Critical Quality Attributes Assessment and Testing Strategy for Biotherapeutics Development

• Sanofi SA

Introduction

Criticality assessment of product quality attributes forms the foundation of the Quality by Design (QbD) approach to drug development. This systematic approach to product development emphasizes product knowledge and process understanding along with process control based on sound science and quality risk management. Due to complexity of the molecular structure as well as the manufacturing process, development of biotherapeutics remains challenging. It is often difficult to fully evaluate the impact of the large number of quality attributes as related to safety and efficacy. The effects of manufacturing process parameters and material attributes on product quality variations are also difficult to fully characterize. For robust manufacturing operations, it is important that an integrated control strategy is developed and improved over time based on systematic process characterization along with implementation of appropriate risk assessment and mitigation throughout the product lifecycle. This article discusses the key concepts of critical quality attributes (CQAs) risk ranking based on prior knowledge and gathering of structure-function relationship information, consideration of process and stability impact on CQAs, and evolving the analytical testing panel as part of the integrated control strategy.

CQA Criticality Assessment

CQA is defined in ICH guidance Q8¹ as “a physical, chemical, biological or microbiological property or characteristics that should be within an appropriate limit, range, or distribution to ensure the desired product quality.” Typically, CQAs for a given product are defined early in development prospectively based on the quality target product profile of the biotherapeutic, and progressively refined as additional product knowledge is gained over time with extensive analytical characterization, animal studies and expanded clinical experience.

Listing of potential CQAs

Identification and selection of CQAs begins with the assembly of a comprehensive list of relevant product attributes that may have potential impact on product quality. The ultimate objective is to establish linkages between specific product attributes of the biotherapeutic to the expected clinical performance. For complex biologicals, this list may be long and daunting to manage. Grouping attributes into assessment categories (e.g., product-specific variants, processrelated impurities and obligatory CQAs; see Table 1) allows some level of simplification and guides the criticality assessment approach depending on the nature/type of the attribute categorization. Obligatory CQAs are not considered through the risk assessment tools as these attributes are typically specified by health authorities for release testing.²

Criticality assessment framework

The basic principles of applying a risk assessment to identify CQAs are based on the quality risk management guidelines as outlined in ICH Q9.³ Application examples using quantitative ranking tools have been published by industry working groups for monoclonal antibodies⁴ and vaccines.⁵ Additional examples from individual companies can be found in publications^2,6 and conference presentations.^7,8 While there are variations in the exact approach and specific scoring tools used by different groups, the common practice is to employ a scoring system based on two factors: impact and uncertainty. Typically, a project team consisting of subject matter experts from multiple disciplines conducts a systematic evaluation of each product quality attribute taking into account the potential impact on safety and efficacy, and also considers the relevance and uncertainty of the available information/prior knowledge. An example scoring approach is summarized in Table 2.

Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special offers
from American Pharmaceutical Review – all delivered right to your inbox! Sign up now!

The impact and uncertainty factors are scored independently against different scales consisting of up to five levels, with higher weighting assigned to the impact factor reflecting its higher importance. The two values are then multiplied to assign a risk score for each product quality attribute. The overall result is a listing of quality attributes along a criticality continuum. This assessment is performed at key points during process development, with studies designed to improve product knowledge and drive down the uncertainty. Overall, the details of the chosen scoring system are less important than ensuring consistency of the assessment approach throughout the product lifecycle. The scientific thought process in deriving the rationale to support the risk-scoring based on product knowledge, platform knowledge and published literature should be clearly documented and may be requested during regulatory submission process.

For impact assessments of product-specific variants, considerations of the severity of potential harm focus on efficacy and safety (Table 1). In the assessment of efficacy, the risk ranking considers the effects of the attribute not only on biological activity with respect to the intended mechanism of action (MOA), but also PK, PD and non-specific off–target effects. The data may come from structure-activity relationship (SAR) studies, forced degradation studies, nonclinical and clinical studies. In the SAR studies, product variant preparations can be made available by a variety of approaches: isolation from product comprising a mixture of variants generated from manipulating cell culture conditions or from different candidate clones/cell lines, application of specific stress conditions to the purified product, and enzymatic treatment to remodel glycosylation as well as generate fragments.² Additionally, analysis of collected serum samples from in vivo studies allows assessment of differential clearance of the different variants as well as potential effects of product modification in vivo.⁹ The experimental range of the CQAs studied in non-clinical models and analytical characterization studies may be broader than actual levels present in clinical trial materials. The key is to understand CQA criticality without taking into account of process capability, stability and manufacture control considerations. For assessment of immunogenicity and direct impact to safety (e.g. adverse events related to drug administration), it is often difficult to have meaningfully relevant data other than actual clinical experience. General concerns about a particular type of variant (e.g. aggregation) may be taken from published literature or from experience with a similar product type,^10,11 with considerations of other factors such as the route of administration, the dose and dosing regimen. In silico algorithms and in vitro methods are mostly used in early development for de risking pending clinical validation.¹²

