The editors and publisher wish to express their gratitude to the colleagues listed below for their valuable contribution to the peer review process for the Generics and Biosimilars Initiative Journal (GaBI Journal)in 2021.
Dr Chiann Chang, Brazil
Associate Professor Dr Devrim Demir Dora, Turkey
Dr Thijs J Giezen, The Netherlands
Dr Weng Fai Lai, Singapore
Dr Gerard Lee, UK
Dr Frits Lekkerkerker, The Netherlands
Dr Robin Thorpe, UK
Dr Philip Travis, USA
Mr Robert Tribe, Australia
Dr Jaana Vesterinen, Finland
Emeritus Professor Arnold Vulto, The Netherlands
The first GaBI Journal issue of 2022 comes at a time of both hope and despair. While the COVID-19 pandemic continues to cause significant global morbidity and mortality; especially in resource-poor countries, both the pace and direction of changes are beginning to produce some encouraging signs. Positive trends are in part the result of the continuing increase in our understanding of the virus itself, its treatment, and public health actions that are effective in limiting its spread. The availability of highly effective vaccines, antibodies, and antiviral medications have greatly limited infections and improved outcomes. Biological products, both innovator and biosimilar versions, as well as generic drug products are decreasing treatment costs and increasing the availability of effective treatments. Of special note is the increasing ability of resource-poor nations to produce their own versions of innovator products without patent constraints. Perhaps the most impressive example of this is, ‘The World’s COVID-19 Vaccine’ developed by the Texas Children’s Hospital at Baylor University and funding support of the Bill and Melinda Gates Foundation. This partnership has demonstrated that it is possible to overcome availability issues inherent in the for-profit pharmaceutical industry.
In contrast to the slowly improving outlook for the COVID-19 pandemic, the world is witnessing an increasing number of brutal hostilities both within, e.g. Syria, Libya, Mali, Ethiopia, and Yemen, and between nations that threaten human health. Especially dangerous examples include the invasion of sovereign nations by neighbouring countries, e.g. Georgia and Ukraine; that has produced a level of destruction and human suffering not seen in decades. These hostilities are causing untold suffering of innocent civilians, as well as of combatants. They have also blocked access to not only medicines, but also to medical treatment, hospitals, and health professionals. Global hostilities are depriving millions of people, including children, the disabled and the elderly of basic human needs. including shelter, food, and water, producing both medical and psychological damage that will consume health care and other resources for decades to come. Hopefully, future access to effective, reasonably priced generic and biosimilar drug products will help countries deal with these issues. This issue of the GaBI Journal contains articles that suggest some ways in which access to such products could be improved.
In a Letter to the Editor, Adjunct Professor Pekka Kurki reviews the history of switch studies done to approve EU marketing of biosimilars. The author concludes that such products should be made interchangeable with their reference products without the need for systematic clinical switch studies. The clinical evidence reviewed by Adjunct Professor Kurki includes 178 biosimilar clinical switch studies that failed to find any evidence of switch-related adverse effects, as well as recent reviews that also failed to find evidence of switch-related adverse effects. The author also cites the very limited power of such switch studies to identify or rule out rare adverse events and suggests that switch studies be replaced by pharmacovigilance and pharmaco-epidemiological studies instead. Readers, including those involved in the development of innovator products and those who simply disagree based on their own experience or their own review of the literature, are encouraged to submit letters presenting any alternative, supportive, or clarifying opinions.
The first Review Article was submitted by Drs Mihaela Buda, Olga Kolaj-Robin, and Emmanuelle Charton from the European Pharmacopoeia Department at the European Directorate for the Quality of Medicines & HealthCare (EDQM). Drs Buda et al., provides an overview of strategies to overcome the challenges of elaborating monographs on biotherapeutics. The opinions expressed in this article were based on discussions with various European stakeholders at scientific meetings, as well as on the experience of the authors in developing standards used to produce biosimilar pharmacopoeial monographs. The article should be very useful both for those who use or utilize the European or other pharmacopoeias. as well as for anyone unfamiliar with these useful publications.
The second Review Article by Adjunct Associate Professor Sia Chong Hock et al., from the National University of Singapore covers the timely topic of the challenges associated with the best-practices manufacture, storage, distribution, and regulation of both new and traditional vaccines. The COVID-19 pandemic has produced many examples of the problems remaining to be solved in, ‘the equitable distribution and availability of safe, efficacious and good quality vaccines’ in a way that increases the efficient use of vaccines to prevent or mitigate the effects of infectious diseases and safeguard public health. The authors discuss how a lack of adequate controls of the manufacture, storage, distribution, and possibly regulation of vaccines contribute to, ‘the continued existence of poor-quality vaccines’ as well as costly waste and decreased efficacy of vaccines. The review should be useful for anyone involved in the testing, evaluation, procurement, or administration of vaccines and vaccination programmes.
The final Review Article by Dr Robert Janknegt and Ms Marloes Dankers, describe the System of Objectified Judgement Analysis (SOJA) and then explains how it was used at Dr Janknegt’s hospital in the Netherlands to provide clinicians with an objective, evidence-based method to evaluate which of the increasing number of competing, long-acting insulin products to prescribe for their diabetic patients. Criteria utilized in the SOJA included: effectiveness, safety, tolerability, and ease of use. Dr Janknegt explains that because application of these criteria to long-acting insulin products produced essentially identical scores, it was decided that acquisition cost and individual patient characteristics be used to determine the choice of formulary products. The manuscript should be helpful for other hospitals when deciding on which long-acting insulins to include on their formularies. It is likely that the SOJA may also be useful to evaluate other medications. Comments are welcomed from readers who disagree with the use of the general approach described, or who want to discuss application of the SOJA to a different multisource product, as well as those familiar with different decision support systems.
The Opinion Article by Dr Pablo Matar raises an important issue concerning the possibility of, ‘loss of bioavailability over time’. Much has been written/claimed about how even minor changes in the manufacturing process can affect the qualities of a biological; whether an innovator or follow-on/biosimilar product. Less has been written about the fact that changes in manufacturing processes over time can also have important effects on originator products. Dr Matar points out that, ‘Current regulations do not require a given biosimilar to remain similar to its reference biological over time’ yet no post-marketing comparative bioavailability studies are required by regulators. Since any change in the manufacturing process can alter the final biological product, either of any two products deemed to be ‘biosimilar’ at one point in time are subject to divergent changes (‘drift, evolution and divergence’) that could result in later, undetected non-equivalence. To deal with this problem the author suggests: 1) pharmacovigilance systems should be strengthened; 2) interchangeability standards are needed to deal with this possibility; and 3) that harmonized regulatory definitions of comparability versus interchangeability need to be established. While the issues are clear, actual evidence-based recommendations for specific, proven effective changes needed in the regulatory approval or monitoring process were not well described. It is also not clear, to this reader at least, which (or even whether) pharmacovigilance programmes can reliably detect all clinically meaningful changes in biological product effects. The effects of manufacturing process changes over time need to be monitored for all innovator products, whether there are any approved biosimilar products available or not. Unfortunately, it is not clear which, if any, in vivo or in vitro methods, including those required by the US Food and Drug Administration to declare interchangeability, are adequate to detect all clinically important differences in product performance. Reader comments on this complicated, important issue are encouraged.
I will close with my wishes for a more peaceful world with more equitable distribution of all resources needed for a long, rewarding, and healthy life for all peoples of the world.
Professor Philip D Walson, MD Editor-in-Chief, GaBI Journal
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
Abstract: Detailed and accurate structural characterization of biopharmaceuticals is of paramount importance for product definition, assessment of impurities and achieving in-depth understanding of manufacturing processes. But how can we have confidence in the data obtained from characterization studies? The expectation from regulatory authorities when assessing methods for product characterization is that there will be some demonstration of their suitability to the task in hand. Here we look at those expectations and what steps can be taken to give confidence to the data produced by these analytical methods.
Submitted: 15 February 2022; Revised: 16 March 2022; Accepted: 22 March 2022; Published online first: 28 March 2022
Structural analysis – the starting point
Structural analysis of a biopharmaceutical product, be it a biosimilar or new drug, inevitably requires the use of a wide range of techniques and technologies. The ICH Q6B [1] guidelines for structural investigation, which are invoked by the regulatory authorities as the document detailing their expectations for structural characterization of biosimilars [2-4], state that all areas of biomolecular structure must be investigated. This requirement covers primary amino acid sequence and post-translational modifications including glycosylation (if present), through to secondary and tertiary structure and aggregation assessments, as well as an assessment and characterization of product related impurities (forms of the product that are not structurally consistent with the product specification). Structural analysis is therefore the starting point in the development of a drug product. It goes hand in hand with developing the product itself, as well as the manufacturing process, to produce the expected end product suitable for use in pharmacokinetics (PK), pharmacodynamics (PD) and clinical trials as necessary.
Structural investigations are performed at the drug development stage to examine and indeed confirm that the product has been made correctly, and also to assess the nature and levels of product related impurities prior to any clinical-based work. It is therefore understandable that analytical methods are liable to evolve as structural data are generated and interpreted, and any further specifically targeted analytical studies are carried out. To this end, there needs to be a degree of flexibility in the details of the analytical methodologies involved, since the precise conditions required for optimal data generation within a specific technique may not yet be defined.
At this point in the development of the product there is therefore no requirement for the application of specifically qualified or validated methods for analysis since the methods, product structure and therefore any product specifications or parameters are not yet defined. This does not mean to say of course that the methods used should not be demonstrated as being fit for purpose and indeed, prior to analysis, investigations need to be performed to optimize a method for application to that particular product. This may include the demonstrated requirement for excipient removal if that is likely to interfere with analysis, e.g. high levels of sugar or amino acid excipients which would interfere with monosaccharide and amino acid analysis, respectively, or assessment of incubation times for a proteolytic digest associated with a peptide mapping study. Purification may be carried out so that both biosimilar and innovator drugs are subsequently analysed from within the same buffer system, which precludes the influence of buffer effects on results. If any up-front purification is required for excipient removal, then an assessment needs to be made as to any possible impact on the active ingredient being assessed [2].
‘Fit for purpose/intended use’
When we consider biosimilars, the requirement for the application of such a broad range of characterization techniques, with their inherent need for a degree of flexibility to suit the molecule under investigation, is recognized by the regulatory agencies as falling outside the area of ‘Quality Control’ since these procedures are used as part of product development and ultimately product definition, see Table 1 for comparison of qualification and validation features.
To this end the methods are not therefore required to be validated in accordance with ICH Q2(R1) [5] but; as stated by the US Food and Drug Administration (FDA) in their Biosimilar guidance document (Quality), methods ‘…should be scientifically sound, fit for their intended use, and provide results that are reproducible and reliable’[3]. The European Medicines Agency (EMA), in their Biosimilar guidance document (Quality) state ‘Methods used in the characterization studies form an integral part of the quality package and should be appropriately qualified for the purpose of comparability. If applicable, standards and reference materials, e.g. from European Pharmacopoeia (Ph. Eur.), World Health Organization (WHO), should be used for method qualification and standardization’ [2]. The UK Medicines and Healthcare products Regulatory Agency (MHRA) Biosimilars guidance document echoes this sentiment, stating ‘Analytical methods need to be sensitive, qualified and sufficiently discriminatory to detect possible differences. Robust data require the application of suitable orthogonal methods’, [4].
The idea of the suitability of methods for their analytical application, i.e. being fit for purpose, is clearly recognized in these regulatory documents but at the same time the need for release method type controls of methodologies is also understood to be unnecessary at this characterization stage. Strength of data and the subsequent conclusions drawn for structural characterization purposes are derived from the use of orthogonal techniques where possible, a point emphasized in the guidelines.
For biosimilar analysis, qualification is useful for the determination of the expected experimental range or error of the method being employed. With this knowledge in hand, a comparison can be made between innovator and biosimilar-derived values, allowing a better assessment of similarity or otherwise between samples for the parameter in question.
Qualification of methods and instrumentation
There is no explicit mention of an expected procedure for qualification in the quoted biosimilar guidelines, and no request drawing upon any principles of the ICH Q2(R1) document on validation of analytical methods. However, when qualifying any method, the application of the principles in this document for generation of a qualified (as opposed to ‘validated’) method in a broad sense would seem wise. Use of appropriate standards to qualify a procedure, in terms of both instrumentation and methodology, can demonstrate its general applicability to a particular analysis with due regard given to whether the method is assessing a particular aspect of the sample in a qualitative or quantitative sense (through considerations of specificity, repeatability, accuracy, linearity, LOD, LOQ, intermediate precision and robustness as appropriate). The method then becomes suitable for use in characterization protocols where the same tenets hold true of the sample as for the standard used in the qualification.
This general method qualification should be repeated on a cycle of one or two years where the same process is repeated to demonstrate that the system (i.e. instrument and methodology) is still performing within the defined parameter range. Where consideration may need to be given to unique features of any biosimilar that requires specific assessment in a method application, qualification may be performed once a method has been developed specifically for that product and structural aspect. An example of this could be a method used to assess the presence of a unique or uncommon post-translational modification.
It should be noted that the basic ‘go-to’ analytical techniques used during product characterization have usually been applied across the analyses of many molecules, including biosimilars, over time and thus have been demonstrated to have basic applicability within the area of analytical structure determination. Therefore, specific qualification of these methods may not be appropriate.
Wherever possible, appropriate checks of the method should be used during the course of sample analysis regardless of whether the method is qualified or not. These checks normally comprise of system suitability tests (SSTs) run at the start and end of an instrument analytical session. The data obtained serve to demonstrate that detection parameters, such as sensitivity and resolution, are optimal and therefore sample results are valid and data can be processed and interpreted. If there are a high number of samples to run, SSTs can also be run at regular intervals throughout the sample runs as an ongoing check that the system has not deviated from the optimal run parameters during the course of analysis. This of course can also be supplemented by interspersed repeat runs of key preparations, such as standard mixtures or blanks, to ensure a consistent and optimal level of performance across the analytical session. The SST may well have its own set of acceptance criteria that must be met, these criteria having come from a qualification of the method and thus a knowledge of how the SST will behave.
One other point to consider is the use of the term ‘qualification’ itself. Qualification does not necessarily only cover the method itself. Qualification is also performed at the time of installation of analytical instruments, in order to demonstrate performance at the expected level for a variety of defined parameters, e.g. sensitivity and resolution. Instrument qualification procedures cover installation, operation and performance qualifications (often described as IQ, OQ and PQ, respectively). On this basis, the instrument itself is then qualified to operate as expected and the method can then be developed with confidence in the system it is being run on.
Method qualification examples: extinction coefficient and assessments of impurities
Extinction coefficient determination One area where method qualification is important at an early stage of product development is in determining the extinction coefficient (e). The extinction coefficient is the measure of light absorption per unit amount of sample and is intimately related to absorbance (A), through the Beer-Lambert Law A=ecl, to the concentration of the sample (c) and pathlength (l) of the sample through which the light passes. Thus, in order to determine the concentration of a biopharmaceutical active ingredient from an absorbance measurement, one must determine the extinction coefficient.
For determination of the extinction coefficient, concentration of the protein/glycoprotein can be measured using amino acid analysis, see Figure 1, and the determined concentration assessed alongside the measured optical density and a value obtained. The significance of this value is that it can be used to measure the concentration of all prepared samples using an ultraviolet (UV) spectrophotometer for studies in subsequent biological analyses. It is critical that the amount of sample used in these studies is accurately known for dosing and safety reasons. Thus, a qualification of the methods used to determine the extinction coefficient (optical density (OD) spectroscopy and amino acid analysis) is important in demonstrating that the value obtained is coming from a reliable method and therefore can itself be analytically supported and relied upon.
Impurities Another area that the ICH Q6B and biosimilarity guidelines highlight is that of assessing impurities present in the product. For biosimilar products it is not expected that the impurities will be the same as the innovator since the manufacturing process will not necessarily be the same. Impurities are defined as being either product or process related.
Product related impurities are the result of differential processing of the product either during biosynthesis or subsequent purification. These are considered part of product characterization itself and will thus naturally be identified and characterized during those analyses. Process related impurities cover all aspects of the drug production process where contaminants are likely to be derived both from cellular and production sources. Removal of these impurities to the extent possible, or at least to below set limits, is critical for product quality and safety.
Methods that assess impurities in the characterization phase can be used in the absence of qualification, however given the importance of impurity clearance as a key consideration in the drug production process it is worth considering qualification of residual testing methods for key process-related impurities, see Figure 2.
This will allow confidence in the method and thus support for the results generated, so that meaningful changes can be made to the production process, if necessary, on the basis of reliable data. This alleviates a potential area of risk that could be very costly and time-consuming to fix if issues with levels of impurities were discovered at a later date. These qualified methods can, if required, be subsequently validated for batch release purposes.
Methods for critical quality attributes
As mentioned above, structural analysis uses a wide variety of instrumentation and techniques to probe all aspects of biopharmaceutical/biosimilar structure. Once the structure is determined (using multiple batches of biosimilar and innovator, if the product is a biosimilar), and the manufacturing process has been tied down, batches can be manufactured for subsequent biological testing.
At this point, the methods that will be used for testing of particular key structural features, or critical quality attributes, should be considered. These methods are designed to demonstrate that the batch being tested is showing the expected structure with no anomalies or aberrations. Tests to consider will vary between different products but should provide unique information on the molecule and be responsive to a change in molecular structure such that the method should reliably detect any change in the product.
