A Parenteral Permissible Daily Exposure for Inactivated Therapeutic Proteins: An Approach Based on Literature Review

HTTPS://WWW.ISTOCKPHOTO.COM

In multiproduct biopharmaceutical manufacturing facilities, cross-contamination with pharmacologically active proteins must be controlled in a good manufacturing practice (GMP) environment (1, 2). Although guidance on control strategies exists for solvents and small-molecule pharmaceutical impurities, that is not directly applicable to inactivated (e.g., denatured and/or degraded) therapeutic proteins (TPs) occurring as impurities in a drug substance (DS) and/or drug product (DP). Small-molecule drugs and TPs differ in their molecular structures, pharmacological mechanisms of action, hazards, and potential impurities, so their cross-contamination control strategies also should be considered differently. Unlike small molecules, TPs are known to denature and degrade when exposed to pH extremes and/or heat and thus are expected to become pharmacologically inactive during the cleaning process (2).

In cleaning activities, permitted daily exposure levels (PDEs) support the amount of residual DS that poses negligible risk to patient safety if it is present as an impurity in another drug. The PDE is a daily dose of a compound that is not expected to cause adverse effects (pharmacological or toxicological). Note that PDEs are established based on the activity of TPs as intact, pharmacologically active molecules (2–4). However, given the inactivation of proteins during cleaning, using PDEs that were established based on the pharmacological activity of a DS is not applicable (1).

Our aim herein is to examine available data to derive and support a protective default PDE for denatured and/or degraded TPs that present in parenteral (intravenous, intramuscular, or subcutaneous) DS and DP. We refer to such denatured and/or degraded TPs as inactivated TPs.

Characteristics of Therapeutic Proteins: Numerous TP modalities are in development and on the market: monoclonal antibodies (MAbs), antibody fragments, fusion proteins, and other biopharmaceuticals such as therapeutic enzymes. TPs exert their pharmacodynamic effects by binding to receptors or targeting particular antigens involved in the pathophysiology of disease. TPs can be fully human, humanized (e.g., with protein sequences modified to increase their similarity to human antibodies), and/or chimeric (consisting of human and nonhuman proteins). Although TPs have varying arrangements of large peptide and/or protein constituents, they are composed of amino acid (AA) chains — which are the building blocks of all proteins.

Proteins have four different levels of structure: primary, secondary, tertiary, and quaternary (Figure 1) (5). Quaternary structure is critical for a TP’s ability to interact with molecules in the body, and the relationship between conformation and function is crucial to ensuring pharmacological activity (6, 7).

Figure 1: Protein structure, degradation, and denaturation; proper folding and conformation of a therapeutic protein (TP) are critical for its ability to exert pharmacological activity. Once disrupted by denaturation and/or degradation, a TP is not anticipated to have pharmacological activity and thus is regarded as being “inactivated.”

Impurities in Therapeutic Proteins: TPs generally consist of three-dimensionally (3D) arranged AA chains produced through expression in biological organisms — e.g., Chinese hamster ovary (CHO) cells or Escherichia coli bacteria (8). During TP manufacturing, a number of impurity types can make their way into a DS: e.g., host-cell proteins (HCPs), cell debris, leachables from equipment, and active/variant TP products (Figure 2). Inactivated TP (after cleaning activities) from previously manufactured TP can carry over into a new batch of TP (whether the same or a different product).

Figure 2: Overview of TP manufacturing and potential process-related impurities, which may be derived at any step of a process and be detected in quality control (QC) samples after cleaning activities; such impurities (as defined by ICH Q6B) include cell-culture media components, host-cell proteins, DNA, leachables from equipment and purification columns, and TP fragments/aggregates. This list of impurities is not extensive; other impurities can be formed and/or present in some processes.

Of all the impurities mentioned, an important carryover risk is DS from a previously manufactured product, which has been concentrated in downstream steps and was intended to have pharmacological activity. Upon completion of cleaning activities, residual TPs are expected to be inactivated. This assessment addresses the acceptability of a level of inactivated TP that is present as an impurity in a DS and/or DP from a previously manufactured batch of the same or a different product.

TP Inactivation
TPs generally are unstable under normal environmental conditions (e.g., exposure to light and ambient temperatures) and sensitive to physical and chemical degradation (9). Therefore, they require strict practices for their handling, administration, and storage — often at specific temperatures in solution with buffers and/or stabilizers. Specific formulations and modifications often are critical to improving their stability and preserving their pharmacological activity in DP development (9–12).

TPs lose their specific pharmacological activity when the molecular structure necessary for their pharmacological effect(s) is altered or destroyed (13). That loss of activity is known as protein inactivation. In principle, it can occur through two distinct processes: denaturation or degradation.

Denaturation of Therapeutic Proteins: Protein denaturation is the disruption and destruction of a protein’s secondary, tertiary, and quaternary structures. Such uncoiling and disruption of higher-order structures typically comes as a consequence of chemical processes or physical stress (Figure 1). Examples of denaturing agents include alcohol, which disrupts hydrogen bonds in secondary and tertiary structures; acids, bases, and heavy-metal salts, which can disrupt salt bridges in tertiary structures; heat, which can disrupt hydrogen bonds and nonpolar hydrophobic interactions; and reducing agents, which can disrupt disulfide bonds (10, 14, 15).

