistent scores on two different MAM capable instruments.

In FIG. 5b is a graph depicting the reduction in inter-instrument/inter-laboratory variation across 57 quality attributes according to the various embodiments disclosed herein. Hydroxylation of K117 shown in FIG. 5a is one of the 57 quality attributes shown in FIG. 5b. The y-axis 560y of chart 550 shows the RSD of intralaboratory precision as a percentage. The x-axis 560x of chart 550 shows 57 quality attributes, among which one such quality attribute is K117 hydroxylation. Each of the 57 quality attributes was determined using the traditional MAM procedure (402) and the new calibrated MAM procedure (404). Diagram 550 in FIG. 5b shows that while the RSD of quality attributes using the traditional MAM procedure (402) was generally in the 350% range, the RSD of quality attributes using the new calibrated MAM procedures (404) was reduced to less than 20% for most of the quality attributes, thus showing the reduction in variability through signal intensity calibrations during operation according to the various embodiments disclosed herein.

This document describes how reducing the variability of MAM-enabled instrument(s) by calibrating signal intensity during operation provides benefits for mass spectrometry (as described for example in FIG. 2). For example, calibration of response factors (eg, ai) as described herein eliminates the requirement for the same response factor across peptide isoforms (k). This methodology can be applied to other types of MAM tools. For example, with respect to traditional methodologies, selected reaction monitoring (SRM) on a triple quadrupole instrument is not suitable for MAM purposes due to potentially different fragmentation efficiencies among peptide isoforms. However, the new calibrated MAM procedure described in this document makes it possible to perform MAM on a three quadrupole instrument, which may be more advantageous than

- 16 039676 the use of an instrument with an orbital trap due to the better accuracy, linearity and dynamic range of a triple quadrupole instrument. However, in such embodiments, it may be necessary to initially set up the instrument with high accuracy for the concentration of each attribute in the reference standard.

Although the disclosure of this document sets forth a detailed description of many different embodiments, it should be understood that the legal scope of the description is determined by the terms of the claims set forth at the end of this patent, and equivalents. The detailed description is to be interpreted as being by way of example only, and does not describe every possible embodiment, since a description of every possible embodiment would be impractical. Numerous alternative embodiments can be implemented using current technology or technology developed after the filing date of this patent, which still fall within the scope of the claims.

The following additional considerations apply to the discussion presented earlier. Throughout this description, multiple examples may implement the components, process steps, or structures described in a single instance. While the individual operations in one or more of the methods are illustrated and described as separate operations, one or more of the individual operations may be performed simultaneously and there is no requirement that the operations be performed in the order presented. Structures and functionality presented as separate components in the exemplary configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component can be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter set forth herein.

Further certain embodiments are described herein as comprising logical elements or a set of procedures, sub-procedures, applications, or commands. They may be either software (eg, code embodied on a computer-readable medium or in a transmission signal) or hardware. In hardware procedures, etc. are concrete blocks that can perform certain operations and can be adapted or arranged in a certain order. In exemplary embodiments, one or more computer systems (e.g., a stand-alone, client, or server computer system) or one or more computer system hardware modules (e.g., a processor or group of processors) may be adapted by software (e.g., an application or part of an application) as a piece of hardware that works to perform certain operations as described in this document.

The various operations of the exemplary methods described herein may be performed at least in part by one or more processors that are temporarily adapted (eg, via software) or permanently adapted to perform the respective operations. Whether temporarily or permanently adapted, such processors may comprise processor-implemented modules that operate to perform one or more operations or functions. The modules referenced herein may, in some exemplary embodiments, include processor-implemented modules.

Likewise, the methods or procedures described herein may be at least partially implemented by processors. For example, at least some of the operations of the method may be performed by one or more processors or software modules implemented by the processors. The execution of certain operations can be distributed among one or more processors located not only in one machine, but located in a number of machines. In some exemplary embodiments, the processor or processors may be located in a single location, while in other embodiments, the processors may be distributed over a number of locations.

The execution of certain operations can be distributed among one or more processors located not only in one machine, but located in a number of machines. In some exemplary embodiments, one or more processors or processor-implemented modules may be located in the same geographic location (eg, in a home environment, an office environment, or a server park). In other embodiments, one or more processors or processor-implemented modules may be distributed over a variety of geographic locations.

This detailed description is to be interpreted as exemplary only, and does not describe every possible embodiment, since a description of every possible embodiment would be impractical, if not impossible. One of ordinary skill in the art can implement many alternative embodiments using either current technology or technology developed after the filing date of this

- 17 039676 applications.

Those skilled in the art will appreciate that numerous modifications, changes, and combinations can be made to the embodiments described above without departing from the scope of the present invention, and that such modifications, changes, and combinations are to be considered as falling within the scope of the present invention.

The claims at the end of this patent application are not intended to be limited under 35 U.S.C. § 112(f) unless conventional language for describing means and functions is explicitly cited, such as the terms means for or step for are clearly cited in the claim(s) . The systems and methods described herein aim to improve computer functionality and improve the performance of standard computers.

Aspects

The following aspects of the present invention are illustrative only and are not intended to limit the scope of the present invention.

1. A calibration system adapted to reduce the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating signal intensity during operation, wherein the calibration system comprises a first MAM-enabled instrument containing a first detector, wherein the first MAM-enabled instrument is characterized state of the first instrument as defined by at least one of (1) the model of the first instrument or (2) the first set of settings, wherein the first MAM-enabled instrument is adapted to receive the first sample and reference standard, and the first MAM-based instrument is further adapted to detect by means of a first isoform detector of the first sample in the first sample and the first isoform of the reference standard in the reference standard; one or more processors associated with the first MAM-enabled device, wherein the one or more processors associated with the first MAM-enabled device are adapted to determine, by a first MAM iteration, a first set of correction factors corresponding to the isoform(s) of the first sample, where the first the set of correction factors is based on the reference standard, and wherein the one or more processors associated with the first MAM-enabled instrument are further adapted to determine the first set of score values in the sample corresponding to the isoform(s) of the first sample, where the first set of score values in the sample is based on the first set of correction factors; a second MAM-based device comprising a second detector, wherein the second MAM-enabled device is characterized by a state of the second device defined by at least one of (1) a model of the second device or (2) a second set of settings, wherein the state of the second device is different from the state a first instrument, wherein the second MAM-enabled instrument is adapted to receive the second sample and the reference standard, and the second MAM-enabled instrument is further adapted to detect, by means of the second detector, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard; and one or more processors associated with the second MAM-enabled device, wherein the one or more processors associated with the second MAM-enabled device are adapted to determine, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors coefficients is based on a reference standard, and wherein one or more processors associated with the second MAM-enabled instrument are further adapted to determine a second set of score values in the sample corresponding to an isoform of the second sample, where the second set of score values in the sample is based on the second set of correction factors, and wherein the variation value for the first set of score values in the sample and the second set of score values in the sample is reduced by taking into account the first set of correction factors and the second set of correction factors patients.

2. The calibration system according to aspect 1, where one or more processors associated with the first device for MS, through the first iteration of the MAM determine the quality attribute.

3. The calibration system according to aspect 2, where the quality attribute is any of the isoform of the first sample, protein, or identified impurity.

4. A sizing system according to any of Aspect 2 or Aspect 3, wherein the quality attribute is determined by any one or more of fragmentation, oxidation, glycation, hydroxylation, sequence variants, isomerization, deamination, C-terminal lysine, O-linked glycans, and/ or N-linked glycans.

5. A calibration system according to any of the previous aspects, wherein one or more processors associated with the first MS instrument are adapted to generate a report containing a quality attribute.

6. The calibration system according to any of the previous aspects, where the model of the first device is different from the model of the second device.

- 18 039676

7. The calibration system according to any of the previous aspects, where the first set of settings is different from the second set of settings.

8. The calibration system according to any of the previous aspects, wherein the first set of correction factors is based on the ion intensity value of the first isoform of the reference standard and the first value of the quantitative index in the reference standard of the first isoform of the reference standard.

9. A calibration system according to any of the preceding aspects, wherein the first set of correction factors calibrates the response factor associated with the first set of score values in the sample to determine the isoform ion intensities of the first sample.

10. The calibration system according to any of the previous aspects, wherein the first set of score values in the sample is further based on the ion intensity value of the isoform of the first sample.

11. A calibration system according to any of the previous aspects, wherein the first MAM-enabled instrument is a mass spectrometry (MS) instrument.

12. A calibration system according to any of the previous aspects, wherein the first MAM-enabled instrument is a tri-quadrupole instrument.

13. A calibration system according to any of the previous aspects, wherein one or more processors associated with the first MS device are communicatively connected over a computer network to one or more processors associated with the second MS device.

14. A calibration system according to any of the preceding aspects, wherein the one or more processors associated with the first MAM capable device is one or more processors associated with the second MAM capable device.

15. A calibration system according to any of the preceding aspects, wherein the first MAM-enabled instrument is located in a first laboratory at a first geographic location and the second MAM-enabled instrument is located in a second laboratory at a second geographic location.

16. Calibration method for reducing the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration method includes receiving in the first device with MAM support containing the first detector, the first sample and the reference standard; detecting, by means of the first detector, the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard; determining, by one or more processors associated with the first MAM-enabled instrument, during the first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard; determining, by means of one or more processors associated with the first MAM-enabled instrument, a first set of score values in the sample corresponding to an isoform of the first sample, where the first set of score values in the sample is based on the first set of correction factors, and wherein the first instrument is capable of MAM is characterized by the state of the first device, defined by at least one of: (1) the first model of the device or (2) the first set of settings; receiving, in a second MAM-enabled instrument containing a second detector, a second sample and a reference standard; detecting, by means of the second detector, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard; determining, by means of one or more processors associated with a second MAM-enabled instrument, during a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard; determining, by means of one or more processors associated with a second MAM-enabled instrument, a second set of score values in the sample corresponding to an isoform of the second sample, where the second set of score values in the sample is based on the second set of correction factors, and the second set of scores in the sample is MAM is characterized by a state of the second instrument defined by at least one of: (1) the model of the second instrument or (2) the second set of settings, where the state of the second instrument differs from the state of the first instrument, and where the variation value for the first set of score values in the sample and of the second set of score values in the sample is reduced by taking into account the first set of correction factors and the second set of correction factors.