Example: Peptide mapping – MS or UV detection? The online LC/ES-MS analysis used for the peptide mapping study during the characterization phase can potentially be reduced down to UV detection rather than MS detection, once structure and the manufacturing process are determined. This means that, with all other parameters, such as column and gradient conditions remaining constant, a simple fingerprint profile could be used, where the identities of the UV peaks are known from the initial MS characterization work. This method can be qualified/validated to demonstrate its applicability prior to use in batch release which would include an assessment of the sensitivity of the method to any structural changes that may occur, for example, variation in the levels of post-translational modifications. Other tests will need to be considered to monitor other aspects of molecular structure, such as intact mass analysis and glycan profiling, since peptide mapping is not a catch-all for assessment of all aspects of a molecule’s structure.
Qualification can de-risk future validation
Ultimately, appropriately selected methods used to assess the quality of a product will need to be validated and performed in a GMP environment. The decision to qualify a method in no way affects the necessity to also validate the method for the later phases of drug development, with validation of methods expected by the regulatory authorities at these stages of drug development. So what good is qualification of a method as a sort of half-way house? This comes down to the need or desire to ensure that the method is producing reliable and meaningful data for the structural aspect of the product being measured, such that the quality of the product and the process is built in from an early stage. A method, qualified through the principles of ICH Q2(R1), should be straightforward to validate since that method will already have gone through the testing phase. It will therefore stand the validation process with less risk than an untried method that has come to be relied upon but whose characteristics have not been tested rigorously.
No hard and fast rules
Method qualification is something that needs to be considered during product development based on the nature of the specific analysis being undertaken and the nature of the methodology being used. For example, qualifications of methods for residual testing and impurities, or other methods used to demonstrate the correct nature of a particular structural feature may be viewed differently compared to qualifications of structural characterization techniques using mass spectrometry. There is no hard and fast rule covering when to qualify a method or what methods should be qualified, but the qualification of selected methods early in product development can help to avoid problems with test processes that may be required for use later but which are shown to have features that preclude meeting validation criteria.
In summary, what is required at the early stage of product development, i.e. at the characterization and definition stage, is that methods are fit for purpose and scientifically accurate such that data derived from these methods are trustworthy and sound. Thus, in biosimilar analysis, the comparative data generated will allow meaningful conclusions to be drawn regarding structural similarities and differences.
Funding sources
This paper is funded by BioPharmaSpec Ltd (www.biopharmaspec.com).
About the author: Richard obtained his PhD in glycoprotein structural characterization using mass spectrometry from Imperial College of Science, Technology and Medicine. He subsequently spent several years there as a postdoctoral research scientist working in the field of glycoprotein structural characterization with emphasis on glycan elucidation. The projects he was involved in required detailed structural analysis of glycoproteins derived from animal, plant and fungal systems, very frequently expressing unusual glycosylation profiles. He moved to GlaxoSmithKline for a short time where he was head of mass spectrometry for the toxicoproteomics and safety assessment group. Richard joined M-Scan Limited (now part of SGS Life Sciences) in 2003 as a biochemist and became the Team Leader for Carbohydrate Analysis before being appointed Principal Scientist. Richard joined BioPharmaSpec in 2016 as Technical Director for Structural Analysis and is responsible for management of all aspects of carbohydrate and glycoprotein characterization at the primary structure level.
References 1. European Medicines Agency. ICH Topic Q6B specifications: test procedures and acceptance criteria for biotechnological/biological products. Guidance Document. August 1999 [homepage on the Internet]. [cited 2022 Mar 28]. Available from: https://www.ema.europa.eu/en/ich-q6b-specifications- test-procedures-acceptance-criteria-biotechnologicalbiological-products 2. European Medicines Agency. Guideline on similar biological medicinal products containing biotechnology-derived proteins as active substance: quality issues (revision 1). EMA/CHMP/BWP/247713/2012. 22 May 2014 [homepage on the Internet]. [cited 2022 Mar 28]. Available from: https://www.ema.europa.eu/en/documents/scientific-guideline/guideline-similar-biological-medicinal-products-containing-biotechnology-derived-proteins-active_en-0.pdf 3. U.S. Food and Drug Administration. Development of therapeutic protein biosimilars: comparative analytical assessment and other quality-related considerations. Guidance for Industry. Draft Guidance. May 2019 [homepage on the Internet]. [cited 2022 Mar 28]. Available from: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/development-therapeutic-protein- biosimilars-comparative-analytical-assessment-and-other-quality 4. Government UK. Guidance on the licensing of biosimilar products [homepage on the Internet]. [cited 2022 Mar 28]. Available from: https://www.gov.uk/government/publications/guidance-on-the- licensing-of-biosimilar-products 5. European Medicines Agency. ICH Q2(R1) validation of analytical procedures: text and methodology. June 1995 [homepage on the Internet]. [cited 2022 Mar 28]. Available from: https://www.ema.europa.eu/en/documents/scientific-guideline/ich-q-2-r1-validation-analytical-procedures-text-methodology-step-5_en.pdf
Author: Richard L Easton, BSC (Hons), DIC, PhD, Technical Director, BioPharmaSpec Ltd, Suite 3.1 Lido Medical Centre, St Saviours Road, St Saviour, Jersey JE2 7LA
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
Author byline as per print journal: Robert Janknegt, PharmD, PhD; Marloes Dankers, MSc
Study objectives: The increasing number of long-acting insulin analogues makes it difficult for general practitioners to have sufficient knowledge of each individual insulin formulation and device. Reducing the number of different insulin analogues used, based on rational criteria, allows physicians and pharmacists to build experience with a more limited set of medicines and to optimize patient information. The costs of newer formulations also need to be considered. Methods: In this study long-acting insulin analogues are compared by means of the System of Objectified Judgement Analysis (SOJA) method. The following selection criteria were applied: effectiveness, safety, tolerability, ease of use, and applicability. Concrete sub-criteria for each selection criteria were defi ned. Results and Conclusions: We found almost identical scores for the long-acting insulin analogues. Therefore, acquisition cost and individual patient characteristics should be used to determine the choice of formulary products. Glargine 100 IU/mL (biosimilar or originator) is the cheapest option in most cases. It should be noted that ‘good old’ Neutral Protamine Hagedorn (NPH) insulin is still a good treatment option from a formulary perspective NPH insulin is also likely to be cheaper than biosimilar glargine in most countries.
Submitted: 24 December 2022; Revised: 22 March 2022; Accepted: 24 March 2022; Published online first: 31 March 2022
Introduction
The incidence and prevalence of both diabetes mellitus type 1 and (especially) type 2 has been steadily increasing. There are now about 60 million people with diabetes in the European Region, or about 10.3% of men and 9.6% of women aged 25 years and over [1].
A substantial number of type 2 diabetes patients eventually require insulin. Different forms of insulin, such as short-, intermediate- and long-acting insulins are available. According to the guidelines of the Dutch College of General Practitioners ‘Type 2 Diabetes Mellitus in primary care’, (long-acting) NPH-insulin is considered the first choice if insulin treatment is indicated. The insulin analogues insulin glargine 100 U/mL and insulin detemir are considered potential alternatives in specific situations, especially when patients experience frequent night-time episodes of hypoglycaemia. The other analogues, insulin glargine 300 U/mL and insulin degludec, are not (yet) recommended, because of the lack of evidence-based advantages compared to other long-acting insulins, insufficient knowledge of long-term safety (in the case of insulin degludec), and costs [2].
Despite the Dutch preference for NPH-insulin, insulin analogues, including biosimilars, are prescribed extensively by European physicians [3-5]. In primary health care, insulins account for a high percentage of drug expenditures. The introduction of the first generation long-acting insulin analogues (insulin glargine 100 U/mL and insulin detemir) in the first decade of the 21st century resulted in rapidly increasing dispensing rates and healthcare costs [6-10]. Because the second generation long-acting insulins (insulin glargine 300 U/mL and insulin degludec) are even more expensive than glargine and detemir, they are thought to be responsible for further increases in total insulin expenditures [9].
Therefore, there is a need for practitioners to make more rational choices between the increasing number of long-acting insulin analogues when they prescribe insulin products.
The following insulin analogues are included in this analysis:
Insulin degludec is a modified form of human insulin in which the amino acid at position B30 is deleted and the lysine at position B29 is conjugated to hexadecanoic acid via a gamma-L-glutamyl spacer.
In insulin detemir a 14-C fatty acid is coupled to the lysine residue on position 29 of the B chain and threonine is removed at position 30 of the B-chain.
In insulin glargine, asparagine has been replaced by glycine at position 21 in the A-chain and by the addition of two arginine residues to the end of the B-chain, at positions 31 and 32. Both ‘manipulations’ result in a slower absorption, leading to a more gradual absorption profile compared to NPH insulin.
Applied methodology The System of Objectified Judgement Analysis (SOJA) method is a decision model used to make formulary recommendations, in which the authors prospectively define selection criteria for a given drug group and assign a relative weight to each criterion by consensus between the authors based on their perceptions of their local situation [11]. The more important a given criterion is judged to be, the higher the relative weight that is assigned to it. The properties of the drugs are evaluated per criterion and each individual drug is assigned a score for each criterion; this score being a percentage of the relative weight of this criterion. The closer the drug approaches the ‘ideal’ drug, the higher the percentage score for that criterion. The drugs with the highest overall scores are considered the most attractive ones for formulary inclusion.
A (English, German and Dutch) Medline and Embase search was performed, as well as a search for studies in the Cochrane library and collection of the references from meta-analyses and systematic reviews to obtain relevant studies regarding long-acting insulin analogues. In addition to these searches, the references of review articles on this subject were obtained and incorporated in the analysis when appropriate and these were included in the manuscript. Search terms were degludec, detemir, glargine and NPH.
The present SOJA score has five sets of (sub-) criteria for the selection of long-acting insulin analogues. These criteria are:
Efficacy (the actual positive outcomes and treatment goals)
Safety (the avoidance of negative outcomes, such as hazardous side effects)
Tolerance (the interruption of the care process due to less hazardous, generally transitory, but disturbing side effects)
Ease of use (ease for the patient, for example, dosing frequency)
Applicability (the scope of the treatment freedom (interactions and such) and the ease for the caregiver)
These criteria are specifically described per selection subject (‘operationalised’). All scores were assigned based on a consensus of the authors, see Table 1.
Selection criteria
1 Effectiveness 1.1 Risk of late complications The ultimate goal of T2DM treatment is the minimization of cardiovascular complications (including renal failure), immortality, and microvascular complications (neuropathy).
The score assigned was dependent on the data available regarding risk reduction by the individual insulin analogues.
The criterion was separated into macrovascular complications and microvascular complications.
1.2 Effect on metabolic control According to the Dutch T2DM guidelines, haemoglobin A1c (HbA1c)-targets depend on age, duration of T2DM and use of other blood glucose lowering drugs. The relationship between HbA1c and cardiovascular outcomes remains to be elucidated, as a result of contradictory study results.
2 Safety The extent and the severity of adverse effects is another important drug selection criterion. A distinction was made between ‘minor’ side effects, such as gastrointestinal disturbances or skin reactions, occurring in clinical trials (scored under tolerability) and severe or even life-threatening adverse reactions observed with large scale studies of the use of the drugs.
The following aspects were taken into consideration: 2.1 Risk of severe hypoglycaemia 2.2 Cancer and heart failure 2.3 Documentation
Obviously, there is less experience with recently introduced medicines compared with medicines that were introduced earlier. This makes it more uncertain whether the newly introduced medicines are as safe as older medicines. Therefore, this item was scored under the main criterion of safety.
The first two sub-criteria below (number of double-blind clinical studies and number of patients in these studies) are indicative of the overall quality and quantity of the randomized, controlled clinical studies of the drugs. A large number of clinical studies and the large number of patients included in these studies leave no doubt about the clinical efficacy and safety profile of these drugs in the populations studied.
The latter two criteria (number of years available and number of patient days) are indicative of the overall clinical experience with the drug in question. These sub-criteria may introduce a bias to the advantage of older drugs, but this is done intentionally. The safety of a newly introduced drug cannot be guaranteed from only the results of clinical studies, in which only a relatively small number of patients were included and most patients at risk for the development of adverse reactions (e.g. patients with diminished renal function) were excluded. Both the number of patients that have been treated on a worldwide basis and the period that a certain drug has been available are of importance, as it may take time until adverse reactions occur or be recognized.
Number of double-blind comparative clinical studies 5% was given for each study. Number of patients in these studies 1% was given for every 10 patients included in the above studies. Number of years available on the market 10% was given for every year that the drug was available on the market. Number of patient days experience 1% was given for every million patient days.
3 Tolerability< The extent and the severity of adverse effects is another important selection criterion for drugs. A distinction was made between ‘minor’ side effects, such as gastrointestinal disturbances or skin reactions, that occurred in clinical trials and the severe or even life-threatening adverse reactions observed with large scale use of the drugs (scored under safety). The evaluation of the ‘minor’ adverse effects was based on the results of double-blind comparative clinical studies.
The following aspects were considered: 3.1 Mild hypoglycaemia 3.2 Weight gain 3.3 Pain at injection site
4 Ease of use Because long-acting insulin analogues must be used lifelong, it is of importance that their use is made as easy as possible.
The following aspects were taken into consideration:
4.1 Dosing frequency A low dosage frequency is of great importance in life-long treatment of diabetes. Patient compliance is best with once daily dosing, although the difference between once and twice daily dosing is not impressive. Patient compliance drops significantly at higher dosage frequencies.
This was scored as follows:
4.2 User-friendly administration form An easy-to-use administration device may increase patient acceptance and compliance. A pen allows more patient-friendly administration than an injection vial for some patients.
5 Applicability This criterion is focused on the caregiver. The following aspects are taken into consideration:
5.1 Medication interactions This criterion is of importance in formulary decision-making since the majority of treated diabetic patients need to take other medications as well. Drug interactions may result in an increased or reduced clinical efficacy of the antidiabetic medicine in question or in a reduction of the clinical efficacy of the other drug, with which the interaction occurs. Interactions may also give rise to increased toxicity of one or both medications. The more frequent these interactions occur and the more serious the consequences are, the lower the score for the drug in question.
5.2 Contra-indications, special warnings, and precautions Contra-indications, warnings, and precautions may hinder the caregiver in the prescription of long-acting insulin analogues to patients. The score is dependent on the extents of contra-indications, warnings, and precautions.
5.3 Dose adjustment Dose adjustments may be necessary in patients with renal or liver function impairment. When dose adjustment is not necessary, this provides an advantage over a medicine for which adjustments are necessary.
Results
1 Effectiveness 1.1 Risk of late complications 1.1.1 Macrovascular complications Both diabetes and older age are important risk factors for cardiovascular complications. Coronary heart diseases are the most important causes of death in patients with type 2 diabetes [12]. The cardiovascular risk profile of diabetes patients is comparable to individuals without diabetes who are 15 years older [13].
One large scale study showed similar effects of glargine and standard care on the risk of myocardial infarction, stroke, or cardiovascular death. The co-primary outcomes were non-fatal myocardial infarction, non-fatal stroke, or death from cardiovascular causes, and any of these events plus revascularization or hospitalization for heart failure. The incidence of cardiovascular outcomes was similar in the insulin-glargine and standard-care groups: 2.94 and 2.85 per 100 person-years, respectively, for the first co-primary outcome (hazard ratio, 1.02; 95% confidence interval [CI], 0.94 to 1.11; p = 0.63) and 5.52 and 5.28 per 100 person-years, respectively, for the second co-primary outcome (hazard ratio, 1.04; 95% CI, 0.97 to 1.11; p = 0.27). Only 11 per cent of the patients in the standard therapy group received insulin. A1C values were comparable at baseline and at study end. About 60 per cent of patients in both groups were treated with statins and 75 per cent with inhibitors of the renin-angiotensin system [14, 15].
All studies with detemir and glargine were of insufficient duration and size to be able to show a relevant beneficial effect (let alone a difference between both drugs) regarding macrovascular complications.
1.1.2 Microvascular complications Microvascular complications were determined based on the long-term effects on metabolic control. These complications can be subdivided into damage to the eye (retinopathy), kidneys (nephropathy), and nerves (neuropathy). In UKPDS 34, a 25% reduction in microvascular complications was found for intensive T2DM therapy compared to conventional treatment. This was mainly correlated with a good metabolic control (HbA1c) and was not found to be associated to the use of any specific blood glucose lowering medications [16]. Other studies came to similar conclusions, both in type 1 diabetes mellitus and in type 2 diabetes mellitus patients [17-20]. A small scale, crossover study between detemir and glargine (n = 42) in diabetes mellitus type 2 patients with macroangiopathy showed a comparable improvement of parameters of endothelial damage for both drugs [21].
1.1.2.1 Retinopathy Diabetic retinopathy is the most important cause of blindness. Elderly diabetic patients also have an increased risk of other ocular diseases, such as glaucoma, cataract, and macular degeneration. Regular eye examinations are therefore important [22].
UKPDS 52 investigated the relationship between the severity of retinopathy and the need for photocoagulation after 3, 6 and 9 years of treatment. In patients without retinopathy at baseline, photocoagulation was necessary in 0.2% after 3 years and in 2.6% after 9 years.
The incidence was higher in patients with a microaneurysm at baseline: 0% after 3 years and 4.7% after 9 years. In patients with retinopathy at baseline, photocoagulation was necessary in 15% after 3 years and in 32% after 9 years [23]. UKPDS 50 showed comparable results after a follow-up of 6 years [24].