Degradation of Therapeutic Proteins: Degradation occurs when the primary structure of a TP — its AA chain — is fragmented, usually through hydrolysis, but degradation may be spontaneous on occasion. The process can be catalyzed by compounds such as enzymes, metal salts, acids, and bases (e.g., sodium and potassium hydroxides) and can be accelerated through heating (16, 17). Once its primary structure has been degraded, then a protein’s secondary, tertiary, and quaternary structures are also disrupted (18).

Denaturation and Degradation During Equipment Cleaning: In practice, a number of denaturation and degradation methods are applied to clean multiuse equipment in biopharmaceutical production facilities. Biopharmaceutical cleaning cycles generally are designed to expose product-contact equipment to extremes of pH (<2 and >13) and temperature (60–80 °C) for several minutes, yielding undefinable proteinaceous material or peptide fragments that lack specific biological activity (19). Upon completion of such cleaning activities, no recognizable TP structures should remain in the equipment. To demonstrate the effectiveness of cleaning procedures, quality control (QC) samples demonstrating control of carryover risk often are taken from the worst-case location(s) in equipment and tested for the presence of residues after final rinsing of cleaning agents (20).

Potential Hazards of Inactivated TPs
Inactivated TPs are an unspecified mixture of undefinable proteinaceous material, essentially endogenous substances (AAs). Studies of such compounds would be unsuitable for setting health-based exposure limits. Additionally, determining a definitive quantitative PDE for variable, inactive proteinaceous material is infeasible. Consequently, a pragmatic exposure limit for inactivated TPs must be based on reasonably conservative assumptions that consider the basic properties of the human immune system.

Biological Activity: For TPs such as antibodies, the intact molecules (sequence, structure, and posttranslational modifications) are necessary to impart full potency (ability to bind to cellular receptors or intended targets) and stability in a person’s bloodstream (21–23). The probability for a component of a denatured and/or degraded TP to refold or present the proper structure and modifications necessary to effect a biological function is considered to be minimal. Therefore, a degraded protein fragment or peptide is not anticipated to undergo specific changes that result in pharmacological activity at another (unintentional) target receptor, so they are expected to be pharmacologically inactive (inactivated).

When comparing general hazards associated with inactivated and intact TPs, the former are considered to be less hazardous for several reasons. Note that proteins and their metabolic products (AAs) are endogenous to all living organisms. In living cells, proteins are constantly synthesized and degraded (e.g., into smaller peptides and individual AAs). That makes classical biotransformation studies such as those performed for small-molecule pharmaceuticals unnecessary for TPs (24, 25). The regulatory position reflects a general belief that degradation products of TPs, unlike those of small-molecule drugs, have limited potential to cause unexpected off-target activity (26, 27). Additionally, biotechnology-derived therapeutics are not tested for genotoxicity or carcinogenicity because they are not expected to interact directly with DNA or other chromosomal material (25). That expectation also could apply to degraded proteins. Additionally, from a safety standpoint, the potential toxicity (e.g., pharmacological activity) of an inactivated TP is expected to be negligible in comparison with the respective active TP.

Absorption, Distribution, Metabolism, and Elimination (ADME): Absorption: Upon parenteral administration, even a small quantity of inactivated TP present as an impurity in a DP can disperse immediately throughout the 3.5–5 L of blood volume (IV) in an average human adult — or slowly from muscle or other tissue (IM, SC) to enter circulation (a lower C_max value). Upon tissue uptake, metabolism/catabolism of inactivated TPs takes place before the remains are excreted as smaller peptides and AA degradants — or are recycled for synthesis into other proteins.

Distribution: Because of their size and hydrophilic nature, circulating inactivated TPs that lack secondary, tertiary, and quaternary structures are expected to have a low ability to bind to cellular receptors and be internalized by cells (28). Tissue distribution is limited for large, inactivated TPs not only because of their size, but also their charge and binding properties. Therefore, inactivated TPs in a parenterally administered drug can be expected to remain in circulation, where specific proteases can degrade them further into smaller protein fragments.

Metabolism: Although proteolysis occurs widely in humans and animals, its kinetics and mechanistic details are poorly understood, especially for large TPs such as MAbs (27, 29). Products of degradation from cellular proteins are transferred from tissue into systemic circulation by the lymphatic system through a highly regulated process that protects endogenous proteins from uncontrolled degradation (27, 30).

Elimination: TPs are cleared through the same catabolic pathways used to eliminate endogenous and dietary proteins, and the same is expected for their inactivated counterparts (30). Immunoglobulin G (IgG) clearance occurs mainly through intracellular catabolism by lysosomal degradation into AAs upon cellular uptake, with a small amount cleared through biliary excretion (30, 31). Renal excretion plays a major role in elimination of protein degradation products smaller in molecular weight (MW) than the glomerular filtration threshold (~55 kDa). Proteins and peptides <30 kDa are filtered most efficiently by our kidneys and have a short half-life in circulation, usually between two and 30 minutes, because of proteolytic degradation and the fact that they are not reabsorbed in the renal tubules (32).

