sustainability of health care expenditure **I. Competition is a more effective mechanism to achieve a long-term predictable price level than regulation**

Biosimilars have the potential to reduce the cost of treatment; this, in turn, strengthens the


Disruption and transition costs occur in both hospital and out-of-hospital (including retail and home care) settings; these differences may need to be considered Note: icons shown on the right represent level of agreement between the stakeholders. The 'consensus' icon indicates that all stakeholders (physicians, payers, policy advisors, **POLICY II. There needs to be incentives for investment in future biosimilars** costs and a portion of the savings are used to meet these costs **PHARMACIST PHYSICIAN** Note: icons shown on the right represent level of agreement between the stakeholders. The 'consensus' icon indicates that all stakeholders (physicians, payers, policy advisors, Note: icons shown on the right represent level of agreement between the stakeholders. The 'consensus' icon indicates that all stakeholders (physicians, payers, policy advisors, manufacturers, pharmacists, and patients) agreed on that point.

**POLICY**

**PHARMACIST**

**PHYSICIAN**

**CONSENSUS**

**CONSENSUS**

**PHARMACIST**

**PHYSICIAN**

**PHARMACIST**

**PHYSICIAN**

**PATIENT**

**III. A sustainable biosimilar market requires collaboration between stakeholders**

Transitioning between biosimilars causes disruption to patient care and health care services. Unnecessary disruptions (i.e., frequent transitions and/or transitions that do not

deliver tangible savings) should be minimized

Disruption caused by biosimilar transition may be unavoidable in some therapeutic areas (e.g., acute vs. chronic conditions); however, switch is not advisable if treatment duration Disruption and transition costs occur in both hospital and out-of-hospital (including retail

and home care) settings; these differences may need to be considered

**III. A sustainable biosimilar market requires collaboration between stakeholders**

is short

manufacturers, pharmacists, and patients) agreed on that point.

manufacturers, pharmacists, and patients) agreed on that point.

extent, therapies with biosimilars already available

Continued investment in biosimilar development and market entry is important to generate competition for biological therapies for which no biosimilar is currently available and, to a lesser

**II. A sustainable biosimilar market must address the needs of all stakeholders**

**Box 3.** Consensus on drivers of and risks to a sustainable biosimilar market.


For key market risks, there was agreement that there is a need for better indicators than those currently available (e.g., the number of biosimilar manufacturers and manufacturing sites) to warn of potential de facto monopoly [30]. Participants agreed that the emergence of monopolies could lead to higher price levels and/or enhanced supply risks (such as poor quality), or supply shortages (e.g., limited production capabilities and poor distribution channels) for biosimilars. This risk also exists for generics, but it would be greater for biosimilars due to the lengthier development and market entry processes, and the much longer lead time in manufacturing (1 year or more). Participants felt that there was a need for more research to identify prospective indicators of market performance; these should be based on a thorough understanding of the role that procurement level (national vs. subnational (procurement is described below)), market size, number of awarded contracts (and market share awarded), and tender criteria may play in ensuring markets perform well. Unfortunately, published evidence on indicator performance or biosimilar supply risks and shortages are scarce making generalizability difficult, but also highlighting the need for establishing validated approaches to long-term quantification of these frameworks.

3.1.4. Drivers and Risks of a Sustainable Biosimilar Market (Procurement Processes)

Issues surrounding procurement processes are summarized in Box 3 and Table 3. Participants agreed that procurement processes should avoid monopolies and minimize patient and health care system disruption, and the principles for procurement should be agreed by all stakeholders. The participants also identified two main goals of procurement design from a multistakeholder perspective. The first goal was to prevent predatory behavior by considering factors in selection criteria other than price or aggressive price discounting; these could include differentiation based on formulation and quality attributes, or stock and distribution channels. The second goal was to minimize disruptions to patient care based on the needs of individual therapy areas, perhaps by setting a contract duration that is proportional to the duration of treatment. Given the potential implications of procurement policies for all stakeholders, participants agreed that all stakeholders should have a voice in setting procurement policies. Participants agreed that there cannot be a "one size fits all" approach to procurement, as the structure and characteristics of health care systems vary; however, procurement policies should be consistent, guided by a common set of principles, and abide with European Union rules on tendering. Participants also advised that biosimilar procurement must be managed carefully over the product lifecycle to preserve competition and promote new investment in biosimilar development.

*Pharmaceuticals* **2020**, *13*, 400 *Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 11 of 19 *Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 11 of 19

*Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 5 of 19

**Table 1.** Consensus on components of a sustainable biosimilar market.

**Table 3.** Consensus on drivers and risks to biosimilar market sustainability (procurement processes). **I. Procurement processes should avoid monopolies and minimize patient and health care system disruption** The emergence of monopolies may lead to higher price levels and/or enhanced supply risks **I. A sustainable biosimilar market must deliver tangible and transparent benefits to the health care system** Biosimilars have the potential to reduce the cost of treatment; this, in turn, strengthens the sustainability of health care expenditure Biosimilar-related savings must be tangible and transparent and should be reinvested efficiently; this may include addressing deficits, and funding innovative therapies, health care or other public services. Biosimilars have the potential to expand access8 Providers (physicians and pharmacists) incur real costs when transitioning to a new biosimilar; transition should only occur if savings substantially exceed these transition costs and a portion of the savings are used to meet these costs **II. A sustainable biosimilar market must address the needs of all stakeholders** Transitioning between biosimilars causes disruption to patient care and health care services. Unnecessary disruptions (i.e., frequent transitions and/or transitions that do not deliver tangible savings) should be minimized Disruption caused by biosimilar transition may be unavoidable in some therapeutic areas (e.g., acute vs. chronic conditions); however, switch is not advisable if treatment duration is short Disruption and transition costs occur in both hospital and out-of-hospital (including retail and home care) settings; these differences may need to be considered **III. A sustainable biosimilar market requires collaboration between stakeholders CONSENSUS CONSENSUS PHARMACIST PHYSICIAN CONSENSUS PHARMACIST PHYSICIAN PATIENT PHARMACIST PHYSICIAN** • There are examples of this in generics, although these issues would be more pronounced for biosimilars due to lengthy development and market entry processes *Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 11 of 19 **Table 3.** Consensus on drivers and risks to biosimilar market sustainability (procurement processes). **I. Procurement processes should avoid monopolies and minimize patient and health care system disruption** The emergence of monopolies may lead to higher price levels and/or enhanced supply risks There are examples of this in generics, although these issues would be more pronounced for biosimilars due to lengthy development and market entry processes Procurement design should aim to: Prevent predatory behaviour, e.g., by considering factors other than price to avoid aggressive price discounting Minimize disruption of patient care, based on the needs of individual therapeutic areas, e.g., by setting contract duration that is proportional to duration of treatment **II. The principles for procurement should be agreed by all stakeholders** There should be a multistakeholder group that sets principles for policy and practice around biosimilar procurement Patients and physicians should have an opportunity for their views to be represented (e.g., in a national forum) and patients should be informed of the rationale behind procurement decisions that impact on their care **CONSENSUS MANUFACTURER POLICY PHARMACIST PAYER PHARMACIST PHYSICIAN PATIENT CONSENSUS PHYSICIAN PATIENT** Procurement design should aim to: • Prevent predatory behaviour, e.g., by considering factors other than price to avoid aggressive price discounting **Table 3.** Consensus on drivers and risks to biosimilar market sustainability (procurement processes). **I. Procurement processes should avoid monopolies and minimize patient and health care system disruption** The emergence of monopolies may lead to higher price levels and/or enhanced supply risks There are examples of this in generics, although these issues would be more pronounced for biosimilars due to lengthy development and market entry processes Procurement design should aim to: Prevent predatory behaviour, e.g., by considering factors other than price to avoid aggressive price discounting Minimize disruption of patient care, based on the needs of individual therapeutic areas, e.g., by setting contract duration that is proportional to duration of treatment **II. The principles for procurement should be agreed by all stakeholders** There should be a multistakeholder group that sets principles for policy and practice around biosimilar procurement Patients and physicians should have an opportunity for their views to be represented (e.g., in a national forum) and patients should be informed of the rationale behind procurement decisions that impact on their care **CONSENSUS MANUFACTURER POLICY PHARMACIST PAYER PHARMACIST PHYSICIAN PATIENT CONSENSUS PHYSICIAN PATIENT** • Minimize disruption of patient care, based on the needs of individual therapeutic areas, e.g., by setting contract duration that is proportional to duration of treatment **Table 3.** Consensus on drivers and risks to biosimilar market sustainability (procurement processes). **I. Procurement processes should avoid monopolies and minimize patient and health care system disruption** The emergence of monopolies may lead to higher price levels and/or enhanced supply risks There are examples of this in generics, although these issues would be more pronounced for biosimilars due to lengthy development and market entry processes Procurement design should aim to: Prevent predatory behaviour, e.g., by considering factors other than price to avoid aggressive price discounting Minimize disruption of patient care, based on the needs of individual therapeutic areas, e.g., by setting contract duration that is proportional to duration of treatment **II. The principles for procurement should be agreed by all stakeholders** There should be a multistakeholder group that sets principles for policy and practice around biosimilar procurement Patients and physicians should have an opportunity for their views to be represented (e.g., in a national forum) and patients should be informed of the rationale behind procurement decisions that impact on their care **CONSENSUS MANUFACTURER POLICY PHARMACIST PAYER PHARMACIST PHYSICIAN PATIENT CONSENSUS PHYSICIAN PATIENT II. The principles for procurement should be agreed by all stakeholders** There should be a multistakeholder group that sets principles for policy and practice around biosimilar procurement *Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 5 of 19 **Table 1.** Consensus on components of a sustainable biosimilar market. **I. A sustainable biosimilar market must deliver tangible and transparent benefits to the health care system** Biosimilars have the potential to reduce the cost of treatment; this, in turn, strengthens the sustainability of health care expenditure Biosimilar-related savings must be tangible and transparent and should be reinvested efficiently; this may include addressing deficits, and funding innovative therapies, health care or other public services. Biosimilars have the potential to expand access8 Providers (physicians and pharmacists) incur real costs when transitioning to a new biosimilar; transition should only occur if savings substantially exceed these transition costs and a portion of the savings are used to meet these costs **II. A sustainable biosimilar market must address the needs of all stakeholders CONSENSUS CONSENSUS PHARMACIST PHYSICIAN** Patients and physicians should have an opportunity for their views to be represented (e.g., in a national forum) and patients should be informed of the rationale behind procurement decisions that impact on their care *Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 11 of 19 **Table 3.** Consensus on drivers and risks to biosimilar market sustainability (procurement processes). **I. Procurement processes should avoid monopolies and minimize patient and health care system disruption** The emergence of monopolies may lead to higher price levels and/or enhanced supply risks There are examples of this in generics, although these issues would be more pronounced for biosimilars due to lengthy development and market entry processes Procurement design should aim to: Prevent predatory behaviour, e.g., by considering factors other than price to avoid aggressive price discounting Minimize disruption of patient care, based on the needs of individual therapeutic areas, e.g., by setting contract duration that is proportional to duration of treatment **II. The principles for procurement should be agreed by all stakeholders CONSENSUS MANUFACTURER POLICY PHARMACIST PAYER PHARMACIST PHYSICIAN PATIENT** There can be no one-size-fits-all approach to procurement, as the structure and characteristics of health care systems vary; however, there should be a consistent approach and a common set of guiding principles *Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 8 of 19 **Table 2.** Consensus on drivers and risks to biosimilar market sustainability (competition and incentives). **I. Competition is a more effective mechanism to achieve a long-term predictable price level than regulation** Increased competition leads to more rapid price reduction and, if procurement policies contribute to business continuity, a sustained lower price level There is a need to develop better prospective indicators to warn about potential risk of *de facto*  Existing indicators, such as the number of manufacturers and manufacturing sites for biosimilars, are imperfect and may only indicate a problem that is too late to reverse Additional indicators that could be explored include procurement design (e.g., contract length), geographic division (national vs. regional) and factors other than cost **CONSENSUS CONSENSUS MANUFACTURER POLICY**