17. Calibration method for reducing the variability of the MAM-supported instrument over several periods of time by calibrating the intensity of the signals during operation, while this calibration method provides for the reception in the MAM-supported instrument during the first period of time of the first sample and the reference standard; detecting, by means of the detector of the MAM-enabled instrument, within a first period of time, the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard; determination by one or more processors during the first MAM iteration during the first time period

- 19 039676 a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard; determining, by one or more processors, by means of a first MAM iteration during a first time period, a first set of score values in the sample corresponding to an isoform of the first sample, where the first set of score values in the sample is based on the first set of correction factors, and where the MAM capable instrument in during the first period of time is characterized by the state of the first device, defined by the first set of settings; reception in the device with MAM support during the second period of time of the second sample and reference standard; detecting, by the detector of the MAM-enabled instrument, within a second time period, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard; determining by one or more processors during the second MAM iteration during the second time period a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard; and determining, by one or more processors, through a second MAM iteration over a second time period, a second set of sample score values corresponding to an isoform of the second sample, wherein the second set of sample score values is based on the second set of correction factors, and where the MAM capable instrument during a second period of time is characterized by the state of the second instrument defined by the second set of settings, where the state of the second instrument differs from the state of the first instrument, and where the value of variation of the first set of sample score values and the second set of sample score values is reduced taking into account the first set of correction coefficients and a second set of correction factors.

18. A calibration system according to aspect 17, wherein one or more processors determine a quality attribute.

19. The calibration system according to aspect 18, where the quality attribute is any of a first sample isoform, a second sample isoform, a protein, or an identified impurity.

20. A calibration system according to any one of aspects 17-19, wherein one or more processors are adapted to generate a report containing a quality attribute.

21. A calibration system according to aspect 1, wherein the variation value for the first set of score values in the sample and the second set of score values in the sample is reduced by at least 25%.

22. The calibration method according to aspect 16, wherein the variation value for the first set of sample score values and the second set of sample score values is reduced by at least 25%.

23. The calibration method according to aspect 17, wherein the variation value for the first set of score values in the sample and the second set of score values in the sample is reduced by at least 25%.

24. The calibration system according to aspect 1, wherein the first sample is a proteolytic peptide, the second sample is a proteolytic peptide, and the reference standard is a proteolytic peptide.

25. The calibration method according to aspect 16, wherein the first sample is a proteolytic peptide, the second sample is a proteolytic peptide, and the reference standard is a proteolytic peptide.

26. The calibration method according to aspect 17, wherein the first sample is a proteolytic peptide, the second sample is a proteolytic peptide, and the reference standard is a proteolytic peptide.

Additional description

As described herein, new calibration systems and methods are provided for quantifying each attribute using the known quantitation of each attribute in a reference standard as the calibrant. Most of the quality attributes in a reference standard remain constant over the entire period of time until the standard degrades and can therefore serve as a calibrant to correct for differences between instruments or sample preparation procedures. Since the reference standard data is usually obtained by the typical MAM method for the purpose of identifying and determining the suitability of the system, no additional work is required by the analytical chemist. With this methodology, instrument model matching is no longer required. At the same time, minor changes in the digestion procedure from laboratory to laboratory, as well as changes due to automation, will not significantly affect the result of the study.

Attribute classification

Depending on whether the attribute score can change during sample preparation, for the purposes of MAM, each attribute is assigned to one of three types, including type 1 attributes, type 2 attributes and type 3 attributes.

- 20 039676

Type 1 attributes are not changed during sample preparation. Examples include sequence variants, hydroxylation of lysine and proline, and so on. The bulk of glycosylation can be attributed to this group when samples are not exposed to extremely low pH during sample preparation.

Type 2 attribute scores may decrease during sample preparation. Examples include phosphorylation due to the instability of this modification. Because reactive (eg, loss of phosphorylation) substrates for type 2 attributes are usually minor components (eg, phosphorylated peptide), the absolute change in type 2 attributes during sample preparation is negligible. A special attribute of type 2 is the residual N-terminal glutamine. When the N-terminus of a protein is represented by a glutamine residue, it usually cyclizes to form pyroglutamic acid as the main component. N-terminal glutamine, which is a minor component, is considered in this work as a modified form of pyroglutamic acid. N-terminal glutamine is a type 2 attribute because cyclization can occur during sample preparation with a decrease in the amount of free N-terminal glutamine.

The relative content of type 3 attributes may increase during sample preparation. Type 3 attributes include oxidation, deamination, aspartic acid isomerization, fragmentation, etc. Since the reaction substrates for type 3 attributes are basic unmodified peptides, their absolute change during sample preparation can be quite high.

As will be discussed further, type 1 and type 2 attributes can often be measured with high accuracy and low limit of quantification because the difference in results is due solely and primarily to LC-MS measurements. Type 3 attributes, on the other hand, have lower accuracy and higher limit of quantitation because differences in sample preparation contribute to different final results.

Notation

The notation used herein can be summarized as follows: A represents the isoform score as a fraction (or percentage). I represents the measured ion intensity (area under the peak in the chromatogram in selective ion detection mode) of the isoform. k represents the response factor, and a represents the correction factor for the response factor. The superscript 0 represents the reference standard (RS). For example, A ⁰ represents the known isoform score in RS and I ⁰ represents the measured ionic intensity of the isoform in RS. The subscripts 0, 1, 2, ... i, ... n represent the n+1 isoforms that are associated with the given residue (including the unmodified form). The subscript 0 represents the most common isoform (usually the unmodified form). For example, Ai represents the score of isoform i, Ii represents the ion intensity of isoform i, ki represents the response factor for isoform i, and ai represents the correction factor for the response factor for isoform i. The value of a ₀ is set to 1. The subscript 1 may be omitted from certain equations when there is only one isoform other than the main isoform (n=1). It should be noted that when the isoforms i=0, 1, ... n are mentioned in this report, they are usually associated with a single residue, whether explicitly stated or not. For example, there may be three isoforms associated with the Asn residue in the peptide, including the unmodified Asn, its deamidated form, and its succinimidated form. An oxidized Met residue on the same peptide, however, may belong to a different set of isoforms (assuming, for example, that these two modifications are random and therefore independent of each other).

Traditional MAM

In the traditional MAM method, each attribute (e.g., different states of modification of an amino acid residue in a peptide) is quantified based on the MS responses (e.g., area under the peak in a selective ion detection chromatogram) of the modified peptide and the unmodified peptide, with the following standard assumptions:

1) In each sample, all isoforms associated with an amino acid residue have the same response coefficient (ratios of all response coefficients = 1)

2) Artificial characterization changes are negligible Assuming that an amino acid residue has n+1 modification states (0, 1, ..., n), there are n+1 peptide isoforms associated with the residue of interest. The most common isoform, usually unmodified, is designated as i=0. The score of each isoform is calculated based on the following equation (3):

kL _: = 1. ,i = 0, n

In equation (3) above, k represents the response coefficient for all isoforms of the peptide

- 21 039676 (for example, a constant value based on the above assumption 1), and Ai and Ii represent the score and MS intensity of isoform i, respectively. The response factor k represents a combination of both peptide recovery and MS response of each isoform.

Solving equation (3) gives equation (4) i

*=Σύ /=0

Equation (4) states that the score for each isoform is calculated by dividing the MS intensity of the isoform by the sum of the MS intensity values of all isoforms.

Due to the assumptions described above, the traditional MAM method must satisfy the following requirements:

1) The splitting efficiency must be reproducible.

2) Instrument states for MS must be the same.

3) Sample preparation should introduce a minimum number of artificial modifications.

However, in reality, the above requirements are difficult to satisfy, at least due to the following reasons:

1) Peptide recovery may vary due to differences in sample preparation procedure, skill of analytical chemists, quality of devices and reagents, etc.

2) Due to differences in MS instrument model, instrument settings, and instrument maintenance patterns, response factors for different peptide isoforms may vary.

3) The amount of artificial modification may vary due to differences in the sample preparation procedure, the skill of the analytical chemist, the quality of the devices and reagents, and the condition of the instruments.

Due to the above difficulties posed by the requirements of the traditional MAM, a new methodology is needed to overcome these shortcomings in order to ensure the long-term success of the MAM.

Calibration of response factors using a reference standard

At MAM, the reference standard is typically analyzed in parallel with the samples to verify system suitability or authenticity. Because most of the quality attributes in a reference standard remain constant throughout the lifetime of the standard, the reference standard can serve as a calibrant to correct for differences between instruments or sample preparation procedures. The scoring of each attribute in the reference standard can be set either using the traditional MAM or the orthogonal method with better accuracy. If an attribute in a reference standard is quantified by an analytical method with an absolute quantification, all subsequent MAM analyzes of the same attribute also become an absolute measurement after the response factor is calibrated.

By using a reference standard as a calibrant in each assay run, most of the requirements for conventional MAM in terms of inter-laboratory and inter-instrument reproducibility can be eliminated. Then, in such embodiments, the conventional two assumptions described herein above take the following form, i.e. corrected assumptions:

1) Within the same LC-MS sequence, the response coefficient ratios among all isoforms remain constant.

2) Man-made attribute changes are minor. It should be noted that in the first revised assumption, instead of requiring that all peptide isoforms have the same response factor (ratios of all response factors = 1) over the life of the product, it only requires that the ratios of response factors be constant within the same LC-MS sequences (per analytical chemist, per instrument, and within one day).

The new revised assumptions remove the assumption that each isoform has the same response factor. Instead, it is assumed that the response factors for isoform i are modified by the same correction factor ai in both the reference standard and the sample. Therefore, the response factor for isoform i in the reference standard is expressed as aik ⁰ (where the superscript 0 denotes the reference standard) and the response factor for isoform i in the sample is expressed as aik. As a result of the analysis of both the reference standard and the sample, equation (3) turns into equation (5), which is the same as equation (2) described in this document

- 22 039676 'a^A = l,i = O,na _i kA _i =Ι _τ Α = ^,η = ¹ (5) η

ΣΛ=1

U=o

In some embodiments, the response rate for the reference standard and sample may differ due to slight differences in sample preparation, differences in instrument sensitivity, and so on. In addition, in order for the equations to be solvable, the coefficient a for one of the isoforms is usually brought to a constant value. It doesn't matter which isoform leads to the constant a. It is expedient to bring to a constant the most common isoform, which is usually the unmodified form (a0=1).

