CN116615232A - Method for producing biotherapeutic agent - Google Patents

Abstract

Methods of making biologic therapeutic agents are described. The methods may include detecting the level of a molecular property of the biologic therapeutic in the formulation, determining the rate of change of the molecular property under storage conditions, and estimating the level of molecular property exposure received by the subject at the time of said administering. Production lots of biotherapeutic agents containing molecular properties may be manufactured based on estimated levels of molecular property exposure, with the production lots containing molecular properties at or below specified specifications for allowable levels of the molecular properties. Methods of developing manufacturing processes for biologic therapeutic agents are described. Methods of assessing the clinical impact of molecular properties of a biologic therapeutic are described.

Description

Method for producing biotherapeutic agent

Cross Reference to Related Applications

The present application claims the benefit of U.S. provisional application No. 63/126,274, entitled "Methods of Manufacturing Biological Therapies [ method of manufacturing a biotherapeutic agent ]" filed on month 12 and 16 of 2020, and U.S. provisional application No. 63/242,395, entitled "Methods of Manufacturing Biological Therapies [ method of manufacturing a biotherapeutic agent ]" filed on month 9 of 2021, each of which is incorporated herein by reference in its entirety.

Technical Field

Embodiments herein relate to methods of manufacturing a biologic therapeutic and molecular property levels of biologic therapeutic.

Background

The natural structure or chemical nature of a biological molecule (such as a therapeutic protein) adapts or changes in response to changes in the molecular environment. Other biological therapeutic agents, including nucleic acid and cell-based therapeutic agents, may also undergo changes within their environment. While this structural or chemical flexibility is required for the biological function of most, if not all, biomolecules and cells, it also presents a number of challenges during the development and manufacture of biotherapeutic agents for pharmaceutical applications. For example, therapeutic proteins are subjected to various conditions in a number of process steps leading to administration to a patient. These many process steps include, for example, one or more of the following: protein production (e.g., recombinant production), harvesting, purification, formulation, filling, packaging, storage, delivery, and final preparation immediately prior to administration to a patient. In each of these steps, the therapeutic protein is placed in one or more environments that may or may not result in a change in its structure or chemical properties. The change in structure or chemical properties may result in the formation of different types of biotherapeutic agents, thereby yielding heterogeneous products. While some species retain their ability to bind to their targets and thus maintain therapeutic efficacy, others lose the target binding ability and thus become functionally inactive. To maximize and maintain quality control of these biotherapeutic agents, the biopharmaceutical industry has concentrated a great deal of effort to understand why some species lose activity while others remain active.

Molecular attributes define the physicochemical properties of therapeutic biomolecules and thus may affect the safety and efficacy of a drug. The level of attributes critical to drug quality, or key quality attributes (CQAs), are well defined by product purity specifications subject to extensive regulatory scrutiny.

Disclosure of Invention

According to some embodiments, a method of manufacturing a biologic therapeutic is described. The method may comprise detecting the level of the molecular property of the biotherapeutic agent in the formulation at one or more time points under storage conditions. The method may include determining a rate of change of the molecular property under the storage conditions. The method may include obtaining data regarding the in vivo safety and/or efficacy of the biologic therapeutic in a subject who has received administration of the biologic therapeutic. The method may include estimating the level of molecular property exposure that the subjects receive at the time of administration of the biotherapeutic agent based on (i) the rate of change of the molecular property during storage of the biotherapeutic agent in the formulation and (ii) the duration of time the biotherapeutic agent in the formulation is in the storage conditions prior to administration of the biotherapeutic agent. The method may include determining whether a correlation exists between the estimated level of molecular property exposure and safety and/or efficacy data of the biologic therapeutic. If the correlation is not present, the method may include manufacturing a production lot of the biologic therapeutic comprising the molecular property at or below a specified allowable level of the molecular property based on the estimated level of molecular property exposure. If the correlation is present, the method may further comprise setting a specification of the molecular property level that does not exceed a maximum allowable level of the molecular property of the biotherapeutic agent at the time of manufacture. The maximum allowable level of the molecular property may be based on a highest estimated level of exposure of the molecular property that is not associated with adverse events and/or efficacy inhibition of the biotherapeutic agent. The manufacturing may further include rejecting the production lot of the biologic therapeutic that contains the molecular property at levels exceeding the maximum allowable levels. In some cases, a specified maximum allowable level of the molecular property is calculated to produce a property exposure level that is less than or equal to 90% -100% of a highest estimated level of the molecular property exposure that is not related to adverse events and/or efficacy inhibition in the subject at the end of the shelf-life.

According to some embodiments, a method of developing a manufacturing process for a biologic therapeutic is described. The method may comprise detecting the level of the molecular property of the biotherapeutic agent in the formulation at one or more time points under storage conditions. The method may include determining a rate of change of the molecular property under the storage conditions. The method may include obtaining data regarding the safety and/or efficacy of the biologic therapeutic in a subject that has received administration of the biologic therapeutic. The method may include estimating the level of molecular property exposure that the subjects receive at the time of administration of the biotherapeutic agent based on (i) the rate of change of the molecular property during storage of the biotherapeutic agent in the formulation and (ii) the duration of time the biotherapeutic agent in the formulation is in the storage conditions prior to administration of the biotherapeutic agent. The method may include determining whether a correlation exists between the estimated level of molecular property exposure and safety and/or efficacy data of the biologic therapeutic. If (a) the correlation is not present, the method may include establishing the manufacturing process to produce the molecular property level at or below a specified allowable level based on the estimated level of molecular property exposure. Alternatively, if (b) the correlation exists, the method may comprise establishing the manufacturing process based on a highest level of the molecular property that is not associated with adverse events and/or efficacy inhibition of the biotherapeutic agent to produce the molecular property level at or below a specified maximum allowable level of the molecular property.

According to some embodiments, a method of assessing the clinical impact of molecular properties of a biologic therapeutic is described. The method may comprise detecting the level of the molecular property of the biotherapeutic agent in the formulation at one or more time points under storage conditions. The method may include determining a rate of change of the molecular property under the storage conditions. The method may include obtaining data regarding the safety and/or efficacy of the biologic therapeutic in a subject that has received administration of the biologic therapeutic. The method may comprise estimating the level of exposure of the molecular property that the subjects receive at the time of administration of the biologic therapeutic based on (i) the rate of change of the molecular property during storage of the biologic therapeutic in the formulation and (ii) the duration of time the biologic therapeutic in the formulation is in the storage conditions prior to said administration. The method may include determining whether a correlation exists between the estimated molecular property exposure and the safety and/or efficacy of the biotherapeutic agent. If (a) the correlation is not present, the method may include determining that the molecular property does not affect neither the clinical safety nor the efficacy of the biologic therapeutic. Alternatively, if (b) the correlation exists, the method may comprise determining that the molecular property would affect the safety and/or efficacy of the biologic therapeutic. In some cases, if (a) the correlation is not present, the method may further comprise setting a specification of allowable levels of the molecular property of the biologic therapeutic, wherein the allowable levels of the molecular property are based on the highest estimated levels of molecular property exposure accepted by the subjects. Alternatively, if (b) the correlation exists, the method may further comprise setting a specification of a maximum allowable level of the molecular property of the biologic therapeutic, wherein the maximum allowable level of the molecular property is based on the level of the molecular property associated with adverse events and/or efficacy inhibition of the biologic therapeutic.

For any of the methods described herein, the estimation may be further based on (iii) the dose of the biologic therapeutic in the administration, and (iv) the amount of molecular property measured at the time of manufacture and/or batch placement.

For any of the methods described herein, the level of the molecular property of the biologic therapeutic in the formulation can be detected at two or more time points under storage conditions.

For any of the methods described herein, the one or more points in time or two or more points in time may include at the time of manufacture and at least two subsequent points in time.

For any of the methods described herein, estimating the level of exposure of the molecular property may include calculating:

wherein A is _t Is this estimated level of molecular property exposure,% A ₀ Is the percentage of the molecular property,% A, at batch placement _Δi Is the rate of change, t, of the percentage of the molecular property level over time under given storage conditions _i Is the time stored under the given conditions, and D is the dose strength of the administration.

For any of the methods described herein, estimating the level of exposure of the molecular property may include calculating:

wherein% _rel. Is the percentage of the relative level of molecular property exposure to dose, and wherein A _t Is the level of exposure of this molecular property calculated using equation 1 or 2, and where D is the dose intensity in the weight dimension of the active pharmaceutical ingredient associated with each treatment.

For any of the methods described herein, the correlation may include a weighted correlation between the attribute exposure and the occurrence of the adverse event.

For any of the methods described herein, determining whether there is a correlation between the estimated level of molecular property exposure and the safety and/or efficacy data of the biotherapeutic agent may include bayesian estimation (Bayesian estimation).

For any of the methods described herein, the biologic therapeutic in the formulation may have been in the storage conditions for different durations in the administrations to different subjects.

For any of the methods described herein, administration (of the biologic therapeutic) can include two or more administration events. In some cases, the estimated level of molecular property exposure that a subject receives at the time of administration (of the biologic therapeutic) may be the maximum or average of the two or more administration events.

For any of the methods described herein, administration (of the biologic therapeutic) may include continuous infusion, and the estimating may include calculating an estimated level of molecular property exposure during two or more intervals of continuous infusion, such as 24 hour intervals.

For any of the methods described herein, the security data may include adverse event data. In some cases, the security data may include adverse event time courses.

For any of the methods described herein, the efficacy data may include clinical endpoint data.

For any of the methods described herein, the molecular property may include at least one of: acidic species, basic species, high molecular weight species, sub-visible particle numbers, low molecular weight, medium molecular weight, glycosylation (such as non-glycosylated heavy or high mannose), non-heavy and light chains, deamidation, deamination, cyclization, oxidation, isomerization, fragmentation/scission, N-and C-terminal variants, reducing species and partial species, folding structures, surface hydrophobicity, chemical modifications, covalent bonds, C-terminal amino acid motif PARG or C-terminal amino acid motif PAR-amides. For any of the methods described herein, the molecular property may include at least one of: acidic species, basic species, high molecular weight species, amino acid isomers or sub-visible particle numbers.

For any of the methods described herein, the biologic therapeutic may be selected from the group consisting of: antibodies, antigen-binding antibody fragments, antibody protein products, bispecific T cell conjugates Molecules, bispecific antibodies, trispecific antibodies, fc fusion proteins, recombinant viruses, recombinant T cells, synthetic peptides, and active fragments of recombinant proteins.

For any of the methods described herein, the formulation may be a pharmaceutically acceptable formulation. For any of the methods described herein, the subject (or patient) may be a human subject (or patient).

For any of the methods described herein, detecting the level of molecular properties of the biological therapeutic agent can include mass spectrometry, chromatography, electrophoresis, spectroscopy, photoresistance, particle methods (such as nanoparticle/visible/micrometer-scale resonance mass or brownian motion), analytical centrifugation, imaging or imaging characterization, or immunoassay.

Drawings

Fig. 1A-C are a series of diagrams illustrating components of a computing method and a statistical method for determining clinical impact of an attribute (Clinical Impact of Attributes, CIA) according to embodiments herein.

Fig. 2A-B are a series of graphs illustrating the effect of properties of mAb a (fig. 1A) and mAb B (fig. 1B) on ADA according to embodiments herein. In each of the graphs of fig. 2B, the results for ADA-negative patients are shown on the left side, and the results for ADA-positive patients are shown on the right side.

FIGS. 3A-B are CIA assays illustrating the incidence of ADA attributes for mAb A (FIG. 3A) and mAb B (FIG. 3B) according to embodiments herein (subjects according to representative A) _t Grouping) a series of graphs.

FIGS. 4A-B are CIA assays illustrating the ADA attribute time course for mAb A (FIG. 4A) and mAb B (FIG. 4B) according to embodiments herein (subjects according to representative A) _t Grouping) a series of graphs.

FIGS. 5A-C are graphs illustrating ADA response of mAb B versus representative A _t A series of graphs of associations between (attribute exposures at the time of treatment administration). In each of the graphs of fig. 5A, the results of all treatments for the ADA positive patients are shown on the left side, the results of all treatments for the ADA negative patients are shown in the center, and the results of all treatments in the ADA-related treatments for the ADA positive patients are shown on the right side.

Fig. 6 is a series of graphs illustrating the effect of factors in treatment on ADA. In each of the graphs of fig. 6, the results for ADA-negative patients are shown on the left side, and the results for ADA-positive patients are shown on the right side.

Fig. 7A-B are a series of graphs illustrating the incidence of CIA analysis on subjects grouped according to review of known clinical outcomes (fig. 7A) and subjects grouped according to representative At (fig. 7B) to determine the effect of mAb a's attributes on ADA. In each of the graphs of fig. 7A, the results for ADA-negative patients are shown on the left side, and the results for ADA-positive patients are shown on the right side.

Fig. 8 is a graph illustrating the effect of a CIA analysis on subjects grouped according to retrospective known clinical outcome using an advanced clinical study to determine mAb a attributes on ADA.

FIG. 9 is a schematic diagram illustrating a method according to representative A _t Analysis in grouped subjects as representative A _t Average or maximum A of (2) _t To determine the profile of the effect of mAb B properties on ADA.

FIGS. 10A-B show product C (a specificationMolecules) and efficacy on 6 adverse events. In each graph of fig. 10A, for each X-axis value class (0,>0-15 or>15 Patients without adverse events are shown on the left and patients with adverse events are shown on the right. In each graph of fig. 10B, for each X-axis value class (0,>0-15 or>15 Patients with no adverse events or no more than 2 adverse events on the left and no more than 3 adverse events on the right).

FIGS. 11A-D are a series of graphs showing the effect of analyzing High Molecular Weight (HMW) species on heat generation using the method of adverse Bayesian estimation as described herein (FIGS. 11A-B) and the method of Bayesian estimation as described herein (FIGS. 11C-D).

Detailed Description

Described herein are methods of making a biologic therapeutic (such as a therapeutic protein, nucleic acid, or cell-containing therapeutic), wherein the safety and efficacy of the biologic therapeutic is controlled by limiting the level of exposure of an attribute of an individual. After manufacturing a production lot of the biotherapeutic agent, the biotherapeutic agent will be stored in the formulation for a period of time before being administered to the subject. The molecular property level of the biotherapeutic agent may vary during storage. In addition, the molecular property level may vary from production lot to production lot, for example reflecting the difference in molecular property level between the initial cell culture and the final drug product of the biologic therapeutic at different production stages. For example, the level of a property such as an acidic substance, a basic substance, a high molecular weight substance, an amino acid isomer, or a sub-visible particle may be increased. Such attributes may lead to reduced efficacy of the biologic therapeutic, and/or may lead to adverse events in a subject receiving the biologic therapeutic. As described herein, changes in molecular property levels in storage can be modeled. Based on the level of molecular property at the time of manufacturing the production lot of the biotherapeutic agent, the amount of time the biotherapeutic agent in the formulation is stored prior to administration to the subject, the rate of change of molecular property in storage and the dosage of the biotherapeutic agent, the level of molecular property at the time of administration to the subject can be calculated. Furthermore, if the actual level of a molecular property is not correlated with an adverse event or loss of efficacy at the time of administration, then the molecular property level may be determined to be safe and effective. Thus, using the methods described herein, a production lot of a biologic therapeutic at or below a specified level can be produced based on a level that is deemed safe and effective at the time of administration.