Immunogenicity is a general concern with the administration of biological materials. Risk factors for potential immunogenicity hazards include the proportion of foreign protein present, the stability of the proteins, and their tendency to aggregate.

Foreign Proteins: The immunogenic potential of a biologic increases with the proportion of foreign protein present. Thus, humanized proteins are less likely to cause a systemic immune response than are chimeric (e.g., murine) antibodies (33, 34).

Protein Stability and Aggregation: Sensitization to a protein allergen generally is anticipated to be more likely when such proteins preserve their native, 3D structure after chemical, physical, or enzymatic interactions (35). Such properties are extremely rare, but they have been reported for some major food allergens.

Aggregates of intact proteins generally have reduced activity and — more important — greater immunogenicity potential because of their multiplicity of epitopes and/or conformational changes. Concentration-dependent antibody aggregation is a great challenge during development of highly concentrated TP formulations. The recommended allowable aggregate level in commercial intravenous immunoglobulin products is limited to <5% (36).

Several features give inactivated TPs lower potential for immunogenicity than that of their intact, active counterparts. Proteins and their metabolic products (AAs) are endogenous to all living things; thus, so are the AAs and proteins resulting from denaturation/degradation of TPs. Note, however, that chimeric proteins have some nonhuman sequences that could be immunogenic. Structural integrity is important for allergens, as demonstrated in studies showing that active and denatured allergens — beta lactoglobulin (BLG), alpha lipoic acid (ALA), and beta casein — had reduced antibody-binding capacity from their loss of conformational epitopes (35). Known allergenic proteins contain certain motifs and conformations that are critical for allergenicity, whereas inactivated TPs do not have tertiary or quaternary structures and thus would not retain such activity. That expectation is consistent with the properties of residues after postprocessing techniques used in the food industry to reduce oral allergies (e.g., enzymatic hydrolysis and heat treatment). For example, heat-treated protein hydrolysates often are described as “hypoallergenic” formulas (37).

In allergies, IgE antibodies are produced against specific epitopes from foreign proteins or glycoproteins. Repeated exposure to the same epitope is required for type 1 hypersensitivity responses. Under the harsh conditions of cleaning methods used in biomanufacturing, the resulting fragmenting and degradation should not produce consistently similar protein epitopes at sufficient concentrations to induce type I hypersensitivity at low exposures (e.g., 100 µg/day). In conjunction with the unlikelihood of de novo epitopes being generated during inactivation, it is reasonable to consider that the allergenicity of degraded TPs is considerably lower than that of common environmental allergens or the parent TP.

Considering Available Limits
It is important to remember that what is in question is the safety of an additional amount of inactivated TP added to a given DP formulation. Essentially, what amount of inactivated TP is not anticipated to pose a safety risk to patients if it is present in a DP? For residual inactivated TPs, the goal is to determine not necessarily the highest level possible, but rather an acceptable level that could be justified using available scientific information that leverages historical safety data.

Risk Assessment Process (RAP) maps published by Jolly et al. in 2022 present a framework to facilitate the establishment of health-based exposure limits (HBELs) for endogenous compounds (38). Because of the general lack of formal toxicological studies and exposure information on endogenous substances, the RAPMAP framework includes evaluating whether an existing limit can be used or adapted to establish an HBEL. What follows is an overview of relevant available limits along with an evaluation as to how each limit could be used and/or adapted to accommodate the anticipated hazards and nature of inactivated TPs. Upon evaluating these limits (Table 1), a protective PDE for inactivated TPs can be established at 100 µg/dose.

Table 1: Published limits and their proposed applicability to inactivated therapeutic proteins (TPs) relative to the proposed permissible daily exposure (PDE) of 100 μg/dose; DP = drug product, DS = drug substance, HCP = host-cell protein, HMW = high molecular weight, IV = intravenous.

Applying Limits for Intact/Active TPs to Residual Inactivated TPs: In 2017, Pfister et al. proposed a default PDE of 10 µg/day for a parenterally administered MAb (Table 1) based on historical evaluation of PDEs for other MAbs (39). To extrapolate from a pharmacologically active dose to a “no observed adverse effect level” (NOAEL), ICH Q3C proposes a factor of 10 (40). Using that 10-µg/day exposure limit for pharmacologically active and intact TPs as a benchmark — and with the conservative assumption that inactivated TPs will be 10-fold less active/potent — gives a default PDE of 100 µg/day. Exposure at or below that limit is expected to pose negligible safety concern for inactivated TPs.

From a toxicological perspective, material derived from degradation and/or denaturation of proteinaceous or peptide TPs other than MAbs (those without a globulin structure) can be regarded to have properties similar to those of inactivated MAbs — provided that those other molecules are completely inactivated. The PDE for inactivated protein residues is independent of the potency or modality of intact TPs.

Applying Limits for Intact/Active Host Cell Proteins to Residual Inactivated TPs: Besides contamination from carryover, proteinaceous impurities in TPs also can derive from biomanufacturing processes (Figure 2). Because intact HCPs sometimes trigger unpredictable immunogenic responses, regulatory guidelines stipulate that such proteins need to be identified and quantified to protect patient safety (41).