**CONSENSUS**

**PHARMACIST**

**PAYER**

**PHARMACIST**

**PHYSICIAN**

**POLICY**

**PHYSICIAN**

**PATIENT**

**PHARMACIST**

**PHYSICIAN**

**CONSENSUS**

**PATIENT**

**PAYER**

services. Unnecessary disruptions (i.e., frequent transitions and/or transitions that do not

has been primarily price-focused and has led to a reduction in "value-add" (e.g., patient support

Patients and physicians should have an opportunity for their views to be represented (e.g., in a national forum) and patients should be informed of the rationale behind Disruption caused by biosimilar transition may be unavoidable in some therapeutic areas (e.g., acute vs. chronic conditions); however, switch is not advisable if treatment duration

Price-setting regulation, if needed to prevent predatory behaviour, should not aim primarily at the

lowest possible prices but at long-term viability of a vibrant and competitive marketplace

procurement decisions that impact on their care

Disruption and transition costs occur in both hospital and out-of-hospital (including retail

**II. There needs to be incentives for investment in future biosimilars**

and home care) settings; these differences may need to be considered

Continued investment in biosimilar development and market entry is important to generate competition for biological therapies for which no biosimilar is currently available and, to a lesser

**III. A sustainable biosimilar market requires collaboration between stakeholders**

extent, therapies with biosimilars already available

is short

deliver tangible savings) should be minimized

programs)

•

•

•

monopoly

•

•

•

•

•

•

•

•

•

•

•

### *3.2. Key Findings from the SLR*

A total of 36 studies were identified in the SLR (Appendix B). Nine publications were identified that discussed (Q1). However, these were too limited to provide any comprehensive evidence and demonstrate the lack of a consistent, comprehensive database of medicine shortages in Europe. Nineteen publications addressed (Q2). None of these reported switching between biosimilars; rather, all considered switches from a reference product to a biosimilar. Nine publications focused on (Q3). These offered insufficient evidence from which to reach generalized conclusions about the effects of different tender models on the outcomes of interest. However, one policy paper concluded that barriers to entry, including the use of single-manufacturer tenders, will limit competition in biosimilars [16]. This paper was considered by the panel, together with additional evidence summarized in Appendix C.

### **4. Discussion**

A Delphi process, involving diverse stakeholders from across Europe, was conducted to achieve a consensus opinion on biosimilar market sustainability in Europe. Divergent views between stakeholder groups, and the reasons for these, were explored through individual, anonymized feedback and facilitated discussion at a roundtable meeting. This important exercise was undertaken to increase our understanding of the current system and to address concerns regarding sustainability, including the unmet need to develop a long-term vision, as highlighted in previous analyses [11,19,20]. Participants agreed that a sustainable biosimilar market must deliver tangible and transparent benefits to the health care system, while meeting the needs of all stakeholders. The definition (as shown in Box 1) was approved by all participants; however, different stakeholder groups emphasized different priorities within this definition, which is consistent with the previous literature on a lack of a unified approach [19,20]. Participants also agreed that, to make this approach work, collaboration between stakeholders is required and a greater awareness of the drivers of and threats to a sustainable market. In brief, strategies around competition, incentives, and procurement policies were identified and discussed with key consensus highlighted in the tables. These areas (notably the need to establish healthy competition, pricing, and market access policies (considering gain sharing and price reductions), government policy and guidance, identification of risks associated with biosimilar drug supply (e.g., quality issues), and patient access to information and education) were highlighted in the previous literature as key areas requiring further improvements [11,19]. Participants in the Delphi process agreed that these key findings should be developed further into a white paper that highlights the need for multistakeholder collaboration on establishing principles for biosimilar procurement in Europe.

Several priorities for future research were identified by stakeholders. First, understanding and measuring the impact of biosimilar transition on hospital and health care services will better enable costs and benefits to be weighed up and help minimize disruption for patients and health care services. Second, there is a need to understand and develop prospective indicators of market sustainability and potential risks to competitive biosimilar markets, particularly the emergence of de facto monopolies and supply risks. Finally, it will also be important to understand the implications of procurement structure and design for biosimilar market sustainability, especially with regard to how the procurement level (national vs. subnational), market size, number of awarded contracts (and market share awarded), and tender criteria affect market sustainability.

There is currently very limited published evidence available to support detailed arguments in the three priority areas described above, largely because there are limited data with which to conduct analyses. Biosimilar markets are still relatively new in Europe, which means that the available data relate to limited time periods and newly emerging trends that may be expected to mature over time. Further, the currently available data (e.g., on supply shortages of biosimilars) are kept at the national level; this allows cross-country comparisons but poses a challenge for pan-European analysis. It is therefore recommended that any further research begins with a scoping phase, in which the available data are reviewed in detail to assess their suitability for the proposed purpose. Further research would

also benefit from a more quantifiable approach to the sustainability framework, allowing us to measure the extent to which a biosimilar market in a specific jurisdiction can be effectively maintained.

Collaboration with stakeholders to develop principles for biosimilar procurement may be progressed in tandem with further research. The objective of establishing processes is to ensure that the concerns of all stakeholders—patients, physicians, pharmacists, payers, policy advisers, and manufacturers—are considered in procurement design. In the absence of evidence, open communication and collaboration between stakeholders may provide the necessary information that procurement decision makers need to prevent risks to biosimilar market sustainability from materializing.

This Delphi process involved a limited number of stakeholders and, as with any Delphi exercise, may also be biased by those who chose to participate [31]. For example, a number of issues were not considered such as the evolution of the biosimilar production process over time. However, the process encompassed evidence from a broad review of available literature and covered a broad range of stakeholder perspectives. Despite a rigorous approach, the findings of the SLR indicated that there was an absence of consistent, comprehensive information about drug shortages (specifically biosimilar shortages) and the costs of switching to biosimilars in Europe; these gaps exacerbate a lack of evidence regarding the impact of different tender models for savings, sustainable competition, and supply risk. The panel identified eight key papers (Appendix C), some of which were not identified by the SLR. The consensus reached by the Delphi process provides further direction for future research into, and implementation of, potential strategies to support these different aspects of sustainability.

### **5. Conclusions**

A sustainable biologics market including biosimilars is essential for ensuring that health care savings are maintained into the future, both for existing molecules and those approaching a loss of exclusivity. This Delphi approach resulted in a consensus definition of biosimilar market sustainability in Europe, specified the components of a sustainable biosimilar market, and identified key drivers and risks to sustainability. Crucially, participants in the Delphi process highlighted the need for multistakeholder collaboration in designing policy and practice relating to biosimilars (including procurement). Further research is required alongside stakeholder collaboration to inform biosimilar policy and practice in alignment with the principles identified in this Delphi process. Failure to care for biosimilar market sustainability may impoverish the biosimilar development and offerings, eventually leading to increased cost for health care systems and patients, with fewer resources for innovation.

**Author Contributions:** Conceptualization, A.G.V. and J.H.; methodology, A.G.V. and J.H.; validation, A.G.V., J.V.-O., M.v.d.G., S.R.A.S. and L.D.; informal analysis, A.G.V., J.V.-O. and B.M.; investigation, A.G.V., M.v.d.G., S.R.A.S. and L.D.; resources, A.G.V. and J.V.-O.; data curation, A.G.V. and J.V.-O.; writing—original draft preparation, J.V.-O. and J.L.; writing—review and editing, All authors; visualization, All authors; supervision, A.G.V. and J.V.-O.; project administration, B.M.; funding acquisition, A.G.V. and J.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received funding from Amgen.

**Acknowledgments:** The authors would like to thank the following for their participation in the panel discussions: Nathalie Deparis (patient advocating, rheumatoid arthritis), Hans-Christian Kolberg (physician), Stuart Parkes and Noemi Martinez Lopez De Castro (hospital pharmacists), Jean-Michel Descoutures and Tim Visser (procurement pharmacists), Jorge Mestre Ferrandiz (policy advisers, health economist), and Stephan Rönninger (manufacturer).

**Conflicts of Interest:** A.G.V. reports personal fees from AbbVie, Accord-Healthcare, Amgen, Biogen Idec, Febelgen, Fresenius-Kabi, Hexal, Medicines for Europe, Mundipharma, Novartis, Pfizer, Samsung, and Sandoz; J.V.-O., R.M., and B.M. are employees of Parexel who were contracted by Amgen; M.v.d.G. reports personal fees from Amgen and BioMarin; L.D. reports personal fees from Abbvie, Amgen, Biogen, BMS, Celltrion, Novartis, Pfizer, Roche, Sanofi-Genzyme, and SOBI; J.L. and J.H. are employees of Amgen; S.G.M. is an Amgen employee and stockholder; S.S. is one of the founders of the KU Leuven Fund on Market Analysis of Biologics and Biosimilars following Loss of Exclusivity (MABEL). He was involved in a European stakeholder roundtable on biologics and biosimilars sponsored by Amgen, Merck Sharp and Dohme, and Pfizer; he has participated in advisory board meetings for Amgen and Pfizer; and he has contributed to studies on biologics and biosimilars for Celltrion, Hospira, Mundipharma, and Pfizer. S.S. is also member of the leadership team of the International Society for Pharmacoeconomics and Outcomes Research Special Interest Group on Biosimilars.

### **Appendix A**


• Are you aware of any specific, direct benefits from biosimilars entering the market?

*Pharmaceuticals* **2020**, *13*, x FOR PEER REVIEW 16 of 19

### **Appendix B**

**Appendix B**

review.

