OR WAIT 15 SECS
The manufacturer should propose stability-indicating methodologies that provide assurance that changes in the identity, purity, and potency of the product will be detected.
The current regulatory guidances governing forced degradation studies of biological pharmaceuticals are extremely general. They itemize broad principles and approaches with few practical instructions. There is no single document that comprehensively addresses issues related to stress studies such as objectives, timing, selection of stress conditions, and extent of degradation. We will attempt to fill that gap by summarizing regulatory guidance for stress studies of biological products and present some examples of their practical applications.
The complexity of biological macromolecules when compared to small molecule therapeutics, differences in manufacturing, and the broad variety of potential degradation pathways lead to special requirements in quality assurance and analytical testing of pharmaceutical proteins. The product-related impurities are molecular variants formed during manufacture, storage, or use, and their properties are different from the desired product with respect to activity, efficacy, and safety.1
Forced degradation studies are designed to generate product-related variants and develop analytical methods to determine the degradation products formed during accelerated pharmaceutical studies and long-term stability studies. Any significant degradation product should be evaluated for potential hazard and the need for characterization and quantitation.2
Regulatory guidance is open to interpretation. Detailed information and specific instructions for conducting forced degradation studies of biologics and biotechnology products are close to non-existent. The most recent workshop was held in 2001 by the Pharmaceutical Research and Manufacturers of America Analytical Research and Development Steering Committee. The conclusion was that there is little information about strategies and principles. The summary provides useful definitions and general comments about degradation studies, while guidance concerning the scope, timing, and best practices is very general.
This article shows a way to reduce these generalities to practice.
Michael Kats, Ph.D.
A one-time forced-degradation study on a single batch is not considered part of the normal stability protocol.4 However, the design of the stress studies and the results are to be provided to regulatory authorities as part of the stability section of the application.5,6
The best samples of product-related degradants for the specificity evaluation would be retrieved throughout the pharmaceutical stability study.
The duration of most stability studies makes this ideal situation untenable. Thus, an analyst is faced with the necessity to artificially generate degraded samples.
Forced degradation or stress testing studies are part of the development strategy5 and are also an integral component of validating analytical methods that indicate stability and detect impurities. This relates to the specificity section of the validation studies.5,6 It is important to recognize that forced degradation studies are not designed to establish qualitative or quantitative limits for change in drug substance (DS) or drug product (DP).8
Testing of stressed samples is required to demonstrate the following abilities of analytical techniques employed in stability studies:1,5,9
The forced degradation studies are also expected
Stress studies may be useful in determining whether accidental exposures to conditions other than normal ranges (e.g., during transportation) are deleterious to the product, and also for evaluating which specific test parameters may be the best indicators of product stability.2,9
Start the search for a stability-indicating analytical procedure that will detect significant changes in the quality of DS at the Phase II stage of IND.
Stress studies on DS and DP should be completed during Phase III and significant impurities should be identified, qualified, and quantified.
Starting forced degradation experiments before Phase II is highly encouraged and should be conducted on DS with multiple aims: to provide timely recommendations for improvements in the manufacturing process; to ensure proper selection of stability-indicating analytical techniques; and to assure sufficient time for degradation product identification, degradation pathways elucidation, and optimization of stress conditions.
Every change in stability-indicating analytical methods, manufacturing processes, or formulation requires re-validation of analytical methods; therefore full validation commences only after the manufacturing process is finalized, formulations established, and test procedures are developed and qualified. However, method validation must be completed before a formal long-term stability study begins. These limitations impose time constraints on all method-validation activities including stressed sample development and testing. Consequently, all preliminary work on optimization of stress conditions must be completed at earlier stages, even though results of forced degradation studies are not required to be reported until the Phase III stage of IND application.
The question of how much degradation is sufficient to meet the objectives of stress studies is widely discussed, especially with respect to conventional therapeutics. A degradation level of 10 to 15% is considered adequate for validation of a chromatographic purity assay.
Carr and Wahlich suggest that chromatographic methods for product-related impurities (including degradants) should be validated by spiking experiments within the range of 0 to 20% if the expected range of impurities is 0 to 10%.
Other authors recommend less than 10% degradation of active ingredient in the stressed samples.
Reynolds and others suggested that DS spiked with a mixture of known degradation products can be used to challenge methods employed for monitoring stability of DP.
The apparent consensus among pharmaceutical scientists is that samples degraded ~10% are optimal for use in analytical method validation. These considerations apply to small organic pharmaceuticals for which stability is dictated by the typical pharmaceutical limit of 90% of label claim.
No such limits for physico-chemical changes, losses of activity, or degradation during shelf life have been established for individual types or groups of biological products.9 In general, international and national regulations for biological products provide little guidance with respect to stability-related issues. These issues should be considered on a case-by-case basis.