The UKPDS 33 studies showed significant differences between intensive and conventional treatment groups on a number of individual endpoints associated with diabetic neuropathy, such as photocoagulation of the retina (29% reduction, from 11.0 to 7.9 per 1,000 patient years, p = 0.0031 and cataract extraction (24% reduction, from 7.4 to 5.6 per 1,000 patient years, p = 0.046). No significant differences were seen in blindness or bleeding into the eye [25]. Regarding individual drugs, only glibenclamide showed a significant reduction in the incidence of photocoagulation. No effects of any other individual drug on any parameter were observed. A significant reduction was seen in some ‘surrogate-parameters’, like the two-step progression of retinopathy [25].
Glargine has a slightly stronger affinity for IGF-1 receptors than do other insulins, theoretically showing an increased risk of mitogenic activity and an unfavourable effect on the progression of retinopathy. The affinity of glargine to the receptor is however only 0.5% of that of IGF-1 and the greater affinity compared to insulin is only seen at concentrations which are above ‘therapeutic’ concentrations [26]. So far, there are neither epidemiological nor clinical data indicating an unfavourable effect of glargine on retinopathy [27]. All studies with detemir and glargine are of insufficient duration and size to be able to show a relevant effect (let alone a difference between both drugs) regarding retinopathy. In an analysis of four comparative studies no relevant difference was seen between glargine and NPH insulin in the progression of early retinopathy.
1.1.2.2 Nephropathy The first sign of diabetic nephropathy is microalbuminuria, which is followed by proteinuria. Diabetic nephropathy is one of the most important indications to start haemodialysis.
Nephropathy is an important complication of diabetes mellitus. About 25% of type 2 diabetes mellitus patients develop microalbuminuria and 5% of these patients develop macroalbuminuria in 10 years. Only relatively few patients (0.8%) show an increased serum creatinine or need renal function replacement therapy. The latter group however has a considerable increase in disease-related mortality [28].
The UKPDS 33 studies showed no significant differences between intensive and conventional treatments on individual endpoints associated with diabetic nephropathy, such as renal failure or death from renal causes. A significant reduction of microalbuminuria and proteinuria was seen in the intensive group after 9 years of treatment. No effect was seen on creatinine clearance [25].
All studies were of insufficient duration and size to be able to show a relevant effect (let alone a difference between both drugs) regarding nephropathy. No comparative studies between insulins on the effect on nephropathy were identified.
1.1.2.3 Neuropathy Foot problems are frequently observed as a complication of diabetes, especially in the elderly. Both vascular and neurogenic factors are involved. The prevalence of diabetic neuropathy in patients above 60 years may be over 50%. Diabetic neuropathy is a feared complication of type 2 diabetes mellitus, resulting in paresthesia, burning feeling, or decreased pain sensation, especially in the feet. This may lead to ulceration, infection and gangrene or amputation of the feet [22]. The UKPDS 33 studies showed no significant differences between the intensive and the conventional treatment on individual endpoints associated with diabetic neuropathy, like amputation [25].
All studies with detemir and glargine were of insufficient duration and size to be able to show a relevant effect (let alone a difference between both drugs) regarding neuropathy.
With respect to macrovascular and microvascular complications, there are no indications of any clinically meaningful differences between degludec, detemir and glargine. All drugs were therefore assigned a value of 50%.
1.2 Effect on metabolic control The direct effect of insulins can be determined by their effect on glucose concentrations. One study showed a significant reduction of mortality for intensive treatment of diabetes (strict glucose regulation, use of renin-angiotensin inhibitors, aspirin and statins) compared to conventional treatment of diabetes after 13 years of treatment. The relative risk of death was 0.54 in the intensively treated group [29].
Two more recent prospective studies concerning the effects of intensive treatment of diabetes mellitus (target value for HbA1c 6.0%, respectively 6.5%) showed contradictory results. In the ACCORD study (n = 10.251), no significant reduction was seen of important cardiovascular events in the group treated intensively for 3.5 years (HbA1c 6.4%) in comparison with the standard treatment group (HbA1c 7.5%). An increased risk of death was seen in the intensive treatment group (Hazard ratio 1.22. p = 0.04), resulting in premature closure of the study after 3.5 years [30].
In the ADVANCE study (n = 11.140), a significant reduction in the combined endpoint of macrovascular and microvascular complications was seen (Hazard ratio 0.90. p = 0.01) after 5 years of intensive treatment (HbA1c 6.5%), but especially in a reduction of nephropathy (Hazard ratio 0.79. p = 0.006), whereas no significant effect on retinopathy was seen. No increased mortality was seen in this study in the intensively treated group (Hazard ratio 0.93. not significant) [31].
1.3 Relative effects on HbA1c Insulin therapy is usually titrated until the target HbA1c is reached [2]. Therefore, no substantial differences between insulins are expected.
There are no double-blind studies available that included patients treated with detemir or glargine. All available studies are open label, randomized studies.
Direct comparison between detemir and glargine One direct comparison study between detemir and glargine, dosed based on the effects on HbA1c, showed (not surprisingly) an identical effect on HbA1c: a decrease of 1.5% compared to baseline. It should be noted that the dosage of detemir (78 IEU) given was higher than that of glargine (44 IU) [32].
In a small-scale crossover study, detemir and glargine showed a similar effect on the 24 h glucose profile [33].
Another study could not demonstrate non-inferiority of detemir compared to glargine. Glargine showed numerically (but not statistically significantly) greater reduction of HbA1c and fasting blood glucose [34].
A meta-analysis did not show relevant differences between detemir and glargine with respect to efficacy and safety [35].
Direct comparisons between degludec and glargine Various studies compared glargine to insulin degludec. The effects of both drugs on HbA1c, fasting blood glucose, and body weight were almost identical [36-72]. It should be noted that a recent meta-analysis showed a slight, but statistically significant difference in the effects on HbA1c between glargine and degludec: 0.1% in favour of glargine [73].
Direct comparison between degludec and glargine 300 IU/mL Some studies compared glargine 300 IU/mL to insulin degludec. The effects of both drugs on HbA1c, fasting blood glucose, and body weight were almost identical [74-78].
Direct comparison between glargine and biosimilar glargine Biosimilar glargine (LY2963016) was as effective as glargine in several direct comparison studies [79-85].
Direct comparison between glargine 100 IU/mL and 300 IU/mL Glargine 300 IU/mL was as effective in lowering HbA1c as glargine 100 IU/mL in several direct comparison studies [72, 86-101] and in a meta-analysis [73].
Comparisons with NPH insulin The effects of detemir and glargine on metabolic control (in comparison with NPH insulin) were investigated in a limited number of studies [102-125]. These studies are included in a number of meta-analyses. The most important conclusions of those meta-analyses are summarized below.
A Cochrane analysis from 2020 included 16 studies with glargine and 8 studies with detemir [126]. In this analysis, only randomized comparative studies between both drugs and NPH insulin were involved with a treatment duration of at least 24 weeks. No significant differences were seen in metabolic control, measured as effect on HbA1c between detemir and NPH insulin and between glargine and NPH insulin. The mean difference in HbA1c for glargine versus NPH was –0.07% (95% CI –0.18 to 0.03; p = 0.17). For insulin detemir compared to NPH insulin, the mean difference was 13% (95% CI –0.02 to 0.28; p = 0.08). Other comparative studies Several other studies were performed between the formulations included in this analysis and other blood glucose lowering agents [127-167].
By far the most studies are performed with glargine, whereas fewer studies were done with detemir. Studies comparing two treatment schedules with glargine [168] or non-randomized studies [169], were not included in this analysis. This is also true for similar studies with detemir [170, 171].
In general, glargine has a stronger effect on HbA1c and fasting blood glucose than increasing the dosage of oral antidiabetics or the addition of other blood glucose lowering medicines [134]. Compared to exenatide, a similar effect on HbA1c was seen, but a more marked effect on fasting blood glucose [130, 131, 172]. In comparison with various insulin lispro schedules, a similar or lesser effect on HbA1c was seen, but a stronger effect on fasting blood glucose [135-138].
Comparable insulin requirements were found for both drugs in a double-blind crossover study [139].
An analysis of the German/Austrian DPV-wss database (over 51,000 patients) showed that the mean daily insulin requirements were significantly lower for glargine (0.29 IU/kg) than for detemir (0.33 IU/kg) [144].
The dosage requirements for glargine 300 lU/mL were higher than those for glargine 100 lU/mL (34 IU vs 30 IU) after transition to glargine 300 lU/mL [145].
There are no indications of clinically meaningful differences between the insulin analogues, as was expected, because all insulins are normally titrated until the target HbA1c is reached. All drugs score 80%.
2 Safety Hypoglycaemia is a well-known side effect of insulins. Severe hypoglycaemia is scored under safety (because it may be life threatening), whereas mild cases of hypoglycaemia are scored under tolerability.
2.1 Risk of severe hypoglycaemia The most recent Cochrane meta-analyses of the incidence of severe hypoglycaemia showed no significant difference between detemir and glargine on one hand, and NPH insulin on the other, although a trend was observed to a lower incidence for both insulin analogues compared to NPH insulin. Detemir showed a small, but significant reduction of serious hypoglycaemia compared to NPH insulin, which was not demonstrated for glargine, despite the larger number of studies [126]. The incidence is however too low to be considered a significant difference in small-size studies. A trend towards lower incidence was found for detemir (RR 0.50; 95% confidence interval 0.18‘’1.38) and glargine (RR 0.70. 95% confidence interval 0.40‘’1.23). There are no indications that clinically relevant differences exist between detemir and glargine in the incidence of severe hypoglycaemia. In the direct comparative study between both drugs, an incidence of severe hypoglycaemia of 2% was found for detemir and 3% for glargine [32].
The same is true for the incidence of severe hypoglycaemia for degludec. The number of events was too small to show a significant difference between degludec and glargine [32].
Biosimilar glargine (LY2963016) was as safe as glargine in several direct comparative studies [79-84].
Glargine 300 IU/mL was as safe as glargine in several direct comparative studies with a lower incidence of hypoglycaemia in some studies [86-128].
A meta-analysis showed a lower incidence of severe nocturnal hypoglycaemia for degludec compared to glargine, whereas glargine 300 IU/mL reduced both daytime and nocturnal hypoglycaemia [73].
One real-life, non-comparative, Finnish study showed a lower incidence of severe hypoglycaemia for detemir, but not glargine compared to NPH insulin [173]. Glargine 300 IU/mL and degludec are awarded 80%. Glargine and detemir score 75%.
2.2 Cancer and heart failure 2.2.1 Cancer There has been debate over whether use of glargine increases the risk of cancer development [174]. However, several studies have now demonstrated that there is no increased risk [175-181]. In addition, no safety signals regarding cancer risk are identified for other insulins.
2.2.2 Heart failure A comparative study did not show differences between degludec and glargine in the incidence of heart failure [182]. No increased risk of heart failure is identified for detemir. All drugs are awarded 90% for the sub-criterion: cancer/heart failure.
2.3 Documentation The documentation of the drugs (only comparative studies with more than 25 patients per treatment arm and a treatment duration of at least 10 weeks) is summarized in Table 2.
Detemir and glargine are more extensively documented than degludec.
Because of the minor differences between classical glargine (100 lU/mL) and the new formulations (300 lU/mL) and biosimilar glargine, the clinical experience of glargine was also considered valid for the latter two compounds. The number of available studies was evaluated separately.
3 Tolerability 3.1 Mild to moderate hypoglycaemia Three meta-analyses showed a significant difference between both insulin analogues and NPH insulin in terms of the incidence of mild to moderate hypoglycaemia and nocturnal hypoglycaemia [126, 183, 184]. The most recent (Cochrane) meta-analysis [126] also showed a lower incidence. The authors found that the relative risk of hypoglycaemia to be glargine 0.88 (95% CI 0.81‘’0.96) and detemir 0.73 (95% CI 0.61‘’0.86). Also, the risk of nocturnal hypoglycaemia was lower with relative risks: glargine 0.74 (95% CI 0.64‘’0.85) and detemir 0.32 (95% CI 0.16‘’0.63).
These results should be interpreted with caution. The authors of the analysis make the following remarks:
‘In the studies, very low blood glucose and HbA1c target values were set. However, doctors often recommend higher targets for people with a long history of type 2 diabetes, who have had a heart attack or stroke, or who are old. With higher target values, hypoglycaemia occurs less frequently, and more people need to be treated with insulin analogues instead of NPH insulin to prevent hypoglycaemia in one person. Therefore, study results are only applicable to people who are treated to such low blood glucose target values. In many studies, an adequate adjustment of NPH insulin was not possible. However, doctors will do that in daily practice. Therefore, a further decrease in the benefit of insulin analogues is expected’.
Comparison between detemir and glargine A meta-analysis did not show relevant differences in hypoglycaemia between detemir and glargine [35].
Comparison between glargine 100 IU/mL and 300 IU/mL Glargine 300 IU/mL was associated with a lower incidence of nocturnal hypoglycaemia than glargine in several direct comparative studies [86-101].
Comparison between glargine and degludec Most comparative studies between glargine and insulin degludec showed that degludec exhibited a lower incidence of hypoglycaemia [36-53]. This was also confirmed in several meta-analyses [146-151].
Other comparative studies with glargine and detemir A planned year one interim analysis of a three-year comparative study of detemir and biphasic insulin twice daily, and shortacting aspart taken three times daily, in insulin-na¡ve patients who had insufficient control of HbA1c on oral treatment with metformin and/or sulfonylurea derivatives showed significantly less hypoglycaemia with detemir than with the other insulin regimes. Here there were 2 cases per year for detemir vs 8.9 for biphasic insulin and 16 for short-acting insulin. No cases of severe hypoglycaemia were observed [129].
Glargine showed a low incidence of hypoglycaemia in comparative studies with exenatide, oral antidiabetics of various insulin-regimens [130-138].
It should be kept in mind that lower blood glucose and HbA1c target values were set in studies comparing degludec and glargine 300 IU/mL to glargine 100 IU/mL than the values used in daily practice. Therefore, the differences are of limited clinical importance.
Degludec scored 80% because of the relatively low incidence of hypoglycaemia, glargine 300 IU/mL scored 75% and glargine 100 IU/mL and detemir scored 70%.
3.2 Weight increase The meta-analysis of Monami [183] showed a significantly lower weight increase for detemir compared to NPH insulin, although the absolute difference in weight increase was small (0.7‘’2.0 kg). No good explanation is available for this observation. The European Medicines Agency’s European public assessment report (EPAR) of detemir also mentions a lower degree of weight increase for detemir compared to NPH insulin [185].
No differences were seen in the extent of weight increase between glargine and NPH insulin in a meta-analysis [183], nor is this reported in the Scientific Discussion of EMA [186].
The more recent Cochrane meta-analysis showed a non-significant weight increase of 0.12 kg for glargine, compared to NPH insulin. No comparison of detemir and NPH insulin was presented in this study [126].
A direct comparative crossover study between both drugs showed a weight increase of 1.6 kg for glargine, versus a weight loss of 0.4 kg for detemir [21].
In a subsequent randomized study of glargine over time no extra weight increase was seen, with an average 2.1 kg after 1 year treatment and 2.0 kg at the end of 39 months follow-up [187].
In summary, it can be concluded that detemir results in significantly less weight increase compared to NPH insulin, but that the absolute differences are small. This has not been demonstrated for glargine.
Clinical studies show there are no clear differences between degludec and glargine in terms of weight gain in clinical studies.
It is unclear whether there are significant and relevant differences between the medicines on this criterion. Detemir is awarded 80% and glargine and degludec 70%.
3.3 Pain at the injection site Pain at the injection site is seen more often for detemir than for NPH insulin [188]. This is, to a lesser extent, also the case for glargine compared to NPH insulin [27]. In the direct comparative study between glargine and detemir, a slightly higher incidence of pain was seen for detemir (4.5%) than for glargine (1.4%). Most cases are mild to moderate.
Lipodystrophy can be prevented by frequently changing the site of injection [26].
Studies with degludec are lacking.
Biosimilar glargine (LY2963016) was as well tolerated as glargine in several direct comparative studies [79-82].
Glargine 300 IU/mL was as well tolerated as glargine in several direct comparative studies [86-93].
Glargine is awarded 70% and detemir and degludec 65%.
4 Ease of use 4.1 Dosage frequency Insulin glargine and degludec can be given once daily. Insulin detemir can be given once or twice daily.
Glargine and degludec are awarded 100% and detemir is awarded 80%.
4.2 Formulations Both pens and pre-filled injections are available for the drugs.
No independent judgements of the pens or the pre-filled injections are available, so we could not make a distinction between the insulins regarding patient-friendliness and ease of use.
All drugs are awarded 80%.
All insulin analogues are awarded 100%.
5 Applicability 5.1 Interactions All insulin preparations show the same drug interactions.
Many drug classes may lower or increase insulin need. These will affect the clinical efficacy and safety of detemir or glargine to the same extent.
All drugs are awarded 80%.
5.2 Contra-indications, warnings and monitoring The drugs are contra-indicated in case of hypersensitivity. There are no indications for relevant differences in the incidence and severity of hypersensitivity reactions for degludec, detemir and glargine.
This is also true for the warning concerning underdosage (hyperglycaemia) or overdosage (hypoglycaemia). Different doses in IU may be necessary for both drugs during a switch from NPH insulin.
Detemir contains metacresol which may cause allergic reactions. All drugs are (arbitrarily) awarded 70%.