Solving equation (5) gives equation (6), which is the same as equation (1) described in this document:

f A ⁰ 4° . L

Equation (6) can be used to calculate a correction factor for each isoform (ai) given the known attribute score for the reference standard (Ai ⁰ ) and calculate the score for each isoform in the sample (Ai). In order for the parameters ai to have uniquely defined values, the denominators should not be close to zero. Therefore, in order to successfully calibrate the response factor, each attribute in the reference standard must be quantified sufficiently high (A _i ⁰ ) to be accurately quantified.

Calibration of artificial modifications using a reference standard

Some attributes, including type 2 attributes such as phosphorylation and type 3 attributes such as oxidation, deamidation, isomerization of aspartic acid residues, and fragmentation, can change their own quantification during sample preparation or due to loss of modification due to their instability ( type 2), or due to the artificial formation of these modifications (type 3). These artificial changes in attributes result in variability between results between analytical chemists and between days and can be adjusted using the reference standard as a calibrant, assuming that the degree of these artificial changes is consistent in the same LC-MS sequence.

In embodiments where the coefficient bi is used to indicate the degree of artificial change for each isoform, after analysis of both the reference standard and the sample, equation (3) becomes equation (7) k[ Α _έ + ] = I _j , i = 0 , n(J)

Σ1='l AO

In equation (7) S _i represents the substrate for artificial modification, which gives the i isoform. For example, for oxidation and deamidation (type 3 attribute), the substrate is an unmodified peptide. For phosphorylation (attribute type 2), however, the substrate is a modified (phosphorylated) peptide due to the possible instability of the modification.

Depending on the nature of the substrate, equation (7) can become complex. For example, using equation (7) for multiple modifications at the same residue, such as N glycosylation at asparagine residues, can lead to a complex version of equation (7).

For residues with a single modification, equation (7) can be reduced to equation (8) and/or equation (9) as described herein, depending on whether the substrate is a modified peptide (type 2) or an unmodified peptide (3 type of). As shown in the embodiments below, for equations (8) and (9), the parts of the equation on the right are solutions of the parts of the equation on the left.

In particular, for type 2 attributes and in relation to equation (8) [ _Г (1-4-М ¹ ') = 4 one

-L__1

A°^+Ι ^ΰ ) I

- 23 039676

C+J)! ⁰

For type 3 attributes and in relation to equation (9)

Ar°[1-A° -0(1-A ⁰ )] = Ζθ ^ ⁰ [A ⁰ +6(1-I ⁰ »7 ⁰ k[A + b(l - A)] = /

(9)

--b \-b

In equation (8), if the amount of artificial modification is significantly less than the ^modified form (i.e. b << 1), which is usually true for most type 2 attributes, equation (8) can be viewed as equation (10) | (1+W°L° -7° (10)

W-A) = C\^ + b)kA^1

Equation (10) is similar to equation (5) when ai=1+b. That is, in such embodiments, the loss of modification during splitting can be modeled by calibrating the response factor, and equations (8) and (5) will give a very similar result.

For parameter b and attribute score A to have uniquely defined values, the denominators must not be close to zero. Therefore, to use equation (8), the attribute score in the reference standard (A ⁰ ) must be much greater than 0. To use equation (9), A ⁰ must be much less than 1 (100%) and b must not be close to 1 In addition, for Equation (9) to be generally valid, the value of b must be much smaller than the uncalibrated attribute score (b << I/(I ₀ +I)). Otherwise, the calibrated quantity A will be close to zero and sometimes negative.

Calibration of both response factors and artificial modifications using two different standards

The reference standard can be used to adjust any of the response factor (a) (ie a-calibration) and/or artificial modification (b) (ie b-calibration). An additional standard is needed to correct each of a and b. To obtain another standard, the reference standard or another sample can be stressed to create another standard containing a higher level of attributes of interest, and then both standards are analyzed along with the samples. The known attribute score for the two standards and their determined MS responses can be used to adjust each of a and b.

A superscript of zero can be used to represent the reference standard, and a superscript of 1 can be used to represent the stressed standard, when considering both standards and sample. Equations (11) and (12) illustrate the characteristic equations depending on the substrate for artificial modification. Note that equations (11) and (12) refer to residues with a single modification.

For type 2 attributes and in relation to equation (11)

[H1- -BA °] - 7 ° ak ^ A ^ + LA °] \u003d 7 ° to ¹ [] - A ¹ - LA ¹ ] \u003d 7d ak ^ + LA ¹ ] ^ ¹ k [1-A-LA] \u003d 7 _d ak [A + BA] - 7

For type 3 attributes and in relation to equation (12)

7^7^-7^ ¹

A°(a7 _d +7° /I +7°) /Η+Ο

A ¹ (I)

- 24 ⁰³⁹⁶⁷⁶

G[1= ak ^l [A' +b(\- A ^l )\ = I ^x k\\-A-S\-A)\ = 1 ₀ ak[A + b(1-A}\ = 1 ( 12) / ₀ °/'(l - 7') -/°/θ(1 - 7°)

As shown above for equations (11) and/or (12), in order for the parameters a and b to have uniquely defined values, the denominator should not be close to zero. Therefore, equation (11) requires I ⁰ 0I ¹ A ⁰ >>I10I ⁰ A ¹ , and equation (12) requires A ¹ >> A ⁰ and A ⁰ << 1. Like equation (9), for equation (12) to have meaning, the value of b must be much less than the attribute score without b-calibration (b << I/(aI ₀ +I)), otherwise the calibrated score A will be zero and sometimes a negative value.

An example of a multivariable assay for an anti-streptavidin IgG2 antibody

As described herein, various exemplary embodiments have been put into practice through specific applications of the calibration systems and methods of the present invention. However, it should be understood that the calibration systems and methods of the present invention are not limited to particular applications. For example, recombinant anti-streptavidin IgG2 has been expressed in the Chinese Hamster Ovary (CHO) cell line. The material representing IgG2 to streptavidin was used as a reference standard. To prepare test samples for analysis, the reference standard was incubated at about 40°C for about 30 days to obtain sample 1, and then samples 2 and 3 were obtained by mixing sample 1 with the reference standard in different ratios (see table. 2).

The IgG2 reference standard and test samples (-120 μg each) were digested with trypsin using the following procedure. First, each sample was treated with 8 mM dithiothreitol at approximately 25°C for approximately 30 min under the conditions of a denaturing solution containing 6.5 M guanidine hydrochloride (Macron Fine Chemicals, Stroudsburg, PA) and 0.2 M Tris (TEKnova, Hollister, CA ) at pH 7.5 to restore disulfide bonds. The reduced IgG2 was then alkylated with 14 mM iodoacetic acid at about 25° for about 20 minutes in the absence of light. The alkylation reaction was quenched using 6 mM DTT.

In order to deliberately create certain differences in the sample preparation procedure, each reduced/alkylated sample was digested with trypsin using two different techniques. According to the first method, the reduced/alkylated sample (IgG2 concentration -1.2 mg/mL) was replaced with a digestion buffer containing about 0.1 M Tris and about 50 mM Methionine (pH 7.5) using a Bio-Spin® 6 column from Bio-Rad (Hercules, CA), for example, following the manufacturer's recommended procedure. After buffer exchange, an appropriate amount of trypsin was added to achieve an enzyme:substrate ratio of about 1:12, followed by incubation at about 37° C. for about 60 minutes. Cleavage was quenched using an equal volume of approximately 0.25 M acetate buffer (pH 4.8) in 8 M guanidine hydrochloride. The final concentration of IgG2 in the cleavage product was -0.5 mg/ml.

According to the second method, each reduced/alkylated sample was processed with the same digestion buffer using a Microcon-30 kDa filter (Millipore Sigma, Burlington, Massachusetts). First, each reduced/alkylated sample was centrifuged at about 14,000 g and the filtrate was discarded. In certain embodiments, the process was repeated three more times after adding 250 µl of digestion buffer each time to the filter. Trypsin digestion was performed on the same filter by adding 140 μl of digestion buffer and 10 μg of trypsin (at 1 mg/ml), followed by incubation at about 37° C. for about 60 minutes. After digestion, an equal volume of quench solution was added to the filter and centrifuged at approximately 14,000 g to collect the peptides in a new collection tube. The final concentration of IgG2 in the cleavage product according to the above procedure is -0.4 mg/ml.

In table. 2 shows anti-streptavidin IgG2 samples used in the examples.

- 25 039676

table 2

Sample name Description Reference standard IgG2 to streptavidin (30 mg/ml) Sample 1 Streptavidin IgG2 stressed at 40°C for 30 days (also used as ^{2'0 and} calibration standard for both a and b) Sample 2 10% sample 1+90% reference standard Sample 3 20% sample 1+80% reference standard

In anti-streptavidin IgG2 and/or similar embodiments, each cleavage product was analyzed by each of three LC-MS/MS systems consisting of an HPLC system from Agilent (Santa Clara, Calif.) connected to a mass spectrometer, e.g. Thermo Scientific Q Exactive Plus Biopharma or Orbitrap Fusion Lumos mass spectrometer. Additional or alternative systems have also been used, as the description in this document is not limited to any type of system, mass spectrometer or otherwise. In some embodiments, two different LC techniques were used to intentionally introduce a certain difference in liquid chromatography conditions (see Table 3).