Savings (implications for health system sustainability)

**Appendix C**

**Figure A1.** PRISMA Flow Diagram. RCT, randomized controlled trial. SLR, systematic literature **Figure A1.** PRISMA Flow Diagram. RCT, randomized controlled trial. SLR, systematic literature review.

Sustainable competition Systematic literature review

Systematic literature review/targeted

Access and pricing Targeted additional search Moorkens E, et al. (2017) [17]

Procurement/purchasing Review of tender documents (2018) Vulto A, et al. (2019) [18] Patient safety/use Targeted additional search Tabernero J, et al. (2016) [4]

additional search Vulto A, et al. (2019) [18]

Mestre-Ferrandiz J, et al. (2016) [16] Dave CV, et al. (2017) [25] Dave CV, et al. (2018) [26]

Kawalec P, et al. (2017) [27]

EULAR PARE (2018) [3]

### **Appendix C**


### **Table A2.** Key literature identified during brainstorming (Delphi process stage 1).

### **Appendix D**


### **Table A3.** Themes and statements (Delphi process stage 2).

### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Analysis of the Regulatory Science Applied to a Single Portfolio of Eight Biosimilar Product Approvals by Four Key Regulatory Authorities**

**Beverly Ingram 1,\*, Rebecca S. Lumsden <sup>2</sup> , Adriana Radosavljevic <sup>1</sup> and Christine Kobryn <sup>3</sup>**


**Abstract:** Slow uptake of biosimilars in some regions is often attributed to a lack of knowledge combined with concerns about safety and efficacy. To alleviate physician and patient apprehensions, regulatory reviews from four major regulatory authorities (RAs) (European Medicines Agency, US Food and Drug Administration, Health Canada, and Japan Pharmaceuticals and Medical Devices Authority) across a portfolio of eight biosimilars were analyzed to provide insight into RA review focus and approach. RA queries were evaluated in an unbiased and systematic manner by major classification (Chemistry, Manufacturing and Controls [CMC], nonclinical, clinical or regulatory) and then via detailed sub-classification. There was a consistent, predominant focus on CMC from all RAs. The review focus based on sub-classification of clinical and regulatory queries was influenced by molecular complexity, with significant differences between categories (monoclonal antibody or protein) in the distribution of query topics; specifically, bioanalytical (*p* = 0.023), comparative safety and efficacy (*p* = 0.023), and statutory (including the justification of extrapolation) (*p* = 0.00033). Each biosimilar had a distinct distribution of clinical query topics, tailored to product-specific data. This analysis elucidated areas of heightened RA interest, and validated their application of regulatory science in the evaluation of biosimilar safety and efficacy.

**Keywords:** biosimilars; regulatory; review; approval; clinical; queries; regulatory science

### **1. Introduction**

Biosimilars represent an increasingly important option in the delivery of high-quality treatments for patients and offer the potential to address one of the greatest access constraints to biologics globally, namely price [1–3]. Since the first biosimilar was approved in 2006 by the European Medicines Agency (EMA) a dedicated regulatory framework for such products has spread rapidly across the world, with biosimilar-specific regulatory paradigms currently established in over 20 countries [4].

The requirement to establish dedicated biosimilar-specific regulatory paradigms by regulatory authorities (RAs) is well documented and is necessary since biosimilars cannot be safely regulated by the pathway used for typical 'small molecule' generic drugs [5,6]. The inherent variation of biological systems means that biosimilars cannot be manufactured to be identical to the originator biologic reference product (i.e., reference product) but are instead structurally and functionally "highly similar" [7]. Building and expanding on scientific principles and methodologies established for novel biologics (i.e., concepts outlined in the International Council for Harmonisation Q5E [8]), the EMA issued the first dedicated biosimilar-specific guidance in 2005. This was followed by the World Health Organization (2009), The Japanese Ministry of Health, Labour and Welfare (2009), Health Canada (HC) (2010) and the US Food and Drug Administration (FDA) (2012) (Initial draft

**Citation:** Ingram, B.; Lumsden, R.S.; Radosavljevic, A.; Kobryn, C. Analysis of the Regulatory Science Applied to a Single Portfolio of Eight Biosimilar Product Approvals by Four Key Regulatory Authorities. *Pharmaceuticals* **2021**, *14*, 306. https:// doi.org/10.3390/ph14040306

Academic Editors: Arnold G. Vulto, Steven Simoens and Isabelle Huys

Received: 18 December 2020 Accepted: 25 March 2021 Published: 1 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

overarching guidance was published in 2012; final guidance was published in 2018). Although developed at different times, these guidances share the same fundamental scientific approach to establishing biosimilarity [9,10]. Major regulators such as the FDA, EMA, HC, and the Pharmaceuticals and Medical Devices Authority (PMDA) have leveraged cross-communication, such as health authority cluster meetings, in order to share learning and foster greater consistency, due to the rapid pace at which the regulatory science has evolved [11]. overarching guidance was published in 2012; final guidance was published in 2018). Although developed at different times, these guidances share the same fundamental scientific approach to establishing biosimilarity [9,10]. Major regulators such as the FDA, EMA, HC, and the Pharmaceuticals and Medical Devices Authority (PMDA) have leveraged cross-communication, such as health authority cluster meetings, in order to share learning and foster greater consistency, due to the rapid pace at which the regulatory science has evolved [11].

The concept of biosimilar development is underpinned by both established scientific knowledge and the application of regulatory science during the assessment by RAs [11,12]. The extent and type of the data required, and the studies conducted during biosimilar development, to meet the regulatory requirements for biosimilarity differ from those required for novel biologics, both in their design and the relative emphasis of contributing parts (Figure 1) [13–15]. RAs also have discretion, as per their respective regulatory guidelines, to determine whether some nonclinical and clinical studies are not required; for example, animal studies may be conducted if residual uncertainties remain following the analytical assessment that need to be resolved prior to conducting a comparative clinical trial [16–18]. The concept of biosimilar development is underpinned by both established scientific knowledge and the application of regulatory science during the assessment by RAs [11,12]. The extent and type of the data required, and the studies conducted during biosimilar development, to meet the regulatory requirements for biosimilarity differ from those required for novel biologics, both in their design and the relative emphasis of contributing parts (Figure 1) [13–15]. RAs also have discretion, as per their respective regulatory guidelines, to determine whether some nonclinical and clinical studies are not required; for example, animal studies may be conducted if residual uncertainties remain following the analytical assessment that need to be resolved prior to conducting a comparative clinical trial [16–18].

**Figure 1.** Development Aim and Impact on Regulatory Information for Novel Biologics and Biosimilars (Adapted from Biosimilar Development and Approval in the EU, European Medicines Agency [15]). <sup>a</sup> While CMC information for a novel biologic is focused solely on the target product, the corresponding information for a biosimilar is highly comparative, with additional content required focusing on both the biosimilar and the reference product. CMC, Chemistry, Manufacturing and Controls; EU, European Union; PD, pharmacodynamic; PK, pharmacokinetic(s); RA, Regulatory Agency. **Figure 1.** Development Aim and Impact on Regulatory Information for Novel Biologics and Biosimilars (Adapted from Biosimilar Development and Approval in the EU, European Medicines Agency [15]). <sup>a</sup> While CMC information for a novel biologic is focused solely on the target product, the corresponding information for a biosimilar is highly comparative, with additional content required focusing on both the biosimilar and the reference product. CMC, Chemistry, Manufacturing and Controls; EU, European Union; PD, pharmacodynamic; PK, pharmacokinetic(s); RA, Regulatory Agency.

> The demonstration of similarity is first and foremost required at a molecular level, by the application of a number of in vitro analytical techniques [19]. These analytical studies are extensive and form the foundation for establishing similarity [13,16,20]. Hence, biosimilar development is focused on the production of a similar molecule to the reference product, analytical (in vitro) assessments that support demonstration of similarity, and manufacturing controls that ensure similarity is maintained [17]. All of these aspects are complemented by a targeted clinical study, sensitive enough with regard to the design, conduct, endpoints and/or population to detect differences should they exist with the reference product [21]. Demonstration of biosimilarity is based on careful consideration of the totality of the information provided [13]. The demonstration of similarity is first and foremost required at a molecular level, by the application of a number of in vitro analytical techniques [19]. These analytical studies are extensive and form the foundation for establishing similarity [13,16,20]. Hence, biosimilar development is focused on the production of a similar molecule to the reference product, analytical (in vitro) assessments that support demonstration of similarity, and manufacturing controls that ensure similarity is maintained [17]. All of these aspects are complemented by a targeted clinical study, sensitive enough with regard to the design, conduct, endpoints and/or population to detect differences should they exist with the reference product [21]. Demonstration of biosimilarity is based on careful consideration of the totality of the information provided [13].

> The biosimilar developer can seek approval of their product for other authorized indications of the reference product via extrapolation of similarity. This is a scientific and regulatory principle that is applied without the need to conduct a comparative clinical study in the extrapolated disease indication(s) [22]. Without this facility, biosimilar development would not follow an abbreviated pathway [23]. Extrapolation of efficacy and The biosimilar developer can seek approval of their product for other authorized indications of the reference product via extrapolation of similarity. This is a scientific and regulatory principle that is applied without the need to conduct a comparative clinical study in the extrapolated disease indication(s) [22]. Without this facility, biosimilar development would not follow an abbreviated pathway [23]. Extrapolation of efficacy and safety data from one indication to another is not a given; it must be thoroughly scientifically justified,

based on data that indicates certain properties of the originator, such as the mechanism of action, PK and immunogenicity, are consistent between the indications [24].

Despite many biosimilars now approved by the EMA [25] and a growing number of biosimilars authorized by the FDA [26], barriers remain to their adoption and use in clinical practice, driven by several issues, including concerns among healthcare providers and patients over their safety and effectiveness [27]. These reservations suggest that gaps may exist between the extent of the evidence required for biosimilars to gain RA approval and the evidence needed to achieve wider acceptance and use by physicians and patients [28]. When the development components and supporting data of a novel biologic and biosimilar are compared (Figure 1), the unique aspects of biosimilar development are revealed as one potential root cause of this gap. The use of an expanded analytical assessment, together with targeted clinical data obtained in a sufficiently sensitive patient population (with justification of extrapolation for additional indications), in place of the more extensive clinical data required for novel biologics, is the foundation of biosimilarity and of the scientific benefit–risk considerations applied by RAs [11].