5.3 Dose adjustments Insulin needs may be decreased by decreased gluconeogenesis and decreased insulin metabolism. There is no data indicating meaningful differences between degludec, detemir and glargine.
All drugs are (arbitrarily) awarded 70%.
There do not seem to be relevant differences on this criterion. All drugs are awarded 80%.
According to the SOJA-matrix, no relevant differences exist between long-acting insulin analogues concerning overall score, with less than 2% difference in score between the highest and lowest scores. Perhaps this was not surprising, because insulins share many characteristics. This is especially the case for the different formulations of insulin glargine, which are almost identical. Since no clinically relevant differences exist in effectiveness, safety, tolerability, ease of use and applicability, other arguments must determine the preferred use and individual patient characteristics may be decisive.
Applied methodology The evaluation of the criteria in the SOJA method is highly standardized in order to promote unbiased judgement of drugs from various drug groups based on clinically relevant criteria. The essence of the SOJA method is that users of the method may assign their own relative weight to each selection criterion. This interactive programme is available on the Internet: www.tablet.sojaonline.nl. It is not likely that major differences between the insulin analogues in scores will result in different weightings of the selection criteria. The SOJA method is intended as a tool for rational drug decision-making, forcing clinicians and pharmacists to include all relevant aspects of a certain group of drugs, thereby preventing formulary decisions being based on only one or two criteria. Besides this, possible ‘hidden criteria’ are excluded from the decision-making process.
There will of course always be room for debate whether or not the correct scoring system was used for each criterion and judgement may be arbitrary for most, if not all, criteria. In this analysis, the weights of the different categories and scores assigned to the different insulins were based on consensus between the authors. The outcome of this study should be seen as the basis for discussions within formulary committees and not as the absolute truth.
Outcomes Minimal differences in score are seen between the insulin analogues (less than 2% between the highest and lowest score). Therefore, no relevant differences were found between the long-acting insulin analogues. The situation may be different in individual patients. When an individual patient shows (severe) hypoglycaemia, degludec or glargine 300 IU/mL may offer advantages over glargine 100 IU/mL. As mentioned above, in spite of the standardized method, the scoring will always depend on the somewhat subjective judgement of the scores. Since the differences in the properties of the insulins are so limited, it seems quite unlikely that assigning other relative weights and or scores to the selection criteria will result in meaningful differences in the scores.
Selection criteria Clinical efficacy and safety are the most important selection criteria for all groups of medicines. Meta-analyses and registry data may be of value in the judgement of efficacy and safety. All data sources have specific strengths and weaknesses.
Acquisition cost was not included as a selection criterion to make the score internationally applicable. The present matrix can be used as a pre-selection tool of the most suitable long-acting insulin analogues from a product specific quality point of view. Because prices may differ between institutions and in different healthcare systems, individual procurement procedures should lead to a selection of the most cost-effective options. Glargine 100 IU/mL (biosimilar or originator) is the cheapest option in most cases.
We did not include NPH insulin in this score, because this analysis was limited to long-acting insulin analogues. If NPH insulin would have been included, it would score similar to the analogues on most selection criteria. It would score slightly lower than glargine 100 IU/mL and detemir for safety and tolerability because of the higher incidence of hypoglycaemia.
Nevertheless, it should not be overlooked that ‘good old’ NPH insulin is still a good option, also from a formulary perspective. NPH insulin is likely to be cheaper than biosimilar glargine in most countries.
Conclusions
We found almost identical scores for the long-acting insulin analogues. Therefore, acquisition cost and individual patient characteristics may determine the choice from a formulary perspective. Glargine 100 IU/mL (biosimilar or originator) is the cheapest option in most cases.
Competing interests: None to declare for both authors.
Provenance and peer review: Not commissioned; externally peer reviewed.
Authors
Robert Janknegt, PharmD, PhD, Hospital Pharmacist, Clinical Pharmacologist, Sittard-Geleen, The Netherlands
Marloes Dankers, MSc, Pharmacist, Institute for Rational Use of Medicine, 11 Churchilllaan; NL-3527 GV Utrecht, The Netherlands
References 1 to 188 are available in GaBI Journal print version.
Author for correspondence: Robert Janknegt, PharmD, PhD, Hospital Pharmacist, Clinical Pharmacologist, Sittard-Geleen, The Netherlands
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
Author byline as per print journal: Adjunct Associate Professor Sia Chong Hock1, BSc (Pharm), MSc; Adelia Pheh1, BSc (Pharm) (Hon); Vimal Sachdeva2, MSc; Associate Professor Chan Lai Wah1, BSc (Pharm) (Hon), PhD
Abstract:
Since the onset of the COVID-19 pandemic, there has been a significant surge in interest of COVID-19 vaccines in particular, and other traditional vaccines in general. This strong interest is expected to continue as the industry strives to manufacture safer and more efficacious vaccines against COVID-19 and other infectious diseases. Vaccines are a unique class of products, being biologicals that are administered to healthy individuals to prevent diseases. The equitable distribution and availability of safe, efficacious and good quality vaccines are of utmost importance in preventing and controlling infections and safeguarding public health. The continued existence of poor-quality vaccines suggests a lack of control of manufacturing, storage, distribution, and possibly, their associated regulation. Nonetheless, all these situations – whether positive or negative, present opportunities for improvements. As regulatory authorities step up efforts in regulating existing traditional vaccines, advancements in vaccine research and development churn out novel vaccines that pose further manufacturing and regulatory challenges. This manuscript provides an overview of vaccines, both traditional and novel, and strives to identify challenges in the manufacture, storage, distribution, handling and their associated regulation. It also evaluates whether current regulatory frameworks are adequate, and where applicable, recommends areas for improvements. International harmonization and convergence of national regulatory framework with the view to facilitate quicker approval of safe, efficacious and good quality vaccines, that are accessible and affordable to patients worldwide, are also explored.
Submitted: 11 January 2022; Revised: 11 February 2022; Accepted: 13 February 2022; Published online first: 25 February 2022
Introduction
Each year, millions of people worldwide receive vaccines to protect themselves from infectious diseases that may otherwise be fatal [1]. Today, a majority of infants are subject to specific vaccines as part of post-birth care [2]. From the well-known influenza and chickenpox vaccines with a relatively longer history of use [3] to the more recently developed COVID-19 vaccines [4], there is a myriad of different types of vaccines in use today. Traditionally, vaccines directly mimic a milder form of the infection to stimulate antibody production in the body of a healthy individual. But, beyond this, there are now novel vaccines that leverage on different technologies to improve their stability and efficacy [5].
Despite the large number of vaccines available, newer vaccines are still being introduced to the regional and global markets [6, 7]. A major reason is the presence of many existing and emerging diseases that still lack a proper preventive vaccine [8] alongside the ever-evolving variants of such diseases, as in the case of the COVID-19 virus which has evolved from the alpha, beta, delta and into the omicron variants of concern. Additionally, the formulations of many current vaccines are also being improved to lengthen their short-lived protection or to expand their coverage of the disease, including the emerging strains of viruses. The potential benefits of improved formulations are very promising. However, there had been an overall drop in the number of vaccines introduced globally in the recent years [2] due largely to challenges in ensuring the safety of vaccines [9].
Vaccines are of paramount importance to the control of infections [5] and studies have shown that they can drastically reduce the rate of infections [10, 11]. They confer immunity against specific potentially fatal diseases such as smallpox, diphtheria, tuberculosis, hepatitis, influenza and COVID-19 [12]. In fact, vaccines are the primary prevention method against many diseases [13] and are administered to healthy individuals as prophylaxis. While there are overwhelming evidence that vaccines can prevent diseases and save lives, vaccines are never completely safe [14] and side effects are inevitable [15].
Being biological in nature, the manufacture, storage, distribution and handling of vaccines require strict temperature control to maintain their quality [16]. They are considered cold chain products as vaccines have components that are very sensitive to temperature changes [17]. The safety, quality and efficacy of vaccines may be significantly compromised if they are not handled under appropriate conditions [18] at any stage of their product life cycle, from manufacture to distribution to their administration to individuals. Across the world, regulatory authorities adopt specific national and international standards to assure the safety, quality and efficacy of vaccines. It is of paramount importance that there are adequate regulatory oversight to ensure that vaccines remain safe and efficacious when they are administered to individuals at the point of use [19].
Although many studies have identified various challenges in the formulation of novel vaccines, few have addressed the challenges in the manufacture and quality assurance of such vaccines, post-formulation [20-23], as well as the challenges that are common to both traditional and novel vaccines.
Hence, this manuscript aims to provide a better understanding of the evolution of vaccines and identify the challenges in vaccine manufacture, storage, distribution and their regulation. This manuscript also intends to evaluate whether existing national and international regulatory frameworks for vaccines are sufficient to address these challenges, and to propose improvements.
Classification of vaccines
There are many different types of vaccines, and their key differences may form the basis for the need of specific control measures during manufacturing and regulatory control. Regardless of its classification, a vaccine is scientifically defined as a pharmacological compound for improving immunity to a specific disease [24, 25]. All vaccines have a general mechanism of action where the body recognizes the vaccine’s components (associated with the disease pathogen) as foreign antigens and thereby stimulates antibody production against the specific antigens. Active immunity is acquired [26] as future exposure to the same pathogen would trigger memory cells to begin a chain of signals leading to suppression and removal of the pathogen [27]. Vaccines may be classified according to the types of pathogenic component that it contains [28-30]. Vaccines can also be broadly classified as traditional or novel. Table 1 summarises some common vaccines according to class, type and composition.
Traditional vaccines
Traditional vaccines were the earliest developed vaccines [59]. Generally, they contain whole pathogens or pathogenic subunits [60] which are directly recognized by the body’s immune cells [61, 62]. Whole pathogen vaccines are the oldest vaccines with many studies supporting their efficacy [63-65], and they may be further classified into live attenuated or inactivated vaccines. Live attenuated vaccines contain modified whole bacteria or viruses with decreased virulence, sufficient to induce an immune response but not cause disease. Such vaccines are occasionally unsuitable for immunocompromised patients due to the risk of reversion to its virulent state [66] and in such cases, inactivated vaccines may be used instead. Inactivated vaccines contain whole bacteria or viruses that have been chemically or heat-killed and are hence unable to replicate. One example of such vaccine often used for children is the Inactivated Polio Vaccine (IPV) administered to pre-school children.
Another class of traditional vaccines comprises subunit vaccines which are acellular. The recombinant vaccines consist of bacterial or viral protein fragments as the antigen for immune cell recognition [67]. Bacterial toxins are also used in vaccines, but they are often inactivated to form toxoids that can trigger an immune response without causing disease. Vaccines using such toxoids are known as toxoid vaccines. Studies have shown that some bacterial polysaccharides used in vaccines are more efficacious in inducing an immune response when conjugated to proteins such as diphtheria or tetanus toxoid proteins due to the toxoid’s high affinity for immune cell recognition. Hence, conjugate vaccines such as the Haemophilus-Influenzae type b vaccine, containing polysaccharides conjugated to the tetanus toxoid, have been developed as well.
Apart from toxoids, newer technology allows a non-infectious component of bacterial outer cell membrane known as outer membrane vesicle (OMV) to be used as an antigen in OMV vaccines. An example of a licensed OMV vaccine is the Bexsero® vaccine used in the US against the Type B meningococcal virus which causes meningitis and sepsis [68]. Another newer traditional vaccine is the virus-like particle (VLP) vaccine which contain naturally occurring or chemically synthesised VLPs as the antigen of interest. In fact, VLP vaccines can also be manufactured with different antigens from multiple pathogens incorporated together.
While vaccines can be separated into the above-mentioned classes according to their characteristics, distinct classification is occasionally impractical such as in the case of combination vaccines. One combination vaccine is the Infanrix Hexa® 6-in-1 vaccine used in the UK [69] containing both inactivated viruses and recombinant viral proteins. The single combination traditional vaccine offers protection against six diseases, namely diphtheria, tetanus, pertussis, polio, influenza B and hepatitis B [70].
Novel vaccines
Traditional approaches to conferring immunity may be ineffective for chronic or newer infections that require more specific focus on certain antigens [71]. Hence, new methods of delivering pathogenic antigens have been developed. Novel vaccines are a recent development and they rely on pathogenic nucleic material or other alternative vector delivery systems instead of the specified pathogen [72]. The most well-known novel vaccines are nucleic acid-based vaccines that use genetic material of the pathogen, such as mRNA and DNA, to elicit an immune response [60, 73]. mRNA vaccines contain lipid enveloped mRNA of the pathogen that are ultimately translated by human cells to produce pathogenic proteins that act as immune cell antigens. The Pfizer BioNTech® and Moderna® vaccines against COVID-19 are the two most recent examples. DNA vaccines contain bacterial or viral DNA, which do not require the protection of any lipid membranes due to its higher stability relative to mRNA. The pathogenic DNA undergoes additional transcription to mRNA before embarking on a pathway similar to mRNA [74].
Live attenuated vaccines are the most widely used traditional vaccines due to the better-established balance between their immune effect and safety [61]. They are also relatively long lasting [75]. On the other hand, mRNA vaccines make up the biggest group of novel vaccines to date [76], having undergone the most extensive research and development among the novel vaccines [77, 78]. For simplicity, this review will focus on live attenuated vaccines and mRNA vaccines as examples of traditional and novel vaccines, respectively.
Vaccine manufacture
The manufacture of vaccines is an elaborate process chain involving many well-coordinated steps [79]. Depending on the composition of a vaccine, the complexity of the steps may differ. It is also more complex to manufacture combined vaccines, e.g. MMR vaccine, than single vaccines. Figure 1 shows the different levels of complexity in the manufacture of different types of traditional vaccines.
Depending on the type of traditional vaccines, the complexity and need for additional steps will vary. However, traditional vaccines tend to have relatively less complicated steps compared to novel vaccines which require a more precise coordination of steps. However, there are general manufacturing steps that are common to most vaccines, both traditional and novel, as summarised in Figure 2.
The first step in the manufacture of traditional vaccines is to generate the antigen used to stimulate antibody production. This antigen is specific to each vaccine and will require specific production conditions. Most traditional vaccines require the growth of a pathogen, such as viruses or bacteria, as the antigen. These pathogens are commonly grown in various cell cultures. Eggs and mammalian cells are most commonly used for viruses. The candidate vaccine virus is injected into these eggs/cells that are later incubated to allow virus replication. Some methods use a chemical bioreactor to provide a favourable environment for growth of bacteria. The manufacture of novel vaccines vary significantly as replication of pathogenic genetic material is required instead.
For mRNA or DNA vaccines, replication of pathogenic DNA is the first step of its manufacture. The biosynthesis of DNA begins when plasmids containing specific viral DNA are inserted into bacterial cells such as Escherichia coli (E.coli), and these genetically modified bacteria are allowed to replicate in bioreactors. For mRNA vaccines, additional transcription of the DNA to mRNA is completed using specific enzymes and chemicals. Novel vaccines using viral vectors will require an additional step as the DNA plasmid will need to be inserted into non-pathogenic vector viral cells before replication.
In traditional vaccines manufacture, the resultant antigen has to be isolated from the culture medium and purified. For most vaccines, this begins with the separation of the pathogen from its cell culture medium. From here, each vaccine type will require specific additional steps. Live attenuated vaccines will require reduction in the pathogen’s virulence through multiple sequential cell cultures or chemical processes to decrease virulence [79]. One example is the Bacillus Calmette-Guerin (BCG) vaccine, where live strains of the bacterium, namely, Mycobacterium bovis (M. bovis) [82] are weakened and used to confer immunity against tuberculosis [83]. The pathogens used in inactivated vaccines undergo inactivation by heat [84], chemicals like hydrogen peroxide [85] or ultraviolet (UV) radiation [86]. Subunit vaccines will require physical disruption of whole pathogens to obtain the specific antigens needed, such as viral proteins or bacterial toxins.
Thereafter, the isolated antigens undergo multiple purification steps, including filtration, chromatography, clarification and concentration [87]. Simple purification methods exploit particle-size difference, where hollow fibres or flat screens are used to filter out antigens of a specific size. The liquid containing the antigens can be flushed in a direction parallel to the filter, known as tangential flow filtration, to ensure continuous filtration and better recovery of the antigens. When the size difference between contaminants and antigens are significantly less distinct, high affinity chromatography is a common method of purification [88]. The antigens and other components are separated based on their ionic charges or hydrophobic interactions instead. For nucleic acid-based novel vaccines, DNA and mRNA have to be isolated and purified.
After purification, the next step is to formulate the vaccine by incorporating the relevant components [89]. Antigens may be combined with an adjuvant to intensify the immune response triggered [90]. Stabilizers, such as surfactants [91], are added to extend the shelf life of vaccines. Some multi-dose vaccine formulations include preservatives to prevent unwanted microbial contamination [92]. All the above-mentioned steps are normally carried out in a segregated cleanroom of the manufacturing facility as an aseptic environment is required to prevent unwanted microbial contamination.
Finally, the manufactured vaccines are filled into sterile depyrogenated vials in an aseptic environment. The freshly manufactured vaccines are sealed with sterile stoppers together with an outer cap to enhance the physical protection against contamination. After filling, the vials are clearly labelled. Upon completion, the sealed, labelled vaccine vials will undergo strict testing and inspection using specialised equipment to ensure container-closure integrity and to eliminate any defects that may compromise the vaccine’s quality. Throughout the entire manufacturing process, the raw materials and products are to be kept strictly at their respective optimal temperatures.
