Biologic characterization

Abstract

A method for characterizing a compound of interest by spectroscopic analysis, in the near infrared region, for instance, addresses uncertainties introduced by current practices of preparing duplicate buffer formulations. An illustrative method involves separating the mAb from the buffer in which it is formulated and obtaining a background spectrum of the buffer. Principal component analysis (PCA) can be performed on the concentrated mAb spectrum, obtained by the separation, and spectra of serially diluted mAb formulations. Subtracting the background spectrum from the spectrum obtained by the PCA yields the spectrum of the pure mAb. In turn, this spectrum can then be used as characterization model in the analysis of complex samples containing the mAb, thus eliminating reliance on the general NIST mAb.

Description

Docket: 0376-0032WO1 BIOLOGIC CHARACTERIZATION RELATED APPLICATIONS

[0001] This application claims the benefit under 35 U.S.C. § 119(e) of U.S. Provisional Application No. 63 / 626,293, filed on January 29, 2024, which is incorporated herein by reference in its entirety. BACKGROUND OF THE INVENTION

[0002] Many processes in the chemical, biochemical, pharmaceutical, food, beverage and in other industries benefit from some type of analysis. Of particular importance is the identification and often the quantification of substances present. One common technique involves obtaining a sample and detecting one or more of its components, sometimes referred to as analytes.

[0003] Analytes can be assessed by various optical spectroscopy approaches. Among these, probably the most common is absorption spectroscopy. Incident light excites electrons of the analyte from a low energy ground state into a high energy, excited state, and the energy can be absorbed by both non-bonding n-electrons and π-electrons within a molecular orbital. Absorption spectroscopy can be performed in the ultraviolet, visible, and / or infrared region, with analytes of varying material phases and composition being interrogated by specific wavelengths or wavelength bands of light. The resulting transmitted light is then used to resolve the absorbed spectra, to determine the analyte's or sample’s composition, temperature, pH and / or other intrinsic properties for applications ranging from medical diagnostics, pharmaceutical developments, food and beverage quality control, to list a few.

[0004] Many existing instruments use light in the visible and / or ultraviolet (UV) region of the electromagnetic spectrum. For the past decade, for example, UV-Visible systems have been the gold standard for measuring protein and monoclonal antibodies (mAb) concentrations. However, the accuracy and reproducibility of UV-visible equipment can suffer from dynamic range limitations due to extremely strong absorption by proteins in the UV region, with typical maximum absorptions of about 3 to 4 absorbance units (AU). A partial solution was the development of systems that utilize variable pathlengths.

[0005] In U.S. Patent Application Publication No. 2019 / 0358632A1, Hassell et al. describe analyses of culture media using near infrared (NIR) spectroscopic techniques.Docket: 0376-0032WO1

[0006] Applications of NIR-based techniques to measurements of samples in a flowcell are described in U.S. Patent Application Publication No. 2020 / 0240902A1, to Hassell et al.

[0007] U.S. Patent Application Publication No. 2021 / 0088433 to Hassell et al. describes robust, hands-free, non-destructive, real-time NIR techniques for identifying and / or quantifying constituents in a given process, using an in-situ probe that can be inserted and / or maintained in a bioreactor.

[0008] Clamshell apparatus configurations and techniques for sample analyses can be found in International Patent Application No. PCT / US2023 / 79110, filed on November 8, 2023, and U.S. Provisional Patent Application No. 63 / 512,996, filed on July 11, 2023, both titled Clamshell Analyzer and both being incorporated herein by this reference in their entirety. SUMMARY OF THE INVENTION

[0009] Biological samples often contain multiple components. These can include not only a compound of interest, e.g., a drug, but other ingredients as well. Protein-containing samples, for instance, might include the protein of interest, along with excipients such as buffers, salts, amino acids, sugars, surfactants, and / or other substances. While High Performance Liquid Chromatography (HPLC) can be used to measure excipients, it cannot be easily applied to some, e.g., surfactants. In many cases, HPLC also requires complex procedures, such as sample preparation, dilutions, and / or shipping to analytical laboratories.

[0010] Therefore, simple approaches for measuring excipients in biological samples remain an unmet need. Particularly desired are techniques for simple, accurate and rapid analyses of compounds of interest along with the excipients present. As an example, a need exists for accurate determinations of constituents in formulations containing monoclonal antibodies produced in bioreactors.