Pfizer has established a portfolio of biosimilars, which differ in their molecular complexity and span disease indications in inflammation and oncology (including supportive care). Proactive engagement with RAs occurred throughout each product's development (via advice procedures) to ensure alignment with expectations and requirements. The RA advice from multiple agencies was incorporated into the respective product's development to inform a global development strategy. This permitted a global dossier preparation and submission approach, whereby the same data for each biosimilar were used to support all submissions (with the inclusion of additional/alternative data to meet a limited number of country-specific requirements). This strategy, and the breadth and extent of regulatory submissions, provides a unique opportunity to analyze the focus of RA review and gain an understanding of the approaches applied by different regulatory bodies in ensuring the requirements for biosimilarity are met. We conducted an analysis of the queries received from multiple RAs in response to license applications for this portfolio of biosimilars. We aimed to bring a greater awareness and appreciation of the RA approach, scientific consistency, and reviewer focus during biosimilar review, to increase confidence in the safety and effectiveness of biosimilars amongst physicians and patients [29].

### **2. Results**

A total of 2438 queries were received from the FDA, EMA, PMDA, and HC in relation to 21 applications for the eight biosimilars. Except for two queries relating to legal matters received from the PMDA, all other queries were retained and included in the analysis.

CMC was the largest category of query assignments received from the FDA (83%), EMA (66%) and PMDA (58%). For HC, 41% of queries were assigned to CMC, which were comparable in number to those assigned to the regulatory category (Figure 2). CMC queries encompassed data supporting the comprehensive in vitro comparative analysis of the biosimilar and its reference product, as well as manufacturing details and quality control aspects, while those assigned to the regulatory category included those focused on the relevance of the reference product, justification of extrapolation of indications, and labeling topics. Nonclinical queries, which related to the limited in vivo studies required, comprised 0.3% or fewer of the overall number of queries received from each RA.

A main focus on CMC-related information was also reflected in the queries associated with the individual biosimilars received from the FDA, EMA and PMDA, which was maintained throughout the duration of review period covered by the first and most recent biosimilar to be authorized (Figure 3A–C). Emphasis on the regulatory classification by HC was apparent across all four biosimilars (Figure 3D). Closer evaluation of the queries assigned to this category for HC found that the majority (90%) were related solely to labeling, with CMC queries representing 65% of the overall share when labeling queries were not included. In contrast, both the FDA and EMA directed the lowest proportion of queries towards regulatory topics, comprising 5% and 3%, respectively, of the total queries received from each RA (Figure 2), irrespective of the particular biosimilar (Figure 3A,B).

*Pharmaceuticals* **2021**, *14*, x FOR PEER REVIEW 4 of 17

**Figure 2.** Regulatory Agency Queries Overall (FDA [*n* = 1397]), EMA [*n* = 791], PMDA [*n* = 608], and HC [*n* = 640]) by Major Classification. CMC, Chemistry, Manufacturing and Controls; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PMDA, Pharmaceuticals and Medical Devices Agency. Number of biosimilars: <sup>a</sup>*n* = 8, <sup>b</sup>*n* = 5, <sup>c</sup>*n* = 4. **Figure 2.** Regulatory Agency Queries Overall (FDA [*n* = 1397]), EMA [*n* = 791], PMDA [*n* = 608], and HC [*n* = 640]) by Major Classification. CMC, Chemistry, Manufacturing and Controls; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PMDA, Pharmaceuticals and Medical Devices Agency. Number of biosimilars: <sup>a</sup> *n* = 8, <sup>b</sup> *n* = 5, <sup>c</sup> *n* = 4. *Pharmaceuticals* **2021**, *14*, x FOR PEER REVIEW 5 of 17

**Figure 3.** Major Classification Queries by Biosimilar across the (**A**) FDA (*n* = 1397), (**B**) EMA (*n* = 791), (**C**) PMDA (*n* = 608), and (**D**) HC (*n* = 640). CMC, Chemistry, Manufacturing and Controls; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PMDA, Pharmaceuticals and Medical Devices Agency. **Figure 3.** Major Classification Queries by Biosimilar across the (**A**) FDA (*n* = 1397), (**B**) EMA (*n* = 791), (**C**) PMDA (*n*= 608), and (**D**) HC (*n* = 640). CMC, Chemistry, Manufacturing and Controls; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PMDA, Pharmaceuticals and Medical Devices Agency.

related to the product mechanism of action.

Analysis of the CMC queries revealed that the RAs showed a consistent focus on

69%, respectively) and drug product (DP) content (38–51% and 22–47%, respectively) to a greater extent than analytical similarity (aspects regarding manufacturing and testing control were highly consistently in their inclusion). The FDA were uniquely interested in DP shipping validation information as part of their focus on DP control. Queries related to facilities/good manufacturing practices (GMP) represented <10% (0–3% and 1–7%, respectively) of the overall CMC queries for any individual biosimilar across both the FDA and EMA (Figures 4A,B). In contrast, the queries related to facilities/GMP represented a far higher share of the CMC queries arising from PMDA review of four biosimilars (19– 36%) (Figure 4C). Neither the EMA nor the PMDA conducted on-site inspections of manufacturing facilities as part of their review process, in contrast to the approach applied by the FDA and HC. Compared with the other RAs a higher proportion of queries related to DP were received from HC (38–80%), with between 23% and 70% of these being related to sample testing questions (namely detailed queries on how to conduct the analytical methods as well as data interpretation) across the biosimilars assessed. Analytical similarity was generally the least frequent CMC category amongst the HC (0–9%) and PMDA queries (2–3%). Extensive in vitro functional data was submitted and categorized under analytical similarity, which always received close attention by all RAs especially when it

*2.1. CMC Category*

For the three biosimilars that were assessed by all four RAs (Figure 3A–D), the analysis found a high focus on CMC topics (PF-trastuzumab [Trazimera™], 25–86% of queries; PFbevacizumab [Zirabev™], 38–86%; and PF-rituximab, 59–81%). While the FDA raised the highest number of queries overall for each biosimilar, the proportions of queries assigned to the clinical (1–23%) and regulatory (2–30%) categories were relatively low across products with respect to the proportion of CMC queries (67–92%). CMC also represented the most frequent category of queries received from the EMA, with the proportion overall (Figure 2) and by individual product being relatively lower than that seen for the FDA. Amongst the three biosimilars assessed by both RAs, the proportions of CMC queries for PF-bevacizumab were the same for the FDA and the EMA (86%).

### *2.1. CMC Category*

Analysis of the CMC queries revealed that the RAs showed a consistent focus on specific aspects, irrespective of molecular complexity or therapy area (Figure 4). The FDA and EMA showed a consistently high focus on both drug substance (DS) (31–54% and 37–69%, respectively) and drug product (DP) content (38–51% and 22–47%, respectively) to a greater extent than analytical similarity (aspects regarding manufacturing and testing control were highly consistently in their inclusion). The FDA were uniquely interested in DP shipping validation information as part of their focus on DP control. Queries related to facilities/good manufacturing practices (GMP) represented <10% (0–3% and 1–7%, respectively) of the overall CMC queries for any individual biosimilar across both the FDA and EMA (Figure 4A,B). In contrast, the queries related to facilities/GMP represented a far higher share of the CMC queries arising from PMDA review of four biosimilars (19–36%) (Figure 4C). Neither the EMA nor the PMDA conducted on-site inspections of manufacturing facilities as part of their review process, in contrast to the approach applied by the FDA and HC. Compared with the other RAs a higher proportion of queries related to DP were received from HC (38–80%), with between 23% and 70% of these being related to sample testing questions (namely detailed queries on how to conduct the analytical methods as well as data interpretation) across the biosimilars assessed. Analytical similarity was generally the least frequent CMC category amongst the HC (0–9%) and PMDA queries (2–3%). Extensive in vitro functional data was submitted and categorized under analytical similarity, which always received close attention by all RAs especially when it related to the product mechanism of action.

### *2.2. Clinical and Regulatory Sub-Classification*

On sub-classification of the clinical and regulatory queries by RA assigned in the major classification, queries from the FDA and EMA were more focused on bioanalytical aspects than on PK/PD or immunogenicity. A high proportion of those received from HC were assigned to labeling (62%), compared with 29% and 24% on this topic amongst those received from the PMDA and FDA, respectively (Figure 5). Labeling represented <5% of the clinical and regulatory queries received from the EMA. The apparent focus on labeling queries by HC and PMDA was further assessed to identify the share of queries directed towards the presentation of specific biosimilar data in the product label and monograph, or non-data-related queries (including formatting, use of reference product trade name vs. international nonproprietary name vs. biosimilar trade name, etc.). The majority of HC labeling queries were not related to the presentation of biosimilar-specific data in the product label, but on non-data-related queries, with 89% of labeling queries being focused on formatting (Supplementary Figure S1). Likewise, the PMDA reviews overall had only 1 out of 68 (1.5%) labeling queries directed towards biosimilar-specific content, with that query being editorial in nature.

**Figure 4.** CMC Queries by Biosimilar across the (**A**) FDA (*n* = 2766), (**B**) EMA (*n* = 1134), (**C**) PMDA (*n* = 588), and (**D**) HC (*n* = 318). CMC, Chemistry, Manufacturing and Controls; DP, drug product; DS, drug substance; EMA, European Medicines Agency; FDA, US Food and Drug Administration; GMP, good manufacturing practices; HC, Health Canada; PMDA, Pharmaceuticals and Medical Devices Agency. **Figure 4.** CMC Queries by Biosimilar across the (**A**) FDA (*n* = 2766), (**B**) EMA (*n* = 1134), (**C**) PMDA (*n* = 588), and (**D**) HC (*n* = 318). CMC, Chemistry, Manufacturing and Controls; DP, drug product; DS, drug substance; EMA, European Medicines Agency; FDA, US Food and Drug Administration; GMP, good manufacturing practices; HC, Health Canada; PMDA, Pharmaceuticals and Medical Devices Agency. *Pharmaceuticals* **2021**, *14*, x FOR PEER REVIEW 7 of 17

**Figure 5.** Sub-Classification of Regulatory (Including Labeling) and Clinical Queries by RA (FDA [*n* = 235]), EMA [*n* = 265], PMDA [*n* = 251], and HC [*n* = 377]). CSE, comparative safety and efficacy; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PK, pharmacokinetic; PMDA, Pharmaceuticals and Medical Devices Agency; RMP, riskmanagement plan. **Figure 5.** Sub-Classification of Regulatory (Including Labeling) and Clinical Queries by RA (FDA [*n* = 235]), EMA [*n* = 265], PMDA [*n* = 251], and HC [*n* = 377]). CSE, comparative safety and efficacy; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PK, pharmacokinetic; PMDA, Pharmaceuticals and Medical Devices Agency; RMP, riskmanagement plan.