Vaccine storage, transport and distribution
Once manufactured, the vaccines are stored within the manufacturing facility at their recommended temperature until they are ready for distribution. The storage, transport and distribution of vaccines are constantly managed under temperature-regulated environments. This is because high temperature can cause denaturation of the vaccine antigen and adjuvants. Hence, post-manufacture handling of vaccines involves a collective and continuous monitoring programme known as the vaccine cold-chain management [93] as all vaccines are cold chain products (CCP). The management of CCPs require high quality temperature control within a stringent temperature range, commonly at 2°C‒8°C for most traditional vaccines [94]. Some traditional vaccines, such as the hepatitis B vaccine and diphtheria vaccine, are prone to freeze damage [95, 96]. Under freezing temperature, the vaccines experience potency loss. Administration of such freeze-damaged vaccines can result in an increased risk of adverse effects such as sterile abscesses [97]. On the other hand, novel vaccines, especially nucleic acid-based vaccines, need to be stored at sub-zero temperatures in order to maintain their potency. The mRNA or DNA in such vaccines are highly susceptible to enzymatic damage and hence ultra-low temperatures are necessary to minimize enzymatic activity and any genetic material damage. Therefore, vaccines have to be kept within the appropriate temperature ranges that are specific to the individual vaccines. Table 2 summarises some optimal temperatures of the different types of vaccines.
The actual storage temperature of a vaccine will vary as it travels from the manufacturing facility to the destination country and vaccination centre. The Pfizer-BioNTech COVID-19 vaccine may be stored long term in ultracold freezers at a temperature of between -90°C and -60°C. The vaccines arrive at the warehouses of destination countries in thermal shippers packed with dry ice, see Figure 3. These thermal shippers maintain the vaccines at the required ultra-cold temperature of between -90°C and -60°C. At the destination country, the temperature monitoring device in the thermal shipper is checked to assure that there are no temperature excursions during transportation.
The vaccines are transferred to higher temperatures in a step-wise manner before use. Before mixing with the diluent for administration, the vaccine may be stored in a pharmaceutical refrigerator between 2°C and 8°C for up to 1 month (31 days). Upon mixing with the diluent (sterile saline), the reconstituted vaccines can be left at room temperature (2°C to 25°C) for up to 6 hours. No refreezing of such vaccines is allowed after reconstitution. Any remaining unused vaccine must be discarded after 6 hours [98, 99].
Regulatory controls
The inherent need for scientifically sound vaccine regulation is acknowledged by regulatory authorities, both globally and nationally [100]. All countries should have an organization that is legally responsible for vaccine regulatory actions. Vaccines are a distinct class of biological products which need to be subject to specific regulations due to their unique characteristics. Since there is currently no single biopharmaceutical products classification system that clearly defines vaccines and their respective scope of regulation [101], national regulatory authorities (NRAs) across different nations have set their own regulations. Table 3 summarises some vaccine-producing countries and the names of their NRAs.
Apart from NRAs, there are other international organizations (IO) that aid in the harmonization of vaccine regulation. Although not defined as a regulatory agency (RA), these IOs are important in vaccine control as they act as a benchmark organization for regulation by setting standards recognized by most countries. These IOs set guidelines which form the basis of regulations enforced by the NRAs. The World Health Organization (WHO) is one IO that is a key player in vaccine regulation by its provision of harmonized standards for NRAs. While different NRAs have slightly varying standards, most vaccines are regulated similarly in accordance with WHO or internationally standardized guidelines [103].
Regulatory control of vaccines begins with the development of the vaccine where multiple clinical trials are required before it is eventually licensed for manufacture. This review focuses on post-licensure regulations relating to vaccines manufacture and post-manufacture handling.
All procedures involving vaccine handling are bound by specific regulatory requirements, and details set by the respective NRAs. Firstly, vaccine manufacturers require a licence to operate, regardless of their country of origin [104]. The licence is granted under the condition that a set of standard manufacturing procedures is established and only this approved set of procedures is permitted at the specific manufacturing facility. In the US, vaccine manufacturers are also required to have a functional department reporting any proposed changes to the US Food and Drug Administration (FDA), Centre for Biologics Evaluation and Research (CBER). Strictly no digression of the standard procedures, raw materials or equipment will be condoned by FDA until it is approved by CBER [105]. Vaccine manufacturers are also required to complete and produce all necessary documentations for inspection at all times [105]. FDA stipulates all manufacturing information and documentations required for the Biologics License Application (BLA) [106].
In general, vaccine manufacture is strictly required to be performed in cleanrooms, that are specifically designed to allow for sterile manufacture of products in accordance with good manufacturing practices (GMP) guidelines [107]. These GMP guidelines have been prepared by WHO and other NRAs and IOs, e.g. US FDA, EMA and PIC/S, and they specify precise measures required to ensure the manufacture of safe and good-quality vaccines. A specific portion of the guidelines is used internationally as a reference for individual countries to set their national GMP requirements in vaccine manufacturing facilities [108]. This means that the cleanrooms have to maintain a certain GMP grade before they can operate. Figure 4 shows some examples of cleanroom grades for the different stages of vaccine manufacturing.
Additionally, the US Centres for Disease Control and Prevention (CDC) will assign the biosafety levels (BSLs) to the vaccine-related facilities after assessment of the level of precautionary measures required. Different vaccine types require different BSLs for its cleanrooms and related facilities. For example, some cell-culture-based Influenza vaccines are assigned BSL 2 due to its large-scale open nature, while other Influenza vaccines are assigned BSL 4 due to the virus’ highly virulent nature [110]. According to the BSL assigned, the amount of safety controls implemented will differ, such as the compulsory use of different personal protective equipment (PPE) or specific training required. Table 4 shows a summary of the different BSL and the respective considerations required.
Airlocks and airflow hoods are necessary to ensure unidirectional air flow and to maintain sterility of the environment. The walls of the facility have to be specially designed and environmental monitoring is mandatory. The facility also needs to be kept at the optimal temperature for manufacture by employing heating, ventilation and air conditioning (HVAC) systems [112]. The workers in the manufacturing facility are also regulated. They need to be dressed in the appropriate PPE and undergo necessary training. The complex nature of vaccine processing and handling necessitate timely inspections of the vaccine facilities and its procedures based on WHO or other international GMP standards.
Additionally, the International Organization for Standardization (ISO) has developed a harmonized standard, namely the ISO classification for cleanrooms and controlled areas, as a standardization of quality assurance across industries, including health care. It is widely used in many countries including the US and European Union (EU) where vaccine cleanrooms are subjected to ISO classifications [113] according to its particulate content as shown in Table 5. Each cleanroom used for handling vaccines has a specific ISO class which determines its respective controls. There are 9 ISO classes according to the particulate level in the air. In cleanrooms, classes 5 to 8 are the most commonly required and their characteristics are summarised in Table 5. Lower ISO classes have more stringent requirements. For example, the areas have a lower maximum concentration of particulates and hence require higher rates of airflow to maintain the air quality [115].
Challenges, safety and quality issues and possible solutions
Despite stringent regulation and post-market surveillance of the vaccine industry, challenges in the manufacture of safe, efficacious and good quality vaccines still prevail.
Currently, there is no single harmonized regulatory system that defines the standards for manufacture, storage and distribution of vaccines across the world, resulting in subjectivity of controls [101]. Scientifically, vaccines have been defined to include biological preparations which are administered to confer immunity against specific diseases [24, 25]. Without a standardized universal definition of vaccines, there are differences in how regulatory authorities and manufacturers emphasize control on vaccines. For instance, RAs tend to focus on safe vaccines and need for vaccines to contain adjuvants that would ensure its quality [116-119]. On the other hand, manufacturers may tend to emphasize on the overall efficacy of vaccines in reducing disease rate or severity [120]. This variation in emphasis may pose complications and challenges in the regulations and manufacture of vaccines.
The lack of a standardised or internationally harmonized framework can lead to differing frameworks and therefore variations in terms of regulatory control. For example, currently, the two dominant regulatory frameworks are the EU GMP Guidance Annex 1: Manufacturing of Sterile Medicinal Products [121], and the US FDA Guidance for Industry: Sterile Drug Products Produced by Aseptic Processing Current Good Manufacturing Practice [122]. Some key differences in the US and EU regulations are summarised in Table 6.
In comparison with US regulations, the EU GMP Guidance has a larger and more stringent scope for cleanroom requirements [124]. Additionally, the EU has set different requirements for cleanrooms during and after operation, respectively. On the other hand, the US FDA has no specifications for cleanroom particulate levels when manufacturing processes are not ongoing. While the EU has specified that cleanroom requirements apply to both aseptic processes and terminal sterilization, the US FDA Guidance specifies only aseptic processes, and does not mention terminal sterilization. Although Annex 1 of the EU GMP Guidance has recently been revised in 2020 to widen its scope [125], the harmonization of standards between the two major jurisdictions remains a more desirable solution.
In 2021, a US vaccine manufacturing plant by the name of Emergent BioSolutions (EBS) had its production operations suspended due to contamination of its vaccine products. The single facility was used concurrently for the manufacture of two different COVID-19 vaccines, one by Johnson & Johnson (JNJ) and the other by AstraZeneca (AZ), leading to a mix-up of distinct starting materials required for each vaccine [126]. JNJ is a company based in the US while AZ is a company based in Belgium; however, both vaccines are manufactured in the same US facility, namely EBS. Although both vaccines involved are novel vaccines using similar viral vectors, the respective vectors used were non-identical and incompatible. It does not help that there are differing cross-jurisdictional regulatory requirements between the US and Belgium.
Additionally, it was discovered that the facility at EBS had previously been found to have a substandard documentation of procedures and training of staff involved [127]. While existing regulations are in place, subsequent review of these regulations and their enforcement are equally crucial. This unfortunate event also points to a need for global convergence and harmonization of international standards for vaccine manufacture and regulation.
Currently, many developing countries still lack a functioning framework for regulating vaccines. In fact, as of 2020, a significant 73% of WHO Member States do not have a mature system for optimal regulation of vaccines [128]. Developing countries also face additional challenges in maintaining the quality of vaccines due to the lack of funding and resources. To be approved as a functional NRA by WHO, the regulatory body must be able to perform regulatory actions such as assuring standards for vaccine licensure and conducting regular inspections of facilities, at least a maturity level of 3 and above [103]. This includes having a national laboratory solely for testing and evaluating the efficacy of vaccine in the country. This poses a challenge to developing countries which are already experiencing a strain on their overall regulatory resources.
With the COVID-19 pandemic driving the need for safer and more efficacious vaccines, new types of vaccines are expected to emerge in the near future. With this evolution, there are also challenges that are bound to arise in the manufacture and quality assurance of both traditional and novel vaccines. The manufacture of traditional whole vaccines is labour and time intensive, which poses the risk of pathogenic shift or drift as the vaccines undergo manufacture. Also, subunit vaccines face the challenge of thorough purification as they contain antigens of relatively smaller sizes. This can limit the degree of purification possible and make it harder for the manufacture of safe vaccines.
Likewise, mRNA vaccines face specific challenges to their novelty. Since most novel vaccines are relatively new, there is still lack of optimization at many stages of their manufacture, which may compromise the quality of vaccines manufactured. The complexity of mRNA vaccines also adds to the challenge of requiring more intricate quality assurance systems that are able to assure the vaccine’s quality at every stage of manufacture [80].
The recent rise in adverse events globally due to the use of poor-quality vaccines suggests the possible inadequacy of current regulatory frameworks and presents opportunity for refinement. In 2013, a batch of Gardasil HPV vaccine was recalled due to contamination with the vaccine glass shards [129], suggesting poor GMP compliance and inadequate enforcement of regulations.
WHO has shown much effort in harmonizing regulatory frameworks with regular review of regulatory guidelines. However, international harmonization of vaccine regulations is the way forward as this would allow cross-border use of all vaccines approved according to such an internationally harmonized regulatory framework for vaccines.
Conclusion
Challenges in the manufacture, storage, distribution and supply chain management, and associated regulation of vaccines are expected to continue. Vaccines have become a crucial weapon in the global war against pandemics. Our reliance on vaccines during the COVID-19 pandemic has clearly illustrated the critical importance of vaccines in the war against this elusive infection. The COVID-19 pandemic has also highlighted the importance of both the industry and the authority to work closely together to assure that safe, good quality and efficacious products are available at vaccination centres and points of use that are located at each and every nook and corner of the world. In a more positive light, the COVID-19 pandemic has presented opportunities for collaboration amongst NRAs, IOs and the industry in vaccine manufacture, storage, distribution, handling, regulation and international convergence of standards. Globally, NRAs should strive towards an internationally harmonized regulatory framework that will facilitate the approval and use of vaccines, whether traditional or novel, across national borders.
Competing interests: None
Provenance and peer review: Not commissioned; externally peer reviewed.
Authors
Adjunct Associate Professor Sia Chong Hocka 1, BSc (Pharm), MSc Adelia Pheha 1, BSc (Pharm) (Hons) Vimal Sachdeva2, MSc Associate Professor Chan Lai Wah1, BSc (Pharm) (Hons), PhD
1Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543 2Technical Officer (Senior GMP Inspector), WHO/HQ/MHP/ RPQ/PQT/INS, 20 Avenue Appia, Geneva CH-1211, Switzerland
Author for correspondence: Adjunct Associate Professor Sia Chong Hock, BSc (Pharm), MSc, Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
Author byline as per print journal:Mihaela Buda, PhD; Olga Kolaj-Robin, PhD; Emmanuelle Charton, PhD
Abstract:
The approach to the elaboration of European Pharmacopoeia (Ph. Eur.) monographs in the field of biotherapeutics has significantly evolved in recent years. In particular, monographs on complex biotherapeutics call for greater flexibility as a means of better addressing the structural complexity and naturally occurring heterogeneity of large biologicals, and of facilitating the use of novel technologies. This manuscript provides an overview of strategies to overcome the challenges of elaborating monographs on biotherapeutics, based on discussions with Ph. Eur. stakeholders. It describes the science-based approach to building flexibility into a public standard, discusses aspects related to the monograph lifecycle and underlines the paramount importance of stakeholder participation in the work of the Ph. Eur.
Submitted: 7 January 2022; Revised: 26 January 2022; Accepted: 31 January 2022; Published online first: 14 February 2022
Introduction
The monographs and associated physical reference standards of the European Pharmacopoeia (Ph. Eur.) are legally binding public standards in the Member States of the European Pharmacopoeia Convention and thus play a major role in ensuring the quality of medicines in Europe – and beyond [1]. While Ph. Eur. monographs on biotherapeutics (covering active substances as well as medicinal products) have existed for nearly three decades, their elaboration has faced challenges in recent years. The difficulties encountered relate mostly to the advent of biosimilars and a misunderstanding of the role of monographs in this context. These challenges were discussed at the international conference entitled European Pharmacopoeia: tackling future challenges of the quality of medicines together in September 2016 in Tallinn, Estonia (hereinafter ‘the Tallinn conference’). Following the event, ways were proposed to identify and elucidate these challenges and determine how they can be overcome [2]. The dialogue with stakeholders on the development of public standards for biotherapeutics continued in June 2019 at the dedicated workshop of the International Conference entitled EDQM & European Pharmacopoeia: state-of-the-art science for tomorrow’s medicines held to mark the publication of the 10th Edition of the Ph. Eur. (hereinafter ‘the Strasbourg conference’) [3]. On the eve of the launch of the 11th Edition of the Ph. Eur. in 2022, this manuscript provides an overview of the key points that emerged from the discussions held with Ph. Eur. users, such as representatives from innovator and biosimilar companies, National Control Laboratories and licensing authorities. It also offers a status update on Ph. Eur. monographs in the field of biotherapeutics.
Pharmacopoeial standards for biotherapeutics
EDQM perspective – case studies Previous discussions revealed that opposition to individual monographs was often a result of their misuse and misinterpretation, for example, their use to assess biosimilarity outside Europe. The European Directorate for the Quality of Medicines & HealthCare (EDQM) has taken action – including wide and regular communication in international conferences, providing a dedicated webpage on the EDQM website [4] and publishing scientific articles [2, 5] – to help ensure that the role played by monographs in defining quality standards for biotherapeutics is correctly understood. Another important milestone in this action plan has been the publication of a revised version of the Technical guide for the elaboration of monographs on synthetic peptides and recombinant DNA proteins[6] with the introduction of a new section on flexibility, see Figure 1 and section 3 for more details. The biotherapeutics covered by monographs of the Ph. Eur. range in size from a 30 amino acid polypeptide (teriparatide) to a 145 kDa glycoprotein (infliximab), see Figure 2. The contents of the monographs for these substances vary significantly, with monographs covering large, complex structures having built-in flexibility in addition to that already allowed by the Ph. Eur. The monograph elaboration or revision process relies on stakeholder participation; the approach is dynamic, collaborative and transparent, with the involvement of interested parties actively sought and encouraged from the early stages of the process. The consultation phase (publication in Pharmeuropa), during which manufacturers should perform testing to verify their substances against the new or revised standard, is of key importance. As compliance with the Ph. Eur. monograph is mandatory, interested parties must be able to resolve any issues before the implementation date of the text: working closely with the EDQM at an early stage and up to the public consultation phase is therefore strongly encouraged. This topic is further explained in section 4.
In sum, Ph. Eur. monographs provide a common framework for setting the quality and for standardisation of the medicinal products available on the market and under development and thus help ensure a better understanding of potential drift and evolution in product quality. This helps ensure that consistently high product quality is maintained.