For the first LC technique (e.g., systems A and B in Table 3), peptides were eluted on an Acquity peptide CSH column (150 x 2.1 mm, particle size 1.7 µm, pore size 170 A) from Waters (Milford, Massachusetts). ) at a flow rate of approximately 0.3 ml/min while maintaining the column temperature at approximately 60°C. Mobile phase A was 0.1% formic acid in water and mobile phase B was 0.1% formic acid in acetonitrile. After an initial hold at about 0.5% B for about 5 minutes, mobile phase B is typically ramped up to about 35% over about 40 minutes. Washing of the column was performed by increasing the amount of mobile phase B to about 99% for about 4 minutes with a retention time of about 1 minute. The column was equilibrated using about 0.5% B for about 15 minutes. For the second LC method (e.g. system C in Table 3), the peptides were eluted on a Waters Acquity BEN C18 column (2.1 x 150 mm, particle size 1.7 µm) at a flow rate of about 0.3 ml/min while maintaining column temperature at approximately 60°C. Mobile phase A was 0.1% formic acid and approximately 0.02% trifluoroacetic acid (TFA) in water, and mobile phase B was 0.1% formic acid and 0.02% TFA in acetonitrile. After an initial hold at about 0.5% B for about 5 minutes, the amount of phase B is typically ramped up to about 40% over about 40 minutes. Washing of the column was performed by increasing the amount of phase B to about 99% for about 4 minutes with a retention time of about 1 minute. The column was equilibrated using about 0.5% B for about 15 minutes.

The HPLC systems were directly connected to a mass spectrometer, such as either a Thermo Scientific Q Exactive Plus Biopharm Mass Spectrometer (eg, System A) or a Thermo Scientific Orbitrap Fusion Lumos Mass Spectrometer (eg, Systems B and C), via an electrospray interface. Although the Thermo Scientific Q Exactive Plus Biopharm mass spectrometer and/or the Thermo Scientific Orbitrap Fusion Lumos mass spectrometer are described in the embodiments of this document, it should be understood that a similar mass spectrometer can be used in accordance with the embodiments disclosed herein. (s). Q Exactive Plus Biopharma was configured to perform MS in full scan mode with a resolution of approximately 70,000 and AGC=1x10 ⁶ followed by five dependent MS/MS scans (high energy collision activated dissociation (HCD), normalized collision energy=27 ) for most common ions. In the case of Fusion Lumos, full scan MS data was collected at approximately 60,000 resolution and AGC=4x10 ⁵ followed by high speed dependent ion trap MS/MS scans (normalized collision energy CID=30). Instrument control and data collection was performed using analytical software such as Thermo Scientific Xcalibur software. Approximately 3 to 4 μg of each tryptic digest product was administered for analysis. Of course, according to the embodiments disclosed herein, software similar to Xcalibur software can be used.

The data obtained by LC-MS/MS was processed by the analytical software, MassAnalyzer, which is supplied by Thermo Scientific under the name BiopharmaFinder™). Analytical software, such as MassAnalyzer, is capable of

- 26 039676 Completely automated characterization, retention time alignment, peptide identification and attribute quantification. Analytical software, such as MassAnalyzer, may rely on comparison of experimental MS/MS data with exactly predicted theoretical MS/MS data to identify peptides. The fitted window function was used to extract the chromatogram in selective ion detection mode to maximize the signal to noise ratio in the chromatogram. As a final result, analysis software such as MassAnalyzer may generate a list of identified variants. It should be understood that, according to the embodiments disclosed herein, analytical software similar to MassAnalyzer may be used.

An example of a multivariate fusion protein analysis

A recombinant fusion protein (e.g., a first protein or protein fragment recombinantly fused to at least a second protein or fragment thereof, often via protein/fragment linkers) expressed in CHO cells may contain multiple glycosylation sites, such as threonine residues and serine. These O-glycans can make the greatest contribution to the heterogeneity of the molecule. In this example, a multi-parameter method was used to quantify these glycoforms in the fusion protein.

For proteolytic digestion, the fusion protein was first denatured in a solution containing about 7.5 M guanidine HCl, about 250 mM Tris (pH 7.5) and about 2 mM EDTA at a protein concentration of about 1 mg/mL. Prior to proteolytic digestion, approximately 2 μl of a 500 mM DTT solution was added to approximately 100 μl of the denatured protein solution, after which it was incubated at approximately 25° C. for approximately 30 minutes to restore disulfide bonds. Approximately 4 μl of 500 mM sodium iodoacetate was then added, followed by incubation at approximately 25° C. for approximately 20 minutes to alkylate the cysteine side chains. Following a buffer exchange procedure with a 100 mM Tris, 50 mM methionine, pH 7.5 solution via a Bio-Rad Bio-Spin desalting column, each 50 µg of the desalted sample was digested using about 5 µg trypsin at about 37°C for about 30 min. . To quench the digestion, approximately 2% formic acid was added to each digestion product to achieve a final acid concentration of approximately 0.2%.

Each tryptic digest (~3 μg) was analyzed using an LC-MS system consisting of an Agilent HPLC directly connected to a mass spectrometer, such as a Thermo Scientific Exactive plus or Q Exactive plus high-resolution mass spectrometer (see Table 3). ). The peptides were eluted on an Agilent Zorbax C18 RR HD column (2.1 x 150 mm, particle size 1.8 μm, pore size 300 A) at a flow rate of approximately 0.2 ml/min while maintaining the column temperature at 50°C. Mobile phase A was about 0.1% formic acid in water and mobile phase B was 0.1% formic acid in acetonitrile.

Starting at about 1.0%, mobile phase B increased linearly to about 40% after about 70 minutes and to about 90% by about 76 minutes. After washing at about 90% for about 5 minutes, the column was equilibrated using about 1% B for about 11 minutes. Data obtained by running MS in full scan mode was collected at a resolution of 140,000 and automatic trap fill control (AGC) was set to 1x10. Instrument control and data collection was performed, for example, using Thermo Scientific Chromeleon software.

Data was analyzed, for example, with Thermo Scientific Pinpoint and Chromeleon software to identify and relative quantify post-translational modifications. The list of O-glycopeptides was characterized by MS/MS using HCD and electron transfer dissociation (ETD), with six glycoforms typically detected. Pinpoint can generate a workbook with accurate mass and isotope distributions for the six forms of O-glycopeptides. The work log was imported into the Chromeleon software, where the MS1 precursor ions were analyzed for retention time, exact mass as well as isotope distribution. A chromatogram in selective ion detection mode was generated in Chromeleon and each peak was integrated for peak area. The integration of each peak was checked manually to ensure accuracy.

In table. 3 shows exemplary LC-MS systems and measurements described.

- 27 039676

Table 3

System HPLC Column for HPLC Method* HPLC MS a IgG2 to streptavidin Q Exactive A Plus Agilent 1290 Agilent waters peptide CSH Additive: 0.1% formic acid Gradient: 0.5% to 35% AcN per Biopharma Fusion AT at 60°C waters peptide CSH 40 min Flow rate: 0.3 ml/min Additive: 0.1% formic acid Gradient: 0.5% to 35% AcN per Lumos Fusion 1290 Agilent at 60°C Waters BEH 40 min Flow rate: 0.3 ml/min Additive: 0.1% formic acid+0.02% TFA FROM Lumos 1200 C18 at 60°C Gradient: 0.5% to 40% AcN per Protein fused to Fc Q Exactive D Plus Biopharma Agilent 1290 Agilent Zorbax C18 40 min Flow rate: 0.3 ml/min Additive: 0.1% formic Exactive E Plus(unit one) Exactive Agilent 1290 Agilent RR HD (2.1x150 mm, particle size 1.8 µm, 300 A) acid Gradient: 1% to 40% AcN in 70 min Flow rate: 0.2 ml/min F Plus (block 2) 1290 at 50°C

Response Calibration Example

Response calibration was performed using analytical software such as Excel® from Microsoft® (Redmond, WA) adapted to implement one or more of the equations described herein. Anti-streptavidin antibody data were processed with analytical software such as MassAnalyzer. Since the MassAnalyzer directly outputs an uncalibrated score for each attribute, and these uncalibrated scores are proportional to the corresponding MS intensities (peak areas), they were treated as MS intensities in the respective calculations. Fc fusion protein data were processed using the Thermo Scientific Chromeleon program, in which the peak area of each characteristic was determined and used as MS intensities for the respective calculations. Standard practice is that a sample of the reference standard is analyzed at least twice (to cover samples). The mean MS intensity for two assays with a reference standard was used to express practical performance in the calculations.

An example of measuring a large number of attributes in IgG2 to streptavidin

In this example, a comprehensive evaluation of the operation of the systems and methods of calibration, which are disclosed in this document, was demonstrated. The reference standard, which is IgG2 to streptavidin, was incubated at approximately 40°C for approximately four weeks. Then about 10% and about 20% of this stressed sample was added to the reference standard, respectively, to create two or more test samples (see, for example, table. 2). Such test samples, together with both the reference standard and the stressed

- 28 039676 total exposure to the sample was trypsin digested in triplicate using two different protocols, with each digestion product through three different LC-MS systems (see e.g. A, B and C shown in Table 3) with different columns, mobile phases, gradients and mass spectrometers. In addition, each of the reference standard and the stressed sample was analyzed by System A six times, and the average measured score of each attribute was used as a reference score. All data were processed with analytical software such as MassAnalyzer to obtain an uncalibrated score for each peptide isoform. These scores were used as MS intensity values for subsequent calibration.

Because systems A and B use the same chromatographic state, data from these runs (eg, a total of 60 LC-MS/MS analysis runs) are processed together because they have consistent retention times. For example, a total of 177 quality attributes were identified and quantified as being in excess of the detection limit (eg, as indicated by a non-zero peak area in all 60 assay runs). These attributes cover a wide range of score levels, including from 0.001 to 39%. Such attributes included sequence variants, hydroxylation, N-linked and O-linked glycans, N-terminal and C-terminal variants, fragmentation, glycation, oxidation, deamidation, succinimide formation, and so on. Fragmentations were distinguished from non-specific trypsin activity by their increased level in the stressed sample (eg, t-test p-value <0.005 and fold change >2.0 compared to reference standard assay cycles).

Example of Response Factor Calibration (a-calibration) Using the reference standard for calibrating the response factor (see Equations (5) and (6)) score values were calculated for each of the 177 attributes in three samples. Fig. 6 is a chart 602 of measured scores for two attributes in sample 2 through two different instrument setups and two different sample preparation procedures. After calibrating the response factor as described herein, variations due to differences in sample preparation and instrument setup are significantly reduced, as indicated by the relative standard deviation (RSD) of the 12 measurements presented in FIG. 6 (ie, A/1, A/2, B/1 and B/2). In particular, the embodiment shown in FIG. 6 represents the score in sample 2, with and without response factor calibration, of the heavy chain variant Cys127Tyr (top) (604) and non-glycosylated Asn289 (bottom) (606), measured by two LC-MS systems (e.g., A and B) and using two splitting protocols (/1 and /2). After calibration of the response factor (see Equations (6)) differences between instruments and sample preparation procedures are significantly reduced, as indicated by RSD values of 605 and 607.