It is contemplated that although the criticality and manufacturing specifications of molecular properties have been determined using conventional methods, the criticality and specifications of conventionally determined molecular properties may hardly exhibit clinical relevance. The criticality is often studied using non-human model systems, and specification limitations reflect low levels of attributes that can be reasonably achieved at the time of manufacture and storage. Alternatively, in a prior knowledge or clinical experience approach, it is not critical to suggest a property when other pharmaceutical products containing that property rarely observe clinical outcome. However, this approach ignores possible product-specific variability in attribute impact. Also, adverse events can be caused by a certain attribute, and if lots containing high enough levels of attributes to cause such events are not frequently distributed in the clinic due to lot-to-lot variability, the adverse events still rarely occur.

The methods described herein can utilize data analysis methods that verify whether there is a correlation between the estimated actual level of patient exposure to a given attribute and the extent to which clinical outcome occurs by analyzing data from clinical studies and product quality analysis studies. This method may be referred to as Clinical Impact of Attributes (CIA). The methods described herein utilize actual clinical data and assessment of the level of exposure of the attribute at the time of administration of the biologic therapeutic. These methods provide a true assessment of molecular property effects, overcoming the shortcomings of conventional methods that have utilized non-human model systems to determine property levels and do not consider property exposure at the time of administration.

Molecular Properties

"molecular properties" and variants of the root term have their ordinary and customary meaning as will be understood by those of ordinary skill in the art in light of this disclosure. "molecular property" refers to a chemically or physically altered structure on a macromolecule (such as a protein or nucleic acid) and can be characterized in terms of its chemical identity or type of property, as well as the position in the sequence of the macromolecule (e.g., the amino acid position on which the property resides). For example, asparagine and glutamine residues are prone to deamidation. Deamidated asparagine at position 10 of the therapeutic protein amino acid sequence is an example of a property. Exemplary molecular property types are described herein. For brevity, molecular properties may be referred to herein simply as "properties". The level of a drug quality critical attribute or Critical Quality Attribute (CQA) may be well defined by the product purity specification. These specifications are typically subject to extensive regulatory scrutiny. In some embodiments, the specification may set an allowable level of one or more molecular attributes in the manufacture of the biologic therapeutic.

In some embodiments, the molecular properties include or consist of one or more of the following: acidic species, basic species, high molecular weight species, sub-visible particle numbers, low molecular weight, medium molecular weight, glycosylation (such as non-glycosylated heavy or high mannose), non-heavy and light chains, deamidation, deamination, cyclization, oxidation, isomerization, fragmentation/scission, N-and C-terminal variants, reducing species and partial species, folding structures, surface hydrophobicity, chemical modifications, covalent bonds, C-terminal amino acid motif PARG or C-terminal amino acid motif PAR-amides.

PARG is an alternative C-terminal variant of an antibody, which may occur due to alternative splicing. It represents 4 amino acids (proline, alanine, arginine, glycine) in which "AR" is genetically inserted into the canonical IgG 2C-terminal sequence. PAR-amides are another C-terminal variant resulting from further processing of PARG. It refers to cleavage of the C-terminal glycine from an antibody ending in PARG, leaving an amide group on the C-terminal arginine.

In some embodiments, the molecular property comprises or consists of at least one of: acidic species, basic species, high molecular weight species, amino acid isomers or sub-visible particle numbers.

Techniques for detecting molecular property levels

Any suitable analytical technique for detecting molecular properties may be used with the methods described herein. Techniques for detecting molecular properties include, but are not limited to, mass spectrometry, chromatography, electrophoresis, spectroscopy, photoresistance, particle methods (nanoparticle/visible/micrometer-scale resonance mass or brownian motion), analytical centrifugation, imaging and imaging characterization, and immunoassays.

Exemplary techniques for detecting molecular properties include reduced and non-reduced peptide mapping (which may detect chemical modifications), chromatography (such as Size Exclusion Chromatography (SEC), ion exchange chromatography (IEX) such as cation exchange Chromatography (CEX), hydrophobic Interaction Chromatography (HIC), affinity chromatography such as protein a-column chromatography or Reversed Phase (RP) chromatography, capillary isoelectric focusing (cif), capillary Zone Electrophoresis (CZE), free Flow Fractionation (FFF) or Ultracentrifugation (UC), HIAC (such as for detecting sub-visible particle count), MFI (such as for detecting sub-visible particle count and morphology), visual inspection (visible particles), SDS-PAGE (such as for detecting fragments, covalent aggregates), color analysis (Trp Ox), rCE-SDS and nrCE-SDS (such as for detecting fragments as part of molecules), nanoparticle size determination, spectroscopy (such as FTIR, CD, intrinsic fluorescence or ANS dye binding), ellman assay (free sulfhydryl), hils, hilc (such as for detecting MALS), and HCP-map (such as for detecting MALS).

Biological therapeutic agent

As used herein, "biotherapeutic agent" and variants of this term have their ordinary and customary meaning as will be understood by those of ordinary skill in the art in view of this disclosure. "biotherapeutic agent" refers to a therapeutic composition comprising a biological macromolecule, such as a gene therapeutic agent, therapeutic protein, nucleic acid, virus or cell or a portion thereof.

In the methods described herein, the biologic therapeutic may be selected from the group consisting of: antibodies, antigen-binding antibody fragments, antibody protein products, bispecific T cell conjugatesMolecules, bispecific antibodies, trispecific antibodies, fc fusion proteins, recombinant viruses, recombinant T cells, synthetic peptides, deoxyribonucleic acids (DNA), ribonucleic acids (RNA), and active fragments of recombinant proteins.

"antibody" has its usual and ordinary meaning as understood by one of ordinary skill in the art in view of this disclosure. It refers to any isotype of immunoglobulin that specifically binds to a target antigen and includes, for example, chimeric antibodies, humanized antibodies, and fully human antibodies. For example, the antibody may be a monoclonal antibody. For example, the human antibody may be of any isotype, including IgG (including IgG1, igG2, igG3, and IgG4 subtypes), igA (including IgA1 and IgA2 subtypes), igM, and IgE. Human IgG antibodies typically comprise two full length heavy chains and two full length light chains. The antibodies may be derived from only a single source, or may be "chimeric" antibodies, i.e., different portions of an antibody may be derived from two or more different antibodies of the same or different species. It will be appreciated that once the antibody is obtained from a source, it may be further engineered, for example, to enhance stability and folding. Thus, it will be appreciated that a "human" antibody may be obtained from a source and may undergo further engineering, for example engineering in the Fc region. Engineered antibodies may still be referred to as one type of human antibody. Similarly, variants of a human antibody, such as those that have undergone affinity maturation, will also be understood to be "human antibodies," unless otherwise indicated. In some embodiments, the antibody comprises, consists essentially of, or consists of: human, humanized or chimeric monoclonal antibodies.

The "heavy chain" of an antigen binding protein (e.g., an antibody) includes a variable region ("VH") and three constant regions: CH1, CH2 and CH3. Exemplary heavy chain constant regions suitable for use in the antigen binding proteins described herein, including, for example, human IgG1, igG2, igG3, and IgG4 constant regions, are shown in fig. 9A-B. The "light chain" of an antigen binding protein, such as an antibody, includes a variable region ("VL") and a constant region ("CL"). Human light chains include kappa chains and lambda chains. FIG. 9C shows exemplary light chain constant regions suitable for use in the antigen binding proteins described herein, including exemplary human lambda and human kappa constant regions.

In various aspects, the biologic therapeutic is an antibody protein product. As used herein, the term "antibody protein product" refers to any of several antibody substitutes that are based in each case on the architecture of an antibody but are not found in nature. In some aspects, the molecular weight of the antibody protein product is in the range of at least about 12-150 kDa. In certain aspects, the antibody protein product has a valence number (n) ranging from monomer (n=1) to dimer (n=2), to trimer (n=3), to tetramer (n=4), if not higher order valence numbers. In some aspects, antibody protein products are those based on intact antibody structures and/or those that mimic antibody fragments that retain intact antigen binding capacity, such as scFv, fab, and VHH/VH (discussed below). The smallest antigen-binding antibody fragment that retains its complete antigen binding site is an Fv fragment, which consists entirely of the variable (V) region. The V region is linked to the scFv fragment (variable single chain fragment) using a soluble flexible amino acid peptide linker to stabilize the molecule, or a constant (C) domain is added to the V region to create a Fab fragment [ antigen binding fragment ]. scFv and Fab fragments can be readily produced in host cells, such as prokaryotic host cells. Other antibody protein products include disulfide stabilized scFv (ds-scFv), single chain Fab (scFab), and dimeric and multimeric antibody forms, such as bifunctional, trifunctional, and tetrafunctional antibodies, or miniantibodies (miniabs) comprising different forms of scFv linked to an oligomerization domain. The smallest fragment is the VHH/VH of the camelidae heavy chain Ab and the single domain Ab (sdAb). The building blocks most commonly used to construct novel antibody formats are single chain variable (V) domain antibody fragments (scFv) comprising V domains (VH and VL domains) from heavy and light chains connected by a peptide linker having about 15 amino acid residues. A peptide body (peptabody) or peptide-Fc fusion is another antibody protein product. The structure of the peptibody consists of a bioactive peptide grafted onto the Fc domain. Peptide bodies are well described in the art. See, e.g., shimamoto et al, mAbs [ monoclonal antibodies ]4 (5): 586-591 (2012).

Biotherapeutic agents suitable for use in the methods described herein may include polypeptides, including those that bind to one or more of the following: these include CD proteins, including CD3, CD4, CD8, CD19, CD20, CD22, CD30 and CD34; including those that interfere with receptor binding. HER receptor family proteins, including HER2, HER3, HER4 and EGF receptors. Cell adhesion molecules such as LFA-I, moI, pl50, 95, VLA-4, ICAM-I, VCAM and αv/β3 integrin. Growth factors such as vascular endothelial growth factor ("VEGF"), growth hormone, thyroid stimulating hormone, follicular stimulating hormone, luteinizing hormone, growth hormone releasing factor, parathyroid hormone, miller tube inhibiting substances (mullerian-inhibiting substance), human macrophage inflammatory protein (MIP-1α), erythropoietin (EPO), nerve growth factors (such as NGF- β), platelet-derived growth factor (PDGF), fibroblast growth factors (including, for example, aFGF and bFGF), epidermal Growth Factor (EGF), transforming Growth Factor (TGF) (including, inter alia, TGF- α and TGF- β, including TGF- βl, TGF- β2, TGF- β3, TGF- β4 or TGF- β5), insulin-like growth factor-I and insulin-like growth factor-II (IGF-I and IGF-II), des (l-3) -IGF-I (brain IGF-I) and osteoinductive factors. Insulin and insulin-related proteins, including insulin, insulin a chain, insulin B chain, proinsulin, and insulin-like growth factor binding proteins. Coagulation proteins and coagulation-related proteins such as, inter alia, factor VIII, tissue factor, von Willebrands factor, protein C, alpha-1-antitrypsin, plasminogen activator (e.g., urokinase) and tissue plasminogen activator ("t-PA"), combazine, thrombin and thrombopoietin; (vii) Other blood and serum proteins, including but not limited to albumin, igE, and blood group antigens. Colony stimulating factors and their receptors, including, inter alia, the following: M-CSF, GM-CSF and G-CSF and their receptors, such as the CSF-1 receptor (c-fms). Receptors and receptor-associated proteins, including, for example, flk2/flt3 receptors, obesity (OB) receptors, LDL receptors, growth hormone receptors, thrombopoietin receptors ("TPO-R", "c-mpl"), glucagon receptors, interleukin receptors, interferon receptors, T cell receptors, stem cell factor receptors (e.g., c-Kit), and other receptors. Receptor ligands, including, for example, OX40L, are ligands for the OX40 receptor. Neurotrophic factors, including Bone Derived Neurotrophic Factor (BDNF) and neurotrophin-3, neurotrophin-4, neurotrophin-5, or neurotrophin-6 (NT-3, NT-4, NT-5, or NT-6). Relaxin a chain, relaxin B chain and relaxin pro-chain; interferon and interferon receptors include, for example, interferon- α, interferon- β and interferon- γ and their receptors. Interleukin and interleukin receptors, including, inter alia, IL-I to IL-33 and IL-I to IL-33 receptors, such as IL-8 receptor, and the like. Viral antigens, including AIDS envelope viral antigen. Lipoproteins, calcitonin, glucagon, atrial natriuretic factors, pulmonary surfactant, tumor necrosis factor-alpha and tumor necrosis factor-beta, enkephalinase, RANTES (activator of normal T cell expression and secretion), mouse gonadotropin-related peptides, dnase, inhibin and activin. Integrins, protein a or D, rheumatoid factors, immunotoxins, bone Morphogenic Proteins (BMP), superoxide dismutase, surface membrane proteins, decay Accelerating Factors (DAF), HIV envelopes, transport proteins, homing receptors, addressees, regulatory proteins, immunoadhesins, antibodies. Myostatin, TALL proteins (including TALL-I), amyloid proteins (including but not limited to amyloid beta protein), thymic stromal lymphopoietin ("TSLP"), RANK ligand ("RANKL" or "OPGL"), c-kit, TNF receptors (including TNF receptor type 1), TRAIL-R2, angiogenin, and biologically active fragments, analogs or variants of any of the foregoing.