Ph. Eur. monographs prescribe the use of reference standards, e.g. biological reference preparation (BRP) and chemical reference substance (CRS): these contribute to defining the quality requirements. The first Ph. Eur. BRP established by the EDQM for a monoclonal antibody, Infliximab BRP for cell-based assay calibration, was the result of a collaborative study involving laboratories from regulatory authorities, official medicine control laboratories and the pharmaceutical sector [7]. The study also shed light on critical parameters and possible contributors to assay variability and helped ensure that the level of method detail and assay conditions were suitably reflected in the monograph. Consequently, the Infliximab concentrated solution (2928) monograph [8] prescribes the use of a ‘suitable cell-based assay based on the inhibitory action of infliximab on the biological activity of TNF-alpha’. The specific assay used in the collaborative study measures the cytotoxic effect of antitumour necrosis factor-alpha (TNF-a) on WEHI-164 cells, and is provided as an example procedure (described in detail) in the monograph, using Infliximab BRP as a standard. It is important to note that Ph. Eur. BRPs are qualified for the use(s) described in Ph. Eur. monographs; they are established as a result of international collaborative studies coordinated through the Biological Standardisation Programme and the full reports are published in Pharmeuropa Bio & Scientific Notes (https://pharmeuropa.edqm.eu/app/BioSN/search/), thereby giving users access to important information on assay procedure, reagents, reference standards and products [9, 10]. There have been significant developments in Ph. Eur. reference standards established to support the physicochemical testing of recombinant glycoproteins in recent years. Reference standards for glycan analysis may serve two functions: 1) to support controlling the performance of a multi-step procedure, and thus serve the purpose of system suitability testing, described in monographs; 2) to assess compliance with both qualitative and quantitative acceptance criteria, therefore acting as a quality benchmark. These functions can be illustrated using two recently established Ph. Eur. CRSs, Infliximab CRS and Etanercept CRS, both of which have been used to support analytical procedure transfer and independent testing.
For the assessors present at the Strasbourg conference, among the challenges identified during monograph elaboration is the question of maturity: is a substance/subject sufficiently well developed for standardisation? A pragmatic approach to the issue would be appropriate: each case will be different and will need to be addressed based on experience. In their views, a monograph should not necessarily be ‘exhaustive’, but should aim at comprising a number of key quality attributes that are both critical and amenable to standardisation. The identification of appropriate, key quality attributes to be covered by the future standard is seen as a major challenge. The recently published monographs for complex recombinant proteins such as etanercept and infliximab demonstrate that this exercise is possible.
On a different topic, from an innovator standpoint, a biopharmaceutical manufacturer must be able to absorb a substantial and challenging workload following implementation of a Ph. Eur. monograph. The following question was raised: can a biotherapeutic product monograph sufficiently describe acceptable quality for market use and be a reliable predictive model of this acceptable quality? From an industry perspective, a monograph can only fulfil this role if it goes through multiple revision cycles, but the regulatory burden these generate would be challenging. In the view of industry representatives present at the Strasbourg conference, a forward-looking strategy should be developed at the level of the Ph. Eur. to move from a ‘product-specific’ toward a ‘modular adaptative’ approach involving class- and performance-based monographs and general texts. In the opinion of biosimilar manufacturers present at the Strasbourg conference, monographs should champion high quality products and analytical procedures but provide sufficient flexibility to address complexity of large biologicals and to facilitate the use of novel technologies.
The following sections describe the approaches taken by the EDQM to address these challenges.
Evolution of monographs: flexibility and approaches to standardisation
The number of Ph. Eur. public standards for biotherapeutics has grown continuously over the last three decades and now covers a broad selection of substances ranging from peptide hormones, interferons and interleukins to growth factors, blood coagulation factors and monoclonal antibodies. In an increasingly evolving multi-product market, the development and evolution of monographs for highly complex biotherapeutics – where the process defines the product – has faced significant challenges. Unlike many other proteins, glycoproteins (including monoclonal antibodies) have complex structures and intrinsic heterogeneities; they display a large diversity of quality attributes, which put high demands on the techniques required for their analysis, far beyond traditional physicochemical procedures. These aspects make it impossible to fully standardise complex biotherapeutics and raise the fundamental question of how to reflect key quality attributes (QAs) and associated testing strategies, and how to establish suitable common expectations in a monograph.
Work on monographs for complex biotherapeutics has shown that additional flexibility (other than that defined in the Ph. Eur. General Notices) is needed to address the structural complexity and process-dependent product heterogeneity of these substances, the complexity and specifics of (often multi-step) procedures for analysis, as well as the potential diversity of the product resulting from different manufacturing processes. A key challenge is how to build this flexibility into a public standard, so that it still provides sufficiently prescriptive requirements (including tools to support analytical procedure control strategy and facilitate successful independent testing, and defined acceptance criteria for key QAs to enable standardisation of functionality), while remaining compatible with the development of follow-on versions.
As a result, the following elements of additional flexibility have been built into monographs for complex biotherapeutic proteins, pioneering a unique approach across pharmacopoeias:
requirements related to process-dependent heterogeneity, e.g. glycosylation, charged variants, are included in the Production section of the monograph, and set in a flexible way
acceptance criteria for QAs are set on the basis of a balance struck between criteria given as figures and ‘as authorised by the competent authority’. Monograph acceptance criteria cannot be expressed numerically for process-dependent QAs, as no ‘one-size-fits-all’ criterion is possible, e.g. for glycosylation or charged variants. Conversely, specific activity and size variants, e.g. aggregates and fragments by SEC, related proteins by CE, are seen as appropriate candidates for standardization, for which defined numerical limits are appropriate
two types of reference preparations (the so-called ‘dual standards’) are used in tests related to process-dependent heterogeneity, e.g. glycan analysis, charged variants determination, namely a Ph. Eur. reference standard for system suitability testing and an in-house reference preparation, shown to be representative of batches tested clinically and batches used to demonstrate consistency of production, for compliance testing
in the case of complex or multi-step procedures (glycan analysis and cell-based assays, for example), an outline of a suitable procedure – if applicable, a generic method for analysis described in Ph. Eur. general chapters – is typically given as a requirement, while a detailed analytical procedure, including system suitability criteria and dedicated reference standard, may be given as an example. This means that the procedure described may be implemented as is or may be replaced by another suitable, validated procedure (without having to demonstrate its equivalence to the ‘example’ procedure), subject to approval by the competent authority.
In combination, these elements provide a means of enhancing monograph flexibility under well-defined conditions, an approach that addresses structural complexity of biotherapeutics and is compatible with development of biosimilars. This approach has been established based on extensive input from the Ph. Eur. experts and the EDQM and relies on scientific and experimental evidence generated through dedicated laboratory studies. Combined with close collaboration with industry parties, these factors have proved essential in finding the best way forward for public standard setting. This concept was also described in detail in the Technical guide for the elaboration of monographs on synthetic peptides and recombinant DNA proteins[6], which was revised in 2018 to take into account these developments.
At the Strasbourg conference, the huge progress in the way monographs address the glycosylation issue was acknowledged and the participants commended both the flexibility offered by the Ph. Eur. and the communication surrounding it. However, this approach was not created overnight – monographs on biotherapeutics are not new: the first texts were published nearly three decades ago and new texts continue to be developed, even for more complex molecules such as monoclonal antibodies. Development of the latest monographs for etanercept or infliximab illustrated the fact that there is an added value for certain standardisation activities subsequent to development of individual monographs: this ‘bottom-up approach’ has allowed the Ph. Eur. to base its approaches and policies on facts and put an end to earlier sterile theoretical debates about whether or not monographs can cover large molecules. Much was learned from these individual cases and it became apparent that monographs should not be expected to cover all quality attributes that are used to characterize the product and should focus on issues that can be standardised. This specific idea is the starting point for standardisation of a particular quality attribute in a multi-product setting. The potency assay is an illustrative example: the bioassay standardisation efforts and establishment of Ph. Eur. BRPs for etanercept and infliximab potency assays enhanced our understanding of the anti-TNF-a neutralization assay that could serve as a multi-product procedure for monoclonal antibodies, exercising the same function. Most importantly, it helped establish a robust assay procedure, with confirmed applicability and transferability, that became part of the infliximab monograph. On the one hand, both regulators and representatives of Official Medicines Control Laboratories considered this to be a very important feature: the monograph sets up the minimum requirements and allows alignment with current expectations. It is important to have a common public standard for all products and robust test procedures. However, on the other hand, some industry representatives claim that individual monographs are underutilized. The point was well-taken and the example described above shows that the knowledge gained in the elaboration of individual monographs can be the basis for future standardisation of general matters and, by extension, to explore flexible concepts of standardisation.
The same bottom-up approach for biological assays could be applied to classes other than TNF-a antagonists, of different complexities, e.g. ErbB/HER, CD20, vascular endothelial growth factor (VEGF) antagonists. As a first step, specific monoclonal antibodies of a given class are to be considered individually in view of their specific quality attributes related to the mechanism of action, e.g. Fc-effector functions; the outcome of such exercise would help better understand how to address specific challenges to bioassay standardisation within a class of monoclonal antibodies, and thus drive forward the development of class-based, general Ph. Eur. texts. Further reflection on the application of this model to the development of monographs for trastuzumab or rituximab, for example, is necessary.
Monograph lifecycle: implementation and revision
Consultation of stakeholders Ph. Eur. texts are public standards legally binding in the 39 states parties to the Convention on the Elaboration of a European Pharmacopoeia and applied in more than 120 countries worldwide.
They become mandatory at the implementation date, six months after their publication. Therefore, users have six months to take appropriate measures to assure compliance of the substances or products covered by the Ph. Eur. Seen from the outside, this period may seem to be short, but it should be noted that all the texts published in the Ph. Eur., whether monographs or general chapters, whether revised or newly elaborated, undergo publication consultation before their official publication and implementation. The forum used for consultation, Pharmeuropa, is freely available online (https://www.edqm.eu/en/How-to-register-for-Pharmeuropa-Online-1457.html). It is essential that the users of the Ph. Eur. examine closely the texts published in Pharmeuropa; they are also strongly encouraged to carry out experimental verification of the described analytical procedures on their own products, after which they might see the need to comment on the texts. Commenting is done either through their national pharmacopoeia authority if they belong to one of the 39 signatory countries to the Ph. Eur. Convention, or directly to the EDQM if they are commenting from outside Europe. The comments are carefully reviewed by the experts of the Ph. Eur., who then – when necessary – take appropriate action. It is important to note that comments submitted after official publication cannot be addressed. Or rather, they will be addressed in a subsequent round of revision.
In this context, the importance of the public consultation period must not be underestimated: the worst-case scenario would be for a user to discover that their product fails to meet all the requirements of the newly implemented text. It can also happen that a user will only perform a paper evaluation of the pharmacopoeia text during the public enquiry, comparing the elements of the new analytical procedure against the in-house procedure without trying out the new procedure experimentally.
During past events, notably at the Strasbourg conference, industry representatives commented that the main issue with biotherapeutic monographs was not the monographs themselves, but the timing at which comments were requested: for example, the commenting period might coincide with the final phase of a dossier submission, or because of the time needed to approach competent authorities, the slot for commenting on a monograph might be missed. The EDQM also received comments from stakeholders who had not yet received a marketing authorization. This grey area during which a product is still the subject of confidential discussions between industry and authorities prevented any contribution to pharmacopoeia efforts.
Implementation of Ph. Eur. monographs Implementation of monographs has an impact on already approved products. During the Strasbourg conference, some industry representatives expressed concerns about the regulatory burden linked with the implementation of new or revised pharmacopoeia texts, especially when a text is revised repeatedly over the years. The Ph. Eur. is keen to reduce the number of revisions rounds for a given text, but this is only possible if contributions from all stakeholders are submitted within the given timelines (for example, public enquiries). If marketing authorization holders reacted too late, then yes, multiple revisions are inevitable.
As regards regulatory updates linked with revisions of Ph. Eur. texts, the EU has introduced the following statement in its guideline on how to deal with variations: ‘There is no need to notify the competent authorities of an updated monograph of the European Pharmacopoeia or a national pharmacopoeia of a Member State in the case that compliance with the updated monograph is implemented within six months of its publication and reference is made to the ‘current edition’ in the dossier of an authorised medicinal product’ [11]. This implies that, in the event that the user does not need to notify the authority, they must ensure compliance with the text. To implement a new or revised text, the user may choose between two options: 1) to implement the text as published; 2) to keep their in-house procedure; in the latter case, to assure compliance with the Ph. Eur., the method will be considered as an alternative procedure, as defined in the General Notices and the user would have to be able to demonstrate, to the satisfaction of the regulatory authority, that the substance or product would comply with the Ph. Eur. if tested. This will most probably require experimental testing and undertaking the necessary regulatory steps with the competent authorities for the respective changes. Industry representatives reported that such changes might be easy to handle in Europe, but 3–5 years may be required to complete a change in test procedure worldwide – a huge burden. At the Strasbourg conference, regulators confirmed that there is regulatory obligation to update the dossier: ‘your analytics have to be up to date’. However, the difficulty of implementation worldwide was unanimously recognized. Reference was made to ICH Q12, which strives for a globally harmonized approach to technical and regulatory considerations. A plea was made for a regulatory pathway that would facilitate the process for biotherapeutic monograph implementation. For a biosimilar manufacturer, life-cycle management is particularly difficult since the appearance of a new standard may call for a change in the biosimilar development.
Stakeholder participation in the work of the Ph. Eur.
One of the main recommendations that emerged from the Tallinn conference was to enhance communication with stakeholders on the activities of the Ph. Eur. Commission in the field of biotherapeutics. Enhanced communication was also viewed necessary to emphasize that, although compliance with the Ph. Eur. monographs is required, it must not be considered as sufficient for proving biosimilarity. Moreover, further feedback from stakeholders on the concept of additional flexibility in monographs for biotherapeutics as proposed in the Infliximab concentrated solution (2928) monograph [8], published at that time for consultation in Pharmeuropa, was to be gathered.
To address these requests, two articles were published shortly after the Tallinn conference, one on the role of Ph. Eur. monographs in setting quality standards for biotherapeutics [2] and another on the process of elaboration of monographs for biotherapeutics using substances and medicinal products from a single source, together with the lessons learnt during this exercise [5]. In addition, in early 2017, the EDQM and the European Medicines Agency (EMA) organized a joint workshop on biosimilars [12]. It was broadcast live to increase its outreach and was highly appreciated by the participants. Two further Ph. Eur. training sessions on biologicals with modules and workshops dedicated to biotherapeutics were also organized in 2017 and 2020 [13, 14]. Finally, a page devoted exclusively to biotherapeutics was created on the EDQM website [4]. In addition to procedural explanations, it gathers all the related news, articles and information on events as well as the regularly updated Ph. Eur. biotherapeutics monograph portfolio.
The participants at the Strasbourg conference were unanimous: the EDQM had done a good job documenting the challenges and answering them. The actions devoted to communication with its stakeholders had contributed to focusing the debate on real, concrete issues. The EDQM has created a safe space for Ph. Eur. experts to debate and discuss different views. The examples given in the presentations shared at the Strasbourg conference support these findings [3].
Achievements and perspectives for the future
The work of the recent years and its many positive outcomes have demonstrated that elaboration of monographs on complex biotherapeutics is not only possible but also very useful. In 2016, with the adoption of Etanercept (2895) and Human coagulation factor IX (rDNA) powder for solution for injection (2994), the Ph. Eur. Commission concluded the so-called P4-Bio pilot phase, thus establishing a routine framework for setting public standards for biotherapeutics using substances and medicinal products from a single source and working in close collaboration with an innovator company. A year later, the adoption of the first monograph on a monoclonal antibody, Infliximab concentrated solution (2928), marked a significant milestone in setting standards for complex biotherapeutics.
Recently, Erythropoietin concentrated solution (1316), the first monograph on a complex glycosylated molecule, first published in Supplement 1999 to the 3rd Edition of the Ph. Eur. in 1999, was revised following this approach (Supplement 10.4).
In addition to regular revision of monographs on biotherapeutics, e.g. Insulin glargine (2571) in Supplement 9.5 or Erythropoietin concentrated solution (1316) in Supplement 9.6, a new monograph on Filgrastim injection (2848), elaborated using the multisource procedure, was published in Supplement 9.8. The commitment of the Ph. Eur. in the field of biotherapeutics is also reflected in the number of items recently added to the work programme. Elaboration of several new monographs via the single source procedure (Golimumab concentrated solution (3103) and Ustekinumab (3165)) as well as via the multisource procedure (Insulin glargine injection (3129), Teriparatide injection (3130), Human coagulation factor VIII (rDNA) powder for injection (3106) and Human coagulation factor VIII (rDNA), B-domain-deleted, powder for injection (3108)) is currently ongoing. Moreover, another individual monograph on a monoclonal antibody, Adalimumab concentrated solution (3147), is being prepared within the framework of the ongoing monoclonal antibodies (mAb) pilot phase. Finally, the mAb WP is also working on the development of horizontal standards applicable to monoclonal antibodies. Three general chapters on methodologies for potency determination for anti-TNF-a antagonists (Cell-based assays for potency determination of TNF-alpha antagonists (2.7.26)) and on physicochemical testing applicable to various monoclonal antibodies (Capillary isoelectric focusing for recombinant therapeutic monoclonal antibodies (2.5.44), Size-exclusion chromatography for recombinant therapeutic monoclonal antibodies (2.5.43)) are currently in preparation, see Figure 3.