Fig. 7 is a chart 702 depicting an illustrative comparison of within-lab precision (eg, indicated by RSD) with and without (704) response factor calibration (706) (see equation (6)) for three attribute types. In FIG. 7 shows that within-lab precision improves significantly after calibration for approximately 86% of the traceable attributes. The shape of the values shown in FIG. 7 represents the type of attributes, and open figure values represent attributes with a sufficiently high score in the reference standard (within the RSD < 10% sequence).

In particular, in FIG. 7 shows the determined RSD (over 12 measurements with two splitting protocols and two instrument setups as shown in figure 6) of all 177 attributes (120 attributes of type 1, 12-2 type and 45-3 type) in three samples, with (704) response factor calibration and without (706) it. In FIG. 7 represents a total of 177x3=531 data points, each with 12 measurements. The RSD of most (457 or 86%) of the data points decreases after the response factor is calibrated. Of the 531 data points, the number of points with RSD < 10% increased from 64 (12%) before calibration to 205 (39%) after calibration. Among them, in the embodiment of FIG. 7, the number of type 1 attributes increased from 55 to 171, type 2 attributes from 2 to 36, and type 3 attributes from 7 to 19. Among these 205 data points with acceptable intralaboratory precision (RSD < 10%), 83% (171 of 205) are type 1 attributes.

As already discussed, response factor calibration requires that the attribute score in the reference standard be generally high enough to be accurately quantified. In FIG. In Figure 7, attributes with a score greater than 10 times the standard deviation (within the RSD < 10% sequence) are denoted by open figure values. For example, at least in the embodiment of FIG. 7, most attributes with an RSD below 10% after calibration have a high score in the reference standard (i.e. values indicated by open figures).

Example of calibration of artificial modifications (b-calibration)

For Type 2 and Type 3 attribute embodiments, differences in artificial modifications under different sample preparation conditions can be calibrated using Equation (8) for Type 2 attributes and Equation (9) for Type 3 attributes.

- 29 039676

In FIG. 8 is a diagram 802 of an embodiment of artificial modification calibration for type 2 attributes showing similar results as for the a-calibration of the response coefficient. In particular, in FIG. 8 shows the improvement in intra-laboratory precision for type 2 attributes (eg, two attributes in three samples got six measurements (806)) after calibration, compared against RSD (804) after calibrating the response factor. In accordance with the embodiments described herein, the two calibration methods produced very similar RSDs.

In FIG. 9 is a graph 902 of the results of b-calibration artificial modification for type 3 attributes versus a-calibration of the response coefficient. In FIG. Figure 9 shows that for type 3 attributes, calibration (904) of artificial modifications improved the in-lab precision for most measurements. For attributes with a calibration parameter b close to the uncalibrated characteristic score I/(I ₀ +I), the calibrated score A (see Equation (9)) can be close to zero, so it can significantly affect the RSD values. For a fair comparison of different calibration methods, the RSD was calculated by dividing the standard deviation of the calibrated score by the mean of the calibrated attribute score. Compared to response factor calibration (a-calibration), artificial modification calibration (b-calibration) gave similar results when the RSD of uncalibrated scores (906) is less than 50%. If the RSD of the uncalibrated score is above 50%, the b-calibration results may become less effective than the a-calibration, and if the RSD of the uncalibrated score is close to 100%, the artificial modification calibration makes the result less consistent (e.g., higher RSD).

Calibration example for both response factors (a) and artificial modifications (b)

In some embodiments, both the response factor and the artificial modification can be calibrated for type 3 attributes when the second standard is easily obtained by stressing the reference standard. In FIG. 10 is an embodiment diagram 1002 showing the results of calibrating with two standards (a and b) for type 3 attributes, compared to a single standard (a) response factor calibration. In particular, in FIG. 10 shows the results of calibration (1004) versus no calibration (1006) using a reference standard and a stressed sample (sample 1) as two standards and samples 2 and 3 as analysis samples. The RSD was calculated by dividing the standard deviation of the calibrated score by the mean a-calibrated attribute score. Fig. 10 demonstrates that two-standard calibration results are generally better than one-standard response factor calibration, but the size of the improvement may not be significant, and in such circumstances, but not in all, may not justify keeping a second reference standard for the life of this product.

Illustrative Response Factor Calibration with Data Generated in Completely Different Systems for LC-MS

In FIG. 11 is a chart 1102 showing the improvement made in the consistency of two sets of data obtained using two different systems for LC-MS by calibrating the response factor. In particular, in FIG. 11 is an illustration of a comparison of performance scores determined by system A (1104) and system C (1106) with and without (1116) response factor calibration. In FIG. Figure 11 also compares the within-laboratory precision (RSD) scores for both datasets with and without calibration. As shown in FIG. 11, values indicated by open figures represent attributes with a score that is at least ten times the standard deviation in the reference standard (eg, within the RSD < 10% sequence).

In other words, in FIG. 11 shows the results of response factor calibration based on LC-MS data analysis, where LC-MS data is collected using two different LC techniques and two different systems for MS (systems A (1104) and C (1106)), where sample preparation is performed by two different procedures. In the embodiment shown in FIG. 11, determined attribute score values were compared with each other with (1114) response coefficient calibration and without (1116) it.

Sample analysis using a different technique may detect and identify a different set of attributes, as shown in FIG. 11, among which 117 are common. In FIG. 11, measuring 117 attributes in three samples yielded 351 data points and corresponding determined scores obtained by two methods, with and without a-calibration. Such values are shown in Fig. 11 (upper graph) (1103). In FIG. 11 shows that the calibration significantly improved the consistency of the measurements, even though the two LC-MS systems were completely different. Fig. 11 (bottom plot) (1113) shows the improvement in intralaboratory precision after calibrating the response factor for 117 attribute data. Many of these attributes can be accurately measured with the same instrument and technique (indicated by the values denoted

- 30 039676 unshaded figures).

However, when such attributes were measured by another system, the results were not consistent, as evidenced by high RSD values of 10 to 100% on the horizontal axis (uncalibrated). Most of these measurements became consistent again after calibration, as evidenced by their RSD values on the vertical axis that are less than 10%.

Illustrative glycan profile of an a-fusion protein

In this example, the fusion protein contains two O-linked glycosylation sites with six different glycoforms. A sample with an unknown glycoform quantitation, along with a reference standard, was analyzed in four runs by three analytical chemists on three LC-MS systems (see Table 3). All data were processed on Chromeleon to obtain peak areas corresponding to each peptide of interest. Each glycoform was quantified using the conventional method (Equation (4)), as well as the response factor calibration (Equation (6)). To quantify each attribute in the reference standard, the reference standard was analyzed by conventional MAM six times, and the average of the six measurements was used as the known quantification in the reference standard.

In FIG. 12 is a chart 1202 showing the measured glycoform (SAHexHexNAc) quantitation (1204) in four sequences (1206), in triplicate each. After calibration of the response factor (equation (6)) the variation between instruments was significantly reduced. In FIG. 12 (lower plot) also shows that Response Factor Calibration for six glycoforms (1216) reduced the RSD (1214) from maximum 21 to maximum 5% (SA: sialic acid or N-acetylneuraminic acid; Hex: hexose; HexNAc: N-acetylhexosamine ). In particular, Figure 12 (top graph) (1203) shows 12 measured scores (triple analysis in 4 sequences 1206) for one of the glycoforms with and without response factor calibration. Although the measured scores were highly reproducible in the same sequence, such measured scores varied significantly from sequence to sequence. Response factor calibration eliminated the difference caused by instruments. In FIG. 12 (bottom plot) (1213) compares the intralaboratory RSD for all six glycoforms with and without calibration. The RSD for these glycoforms ranged from 2.4 to 21% without calibration. After calibration, it was reduced to no more than 5%.

Calibration

In some, but not all, embodiments, attribute scoring using calibration methods may require additional measurement of the ion intensities of each isoform in the reference standard. These extra dimensions can introduce additional errors into the final attribute score calculation. An improvement in intralaboratory precision can be achieved if the errors caused by these additional measurements are smaller than the difference between the laboratory and the sample.

In some, but not all, embodiments, it is necessary to determine the ion intensity of each isoform in the reference standard after calibration to ensure improved measurement precision. This requires that the attribute of interest be contained in a sufficiently high amount in the reference standard. Typically, an attribute score should be at least ten times the standard deviation of the measurement (shown, for example, as open figure values in FIGS. 7, 8, 9, and 10 described herein).

For embodiments involving Type 3 b-calibration (Equation (9)), calculating the attribute score A may include taking the difference between the uncalibrated attribute score I/(I ₀ +I) and b. If the value of b is close to I/(I ₀ +I), then the calculation of A involves taking the difference of two large numbers to produce a very small number, potentially introducing large errors. In extreme cases, the value of b can be greater than I/(I ₀ +I), resulting in a negative value of A. This can make b-sizing for type 3 attributes less reliable. The same is true for the a- and b-sizing for type 3 attributes shown in equation (12).

In some embodiments, the difference in response factors, which include both digestion efficiency and instrument response, may be of particular importance because the use of HPLC and MS tools, as well as sample preparation automation, are unavoidable. Artificial modifications can generally be controlled (as described herein) and are less of a problem. Therefore, when comparing the three calibration methods, the response factor calibration (a-calibration) is usually the preferred one due to its reliability and applicability to all three types of attributes. Artificial modification calibration (b-calibration) for type 3 attributes, although it can correct sample preparation inconsistencies not corrected by a-calibration, is usually less reliable,

- 31 039676 if the correction level is close to the attribute score. In some embodiments, b-sizing on a type 2 attribute, due to the similarity in mathematical aspects, gives the same results as a-sizing, so it can be replaced by a-sizing. Calibration to two standards for both response factor and artificial modification (a- and b-calibration), on the other hand, in some, but not all, embodiments, may be less practical due to the need for an additional standard over time. product use. In addition, two more ion intensity measurements may be needed for a- and b-calibration, further increasing the variance in the final attribute score result. Therefore, response factor calibration using a single standard is commonly used to implement new MAM systems and methods, as described in this document, under current GMP rules.