Examples of biotherapeutic agents suitable for use in the methods described herein include antibodies, such as infliximab, bevacizumab, cetuximab (cetuximab), ranibizumab (ranibizumab), palivizumab (palivizumab), aba Fu Shan-antibody (abagnovimab), abciximab (abciximab), aclostrobiumab (actoxymab), adalimumab (adalimumab), aftimumab (afeimimab), amoizumab (afutuzumab), alzhuzumab (alacizumab), alemtuzumab (alacizumab pel), ald518, alemtuzumab (alemtuzumab), alemtuzumab (alicicumab), altimumab, altimomab (alatimomab), amatimomab), ma Anna-antibody (anatumomab mafenatox), kinumab (kinumab) Abolizumab (apolizumab), acipimab (arcitumomab), asenapuzumab (aselizumab), atino mab (altitumab), atizumab (atlizumab), atolizumab (atoleiumab), tozumab (tobalizumab), bapiduzumab (bapineuzumab), basiliximab (basiliximab), bavaluximab (bavituximab), bei Tuo momab (beclomab), belimumab (belimumab), bei Mali tozumab (bemaritumab), benzolizumab (benralizumab), bai Ti mumab (bertilimumab), bei Suoshan ab (besileumab), bevacizumab (bevacizumab), bezozumab (beziumab), bicetrimab (biciromab), bivalizumab (bivatuzumab), moxidecuzumab (bivatuzumab mertansine), boluzumab (blinatumomab), busuzumab (blosozumab), cetuximab (brentuximab vedotin), baziram mab (brikinumab), (brodalumab), connakuumab (canokinumab), 823-bezumab (cantuzumab mertansine), mo Kantuo-bemtuzumab (cantuzumab mertansine), kaposizumab (cappucizumab), carlo monoclonal antibody (capromab pendetide), caluzumab (carlumab), katuzumab (cataxomab), 49, cetrimab (cedlizumab), cetuximab (certolizumab pegol), cetuximab (cetuximab), cetuximab (citatuzumab bogatox) cetuximab (cixuumumab), cladazazumab (clazakizumab), celecoxib (clenoliximab), cervacizumab (clivatuzumab tetraxetan), corumumab (conatumumab), keramitraz monoclonal antibody (crenezumab), cr6261, daclizumab (dacuzumab), daclizumab (daclizumab), daclizumab Luo Tuzhu (dalotuzumab), daratuzumab (daclizumab), dimuzumab (demcizumab), denouzumab (delumumab), delumomab (delumomab), atovatuzumab (dorlimomab aritox), dec Luo Tushan (drozitumab), du Lige (dugoltumab), du Pilu monoclonal antibody (dulumab), exelizumab (ecomeximab), irituzumab (eclicumab), ebolizumab (edobamab), edelomab (edolomab), efalizumab (efalizumab), ifenesin mab (efalizumab), enotuzumab (elotuzumab), ai Ximo mab (elsilimomab), emamumab (enavatuzumab), pezilimumab (enlimomab pegol), enrouzumab (enokizumab), enotikumab (enotikumab), epotuzumab (etatuzumab), epotuzumab (epitumomab cituxetan), epratuzumab (epratuzumab), eretimumab (ereniumab), erluzumab (ertuzumab), idauzumab (etalizumab), etomizumab (etomizumab), etomizumab Zhuo Lizhu mab (etomizumab), etomizumab (35 vokuumab), valuzumab (35 vokuumab) Fascizumab (fanolesomab), faraday-bevacizumab (farletuzumab), farnesumab (fasinumab), fbta05, non-valuzumab (felvizumab), non-zakuumab (fezakinumab), fevernacuzumab (ficlatuzumab), phenytoin (figitumab), fratuzumab (flanvotumab), aryltuzumab (fontolizumab), fre Luo Lushan antibody (foralemab), fula Wei Shankang (foravirumab), fremizumab (fresolumab), framizumab (fulranumab), futuzumab (futuximab), ganimumab), ganitumumab (ganitumab), galitumumab (ganitumab), galuzumab (gandrum), gandrum ab (gandrum ab) and ganciclizumab (gandrum ab), jituzumab ozagrimocin (gemtuzumab ozogamicin), ji Fuzhu mab (gevokizumab), ji Rentu mab (givantuximab), vildazumab (glembatumumab vedotin), golimumab (golimumab), golimumab (gomiliximab), gs6624, irinotecan mab (ibalizumab), irinotecan mab (ibritumomab tiuxetan), elkumab (icrucumab), icofumab (igovimab), infliximab (imcriromab), enguzumab (imgazumab), elkuumab (incleaumab), etallimumab (indatuximab ravtansine), infliximab (infliximab), intuximab (etumumab), inomomab (inolimumab), irimizumab (inotuzumab ozogamicin) ipilimumab (ipilimumab), (iratuumab), etomizumab (itolizumab), mikulizumab (ixekizumab), keliximab (keliximab), la Bei Zhushan antibody (labtuzumab), lebrikizumab (lebikizumab), leybirbesumab, lemaleumab, ledymiumab (lerdileimiumab), lesimab (lexatumumab), li Weishan antibody (libivirus mab), li Gezhu antibody (ligelizumab), lytuzumab (lintuzumab), lilutumab (lirilumab), mo Fu Zhushan antibody (lorvotuzumab mertansine), lu Kamu mab (lucatumab), lu Xishan antibody (lumill ximab), ma Pamu mab (mapapumab), ma Simo mab (shimumab), melimumab (lizumab) and melizumab (lilimumab), matuzumab, mepolizumab, mevaluzumab, mi Lazhu mab (milatuzumab), merozolizumab (minutumomab), mi Tuomo mab (mitumomab), mo Gemu mab (mogamulizumab), moruzumab (moruzumab), motuzumab (motuzumab), motuzumab (motavizumab), mo Xitu mab (moxetumomab pasudotox), motuzumab-cd 3 (muromonab-cd 3), taluzumab (nacolomab tafenatox), nanoruzumab (namimumab), tamuzumab (naptumomab estafenatox), naltretamuzumab (namatumab), natalizumab (natuzumab), natuzumab (natuzumab), nabuzumab (natuzumab), nabapuzumab (neubacumab), nesuzumab (neuuzumab), retuzumab (neuuzumab) West-resistant monoclonal antibody (nesvacumab), nimuzumab (nimotuzumab), nivolumab (nivolumab), minolimumab (nofetumomab merpentan), oxcarbazelizumab (ocurauuzumab), oxlizumab (ocrelizumab), oxlimumab (odulimomab), oxlimumab (ofatumumab), olamagumab (olamatumab), omut2 monoclonal antibody (olokuzumab), omalizumab (omalizumab), onauzumab (onartuzumab), mo Aozhu monoclonal antibody (oportuzumab monatox), og Fu Shan antibody (orevomab), oxlimumab (ortitumab), oxlizumab (oxylizumab), oxuzumab (ozuzumab), oxzanuzumab (ozuzumab), omuzumab (otuzumab), omuzumab (ozlizumab) and omuzolizumab (omalizumab), parcabiximab (palibaximab), panacimumab (palivizumab), panitumumab (panitumumab), panorakuzumab (panobacumab), paspaluzumab (pamituzumab), panacolizumab (pascoluzumab), panacimumab (pasmodizumab), panitumumab (parthenoximab), panitumumab (patitumab), panitumumab (pamitumab), peziumab (perakizumab), pertuzumab (pertuzumab), pezizumab (pexelizumab), picolizumab (pimitumab), pertuzumab (pintumomab), prakuuzumab (plakuzumab), pertuzumab (ponemab), priliximab (priliximab), primu mab (pritumab), PRO 140 kunzirumab (quick mab), lei Kutu mab (racotuzumab), rad-image mab (radretumab), lei Weishan antibody (rafimumab), ramucirumab (ramucirumab), ranibizumab (ranibizumab), lei Xiku mab (raxibacumab), regasifi Wei Shankang (regavirumab), rayleigh mab (reslizumab), rituximab (rilotuzumab), rituximab (rituximab), luo Tuomu mab (robatumab), luo Ledu mab (roledumab), luo Msu mab (romasozumab), (rontalizumab), rosilizumab (rovelizumab), lu Lizhu mab (ruplizumab), sha Mali mab (samanimab), sha Lushan antibody (sabuzumab), sha Tuo mab (satumomab pendetide), and seuzumab (Wei Shankang), sibrotuzumab (sibrotuzumab), sibrotuzumab (sibdulumab), steuximab (siltuximab), sibutruzumab (sibuzumab), sibirizumab (siblizumab), sirtuin monoclonal antibody (sirukumab), su Lanzhu monoclonal antibody (sollanuzumab), thremzumab (solituab), pinepuzumab (sonepuzumab), pintuzumab (sontuzumab), stavtuzumab (stamulumab), thioxouzumab (sulesumab), sovisuzumab (suvizumab), ta Bei Lushan antibody (tabuzumab), tazuumab (tacatuzumab tetraxetan), tazuzumab (taduzumab), tazuumab (taluzumab), tauzumab (tanuzumab), pamuzumab (taplitumomab paptox), tezumab (tezuab) anammox (telimob aritox), tetomimumab (tenatumomab), tibezumab (tefibuzumab), tenectuzumab (teneliximab), toxib mab (toralizumab), toxib mab (toplizumab), teprimumab (tefuzumab), tizepelumab (tezepelumab), TGN1412, tremelimumab (tremelimumab), tizetimumab (ticlimumab), tilazumab (tildrakuzumab), tigazumab (tigatuzumab), TNX-650, toxilizumab, tolizumab (toralizumab), tositumomab (tositumab), qu Luoqing wooden mab (trastuzumab), trastuzumab (trastuzumab), TRBS, qu Jiali mab (tregzumab), 35 uzumab (35 Ezetimab), 67. Cine, to Wei Shankang (tuvirumab), wubolizumab (ublituximab), wu Ruilu mab (urelomab), wu Zhushan anti (urtoxazumab), you-t-tuzumab (ustekumab), valaciumab (vanoliximab), viterbi mab (vatelizumab), vedolizumab (vedolizumab), veltuzumab (veltuzumab), velumomab (vepalimomab), velocimab (veseneumab), viscizumab (visilizumab), fu Luoxi mab (volociximab), votuzumab (vorsetuzumab mafodotin), votumomab (votumab), zafimumab (zanolimumab), zafiumab (nolimumab), zafimbrumab (zafiumab), ji Lamu mab (ziramumab) or azamomab (lizomib).

In some embodiments, the biologic therapeutic isA molecule. />The molecules are engineered bispecific antigen binding constructs that direct the cytotoxic activity of T cells against cancer cells. They are fusions of two single chain variable fragments (scFv) of different antibodies or amino acid sequences from four different genes on a single peptide chain of about 55 kilodaltons. One of the scFv binds to T cells via CD3 receptor, while the other binds to tumor cells via tumor specific molecules. Bonauzumab ∈ ->Is specific for CD19>Examples of molecules. Modified->Molecules such as those modified to extend their half-life may also be used in the disclosed methods. In various aspects, the polypeptide is an antigen binding protein, e.g. +.>A molecule. In some embodiments, the antibody protein product comprises +.>A molecule.

In some embodiments, the biologic therapeutic is in a formulation. The formulation may be a pharmaceutically acceptable formulation. The formulation may comprise the biologic therapeutic together with pharmaceutically acceptable diluents, carriers, solubilizers, emulsifiers, preservatives and/or adjuvants.

Acceptable formulation materials for use in biotherapeutic agents as described herein are preferably non-toxic to the recipient at the dosages and concentrations employed. In certain embodiments, the pharmaceutical composition may contain a formulation for altering, maintaining or maintaining, for example, the pH, osmolarity, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption or penetration of the composition. In such embodiments, suitable formulation materials include, but are not limited to, amino acids (such as glycine, glutamine, asparagine, arginine, or lysine); an antimicrobial agent; antioxidants (such as ascorbic acid, sodium sulfite or sodium bisulfite); buffers (such as borates, bicarbonates, tris-HCl, citrates, phosphates or other organic acids); bulking agents (such as mannitol or glycine); chelating agents such as ethylenediamine tetraacetic acid (EDTA); complexing agents (such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin or hydroxypropyl-beta-cyclodextrin); a filler; a monosaccharide; disaccharides; and other carbohydrates (such as glucose, sucrose, mannose or dextrins); proteins (such as serum albumin, gelatin, or immunoglobulins); coloring agents, flavoring agents, and diluents; an emulsifying agent; hydrophilic polymers (such as polyvinylpyrrolidone); a low molecular weight polypeptide; salt-forming counterions (such as sodium); preservatives (such as benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid, or hydrogen peroxide); solvents (such as glycerol, propylene glycol or polyethylene glycol); sugar alcohols (such as mannitol or sorbitol); a suspending agent; surfactants or wetting agents (e.g., pluronics), PEG, sorbitan, polysorbates (e.g., polysorbate 20, polysorbate), tritium, tromethamine, lecithin, cholesterol, tyloxapol); stability enhancers (such as sucrose or sorbitol); tonicity enhancing agents (such as alkali metal halides, preferably sodium or potassium chloride, mannitol, sorbitol); a delivery vehicle; a diluent; excipients and/or pharmaceutically acceptable adjuvants. See, e.g., REMINGTON' S PHARMACEUTICAL SCIENCES [ Lemington pharmaceutical science ], 18 th edition, (A.R. Genrmo editions), 1990,Mack Publishing Company [ Mich.publishing company ].

Suitable vehicles or carriers for such formulations may be water for injection, physiological saline solution or artificial cerebrospinal fluid, possibly supplemented with other substances common in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are additional exemplary vehicles. In particular embodiments, the pharmaceutical composition comprises a Tris buffer at about pH7.0-8.5 or an acetate buffer at about pH 4.0-5.5, and may further comprise sorbitol or a suitable substitute thereof.

The formulation components are preferably present in a concentration acceptable for the site of application. In certain embodiments, buffers are used in order to maintain the composition at physiological pH or at a slightly lower pH, typically in the pH range of about 5 to about 8. Including about 5.1, about 5.2, about 5.3, about 5.4, about 5.5, about 5.6, about 5.7, about 5.8, about 5.9, about 6.0, about 6.1, about 6.2, about 6.3, about 6.4, about 6.5, about 6.6, about 6.7, about 6.8, about 6.9, about 7.0, about 7.1, about 7.2, about 7.3, about 7.4, about 7.5, about 7.6, about 7.7, about 7.8, about 7.9, and about 8.0.

It should be noted that some biotherapeutic agents may be self-administered by a subject directly or via an auto-injector, while some biotherapeutic agents may be administered to a subject by another individual (such as a health care provider). Thus, as used herein, a subject "receiving and administering" or "administering" a biologic therapeutic to a subject, as well as variations of these root terms, may refer to self-administration of the biologic therapeutic to the subject by the subject to itself (either directly or via a device such as an automatic injector) and/or by other individuals such as a health care provider.

Method for producing biotherapeutic agent

It is contemplated that the methods described herein can be used to evaluate whether a molecular property would affect the efficacy of a biologic therapeutic, and/or whether the safety profile of a biologic therapeutic would be affected. The methods may include determining an estimated actual level of exposure of the molecular property when the subject receives administration of the biologic therapeutic. Such methods may determine molecular property levels or ranges suitable for manufacturing, batch release, and/or application time.