The experts of the Ph. Eur. present at the Strasbourg conference considered that the monograph development exercise enhances product understanding. Monographs based on a single product bring a lot of insight; a monograph based on many products helps build a bigger picture, a robust standard and thus a foundation for the rationalization of test procedures in a multi-product setting. Overall, it is beneficial for the field to work on and add to this understanding together. The resulting standards are more robust and of greatest benefit to patients. The Ph. Eur. portfolio of quality requirements for biotherapeutics will continue to rely on the experience gathered from product-specific cases and use it as a basis for driving general, transversal matters. A number of studies are underway, and the Ph. Eur. looks forward to making further progress in the field in the years to come.
Competing interests: Mihaela Buda, Olga Kolaj-Robin and Emmanuelle Charton are employees of the EDMQ, Council of Europe. All authors declare that they have no conflicts of interest that might be relevant to the contents of this manuscript.
Provenance and peer review: Not commissioned; externally peer reviewed.
Authors
Mihaela Buda, PhD Olga Kolaj-Robin, PhD Emmanuelle Charton, PhD
European Pharmacopoeia Department, European Directorate for the Quality of Medicines & HealthCare (EDQM), Council of Europe, Strasbourg, France
References 1. Council of Europe. European Directorate for the Quality of Medicines & HealthCare. The European Pharmacopoeia (Ph. Eur.). Background Mission [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://www.edqm.eu/en/European-Pharmacopoeia-Background-Mission 2. Charton E. The role of European Pharmacopoeia monographs in setting quality standards for biotherapeutic products. Generics and Biosimilars Initiative Journal (GaBI Journal). 2016;5(4):174-9. doi: 10.5639/gabij.2016.0504.045 3. Council of Europe. European Directorate for the Quality of Medicines & HealthCare, EDQM & European Pharmacopoeia: State-of-the-art science for tomorrow’s medicines, June 2019, Strasbourg, France. Workshop on biotherapeutics [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://www.edqm.eu/sites/default/files/medias/fichiers/Events/edqm_european_pharmacopoeia_state-of-the-art_science_for_tomorrows_medicines_-_workshop_on_biotherapeutics.pdf 4. Council of Europe. European Directorate for the Quality of Medicines & HealthCare. Biotherapeutics [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://www.edqm.eu/en/biotherapeutics 5. Buda M, Wicks S, Charton E. Elaborating European Pharmacopoeia monographs for biotherapeutic proteins using substances from a single source. Pharmeur Bio Sci Notes. 2016:129-34. 6. Council of Europe. European Directorate for the Quality of Medicines & HealthCare. Technical guide for the elaboration of monographs on synthetic peptides and recombinant DNA proteins. 2018 [homepage on the Internet]. [cited 2022 Jan 26]. Available from https://www.edqm.eu/sites/default/files/medias/fichiers/European_Pharmacopoeia/Find_information_on/Technical_Guides/guide_ph_eur_synthetic_peptides_and_rdna_proteins_2018.pdf 7. Metcalfe C, Dougall T, Bird C, Rigsby P, Behr-Gross M-E, Wadhwa M, et al. The first World Health Organization International Standard for infliximab products: a step towards maintaining harmonized biological activity. MAbs. 2019;11(1):13-25. 8. Infliximab concentrated solution, monograph 2928. Ph. Eur. 10th Edition. Strasbourg, France: Council of Europe 2019. 9. Wadhwa M, Rigsby P, Behr-Gross M-E. Collaborative study for the establishment of infliximab biological reference preparation batch 1. Pharmeuropa Bio Sci Notes. 2020;2020:49-52. 10. Wadhwa M, Rigsby P, Behr-Gross M-E. Collaborative study for the establishment of etanercept biological reference preparation batch 1. Pharmeuropa Bio Sci Notes. 2020;2020:203-5. 11. Official Journal of the European Union. C 223. Guidelines on the details of the various categories of variations, on the operation of the procedures laid down in Chapters II, IIa, III and IV of Commission Regulation (EC) No 1234/2008 of 24 November 2008 concerning the examination of variations to the terms of marketing authorisations for medicinal products for human use and veterinary medicinal products and on the documentation to be submitted pursuant to those procedures [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:C:2013:223:FULL:EN:PDF 12. Council of Europe. European Directorate for the Quality of Medicines & HealthCare. Joint EDQM-EMA Session on Biosimilars. February 2017, Strasbourg, France [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://www.edqm.eu/en/european-pharmacopoeia-training-resources#Biologicals 13. Council of Europe. European Directorate for the Quality of Medicines & HealthCare. European Pharmacopoeia training session on Biologicals. February 2017, Strasbourg, France [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://www.edqm.eu/en/european-pharmacopoeia-training-resources#Biologicals 14. Council of Europe, European Directorate for the Quality of Medicines & HealthCare. EDQM Training Session – Biologicals. February 2020, Strasbourg, France. [homepage on the Internet]. [cited 2022 Jan 26]. Available from: https://www.edqm.eu/en/european-pharmacopoeia-training-resources#Biologicals
Author for correspondence: Emmanuelle Charton, PhD, European Pharmacopoeia Department, European Directorate for the Quality of Medicines & HealthCare (EDQM), Council of Europe, 7 allée Kastner, CS 30026, FR-67081 Strasbourg, France
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
Abstract:
Current regulations do not require a given biosimilar to remain similar to its reference biological over time. However, two products that were initially deemed biosimilar or interchangeable could each undergo unique patterns of drift and evolution in their manufacturing processes (divergence), ultimately resulting in two products that would be no longer biosimilar. In cases where divergence in potency, safety and immunogenicity may be present, care should be taken with multiple switches between reference and biosimilar products: each time a switch occurs, the difference between products could be greater. Taking into account that post-marketing comparative biosimilarity validation is not required, drift, evolution and divergence may present greater challenges when assessing biosimilar. In a marketplace with multiple biosimilars of a given reference product and in the context of interchangeability with drift and divergence, pharmacovigilance systems should be strengthened.
Submitted: 25 October 2021; Revised: 9 December 2021; Accepted: 14 December 2021; Published online first: 27 December 2021
Background
Currently, regulatory agencies in most of the world have established the requirements to achieve biosimilarity between two biological products. However, there is no mandatory legal obligation to perform quality or clinical studies that directly compare the biosimilar versus originator products in the post-approval period. Loss of biosimilarity over time could have important implications for the way in which regulators and healthcare providers handle safety surveillance, product naming, interchangeability, and medical records. Thus, biosimilars introduce new challenges because two products that were initially deemed biosimilar (or interchangeable) could each undergo unique patterns of variation resulting in two products that are no longer biosimilar (nor interchangeable). In this context, it would be essential that regulatory agencies adopt measures to minimize the risk of possible adverse events or lack of efficacy of treatments with biologicals, such as to determine extended biosimilarity and interchangeability standards and to strengthen pharmacovigilance systems.
Changes in the production process: comparability concept
During manufacturing process, the cell culture and fermentation processes are particularly critical and sensitive in terms of defining the identity, purity and potency of the approved biological. Modifications of parameters in any of these steps may impact cell culture performance, leading to variability in the quality of the recombinant protein [1].
Throughout the product life cycle of an approved biological molecule, a manufacturer may implement process changes to incorporate technological advances or efficiencies. Regulatory agencies evaluate these changes carefully and use scientific comparability criteria to determine whether there is a potential impact on the safety or efficacy that underlies its approval. The evolution of the changes introduced by the manufacturer follows a comparability exercise between the pre- and post-change product; and depending on the nature and extent of the manufacturing change, routine control measures and analytical tests may not be sufficient to assess the impact of the change on a product’s quality, safety and efficacy; this may necessitate non-clinical and clinical evaluations [2].
Until the mid-1990s, manufacturers of innovative biological products faced significant regulatory hurdles in making changes to their own manufacturing processes. But, in 1996, the US Food and Drug Administration (FDA) changed the paradigm for conducting comparability assessments of biological products in order to facilitate this approach. The agency’s justification for an increase in regulatory flexibility was based on recognition of the advances in analytical methodology and, perhaps more important, on the reasoning that ‘knowledge of the process involved in the manufacture of the product is an integral component in determining the design of an appropriate comparability assessment program’ [3].
The evolution of the regulations in Europe was certainly different from those of FDA. In 2001, the European Medicines Agency (EMA) established a comparability approach with the adoption of the Committee for Proprietary Medicinal Products (CPMP) by the ‘Guideline on comparability of medicinal products containing biotechnology-derived protein as active substances’ [4]. This guidance focused on comparability in the context of a change in the manufacturing process of a given product, but also, at the same time, on ‘comparability exercise’ that would need to be conducted to support an application for a product claimed to be similar to an already marketed product, with the recommendation that, in this latter case, additional preclinical and clinical studies (potentially a full data package) would be required. However, no ‘essentially similar’ product was approved based on this guidance, until, in 2005, a new independent pathway for the approval of ‘biosimilars medicinal products’ was introduced [5].
Later, other regulatory bodies provided analogous guidelines, culminating in the International Conference on Harmonization Comparability Guidance ICH Q5E, which acknowledges that ‘the demonstration of comparability does not necessarily mean that the quality attributes of the pre-change and post-change product are identical, but that they are highly similar and that the existing knowledge is sufficiently predictive to ensure that any differences in quality attributes have no adverse impact upon safety or efficacy of the drug product’. Currently, the principles of the comparability exercise established on ICH Q5E are recognized by regulatory authorities throughout the world [6].
In addition to the changes that manufacturers usually implement in manufacturing processes of an approved biological for a variety of reasons (including the need to comply with regulatory commitments, improve product quality and yield, and improve manufacturing efficiency and reliability), the technology transfer between different manufacturers is considered another potential scenario where comparability exercise would be performed. In this scenario, the company that developed the innovator product transfers the know-how and the full history of the manufacturing process to another manufacturer, similar to when a company opens its own second manufacturing site. All the information regarding critical quality attributes (CQAs), raw material, excipient suppliers, purification, formulation studies, containers, stability data, analytical methods, and product packaging would be available for consideration by the other manufacturer. Access to the full range of innovator manufacturing information fundamentally distinguishes this comparability approach from the situation facing the biosimilar product manufacturer [7].
Comparability versus biosimilarity
A biosimilar is a biopharmaceutical that has demonstrated similar CQAs, biological function, clinical efficacy and safety to that of an already licensed biological reference product. Then, biosimilarity must first be proved in an extensive analytical comparability exercise, systematically evaluating the quality and similarity of the biosimilar product and the originator product across dozens of physicochemical, biological and pharmacological CQAs, before establishing equivalence in clinical efficacy and safety [8].
Therefore, the scientific principles to establish the impact of a change in manufacturing process of a biological product (comparability) and those necessary to the generation of a biosimilar taking an innovator biological as a reference product (biosimilarity) are not the same. The potential for differences between an innovator biological and a biosimilar is greater than that between a biological before and after a manufacturing change [9].
The comparability practice as described within ICH Q5E applies to a single product before and after process changes within a single manufacturer. ICH Q5E would not sufficiently cover differences in the manufacturing process of the biosimilar compared to that of the reference product including expression system, recombinant DNA plasmid, fermentation system, control strategy, and purification process, process-related and product-related, formulation, container-closures system, drug product manufacturing and storage [10, 11].
The regulatory agency that most appropriately establishes the differences between biosimilarity and comparability is FDA in the ‘Scientific considerations in demonstrating biosimilarity to a reference protein product; Guidance for industry’, which states that: ‘Demonstrating that a proposed product is biosimilar to a reference product typically will be more complex than assessing the comparability of a product before and after manufacturing changes made by the same manufacturer. Even though some of the scientific principles described in ICH Q5E may also apply in the demonstration of biosimilarity, in general, FDA anticipates that more data and information will be needed to establish biosimilarity than would be needed to establish that a manufacturer’s post-manufacturing change product is comparable to the pre-manufacturing change product’ [12].
By contrast, from 2003, EMA uses the term comparability when evaluating both inter- as well as intra-manufacturing changes and as the explicit basis for biosimilars development [4]. More recently, EMA has used the expression ‘biosimilar comparability’ to clarify the context but not to change the concept as a scientific matter. Furthermore, EMA considers that, with the extensive experience of regulators and sponsors in highly regulated markets, comparability is the universal standard for judging interchangeability of pre- and post-manufacturing changes of any biological. With the development of the biosimilar approval pathway, the scientific approach underlying interchangeability can be broadly applied to an originator product undergoing a manufacturing change or a biosimilar at initial approval or a biosimilar undergoing a manufacturing change [13].
On interchangeability, the US legislation is much stricter and more specific, and differs substantially from EMA’s position. In 2019, FDA issued the guideline ‘Considerations in demonstrating interchangeability with a reference product; Guidance for industry’, in which it established that a biosimilar is not interchangeable with the innovative biological until the sponsor provides scientific and clinical evidence that supports such property of the biosimilar. In this guideline, FDA states that the term interchangeable or interchangeability means that the biological product may be substituted for the reference product without the intervention of the healthcare provider who prescribed the reference product. In this way, interchangeability is directly related to automatic substitution at the pharmacy level [14].
The interchangeability between innovator biological and biosimilar is not regulated in other countries of the world. However, it is an important scientific issue that is under constant debate.
Impact of drift, evolution and divergence on biosimilarity and interchangeability
Currently, one important question under debate is how to manage the oversight of a biosimilar if its reference product undergoes a change in its quality profile (or vice versa, if a biosimilar undergoes a change). In other words, is a biosimilar a ‘biosimilar forever’ or just a ‘biosimilar for licensing purposes’ that has a life cycle of its own after approval?
Biological product quality changes resulting from process variation may be unintended or intended. Unintended process variation may occur owing to the impact of uncontrolled variables and can result in gradual changes over time or in a sudden shift in a quality attribute, a process called manufacturing drift. The source of the change may not be well understood and may be an unintended result of changes outside of the manufacturer’s control [15].
As mentioned in the previous section, additional changes in product quality may be the result of intentional changes made by the manufacturers of biological medicines to the manufacturing process and can range from changes in manufacturing sites to changes in suppliers or cell culture media. Also, changes to a manufacturing process are sometimes made to introduce new technologies that can improve productivity. This type of change in the manufacturing process, called evolution, has been observed in most, if not all, approved biologicals on the market today since their initial approval [16].
Put together, normal variability, drift and evolution may present greater challenges when assessing biosimilars, and much more when they are evaluated as possible interchangeable products. Two products that were initially deemed biosimilar or interchangeable could each undergo unique patterns of drift and evolution, ultimately resulting in two products that are no longer biosimilar nor interchangeable. This process is defined as divergence [17].
Divergence is not just a hypothetical phenomenon. In some cases, divergence can occur for biologicals transferred between licensing partners, where the partners retain some right of reference to the originator’s development data, and also divergence can certainly occur and is arguably more likely with completely independent entities that have no right of reference to proprietary information collected during development, such as biosimilars manufacturers.
The case of epoetin alfa is an example of both types of divergence. Epoetin alfa is manufactured by separate entities for the US, Japan and Europe. Subcutaneous administration of Eprex® (epoetin alfa) in patients with chronic kidney disease (CKD) was banned in Europe between 2002 and 2006 after increasing reports of anti-erythropoietin (EPO) antibody-mediated pure red cell aplasia (PRCA) [18]. An investigation revealed that the transient increase of anti-EPO antibody mediated PRCA was associated with a change in the formulation/composition of the product. More precisely, the excipient of the formulation, human serum albumin, was replaced with polysorbate-80. This route of administration was subsequently restored after the sponsor addressed the manufacturing issue. Meanwhile, the corresponding US product did not implement the formulation change and retained the original route of administration on its label. The reason for the increase in PRCA observed with Eprex® has been associated with safety issues become apparent only in the post-marketing setting when larger numbers of patients are being treated [19].
Currently, a wider group of innovator, biosimilar and second-generation epoetin products are available across different markets [20]. Epoetins are heavily glycosylated proteins. Glycosylation profile is a CQA of epoetins, as it has a crucial influence upon in vivo biological and clinical activity [21]. Marketing authorization of biosimilar epoetin alfa products, e.g. Binocrit® and Silapo®, by EMA was based upon detailed biosimilarity exercises with the innovator product, Eprex®. In a recent study, the glycosylation profiles of Eprex® and the two approved biosimilars Binocrit® and Silapo® were characterized and compared. The products exhibit notable differences in N- and O-glycosylation, including attributes, such as sialic acid occupation, O-acetylation, N-acetyllactosamine extended antennae and sulphated/penta-sialylated N-glycans, which have the potential to cause divergency. The study highlights the need for continued monitoring of epoetin glycosylation, ideally allied to pharmacological data, in order to ensure consistency and therapeutic equivalence between products over time. In a marketplace where multiple epoetins are available, there exists the potential for divergence of glycosylation profiles, and therefore therapeutic potencies. It was evidenced that, post-authorization product surveillance and life-cycle management of epoetin alfa biosimilars, which may involve process manufacturing changes, can occur independently of Eprex® and to produce divergence in their clinical performance [22].
Regulatory controls are in place to ensure comparability of stand-alone biologicals before and after manufacturing changes. However, manufacturing changes to the biosimilar will not trigger repeated biosimilarity testing with the innovator; therefore, a standard of biosimilarity that is achieved at the time of approval of the biosimilar may not be maintained over time. Taking into account the possibility that divergence occurs, the similarity assessment should be an ongoing exercise that requires the biosimilar candidate to be assessed throughout the life cycle of the product. In addition to continuous biosimilarity, one important challenge for regulatory agencies is to demonstrate whether interchangeability is maintained in the longer term, particularly following changes to either the originator or (multiple) biosimilar product versions, see Figure 1[6].