For example, production of TPs in CHO cells yields low levels of CHO proteins (CHOPs, considered to be process-related impurities) in resulting DPs. Specifications placed on final DPs thus include HCP levels of <100 ppm (36). A recent report indicates that the most likely range of HCPs in biologic products reviewed by the US Food and Drug Administration (FDA) is 1–100 ppm (42–44). If the dose of a TP is 1,000 mg (1 g), then the acceptable tolerance limit of 100 ppm is consistent with the PDE of 100 μg/dose (42, 45).

A limit of 0.1 mg/dose (100 µg/dose) has been proposed for residual HCPs based on the NOAEL from a keyhole limpet hemocyanin (KLH) antigen study in monkeys with CHOPs (46). The proposed PDE of 100 μg/dose should be protective for inactivated TPs given that the same limit has been proposed to be safe for residual, intact HCPs.

Applying Limits for Protein Fragments to Residual Inactivated TPs: Low–molecular-weight (LMW) species (e.g., truncated protein-backbone fragments) and high–molecular-weight (HMW) species (e.g., antibody dimers) are both examples of common TP-related impurities. Aggregation-formed HMW species within a DP can compromise both drug efficacy and safety. Additionally, LMW species often have low or substantially reduced activity relative to a TP’s monomeric form. Thus, both types of impurities are considered to be critical quality attributes (CQAs) that must be monitored routinely during drug development and as part of release testing for purified DPs (47). A level of ≤5–10% of soluble protein aggregates in a TP DP has been recommended (48); a level of <5% of HMW immunogenic aggregates is recommended (36). For a TP administered at 1 mg/kg (IV) to a subject weighing 50 kg, with the assumption that ≤5% of the dose consists of product-related impurities, the resulting 2.5-mg/dose mixture of protein impurities would be more immunogenic than residual inactivated TPs. The proposed PDE of 100 µg/dose is 25-fold lower than the recommended level for aggregate impurities present in the TP dose.

In an attempt to develop an acceptable limit for pharmacologically inactive fragments of human TPs, Sharnez et al. reported on their studies with gelatin in 2013 (49). They chose gelatin because it is a complex protein with fragments (15–400 kDa) and is of animal origin (which should be more immunogenic than degraded human TPs would be). Also, given that gelatin is derived by exposing collagen to pH and temperature extremes, its protein fragments are considered to be chemically comparable with the TP fragments in bioprocess residues after cleaning and sterilization (49). Gelatin also is used in blood infusions and a number of vaccines. Based on clinical experience, the safe and acceptable limit for inactive gelatin fragments was ascertained to be
650 µg/dose. Given the nonhuman nature of the protein, that provides greater confidence that a PDE of 100 μg/dose is a protective and acceptable exposure limit for inactivated residual TPs.

Applying Threshold of Toxicological Concern (TTC) Approaches to Residual Inactivated TPs: The TTC approach presented by Dolan et al. in 2005 proposed and supported exposure limits of 1, 10, and 100 µg/day respectively for compounds that are likely to be carcinogenic, those expected to be potent or highly toxic, and those that are neither (50). Originally established for pharmacologically active, small-molecule APIs, the approach also is commonly used in setting PDEs for other data-poor substances (51). Inactivated TPs are unlikely to be potent, highly toxic, or carcinogenic — giving them
an acceptable daily intake (ADI) of 100 μg/day. That limit is consistent with the PDE of 100 μg/dose proposed herein for inactivated TPs (assuming daily administration).

In 1998, Munro and Kroes proposed a similar approach based on the Cramer structural classification scheme and evaluation of NOAELs for >600 substances tested in repeat-dose toxicity studies (52–54). Briefly, Cramer class I substances have simple chemical structures, known metabolic pathways, and low potential toxicity. Normal biological constituents (aside from hormones) thus are included in that class (52, 53). Cramer class II substances have less-innocuous structures than those in class I but no positive indication of toxicity. Cramer class III substances contain structural features that suggest the potential for significant toxicity.

The TTC values established were 90, 540, and 1,800 μg/person/day for Cramer class III, II, and I substances based on a recipient’s body weight of 60 kg (54). Consistent with Cramer class I compounds (1,800 μg/person/day), AAs and inactivated TPs are not expected to pose a risk of significant toxicity. As a protective measure, if there is uncertainty regarding immunogenicity potential, denatured and degraded TPs also can be regarded as Cramer class II (540 μg/person/day) with the proposed PDE over fivefold lower. Even considering the most stringent class (Cramer class III, 90 µg/day), which is associated with a clearly positive indication of toxicity and data-poor substances, the PDE proposed herein is conservative and within an order of magnitude.

Applying ICH Guidance for Impurities in Small-Molecule Therapeutics to Residual Inactivated TPs: Small molecules generally are considered to be those with a molecular weight of <900 Da. In the case of degraded proteins, fragments can consist of single to multiple AAs, which range 57–186 Da in MW. Thus, an AA or a peptide fragment could be thought of as a small molecule.