[0011] Different compounds, and in particular different biologicals (antibodies, gene therapies and / or cell therapies) have different spectroscopic properties, different absorption spectra, for example. In many cases, the spectrum of a drug product is not known a priori and a need exists for simple and, preferably, also fast protocols for characterizing the drug product spectroscopically, e.g., by optical absorption. Once determined, the spectrum can be used to quantitatively evaluate complex samples that contain the drug product.Docket: 0376-0032WO1

[0012] In a conventional approach, the characterization of excipients in a formulation containing a monoclonal antibody (mAb) relies on the general “NIST mAb” material (a recombinant humanized IgG1κ expressed in murine suspension culture, as known in the art). It was discovered, however, that using the NIST mAb model did not appear to generate an accurate determination of other excipients present in the formulation. For instance, concentrations of histidine and histidine HCl were found to fluctuate in correlation with the mAb concentration (based on the general NIST mAb spectrum) when they should have remained constant.

[0013] Embodiments of the invention recognize that each mAb has a unique spectral (e.g., NIR) signature, due to variable regions of the antibody, and that reliance on a generalized spectral signature can yield misleading results.

[0014] Also addressed in some embodiments is the problem of faithfully reproducing a drug formulation. In the case of a mAb, for instance, remaking a formulation to correspond exactly to the original formulation presents significant, if not insurmountable challenges. Yet even slight variations in formulations will imprint on the newly acquired mAb spectrum.

[0015] To address these and other problems, approaches described herein focus on the formulation on hand. In specific embodiments, a sample is extracted from a bioreactor or analyzed in situ in the bioreactor. The compound of interest, a mAb, for example, is separated as a concentrate from the remaining ingredients, often the supernatant. Scanning this supernatant yields a background spectrum for the actual medium in the actual mAb formulation.

[0016] One implementation employs centrifuge concentration, using a filter designed to retain the compound of interest in an upper region of a suitable tube and pass through the supernatant, which settles at a bottom region. In an illustration, a mAb of around 150 kilodaltons (kDa), is concentrated in a tube having a molecular weight cutoff of 50 kDa.

[0017] In further embodiments, the concentrated mAb and the supernatant, both obtained from the original formulation, are used to prepare one and typically more than one sample, diluted to desired concentrations. As used herein, the term “serially diluted samples” encompasses the most concentrated through the most diluted samples derived from the two fractions (pools) obtained through the separation.Docket: 0376-0032WO1

[0018] In specific implementations, the spectrum of the concentrated and / or of diluted formulations can be obtained using techniques and equipment described in International Patent Application No. PCT / US2023 / 79110 and / or U.S. Provisional Patent Application No. 63 / 512,996, which applications are incorporated herein by this reference. Techniques and equipment such as described in International Patent Application No. PCT / US2023 / 79110 and / or U.S. Provisional Patent Application No. 63 / 512,996 also can be employed to scan the supernatant and generate a true (ideal) background spectrum.

[0019] In embodiments, this true (ideal) background spectrum is subtracted from the spectra of the serially diluted samples. Applying principal component analysis (PCA) to the resulting spectra can then provide a very accurate spectrum of the mAb.

[0020] Exploiting advantages associated with NIR spectroscopic techniques, practicing embodiments of the invention can provide spectra of an unknown compound of interest, a mAb, for instance, along with spectra of excipients present in a formulation with high accuracy. This can be carried out without any need for reproducing the original formulation, thus bypassing errors inherent in preparing duplicate samples.

[0021] In addition to its high accuracy, the method described herein is fast and straightforward. It does not require shipping to offsite analytical labs and can address most if not all types of common excipients found, for instance, in mAb formulations.

[0022] Dilutions and / or PCA techniques can address small variations that may occur when relying solely on the spectrum of the concentrated mAb fraction.

[0023] Principles described with reference to NIR spectroscopy can be adapted or extended to other spectroscopic analysis techniques (Raman, ultraviolet, visible, other IR regions, etc.). Similarly, techniques illustrated here for monoclonal antibodies can be adapted or extended to other compounds of interest, drug products for vaccines, enzymes, other proteins, gene or cell therapies, for instance.

[0024] In general, according to one aspect, the invention features a method for characterizing a compound in a sample from a bioreactor. The method comprises separating the sample into a compound-containing fraction and a supernatant fraction, obtaining a background spectrum using spectroscopic analysis of the supernatant fraction, obtaining a concentrated spectrum using the spectroscopic analysis of the compound- containing fraction, and performing an analysis on the concentrated spectrum of theDocket: 0376-0032WO1 compound-containing fraction and the background spectrum of the supernatant fraction to obtain a spectrum of the compound to characterize a drug product in the bioreactor.