In some embodiments, a method of manufacturing a biologic therapeutic is described. The method may comprise detecting the level of the molecular property of the biological therapeutic in the formulation at one or more time points under storage conditions (optionally, at two or more time points under storage conditions). Optionally, it may be sufficient to detect the level of molecular properties at a point in time, for example in the case of no change in molecular properties upon storage (e.g. for some pharmaceutical products stored frozen). It is further contemplated that for levels of molecular properties that may change during storage, detecting levels of molecular properties at two or more points in time under storage conditions may be used to calculate the rate of change of molecular properties as described herein. For example, the point in time may include at the time of manufacture. For example, the points in time may include at the time of manufacture and at least one other point in time. The method may include determining a rate of change of a molecular property under storage conditions (which may be calculated to be zero for some molecular properties of some biotherapeutic agents under some storage conditions (e.g., some biotherapeutic agents stored frozen). The method may include obtaining data regarding the in vivo safety and/or efficacy of the biologic therapeutic in a subject who has received administration of the biologic therapeutic. The method may comprise estimating the level of molecular property exposure that the subjects receive at said administration based on (i) the rate of change of the molecular property during storage and (ii) the duration of time that the biotherapeutic agent in the formulation was under the storage conditions prior to said administration. The estimation may also take into account (iii) the level of molecular properties at the time of manufacture and/or at the time of release of the batch, and/or (iv) the dose of the biologic product administered to the subject. For example, the estimation may use equation 1 or equation 2 or equation 3. The method may include determining whether a correlation exists between the estimated level of molecular property exposure received by the subject and safety and/or efficacy data for the biologic therapeutic. If the correlation is not present, the method may include manufacturing a production lot of the biologic therapeutic comprising the molecular property at or below a specified allowable level of the molecular property based on the estimated level of molecular property exposure. For example, the specified allowable level may be a molecular property level calculated to be less than or equal to a highest estimated level of molecular property exposure in the subject at the end of the shelf life. For example, the specified allowable level may be a molecular property level calculated to produce a property exposure level that is less than or equal to 90% -100% of the highest estimated level of molecular property exposure in the subject at the end of the shelf-life. Thus, the allowable levels are expected to be demonstrated as safe and effective levels of attribute exposure in the subject throughout the shelf life of the biologic therapeutic. Production lots whose molecular property levels exceed a specified allowable level may be rejected.

If the correlation is present, the method may further comprise setting a specification of the molecular property level that does not exceed a maximum allowable level of the molecular property of the biotherapeutic agent at the time of manufacture. The maximum allowable level of the molecular property may be based on a highest estimated level of molecular property exposure in the subject that is not associated with adverse events and/or efficacy inhibition of the biologic therapeutic. The method of manufacturing may further comprise rejecting the production lot of the biologic therapeutic comprising the level of the molecular property exceeding a maximum allowable level (and thus out of specification). Production lots may be received having molecular property levels that do not exceed a specified maximum allowable level. For example, the specified maximum allowable level may be a level of molecular property calculated to produce a property exposure at the end of the shelf life that is less than or equal to a highest estimated level of the molecular property exposure that is not related to adverse events and/or efficacy inhibition in the subject. The calculation may use equation 1 or equation 2 or equation 3, and may take into account the rate of change of the molecular property in the formulation during storage, as well as the dose of the molecular property to be administered to the subject. For example, the specified maximum allowable level may be a level of molecular property calculated to produce a property exposure level at the end of the shelf life that is less than or equal to 90% -100% of the highest estimated level of the molecular property exposure that is not associated with adverse events and/or efficacy inhibition in the subject.

As used herein, an "allowable level" of a molecular property refers to a level of the molecular property of a biologic therapeutic in a formulation at the time of manufacture that is within specification. As described herein, the allowable level may be calculated to obtain a specified level of molecular property exposure (for a subject to whom the biologic therapeutic in the formulation is to be administered) that is less than or equal to the estimated level of molecular property exposure that is not associated with adverse events and/or loss of efficacy at the end of or upon expiration of the shelf-life (this may be referred to as an allowable level that is "based on" the estimated level of property exposure). Thus, the allowable level of molecular property may provide confidence that the level of exposure of the molecular property from the biologic therapeutic in the formulation is safe and effective when administered to a subject even if administered very close to its shelf-life or expiration date. "maximum allowable level" refers to the situation when there is a correlation between the estimated level of exposure of a molecular property and safety and/or efficacy data. By "maximum allowable level" is meant the highest level of molecular property in the formulation that results in a level of molecular property exposure (for the subject to whom the biotherapeutic agent in the formulation is to be administered) that is less than or equal to the highest level of molecular property exposure that is not associated with adverse events at the end of or at the expiration of the shelf life. The allowable level or maximum allowable level may be calculated based on the estimated level of attribute exposure unrelated to the adverse event, the duration until the end or expiration of the shelf life, the rate of change of the molecular attribute level, and the dosage of the biologic therapeutic using equation 1 or equation 2 or equation 3. That is, using the level of exposure of the molecular property, the rate of change of the molecular property in the formulation, and the duration that are not associated with adverse events and/or efficacy inhibition, the allowable level (and, if applicable, the maximum allowable level) may be determined. Thus, if a biologic therapeutic is manufactured with molecular properties at or below an acceptable level (at or below a maximum acceptable level, if applicable), it is expected that upon administration to a subject, the estimated level of molecular property exposure to the subject will be at or below a level that is not associated with an adverse event and/or loss of efficacy.

As used herein, an allowable level or maximum allowable level of an estimated level of exposure of a "based on" molecular property refers to an allowable level (or maximum allowable level) calculated to not exceed the estimated level of molecular property at the end or expiration of the shelf-life. If more than one dose is appropriate for the biologic therapeutic in the formulation, the highest appropriate dose may be used to calculate the allowable or maximum allowable level (as lower doses will have even lower levels of molecular property exposure) based on the estimated level of molecular property. For example, the estimated level of attribute exposure may be selected to fall within a distributed or scattered confidence interval of estimated molecular attribute exposure levels determined for a group of subjects. Considering that the estimated level of attribute exposure that a subject receives does not necessarily imply that each subject receives the same numerical level of attribute exposure. Conversely, the estimated level of exposure to the attribute that those subjects receive may include a distribution of estimated attribute levels that the individual subjects receive. Thus, the confidence interval may be used to select the maximum allowable level such that there is a probability of at least 85%, 90%, 95%, 97% or 99% that the level of attribute exposure at the end of the shelf life is less than or equal to the highest level of attribute exposure that is not associated with loss of safety or efficacy. It is further contemplated that actual values of allowable levels or maximum allowable levels of exposure levels of the "based on" molecular properties may be rounded. This rounding may provide administrative or mathematical convenience, such as rounding to one, two, or three significant digits in a suitable means for measuring a property of a molecule. To increase caution, rounding may be downward rounding. For example, an attribute tolerance level or maximum tolerance level that is "based on" the estimated level of attribute exposure may be calculated to result in an attribute exposure level that does not exceed 99%, 97%, 95%, 90%, 85%, or 80% of the reference level of attribute exposure (unrelated to loss of safety and/or efficacy) calculated using equation 1 or equation 2 as described herein at the end of the shelf life. Such downward rounding may provide additional assurance that the level of attribute exposure from the biologic therapeutic at the end of the shelf life will still be safe and/or effective.

In some embodiments, production lots are manufactured that each have a level of more than one molecular property at or below its allowable (or maximum allowable) level. For example, production lots having at least one, two, three, four, five, six, seven, eight, nine, or ten molecular attributes may be manufactured, including ranges between any two of the values listed, such as one-ten, one-five, two-ten, two-five, three-ten, three-five, or five-ten molecular attributes, each at or below a specified tolerance (or maximum tolerance) level.

Manufacturing technique

Methods for making a biologic therapeutic (e.g., therapeutic protein) as described herein can utilize recombinant DNA techniques. Recombinant DNA methods for producing therapeutic proteins such as antibodies or antibody protein products are well known. The DNA may encode a therapeutic protein. For example, the DNA may encode an antibody, e.g., DNA (target polynucleotide) encoding a VH domain, VL domain, single chain variable fragment (scFv), or fragments and combinations thereof, may be inserted into a suitable expression vector, which may then be transfected into a suitable host cell, such as an Escherichia coli cell, a COS cell, a Chinese Hamster Ovary (CHO) cell, or a myeloma cell, that would not otherwise produce the antibody, to obtain the desired antibody.

Suitable expression vectors are known in the art and comprise, for example, polynucleotides encoding target polypeptides linked to promoters. Such vectors may include nucleotide sequences encoding the constant regions of the antibody molecules, and the variable domains of the antibodies may be cloned into such vectors to express the heavy chain, the entire light chain, or the entire heavy and light chains (or fragments thereof). The expression vector may be transferred into a host cell by conventional techniques, and the transfected cells may be cultured to produce antibodies.

Any cell line that can express or be engineered to express a protein, such as a functional antibody or antibody fragment, can be used. For example, suitable mammalian cell lines include immortalized cell lines available from the American type culture Collection (Manassas, va.) including Chinese hamster ovary (CH) cells, sea Law cells, baby Hamster Kidney (BHK) cells, monkey kidney Cells (COS), human hepatocellular carcinoma cells (e.g., hep G2), and human epithelial kidney 293 cells. In addition, the cell line or host system may be selected to ensure proper modification and processing of the antibody. Eukaryotic host cells having cellular machinery for proper processing of the primary transcript, glycosylation and phosphorylation of the gene product may be used. These include CHO, VERY, BHK, hela, COS, MDCK, 293, 3T3, W138, BT483, hs578T, HTB2, BT20 and T47D, NS0 (not included) Murine myeloma cell line that produces any functional immunoglobulin chain), SP20, CRL7030 and HsS Bst cells. Human cell lines produced by immortalizing human lymphocytes may also be used. Human cell lines can be used(Yansen; titusville, NJ) to recombinantly produce monoclonal antibodies. Examples of non-mammalian cells that may also be used include insect cells (e.g., sf21/Sf9, trichoplusia ni (Trichoplusia ni) Bti-Tn5 bl-4), or yeast cells (e.g., saccharomyces (s. Cerevisiae), pichia (Pichia), etc.), plant cells, or chicken cells.

Proteins such as antibodies can be stably expressed in cell lines using conventional methods. Stable expression can be used for long-term, high-yield production of recombinant proteins. For stable expression, the host cell may be transformed with a suitably engineered vector comprising an expression control element (e.g., promoter, enhancer, transcription terminator, polyadenylation site, etc.), and a selectable marker gene. Methods for producing stable cell lines in high yield are known in the art, and reagents are commercially available. Transient expression may also be achieved by conventional methods.

Cell lines expressing proteins such as antibodies can be maintained in cell culture media and culture conditions that cause expression and production of the antibodies. Cell culture media can be based on commercially available media formulations including, for example, DMEM or hami F12 (Ham's F12). In addition, the cell culture medium may be modified to support cell growth and increased expression of biological proteins. Of course, cell culture media can be optimized for a particular cell culture, including cell culture growth media formulated to promote cell growth or cell culture production media formulated to promote recombinant protein production.

Many cell culture media and cell culture nutrients and supplements are known. For example, suitable basal media include Dulbecco's Modified Eagle's Medium (DMEM), DME/F12, minimal Essential Medium (MEM), basal Medium Igor (Basal Medium Eagle, BME), RPMI 1640, F-10, F-12, a-minimal essential Medium (a-MEM), grassow minimal essential Medium (Glasgow's Minimal Essential Medium, G-MEM), PF and Iscove's Modified Dulbecco's Medium (Iscove's Modified Dulbecco's Medium). Examples of other basal media that may be used include BME basal media, dulbeck's modified eagle's medium.

The basal medium may be serum-free, which means that the medium does not contain serum (e.g., fetal Bovine Serum (FBS)), or animal protein-free medium or chemically defined medium. The basal medium may be modified to remove certain non-nutritive components found in the basal medium, such as various inorganic and organic buffers, one or more surfactants, and sodium chloride. The cell culture medium may contain a basal cell culture medium (modified or unmodified) and at least one of the following: an iron source, recombinant growth factor; a buffer; a surfactant; osmotic pressure regulator; an energy source; and non-animal hydrolysates. In addition, the modified basal cell culture medium may optionally contain amino acids, vitamins, or a combination of both amino acids and vitamins. The modified basal medium may further contain glutamine, such as L-glutamine and/or methotrexate.

Once the therapeutic protein (e.g., antibody or antibody protein product) has been produced, it may be purified by conventional methods, for example, by chromatography (e.g., ion exchange, affinity (particularly by affinity to specific antigens, protein a, protein G) or column chromatography), centrifugation, differential solubility, or by any other standard technique for purifying proteins. In addition, the protein may be fused to a heterologous polypeptide sequence ("tag") to facilitate purification.

The purified protein is typically formulated with excipients to produce a sterile solution that can be injected or infused. For example, the purified protein may be formulated into a formulation as described herein. After formulation, it may be filled, packaged, stored, delivered and finally prepared immediately prior to administration to a patient.

Method of developing manufacturing processes for biotherapeutic agents

It is contemplated that the methods described herein may also be used to develop methods of manufacturing biological therapeutics. In some embodiments, a method of developing a manufacturing process for a biologic therapeutic is described. The method may comprise detecting the level of the molecular property of the biological therapeutic in the formulation at one or more time points under storage conditions (optionally at two or more time points under storage conditions). For example, the point in time may include at the time of manufacture. For example, the points in time may include at the time of manufacture and at least one other point in time. The method may include determining a rate of change of the molecular property under the storage conditions. The method may include obtaining data regarding the safety and/or efficacy of the biologic therapeutic in a subject that has received administration of the biologic therapeutic. The method may comprise estimating the level of molecular property exposure that the subjects receive at the time of administration based on (i) the rate of change of the molecular property during storage of the biological therapeutic agent in the formulation and (ii) the duration of time that the biological therapeutic agent in the formulation was in the storage conditions prior to said administration. The estimation may also take into account (iii) the level of molecular properties at the time of manufacture and/or at the time of release of the batch, and/or (iv) the dose of the biologic product administered to the subject. For example, the estimation may use equation 1 or equation 2 or equation 3. The method may include determining whether a correlation exists between the estimated level of molecular property exposure and safety and/or efficacy data of the biologic therapeutic.

If (a) the correlation is not present, the method may include establishing the manufacturing process to produce the molecular property level at or below a specified allowable level based on the estimated level of molecular property exposure. For example, the specified allowable level may be a molecular property level calculated to be less than or equal to a highest estimated level of molecular property exposure in the subject at the end of the shelf life. The calculation may use equation 1 or equation 2 or equation 3. For example, the specified allowable level may be a molecular property level calculated to produce a property exposure level that is less than or equal to 90% -100% of the highest estimated level of molecular property exposure in the subject at the end of the shelf-life. Production lots whose molecular property levels exceed a specified allowable level may be rejected.