As a group, biological products form a wide variety of product-related degradants under stress conditions. In cases with multiple degradation pathways, it appears to be beneficial to develop multiple product-related variants to challenge the specificity of analytical methods, even when some of degradants may be present at concentrations exceeding 10%. Do this when accelerated stability studies do not provide clear indication of the degradation pathways. When a stress factor generates only one degradation product, for example higher molecular weight non-covalent aggregates, 10 to 15% level of aggregation may be sufficient to challenge the specificity of such methods as size exclusion chromatography (SEC) or light scattering.
The forced degradation experiments do not necessarily result in product decomposition. The study can be stopped if no degradation is observed after DS or DP has been exposed to a stress that exceeds conditions of accelerated stability protocol.8 Protocols for generation of product-related degradation may differ for DS and DP due to differences in matrices and concentrations. For example, sugar additives often present in DP are known to stabilize proteins vis-a-vis denaturing conditions.12
Forced degradation is normally carried out under more severe conditions than those used for accelerated studies.
The choice of stress conditions should be consistent with the product's decomposition under normal manufacturing, storage, and use conditions which are specific in each case.
The ICH guidance recognizes it is impossible to provide strict degradation guidelines and allows certain freedom in selecting stress conditions for biologics.
The choice of forced degradation conditions should be based on data from accelerated pharmaceutical studies and sound scientific understanding of the product's decomposition mechanism under typical use conditions.
A minimal list of stress factors suggested for forced degradation studies must include acid and base hydrolysis, thermal degradation, photolysis, oxidation,
and may include freeze-thaw cycles and shear.
Regulatory guidance does not specify pH, temperature ranges, specific oxidizing agents, or conditions to use, the number of freeze-thaw cycles, or specific wavelengths and light intensities. The design of photolysis studies is left to the applicant's discretion although Q1B recommends that the light source should produce combined visible and ultraviolet (UV, 320-400 nm) outputs, and that exposure levels should be justified.8 Consult the appropriate regulatory authorities on a case-by-case basis to determine guidance for light-induced stress.9
Degradation products arising in significant amounts during manufacture and storage should be identified, tested for, and monitored against appropriately established acceptance criteria.
Examination of some degradation products generated under stress conditions may not be necessary for certain degradants if it has been demonstrated they are not formed under accelerated or long-term storage conditions.
Design your forced degradation studies to be part of impurity characterization. When identification of the impurity is not feasible, incorporate the description of unsuccessful experiments (including those conducted in stress testing studies) in the text of the application.15 The most frequently encountered protein variants include truncated fragments, deamidated, oxidized, isomerized, aggregated forms, and mismatched disulfide links.1
Degradation pathways for proteins can be separated into two distinct classes involving chemical instability and physical instability. Chemical instability is any process that yields a new chemical entity including modification of the protein (via individual amino acid alteration), covalent bond formation, or cleavage. Physical instability refers to changes in the higher order structures (secondary and above). Non-covalent aggregation usually results from partial or full unfolding, which enhances hydrophobic interactions between protein molecules. It may also lead to denaturation, adsorption to surfaces, and precipitation.12 Aggregation presents a significant patient risk because protein aggregates are frequently im-munogenic, therefore analytical methods employed in stability testing should detect low concentrations of aggregates.16 Stress the placebo in parallel with DP as a control for excipients' decomposition and to monitor the decomposition's effect on degradation pathways of active ingredients. Potential degradation pathways are extensively researched, and methods for their detection are well established. A number of comprehensive reviews on this topic are available in the literature.12,17,18,19
The manufacturer should propose stability-indicating methodologies that provide assurance that changes in the identity, purity, and potency of the product will be detected.
The selection of tests is product specific. Stability-indicating methods will characterize potency, purity, and biological activity.
As examples, stability indicating methods may include electrophoresis (SDS-PAGE, immunoelectrophoresis, Western blot, isoelec-trofocusing), high-resolution chromatography (e.g., reversed phase chromatography, SEC, gel filtration, ion exchange, and affinity chromatography), and peptide mapping.
The selected set of methods must be able to detect, separate, and quantitate all observed degradation products, although it is recognized that identification and characterization of the appropriate variants may require use of additional analytical methodologies. New analytical technologies and modifications of existing technologies are continuously being developed and should be utilized when appropriate.1 The list of assays challenged by stressed samples should include analytical methods employed in the stability program and those monitoring impurities.
It is one thing to be theoretical and quite another to implement the action plan. This set of stress studies experiments shows the application in practice. The model protein is a covalent homo-dimer (AA) containing 356 amino acids in each chain with a total molecular weight ~92,000 Da of which ~14,000 Da are attributable to carbohydrate attachments.