There was no clear focus on any specific clinical and regulatory sub-category when

of the analysis appeared to reflect that safety/risk management plan (RMP) and labeling topics were of consistent interest to PMDA, with the former category comprising 37% of clinical and regulatory queries overall (Figure 6C). The highest frequency of label queries was observed with HC reviews (Figure 6D) comprising 62% of those received across all biosimilars submitted to this RA. Queries directed towards pharmacokinetic (PK)/pharmacodynamic (PD) data were relatively low across all four RAs (Figure 6A–D) and biosimilar products, ranging from 2% to 11%, while bioanalytical assays used to derive the clinical data, comprised 27% and 22% of the clinical and regulatory queries received from the FDA and EMA, respectively (Figure 6A,B). For the three biosimilars (PF-trastuzumab, PF-bevacizumab and PF-rituximab) assessed by all four RAs, each authority raised a different composition of queries on the clinical and regulatory content when presented with

essentially the same data (Figure 6A–D).

There was no clear focus on any specific clinical and regulatory sub-category when assessing the queries by biosimilar product for the FDA and EMA (Figure 6A,B). Results of the analysis appeared to reflect that safety/risk management plan (RMP) and labeling topics were of consistent interest to PMDA, with the former category comprising 37% of clinical and regulatory queries overall (Figure 6C). The highest frequency of label queries was observed with HC reviews (Figure 6D) comprising 62% of those received across all biosimilars submitted to this RA. Queries directed towards pharmacokinetic (PK)/pharmacodynamic (PD) data were relatively low across all four RAs (Figure 6A–D) and biosimilar products, ranging from 2% to 11%, while bioanalytical assays used to derive the clinical data, comprised 27% and 22% of the clinical and regulatory queries received from the FDA and EMA, respectively (Figure 6A,B). For the three biosimilars (PFtrastuzumab, PF-bevacizumab and PF-rituximab) assessed by all four RAs, each authority raised a different composition of queries on the clinical and regulatory content when presented with essentially the same data (Figure 6A–D). *Pharmaceuticals* **2021**, *14*, x FOR PEER REVIEW 8 of 17

**Figure 6.** Sub-Classification of Regulatory (Including Labeling) and Clinical Queries by Biosimilar from the (**A**) FDA (*n* = 235), (**B**) EMA (*n* = 265), (**C**) PMDA (*n* = 251), and (**D**) HC (*n* = 377). CSE, comparative safety and efficacy; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PK, pharmacokinetic; PMDA, Pharmaceuticals and Medical Devices Agency; RMP, risk-management plan. **Figure 6.** Sub-Classification of Regulatory (Including Labeling) and Clinical Queries by Biosimilar from the (**A**) FDA (*n* = 235), (**B**) EMA (*n* = 265), (**C**) PMDA (*n* = 251), and (**D**) HC (*n* = 377). CSE, comparative safety and efficacy; EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; PK, pharmacokinetic; PMDA, Pharmaceuticals and Medical Devices Agency; RMP, risk-management plan.

As indicated in Table 1, the eight biosimilars assessed in this analysis differ in their molecular complexity, which is reflected in their distinct development programs and the data accumulated to support their regulatory approval. They are approved for specific indications within oncology (including supportive care) and inflammatory disease. PFrituximab is approved for disease indications in both therapy areas. As indicated in Table 1, the eight biosimilars assessed in this analysis differ in their molecular complexity, which is reflected in their distinct development programs and the data accumulated to support their regulatory approval. They are approved for specific indications within oncology (including supportive care) and inflammatory disease. PFrituximab is approved for disease indications in both therapy areas.

(3, *n* = 3436) = 2.14, *p* = 0.54] was found to have a significant relationship

(6, *n* = 277) = 0.31, *p* = 1.0]. On the other hand, a chi-square test of independence

performed to examine the relationship between molecular complexity and clinical and regulatory sub-classification (Supplementary Figure S2B) found the relationship between

vidual clinical and regulatory sub-classifications confirmed that comparative safety and

(1, *n* = 1128) = 5.13, *p* = 0.023], bioanalytical [<sup>2</sup>

demonstrated significant relationships with molecular complexity.

(3, *n* = 1623) = 1.12, *p* = 0.78] nor molecular

(1, *n* = 1128) = 5.11, *p* =

(6, *n* = 1128) = 12.62, *p* = 0.049]. Assessment of the indi-

(1, *n* = 1128) = 12.88, *p* = 0.00033] query sub-classifications each

Across all RAs, neither therapy area [<sup>2</sup>

these variables to be significant [<sup>2</sup>

complexity [<sup>2</sup>

efficacy (CSE) [<sup>2</sup>

0.023], and statutory [<sup>2</sup>

icant [<sup>2</sup>



<sup>a</sup> Known the US as Nivestym® and in the RoW as Nivestim. <sup>b</sup> Known in the US and in the RoW as Retacrit®, with the products designated by the INNs epoetin alfa-epbx and epoetin zeta, respectively. <sup>c</sup> Known as Abrilada™ in the US and as Amsparity in the EU. <sup>d</sup> Not all biosimilars have been submitted in all domains. Some applications are ongoing. <sup>e</sup> Excluded from the analysis as the reference product was developed to include several different aspects in the US and EU (e.g., strength and presentation). <sup>f</sup> Licensing by the EMA was divested by Pfizer in February 2016. <sup>g</sup> Market authorization for reference epoetin alfa is held by two companies (Epogen®; Amgen Inc., Thousand Oaks, CA, USA and Procrit®; Janssen Products, LP, Horsham, PA, USA). <sup>h</sup> Approved for the inflammatory disease indications of granulomatosis with polyangiitis and microscopic polyangiitis in the US; not approved for rheumatoid arthritis in the US. EMA, European Medicines Agency; FDA, US Food and Drug Administration; HC, Health Canada; INN, international nonproprietary name; MAb, monoclonal antibody; PMDA, Pharmaceuticals and Medical Devices Authority; RoW, Rest of the World.

> Across all RAs, neither therapy area [χ 2 (3, *n* = 1623) = 1.12, *p* = 0.78] nor molecular complexity [χ 2 (3, *n* = 3436) = 2.14, *p* = 0.54] was found to have a significant relationship with major classification category (Supplementary Figure S2A). The relationship between therapy area and clinical and regulatory sub-classification was also found to be not significant [χ 2 (6, *n* = 277) = 0.31, *p* = 1.0]. On the other hand, a chi-square test of independence performed to examine the relationship between molecular complexity and clinical and regulatory sub-classification (Supplementary Figure S2B) found the relationship between these variables to be significant [χ 2 (6, *n* = 1128) = 12.62, *p* = 0.049]. Assessment of the individual clinical and regulatory sub-classifications confirmed that comparative safety and efficacy (CSE) [χ 2 (1, *n* = 1128) = 5.13, *p* = 0.023], bioanalytical [χ 2 (1, *n* = 1128) = 5.11, *p* = 0.023], and statutory [χ 2 (1, *n* = 1128) = 12.88, *p* = 0.00033] query sub-classifications each demonstrated significant relationships with molecular complexity.

### **3. Discussion**

In line with the foundational role analytical data plays in the biosimilar development pathway, this analysis established there was a consistently high focus by all RAs on CMC information across all biosimilars, irrespective of their molecular complexity or therapy area (Figure 2).

Analysis of the assignment of queries to CMC categories showed RAs were all highly focused on aspects related to the DS and/or DP including manufacturing/testing controls reflecting the interdependence of this content with analytical similarity. The DS and DP content includes the controls upon which analytical similarity is based; the level of

interest by RAs may stem from their aim to establish sufficient rigor will be applied by the manufacturer in maintaining analytical similarity throughout product development. The DS and DP content also includes the control measures to be applied to subsequent commercial manufacturing of the approved biosimilar, and again, review focus on this aspect would ensure similarity should be maintained in the future. The content of the DS and DP release specifications was an area of focus for all RAs, although the attributes and expectations were not identical. Despite the different queries received, the RA intent was clearly to ensure analytical similarity was measured and controlled to meet high expectations.

From the sub-classification of the clinical and regulatory queries there was a relatively greater focus on bioanalytical than on PK/PD or immunogenicity aspects by the FDA and EMA compared with the other RAs. This review approach was applied across all submitted products but not at consistent levels, suggesting it may have been influenced by individual reviewer preference. However, it should be noted that interrogation of the bioanalytical methods by reviewers can be considered an indirect assessment of the validity of the clinical (i.e., PK/PD or immunogenicity) data and this approach may be reflected in the findings for the FDA and EMA. During review, HC typically request visibility of FDA and EMA queries, if available, which may allow them to focus their review elsewhere on elements related to national regulatory (labeling) requirements. In the present analysis, based on the CMC categorization, the PMDA and HC both issued a higher proportion of queries related to facilities/GMP, compared with FDA and EMA, which is most likely due to differences in approval procedures rather than fundamental differences in GMP expectations. Both the FDA and HC review procedures can include site inspections; this only occurs during EMA and PMDA application assessments if a site has not previously been inspected within an acceptable time frame. HC was the only RA of the four covered in this analysis that routinely conducted sample testing as part of their review, which required DP samples and information on the analytical testing method to be provided. The high proportion of DP-related queries issued by HC amongst the CMC categorization across biosimilars could be attributed to sample testing activities, including those related to the transfer of analytical testing methods.

One area that received little attention from the RAs was nonclinical information, comprising <0.3% of queries in our analysis, which is consistent with the less significant role such studies play in biosimilar development compared with a novel biologic. It also supports the ongoing regulatory focus on the principle of the 3Rs (Reduce, Replace, Refine) that has been applied in updates to the earlier biosimilar guidance in some regions [16]. The low number of nonclinical queries received from the RAs that require animal studies (i.e., FDA, PMDA) supports the view that little or no nonclinical information should be necessary for demonstrating biosimilarity [30].

The proportion of clinical queries overall was low compared with requests for CMC information. Differences in the distribution of CMC and clinical queries between RAs is not unexpected and may reflect the experience of the RA authority or reviewer with biosimilars, their expectations, and/or discrete approach to reviewing data. Analysis of the query topics within the clinical sub-classification demonstrated alignment in the focus of the reviews and the principles of the scientific guidance for biosimilars across all four RAs (Figure 1). Analysis of the distribution of clinical queries following sub-classification revealed that the RA questions were tailored to individual biosimilars. Moreover, no prescribed review approach was evident from any of the RAs, with a different composition and distribution of queries being received on clinical topics for each biosimilar. Based on sub-classification of the clinical queries, there was no evidence that the tailored RA review approach was influenced by the therapy area (oncology or inflammation) of the biosimilar. However, the molecular complexity of the biosimilar showed significant association with queries subclassified to bioanalytical, CSE, and statutory topics (which encompassed biosimilar-specific requirements, including the justification of reference product selection and extrapolation of indications).