If (b) the correlation is present, the method may include establishing the manufacturing process based on a highest level of the molecular property that is not associated with adverse events and/or efficacy inhibition of the biologic therapeutic to produce the molecular property level at or below a specified maximum allowable level of the molecular property. For example, the specified maximum allowable level may be a level of molecular property calculated to produce a property exposure at the end of the shelf life that is less than or equal to a highest estimated level of the molecular property exposure that is not related to adverse events and/or efficacy inhibition in the subject. The calculation may use equation 1 or equation 2 or equation 3. For example, the specified maximum allowable level may be a level of molecular property calculated to produce a property exposure level at the end of the shelf life that is less than or equal to 90% -100% of the highest estimated level of the molecular property exposure that is not associated with adverse events and/or efficacy inhibition in the subject. The tolerance level or maximum tolerance level may be part of the specifications at the time of manufacture or at the time of batch placement.

Methods of assessing clinical impact of molecular properties of biotherapeutic agents

In some embodiments, a method of assessing clinical impact of a molecular property is described. Such methods may further be used in methods of manufacturing a biologic therapeutic as described herein, as well as in methods of developing manufacturing processes for biologic therapeutic. The method may comprise detecting the level of the molecular property of the biotherapeutic agent in the formulation at one or more time points under storage conditions. The level of attribute exposure may be detected at two or more points in time under storage conditions. For example, the time point for detecting the level of the molecular property may comprise at the time of manufacture or at least one other time point. For example, the two or more time points for detecting the level of the molecular property may include at the time of manufacture and at least one other time point. The method may include determining a rate of change of the molecular property under the storage conditions. The method may include obtaining data regarding the safety and/or efficacy of the biologic therapeutic in a subject that has received administration of the biologic therapeutic. The method may comprise estimating the level of molecular property exposure that the subjects receive at said administration based on (i) the rate of change of the molecular property during storage of the biological therapeutic agent in the formulation and (ii) the duration of time that the biological therapeutic agent in the formulation was in the storage conditions prior to said administration. The estimation may also take into account (iii) the level of molecular properties at the time of manufacture and/or at the time of release of the batch, and/or (iv) the dose of the biologic product administered to the subject. For example, the estimation may use equation 1 or equation 2 or equation 3 as described herein. The method may include determining whether a correlation exists between the estimated molecular property exposure and the safety and/or efficacy of the biotherapeutic agent. If (a) the correlation is absent, it can be determined that the molecular property does not affect neither the clinical safety nor the efficacy of the biologic therapeutic. If (b) the correlation is present, it can be determined that the molecular property affects the clinical safety and/or efficacy of the biologic therapeutic.

If (a) the correlation is not present, the method may further comprise setting a specification of allowable levels of the molecular property of the biologic therapeutic, wherein the allowable levels of the molecular property are based on the highest estimated levels of molecular property accepted by the subjects. For example, the specification of the allowable level may be a molecular property level at the time of manufacture, calculated to be a property exposure level that is less than or equal to the highest estimated level of molecular property exposure in the subject at the end of the shelf life. For example, the specification of the allowable level may be a molecular property level calculated to produce a property exposure level of less than or equal to 90% -100% of the highest estimated level of molecular property exposure in the subject at the end of the shelf life. The specification may be used for manufacturing and/or release testing of the biologic therapeutic. If a batch or product of a biologic therapeutic contains a molecular property level that exceeds an acceptable level, the batch or product may be considered to be out of specification and rejected. The biologic therapeutic may be manufactured according to the specifications.

If (b) the correlation exists, the method may further comprise setting a specification of a maximum allowable level of the molecular property of the biologic therapeutic, wherein the maximum allowable level of the molecular property is based on a highest estimated molecular property exposure level that is not related to adverse events and/or efficacy inhibition of the biologic therapeutic. The specification may be used for manufacturing and/or release testing of the biologic therapeutic. For example, the specification of the maximum allowable level may be a level of molecular property calculated to produce a property exposure level at the end of the shelf life that is less than or equal to a highest estimated level of the molecular property exposure that is not related to adverse events and/or efficacy inhibition in the subject. The calculation may use equation 1 or equation 2 or equation 3. For example, the specified maximum allowable level may be a level of molecular property calculated to produce a property exposure level at the end of the shelf life that is less than or equal to 90% -100% of the highest estimated level of the molecular property exposure that is not associated with adverse events and/or efficacy inhibition in the subject. If a manufacturing lot or product of a biologic therapeutic contains a level of molecular attributes that exceeds a maximum allowable level, the lot or product may be considered out of specification and rejected. The biologic therapeutic may be manufactured according to the specifications.

Other aspects of the method

Any of the methods described herein may include one or more other aspects.

In some embodiments, for any of the methods described herein, the estimation is further based on (iii) the dose of the biologic therapeutic in said administration, and (iv) the amount of the molecular property measured at the time of manufacture and/or batch placement. Regarding (iii), note that a larger dose of the biologic will administer a greater amount of molecular property than a smaller dose with the same percentage content of molecular property. It is also contemplated that in some cases the molecular property level and/or dose at the time of manufacture may not vary significantly and thus can be estimated from (i) and (ii) alone. For example, manufacturing processes can be established and tightly controlled and show consistent levels of molecular properties at the time of manufacture. For example, the biologic therapeutic may be administered in only a single dose, such that weighting of the calculations by dose is not required.

In some embodiments, for any of the methods described herein, the molecular property level of the biologic therapeutic in the formulation is measured at one or more time points at or after manufacture, e.g., at least one, two, three, four, five, six, seven, eight, nine, or ten time points, including ranges between any two of the listed values, e.g., one-two time points, one-five time points, or one-ten time points. Without being limited by theory, it is contemplated that for some biotherapeutic agents in a formulation, the level of molecular properties does not change during storage, as may be the case for some biotherapeutic agents in a frozen stored formulation, it may be sufficient to measure the level of molecular properties at a single point in time. In some embodiments, for any of the methods described herein, the molecular property level of the biologic therapeutic in the formulation is measured at two or more time points, e.g., at least two, three, four, five, six, seven, eight, nine, or ten time points, including ranges between any two of the listed values, e.g., two-five time points, two-ten time points, three-five time points, three-ten time points, or five-ten time points. The molecular property level at two or more time points can be used to calculate the rate of change of the molecular property level under storage conditions. The point in time may be at the time of manufacture or after manufacture. In some embodiments, for any of the methods described herein, the two or more points in time comprise at the time of manufacture. In some embodiments, for any of the methods described herein, the two or more time points comprise at the time of manufacture and at least one subsequent time point. In some embodiments, for any of the methods described herein, the two or more time points comprise at the time of manufacture and at least two subsequent time points.

In some embodiments, for any of the methods described herein, estimating the level of molecular property exposure comprises using an equation as described herein. In some embodiments, estimating the level of molecular property exposure includes using equation 1:

A _t ＝(％A ₀ +％A _Δ ×t)D (1)

wherein A is _t Is an estimated level of molecular property exposure,% A ₀ Is the attribute percentage of the batch run,% A _Δ Is the rate of change of the percentage of the property level, t is the storage time between the release of the batch and the administration of the treatment, and D is the dose strength in the weight dimension of the active pharmaceutical ingredient associated with each treatment.

In some embodiments, estimating the level of molecular property exposure includes using equation 2:

wherein A is _t Is this estimated level of molecular property exposure,% A ₀ Is the percentage of the molecular property,% A, at batch placement _Δi Is the rate of change, t, of the percentage of the molecular property level over time under given storage conditions _i Is the time stored under the given conditions, and D is the dose strength of the administration.

In some embodiments, estimating the level of molecular property exposure includes using equation 3:

wherein% _rel Is the percentage of the relative level of molecular property exposure to dose, where A _t Is the level of exposure of this molecular property calculated using equation 1 or 2, and D is the dose intensity in the weight dimension of the active pharmaceutical ingredient associated with each treatment. Note that equation 3 can explain the relative level of molecular property exposure versus dose intensity. Without being limited by theory, if a molecular property affects efficacy, it can be inferred that a non-property may also affect efficacy. For example, the relative concentration of the Active Pharmaceutical Ingredient (API) with respect to the dose intensity may affect efficacy. Thus, the% molecular property relative to dose intensity can be used to measure the effect of the molecular property on efficacy . Thus, in the method of some embodiments, equation 3 may be used when determining whether there is a correlation between the estimated level of molecular property exposure and the efficacy data.

In some embodiments, for the methods described herein, estimating the level of molecular property exposure that the subject receives comprises assigning the estimated level of property exposure as follows:

(A) A is first determined using equation 1 or equation 2 or equation 3 for all relevant treatments as described below _t 。

(1) For patients that did not show clinical outcome of interest throughout the clinical study, the relevant treatments were all treatments administered to the patient throughout the study.

(2) For patients exhibiting clinical outcome, the relevant treatment is all treatments administered to the patient prior to the occurrence of the clinical outcome.

(3) For patients who develop clinical outcome prior to receiving any treatment, there is no relevant treatment. A of these patients _t Is 0.

(4) For biologic therapeutic administered via continuous infusion, a occurs during relevant portions of the infusion _t The determination (for a biologic therapeutic that is not administered via continuous infusion, item (4) may be omitted)) may be as follows:

a. for infusions administered continuously over 12 hours or 24 hours, a for each 12 hour or 24 hour period of relevant infusions prior to the occurrence of clinical outcome was determined _t 。

b. For infusions less than 12 hours long (or 24 hours long, as applicable), A associated with each associated infusion before clinical outcome occurs is determined _t 。

c. For any of the 2 cases (a and b) above, if a clinical outcome occurs during infusion, determining a related to the portion of the infusion prior to the clinical outcome occurring _t 。

(B) Once all relevant treatments A for each patient with or without clinical outcome are determined _t Then all A's associated with the relevant portion of the relevant treatment or infusion are calculated as defined above _t Is a flat part of (2)Mean or maximum.

Note that additional mathematical operations may be applied to further refine the determination of correlation between the molecular property exposure level and data regarding safety and efficacy. In some embodiments, for any of the methods described herein, the correlation comprises a weighted correlation between the attribute exposure and the occurrence of the adverse event. The weighting may use the size of each group of subjects. Without being limited by theory, it is contemplated that if subjects experience adverse events in a multi-dose regimen of a biologic therapeutic, those subjects may discontinue use of the biologic therapeutic. Overall, subjects experiencing adverse events may be less data than subjects not experiencing adverse events, which may affect the determination of correlation (or lack thereof). Thus, the data may optionally be weighted to account for subjects who discontinued the regimen prior to completion of the multi-dose regimen of the biologic therapeutic, such as subjects experiencing adverse events.

It is further contemplated that the biologic therapeutic in the formulation may be in storage conditions for different subjects for different durations. Thus, there may not be a single duration of time that the biologic therapeutic in the formulation is in storage conditions. Instead, different durations may be applied for different administrations of the biologic therapeutic, as appropriate. By selecting in each case the appropriate storage time (t _i ) Estimating the level of exposure of the molecular property may account for these different durations. Thus, in the methods of some embodiments, the biologic therapeutic is in storage conditions for different durations in administration to different subjects.

It is further contemplated that for a method as described herein, administration of a biologic therapeutic may include two or more administration events, e.g., at least two, three, four, five, or ten administration events, including a range between any two of the listed values, such as two-three, two-five, two-ten, three-five, three-ten, or five-ten administration events. Thus, in some embodiments, the estimated level of exposure of the attribute that the subjects receive at the time of administration is the maximum or average of the two or more administration events.

It is further contemplated that for the methods described herein, administration of the biologic therapeutic may comprise continuous infusion. Continuous infusion may be for hours or days. In some embodiments, the administering comprises continuous infusion, and estimating the level of exposure of the property comprises calculating an estimated level of exposure of the property of the molecule during two or more intervals of continuous infusion, such as 12 hours or 24 hours intervals.

The methods described herein can utilize clinical study data. These include, but are not limited to, subject number, clinical outcome information (date, time and extent (severity or grade)), treatment lot number, treatment (date, time and duration). The data may be grouped according to clinical outcome (e.g., clinical endpoint and/or adverse event). Adverse events may indicate safety data and clinical endpoints may indicate validity data. In groupings based on clinical outcome, subjects are grouped according to the occurrence of clinical outcome and the severity or extent of interest. For example, to analyze the impact of molecular attributes on the severity of a safety adverse event, subjects are grouped according to the occurrence of AEs with severity levels of interest (e.g., a severe fever positive group would include patients with grade 3 or higher fever, while a severe fever negative group would include subjects without grade 3 or higher fever). Then, an estimated level of attribute exposure for each subject is determined as described.

For example, security data for the methods described herein may include adverse event data. Adverse events may include treatment-related adverse events. For example, adverse events related to treatment with a biologic therapeutic may include immune adverse events, such as immune responses to biologic therapeutic (e.g., anti-drug antibody (ADA)). Other examples of adverse events include anemia, cytokine Release Syndrome (CRS), fever, infusion-related reactions (IRR), lymphopenia, or neurological events. In some embodiments, the security data includes adverse event data for any of the methods described herein. For example, the security data may include adverse event time courses.

For example, efficacy data for the methods described herein can include clinical evidence of efficacy, such as treatment, prevention, amelioration, or delay of onset of a disease or disorder. Clinical endpoint data may indicate efficacy. Thus, in some embodiments, the efficacy data comprises clinical endpoint data.

It is contemplated that for any of the methods herein, statistical methods may be used to calculate whether there is a correlation between the estimated level of molecular property exposure and the safety and/or efficacy data of the biologic therapeutic. For example, a linear regression correlation between the ratio of clinical events from safety and/or efficacy data and the attribute exposure level may be calculated, and a p-value may be determined (fig. 3A, 3B, 7B, and 9). For example, the p-value may be calculated from a t-test or an F-test (fig. 2A, 2B, 5A, 5B, 6, 7A, 8, 10A, and 10B). For example, a log rank test or Mantel-Cox test can be performed to calculate p-values to test early clinical safety event episodes for a group of clinical subjects with higher levels of attribute exposure (fig. 4A and 4B).