The new MAM systems and methods have significant advantages over traditional MAM because the new MAM systems and methods significantly reduce inter-laboratory variability through on-line calibration of responses. The new MAM systems and methods virtually eliminate the requirement for MAM to use a single piece of equipment, which is a major problem in the current MAM workflow. In addition, since the reference standard should normally already be analyzed in parallel with the sample in the current run, no additional work is required by the analytical chemist.

As described herein, using equation (6) to calibrate response factors eliminates the requirement that different peptide isoforms have the same response factor. Therefore, new MAM systems and methods can be used for other types of MAM tools. For example, due to the requirement of equivalence of response coefficients for different isoforms in traditional MAM, selected response monitoring (SRM) on a triple quadrupole instrument is unacceptable due to the potentially very different fragmentation efficiency among peptide isoforms. New MAM systems and methods, on the other hand, allow the advantage of triple quadrupole instruments because such new MAM systems and methods do not require different peptide isoforms to have equivalent response factors. The score of each attribute in the reference standard, however, must initially be set on the high resolution instrument.

One disadvantage of this conventional MAM methodology is that the measured attribute score is not absolute due to the assumption that all isoforms have the same response coefficients, which may not be true for modifications that include changes in charge, hydrophobicity, or length of the peptide. This disadvantage usually does not present a particular problem as long as the response factors are consistent throughout the lifetime of the method. However, if an attribute score in the reference standard is determined by an absolute scoring methodology, then the attribute score determined in accordance with the new MAM methodology also becomes an absolute value.

The data shown for the various embodiments herein can also be used to gain insight into the limit of quantification (LOQ) of the MAM platform. LOQ can be defined as the minimum concentration of an attribute with an RSD of less than 10%. For example, in FIG. 13 is a chart 1302 showing the relationship of an attribute score (1306 and/or 1316) and RSD within a sequence (1304 and/or 1314) for three attribute types. In the embodiment shown in FIG. 13, for type 1 attributes (top chart) (1303) the RSD (1304) for most attributes is less than 10% with scores (1306) not exceeding 0.003% (LOQ=0.003%) if the minor minor isoform peak is well separated from other peaks corresponding to the main peptides. If there is poor separation of the isoform peak from the main peak, the LOQ will be higher due to the limited dynamic range of the mass spectrometers used in this work within the scan. However, for type 2 and type 3 attributes (lower chart) (1313), most attributes below 0.1% have an RSD (1314) >10%, and most attributes above 1% have an RSD (1314) < 10%, assuming values LOQs are typically between ~0.1 and ~1% depending on the amount of difference introduced during sample preparation.

In particular, in FIG. 13 shows the relationship of score (1306 and/or 1316) and RSD within sequence (n=6) (1304 and/or 1314) for different attribute types. For most Type 1 attributes in FIG. 13 (upper chart) (1303) the RSD values within the sequence are below 10%, with a scoring down to 0.003%, indicating that the quantification limit of the system for LC-MS only reaches 0.003% if the attribute is not changed in sample preparation. As also shown in FIG. 13 (bottom chart) (1313), quantitation limits are much higher (0.1-1%) for most type 2 and 3 attributes.

- 32 039676

Additional aspects

The following additional aspects of the present invention are illustrative only and are not intended to limit the scope of the present invention. The following additional aspects can be considered as part of or in addition to the aspects of the present invention, or separately from them, including as a non-limiting example, any of the aspects 1-26 described in this document.

27. A calibration system adapted to reduce the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating signal intensity during operation, wherein the calibration system comprises a first MAM-enabled instrument containing a first detector, wherein the first MAM-enabled instrument is characterized state of the first instrument as defined by at least one of (1) the model of the first instrument or (2) the first set of settings, wherein the first MAM-enabled instrument is adapted to receive the first sample and reference standard, and the first MAM-enabled instrument is further adapted to detect by means of the first detector of the first isoform of the sample in the first sample and the first isoform of the reference standard in the reference standard, where the first sample is prepared by the first preparation type; one or more processors associated with the first MAM-enabled device, wherein the one or more processors associated with the first MAM-enabled device are adapted to determine, by a first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard, and one or more processors associated with the first MAM-enabled instrument are further adapted to determine a first set of sample score values corresponding to the isoform of the first sample, where the first set of sample score values is based on the first set of correction factors ; a second MAM-enabled device comprising a second detector, wherein the second MAM-enabled device is characterized by a state of the second device defined by at least one of: (1) a model of the second device or (2) a second set of settings, where the state of the second device is different from the state of the first instrument, wherein the second MAM-enabled instrument is adapted to receive the second sample and the reference standard, and the second MAM-enabled instrument is further adapted by a second detector to detect the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard, where the second sample is prepared by a second preparation type, and wherein the first preparation type is different from the second sample preparation type, resulting in a variation between the first sample and the second sample, the variation being caused by artificially changing the attribute score during the sample preparation; and one or more processors associated with the second MAM-enabled device, wherein the one or more processors associated with the second MAM-enabled device are adapted to determine, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors coefficients is based on the reference standard, and one or more processors associated with the second MAM-enabled instrument are further adapted to determine a second set of sample score values corresponding to an isoform of the second sample, where the second set of sample score values is based on the second set of correction values. coefficients, and wherein the value of variation for the first set of score values in the sample and the second set of score values in the sample is reduced taking into account the first set of correction factors and the second set of correction factors, where at least one of (1) one or more processors associated with the first MAM-enabled device through the first MAM iteration or (2) one or more processors associated with the second MAM-enabled device through the second MAM iteration determines the quality attribute with decreasing variation between the first sample and the second sample, wherein the quality attribute is associated with a type 2 attribute or a type 3 attribute.

28. The calibration system according to aspect 27, where the type 2 attribute causes a decrease in the quantitative indicator during the preparation of the first sample or the second sample.

29. The calibration system according to aspect 27, where the type 3 attribute causes an increase in the quantitative indicator during the preparation of the first sample or the second sample.

30. Calibration method for reducing the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration method provides for the reception in the first device with MAM support of the first sample and the reference standard, while the first device with MAM support comprises a first detector, and the first MAM-enabled device is characterized by a first device state defined by at least one of (1) the first device model or (2) the first set of settings; detection by the first detector based on the MAM device of the first isoform of the sample from the first sample and the first isoform of the reference standard from the reference standard, where the first sample is prepared by the first type of preparation; determination by one or more

- 33 039676 processors associated with the first MAM-enabled instrument through the first MAM iteration of the first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard; determining, by means of one or more processors associated with the first MAM-enabled device, a first set of sample score values corresponding to an isoform of the first sample, wherein the first set of sample score values is based on the first set of correction factors; receiving by the second MAM-enabled instrument a second sample and a reference standard, wherein the second MAM-enabled instrument comprises a second detector, wherein the second MAM-enabled instrument is characterized by a state of the second instrument defined by at least one of: (1) the model of the second instrument, or (2) a second set of settings, where the state of the second device is different from the state of the first device; detection by the second detector of the second MAM-enabled instrument of the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard, where the second sample is prepared by the second preparation type, and where the first preparation type is different from the second preparation type, resulting in a variation between a first sample and a second sample, the variation being caused by an artificial change in the attribute score during sample preparation; determining, by means of one or more processors associated with a second MAM-enabled instrument, by means of a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard; and determining, by means of one or more processors associated with the second MAM-enabled instrument, a second set of sample score values corresponding to an isoform of the second sample, wherein the second set of sample score values is based on a second set of correction factors, where the variation value for the first set is the sample score values and the second set of sample score values are reduced with respect to the first set of correction factors and the second set of correction factors, and where at least one of (1) one or more processors associated with the first MAM-enabled instrument, by means of the first iteration of the MAM, or (2) one or more processors associated with the second MAM-based device, through the second iteration of the MAM determines a quality attribute with a decrease in variation between the first sample and the second sample, while the quality attribute is associated with a type 2 attribute or at ributome type 3.

31. The calibration method according to aspect 30, wherein the type 2 attribute causes the score to decrease during the preparation of the first sample or the second sample.

32. The calibration method according to aspect 30, wherein the type 3 attribute causes the score to increase during the preparation of the first sample or the second sample.

33. A calibration system adapted to reduce the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration system includes a first MAM-enabled instrument containing a first detector, while the first MAM-enabled instrument is characterized state of the first instrument, as defined by at least one of: (1) the model of the first instrument, or (2) the first set of settings, wherein the first MAM-enabled instrument is adapted to receive the first sample, reference standard, and stressed standard, and wherein the first the MAM-enabled instrument is further adapted by a first detector to detect the first sample isoform in the first sample, the first reference standard isoform in the reference standard, and the stressed reference standard isoform in the stressed standard; one or more processors associated with the first MAM-enabled instrument, wherein the one or more processors associated with the first MAM-enabled based instrument are adapted by a first MAM iteration to determine a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors coefficients is based on the reference standard and the stressed standard, and one or more processors associated with the first MAM-enabled instrument are further adapted to determine a first set of score values in the sample corresponding to the isoform of the first sample, where the first set of score values in the sample is based on the first set of correction factors; a second MAM-enabled device comprising a second detector, wherein the second MAM-enabled device has a second device state defined by at least one of (1) a model of the second device or (2) a second set of settings, wherein the state of the second device is different from the state of the first instrument, wherein the second MAM-enabled instrument is adapted to receive the second sample, the reference standard and the stressed standard, and the second MAM-enabled instrument is further adapted by a second detector to detect the isoform of the second sample in the second sample, the second isoform of the reference standard in the reference the standard and the second isoform of the stressed standard in the stressed standard; and one or more processors associated with a second device with support

- 34 039676

MAM, wherein one or more processors associated with a second MAM-enabled instrument are adapted by a second MAM iteration to determine a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on the reference standard and the stressed standard, and wherein the one or more processors associated with the second MAM-enabled instrument are further adapted to determine a second set of sample score values corresponding to an isoform of the second sample, wherein the second set of sample score values is based on the second set of correction factors, and wherein the variation value of the first set of score values in the sample and the second set of score values in the sample is reduced by considering the first set of correction factors and the second set of correction factors.