[0025] In examples, the drug product is or includes monoclonal antibodies and / or wherein the supernatant fraction includes an excipient selected from the group consisting of an amino acid, buffer, sugar, surfactant, salt, chelator and any combination thereof and the analysis yields a spectrum of the pure compound or drug product.

[0026] In a current embodiment, the method further includes generating serially diluted samples have different amounts of the compound and the analysis further includes obtaining spectra of the serially diluted samples to be used in the analysis. The analysis can include a principal component analysis on the spectrum of the concentrated preparation and spectra of the serially diluted samples. The method can also include subtracting the background spectrum from the spectrum obtained by the analysis to obtain a spectrum of the pure compound.

[0027] In a current embodiment, the spectra are collected in the wavelength range of 1.4–3 μm, such as in about 2000-2500 nm.

[0028] Typically, the separating includes centrifugation of the sample. The supernatant fraction and the compound-containing fraction are separated in tubes containing a membrane.

[0029] In general, according to one aspect, the invention features a system for characterizing a compound in a sample from a bioreactor. This system comprises a separator for separating the sample into a compound-containing fraction and a supernatant fraction and an analyzer for obtaining a background spectrum using spectroscopic analysis of the supernatant fraction and obtaining a concentrated spectrum using the spectroscopic analysis of the compound-containing fraction and performing an analysis on the concentrated spectrum of the compound-containing fraction and the background spectrum of the supernatant fraction to obtain a spectrum of the compound to characterize a drug product in the bioreactor.

[0030] The above and other features of the invention including various novel details of construction and combinations of parts, and other advantages, will now be more particularly described with reference to the accompanying drawings and pointed out in the claims. It will be understood that the particular method and device embodying theDocket: 0376-0032WO1 invention are shown by way of illustration and not as a limitation of the invention. The principles and features of this invention may be employed in various and numerous embodiments without departing from the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In the accompanying drawings, reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale; emphasis has instead been placed upon illustrating the principles of the invention. Of the drawings:

[0032] FIG. 1 is a schematic diagram illustrating the separation and concentration of a mAb using a centrifugal concentrator;

[0033] FIG. 2 is a schematic diagram illustrating serial dilutions of a concentrated mAb in the buffer separated from the original formulation;

[0034] FIG. 3 shows the result of principal component analysis with two components applied to the spectra of the serially diluted samples of FIG. 2;

[0035] FIG. 4 is a schematic diagram depicting an overall process for characterizing a drug product, a mAb, for instance;

[0036] FIG. 5 is a flow chart of a protocol that can be used to characterize a drug product, e.g., a mAb;

[0037] FIGS. 6, 7 and 8 show the results of an analysis of a commercial unknown mAb in 20 mM histidine buffer using NIST mAb as model mAb;

[0038] FIGS. 9, 10 and 11 show the results of an analysis of a commercial unknown mAb in 20 mM histidine buffer, performed with a characterized model mAb. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039] The invention now will be described more fully hereinafter with reference to the accompanying drawings, in which illustrative embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.Docket: 0376-0032WO1

[0040] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. Further, the singular forms and the articles "a", "an" and "the" are intended to include the plural forms as well, unless expressly stated otherwise. It will be further understood that the terms: includes, comprises, including and / or comprising, when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Further, it will be understood that when an element, including component or subsystem, is referred to and / or shown as being connected or coupled to another element, it can be directly connected or coupled to the other element or intervening elements may be present.

[0041] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0042] Techniques described herein rely on spectroscopic approaches for determining the spectral response of analytes in one or more of the following regions of the electromagnetic spectrum: infrared (IR), visible, and / or ultraviolet (UV). Further, spectroscopic investigations can measure different characteristics, such as absorption spectra, emission (including blackbody or fluorescence) spectra, elastic scattering and reflection spectra, impedance (e.g., index of refraction) spectra, and / or inelastic scattering (e.g., Raman and Compton scattering) spectra.

[0043] For illustration purposes, embodiments of the invention rely on infrared spectroscopy and particularly absorption spectroscopy, which generally covers the near infrared (0.75–1.4 μm, NIR), short-wavelength infrared (1.4–3 μm, SWIR), mid- wavelength infrared (3–8 μm, MWIR), long-wavelength infrared (8–15 μm, LWIR), and / or the far infrared (15–1000 μm, FIR) of the spectrum. In one specific example, the absorption spectroscopy is performed in a range falling in the band of about 2000- 2500 nm. More specifically, the absorption spectroscopy is performed in a range falling in the band of about 2100-2400nm, or even 2150-2350 nm.Docket: 0376-0032WO1