In some implementations of any of the methods described herein, a bayesian estimation method is used. Bayesian estimation methods can be used to determine if there is a correlation between the estimated level of molecular property exposure and safety and/or efficacy data for the biological therapeutic. Due to the progressive learning nature of bayesian methods, all evidence can be used entirely a priori in an unbiased manner. In particular, when a bayesian estimation method is applied to determine if there is a correlation between the estimated level of molecular property exposure and the safety and/or efficacy data of the biologic therapeutic, the method may modify the probability density function of the clinical impact of each property to take on a certain value when new evidence appears until all evidence is used. At the end of the analysis, the probability of each attribute being associated with each adverse event is derived. Computer programs have been developed to take advantage of the high computational power required by bayesian methods (see example 12). The program provides an interactive platform to visualize the results of the bayesian estimation method using a user-provided spreadsheet. The program also provides a parameter adjustment module to define an initial prior, remove outliers and select a plurality of attributes. Several sets of authentic clinical data from CIA previously performed on mAb D were systematically examined, along with some modeling data based on mAb D CIA data. The Bayesian estimation method generates expected results, which prove the effectiveness of the system.

Examples

Example 1: molecular attribute analysis method

Here, we examined the clinically relevant immunoprotection effects of 15 molecular attributes between 2 immunoglobulin G2 (IgG 2) monoclonal antibody (mAb) drug products mAb a and mAb B using a data analysis method called clinical impact analysis (Clinical Impact Analysis, CIA). CIA verifies whether there is a correlation between the level of exposure of a patient to a given attribute and the extent of occurrence of clinical outcome by analyzing data from clinical studies and product quality analysis studies. No correlation was found between the level of exposure of the patient to any of the attributes and the ADA response, indicating that the analyzed attributes did not elicit ADA response at the examined exposure level. Moreover, the upper range of exposure levels examined was above or near the purity specification limit of the mAb a attributes, indicating the feasibility of producing clinically relevant evidence to demonstrate that CIA made these attributes possible. These results show that integrating clinical and analytical information can provide clinically relevant insight into the impact of attributes suitable for drug development and manufacture.

We evaluate the clinically relevant safety impact of attributes by integrating the clinical and attribute analysis data (fig. 1, equation 1). Briefly, the attribute exposure level at the time of treatment administration was approximated by combining the measured attribute level at the time of batch release with the change in attribute level between batch release and treatment administration (fig. 1A). Attribute exposure of the individual treatments to be administered to the patient during the treatment regimen is then determined. If a certain attribute causes a clinical event, then a patient with that event will be found to have a higher correlation with the average exposure level of that attribute (FIG. 1B), or a positive correlation will be found between the attribute exposure level and the incidence or time course of the clinical event (FIG. 1C, left) or (FIG. 1C, right). Conversely, the absence of an association between an attribute exposure and a clinical event will indicate that the attribute does not cause the event at the examined exposure level.

As shown in FIG. 1A, the level of exposure A of a property at the time of treatment administration is approximated by taking into account the measured property level at the time of manufacture and the change in that level over time _t . In retrospective analysis, patients are grouped according to retrospective known clinical outcomes, such as ADA test outcomes (fig. 1B). Then, as shown, A representing individual patients between groups is compared _t . In the prospective analysis, according to representative A _t Patients were grouped (fig. 1C). The incidence (left) or time course (right) of clinical outcome for each group is compared to the average representative A for that group _t A comparison is made.

We studied the immune response of patients exposed to 15 attributes between 2 independent drug products, mAb a and mAb B, by examining attribute exposure and ADA test results for 1,398 patients from 8 separate clinical studies of mAb a and 5,439 patients from 2 combined clinical studies of mAb B. ADA remains a critical regulatory issue due to a range of effects on safety and efficacy, including neutralization of target binding functions by drugs and other more serious adverse events ^17,18 。

The properties analyzed for mAb A are High Molecular Weight (HMW) species in Size Exclusion Chromatography (SEC), acidic species in Cation Exchange (CEX), non-heavy and light chain (non-HC+LC) species in capillary electrophoresis under reducing/denaturing conditions (rCE-SDS), and hydrophilic species in the front peak eluting before the main species in Hydrophobic Interaction Chromatography (HIC). The attribute analyzed for mAb B was SEC HMW; rCE-SDS separable Low Molecular Weight (LMW) species, medium Molecular Weight (MMW) species, non-glycosylated heavy chain (NGHC) species, sum of LMW MMW and NGHC (L+M+NG), and non-HC+LC; high Mannose (HM) species, CEX-separable alkaline species 3 and all alkaline species analyzed by hydrophilic interaction chromatography; and C-terminal sequence variant 1 (CSV 1) and C-terminal sequence variant 2 (CSV 2) analyzed by mass spectrometry. It is contemplated that any attribute or combination of attributes may be analyzed for the methods described herein.

At the examined attribute exposure level, we found that none of the 4 attributes of mAb a was a determinant of ADA. None of the 11 properties of mAb B was also found to be a determinant of ADA. Moreover, the upper range of exposure levels for the examined attributes is above or near the recommended purity specification for the corresponding attributes of mAb a, demonstrating that CIA can be used to generate clinically relevant evidence justifying the recommended specification.

CIA requires analysis of popularity or time course of clinical outcome for each given variable, such as different levels of exposure to a patient's attributes. The following methods were performed to integrate ADA test results and attribute exposure levels for individual patients when CIA was performed.

Selection of clinical studies

Criteria for selecting CIA clinical studies were (1) the presence of large variability in the attribute exposure levels, and (2) the large number of patients enrolled. Thus, analysis of clinical effects in the context of broad levels of attribute exposure can yield statistically significant results. Thus, dose escalation studies or studies using multiple treatment batches are selected and, if a greater number of patients is required, the patients from the equivalent study are combined for analysis.

ADA test results of clinical study

Analysis of bridging immunoassays from Electrochemiluminescence (ECL) based ¹⁹ Each of these bridging immunoassays was developed for the specific detection of ADA binding to mAb a or mAb B in patient serum. Note that binding ADA encompasses both neutralizing and non-neutralizing ADA. However, neutralizing ADA is far less common to both mAb a and mAb B than binding ADA, requiring a much larger number of patients for statistically significant analysis, and analyzing binding ADA is suitable for this example, which verifies whether any clinical outcome is relevant to the variable in treatment.

The clinical study selected for mAb a was 4 early dose escalation studies for up to 218 patients, 2 phase 2 studies for up to 627 patients, and 2 phase 3 studies for up to 542 patients. These 3 groups of studies were analyzed individually by retrospective or prospective CIA as follows. The clinical study selected for mAb B was a 2-phase 2 study with a maximum of 5,440 patients. The study set was analyzed by retrospective and prospective CIA.

Attribute analysis and stability study data

Drug lots administered in a clinical study selected for CIA are tracked using lot tracking records to determine drug lots reviewed in a product quality analysis test, so that the level of attributes determined at the time of lot placement for a single treatment lot can be obtained.

Assays for batch let-down analysis were SEC for HMW (size exclusion chromatography), rCE-SDS for HC, LC, LMW, MMW and NGHC (capillary electrophoresis under reducing and SDS denaturing conditions), and CEX for acidic or basic species (cation exchange chromatography), which were developed specifically for mAb a or mAb B alone. Other analytical data from HIC (hydrophobic interaction chromatography), hydrophilic chromatography for HM (HIC) and peptide mapping by lcms (liquid chromatography-tandem mass spectrometry) for C-terminal sequence variant species (CSV 1 and CSV 2) can be used for limited batches of mAb a or mAb B, as they are not in the group of batch release assays.

Attribute stability study data for each drug product was also mined. Stability data represents the change in individual attribute levels over time under storage conditions, which is critical to determining the attribute exposure level at the time of treatment administration as follows.

Determination of attribute exposure

Determining the attribute exposure A at the time of treatment administration associated with each treatment from the above information using equation 1 _t ：

A _t ＝(％A ₀ +％A _Δ X t) D (equation 1)

Wherein% ₀ Is the attribute percentage of the batch run,% A _Δ Is the rate of change of the percentage of the property level, t is the storage time between the release of the batch and the administration of the treatment, and D is the dose strength in the weight dimension of the active pharmaceutical ingredient associated with each treatment.

Each patient is then assigned a representative A _t . For early studies with single treatment administered to each patient, representative A _t Is used for treatingRelated A _t 。

For studies involving multiple treatment regimens, the treatment administered after the first test patient ADA positive is independent of the cause of the study ADA. Thus, for ADA-positive patients, representative A _t Is the average or maximum of the exposure levels of the attribute associated with the treatment administered prior to the first ADA test being positive. For ADA negative patients, all treatments in the whole regimen were included in determining mean or maximum exposure to obtain representative a _t 。

Patient group for CIA

Of the patients from the study selected as described above, only patients meeting the following criteria were included in CIA: (1) Batch information is available for all treatments in the regimen; (2) Attribute exposure for all treatments in the regimen can be determined; (3) ADA test results are available; and (4) no positive results were obtained from ADA tests performed prior to the administration of the first treatment. For example, one patient from the mAb B clinical study was excluded from CIA because one of the treatments lacks an entry for the drug product lot number. Likewise, 29 patients in the mAb B study were excluded because they were tested positive for ADA prior to treatment.

The application of these criteria resulted in the following patient numbers: the CIA for HMW, non-hc+lc or acidic species of mAb a was analyzed alone for 218 patients from 4 combined early studies and 627 patients from 2 combined phase 2 studies. CIA for the HIC front peak of mAb a was analyzed separately for 90 patients from 2 combined early studies and 24 patients from 2 combined phase 2 studies. The individual CIA for HMW of mAb a was additionally analyzed for 553 patients from 2 combined phase 3 studies.

For HMW, LMW, MMW, NGHC, M +n+ng, non-hc+lc, alkaline 3 and all alkaline for mAb B, 5,439 patients from 2 combined phase 2 studies were analyzed by all 3 CIA methods (fig. 1B and 1C, see below). CIA of the HM of mAb B was performed by analysis of 127 patients and, for CSV1 and CSV2, 1,671 patients from the same group of clinical studies were analyzed.

Example 2: analysis of the influence of the properties of individual patients

If the attribute elicits an ADA response, there will be a positive correlation, and thus ADA positive patients will be found to correlate with higher levels of attribute exposure than ADA negative patients (FIGS. 1A-C). Finding associations does not prove causal relationships, but such findings may suggest further research to test causal relationships. However, the absence of such association was found to eliminate the causal relationship of this attribute to ADA at the examined exposure level. Three independent assays were performed to test the correlation between the level of attribute exposure and the occurrence of ADA responses.

Representative attribute exposure levels analyzed for individual patients were determined from single exposure of patients in phase 1 studies of mAb a, maximum exposure of patients in phase 2 and 3 studies of mAb B, and maximum exposure of patients of mAb B (as described in example 1). Average A _t Is a statistical representation of multiple attribute exposures. Maximum A _t Is immunologically relevant because patients may produce ADA only when exposed to immunogens exceeding a certain threshold level ^20,21 。

Example 3: retrospective CIA of mAb a and mAb B properties

Retrospective analysis (subjects grouped according to retrospective known clinical outcome) was developed to test whether ADA positive patients correlated with higher levels of attribute exposure than ADA negative patients. Patients were grouped according to review of known ADA test results (fig. 1B). Then, representative attribute exposure level profiles for each attribute between the ADA positive patient group and the ADA negative patient group were compared. If the representative attribute exposure level of the ADA positive patient group is on average higher than the representative attribute exposure level of the ADA negative patient group, a student t-test is performed to evaluate whether the difference is statistically significant.

There was statistically no correlation between ADA positive patients and higher exposure to any of the attributes of mAb a or mAb B (fig. 2), indicating that none of these attributes caused ADA at the detected exposure level. This finding holds true whether a single exposure (fig. 2A), average exposure (fig. 1), or maximum exposure (fig. 2B) is used as a representative exposure level (see methods section). Examination of multiple sets of mAb a clinical studies at different stages synonymously indicated that none of the 4 attributes was a determinant of ADA (fig. 2A, fig. 7A, and fig. 8).

Fig. 2A-B illustrate that a comparison of the exposure levels of the attributes between ADA positive and ADA negative patients shows that there is no statistically significant correlation between high exposure levels to ADA positive patients for all the attributes analyzed, indicating that these attributes do not cause ADA in the clinical condition examined. The numbers in brackets are p values determined when needed. The line is the mean ± standard deviation of representative attribute exposures for each group of individual patients.

Example 4: prospective CIA for incidence of ADA attribute of mAb a and mAb B

According to representative A _t The incidence of grouped subjects was analyzed to test whether there was a positive correlation between the attribute exposure level and the incidence of ADA responses. Patients were grouped in quartiles according to attribute-exposure levels, and group averages of% ADA positive patients determined for each group relative to representative attribute-exposure levels were compared to each other (fig. 1C, left panel).

The exposure level range applied to create the quartile bins (bin) is optimized to achieve minimum variability in exposure levels within each patient group while ensuring an adequate number of patients making up each group. If a positive correlation is observed between the incidence of ADA and the level of attribute exposure, the number of patients in the individual group is used as a weight ²² To calculate a weighted correlation to test whether the correlation is statistically significant.

Consistent with the results of the CIA analysis of example 3, in which subjects were grouped according to retrospective known clinical results, no statistically positive correlation was found between the incidence of ADA and the exposure levels for all properties analyzed for mAb a or mAb B (fig. 3A-B, fig. 7B). In fact, lower exposure appears to be associated with ADA-positive patients with mAb B, in terms of representative A _t The incidence of grouped subjects showed a negative correlation (fig. 3B). The negative correlation is a statistical informative result that additionally confirms that no causal relationship exists (see below; FIGS. 5A-B, FIG. 9).

Fig. 3A-B show that scores of ADA positive patients in patient quartiles grouped according to representative attribute exposure levels ranging from low to high are plotted against a group average of representative exposure levels. The numbers in brackets are weighted p values determined when a positive correlation is found. The number of patients in each quartile is adjacent to the figure. Error bars are standard deviations. The results for mAb a are shown in fig. 3A and mAb B in fig. 3B.

Example 5: prospective CIA analysis of ADA time course for mAb a and mAb B attributes

Another prospective analysis was developed (based on representation a _t Grouped subjects; see fig. 1B, right panel) compares the time course of ADA responses in patients grouped in the same manner as the CIA incidence in example 4 above. If higher levels of attribute exposure correlate with a broader occurrence over time, a Mantel-Cox log rank test 23,24 is performed to test if the differences in time course are statistically significant.

Consistent with the above analysis, no correlation was found between the time course of ADA and the exposure levels of all properties of mAb a or mAb B (fig. 4A-B). For the HIC front peak of mAb a (fig. 4A), the results showing no correlation with ADA were non-informative due to the limited number of patients analyzed. The other analyses described above, performed in a different set of clinical studies (fig. 2A and 3A) using data from a large number of patients, indicated that the HIC front peak was not actually a determinant of ADA.

_t Example 6: correlation between ADA response and lower representative A

Lower A was found in both the incidence and time course of mAb A and mAb B _t Correlation or negative correlation with ADA response (fig. 2B, 3B, 4A and 4B). Maximum A was analyzed when in CIA of mAb B _t This negative correlation becomes apparent (fig. 2B, 3B, and 4B).