Although not directly applicable to impurities in biologics, ICH Q3A recommends qualification of impurities present at a concentration threshold of 0.15% in a DS dosed at 2 g/day for nonmutagenic small-molecule impurities or to an impurity limit (per impurity) of 1 mg/day, whichever is lower (55). Note that these impurities can include pharmacologically active molecules. Notably, for a TP administered once daily, the level of 100 µg/dose is 10-fold below the 1-mg/day threshold. Additionally, if a DS is dosed at 2 g, then 0.15% would be 3 mg, which is 30-fold higher than the proposed PDE of 100 µg/dose of inactivated TP.

ICH Q3B recommends qualification of impurities present at a concentration threshold of 1.0% for nonmutagenic small-molecule impurities present in drug products dosed at <10 mg/day or to a threshold (per impurity) of 50 µg/day, whichever is lower (56). Such limits apply to each impurity, not to the total amount of all nonmutagenic small-molecule impurities present. Additionally, those impurities can be pharmacologically active. With that in mind — and the expectation that inactivated, denatured, and degraded TPs are mixtures of many proteinaceous compounds — the level of 100 µg/dose would be conservative. Note that the thresholds presented in ICH Q3A and Q3B are related more to product quality than patient safety.

Additional Considerations
Residual Circulating Inactive TPs: An important goal is protecting patients from potential acute effects driven by endogenous substances at peak systemic exposures (e.g., upon TP administration). Circulating endogenous IgG levels vary ~9.5–12.5 mg/mL (57). Considering a plasma volume of 3.5 L for a 70-kg adult, a residual amount of 100 µg degraded protein in circulation would be at 0.029-µg/mL concentration (29 ng/mL). A nanogram-level change is anticipated to be marginal compared with the total amount of degraded proteins present from physiological processes in our bodies. Therefore, an additional parenteral exposure of 100 µg inactivated TPs would be at 0.0002–0.0003% of circulating IgG levels, which is not anticipated to have a significant impact.

Dose, Frequency, and Duration: TP administration schedules vary, and low-level chronic exposures to protein degradants (as impurities in TPs) are unlikely to occur daily because most TPs are administered weekly or less frequently. From this standpoint (as elsewhere herein), the proposed PDE of 100 µg/dose to inactivated TP impurities is conservative.

Some individuals receive multiple drugs per day. Our evaluation focuses on additional risk to patients posed by a level of 100 µg of inactivated TP if present in an administered TP. It is important to note that the PDE for inactivated TPs is a conservative estimate. Exposure to such a level of inactivated TPs in more than one DP administered at roughly the same time should be acceptable for each TP administered with negligible safety concern. Similarly, impurity assessments for small-molecule drugs focus on the DS or DP at hand and not the possible agglomerate of impurities if multiple drugs are administered on the same day (40, 50, 55, 56).

Analytical Considerations: Applying a PDE in cleaning validation requires consideration of analytical feasibility. Inactivation studies usually are based on bioassays — e.g., enzyme-linked immunosorbent assays (ELISAs) — which measure the relative amount of biologically active product by investigating binding sites that are functionally intact (49). Measuring inactivated TPs does not require such specific analytical methods. Commonly used analytical methodologies detect all proteinaceous material and generally cannot tease apart whether a detected degraded protein comes from a TP or other sources (e.g., HCP). The standard method is a combination of total organic carbon (TOC) analysis for impurity quantification and a sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) assay for analyzing fractions of proteinaceous material (58, 59). TOC has a limit of quantification (LoQ) of ~0.2 ppm, which is sufficient for most cleaning validation applications.

Limitations and Applicability
In certain circumstances, application of the PDE proposed herein could benefit from a risk assessment. For example, that could be helpful for evaluating chimeric TPs that include foreign sequences and thus might present increased risk of immunogenicity. Another scenario that calls for additional risk assessment is when a TP is administered by a less common route such as intravitreal injection (60, 61). Highly stable TPs that can be administered orally also could benefit from in-depth assessment because the digestion processes (stomach acids and enzymatic activity) to which they are subject can be similar to harsh conditions of some cleaning processes.

Our proposed PDE of 100 µg/dose is not applicable to inactivated proteins from antibody–drug conjugates, protease inhibitors, enzymes, or plasma-derived TPs. For TPs dosed below the PDE threshold level, product-quality concerns come into question. The proposed PDE is based on safety, with product quality considerations aside.

After reviewing data and limits from available literature, we anticipate that a parenteral PDE for inactivated (denatured and/or degraded) TPs in the range of 100–3,000 µg/dose is generally acceptable within the above constraints. Available information supports that an exposure limit of 100 µg/dose is protective for inactivated TPs.

References
1 Annex 15: Qualification and Validation. EU Guidelines GMP for Medicinal Products for Human and Veterinary Use. European Commission: Brussels, Belgium, 30 March 2015; https://health.ec.europa.eu/system/files/2016-11/2015-10_annex15_0.pdf.

2 EMA/CHMP/CVMP/SWP/169430/2012. Guideline on Setting Health Based Exposure Limits for Use in Risk Identification in the Manufacture of Different Medicinal Products in Shared Facilities. European Medicines Agency: London, UK, 20 November 2014; https://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2014/11/WC500177735.pdf.