[0044] Probing molecular overtone and combination vibrations, NIR-SWIR spectroscopy covers a region of from 780 nanometer (nm) to 2500 nm of the electromagnetic spectrum. In a shorthand approach, this region between 780 nm to 2500 nm can be simply referred to by the abbreviation “NIR”. An overview of NIR spectroscopy can be found, for example, in an article by A.M.C. Davies in “An Introduction to Near Infrared (NIR) Spectroscopy”, http: / / www.impublications.com / content / introduction-near-infrared-nir-spectroscopy. See also, Cervera, A. E., Petersen, N., Lantz, A. E., Larsen, A. & Gernaey, K. V. “Application of near-infrared spectroscopy for monitoring and control of cell culture and fermentation”, Biotechnol. Prog. 25, 1561–1581 (2009); and Roggo Y, et al., “A review of near infrared spectroscopy and chemometrics in pharmaceutical technologies”, Journal of Pharmaceutical and Biomedical Analysis, Volume 44, Issue 3, 2007.

[0045] The invention generally relates to approaches for detecting and often quantifying compounds (analytes) present in a sample from a bioreactor. Materials that can be investigated include but are not limited to components in culture media, nutrients, metabolites, enzymes, hormones, cytokines, proteins, and so forth. In specific examples, applications of the invention are directed to the analysis of drug product formulations such as those used in vaccines, gene therapy, cell therapy or other applications.

[0046] Of particular interest are monoclonal antibodies (mAb). Since most mAbs are prepared in a complex formulation containing salts, sugars, surfactants, etc.), it is often important to also analyze other constituents present in the formulation, generally referred to herein as “excipients”. These excipients will encompass substances other than the mAb, or, more generally, other than the active drug or compound.

[0047] In the analysis of the formulation, it is often important to obtain a “background” spectrum, namely the spectrum of the formulation without the active compound, e.g., without the mAb in the present example). In the absence of the active compound (e.g., in the absence of the mAb), the remaining material can be thought of as the “solvent” in the formulation; the solvent (also referred to herein as “buffer” or “supernatant”) will typically include the excipients discussed above in a liquid medium, in water, for instance.

[0048] As noted above, the conventional approach of preparing a duplicate solvent cannot reproduce exactly the actual solvent composition in the original formulation. Even slight variations will be reflected in the background spectrum of the duplicate solventDocket: 0376-0032WO1 which will differ from the “ideal” background spectrum, i.e., the spectrum of the actual solvent in the actual formulation being analyzed.

[0049] Rather than relying on preparing solvent duplicates, as in the conventional approach, embodiments of the invention relate to obtaining a background spectrum of the actual solvent in the original formulation. To this end, the mAb is separated from the other materials in the formulation, i.e., the solvent. A spectrum of the separated solvent will accurately measure the excipients in the actual formulation, rather than the excipients in a duplicate.

[0050] The separation can be conducted in a number of ways. An illustrative approach employs a centrifugal device, typically a concentration tube. Under centrifugal forces, the target molecule in a sample is separated and concentrated, while the remaining sample constituents pass through a membrane (also referred to herein as a “filter”) having a suitable pore size for the desired separation. Centrifugal concentrators of various specifications are widely available from different suppliers, such as, for instance, Vivaspin® tubes from Sartorius Stedim Biotech GmbHin, Thermo Fisher - Pierce™ Protein Concentrators PES ( thermofisher.com / order / catalog / product / 88504), Millipore Amicon® (emdmillipore.com / US / en / life-science-research / protein-sample- preparation / protein-concentration / amicon-ultra-centrifugal-filters / 15- ml / Q5ab.qB.pKIAAAFBuLplvyxp,nav), to name a few.

[0051] A schematic illustration is shown in FIG. 1. Sample 10 of the characterization formulation, containing an mAb at a concentration often within a range of about 5 to about 20 milligram per milliliter (mg / mL), for example, is placed in tube 12, configured to hold a specific sample amount, e.g., 100 to 500 microliters (μL). In some implementations, sample 10, used in the characterization procedure described herein, is optimized to avoid high molecular weight species.

[0052] Polysorbate (PS), for example, can have a molecular weight of ~76 kDa. Using a filter with a molecular weight cutoff (MWCO) of 50 kDa would result in the PS getting trapped and thus concentrated as well. Selecting a MWCO of the spin filter that suits the formulation can address this issue in some cases. However, it was discovered that even with filters that are 100 kDa MWCO, PS80 still appears to get trapped. Accordingly, performing the characterization on samples that are free of PS20 / 80, or any larger molecules for that matter, entirely circumvents this problem. For similar reasons, inDocket: 0376-0032WO1 specific implementations, the sample employed does not contain proteins other than the one being characterized.