Using HMW exposure in mAb B-treated patients, we demonstrate that negative correlation potentiates CIA results, showing that attribute exposure does not elicit ADA responses. Absence of maximum HMW exposure between ADA-negative and ADA-positive patientsIn bias: maximum a between ADA negative and ADA positive patients _t Levels were generally not different (fig. 5A); moreover, the treatment time at which maximum HMW exposure occurred was also similarly distributed between ADA negative and ADA positive patients (fig. 5B). However, most treatments were associated with HMW exposure, particularly HMW exposure below the average of the maximum HMW exposure of 2mg in fig. 5A (fig. 5C). Without being limited by theory, it is contemplated that excluding treatment at some late administration times (due to lack of causal relationship with ADA occurrence) may result in analysis of the next highest HMW exposure as representative exposure for ADA positive patients. The use of these next-highest HMW exposures may create a bias that pushes the correlation in the negative direction. This potential bias can be considered in conducting the analysis.

In FIG. 5A, maximum A _t Is determined using all treatments or is plotted for ADA negative patients or ADA positive patients. Indicating maximum a determined using all treatments for ADA negative patients _t Mean ± standard deviation of (a); and those of the other groups are analyzed as for the same corresponding group in fig. 2B. For the comparison indicated as n.s. (insignificant) the p-value from the t-test was higher than 0.05, unless indicated otherwise. Error bars are standard deviations. In fig. 5B, the frequency of monthly treatment times associated with maximum HMW exposure in individual patients is shown for the ADA negative and ADA positive groups. In fig. 5C, the frequency of treatment resulting in different HMW exposure levels for individual mAb2 patients is plotted as a cumulative percentage of treatment times.

Only the attribute exposure of the treatment administered prior to ADA response was related to ADA response. Also, analyzing the maximal property exposure considers immunological mechanisms, as ADA may only be induced by immunogens exceeding a threshold concentration ^20,21 . Overall, the negative correlation found here provides additional evidence that attribute exposure does not cause ADA (see example 8).

Example 7: application of CIA methods in analyzing clinical impact of treatment regimens

In addition to analyzing the effects of the attributes, we tested whether there was a correlation between ADA response and various factors of the treatment regimen, including the number of treatment administrations, the dose intensity per treatment, and the number of treatment batches associated with the course of treatment for each patient.

These factors were not found to be related to the incidence of ADA. Indeed, ADA positive patients are associated with lower treatment times and durations and fewer treatment batches, since the analysis only includes treatments administered before their first positive ADA test.

This demonstrates that the CIA method can be applied to other aspects of treatment in addition to attribute exposure. We contemplate that a powerful analysis of extended drug safety and efficacy knowledge can be performed by combining attribute exposure information with patient information, such as demographic patient information or even genotype.

Example 8: additional analysis of prospective CIA occurrence

In accordance with representative A _t Of the CIA occurrence rates of the grouped subjects (fig. 9A-B), it was checked whether the slope of the linear regression was significantly different from zero by using the F-test, which calculated randomly selected points would result in or exceeded the observed linear regression correlation coefficient R ² Is a probability p of (c). Using at most A _t The p-values of the F-test, which determine significant non-zero slopes in regression of rCE-SDS non-hc+lc, CEX alkaline 3, CEX all alkaline, and CSV2, were 0.017, 0.049, 0.031, and 0.045, respectively. Other slopes were not significantly different from zero.

Summarizing FIG. 9B, and using average A _t Rather than the maximum a of fig. 9B _t The same set of data re-analyzed is compared. The line is a non-weighted linear regression and the slope of all regressions is statistically zero except those: rCE-SDS non-HC+LC using maximum At analysis, CEX alkaline substance 3 using maximum At analysis, maximum A _t All alkaline substances of CEX analyzed and use of maximum A _t And CSV2 analyzed. * The number of patients binned in the quartiles for re-analysis is the same as the corresponding quartiles in fig. 9B, unless indicated otherwise. Except for HM and CSV1 (all of their fit slopes are statistically zero)The correlation of the re-analyzed data is typically less negative.

Example 9: CIA using additional clinical studies

CIA analysis of the effect of mAb a attributes on ADA using data from phase 2 clinical studies indicated that no CQA was a determinant of ADA response at the examined exposure levels (fig. 7). The level of attribute exposure in phase 3 clinical studies of mAb a was lower than in early studies, and as per representative a _t CIA analysis of the grouped subjects showed that ADA was not induced (fig. 8).

As shown in fig. 7-B, patients receiving the regimen in clinical studies different from those analyzed in fig. 2A were analyzed. Average A for each treatment _t Representative A as monotherapy _t . Error bars are averaged from individual A _t Uncertainty of the propagation of the relevant standard deviation. The quartiles analyzed in (fig. 7B) and the attribute exposure ranges for grouping the quartiles are the same as in fig. 4A.

Low A _t Association with ADA

In the maximum attribute exposure between ADA negative and ADA positive patients, bias will be found if these attributes do not cause ADA, and ADA-related therapies are analyzed only in CIA of ADA positive patients, as established herein. Such bias will decrease when the average exposure per treatment is analyzed as representative exposure for the individual patient. For example, plausibly, bias that produces reduced maximum HMW exposure when ADA-independent treatment is excluded from CIA of ADA-positive patients may be offset by similar bias that produces increased minimum HMW exposure. Consistent with this illustration, using average exposure as a representative exposure for CIA generally reduces or eliminates these negative correlations (fig. 9). Because of A analyzed here _t The correlation with ADA and analysis of maximal exposure was immunologically significant 25,26, so the negative correlation described herein provides additional evidence that these attributes do not elicit ADA under the clinical conditions examined.

Example 10: analysis and conclusion of examples 1-9

We demonstrate that integrating information about treatment batches, clinical results, and attribute analysis data enables approximate estimation of the level of attribute exposure associated with individual treatments administered to a patient. Then, it can be checked whether the attribute would cause ADA by statistical tests that check for higher A _t The existence of a correlation with the occurrence of ADA responses. While the existence of a relationship requires further investigation of emphasis to establish a causal relationship, the absence of a relationship eliminates the causal relationship.

The work herein describes efforts to integrate treatment information for individual patients and evaluate clinically relevant effects of molecular properties of therapeutic agents. Here, no significant positive correlation was found between the ADA response examined between 2 drug products mAb a and mAb B and the exposure level of any of the 15 attributes, indicating that none of these attributes is a determinant of ADA at the examined level of the corresponding drug product (table 1). This work did not establish that these attributes do not pose an increased safety risk at different exposure levels or for other therapeutic mAb molecules.

The method of the present invention analyzes passive data mined from clinical studies originally aimed at establishing the safety or efficacy of pharmaceutical products. Ethical issues when exposing patients to arbitrary levels of attributes preclude the generation of valid data by designing studies that study the effects of attributes in patients specifically. However, CIA of passive data provides rich insight and insight into the application and basic science pursuits in biotechnology.

For example, CIA can help determine patient-centric product purity specification limits. As established above, the lack of causal relationship of attribute exposure to ADA indicates that these attributes do not cause an increase in the immune safety risk at the exposure level examined. Notably, the upper limit of exposure levels of these attributes for mAb a analysis (table 1) was close to or above the limit imposed on the corresponding attributes by the specifications of the drug product (not shown), demonstrating that patient-focused evidence produced by CIA can be used to determine the limits of therapeutic specifications.

Fig. 6 illustrates the effect of factors in treatment on ADA. Various factors of mAb B treatment regimens were compared between ADA negative patients and ADA positive patients: average dose intensity per treatment, duration of treatment, number of treatments and treatment lot number. These factors pertain to all treatments of ADA negative patients, or to treatments of ADA positive patients prior to the first ADA positive test results. The error is the standard deviation.

Table 1. Means and maxima (mg) of representative attribute exposures analyzed for individual patients in cia.

The values for mAb a are from fig. 2A and fig. 3A. />

The values for mAb B are from fig. 2B, fig. 3B and fig. 4B.

CIA can also deepen the basic knowledge of clinical immunology. Any association found between ADA response and attribute exposure levels may suggest focusing on in vitro or in vivo studies to investigate causal relationships or mechanisms of ADA production. The assumption may also be narrowed to verify whether repeated high-attribute exposures or single high-attribute exposures are critical for inducing ADA responses, depending on whether average or maximum better represents any attribute exposures found to be associated with ADA.

We further demonstrate that CIA can be combined with other assays to provide powerful implications in therapeutic agent development and to address unknown key issues. Is the attribute affecting the clinical outcome of a subset of the population defined by its demographic or genomic information? Is the resulting demographic and genomic knowledge able to drive clinical trial design? Whether the properties will alter the biophysical behavior of different therapeutic protein molecules such as structure and conformational dynamics, thereby affecting clinical outcome? The present work reports data and analysis that acts as a pioneer for these and further discussion.

In summary, understanding the impact of molecular attributes on drug safety and efficacy is critical to therapeutic development. For example, some attributes may lead to a patient developing a drug-resistant antibody (ADA) response, which remains an important regulatory issue due to potential clinical effects, including neutralization of drugs. Non-human model systems, such as cells, tissues, and animals, are viable experimental systems for testing the safety impact of attributes, but they provide limited insight into the clinical outcome of patient exposure to attributes. Here, we examined the clinically relevant immune safety impact of 15 attributes between 2 immunoglobulin G2 (IgG 2) monoclonal antibody (mAb) drug products mAb a and mAb B using a data analysis method called Clinical Impact Analysis (CIA). CIA verifies whether there is a correlation between the level of exposure of a patient to a given attribute and the extent of occurrence of clinical outcome by analyzing data from clinical studies and product quality analysis studies. No correlation was found between the level of exposure of the patient to any of the attributes and the ADA response, indicating that the analyzed attributes did not elicit ADA response at the examined exposure level. Moreover, the upper range of exposure levels examined was above or near the purity specification limit of the mAb a attributes, indicating the feasibility of producing clinically relevant evidence to demonstrate that CIA made these attributes possible. These results show that integrating clinical and analytical information can provide clinically relevant insight into the impact of attributes suitable for drug development.

Example 11: product C (a specification Molecules) analysis of

For product C (a specificationMolecules) were further analyzed. The measured molecular properties are Tyr sulfation, Asp isomerisation, lys hydroxylysine, lys glucosyl-galactosyl hydroxylysine, met oxidation and SE-HPLC aggregates. Efficacy was also assessed. The security data includes determining the following adverse events: anemia, cytokine Release Syndrome (CRS), fever, infusion-related reactions (IRR), lymphopenia, and neurological events.

The attribute exposure is estimated based on: 1) the% level of the property of the different drug product batches at the time of manufacture, 2) the stability of the property, or the rate at which the property level changes during the duration of storage of the Drug Product (DP), 3) the date and time of administration of the individual treatments to each patient, and a record of the drug product batches administered per treatment. Then, 4) the date and time of occurrence of a given clinical outcome with a grade (or severity) of interest can be incorporated into the CIA for the next step (fig. 1), wherein 5) the CIA combines this information to determine attribute exposure, and 6) analyze whether there is a correlation between the attribute exposure level and the occurrence of the clinical outcome of interest.

(A) For all relevant treatments as defined below, a is first determined using equation 1 _t 。

(1) For patients that did not show clinical outcome of interest throughout the clinical study, the relevant treatments were all treatments administered to the patient throughout the study.

(2) For patients exhibiting clinical outcome, the relevant treatment is all treatments administered to the patient prior to the occurrence of the clinical outcome.

(3) For patients who develop clinical outcome prior to receiving any treatment, there is no relevant treatment. A of these patients _t Is 0.

(4) For administration via continuous infusion, a occurring during the relevant part of the infusion is determined as follows _t ：

a. For infusions administered continuously over 24 hours, A was determined for each 24 hour period of relevant infusions prior to the occurrence of clinical outcome _t 。

b. For infusions less than 24 hours long, A associated with each associated infusion before clinical outcome occurs is determined _t 。

c. For the upper partAny of the 2 cases, if a clinical outcome occurs during infusion, determining a related to the portion of the infusion prior to the occurrence of the clinical outcome _t 。

The analysis of product C is shown in FIGS. 10A-B. In addition to the 6 CQAs listed for CIA for product C, the effect of efficacy on AE was also analyzed for patients not administered, administered medium dose levels, or administered high dose levels by testing whether there was a correlation between product C dose scaled by efficacy value and AE incidence. As shown in fig. 10A, the correlation of efficacy with any grade of AE incidence was analyzed. As shown in fig. 10B, the correlation of efficacy with the occurrence of AE of grade 3 or higher was analyzed.

No attribute was observed to be associated with any adverse events tested for. Thus, it is safe to determine an estimated level of attribute exposure.

Example 12: computer implementation of bayesian estimation

Computer programs were developed to take advantage of the high computational power required by bayesian methods. The program provides an interactive platform to visualize the results of the bayesian estimation method using a user-provided spreadsheet. The program also provides a parameter adjustment module to define an initial prior, remove outliers and select a plurality of attributes. A set of processed clinical data for a method of assessing the clinical impact of the molecular properties of mAb D and some modeling data based on mAb D CIA data were examined systematically. The method is performed to verify whether there is a positive correlation between mAb D HMW exposure and the incidence of febrile adverse events by comparing the maximum daily HMW exposure between adverse event negative and positive subjects. This method did not find a correlation using the conventional method (fig. 11A). The bayesian approach (fig. 11C-D) produces similar results to the conventional system (fig. 11A-B), demonstrating the effectiveness of the system:

(1) When analyzing clinical data of mAb D clinical trial, bayesian estimation method indicated that there was no possibility of attribute correlation with clinical outcome (fig. 11C). When data is processed using conventional statistics, the conventional statistics indicate the same result (FIG. 11A), whereas

(2) When the modeled dataset is analyzed (based on the clinical data), but where the exposure level of the attribute of the clinical study subject increases arbitrarily, the bayesian estimation method indicates the likelihood that the attribute is correlated with the clinical outcome (fig. 11D), as is conventional statistics (fig. 11B). Thus, both conventional statistical and bayesian estimation methods indicate a correlation between clinical outcome and any increased level of attribute exposure.

Reference to the literature

The following documents are incorporated herein by reference in their entirety.

1Rosenberg,A.S.,Verthelyi,D.&Cherney,B.W.Managing uncertainty:aperspective on risk pertaining to product quality attributes as they bear on immunogenicity of therapeutic proteins.J Pharm Sci 101,3560-3567,doi:10.1002/jps.23244(2012).