3 Report No. PI 046-1. Guideline on Setting Health Based Exposure Limits for Use in Risk Identification in the Manufacture of Different Medicinal Products in Shared Facilities. Pharmaceutical Inspection Co-Operation Scheme: Geneva, Switzerland, July 2018; https://picscheme.org/docview/2467.

4 Baseline® Pharmaceutical Engineering Guide: Risk-Based Manufacture of Pharmaceutical Products: A Guide to Managing Risks Associated with Cross-Contamination. Second Edition. International Society of Pharmaceutical Engineers: Tampa, FL, July 2017.

5 Protein Structure Particle Sciences. Lubrizol Life Sciences: Bethlehem, PA, 2019; https://lubrizolcdmo.com/technical-briefs/protein-structure.

6 Berendsen HJ, Hayward S. Collective Protein Dynamics in Relation to Function. Curr. Opin. Struct. Biol. 10(2) 2000: 165–169; https://doi.org/10.1016/s0959-440x(00)00061-0.

7 Grant BJ, Gorfe AA, McCammon JA. Large Conformational Changes in Proteins: Signaling and Other Functions. Curr. Opin. Struct. Biol. 20(2) 2010: 142–147; https://doi.org/10.1016/j.sbi.2009.12.004.

8 Graham JC, Yao H, Franklin E. Occupational Exposure Risks When Working with Protein Therapeutics and the Development of a Biologics Banding System. Applied Biosafety 26(4) 2021: 193–204; https://www.liebertpub.com/doi/10.1089/apb.2021.0004.

9 Krause ME, Sahin E. Chemical and Physical Instabilities in Manufacturing and Storage of Therapeutic Proteins. Curr. Opin. Biotechnol. 60, 2019: 159–167; https://doi.org/10.1016/j.copbio.2019.01.014.

10 Wang W, et al. Antibody Structure, Instability, and Formulation. J. Pharmaceut. Sci. 96(1) 2007: 1–26; https://doi.org/10.1002/jps.20727.

11 CBER/CDER. Guidance for Industry: Immunogenicity Assessment for Therapeutic Protein Products. US Food and Drug Administration: Rockville, MD, 2014; https://www.fda.gov/media/85017/download.

12 CBER/CDER. Guidance for Industry: Development of Therapeutic Protein Biosimilars: Comparative Analytical Assessment and Other Quality-Related Considerations. US Food and Drug Administration: Rockville, MD, 2019; https://www.fda.gov/media/159261/download.

13 Manning MC, Patel K, Borchardt RT. Stability of Protein Pharmaceuticals. Pharm. Res. 6(11) 1989: 903–918; https://doi.org/10.1023/a:1015929109894.

14 Taschner N, et al. Modulation of Antigenicity Related to Changes in Antibody Flexibility Upon Lyophilization. J. Mol. Biol. 310(1) 2001: 169–179; https://doi.org/10.1006/jmbi.2001.4736.

15 Ophardt C. Denaturation of Protein. Virtual Chembook. Elmhurst University: Elmhurst, IL, 2003.

16 Kaye G, Weber P, Wetzel W. The Alkaline Hydrolysis Process. ALN Mag. 108, 2004.

17 Alkaline Hydrolysis Deep Dive. BioSafe Engineering: New York, NY, 2017; https://biosafeeng.com/research.

18 Vlasak J, Ionescu R. Fragmentation of Monoclonal Antibodies. mAbs 3(3) 2011: 253–263; https://doi.org/10.4161/mabs.3.3.15608.

19 Sharnez R. Methodology for Assessing Product Inactivation During Cleaning, Part 1: Experimental Approach and Analytical Methods. J. Valid. Technol. 16, 2015; http://www.pda.org/docs/default-source/website-document-library/chapters/presentations/capital-area/cleaning-validation-2015/cleaning-validation-for-biopharmaceuticals.pdf?sfvrsn=4.

20 Mott A, et al. Methodology for Assessing Product Inactivation During Cleaning, Part 2: Setting Acceptance Limits of Biopharmaceutical Product Carryover for Equipment Cleaning. J. Valid. Technol. 19(4) 2013; http://www.pda.org/docs/default-source/website-document-library/chapters/presentations/capital-area/cleaning-validation-2015/cleaning-validation-for-biopharmaceuticals.pdf?sfvrsn=4.

21 Hruby VJ, Patel D. Structure–Function Studies of Peptide Hormones: An Overview. Peptides: Synthesis, Structures, and Applications. Academic Press: San Diego, CA, 1995: 247–286; https://doi.org/10.1016/B978-012310920-0%2F50007-3.

22 Matsubayashi Y. Post-Translational Modifications in Secreted Peptide Hormones in Plants. Plant Cell Physiol. 52(1) 2011: 5–13; https://doi.org/10.1093/pcp/pcq169.

23 Reichert JM. Antibodies To Watch in 2010. mAbs 2(1) 2010: 84–100; https://doi.org/10.4161/mabs.2.1.10677.

24 Guidelines on the Quality, Safety, and Efficacy of Biotherapeutic Protein Products Prepared By Recombinant DNA Technology. World Health Organization: Geneva, Switzerland, 2013; https://www.who.int/biologicals/biotherapeutics/rDNA_DB_final_19_Nov_2013.pdf. 2013.