[0053] Sample 10 is centrifuged (represented by the arrow), using a suitable machine, such as Beckman Coulter (Allegra X-30 or Microfuge 20), and yields a concentrated mAb fraction (pool) 14 at an upper region 16 of the tube. Due to the filter characteristics, the other constituents in the sample, i.e., the solvent, represented in this case by the supernatant, pass through the filter and collect as fraction (pool) 18 at a bottom section 20 of tube 12. With buffer components flowing freely through the filter, their concentration in the mAb concentrated fraction 14 and in the mAb-free fraction 18 will be the same.

[0054] In some cases, the filter is flushed and spun down, with deionized water (DI), for instance, to remove unwanted species such as glycerin, sodium azide, etc. that might be present in the membrane.

[0055] With the supernatant representing the exact solvent in the original formulation, its spectrum is the accurate (“ideal”) background spectrum of the sample and can be subtracted from a spectrum of the sample, thus providing an accurate spectroscopic fingerprint of the pure mAb.

[0056] Scanning can be conducted using any suitable equipment. One embodiment of the invention employs apparatus or analyzer 11, having a "clamshell" design, in which the apparatus opens to receive the sample and then closes over the sample to perform the analysis. An example is described in incorporated application International Patent Application No. PCT / US2023 / 79110. The apparatus can be constructed from two sections, halves, parts or portions, at least one section being configured for opening and closing in clamshell (or flip) fashion. The movement can be realized through a connection that joins the two sections directly or via an intermediary support or base. The analyzer 11 further includes computing resources such as a computer for analyzing the spectral responses obtained.

[0057] In specific embodiments, some components are arranged in a first section, often the lower or bottom part, that is fixed onto a support. A second section, often the upper, top or “lid”-like part houses other components and is configured to move or “flip” between an open and closed position. During operation, the apparatus can be placed in the open position to introduce a sample, for example. Closing the second (e.g., upper) section brings the two sections together, forming a sample detection region, also referred to as aDocket: 0376-0032WO1 sample “gap”. Absorption spectra of sample analytes present in the sample detection region can be obtained while the apparatus is in the closed configuration.

[0058] The sample gap can be formed between rods that transmit light in the desired region of the electromagnetic spectrum, e.g., NIR or MIR. Some approaches employ rods designed and / or oriented to reduce or minimize reflections / etalons. In specific implementations, the gap (and thus also the light pathway) has fixed dimensions. In one example, the gap is set and maintained at a fixed pathlength for a given measurement. Some construction details allow for resetting the pathlength as a new or different sample is analyzed. The pathlength (gap) can have a value within a range of from about 0.010 millimeters (mm) to about 10 mm. In illustrative implementations, the pathlength has a value within a range of from about 0.010 mm to about 5 mm.

[0059] For many arrangements, the first (e.g., lower) section houses elements for directing light, typically from a light source, e.g., a laser, to the sample gap, while the second (e.g., lid-like) section houses elements for detecting the light once it has traversed the sample gap.

[0060] In some embodiments, the apparatus is part of a system that also includes a laser, a tunable laser, for example. Thus, in one of its aspects, the invention features a system that includes a clamshell apparatus and a tunable laser for generating a swept wavelength signal. A first section of the apparatus includes components for directing light from the laser to a sample detection region; a second section includes a photodetector for detecting the swept wavelength signal after transmission through the sample detection region. For specific implementations, the first section further includes a detector for detecting the swept wavelength signal prior to transmission through the sample.

[0061] In some cases, the tunable laser is configured for a specific wavelength range which contains relevant, protein-critical chemical information (C-H, O-H, etc.). For a laser that is external to the clamshell apparatus, the system can employ fiber optic technology to transmit light to the first section of the apparatus. In one example, the tunable laser sweeps in range falling in the band of about 2000-2500 nm. More specifically, the tunable laser sweeps in a range falling in the band of about 2100-2400nm, or even 2150-2350 nm.

[0062] A method for analyzing the supernatant or the sample comprises generating a swept wavelength signal, transmitting the swept wavelength signal through a first section of a clamshell apparatus, to and though a sample detection region; detecting the sweptDocket: 0376-0032WO1 wavelength signal after transmission through the sample detection region in a second section of the clamshell apparatus, and resolving an absorption spectrum of the sample. These steps are conducted with the apparatus in a closed configuration. Loading the sample takes place in the open configuration.

[0063] The method can also include detecting the swept wavelength signal prior to transmission through the sample detection region and resolving an absorption spectrum of the sample with reference to the swept wavelength signal before and after transmission through the sample detection region.