34. The calibration system according to aspect 33, where the stressed standard contains a higher level of quality attributes compared to the reference standard.

35. A calibration method for reducing the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration method includes receiving in the first device with MAM support the first sample, the reference standard and the stressed standard, while the first MAM-enabled device comprises a first detector, and the first MAM-enabled device is characterized by a state of the first device defined by at least one of (1) the model of the first device or (2) the first set of settings; detecting by the first detector of the MAM-enabled instrument the isoform of the first sample in the first sample, the first isoform of the reference standard in the reference standard, and the isoform of the stressed reference standard in the stressed standard; determining, by one or more processors associated with the first MAM-enabled instrument, by means of a first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard and a stressed standard; determining, by means of one or more processors associated with the first MAM-enabled device, a first set of sample score values corresponding to an isoform of the first sample, wherein the first set of sample score values is based on the first set of correction factors; receiving in a second MAM-enabled instrument a second sample, a reference standard, and a stressed standard, wherein the second MAM-enabled instrument comprises a second detector, wherein the second MAM-enabled instrument has a second instrument state defined by at least one of (1) a second instrument model or (2) a second set of settings, wherein the state of the second instrument is different from that of the first instrument; detecting, by means of a second detector of a second MAM-enabled instrument, a second sample isoform in the second sample, a second isoform of the reference standard in the reference standard, and a second isoform of the stressed standard in the stressed standard; determining, by one or more processors associated with a second MAM-enabled instrument, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on the reference standard and the stressed standard; and determining, by means of one or more processors associated with the second MAM-enabled instrument, a second set of sample score values corresponding to an isoform of the second sample, wherein the second set of sample score values is based on the second set of correction factors, wherein the variation value of the first set is the sample score values and the second set of sample score values are reduced by the first set of correction factors and the second set of correction factors.

36. The calibration method according to aspect 35, wherein the stressed standard contains a higher level of quality attributes compared to the reference standard.

Claims (36)