We optimized five stress conditions including elevated temperature, UV-VIS (320 to 400 nm) light exposure, H2O2-induced oxidation, and high and low pH to generate partial degradation (10 to 30%) of model protein via various degradation pathways. DS buffer solution contained 50 mM sodium chloride and 25 mM phosphate buffer, pH 7.5. In addition, reconstituted DP contained 100 mg/mL of maltose. The purposely degraded samples were used to challenge the ability of stability-indicating methods to separate, detect, quantify, and identify de-gradants. Table 1 is a summary of the results of analytical tests.
Table 1. Summary of Analyses of the Stressed Model Protein Drug Substance
Results from SDS-PAGE are shown in Figure 1. The non-reduced model protein produced a main broad band at ~100 kDa, which represents the intact AA monomer. The reduced subunit (A) produced a main broad band at ~50 kDa. The electrophoretic band migration of the light-stressed protein exhibited additional bands of higher molecular weight species on both reduced and non-reduced SDS gels, indicating formation of higher-molecular-weight covalent aggregates (Figure 1, lanes 5 and 10). Other types of forced degradation did not have a noticeable effect on the protein migration in the gels.
Figure 1. SDS-PAGE Analysis of Reduced (R) and Non-reduced (NR) Model protein DP Samples Stressed with UV light and Low pH Markers: Lane 1, Blank: Lanes 2, 7 and 12; Reference material: Lanes 3(R), 6(R), 8(NR) and 11(NR), Low pH stressed: Lanes 4(R) and 9(NR), UV stressed: Lanes 5(R) and 10(NR).
The SEC method separated and quantitated AA monomer, multimeric forms, and lower-molecular-weight species. SEC exhibited almost exclusively an AA form of the intact molecule, usually 99.2 to 99.5% (Figure 2, curve 1). Analyses of the stressed samples demonstrated the presence of elevated amounts of multimeric species for every type of the applied stress condition. The extent of aggregation ranged from 2.6% for the H2O2-oxidized sample to 28% for low-pH-stressed protein. Some differences in the relative abundance of higher molecular mass species may indicate different pathways of aggregate formation.
Figure 2. Size Exclusion Chromatography of the Stressed Model Protein
Results of the peptide map analysis of H2O2-treated model protein DP are presented in Figure 3 and Table 2. Methionine (Met) residue is the most reactive in oxidation as compared to other potential targets such as cysteine, histidine, tryptophan, and tyrosine. Met oxidation can spontaneously occur in the presence of atmospheric oxygen during the purification and storage of proteins.
Figure 3. Tryptic Peptide Maps of the Model Protein DP Samples (a) Reference Material (b) H2O2âstressed (c) High-pHâstressed
The intact T5 fragment represents a glycopeptide with multiple glycoforms. On the peptide map it eluted as a broad, irregular-shaped peak in the retention time interval 67 to 70 min (Figure 3a). When treated with H2O2, this peptide, which contains two Met residues, formed a mixture of degradants with one or two Met-sulfoxides. This mixture of oxidized glycopeptides eluted over broad and overlapping retention intervals, which made chromatographic separation and quantitation of these species not possible.
The H2O2 treatment induced Met-sulfoxide formation in every Met position in the model protein (Figure 3b), with the extent of degradation dependent on the Met location in the sequence. The highest level of degradation was observed for the T6 fragment with Met-sulfoxide at 85%, while the N-terminal T1 fragment was oxidized less than 10% (Table 2).
Table 2. Quantitation of Intact and Sulfoxidized Peptides of the Model Protein Tryptic Map
Elevated temperatures and a shift to basic pH create favorable conditions for side chain deamidation of Asn and Gln amino acids in proteins. The Asn-Gly sequence is very sensitive to this reaction, which results in substitution of Asn for Asp or iso-Asp residues. The peptide map profile of a high-pH-stressed sample shows formation of two variants of the T26 fragment (Figure 3c). MS/MS analysis confirmed that these peaks represent T26 peptide deamidated at one or two Asn positions. Another deamidation site was found in the T15 peptide. In addition to these peptides, a deamidated variant of the T12 fragment was found to be eluting after T12. A detected elevated level of Met-oxidation in the T6 fragment was apparently the result of a parallel oxidation reaction during the prolonged incubation.
With the exception of oxidation, other types of stress had little or no effect on specific binding as determined by BIACore and competitive ELISA (Table 1). At the same time, increase of in vitro bioactivity was observed for samples after high- and low-pH stress and UV light exposure. The results indicating sensitivity to light and elevated temperature exposure were used to determine handling, storage, and transportation conditions for the product.