[0064] At least some of the system operations and / or data analysis can be controlled by a controller.

[0065] Additional details describing configurations and techniques pertaining to clamshell arrangements can be found in International Patent Application No. PCT / US2023 / 79110, filed on November 8, 2023, and U.S. Provisional Patent Application No. 63 / 512,996, filed on July 11, 2023.

[0066] Some aspects of the invention rely on

[0067] In one approach, the supernatant 18 (e.g., from the bottom section 20 of tube 12 in FIG. 1) and possibly the concentrated mAb (e.g., from the top region of 12 of tube 10 in FIG. 1) are used to prepare a set of diluted samples, as illustrated in FIG. 2. For instance, the highest concentration of the mAb in Cal 1 can be diluted with the actual solvent that was separated from the original formulation, e.g., supernatant 18 in FIG. 1, to prepare diluted samples, Cal 2, Cal 3, Cal 4 and Cal 5, for example. In one illustration, a specified volume of supernatant 18 is added to the most concentrated Cal 1, followed by mixing to prepare Cal 2. A subsequent addition and mixing of the specified supernatant volume with contents of Cal 2 will form Cal 3. The dilution process can be repeated several times (as determined by routine experimentation, prior experience or other factors) to generate a suitable sample set for PCA. Many situations will require 3 to 5 serially diluted samples.Docket: 0376-0032WO1

[0068] FIG 3 illustrates the PCA process (represented by the arrow). The plots (A) are absorbance spectra as a function of wavelength of the serially diluted samples Cal 1 (uppermost plot) through Cal 5 (the bottom plot).

[0069] Performing PCA on spectra of the serially diluted samples (Cal 1 through Cal 5; plots (A) in FIG. 3, for example) with two components generates two spectra. The first spectrum corresponds to the PCA of Component 1, theoretically the mAb (or, in further examples, of another protein, etc.), without the “noise” caused by other ingredients that might also be present. In the specific illustration of FIG. 3, plot (B) is the first component from the PCA analysis, deemed the “pure” mAb spectrum.

[0070] The PCA of Component 2 (plot (C) in the example of FIG. 3) can inform on how perfect the background was. For example, plot (C) could indicate if some other substance (e.g., PS80) was diluted along with the compound of interest, the mAb, in this example. Information provided by the second component might be used to adjust future spinning parameters or redo the entire protocol to improve the separation of Compound 1 (the mAb, in this example) from other ingredients.

[0071] The overall process is further illustrated with reference to FIGS. 4 and 5. Illustrative materials and equipment that can be utilized include: a Vivaspin® tube 500 50kda MWCO; ultra-pure water such as the Millipore Milli-Q® water; pipettes and tips (P1000, P200, P20); a centrifuge such as Beckman Coulter (Allegra X-30 or Microfuge 20); six Eppendorf® tubes, labeled 1, 2, 3, 4, 5, buffer; DI water for the flush / spin the filer prior to use, to remove any glycerin / sodium azide present in the membrane; clamshell analyzer. The mAb characterization sample is free of PS 20 / 80 and is provided in a volume of at least 100 to 500 μL, in a volume larger than 250 μL, in one example.

[0072] As shown in FIG. 4, an initial, characterization sample 10 contains a drug product 50, such as a mAb, a buffer component 52 (e.g., a sugar) and a buffer component 54 (e.g., an amino acid). In the initial sample, the concentration of the drug product 50 (e.g., mAb) is C50(1); the concentration of buffer component 52 is C52(1); and the concentration of buffer component 54 is C54(1). Separation 60 (by centrifuging through a suitable filter, for example) produces concentrated fraction 14 and supernatant 18 (see FIG. 1). In the concentrated fraction 14, the concentration of the drug product 50 is C550(2); the concentration of the buffer component 52 is C52(2); and the concentration of the bufferDocket: 0376-0032WO1 component 54 is C54(2). The component concentrations in fraction 14 relative to concentrations in the initial sample are: C50(2) > C50(1); C52(2) = C52(1); and C54(2) = C54(1), while in the supernatant 18: C50(2) = 0; C52(2) = C52(1); and C54(2) = C54(1).

[0073] Scanning (arrow 62) the supernatant 18 (obtained after the separation) generates a background spectrum 70 (Scan 0 in FIG. 1). Scanning (arrow 64) the concentrated fraction 14 (Cal 1 in FIG. 2) and serially diluted samples (e.g., Cal 2 through Cal 5 in FIG. 2) produces spectrum 72 of the concentrated drug product 50 (in the presence of buffer components C52 and C54). Spectrum 70 of the supernatant (Scan 0 in FIG. 1) is subtracted from the spectrum 72, to obtain the spectrum 74 of drug product 50.