When changes in CQA levels are interrelated, the risk assessment should seek to clarify the linkages. A common example is aggregation which is often categorized as a CQA due to its high risk of immunogenicity. Mechanisms of aggregate and particulate formation are complex and may be triggered by various structural perturbations such as conformational changes as well as oxidation, deamidation and disulfide scrambling.^13,14 The criticality assessment should differentiate between attributes that are directly associated with an adverse clinical impact (i.e. aggregation), and attributes that have a relationship with the critical attribute, which should be assessed independently of their potential to promote aggregation.6 Another example is charge heterogeneity. Charge isoforms may arise from different posttranslational modifications (PTMs) of the protein such as N/C terminal processing, deamidation, oxidation and disulfide heterogeneity.¹⁵

Depending on the location of the affected residues, potency may be affected.16 In some cases, acidic and basic protein species may exhibit different clearance rates from the main species.¹⁷

In practice, it is often not possible to generate and isolate product variants with a single PTM for direct assessment of biological impact.

Some attributes may only be evaluated through profile-based analysis (e.g. charge variants). Product variants never detected or observed only at low abundance during development may not pose significant risk to a program to warrant further tracking. Overall, the CQA risk ranking is used to drive execution of additional structure-function studies for further delineation of the impact on safety and efficacy of the drug and progressively decrease uncertainty. Preliminary acceptable ranges for CQAs can be defined from findings of these studies.

For impact assessment of process-related impurities, the focus is often on safety and toxicity concerns. However, in some cases impurities such as intracellular enzymes, metal ions and nonmetallic leachables may cause product degradation or protein modifications, which in turn might impact function.^18,19 For host cell proteins (HCPs), proteomics profiling by LC-MS techniques allow identification and impact assessment of individual HCPs.²⁰ Process additives undergo an assessment based on worst-case assumptions of their level in a product dose, while toxicology data and the additives’ history of use in the industry is also considered.⁴

For the uncertainty factor, the level of reliance on in vitro vs. in vivo data should be considered, as well as the availability of molecule specific data pertaining to potency and PK, the relevance of data leveraged from related molecules, and the range of clinical exposure.

By evaluating attribute criticality solely on the basis of impact and uncertainty to safety and efficacy, the product risk assessment only needs to be revised when new information is discovered regarding the properties of the attributes themselves, and not every time a process change is made.

Process Capability and Stability Information

With the CQA criticality ranking as a foundation, a second level of risk assessment can be performed to consider occurrence.⁴ This takes into account the likelihood (or frequency of failure) of maintaining a CQA within its acceptable limits under the range of process and storage conditions of interest. Manufacturing process and formulation development eff orts are thus focused on improving process performance via controlling the process parameters, material attributes and procedural controls that are linked to the CQAs.

Results from stability studies (real-time, accelerated and forced degradation) demonstrate the primary degradation mechanisms of the biotherapeutic and inform understanding of the impact to CQAs during routine storage and handling as well as excursions.^21,22 The process/stability impact ranking tool may consist of 3-5 levels, and the resultant occurrence scores are combined (via multiplication) with the criticality score for each CQA. The overall output is a listing of CQAs based on residual risk after process control considerations and accordingly used to define the analytical control elements.²³ CQAs with higher risk scores often require more stringent controls. However, a highly critical attribute that is well controlled through the process and formulated appropriately to ensure stability may not require routine testing.

Evolving the Testing Strategy

In the QbD approach to control strategy design, the overall analytical strategy should focus on implementation of broad and sophisticated testing for process and product characterization during the development phase while simplifying the GMP analytical monitoring program upon licensure.^24,25 End-product testing can be focused on a limited number of attributes (example in Table 3) after demonstration of adequate process control. For process impurities (ex. HCP and residual DNA) with well-understood mechanisms of removal, if robust and consistent process capability is demonstrated (e.g. through spiking studies), control through release testing and specification limits may not be necessary. Stability testing for attributes which are not impacted by known mechanisms of degradation and demonstrated to not change significantly over time are not necessary.^23-25

In addition, the frequency of testing and the application of specification ranges may be varied according to the combined scores from the criticality and occurrence ranking with scientific and risk-based justifications.^22,23 A more extensive set of characterization tests for the product may be maintained for occasional process investigations and when performing comparability studies in support of process change implementations.

Concluding Remarks

Criticality assessment of quality attributes forms the foundation of CMC development of biotherapeutics, informing process, formulation and analytical development as well as the design of an integrated control strategy for robust manufacturing. The risk ranking goes through iterative refinements as additional product knowledge is gained. However, it remains challenging to establish clear linkages of particular molecule-specific attributes to safety and efficacy when the mechanisms of action and the structure of the biotherapeutics are complex. Implementation of appropriate biochemical, biophysical and biological assays that are fi t-for-purpose in terms of sensitivity and specificity is essential in supporting structure-function studies.

With broad availability of high resolution analytical technologies as well as implementation of high throughput analytics, advancement in expanding product knowledge and process understanding continues.