Stress testing studies are conducted to challenge the specificity of stability-indicating and impurity-monitoring methods as part of the validation protocol. Another major goal is to investigate degradation products and pathways. The results of the forced degradation studies are required to be included in a Phase III IND filing. We recommend that you start the study as early as possible to be able to provide valuable information that can be used to improve formulations and the manufacturing process.
The choice of stress conditions should be consistent with product decomposition under normal manufacturing, storage, and use conditions. Recommended stress factors include high and low pH, elevated temperature, photolysis, and oxidation. The extent of the stress applied in forced degradation studies should ensure formation of the desired amount (usually 10 to 20%) of degradation.
In our experiments, the purposefully degraded protein was used to challenge the ability of stability-indicating methods to separate, detect, quantify, and identify degradation products. The results demonstrated that a number of analytical techniques were able to detect and characterize various alterations in primary structure, changes in physical stability, binding to a specific target, and biological activity of the model protein. Low levels of deamidated and oxidized degradants were detected and identified as well as products of covalent and non-covalent aggregation. ?
Analytical testing of the stressed samples was conducted by the following BMS scientists: R. Abraham, Ph.D., L. Ingraham, S. Musial, T. Ropchak, Z. Wei, Ph.D., Z. Zhang, Ph.D. The author would like to thank Drs. P. Thammana and K. Venkat for useful discussions.
1. ICH. Guidance for Industry, Q6B. "Specifications: Test Procedures and Acceptance Criteria for Bio-technological/Biological Products." ICH-Q6B. 1999 August 18.
2. ICH. Guidance for Industry, Q5C. "Quality of Biotechnological Products: Stability Testing of Biotechnological/Biological Products." ICH-Q5C. 1996 July 10.
3. Reynolds DW, et al. Available guidance and best practices for conducting forced degradation studies. Pharmaceutical Technology 2002 Feb. 1.
4. FDA. Guidance for Industry. "INDs for Phase 2 and 3 Studies of Drugs, Including Specified Therapeutic Biotechnology-Derived Products." Draft Guidance, 1999 February.
5. ICH. Guidance for Industry. "Q1A Stability Testing of New Drug Substances and Products" ICH-Q1A. 2001 August.
6. FDA. Guidance for Industry. "Analytical Procedures and Methods Validation: Chemistry, Manufacturing, and Controls Documentation", Draft Guidance, 2000 August.
7. Jenke DR. Chromatographic method validation: A review of current practices and procedures. II. Guidelines for primary validation parameters. J. Liq. Chromatogr. 1996; 19:737-757.
8. ICH. Guidance for Industry, Q1B "Photostability Testing of New Drug Substances and Product", ICH-Q1B. 1996 November.
9. ICH. "Final Guidance on Stability Testing of Biotechnological/Biological Products; Availability." Federal Register 61FR p. 36466-9. 1996.
10. Szepesi G. et al. Selection of high-performance liquid chromatographic methods in pharmaceutical analysis. III Method validation. J. Chromatogr. 1989; 464:265-278.
11. Carr GP, Wahlich JC. A practical approach to method validation in pharmaceutical analysis. J. Pharm. Biomed. Anal. 1990; 86 8):613-618.
12. Manning MC et al. Stability of protein pharmaceuticals. Pharmaceutical Res. 1989; 6: 903-918.
13. CDER. Reviewer Guidance. "Validation of Chromatographic Methods." 1994 November.
14. ICH. Guidance for Industry, Q2B "Validation of Analytical Procedures: Methodology." ICH-Q2B. 1996 November.
15. ICH. Guidance for Industry, Q3A "Impurities in New Drug Substances." ICH-Q3A. 2003 February 11.
16. ICH. Guidance for Industry, Q3B(R). "Impurities in New Drug Products." ICH-Q3B. 2003 November 14.
17. Reubsaet JLE, et al. Analytical techniques used to study the degradation of proteins and peptides: Chemical instability. J. Pharm. and Biomed. Anal. 1988; 17:955-978.
18. Volkin DB, et al. "Degradative covalent reactions important to protein stability. Mol. Biotechnol. 1997; 8:105-122.
19. Reubsaet JLE, et al. Analytical techniques used to study the degradation of proteins and peptides: physical instability. J. Pharm. and Biomed. Anal. 1998; 17:979-984.
20. Bichsel E, et al. Requirements for the quality control of chemically synthesized peptides and biotechnologically produced proteins. Pharmaceutica Acta Helvetiae 1996; 71:439-446.
Michael Kats, Ph.D., is senior scientist, Biologics Quality Control, Bristol-Myers Squibb Co., P. O. Box 4755, Syracuse, NY. 13221, 315.432.2563, fax 315.432.2948, firstname.lastname@example.org.