1. A calibration system adapted to reduce the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating signal intensity during operation, wherein the calibration system comprises a first MAM-enabled instrument containing a first detector, wherein the first MAM-enabled instrument is characterized state of the first instrument as defined by at least one of (1) the model of the first instrument or (2) the first set of settings, wherein the first MAM-enabled instrument is adapted to receive the first sample and reference standard, and the first MAM-enabled instrument is further adapted to detect by means of a first isoform detector of the first sample in the first sample and the first isoform of the reference standard in the reference standard; - 35 039676 one or more processors associated with the first MAM-enabled device, wherein one or more processors associated with the first MAM-enabled device are adapted to determine, by the first MAM iteration, the first set of correction factors corresponding to the isoform of the first sample, where the first the set of correction factors is based on the reference standard, and one or more processors associated with the first MAM-enabled instrument are further adapted to determine a first set of score values in the sample corresponding to the isoform of the first sample, where the first set of score values in the sample is based on the first a set of correction factors; a second MAM-enabled device comprising a second detector, wherein the second MAM-enabled device is characterized by a state of the second device defined by at least one of (1) a model of the second device or (2) a second set of settings, where the state of the second device is different from the state of the first an instrument, wherein the second MAM-enabled instrument is adapted to receive the second sample and the reference standard, and the second MAM-enabled instrument is further adapted to detect, by means of the second detector, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard; and one or more processors associated with the second MAM-enabled device, wherein the one or more processors associated with the second MAM-enabled device are adapted to determine, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors coefficients is based on the reference standard, and one or more processors associated with the second MAM-enabled instrument are further adapted to determine a second set of sample score values corresponding to an isoform of the second sample, where the second set of sample score values is based on the second set of correction values. coefficients, and where the variation value for the first set of score values in the sample and the second set of score values in the sample is reduced by taking into account the first set of correction factors and the second set of correction factors. 2. The calibration system of claim 1, wherein one or more processors associated with the first MAM-enabled device determine a quality attribute through a first MAM iteration. 3. The calibration system according to claim 2, where the quality attribute is any of the isoform of the first sample, protein, or identified impurity. 4. The sizing system of claim 2, wherein the quality attribute is defined by one or more of fragmentation, oxidation, glycation, hydroxylation, sequence variant, isomerization, deamination, C-terminal lysine, O-linked glycan, and/or N-linked glycan. 5. The calibration system of claim 2, wherein one or more processors associated with the first MS instrument are adapted to generate a report containing a quality attribute. 6. The calibration system according to claim 1, where the model of the first instrument is different from the model of the second instrument. 7. The calibration system of claim 1, wherein the first set of settings is different from the second set of settings. 8. The calibration system of claim 1, wherein the first set of correction factors is based on the ion intensity value of the first isoform of the reference standard and the first score value in the reference standard of the first isoform of the reference standard. 9. The calibration system of claim 1, wherein the first set of correction factors calibrates the response factor associated with the first set of sample score values to determine an isoform ion intensity value of the first sample. 10. The calibration system of claim 1, wherein the first set of score values in the sample is further based on an isoform ion intensity value of the first sample. 11. The calibration system of claim 1, wherein the first MAM-enabled instrument is a mass spectrometry (MS) instrument. 12. The calibration system of claim 1, wherein the first MAM-enabled instrument is a tri-quadrupole instrument. 13. The calibration system of claim 1, wherein one or more processors associated with the first MS instrument are communicatively connected over a computer network to one or more processors associated with the second MS instrument. 14. The calibration system of claim 1, wherein the one or more processors associated with the first MAM-enabled device are one or more processors associated with the second MAM-enabled device. 15. The calibration system of claim 1, wherein the first MAM-enabled instrument is located in the first laboratory at the first geographic location and the second MAM-enabled instrument is located in the second laboratory at the second geographic location. 16. Calibration method for reducing the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration method provides - 36 039676 reception in the first MAM-enabled instrument containing the first detector, the first sample and the reference standard; detecting by the first detector the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard; determining, by one or more processors associated with the first MAM-enabled instrument, during the first MAM iteration, a first set of correction factors corresponding to the first isoform of the sample, where the first set of correction factors is based on a reference standard; determining, by means of one or more processors associated with the first MAM-enabled instrument, a first set of score values in the sample corresponding to an isoform of the first sample, where the first set of score values in the sample is based on the first set of correction factors, and where the first MAM-enabled instrument characterized by the state of the first device, defined by at least one of (1) the model of the first device or (2) the first set of settings; receiving, in a second MAM-enabled instrument containing a second detector, a second sample and a reference standard; detecting by the second detector the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard; determining, by one or more processors associated with a second MAM-enabled instrument, during a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard; determining, by one or more processors associated with the second MAM-enabled device, a second set of sample score values corresponding to an isoform of the second sample, where the second set of sample score values is based on the second set of correction factors, and where the second MAM-enabled device is characterized a state of the second instrument defined by at least one of (1) the model of the second instrument or (2) the second set of settings, where the state of the second instrument differs from the state of the first instrument, and where the variation value for the first set of score values in the sample and the second set of score values indicators in the sample is reduced taking into account the first set of correction factors and the second set of correction factors. 17. Calibration method for reducing the variability of the MAM-supported instrument over several time periods by calibrating the signal intensity during operation, wherein the calibration method includes receiving in the MAM-supported instrument during a first time period a first sample and a reference standard; detecting, by means of the detector of the MAM-enabled instrument, within a first period of time, the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard; determining by one or more processors during the first iteration of MAM during the first period of time the first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard; determining, by one or more processors, by means of a first MAM iteration during a first time period, a first set of score values in the sample corresponding to an isoform of the first sample, where the first set of score values in the sample is based on the first set of correction factors, and where the MAM-enabled instrument for the first period of time is characterized by the state of the first device, defined by the first set of settings; reception in the device with MAM support during the second period of time of the second sample and reference standard; detecting, by the detector of the MAM-enabled instrument, within a second time period, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard; determining by one or more processors during the second MAM iteration during the second time period a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard; and determining, by one or more processors, through a second MAM iteration over a second time period, a second set of sample score values corresponding to an isoform of the second sample, where the second set of sample score values is based on the second set of correction factors, and where the MAM capable instrument is those The value of the second time period is characterized by the state of the second instrument defined by the second set of settings, where the state of the second instrument is different from the state of the first instrument, and where the value of variation for the first set of quantitative values in the sample and the second set of quantitative values in the sample is reduced taking into account the first a set of correction factors; and a second set of correction factors. 18. The calibration method of claim 17, wherein one or more processors determine a quality attribute. 19. The calibration method of claim 18, wherein the quality attribute is any of a first sample isoform, a second sample isoform, a protein, or an identified contaminant. 20. The calibration method of claim 18, wherein the one or more processors are adapted to generate a quality attribute report. 21. The calibration system of claim 1, wherein the variation value for the first set of sample score values and the second set of sample score values is reduced by at least 25%. 22. The calibration method of claim 16, wherein the variation value for the first set of sample score values and the second set of sample score values is reduced by at least 25%. 23. The calibration method of claim 17, wherein the variation value for the first set of sample score values and the second set of sample score values is reduced by at least 25%. 24. The calibration system of claim 1, wherein the first sample is a proteolytic peptide, the second sample is a proteolytic peptide, and the reference standard is a proteolytic peptide. 25. The calibration method of claim 16, wherein the first sample is a proteolytic peptide, the second sample is a proteolytic peptide, and the reference standard is a proteolytic peptide. 26. The calibration method of claim 17, wherein the first sample is a proteolytic peptide, the second sample is a proteolytic peptide, and the reference standard is a proteolytic peptide. 27. A calibration system adapted to reduce the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating signal intensity during operation, wherein the calibration system comprises a first MAM-enabled instrument containing a first detector, wherein the first MAM-enabled instrument is characterized state of the first instrument as defined by at least one of (1) the model of the first instrument or (2) the first set of settings, wherein the first MAM-enabled instrument is adapted to receive the first sample and reference standard, and the first MAM-enabled instrument is further adapted to detect by means of the first isoform detector of the first sample in the first sample and the first isoform of the reference standard in the reference standard, where the first sample is prepared by the first preparation type; one or more processors associated with the first MAM-enabled device, wherein the one or more processors associated with the first MAM-enabled device are adapted to determine, by a first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard and one or more processors associated with the first MAM-enabled instrument are further adapted to determine a first set of score values in the sample corresponding to the isoform of the first sample, where the first set of score values in the sample is based on the first set of correction factors; a second MAM-enabled device comprising a second detector, wherein the second MAM-enabled device is characterized by a state of the second device defined by at least one of (1) a model of the second device or (2) a second set of settings, where the state of the second device is different from the state of the first instrument, wherein the second MAM-enabled instrument is adapted to receive the second sample and the reference standard, and the second MAM-enabled instrument is further adapted to detect, by means of the second detector, the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard, where the second sample is prepared by a second type of preparation, and where the first type of preparation is different from the second type of sample preparation, resulting in variation between the first sample and the second sample, the variation being caused by artificially changing the attribute score during sample preparation; and one or more processors associated with the second MAM-enabled device, wherein the one or more processors associated with the second MAM-enabled device are adapted - 38 039676 to determine, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard, and one or more processors associated with the second MAM-enabled device are further adapted to determine the second set of values of the sample score corresponding to the isoform of the second sample, where the second set of sample score values is based on the second set of correction factors, and where the value of variation for the first set of sample score values and the second set of sample score values is reduced given the first set correction factors and a second set of correction factors, where at least one of (1) one or more processors associated with the first MAM-enabled device through the first MAM iteration or from (2) one or more of the processors associated with the second MAM-enabled device, through a second iteration, the MAM determines a quality attribute with reduced variation between the first sample and the second sample, the quality attribute being associated with a type 2 attribute or a type 3 attribute. 28. The calibration system of claim 27, wherein the type 2 attribute causes the score to decrease during the preparation of the first sample or the second sample. 29. The calibration system of claim 27, wherein the type 3 attribute causes the score to increase during the preparation of the first sample or the second sample. 30. Calibration method for reducing the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration method provides for the reception in the first device with MAM support of the first sample and the reference standard, while the first device with MAM support comprises a first detector and a first MAM-enabled device characterized by a state of the first device defined by at least one of (1) the model of the first device or (2) the first set of settings; determining, by the first detector of the first MAM-enabled instrument, the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard, where the first sample is prepared by the first type of preparation; determining, by one or more processors associated with the first MAM-enabled instrument, during the first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard; determining, by means of one or more processors associated with the first MAM-enabled device, a first set of sample score values corresponding to an isoform of the first sample, wherein the first set of sample score values is based on the first set of correction factors; receiving in the second MAM-enabled instrument a second sample and a reference standard, wherein the second MAM-enabled instrument comprises a second detector, wherein the second MAM-enabled instrument is characterized by a state of the second instrument defined by at least one of (1) the model of the second instrument or ( 2) the second set of settings, where the state of the second device differs from the state of the first device; detection by the second detector of the second MAM-enabled instrument of the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard, where the second sample is prepared by a second preparation type and where the first preparation type is different from the second sample preparation type, resulting in variation between the first a sample and a second sample, the variation being caused by an artificial change in the attribute score during sample preparation; determining, by means of one or more processors associated with a second MAM-enabled instrument, during a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on a reference standard; and determining, by means of one or more processors associated with the second MAM-enabled device, a second set of sample score values corresponding to an isoform of the second sample, wherein the second set of sample score values is based on the second set of correction factors; where the variation value for the first set of sample score values and the second set of sample score values is reduced given the first set of correction factors and the second set of correction factors, and where at least one of (1) one or more processors associated with the first instrument with MAM support, through a first MAM iteration, or from (2) one or more processors associated with a second MAM-enabled device, through a second MAM iteration, determines a quality attribute with reduced variation between the first sample and the second sample, wherein the attribute - 39 039676 quality bottle is associated with a type 2 attribute or a type 3 attribute. 31. The calibration method of claim 30, wherein the type 2 attribute causes the score to decrease during the preparation of the first sample or the second sample. 32. The calibration method of claim 30, wherein the type 3 attribute causes the score to increase during the preparation of the first sample or the second sample. 33. A calibration system adapted to reduce the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration system includes a first MAM-enabled instrument containing a first detector, while the first MAM-enabled instrument is characterized state of the first instrument as defined by at least one of (1) the model of the first instrument or (2) the first set of settings, wherein the first MAM-enabled instrument is configured to receive the first sample, reference standard, and stressed standard, and wherein the first instrument with MAM support further adapted to detect by the first detector the isoform of the first sample in the first sample, the first isoform of the reference standard in the reference standard and the isoform of the stressed reference standard in the stressed standard; one or more processors associated with the first MAM-enabled device, wherein the one or more processors associated with the first MAM-enabled device are adapted to determine, by a first MAM iteration, a first set of correction factors corresponding to the isoform of the first sample, where the first set of correction factors is based on a reference standard and a stressed standard, and one or more processors associated with the first MAM-enabled instrument are further adapted to determine a first set of sample score values corresponding to an isoform of the first sample, where the first set of sample score values is based on the first set of correction factors; a second MAM-enabled device comprising a second detector, wherein the second MAM-enabled device is characterized by a state of the second device defined by at least one of (1) a model of the second device or (2) a second set of settings, where the state of the second device is different from the state of the first instrument, wherein the second MAM-enabled instrument is adapted to receive the second sample and the reference standard and the stressed standard, and the second MAM-enabled instrument is further adapted to detect, through the second detector, the isoform of the second sample in the second sample, the second isoform of the reference standard in the reference standard , and the second isoform of the stressed standard in the stressed standard; and one or more processors associated with the second MAM-enabled device, wherein the one or more processors associated with the second MAM-enabled device are adapted to determine, by a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors coefficients is based on the reference standard and the stressed standard, and one or more processors associated with the second MAM-enabled instrument are further adapted to determine a second set of sample score values corresponding to an isoform of the second sample, where the second set of sample score values is based on a second set of correction factors, and where the value of variation for the first set of score values in the sample and the second set of score values in the sample is reduced by the first set of correction factors and a second set of correction factors. 34. The calibration system of claim 33, wherein the stressed standard contains a higher level of quality attributes than the reference standard. 35. A calibration method for reducing the variability of multi-parameter methods (MAM) between laboratories or between instruments by calibrating the intensity of signals during operation, while the calibration method includes receiving in the first device with MAM support the first sample, the reference standard and the stressed standard, while the first MAM-enabled device comprises a first detector, and the first MAM-enabled device is characterized by a state of the first device defined by at least one of (1) the model of the first device or (2) the first set of settings; determining by means of the first detector of the first MAM-enabled instrument the isoform of the first sample in the first sample, the first isoform of the reference standard in the reference standard, and the isoform of the stressed reference standard in the stressed standard; determination by one or more processors associated with the first MAM-enabled device, during the first MAM iteration, of the first set of correction factors, respectively - 40 039676 corresponding to the isoform of the first sample, where the first set of correction factors is based on the reference standard and the stressed standard; determining, by means of one or more processors associated with the first MAM-enabled device, a first set of sample score values corresponding to an isoform of the first sample, wherein the first set of sample score values is based on the first set of correction factors; the second MAM-enabled instrument receiving a second sample, a reference standard, and a stressed standard, wherein the MAM-enabled instrument comprises a second detector, and the second MAM-enabled instrument has a state of the second instrument determined by at least one of (1) models of the second instrument or (2) a second set of settings, where the state of the second device is different from the state of the first device; detecting, by means of a second detector of a second MAM-enabled instrument, a second sample isoform in the second sample, a second isoform of the reference standard in the reference standard, and a second isoform of the stressed standard in the stressed standard; determining, by means of one or more processors associated with a second MAM-enabled instrument, during a second MAM iteration, a second set of correction factors corresponding to the isoform of the second sample, where the second set of correction factors is based on the reference standard and the stressed standard; and determining, by means of one or more processors associated with the second MAM-enabled device, a second set of sample score values corresponding to an isoform of the second sample, wherein the second set of sample score values is based on the second set of correction factors; where the variation value for the first set of score values in the sample and the second set of score values in the sample is reduced by taking into account the first set of correction factors and the second set of correction factors. 36. The calibration method of claim 35, wherein the stressed standard contains a higher level of quality attributes than the reference standard. Fig. one Fig. 2 - 41 039676 300 //--------------------------------> : : ( Start j ......7........' Reception in the device with MAM support for the first period of time of the first sample of the proteolytic peptide and the reference standard of the proteolytic peptide ' Determination by the detector of the MAM-enabled instrument during the first period of time of the isoform of the first sample in the first sample and the first isoform of the reference standard in the reference standard ................................................. ........I......................................... .... χ-308 Determination by one or more processors of the first MAM iteration during the first period of time of the first correction factor corresponding to the isoform of the first sample, where the first correction factor is based on the reference standard ___ΖΞΞΞΞΞΞΞΖΞΞΙ^^__ Determining by one or more processors, by means of a first iteration of MAM, during a first time period, a first sample score value corresponding to an isoform of the first sample, where the first sample score value is based on the first correction factor, and where the MAM-enabled instrument during the first period time is characterized by _______ the state of the first device, determined by the first set of settings _______ j _g 312 Reception in the device with the support of MAM during the second period of time of the second sample of the proteolytic peptide and the reference standard of the proteolytic peptide Determination by the detector of the MAM-enabled instrument during the second period of time of the isoform of the second sample in the second sample and the second isoform of the reference standard in the reference standard Determination by one or more processors during the second MAM iteration during the second time period of the second correction factor corresponding to the isoform of the second sample, where the second correction factor is based on the reference standard Determining, by one or more processors, by a second iteration of MAM over a second period of time, a second sample score value corresponding to an isoform of the second sample, where the second sample score value is based on the second correction factor, and where the MAM-enabled instrument during the second period time is characterized by the state of the second instrument determined by the second set of settings, wherein the state of the second instrument differs from the state of the first instrument, and the variation value of the first quantitative index value in the sample and the second quantitative index value in the sample is reduced taking into account the first correction factor and the second correction factor