Amongst the clinical and regulatory queries following sub-classification, a relatively low proportion were directed towards PK/PD data for all four RAs (Figure 6A–D). This finding was somewhat surprising since comparative PK studies are considered a key component of biosimilar development, due to their role in addressing residual uncertainties arising from the analytical assessment and in establishing there will be no clinically meaningful differences between the proposed biosimilar and reference product. They are also key to guiding the requirement for, and nature of, subsequent comparative clinical studies [31]. The proportion of PK/PD queries was unaffected whether the study was conducted in patients (such as for PF-rituximab [32]) or in healthy subjects (PF-infliximab [Ixifi™] [33], PF-epoetin [34,35], PF-filgrastim [36], PF-trastuzumab [37], PF-bevacizumab [38], PFadalimumab [Abrilada™/Amsparity] [39], PF-pegfilgrastim [Nyvepria™] [40]). The generally low focus on PK information across RAs may reflect the proactive engagement of the applicant with the relevant RA during the study design process, the transparency of the data disclosure and interpretation. The approach to review of the clinical information was often two-pronged; indirect, via queries related to bioanalytical assays that supported the clinical study conclusions (PK/PD, CSE, and immunogenicity), and direct, with questions focused on the clinical data. In particular, the FDA and EMA appeared to favor focusing on bioanalytical assays as an indirect assessment tool to complement assessment of the data generated in clinical studies (PK/PD, CSE, and immunogenicity) (Figure 5). In the case of the FDA this included routinely requesting internal method validation reports and sample management records.

In general, neither the EMA nor the FDA allow inclusion of clinical data for the biosimilar on the product label, but instead use the data already provided in the reference product label [41,42]. In contrast, the PMDA and HC permit certain data for the biosimilar generated in comparative clinical trials to be included in the label [43,44]. It might be expected that these differences between RAs in relation to inclusion of data in the product label would be reflected in the distribution of the labeling and statutory topics following sub-classification of the regulatory and clinical queries. While our analysis found that both HC and the PMDA placed a greater focus on labeling queries compared with the FDA and EMA, in-depth analysis of their labeling questions determined that biosimilar data-related queries comprised only 11% and 2% of the overall labeling queries received from HC and the PMDA, respectively. This suggests that there may be opportunity for RAs to provide more operationally focused labeling guidance for biosimilars to reduce their effort and resources on raising queries on this topic.

The biosimilar guidance and regulatory requirements for the four RAs covered by this analysis were largely aligned, with only minimal divergence to meet country-specific content requirements. This allowed for submission of consistent content across the different RAs. Each RA showed a broadly similar approach in implementing their guidance, which was evident in their high focus on CMC content (ranging, on average from 41–83%), minimal nonclinical queries (<0.3%), and between 17% and 59% of queries directed to clinical and regulatory topics combined. Comparison of the clinical and regulatory subclassification queries of the three biosimilars assessed by all four RAs suggested a tailored approach to their review, focused on different topics and with varying frequency across RAs for each biosimilar. Despite aligned guidance and shared high-level expectations across RAs on biosimilar development, the differences in the review focus highlighted here may reflect RA-specific guidance implementation approaches with regard to sub-classification content. For the biosimilars assessed by more than one RA, while their review decision was the same, with approval granted for all requested indications via the justification of extrapolation, their approach to data review and assessment of benefit–risk differed. No common review strategy was observed across the four RAs; however, the approach for each RA reflected a robust assessment of data, and the justification of that data to support biosimilarity and extrapolation of indications. Interestingly, while the RAs were equally effective, the level of review of the sub-classification topics may reflect their differing implementation of the biosimilarity concept.

The biosimilars assessed in this analysis formed a single portfolio of products developed following a global strategy, guided by proactive RA engagement, whereby the same data for each biosimilar were used to support all submissions. Our analysis revealed that the RAs do not have a pre-set approach to data review and their focus is influenced by the individual submission data per product. Although we applied the same global development strategies, the data provided was unique to each product. Therefore, it is anticipated that when assessing other biosimilars of the same reference products covered in our analysis, RAs will focus on the same topics (according to the classification and categorization system used here).

Not all biosimilars were submitted to all RAs. The data submitted to the RAs reflected product-specific information and the particular studies conducted during the biosimilar development program (e.g., comparative PK/PD and immunogenicity for PF-filgrastim and PF-pegfilgrastim were derived from studies conducted in healthy volunteers, in lieu of CSE; data for delivery by autoinjector formed part of the submission package for PFadalimumab).

The volume of queries served solely as a proxy measure of areas of the biosimilar product dossier that received attention by the RAs. It was not possible to weight the queries to reflect the importance of the questions to the regulators or the complexity of the response required.

### **4. Materials and Methods**

From 2017 to November 2020, 21 regulatory submissions for market authorization were undertaken for eight biosimilars in one or more of the major domains comprising the USA, EU, Japan, or Canada, in advance of cascading to further submissions globally. Contemporaneous submissions of the same data content for each biosimilar allows comparison of the review approaches between four RAs: US FDA, EMA, PMDA, and HC. These RAs were selected due to the consistency of their biosimilar guidance and their regulatory requirements, as well as the leading role their guidance has played in shaping regulatory expectations in other countries.

Details of the biosimilars included are provided in Table 1, together with the approval dates in these four RA domains. To ensure comparisons were based on consistent submission content and time frame, EMA approvals of the filgrastim (PF-filgrastim; Nivestim) and epoetin (PF-epoetin: Retacrit®) biosimilars in 2010 and 2007, respectively, were excluded from this analysis. The reference products, Neupogen® and Epogen®/Procrit®, respectively, were developed to include several different aspects in the US and EU (e.g., strength and presentation), and the biosimilars were developed to align with the EMA-approved reference products. As a result, the differences in submission information between EU and US meant that direct comparison could not be made.

Queries received during the course of review of all eight biosimilars by the four RAs were collated and each was categorized retrospectively by the authors using the methods of classification outlined below. Any uncertainties in category assignments were resolved by author calibration meetings. In instances where a query included subsidiary questions (e.g., query X, parts i-v) the main query and subsidiary components were counted separately. In some cases, a query related to a single topic, while others covered multiple topics (e.g., one CMC query could have included aspects that related to both DS and DP). In such instances, a single question could be assigned to more than one sub-classification category.

### *4.1. Major Classification*

Initially, queries were assigned to one of four major categories: CMC, nonclinical, clinical, and regulatory (encompassing all statutory requirements such as labeling and justification of extrapolation of indications) (Table 2). The relative frequencies of each category were determined as a percentage of the overall number of queries received from

each RA and for each biosimilar. There were relatively few queries assigned to the nonclinical category. Therefore, queries assigned to this category were not sub-classified further.

**Table 2.** Query Classification and Category Assignment Criteria.


<sup>a</sup> Queries received from HC and PMDA assigned to labeling/monograph sub-class were further subdivided by assignment to either data (queries related to the presentation of biosimilar data from comparative clinical studies) or to format (queries related to label text and unrelated to biosimilar clinical data). CMC, Chemistry, Manufacturing and Controls; EMA, European Medicines Agency; EU, European Union; FDA, US Food and Drug Administration; GMP, good manufacturing practice(s); HC, Health Canada; PD, pharmacodynamic(s); PK, pharmacokinetic(s); PMDA, Pharmaceuticals and Medical Devices Authority; RMP, risk management plan; SmPC, Summary of Product Characteristics; US PI, United States Prescribing Information.

### *4.2. CMC Sub-Classification*

All queries assigned to the CMC category in the major classification were further assigned to four CMC categories according to the criteria in Table 2. Since individual CMC queries may have not have been related to a single topic they could be assigned to more than one CMC category. Relative frequencies of each topic were determined as a percentage of the total volume of CMC queries by RA and by biosimilar.

### *4.3. Clinical and Regulatory Sub-Classification*

All queries that were assigned to the clinical and regulatory categories in the major classification were further sub-classified to one of seven categories according to the assignment criteria in Table 2. Relative frequencies of each category were determined as a percentage of the total number of queries received from each RA.

Labeling queries received from HC and PMDA were further subdivided by assignment to either 'data' (queries related to the presentation of biosimilar data from comparative clinical studies) or 'format' (queries related to label text and unrelated to biosimilar clinical data).

### *4.4. Statistical Analysis*

Chi-square tests of independence were performed between molecular complexity (monoclonal antibody [MAb] or protein) and therapy area (oncology or inflammation) versus major query classification for each RA. The rituximab biosimilar (PF-rituximab; Ruxience™) was included under inflammation and oncology therapy areas in this analysis since it is approved for disease indications in both.

Chi-square tests of independence were also performed on the basis of molecular complexity and query sub-classification, as well as on the basis of therapy area and query sub-classification.

### **5. Conclusions**

Analysis of the focus of the FDA, EMA, HC, and PMDA review of the biosimilars described here gives an indication of the practical application of the regulatory science underpinning the robust regulatory standards that exist in the countries and region served by these RAs. The distinct distribution of queries received for three biosimilars assessed by all four RAs may reflect a different approach in assessing benefit–risk, while still ultimately reaching the same regulatory decision.

Analysis of the focus of RAs on specific query topics identified areas of heightened interest and gave some insight as to their significance. When provided with essentially the same data, aside from country-specific content, all four RAs focused primarily on CMCrelated topics, irrespective of the molecular complexity or therapy area of the biosimilar. The level of focus on CMC information was consistent with the fundamental importance of data in this domain to the demonstration of similarity, as the basis for extrapolation of indications, and to the controls applied to biosimilar manufacturing and testing.

The clinical and regulatory data review was tailored and product-specific, irrespective of therapy area, but the focus of the queries based on their sub-classification was significantly associated with the category of molecular complexity. Nevertheless, the proportion of queries on clinical topics overall was relatively low, confirming that the information from clinical studies is deemed by RAs to be largely supportive in demonstrating biosimilarity. The greatest area of RA focus was consistently placed on the assessment of data that represented the most sensitive information in the demonstration of biosimilarity, namely CMC, and the justification for extrapolation of indications.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/ph14040306/s1, Figure S1: Sub-classification of Labeling and Monograph Queries for HC, Figure S2: Major Classification Queries and Sub-classification of Regulatory (Including Labeling) and Clinical Queries by Molecular Complexity for all RAs, Table S1: Relationship of therapy area and molecular complexity with major classification, and clinical and regulatory sub-classification.

**Author Contributions:** All authors were involved in the conception and design of the study. B.I. and C.K. were involved in the assignment of the queries (except for those received from the PMDA and for sub-classification of the CMC queries). All authors were involved in analysis and interpretation of the data. All authors were involved in the drafting of and revising the manuscript, and approved the final version for submission. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was funded by Pfizer.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data are contained within the article or supplementary material.

**Acknowledgments:** The authors wish to thank their regulatory colleagues in Japan, Yuko Yao and Ayako Ohno, for supporting the assignment of PMDA queries. The authors also thank Pei-Li Wang for advice on the statistics analysis, and their colleagues Heather Hufnagel, Scott Tennyson, and Lisa LeSueur for assistance in assignment of the CMC queries. Medical writing support was provided by Iain McDonald of Engage Scientific Solutions and was funded by Pfizer.