25 ICH S6(R1). Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use: Geneva, Switzerland, 12 June 2011; https://database.ich.org/sites/default/files/S6_R1_Guideline_0.pdf.

26 Ball K, et al. Characterizing the Pharmacokinetics and Biodistribution of Therapeutic Proteins: An Industry White Paper. Drug Metab. Dispos. 50(6) 2022: 858–866; https://doi.org/10.1124/dmd.121.000463.

27 Schadt S, et al. Are Biotransformation Studies of Therapeutic Proteins Needed? Scientific Considerations and Technical Challenges. Drug Metab. Dispos. 47(12) 2019: 1443–1456; https://doi.org/10.1124/dmd.119.088997.

28 Unit 3: How Are Eukaryotic Cells Organized into Smaller Parts? Scitable: Essentials of Cell Biology. Nature Education: Cambridge, MA, 2014; https://www.nature.com/scitable/ebooks/essentials-of-cell-biology-14749010/122997196.

29 Vugmeyster Y, Harrold J, Xu X. Absorption, Distribution, Metabolism, and Excretion (ADME) Studies of Biotherapeutics for Autoimmune and Inflammatory Conditions. AAPS J. 14(4) 2012: 714–727; https://doi.org/10.1208%2Fs12248-012-9385-y.

30 Taft DR. Chapter 9: Drug Excretion. Pharmacology: Principles and Practice. Hacker M, Bachmann K, Messer W, Eds. Academic Press: New York, 2009; 175–199.

31 Ryman JT, Meibohm B. Pharmacokinetics of Monoclonal Antibodies. CPT Pharmacomet. Syst. Pharmacol. 6(9) 2017: 576–588; https://doi.org/10.1002/psp4.12224.

32 Di L. Strategic Approaches to Optimizing Peptide ADME Properties. AAPS J. 17(1) 2015: 134–143; https://doi.org/10.1208/s12248-014-9687-3.

33 Halsen G, Kramer I. Assessing the Risk to Health Care Staff from Long-Term Exposure to Anticancer Drugs: The Case of Monoclonal Antibodies. J. Oncol. Pharm. Pract. 17(1) 2011: 68–80; https://doi.org/10.1177/1078155210376847.

34 Gülsen A, Wedi B, Jappe U. Hypersensitivity Reactions to Biologics (Part 2): Classifications and Current Diagnostic and Treatment Approaches. Allergo J. Int. 29(5) 2020: 139–154; https://link.springer.com/article/10.1007/s40629-020-00127-5.

35 Pekar J, Ret D, Untersmayr E. Stability of Allergens. Mol. Immunol. 100, 2018: 14–20; https://doi.org/10.1016/j.molimm.2018.03.017.

36 Chon JH, Zarbis-Papastoitsis G. Advances in the Production and Downstream Processing of Antibodies. New Biotechnol. 28(5) 2011: 458–463; https://doi.org/10.1016/j.nbt.2011.03.015.

37 Høst A, Halken S. Hypoallergenic Formulas: When, to Whom and How Long: After More Than 15 Years We Know the Right Indication! Allergy 59(S78) 2004: 45–52; https://doi.org/10.1111/j.1398-9995.2004.00574.x.

38 Jolly RA, et al. Setting Impurity Limits for Endogenous Substances: Recommendations for a Harmonized Procedure and an Example Using Fatty Acids. Reg. Toxicol. Pharmacol. 134, 2022:105242; https://doi.org/10.1016/j.yrtph.2022.105242.

39 Pfister MP, et al. Proposal of a Default PDE Value for Data-Poor Therapeutic Monoclonal Antibodies. SwissTox: Annual Meeting of the Swiss Society of Toxicology, 2017, Basel, Switzerland.

40 ICH Q3C(R8). Impurities: Guideline for Residual Solvents. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use: Geneva, Switzerland, 20 October 2016; https://database.ich.org/sites/default/files/ICH_Q3C-R8_Guideline_Step4_2021_0422.pdf.

41 Jawa V, et al. Evaluating Immunogenicity Risk Due to Host Cell Protein Impurities in Antibody-Based Biotherapeutics. AAPS J. 18(6) 2016: 1439–1452; https://doi.org/10.1208/s12248-016-9948-4.

42 Wang X, Hunter AK, Mozier NM. Host Cell Proteins in Biologics Development: Identification, Quantitation and Risk Assessment. Biotechnol. Bioeng. 103(3) 2009: 446–458; https://doi.org/10.1002/bit.22304.

43 Champion K, et al. Defining Your Product Profile and Maintaining Control Over It, Part 2. BioProcess Int. 3(8) 2005: 52–57; https://bioprocessintl.com/2005/september-2005/defining-product-profile-maintaining-control-part-2.

44 Pilely K, et al. Monitoring Process-Related Impurities in Biologics: Host Cell Protein Analysis. Anal. Bioanal. Chem. 414(2) 2022: 747–758; https://doi.org/10.1007/s00216-021-03648-2.