A systematic and consistent application of science-based risk assessment during development and continued improvement of the manufacturing control strategy remains crucial in ensuring product quality throughout the product lifecycle.

References

1. ICH Q8 (R2) Pharmaceutical development. 2009
2. Atl N, Zhang TY, Motchnik P, et al. Determination of critical quality attributes for monoclonal antibodies using quality by design principles. Biologicals 2016; 44: 291-305.
3. ICH Q9 Quality risk management. 2006
4. CMC Biotech working group. A-mAb: a case study in bioprocess development. 2009
5. CMC Vaccine working group. A-vax quality by design case study. 2012
6. Vandekerckhove K, Seidl A, Gutka H, et al. Rational selection, criticality assessment, and tiering of quality attributes and test methods for analytical similarity evaluation of biosimilars. AAPS Journal 2018; 20: 68.
7. Stangler T. What to control? CQAs and CPPs. BWP workshop on setting specifications. London, UK. September 9, 2011. https://www.ema.europa.eu/documents/presentation/presentation-what-control-cqas-cpps-thomas-stangler-behalf-ega_en.pdf
8. Barnes CAS The role of the critical quality attribute assessment in biotherapeutic product development. CASSS Midwest discussion group, St. Louis, MO, October 20, 2016. https://cdn.ymaws.com/www.casss.org/resource/resmgr/mwdg/1016_mwdg_SrebalusBarnes.pdf
9. Li Y, Huang Y, Ferrant J, Lyubarskaya Y, et al. Assessing in vivo dynamics of multiple quality attributes from a therapeutic IgG4 monoclonal antibody circulating in cynomolgus monkey. mAbs 2016; 8 (5): 961-968.
10. Moussa EM, Panchal JP, Moorthy BS, et al. Immunogenicity of therapeutic protein aggregates. J Pharm Sci 2016; 105: 417-430.
11. Jefferis R. Posttranslational modifications and the immunogenicity of biotherapeutics. J Immunology Res 2016; article ID 5358272
12. Gokemeijer J, Jawa V, Mitra-Kaushik S. How close are we to profiling immunogenicity risk using in silico algorithms and in vitro methods: an industry perspective. AAPS J. 2017; 19 (6): 1587-1592
13. Roberts CJ. Therapeutic protein aggregation: mechanisms, design, and control. Trends Biotechnol. 2014: 32 (7): 372–380.
14. Yi Y, Zang L. Factors influencing biotherapeutic monoclonal antibody aggregation. Amer Pharm Rev 2018; March
15. Vlasak V, Ionescu R. Heterogeneity of monoclonal antibodies revealed by charge-sensitive methods. Curr. Pharmaceutical Biotech 2008; 9(6):468-81
16. Schmid I, Bonnington L, Gerl M, et al. Assessment of susceptible chemical modification sites of trastuzumab and endogenous human immunoglobulins at physiological conditions. Communications Biology 2018; 1, article number 28
17. Cheung S, Tian J, Tan Z, et al. Industrial bioprocessing perspectives on managing therapeutic protein charge variant profiles. Biotech. Bioeng. 2018; 115: 1646-1665
18. Wang W, Ignatius AA, Thakkar SV Impact of residual impurities and contaminants on protein stability. J Pharm Sci 2014; 103: 1315-1330.
19. Zhang K, Pellett JD, Narang AS, Wang YJ, Zhang YT Reactive impurities in large and small molecule pharmaceutical excipients – A review. Trends in Anal Chem 2018; 101: 34-42.
20. Wang F, Richardson D, Mueller HM, et al. Host-cell protein risk management and control during bioprocess development. Part 1 & 2. BioProcess International 2018; 16 (5) 18-25 & 16(6) 42-47.
21. Manning MC, Chou DK, Murphy BM, et al. Stability of protein pharmaceuticals: an update. Pharm Res 2010; 27:544–575.
22. Chan CP. Forced degradation studies: current trends and future perspectives for protein based therapeutics. 2016; Expert Rev. Proteomic 13 (7): 651-658.
23. Kepert JF, Cromwell M, Engler N, et al. Establishing a control system using QbD principles. Biologicals 2016: 44: 319-331.
24. Kendrick B. New approaches to specifications based on novel QC paradigms. International Alliance for Biological Standardization conference, Lister Hill, MD, September 25-26, 2013.
25. Cherney B. Application of right size testing, regulatory responses and future directions. Well-Characterized Biological Products conference. Washington, DC, January 2016

Author Biographies

Dr. Christine P. Chan is currently principal scientist in Global Manufacturing Science & Technology, Industrial Affairs at Sanofi. She is a protein biochemist with broad experience in the biopharmaceutical industry, including prior experience at Sandoz Pharmaceuticals and Merck & Co., Inc. She specializes in the analysis of recombinant products produced from mammalian cells for vaccines and biologics development. Her application experience includes expression cell line selection, drug substance and drug product process development, manufacturing tech transfers and lifecycle management of commercialized products.