**Conflicts of Interest:** Beverly Ingram, Rebecca S. Lumsden, and Christine Kobryn are full-time employees, and hold stock or stock options in Pfizer. Adriana Radosavljevic participated in this work while completing a student internship at Pfizer.

### **References**


### *Article* **Type and Extent of Information on (Potentially Critical) Quality Attributes Described in European Public Assessment Reports for Adalimumab Biosimilars**

**Ali M. Alsamil 1,2, Thijs J. Giezen 3,4, Toine C. Egberts 1,5 , Hubert G. Leufkens <sup>1</sup> and Helga Gardarsdottir 1,5,6,\***


**Abstract:** Regulatory approval of biosimilars predominantly relies on biosimilarity assessments of quality attributes (QAs), particularly the potentially critical QAs (pCQAs) that may affect the clinical profile. However, a limited understanding exists concerning how EU regulators reflect the biosimilarity assessments of (pC)QAs in European public assessment reports (EPARs) by different stakeholders. The type and extent of information on QAs and pCQAs in EPARs were evaluated for seven adalimumab biosimilars. Seventy-seven QAs, including 31 pCQAs, were classified and assessed for type (structural and functional attributes) and extent (biosimilarity interpretation and/or test results) of information in EPARs. Reporting on the QAs (35–75%) varied between EPARs, where the most emphasis was placed on pCQAs (65–87%). Functional attributes (54% QAs and 92% pCQAs) were reported more frequently than structural attributes (8% QAs and 22% pCQAs). About 50% (4 structural and 12 functional attributes) of pCQAs were consistently reported in all EPARs. Regulators often provided biosimilarity interpretation (QAs: 83% structural and 80% functional; pCQAs: 81% structural and 78% functional) but rarely include test results (QAs: 1% structural and 9% functional and pCQAs: 3% structural and 9% functional). Minor differences in structural attributes, commonly in glycoforms and charge variants, were often observed in adalimumab biosimilars but did not affect the functions and clinical profile. Despite the variability in reporting QAs in EPARs, the minor observed differences were largely quantitative and not essentially meaningful for the overall conclusion of biosimilarity of the seven adalimumab biosimilars.

**Keywords:** adalimumab; biosimilar; biosimilarity assessment; quality attributes (QAs); potentially critical quality attributes (pCQAs); European public assessment reports (EPARs)

### **Highlights**


**Citation:** Alsamil, A.M.; Giezen, T.J.; Egberts, T.C.; Leufkens, H.G.; Gardarsdottir, H. Type and Extent of Information on (Potentially Critical) Quality Attributes Described in European Public Assessment Reports for Adalimumab Biosimilars. *Pharmaceuticals* **2021**, *14*, 189. https://doi.org/10.3390/ph14030189

Academic Editor: Steven Simoens

Received: 31 December 2020 Accepted: 22 February 2021 Published: 25 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

though this had no effect on the functions and clinical profiles and did not preclude biosimilarity.

• Regulators provided a biosimilarity interpretation but rarely reported test results for QAs in EPARs, impeding the interpretation by EPAR users.

### **1. Introduction**

Biological drugs have become important treatment options for numerous diseases, including cancer and inflammatory diseases [1]. After patent expiration of the reference biologicals, biosimilars contribute to improved patient access to treatment due to competition, resulting in lower prices. Unlike small molecule drugs, biological drugs, including biosimilars, are large and complicated molecules produced through a complex process using living microorganisms. Variability within and between batches is an inherent feature of the production of biologicals [2,3]. Therefore, biosimilars are, generally, not exact replications of the reference biological but are highly similar [4].

The leading regulatory and health authorities in highly regulated markets, such as the European Medicines Agency (EMA), the United States Food and Drug Administration (US FDA), and the World Health Organization (WHO), have established frameworks and guidelines for the development, assessment, and approval of biosimilars [5–8]. Biosimilar development and regulatory approval predominantly rely on demonstrating the biosimilarity to the reference biological, which involves a stepwise comparability assessment. The comparability assessment of quality attributes (QAs) is a fundamental step, and it forms the basis for establishing biosimilarity and determining the scope and range of the in-vitro and clinical studies needed for biosimilar approval [9–12]. Minor differences in QAs between the biosimilar and reference biological may exist but should not be clinically relevant to obtaining regulatory approval.

Quality attributes are measurable molecular characteristics that describe the physical, chemical, biological, and microbiological properties of a drug molecule [13]. Some QAs are classified as potentially critical QAs (pCQAs) because they may affect the biological activity (potency) and the clinical drug profile, which includes pharmacokinetics (PK), pharmacodynamics (PD), safety, immunogenicity, and efficacy [14]. This criticality can be illustrated by a recent example where a biosimilar company discovered a drift in antibodydependent cell-mediated cytotoxicity (ADCC) activity due to shifts in afucosylated glycans of the reference biological trastuzumab [15], which was associated with a reduced eventfree survival rate [16]. Several studies have provided valuable insight into various risk assessment tools for identifying pCQAs [17–22]. Some pCQAs apply to all biologicals, but some pCQAs are specific to a biological and information about these may (d)evolve over time as more knowledge of the product and manufacturing process becomes available. The pharmaceutical industry generally defines which QAs are considered pCQAs based on the available information and the manufacturer risk assessment [23–32]. For biosimilars, the test results of all QAs must remain within the range of variability set by analyzing different batches of the reference biological. Scientific justification is needed if any deviation occurs in the QAs, especially in pCQAs. This rigorous assessment should also be followed when changes are introduced to the manufacturing processes of approved biologicals, including biosimilars [33–36].

Since the regulatory approval of the first biosimilar in Europe in 2006, 49 unique biosimilars marketed under 69 brand names for 15 reference biologicals have received a positive opinion from the EMA's Committee for Medicinal Products for Human Use (CHMP) as of November 2020 [37]. Currently, the reference biological adalimumab, sold under the brand name Humira® by AbbVie Corporation, USA, has the largest number of biosimilars approved in the EU market. Adalimumab is an anti-tumor necrosis factor-α (TNF-α) monoclonal antibody that prevents the interaction of TNF-α with its receptors and is indicated for the treatment of various immune-mediated inflammatory diseases [23,38,39].

Despite the established and stringent regulatory pathway of biosimilars in Europe, the adoption of biosimilars in clinical practice is challenged by a lack of knowledge and

understanding of the scientific rationale behind their approval [40–42]. In Europe, regulators have taken actions to increase transparency for the biosimilar approval process to improve stakeholder understanding of biosimilars through various communication media. The European public assessment report (EPAR) is an unbiased source through which the EMA publishes and broadcasts information to stakeholders about regulatory assessments for all medicinal products approved by the European Commission (EC) [37]. Previous studies have provided an in-depth overview of the clinical evidence reported in EPARs that supports approval of biosimilars in general [43,44] and approval of adalimumab biosimilars in particular [45]. These studies have shown that variations exist in reporting clinical data that confirm the biosimilarity of biosimilars to a reference biological, but they have not explored the reporting of the QAs that are the basis of biosimilar approval. The biosimilarity assessment of QAs is increasingly reported in scientific publications of biosimilars [46], which needed to be systematically consulted with the corresponding EPARs to obtain comprehensive information on biosimilarity at the quality level [47]. However, a limited understanding exists concerning how EU regulators reflect the biosimilarity assessment of (pC)QAs in EPARs by different stakeholders.

Therefore, this study aims to evaluate the QAs and pCQAs reported in EPARs using adalimumab biosimilars as a case study in terms of (1) consistency of QA and pCQA reporting between biosimilars of the same reference biological (i.e., adalimumab), (2) Type of the reported QAs and pCQAs (i.e., structural or functional attributes), and (3) how biosimilarity interpretation and test results were described for the reported (pC)QAs. We hypothesized that EU regulators are more focused in the reporting of pCQAs and the biosimilar interpretation because these are more likely to be of clinical relevance.

### **2. Results**

### *2.1. Characteristics of the Included European Public Assessment Reports of Adalimumab Biosimilars*

As of 30 November 2020, seven unique adalimumab biosimilars (11 brand names) had received marketing authorization from the EC. Three of the seven biosimilars (i.e., ABP501, GP2017, and MSB11022) were marketed under more than one brand name. Rapporteurs from 11 member states prepared the initial EPARs of the seven adalimumab biosimilars. Rapporteurs from two (Finland and Austria) of the 11 member states were involved in more than one EPAR of adalimumab biosimilars (Table 1).


**Table 1.** Characteristics of the included initial European public assessment reports (EPARs) of adalimumab biosimilars [48–58].

\* Solymbic®, Cyltezo® and Kromeya® were approved by the European Medicines Agency (EMA) but voluntarily withdrawn by the applicant for commercial reasons.

### *2.2. Types of Reported (Potentially Critical) Quality Attributes*

In general, the frequency of reported QAs (range: 27 (35%)–58 (75%)) varied between EPARs of adalimumab biosimilars, with most emphasis placed on the reporting of the pCQAs (range: 20 (65%)–27 (87%)). The proportion of reported pCQAs was comparable for all biosimilars. Overall, 16 (21%) of all QAs were reported in all EPARs of adalimumab biosimilars. Of the 31 pCQAs, 29 (94%) were reported at least in one EPAR, and 16 (52%) were consistently reported in all included EPARs (Table 2). Two (6%) pCQAs related to structural attributes were not reported in any included EPAR: post-translation modifications (PTMs) including neuraminic N-glycolyl acid and galactose alpha-1,3-galactose (Figure S1).

**Table 2.** Reporting of the quality attributes (QAs) and potentially critical quality attributes (pCQAs) stratified by structural and functional attributes and the company code of adalimumab biosimilars in the included European public assessment reports (EPARs).


Overall, functional attributes (54% QAs and 92% pCQAs) were more often consistently reported than structural attributes (8% QAs and 22% pCQAs) in EPARs of adalimumab biosimilars (Table 2). Consistent reporting of functional pCQAs was high, with 12 (92%) out of 13 pCQAs reported in all EPARs, including binding to soluble- and transmembrane-TNFα (s-TNFα and tm-TNFα), (ADCC), and complement-dependent cytotoxicity (CDC) activity and binding to complement component 1q (C1q), neonatal Fc receptor (FcRn), and six Fcγ-receptors. Of the 18 structural pCQAs, only four (22%) were consistently reported in all EPARs, including amino acid sequence and disulfide bridges, glycosylation, and aggregates (Figure S1).

### *2.3. Extent of Information on Reported (Potentially Critical) Quality Attributes*

In general, no differences were observed in the extent of the reported information between the QAs and pCQAs in all EPARs of adalimumab biosimilars. Regulators frequently provided biosimilarity interpretations of the reported QAs (83% structural and 80% functional) and pCQAs (81% structural and 78% functional) but rarely included test results with or without biosimilarity interpretations of the reported QAs (1% structural and 9% functional) and pCQAs (3% structural and 9% functional) (Figure 1).