[0074] A flow diagram for overall protocol 100 for characterizing a mAb can include the steps shown in FIG. 5. In a first step 102, an initial sample volume (a volume of at least 100 - 500 μL, e.g., a volume larger than 250 μL), containing the mAb, is transferred into a tube such as tube 12 in FIG. 1. In preferred implementations, the sample is free of PS20 / 80, proteins other than the one being characterized, and / or any other high molecular weight species.

[0075] Step 104 involves a centrifuge operation, while step 106 ascertains whether the sample did indeed separate into two pools (fractions). If the separation is found to be unsuccessful, step 104 is repeated. Otherwise, content from the tube section above the filter, namely from the fraction containing the highest level of mAb, is transferred to a first container, e.g., Eppendorf® tube labeled “1”, in step 108.

[0076] In step 110, content from the tube fraction formed below the filter, namely from the supernatant, is transferred to a buffer container, e.g., Eppendorf® tube labeled “buffer”.

[0077] Eppendorf® tube “1”, considered the initial mAb concentration “i” (stage 112), is subjected to serial dilutions, iterations “i”+1, using the supernatant, e.g., from theDocket: 0376-0032WO1 “buffer” Eppendorf® tube. In one illustration, “i” is diluted into “i+1” by adding 20 μL from the “buffer” tube to contents of Eppendorf® tube labeled “1”, followed by mixing.

[0078] The resulting solution can be further diluted by adding and mixing in another 20 μL from the “buffer” container and the dilution procedure is repeated for a desired number of times. As an illustration, loop 114 inquires whether “i” is ≤ 4. If the answer is “no”, serial dilutions continue. With a “yes” answer, the protocol proceeds to step 116, which involves scanning the contents of the five Eppendorf® tubes (containing the different concentrations of mAb) to obtain spectra such as shown in FIG.3 (plots A).

[0079] Step 118 involves the PCA on the spectra of Eppendorf® tubes labeled 1-5 by the analyzer 11. As already discussed, the buffer is scanned (step 120 in protocol 100) to generate the background spectrum, which is then subtracted by the analyzer 11 before the dilutions are carried out. Thus, the spectra fed to the PCA (step 120) executed by the analyzer 11 are spectra obtained once the buffer has been subtracted.

[0080] Moreover, on an ongoing basis as an extension of step 118, the background spectrum is used as part of the continuing monitoring of the bioreactor. That is, this background spectra is reused to analyze the bioreactor’s progress over time by simply subtracting the background spectra from each new spectra of each new sample extracted from the bioreactor to obtain a spectrum of the compound of interest and thereby characterize the progress of a drug product being synthesized in the bioreactor.

[0081] In many cases, a typical scan can be performed in as little as 30 seconds, with the entire protocol 100 taking about 30 minutes.

[0082] Once the mAb has been characterized by the analyzer 11, it can be used as a characterized model in the analysis of complex samples, samples that contain the mAb in the presence of any number of excipients (including, but not limited to polysorbate, histidine, other amino acids, sugars, surfactants, salts, chelators etc.). Thus, in further implementations, the spectrum of the pure mAb, obtained by the analyzer 11 as described above, is used in the analysis of a complex sample containing the mAb by the analyzer 11, thus eliminating the inaccuracies associated with using the NIST mAb model.

[0083] The invention is further illustrated by the following non-limiting examples. Example 1Docket: 0376-0032WO1

[0084] Pure and serial dilutions of a commercial unknown mAb were prepared in a histidine buffer. The mAb content increased from 10 to 100 mg / mL; the amount of total histidine in each sample was 20 mM. Total histidine represents the sum of histidine and histidine HCl; each of histidine and histidine chloride can be measured individually, applying techniques described above. Subset Samples 1, 3 and 5, which had, respectively, 100, 50, 25 mg / mL mAb in 20 mM hist were analyzed using the NIST mAb as model mAb. Results are shown in FIGS. 6, 7 and 8, each of the figures presenting a series of spectra. In each case, the upper left plot shows the spectrum of the sample; the upper right plot shows the sample residual – library fit; the lower left plots show the histidine residual; and plots in the lower right show the spectra of histidine chloride residual.

[0085] As mAb concentration decreases (from Sample 1 to Sample 3, then to Sample 5), the impact of incorrect spectra decreases as well and the histidine measurements became more accurate with lower mAb concentrations. Example 2

[0086] Samples were prepared as in Example 1 above but the analysis was performed using the characterized mAb model obtained according to embodiments of the invention.