2 Goetze,A.M.,Schenauer,M.R.&Flynn,G.C.Assessing monoclonal antibodyproduct quality attribute criticality through clinical studies.MAbs 2,500-507(2010).

3 Kelley,B.,Cromwell,M.&Jerkins,J.Integration of QbD risk assessment toolsand overall risk management.Biologicals 44,341-351,doi:10.1016/j.biologicals.2016.06.001(2016).

4 Bessa,J.et al.The immunogenicity of antibody aggregates in a novel transgenicmouse model.Pharm Res 32,2344-2359,doi:10.1007/s11095-015-1627-0(2015).

5 Bi,V.et al.Development of a human antibody tolerant mouse model to assessthe immunogenicity risk due to aggregated biotherapeutics.J Pharm Sci 102,3545-3555,doi:10.1002/jps.23663(2013).

6 Bee,J.S.,Goletz,T.J.&Ragheb,J.A.The future of protein particlecharacterization and understanding its potential to diminish the immunogenicity ofbiopharmaceuticals:a shared perspective.J Pharm Sci 101,3580-3585,doi:10.1002/jps.23247(2012).

7 Goetze,A.M.,Liu,Y.D.,Arroll,T.,Chu,L.&Flynn,G.C.Rates and impact ofhuman antibody glycation in vivo.Glycobiology 22,221-234,doi:10.1093/glycob/cwr141(2012).

8 Jawa,V.et al.Evaluating Immunogenicity Risk Due to Host Cell ProteinImpurities in Antibody-Based Biotherapeutics.AAPS J 18,1439-1452,doi:10.1208/s12248-016-9948-4(2016).

9 Joubert,M.K.et al.Use of In Vitro Assays to Assess Immunogenicity Risk ofAntibody-Based Biotherapeutics.PLoS One 11,e0159328,doi:10.1371/journal.pone.0159328(2016).

10 Liu,Y.D.et al.Human IgG2 antibody disulfide rearrangement in vivo.J BiolChem 283,29266-29272,doi:10.1074/jbc.M804787200(2008).

11 Liu,Y.D.,van Enk,J.Z.&Flynn,G.C.Human antibody Fc deamidation in vivo.Biologicals 37,313-322,doi:10.1016/j.biologicals.2009.06.001(2009).

12 Yang,J.,Goetze,A.M.&Flynn,G.C.Assessment of naturally occurringcovalent and total dimer levels in human IgG1 and IgG2.Mol Immunol 58,108-115,doi:10.1016/j.molimm.2013.11.011(2014).

13 Zhang,Q.et al.Characterization of the co-elution of host cell proteins withmonoclonal antibodies during protein A purification.Biotechnol Prog 32,708-717,doi:10.1002/btpr.2272(2016).

14 Seamon,K.B.Specifications for biotechnology-derived protein drugs.CurrOpin Biotechnol 9,319-325(1998).

15 Rathore,A.S.Setting Specifications for a Biotech Therapeutic Product in theQuality by Design Paradigm.BioPharm Internaltional 23(2010).accessible on the worldwide web at www dot biopharminternational dotcom/setting-specifications-biotech-therapeutic-product-quality-design-paradigmid＝&pageID＝1&sk＝&date＝.

16 Rathore,A.S.Roadmap for implementation of quality by design(QbD)forbiotechnology products.Trends Biotechnol 27,546-553,doi:10.1016/j.tibtech.2009.06.006(2009).

17 Casadevall,N.et al.Pure red-cell aplasia and antierythropoietin antibodies inpatients treated with recombinant erythropoietin.N Engl J Med 346,469-475,doi:10.1056/NEJMoa011931(2002).

18 Li,J.et al.Thrombocytopenia caused by the development of antibodies tothrombopoietin.Blood 98,3241-3248(2001).

19 Bautista,A.C.,Salimi-Moosavi,H.&Jawa,V.Universal immunoassay appliedduring early development of large molecules to understand impact of immunogenicity onbiotherapeutic exposure.AAPS J 14,843-849,doi:10.1208/s12248-012-9403-0(2012).

20 Sauerborn,M.,Brinks,V.,Jiskoot,W.&Schellekens,H.Immunologicalmechanism underlying the immune response to recombinant human protein therapeutics.Trends Pharmacol Sci 31,53-59,doi:10.1016/j.tips.2009.11.001(2010).

21 Baker,M.P.,Reynolds,H.M.,Lumicisi,B.&Bryson,C.J.Immunogenicity ofprotein therapeutics:The key causes,consequences and challenges.Self Nonself 1,314-322,doi:10.4161/self.1.4.13904(2010).

22 Sauna,Z.E.,Lagassé,D.,Pedras-Vasconcelos,J.,Golding,B.&Rosenberg,A.S.Evaluating and Mitigating the Immunogenicity of Therapeutic Proteins.Trends Biotechnol36,1068-1084,doi:10.1016/j.tibtech.2018.05.008(2018).

23 Mantel,N.Evaluation of survival data and two new rank order statistics arisingin its consideration.Cancer Chemother Rep 50,163-170(1966).

24Peto,R.&Peto,J.Asymptotically Efficient Rank Invariant Test Procedures.Journal of the Royal Statistical Society.Series A(General)135,185-207,doi:10.2307/2344317(1972).

25Sauerborn,M.,Brinks,V.,Jiskoot,W.&Schellekens,H.Immunological mechanism underlying the immune response to recombinant human protein therapeutics.Trends Pharmacol Sci 31,53-59,doi:10.1016/j.tips.2009.11.001(2010).

26Baker,M.P.,Reynolds,H.M.,Lumicisi,B.&Bryson,C.J.Immunogenicity of protein therapeutics:The key causes,consequences and challenges.Self Nonself 1,314-322,doi:10.4161/self.1.4.13904(2010).

All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein.

The use of the terms "a" and "an" and "the" and similar referents in the context of describing the disclosure (especially in the context of the following claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. Unless otherwise indicated, the terms "comprising," "having," "including," and "containing" are to be construed as open-ended terms (i.e., meaning "including, but not limited to").

The terms "patient" and "subject" are used interchangeably herein. Generally, these terms are understood to mean a human. In some embodiments of the methods, the patient or subject is a human.

Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range and each endpoint, unless otherwise indicated herein, and each separate value and endpoint is incorporated into the specification as if it were individually recited herein.

All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., "such as") provided herein, is intended merely to better illuminate the disclosure and does not pose a limitation on the scope of the disclosure unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the disclosure.

Preferred embodiments of this disclosure are described herein, including the best mode known to the inventors for carrying out the disclosure. Variations of those preferred embodiments may become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the disclosure to be practiced otherwise than as specifically described herein. Accordingly, this disclosure includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, unless indicated otherwise or clearly contradicted by context, the present disclosure covers any combination of all possible variations of the elements described herein.

Claims (25)

1. A method of manufacturing a biologic therapeutic, the method comprising:

detecting the level of the molecular property of the biological therapeutic in the formulation at one or more time points under storage conditions;

determining the rate of change of the molecular property under the storage conditions;

obtaining data regarding the in vivo safety and/or efficacy of the biologic therapeutic in a subject who has received administration of the biologic therapeutic;

estimating the level of molecular property exposure that the subjects receive at said administration based on (i) the rate of change of the molecular property during storage of the biological therapeutic in the formulation and (ii) the duration of time the biological therapeutic in the formulation is under the storage conditions prior to said administration;

determining whether there is a correlation between the estimated level of molecular property exposure and safety and/or efficacy data for the biological therapeutic; and

if the correlation is not present, a production lot of the biologic therapeutic containing the molecular property at or below a specified allowable level of the molecular property is manufactured based on the estimated level of molecular property exposure.

2. The method of claim 1, wherein if the correlation is present, the method further comprises setting a specification of the molecular property level that does not exceed a maximum allowable level of the molecular property of the biological therapeutic at the time of manufacture,

Wherein said maximum allowable level of the molecular property is based on a highest estimated level of molecular property exposure that is not associated with adverse events and/or efficacy inhibition of the biologic therapeutic,

the manufacturing further includes rejecting the production lot of the biologic therapeutic that contains the level of the molecular property that exceeds the maximum allowable levels.

3. The method of claim 2, wherein a specified maximum allowable level of the molecular property is calculated to produce a property exposure level at the end of the shelf-life that is less than or equal to 90% -100% of the highest estimated level of the molecular property exposure that is not related to adverse events and/or efficacy inhibition in the subject.

4. A method of developing a manufacturing process for a biologic therapeutic, the method comprising:

detecting the level of the molecular property of the biological therapeutic in the formulation at one or more time points under storage conditions;

determining the rate of change of the molecular property under the storage conditions;

obtaining data regarding the safety and/or efficacy of the biologic therapeutic in a subject that has received administration of the biologic therapeutic;

estimating the level of molecular property exposure that the subjects receive at said administration based on (i) the rate of change of the molecular property during storage of the biological therapeutic in the formulation and (ii) the duration of time the biological therapeutic in the formulation is under the storage conditions prior to said administration;

Determining whether there is a correlation between the estimated level of molecular property exposure and safety and/or efficacy data for the biological therapeutic; and

(a) If the correlation is not present, establishing the manufacturing process based on the estimated level of molecular property exposure to produce the molecular property level at or below a specified allowable level; or alternatively

(b) If the correlation is present, the manufacturing process is established based on a highest level of the molecular property that is not associated with adverse events and/or efficacy inhibition of the biotherapeutic agent to produce the molecular property level at or below a specified maximum allowable level of the molecular property.

5. A method of assessing the clinical impact of a molecular profile of a biologic therapeutic, the method comprising:

detecting the level of the molecular property of the biological therapeutic in the formulation at one or more time points under storage conditions;

determining the rate of change of the molecular property under the storage conditions;

obtaining data regarding the safety and/or efficacy of the biologic therapeutic in a subject that has received administration of the biologic therapeutic;

estimating the level of molecular property exposure that the subjects receive at said administration based on (i) the rate of change of the molecular property during storage of the biological therapeutic in the formulation and (ii) the duration of time the biological therapeutic in the formulation is under the storage conditions prior to said administration;

Determining whether there is a correlation between the estimated molecular property exposure and the safety and/or efficacy of the biotherapeutic agent; and

if (a) the correlation is not present, determining that the molecular property does not affect neither the clinical safety nor the efficacy of the biologic therapeutic; or alternatively

If (b) the correlation is present, determining the molecular property may affect the safety and/or efficacy of the biologic therapeutic.

6. The method of claim 5, wherein if (a) the correlation is not present, the method further comprises setting a specification of allowable levels of molecular properties of the biologic therapeutic, wherein the allowable levels of molecular properties are based on the highest estimated levels of molecular property exposure accepted by the subjects, or

Wherein (b) if the correlation is present, the method further comprises setting a specification of a maximum allowable level of a molecular property of the biologic therapeutic, wherein said maximum allowable level of the molecular property is based on the level of the molecular property associated with adverse events and/or efficacy inhibition of the biologic therapeutic.

7. The method of any one of the preceding claims, wherein the estimation is further based on (iii) the dose of the biologic therapeutic in said administration, and (iv) the amount of the molecular property measured at the time of manufacture and/or batch release.

8. The method of any one of the preceding claims, wherein the level of the molecular property of the biologic therapeutic in the formulation is detected at two or more time points under storage conditions.

9. The method of any one of the preceding claims, wherein the one or more points in time or two or more points in time comprise at the time of manufacture and at least two subsequent points in time.

10. The method of any one of the preceding claims, wherein estimating the level of exposure of the molecular property comprises the following calculations:

wherein A is _t Is an estimated level of molecular property exposure,% A ₀ Is the percentage of the molecular property,% A, at batch placement _Δi Is the rate of change, t, of the percentage of the molecular property level over time under given storage conditions _i Is the time stored under the given conditions, and D is the dose strength of the administration.

11. The method of any one of the preceding claims, wherein estimating the level of exposure of the molecular property comprises the following calculations:

wherein% _rel. Is the molecular attributeThe relative level of exposure to dose, where A is the percentage of _t Is the level of exposure of this molecular property calculated using equation 1 or 2, and where D is the dose intensity in the weight dimension of the active pharmaceutical ingredient associated with each treatment.

12. The method of any one of the preceding claims, wherein the correlation comprises a weighted correlation between attribute exposure and occurrence of adverse events.

13. The method of any one of the preceding claims, wherein the biologic therapeutic in the formulation is in the storage conditions for different durations in the administrations to different subjects.

14. The method of any one of the preceding claims, wherein the administration comprises two or more administration events.

15. The method of claim 14, wherein the estimated level of molecular property exposure that the subjects receive at said administering is the maximum or average of the two or more administration events.

16. The method of any one of the preceding claims, wherein the administering comprises continuous infusion and the estimating comprises calculating an estimated level of molecular property exposure during two or more intervals of the continuous infusion, such as 24 hour intervals.

17. A method as claimed in any one of the preceding claims, wherein the security data comprises adverse event data.

18. The method of claim 17, wherein the security data includes adverse event time courses.

19. The method of any one of the preceding claims, wherein the efficacy data comprises clinical endpoint data.

20. The method of any one of the preceding claims, wherein the molecular property comprises at least one of: acidic species, basic species, high molecular weight species, sub-visible particle numbers, low molecular weight, medium molecular weight, glycosylation (such as non-glycosylated heavy or high mannose), non-heavy and light chains, deamidation, deamination, cyclization, oxidation, isomerization, fragmentation/scission, N-and C-terminal variants, reducing species and partial species, folding structures, surface hydrophobicity, chemical modifications, covalent bonds, C-terminal amino acid motif PARG or C-terminal amino acid motif PAR-amides.

21. The method of any one of the preceding claims, wherein the molecular property comprises at least one of: acidic species, basic species, high molecular weight species, amino acid isomers or sub-visible particle numbers.

22. The method of any one of the preceding claims, wherein the biologic therapeutic is selected from the group consisting of: antibodies, antigen-binding antibody fragments, antibody protein products, bispecific T cell conjugates Molecules, bispecific antibodies, trispecific antibodies, fc fusion proteins, recombinant viruses, recombinant T cells, synthetic peptides, and active fragments of recombinant proteins.

23. The method of any one of the preceding claims, wherein the formulation is a pharmaceutically acceptable formulation.

24. The method of any one of the preceding claims, wherein detecting the level of molecular properties of the biological therapeutic agent comprises mass spectrometry, chromatography, electrophoresis, spectrometry, photoresistance, particle methods (such as nanoparticle/visible/micron-scale resonance mass or brownian motion), analytical centrifugation, imaging or imaging characterization, or immunoassay.

25. The method of any one of the preceding claims, wherein determining whether there is a correlation between the estimated level of molecular property exposure and safety and/or efficacy data of the biologic therapeutic comprises bayesian estimation.