45 Hogwood CE, Bracewell DG, Smales CM. Measurement and Control of Host Cell Proteins (HCPs) in CHO Cell Bioprocesses. Curr. Opin. Biotechnol. 30, 2014: 153–160; https://doi.org/10.1016/j.copbio.2014.06.017.

46 Sharnez R, Spencer A, Horner M. Biopharmaceutical Cleaning Validation: Leveraging Acceptable Exposure of Host-Cell Protein to Set Acceptance Limits for Inactivated Product. J. Valid. Technol. Summer 2012: 38–44.

47 Wang S, et al. Characterization of Product-Related Low Molecular Weight Impurities in Therapeutic Monoclonal Antibodies Using Hydrophilic Interaction Chromatography Coupled with Mass Spectrometry. J. Pharm. Biomed. Anal. 154, 2018: 468–475; https://doi.org/10.1016/j.jpba.2018.03.034.

48 Wang W, et al. Immunogenicity of Protein Aggregates: Concerns and Realities. Int. J. Pharmaceut. 431(1–2) 2012: 1–11; https://doi.org/10.1016/j.ijpharm.2012.04.040.

49 Sharnez R, et al. Biopharmaceutical Cleaning Validation: Acceptance Limits for Inactivated Product Based on Gelatin as a Reference Impurity. J. Valid. Technol. Winter 2013; ; https://www.researchgate.net/publication/313361609_Biopharmaceutical_Cleaning_Validation_Acceptance_Limits_for_Inactivated_Product_Based_on_Gelatin_as_a_Reference_Impurity.

50 Dolan DG, et al. Application of the Threshold of Toxicological Concern Concept to Pharmaceutical Manufacturing Operations. Regul. Toxicol. Pharmacol. 43(1) 2005: 1–9; https://doi.org/10.1016/j.yrtph.2005.06.010.

51 Faria EC, et al. Using Default Methodologies To Derive an Acceptable Daily Exposure (ADE). Regul. Toxicol. Pharmacol. 79(S1) 2016: S28–S38; https://doi.org/10.1016/j.yrtph.2016.05.026.

52 Cramer G, Ford R, Hall R. Estimation of Toxic Hazard: A Decision Tree Approach. Food Cosmet. Toxicol. 16(3) 1976: 255–276; https://doi.org/10.1016/s0015-6264(76)80522-6.

53 Munro IC, et al. Correlation of Structural Class with No-Observed-Effect Levels: A Proposal for Establishing a Threshold of Concern. Food Chem. Toxicol. 34(9) 1996: 829–867; https://doi.org/10.1016/s0278-6915(96)00049-x.

54 Munro I, Kroes R. Application of a Threshold of Toxicological Concern in the Safety Evaluation of Certain Flavouring Substances. Safety Evaluation of Certain Food Additives and Contaminants (Annex 5). WHO Food Additives Series: Geneva, Switzerland, 1998; https://apps.who.int/iris/bitstream/10665/43645/1/9789241660587_eng.pdf.

55 ICH Q3A(R2). Impurities in New Drug Substances. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use: Geneva, Switzerland, 25 October 2006; https://database.ich.org/sites/default/files/Q3A%28R2%29%20Guideline.pdf.

56 ICH Q3B(R2). Impurities in New Drug Products. International Council on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use: Geneva, Switzerland, 2 June 2006; https://database.ich.org/sites/default/files/Q3B%28R2%29%20Guideline.pdf.

57 Lobo ED, Hansen RJ, Balthasar JP. Antibody Pharmacokinetics and Pharmacodynamics. J. Pharm. Sci. 93(11) 2004: 2645–2668; https://doi.org/10.1002/jps.20178.

58 Wang XD, et al. Development of a Technique for Quantifying Protein Degradation. BioPharm Int. 29(11) 2017: 38–44, 53; https://www.biopharminternational.com/view/development-technique-quantifying-protein-degradation.

59 Tanyous JN. Cleaning Validation: Complete Guide for Health-Based Approach in Chemical Cross-Contamination Risk Assessment. PDA J. Pharm. Sci. Technol. 73(2) 2019: 204–210; https://doi.org/10.5731/pdajpst.2018.008946.

60 Lovsin Barle E, et al. Determination and Application of the Permitted Daily Exposure (PDE) for Topical Ocular Drugs in Multipurpose Manufacturing Facilities. Pharm. Dev. Technol. 23(3) 2018: 225–230; https://doi.org/10.1080/10837450.2017.1312442.

61 Pohl L, et al. Impurities in Drug Vials Intended for Intravitreal Medication. Case Rep. Ophthalmol. Medi. 2020: 8824585; https://doi.org/10.1155/2020/8824585.

Corresponding author Jessica Graham is director and head of product quality and occupational toxicology at Genentech, Inc. (a member of the Roche Group) in South San Francisco, CA; [email protected]. During the time of this work, Selene Araya was managing toxicologist at Lonza in Basel, Switzerland. Kamila Blum is a corporate toxicologist at GlaxoSmithKline, Munich, Germany. Janet Gould is principal toxicologist at Safebridge in New York, NY. Thomas Pfister is a senior occupational toxicologist at F. Hoffmann-La Roche in Basel, Switzerland.