Biosimilarity is not a transitive property
The relationship between a given biosimilar product and its reference product is unique and not transitive to other biosimilars. This is a consequence of the fact that biosimilars are not structurally identical to their reference biological products or to each other. Although differences between a biosimilar and its reference product are evaluated for equivalent clinical effects during biosimilarity assessment, it is unlikely that potential differences between any two indirectly related biosimilars will be formally evaluated. Indeed, there is no regulatory requirement to ensure that all biosimilars of a particular reference biological differ in a similar qualitative manner or to the same extent. Furthermore, biosimilar pathways permit variations in pharmaceutical attributes, clinical development approaches, and regulatory outcomes, resulting in further diversity of attributes among approved biosimilars [23].
The more important implication of this diversity is that biosimilars should not be used in practice in the same manner as multiple-source generic drugs. By definition, a generic medicine is interchangeable. A prescriber need not select any particular version, i.e. they are prescribed by the International Nonproprietary Name (INN) or generic name, and substitution among generic equivalents is commonly practiced at the pharmacy level without prescriber involvement [24]. Because biosimilars may vary across the ranges of structural and functional acceptance criteria, they should not be treated like multisource drugs and then none of the generic drug practices are advisable for biosimilars. Rather, once approved, they should be considered as individual therapeutic alternatives, stand-alone products, with all of the associated regulatory requirements. In practice, and in the context of multiple biosimilar versions of a single biological reference product, this means that biosimilars should be prescribed and tracked in medical records by a unique name, and clinicians should be involved in decisions to switch patients from one biological to another, particularly when a given biosimilar has not been qualified as interchangeable with the prescribed biological.
As was described above, uncorrected manufacturing process drift and/or evolution of one or both products, originator and biosimilar, could result in product divergence, and this divergence can occur between an originator and a biosimilar and between multiple biosimilar products. Divergence should not lead to a change in the safety or efficacy of each single product, originator and biosimilar(s), but could potentially result in clinically meaningful differences, e.g. potency, safety, or immunogenicity profile, during a chronic treatment interchanging biological. In cases where divergence may be present, care should be taken with multiple switches: each time a switch occurs, the difference between products will be greater. The most relevant cases may be divergence in potency (although both approved products are still effective, switching between them could cause a disruption in dosing), and divergence in immunogenicity profile (a patient could be exposed to one less immunogenic product and then be switched to the more immunogenic product) [25].
Concluding remarks
Three dynamic actions could be taken by regulatory agencies in order to control, at least partially, the clinical impact of divergence between innovator biologicals and biosimilars:
(1) Strengthen pharmacovigilance systems
Pharmacovigilance is especially important for biologicals because of their susceptibility to changes in the manufacturing process and the possibility that drift, evolution, or divergence may have adverse consequences for patients. A robust, product-specific pharmacovigilance system for biologicals may require special policy measures such as mandatory use of distinguishable names for prescribing.
(2) Determine interchangeability standards
Currently, only FDA has established the scientific basis to determine if a biosimilar should be deemed interchangeable with an innovator biological [14]. However, no regulatory mechanisms are currently in place to ensure continued interchangeability in the event of product drift, evolution or divergence. Thus, products that were interchangeable at the time of approval might continue to be considered interchangeable by regulators even though the quality attributes of the originator and biosimilar products have diverged.
(3) Establish the differences between comparability and biosimilarity
The impact of a particular change and any product evolution can be readily evaluated by comparability exercise. To the contrary, in biosimilar development, almost every aspect of the manufacturing process may have changed, and the only point of reference is the reference drug product. As biological manufacturers do not have access to the originator manufacturing process as a point of reference, comparing the biosimilar product to the reference product (biosimilarity) is necessarily a more complex process.
Competing interests: None.
Provenance and peer review: Not commissioned; externally peer reviewed.
Author: Pablo Matar, PhD, Institute of Experimental Genetics, School of Medical Sciences, National University of Rosario (UNR), Santa Fe 3100, (2000) Rosario, Argentina; and National Scientific and Technical Research Council (CONICET), Rosario, Argentina
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
Abstract:
Systematic clinical switch studies to demonstrate interchangeability of biosimilars are required in some jurisdictions while theoretical considerations and current clinical data question the feasibility and relevance of such studies.
Submitted: 27 November 2021; Revised: 15 December 2021; Accepted: 15 December 2021; Published online first: 28 December 2021
Biosimilars have been used for 15 years in the European Union (EU). They are able to trigger price competition and reduce costs as well as to increase access to important biological medicines [1]. In spite of the long experience and excellent safety record, many prescribers and other healthcare professionals still have concerns about the quality, safety, efficacy, and especially interchangeability of biosimilars [2]. Systematic clinical switch studies to demonstrate interchangeability of biosimilars are required in some jurisdictions while theoretical considerations and current clinical data question the feasibility and relevance of such studies.
Differences in terminology and regulatory approach hurt biosimilars
In the EU, interchangeability is a scientific and clinical concept meaning that switching between a biosimilar and its reference product does not change the efficacy or safety [3]. Interchangeable products can be switched under the control of the prescriber (physician-led switch) or at the pharmacy without consultation of the prescriber (automatic substitution). Both scenarios are practical applications of interchangeability. Several European national regulatory authorities have issued position papers that endorse physician-led switching [4]. Automatic substitution is a political, administrative, and practical measure that may require changes to the legislation and needs to be adapted to local circumstances.
In the US, interchangeability is regarded as a higher level of biosimilarity that allows automatic substitution. The US Food and Drug Administration (FDA) may grant interchangeable status to already licensed biosimilars if they fulfill the requirements of its interchangeability guideline [5]. The main hurdle is the conduct of a controlled clinical trial consisting of three switches. The switch studies constitute a significant burden to developers and patients. In addition, the results may not give a definite answer [6]. Thus far, the interchangeable status has been granted to only two products by FDA; Semglee (insulin glargine-yfgn) in July 2021 and Cyltezo (adalimumab-adbm) in November 2021.
In spite the fact that switch studies are not required in the EU, the concept a clinical switch study is deeply rooted in the minds of prescribers and policymakers also in the EU, as seen in some position papers of national regulatory authorities and learned societies [4] and in the hesitance among prescribers [2]. Manufacturers whose old blockbuster products are threatened by biosimilars support requirements for systematic multiple switch studies and even switch studies between biosimilars and repeated demonstration of interchangeability [7-9] – obviously because extensive switch studies discourage biosimilar development. The feasibility of conducting extensive switch studies has been questioned [6, 10]. Moreover, it is difficult to find detailed theoretical justifications and clinical data to support systematic switch studies. Thus, the crucial question is not what kind of comparative switch studies should be done but whether these studies are justified and feasible at all.
The hypothesis of switch-related immunogenicity
Immunogenicity is presented as the risk of switching between a biosimilar and its reference product. The FDA interchangeability guideline states,‘in the context of switching between the products, multiple exposures to each product may potentially prime the immune system to recognize subtle differences in structural features between products’ and‘the overall immune response could be increased under these conditions’ [5]. Some proponents of switch studies explain this hypothesis by the discontinuation theory [7]. Thus,‘subtle’ differences in the structure, notably the glycosylation profiles of active substances of biosimilars and their reference product seem to be a cornerstone for the immunogenicity hypothesis [8]. This theory may be relevant to vaccines that are designed to be‘foreign’ to human immune system and immune response is amplified by adjuvants ― a completely different situation compared to therapeutic proteins that are made highly similar. Thus, this possibility is also highly unlikely after switching between a biosimilar and its reference product that have already been shown to be highly similar and which are expected to have some degree of immunological tolerance due to their resemblance to human proteins [11].
It is also suggested that exposure to different sets of antigenic epitopes upon switching might enhance anti-drug immunity by epitope spreading. This situation has been observed in epoetin alpha (epoetin-α) products where a poor product formulation or inappropriate storage may lead to cross-reacting neutralizing antibodies to endogenous erythropoietin. The root cause of the induction of neutralizing erythropoietin antibodies is aggregation of proteins in a poor formulation [12]. Thus far, there is no clinical data to support above-mentioned immunogenicity theories in the context of switching between a reference product and its licensed biosimilar version.
Switch-related immunogenicity: a remote risk
Switch-related immunogenicity is often seen as an unpredictable reaction of the immune system. However, the factors triggering of anti-drug immunity are well known. Immune recognition is essential for identifying proteins as‘self’ or‘non-self’. Recognition of the protein as‘non-self’ may be associated with transient or persistent immune reactions. The switching between highly similar versions of the same active substance is not sufficient to break tolerance and to raise persisting and strong T cell-dependent immunogenicity such as affinity maturation, isotype switch or strong recall response due to the identical amino acid sequence and highly similar secondary structure. Differences in glycosylation profiles have never triggered immunogenicity of different versions of therapeutic proteins [10]. Such differences are found after changes in manufacturing process and even between different batches of therapeutic proteins without immunological problems [13]. Studies of originator and biosimilar infliximab and adalimumab support the similarity of important antigenic epitopes in biosimilars and their reference products [14-16].
B cells may be stimulated independently by impurities, such as bacterial endotoxins as well as by degraded and aggregated proteins [11]. Such problems may be due to inappropriate storage or poor formulations, i.e. not due to‘subtle differences’ in the active substances of highly similar products in normal use. These cases, such as the formulation change of originator epoetin-α or one of its biosimilar versions have made developers and regulators very careful in assessing formulations and stability studies of biosimilars [17]. Therefore, it is unlikely that licensed biological products will have immunogenicity triggered by impurities [13].
The clinical evidence
Review of 178 clinical switch studies of biosimilars and their reference products found no evidence of switch-related adverse effects, including harmful immunogenicity [18]. More recent reviews of newer biosimilars confirm the lack of switch-related adverse effects [19, 20]. From a theoretical perspective, the lack of adverse effects is the expected outcome whereas the hypothesis of switch-related adverse effects is not supported by current evidence [6, 11, 21]. Furthermore, clinical switch studies will not be able to detect potential rare adverse effects. Thus, the risk management of switching should be based on pharmacovigilance and pharmaco-epidemiological studies.
Clinical switch studies: time for a re-evaluation
The concept of systematic switch studies is becoming obsolete, but FDA keeps requiring extensive studies due to US legislation. The innovative industry is promoting even more extensive switch studies due to their commercial interests, experts do not dare to challenge FDA and the existing dogma, and prescribers are confused. Ironically, a similar situation has been described in the old tale‘The Emperor’s New Clothes’ [22].
The moral of the tale of systematic switch studies is that EU regulators should have the courage to issuing a common position in interchangeability of biosimilars [23]. The message should be that EU-licensed biosimilars are interchangeable with its reference product without systematic clinical switch studies.
Funding sources
No financial support was received for conducting this research.
Disclaimer
The opinions expressed in this article are personal views of the author and should not be understood being made on behalf of or reflecting the position of the agencies or organizations with which the author is or has been affiliated.
Competing interests: None.
Provenance and peer review: Not commissioned; externally peer reviewed.
References 1. Troein P, Newton M, Scott K. The impact of biosimilar competition in Europe. December 2020 [homepage on the Internet]. [cited 2021 Dec 15]. Available from: https://ec.europa.eu/health/sites/health/files/human-use/docs/biosimilar_competition_en.pdf 2. Barbier L, Simoens S, Vulto AG, Huys I. European stakeholder learnings regarding biosimilars: Part I—improving biosimilar understanding and adoption. BioDrugs 2020;34:783-96. 3. European Medicines Agency and European Commission. Biosimilars in the EU. Information guide for healthcare professionals [homepage on the Internet]. [cited 2021 Dec 15]. Available from: https://www.ema.europa.eu/en/documents/leaflet/biosimilars-eu-information-guide-healthcare-professionals_en.pdf 4. Medicines for Europe. Positioning statements on physician-led switching for biosimilar medicines. 2020 [homepage on the Internet]. [cited 2021 Dec 15]. Available from: https://www.medicinesforeurope.com/docs/M-Biosimilars-Overview-of-positions-on-physician-led-switching.pdf 5. U.S. Food and Drug Administration. Considerations in demonstrating interchangeability with a reference product. Guidance for Industry. May 2019 [homepage on the Internet]. [cited 2021 Dec 15]. Available from: https://www.fda.gov/media/124907/download 6. Alvarez DF, Wolbink G, Cronenberger C, Orazem J, Kay J. Interchangeability of biosimilars: what level of clinical evidence is needed to support the interchangeability designation in the United States? BioDrugs. 2020;34(6):723-32. 7. Faccin F, Tebbey P, Alexander E, Wang X, Cui L, Albuquerque T. The design of clinical trials to support the switching and alternation of biosimilars. Expert Opin Biol Ther. 2016;16(12):1445-53. 8. Feagan BG, Marabani M, Wu JJ, Freddy Faccin F, Spronk C, Castañeda-Hernández G. The challenges of switching therapies in an evolving multiple biosimilars landscape: a narrative review of current evidence. Adv Ther. 2020;37(11):4491-518. 9. Afzali A, Furtner D, Melsheimer R, Molloy PJ. The automatic substitution of biosimilars: definitions of interchangeability are not interchangeable. Adv Ther. 2021;38(5):2077-93. 10. Mysler E, Azevedo VF, Danese S, Alvarez D, Iikuni N, Ingram B, et al. Biosimilar-to-biosimilar switching: what is the rationale and current experience? Drugs. 2021;81(16):1859-79. 11. Doevendans E, Schellekens H. Immunogenicity of innovative and biosimilar monoclonal antibodies. Antibodies (Basel). 2019;8(1):21. 12. Rubic-Schneider T, Kuwana M, Christen B, Aßenmacher M, Hainzl O, Zimmermann F, et al. T-cell assays confirm immunogenicity of tungsten-induced erythropoietin aggregates associated with pure red cell aplasia. Blood Adv. 2017;1(6):367-79. 13. Lamanna WC, Holzmann J, Cohen HP, Guo X, Schweigler M, Stangler T, et al. Maintaining consistent quality and clinical performance of biopharmaceuticals. Expert Opin Biol Ther. 2018;18(4):369-79. 14. Reinisch W, Jahnsen J, Schreiber S, Danese S, Panés J, Alejandro Balsa A, et al. Evaluation of the cross-reactivity of antidrug antibodies to CT-P13 and infliximab reference product (Remicade): an analysis using immunoassays tagged with both agents. BioDrugs. 2017;31(3):223-37. 15. Ben-Horin S. Yavzori M, Benhar I, Fudim E, Picard O. Ungar B, et al. Cross-immunogenicity: antibodies to infliximab in Remicade-treated patient with IBD similarly recognise the biosimilar Remsima. Gut. 2016;65(7):1132-8. 16. Goncalves J, Myung G, Park M, Jeong D, Ghil J. SB5 shows cross-immunogenicity to adalimumab but not infliximab: results in patients with inflammatory bowel disease or rheumatoid arthritis. Therap Adv Gastroenterol. 2019;12:1756284819891081. 17. European Medicines Agency. Guideline on similar biological medicinal products containing biotechnology-derived proteins as active substance: quality issues (revision 1). 22 May 2014 [homepage on the Internet]. [cited 2021 Dec 15]. Available from:https://www.ema.europa.eu/en/documents/ scientific-guideline/guideline-similar-biologicalmedicinal- products-containing-biotechnologyderived- proteins-active_en-0.pdf 18. Barbier L, Ebbers H, Declerck P, Simoens S, Vulto A, Huys I. The effi cacy, safety, and immunogenicity of switching between reference biopharmaceuticals and biosimilars: a systematic review. Clin Pharmacol Ther. 2020;108(4):734-5. 19. Strand V, Gonçalves J, Hickling TP, Jones HE, Marshall L, Isaacs JD. Immunogenicity of biosimilars for rheumatic diseases, plaque psoriasis, and infammatory bowel disease: a review from clinical trials and regulatory documents. BioDrugs. 2020;34(1):27-37. 20. Kurki P, Barry S, Bourges I, Tsantili P, Wolff- Holz E. Safety, immunogenicity and interchangeability of biosimilar monoclonal antibodies and fusion proteins: a regulatory perspective. Drugs. 2021;81(16):1881-96. 21. Chamberlain P. Multidisciplinary approach to evaluating immunogenicity of biosimilars: lessons learnt and open questions based on 10 years’ experience of the European Union regulatory pathway. Biosimilars. 2014;4:23-43. 22. Andersen HC. The emperor’s new clothes (“Kejserens nye klæder”). In: Reitzel CA. Fairy tales told for children. Third Booklet. 1837. 23. Ebbers HC, Schellekens H. Are we ready to close the discussion on the interchangeability of biosimilars? Drug Discov Today. 2019;24(10):1963-7.
Author: Adjunct Professor Pekka Kurki, MD, PhD, University of Helsinki, 19 Lukupolku, FI-00680 Helsinki, Finland
Disclosure of Conflict of Interest Statement is available upon request.
Permission granted to reproduce for personal and non-commercial use only. All other reproduction, copy or reprinting of all or part of any ‘Content’ found on this website is strictly prohibited without the prior consent of the publisher. Contact the publisher to obtain permission before redistributing.
GaBI Journal is an independent and peer reviewed academic journal. GaBI Journal encompasses all aspects of generic and biosimilar medicines development and use, from fundamental research up to clinical application and policies.