[0087] The results for selected Samples 1, 3 and 4 are shown in FIGS. 9, 10 and 11. As in Example 1, each of the figures presents a series of spectra. In each case, the upper left plot shows the spectrum of the sample; the upper right plot shows the sample residual – library fit; the lower left plots show the histidine residual; and plots in the lower right show the spectra of histidine chloride residual.

[0088] Thus, as seen in Examples 1 and 2 and FIGS. 6-11, when using the NIST mAb as a model at high mAb concentrations (such as, for instance, 100 mg / mL), the NXT SW cannot detect the correct amount of histidine, which should be total 20 mM. In contrast, using the PCA mAb from the protocol disclosed herein accounts for the signal perfectly (Fig 9). Using the properly characterized mAb can result in more accurate measurements of the mAb change, while the total histidine remained constant (normal + HCl) at around 20 mM.

[0089] While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art thatDocket: 0376-0032WO1 various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.

Claims

Docket: 0376-0032WO1 CLAIMS What is claimed is:

1. A method for characterizing a compound in a sample from a bioreactor, the method comprising: separating the sample into a compound-containing fraction and a supernatant fraction; obtaining a background spectrum using spectroscopic analysis of the supernatant fraction; obtaining a concentrated spectrum using the spectroscopic analysis of the compound-containing fraction; and performing an analysis on the concentrated spectrum of the compound- containing fraction and the background spectrum of the supernatant fraction to obtain a spectrum of the compound to characterize a drug product in the bioreactor.

2. The method of claim 1, wherein the drug product is or includes monoclonal antibodies and / or wherein the supernatant fraction includes an excipient selected from the group consisting of an amino acid, buffer, sugar, surfactant, salt, chelator and any combination thereof.

3. The method of either of claims 1 or 2, wherein the analysis yields a spectrum of the pure compound.

4. The method of any of claims 1-3, further comprising generating serially diluted samples have different amounts of the compound and the analysis further includes obtaining spectra of the serially diluted samples to be used in the analysis.

5. The method of claim 4, wherein the analysis includes a principal component analysis on the spectrum of the concentrated preparation and spectra of the serially diluted samples.

6. The method of any of claims 1-5, further comprising subtracting the background spectrum from the spectrum obtained by the analysis to obtain a spectrum of the pure compound.Docket: 0376-0032WO1 7. The method of any of claims 1-6, wherein the spectra are collected in the wavelength range of 1.4–3 μm.

8. The method of any of claims 1-7, wherein the spectra are collected in the wavelength range of about 2000- 2500 nm.

9. The method of any of claims 1-8, wherein the separating includes centrifugation of the sample.

10. The method of any of claims 1-8, wherein the supernatant fraction and the compound-containing fraction are separated in tubes containing a membrane.

11. A system for characterizing a compound in a sample from a bioreactor, the system comprising: a separator for separating the sample into a compound-containing fraction and a supernatant fraction; and an analyzer for obtaining a background spectrum using spectroscopic analysis of the supernatant fraction and obtaining a concentrated spectrum using the spectroscopic analysis of the compound-containing fraction and performing an analysis on the concentrated spectrum of the compound-containing fraction and the background spectrum of the supernatant fraction to obtain a spectrum of the compound to characterize a drug product in the bioreactor.

12. The system of claim 11, wherein the drug product is or includes monoclonal antibodies and / or wherein the supernatant fraction includes an excipient selected from the group consisting of an amino acid, buffer, sugar, surfactant, salt, chelator and any combination thereof.

13. The system of either of claims 11 or 12, wherein the analysis by the analyzer yields a spectrum of the pure compound.

14. The system of any of claims 11-13, wherein the separator generates serially diluted samples having different amounts of the compound and the analyzer further obtaines spectra of the serially diluted samples to be used in the analysis.Docket: 0376-0032WO1 15. The system of claim 14, wherein the analyzer performs a principal component analysis on the spectrum of the concentrated preparation and spectra of the serially diluted samples.

16. The system of any of claims 11-15, wherein the analyzer subtracts the background spectrum from the spectrum obtained by the analysis to obtain a spectrum of the pure compound.

17. The system of any of claims 11-16, wherein the analyzer collects spectra in the wavelength range of 1.4–3 μm.

18. The system of any of claims 11-17, wherein the analyzer collects spectra in the wavelength range of about 2000- 2500 nm.

19. The system of any of claims 11-18, wherein the separator is a centrifuge.

20. The system of any of claims 11-18, wherein the supernatant fraction and the compound-containing fraction are separated in tubes containing a membrane.

Images