The total number of reported QAs included with a biosimilarity interpretation in EPARs was 69 QAs and the number varied (range: 10–58 QAs) for adalimumab biosimilars. The interpretation of the biosimilarity of the reported QAs was most frequently reported as being similar (range: 7–44 QAs) than having minor differences (range: 1–18 QAs) (Table S1). Thirty-one QAs, including fifteen pCQAs, were observed with minor differences in at least one adalimumab biosimilar. The most common structural pCQAs with minor differences were the four glycoforms (galactosylated glycans, high mannose glycans, afucosylated glycans, and sialylated glycans) and the charge variants (acidic and basic variants). While functional pCQAs were more often similar between the biosimilar and reference biological, minor differences were observed for the functional pCQAs tm-TNFα

and PF-06410293 (Figure S1).

*Pharmaceuticals* **2021**, *14*, 189 5 of 15

*2.3. Extent of Information on Reported (Potentially Critical) Quality Attributes*

and 9% functional) and pCQAs (3% structural and 9% functional) (Figure 1).

In general, no differences were observed in the extent of the reported information between the QAs and pCQAs in all EPARs of adalimumab biosimilars. Regulators frequently provided biosimilarity interpretations of the reported QAs (83% structural and 80% functional) and pCQAs (81% structural and 78% functional) but rarely included test results with or without biosimilarity interpretations of the reported QAs (1% structural

The total number of reported QAs included with a biosimilarity interpretation in EPARs was 69 QAs and the number varied (range: 10–58 QAs) for adalimumab biosimilars. The interpretation of the biosimilarity of the reported QAs was most frequently reported as being similar (range: 7–44 QAs) than having minor differences (range: 1–18 QAs) (Table S1). Thirty-one QAs, including fifteen pCQAs, were observed with minor differences in at least one adalimumab biosimilar. The most common structural pCQAs with minor differences were the four glycoforms (galactosylated glycans, high mannose glycans, afucosylated glycans, and sialylated glycans) and the charge variants (acidic and basic variants). While functional pCQAs were more often similar between the biosimilar and reference biological, minor differences were observed for the functional pCQAs tm-TNFα binding, ADCC activity, and C1q binding in two adalimumab biosimilars: GP2017

### binding, ADCC activity, and C1q binding in two adalimumab biosimilars: GP2017 and PF-06410293 (Figure S1). biological (range = 5.3–12.0%), were interpreted by the regulators as minor difference. Figure S1 shows reporting of the type and extent of information on QAs and pCQAs described in the EPARs of adalimumab biosimilars included.

Regulators provided both biosimilarity interpretations and test results in EPARs for only five pCQAs, including the protein concentration and binding to FcγRIIIa for ABP501 and the high mannose glycans, ADCC activity, and binding to FcγRIIIa for MSB11022 (Table S2). Of those five pCQAs, only the test results of high mannose glycans, which were slightly lower in the MSB11022 biosimilar (range = 1.9–2.5%) compared to the reference

**Figure 1.** Comparison of the extent of reported information on quality attributes (QAs) and potentially critical quality attributes (pCQAs) stratified by the types of QAs and pCQAs (structural and functional) reported in all EPARs of adalimumab biosimilars included. **Figure 1.** Comparison of the extent of reported information on quality attributes (QAs) and potentially critical quality attributes (pCQAs) stratified by the types of QAs and pCQAs (structural and functional) reported in all EPARs of adalimumab biosimilars included.

> Regulators provided both biosimilarity interpretations and test results in EPARs for only five pCQAs, including the protein concentration and binding to FcγRIIIa for ABP501 and the high mannose glycans, ADCC activity, and binding to FcγRIIIa for MSB11022 (Table S2). Of those five pCQAs, only the test results of high mannose glycans, which were slightly lower in the MSB11022 biosimilar (range = 1.9–2.5%) compared to the reference biological (range = 5.3–12.0%), were interpreted by the regulators as minor difference. Figure S1 shows reporting of the type and extent of information on QAs and pCQAs described in the EPARs of adalimumab biosimilars included.

### **3. Discussion**

The present study evaluated the type and extent of information on QAs and pCQAs reported in EPARs by EU regulators for seven adalimumab biosimilars approved in Europe as of November 2020. In general, reporting of QAs (ranging from 27 (35%) to 58 (75%)) varied between EPARs of adalimumab biosimilars, where the most emphasis was on reporting pCQAs (ranging from 20 (65%) to 27 (87%)). About 50% (4 structural and 12 functional attributes) of pCQAs were consistently reported in all EPARs. Functional attributes (54% QAs and 92% pCQAs) were more frequently and consistently reported than structural attributes (8% QAs and 22% pCQAs). Minor differences between adalimumab biosimilars and the reference biological in certain structural attributes, most commonly in glycoforms and charge variants, were often observed by regulators. Regulators reported on the biosimilarity interpretation but rarely presented the test results underlying their interpretation in EPARs. However, QA and pCQA data not reported in the EPARs do not necessarily indicate that they were neither submitted by companies nor assessed by regulators during the stringent regulatory process.

This study highlights some variations in reporting biosimilarity assessments at the quality level in EPARs. Despite this variability in QA reporting, pCQAs were most frequently and consistently reported by EU regulators in EPARs. The variation in QA report-

ing between EPARs is consistent with the variability in reporting clinical data, which was explained by the flexibility in regulatory requirements (i.e., a case-by-case basis) [43,44]. However, such flexibility cannot explain the variability in reporting of QAs and pCQAs for biosimilars, particularly those containing the same active substance and compared to the same reference biological (e.g., Humira® in the case of adalimumab), that were assessed based on the same regulatory standards for establishing biosimilarity. The variability in QA reporting may be explained by the fact that the EPARs are prepared by various rapporteurs (i.e., regulators) from different member states. Nevertheless, regulators diligently reported the pCQAs, which are all considered to be of relevance because these may potentially affect functions (biological and immunochemical activity) and the clinical profile, including the pharmacokinetics, pharmacodynamics, safety, immunogenicity and efficacy of the drug. It is, however, important to note that learning on pCQAs is an ongoing process, which will likely result in changes to the current list over time.

The direct or indirect relationship between structural and functional QAs and the clinical profile influences the determination of pCQAs [19]. This relationship can be illustrated by the four structural pCQAs, including the amino acid sequence, disulfide bridges, aggregates, and glycosylation, which were consistently reported in EPARs. A mismatch in amino acid sequence and disulfide bridges can change the structural conformation affecting the biological activity and clinical performance, which were identical to the reference biological for all adalimumab biosimilars. Aggregates can elicit immunogenic responses by inducing neutralizing antibodies, hypersensitivity reactions, and infusion-related reactions in vivo. The propensity of aggregation may increase with some structural attributes (e.g., disulfide bridges, oxidation, and deamidation) if these are inadequately controlled. For all adalimumab biosimilars, aggregate levels were similar to the reference biological. Glycosylation is a PTM that occurs through an enzymatic process at specific sites in a protein drug and can influence the biological activity (potency and efficacy), serum half-life clearance (pharmacokinetics), and immunogenicity (safety). Minor differences in glycosylation were observed in adalimumab biosimilars, which are the most frequent notable differences in biosimilars and reference biologicals in general [9–12].

In practice, minor differences in QAs and pCQAs are expected for biosimilars due to the use of various manufacturing processes, cell lines, and materials [35]. These minor differences have also been observed between batches of a reference biological, primarily when a company introduces manufacturing changes [2,3,23]. The galactosylated glycans, high mannose glycans, afucosylated glycans, and sialylated glycans are types of glycoforms where minor differences have most commonly been reported (Figure S1). Galactosylated glycans may influence C1q binding and CDC activity, whereas high mannose glycans may influence pharmacokinetics parameters. However, structure-activity relationship studies and pivotal pharmacokinetics trials indicate that these are not affected by minor differences in galactosylated and high mannose glycans [48,49,51–57]. The same applies to afucosylated and sialylated glycans, which may influence Fcγ-receptors and ADCC activity [51–58]. These examples demonstrate the importance of structure-activity relationship studies and pharmacokinetics and pharmacodynamics trials in assessing the potential effect of minor differences in pCQAs in biosimilarity assessments. Minor differences in acidic and basic variants in several adalimumab biosimilars were attributed to changes in c-terminal lysin [48,49,51–54,58], which is generally cleaved in human serum with no effect on clinical profiles, and were thus considered noncritical QAs. Minor differences for certain functional pCQAs were attributed to minor differences in certain structural QAs and pCQAs, which were observed and reported by EU regulators in EPARs for GP2017 and PF06410293. For both biosimilars, the minor differences in ADCC activity disappeared when using an in-vitro assay with more physiological conditions in peripheral blood mononuclear cells. For GP2017, the aggregate levels were slightly higher using size-exclusion chromatography and slightly lower using analytical ultracentrifugation than the reference biological, which was considered a minor and clinically irrelevant difference by regulators. This ADCC and aggregate example indicates the importance of using orthogonal methods to assess the

(dis)similarity of QAs. Based on these observations, minor differences in these pCQAs seem to be quantitative (i.e., numerical values) but do not preclude the overall conclusion for biosimilarity and are considered clinically irrelevant.

The underlying reason functional pCQAs are more frequently and consistently reported in EPARs could relate to their direct relationship with the mechanisms of action (MoAs). The primary MoA of adalimumab involves binding to, and neutralizing TNF-α. Adalimumab also mediates effector functions, such as ADCC and CDC activity, by binding to tm-TNF-α, C1q (for CDC), and Fcγ-receptors. The relevance of ADCC or CDC activity to the primary MoA and efficacy of adalimumab is not well established but may be important, particularly in inflammatory bowel disease [45]. Binding to tm-TNFα can trigger potential biological functions known as "referred signaling," which may play a role in some therapeutic indications (e.g., inflammatory bowel disease). For GP2017, regulators reported minor differences in the binding to tm-TNFα, for which the scientific justifications provided by the company were not available in the EPAR for GP2017. However, the developer company of GP2017 reported functional and pharmacological characterizations demonstrating indistinguishable binding profiles and subsequent induction of reverse signaling to support the rationale for extrapolation across indications [28]. Therefore, functional pCQAs provide the final insight into the (dis)similarity at the quality level and useful information in predicting the outcomes of clinical studies [9–11], forming the basis for supporting the extrapolation of biosimilars across all indications authorized for the reference biological [59–62].

Regulators frequently describe the biosimilarity interpretation of reported QAs and pCQAs but rarely present the test result data, impeding the interpretation by EPAR users. For example, in EPARs, minor differences are frequently expressed subjectively as "slightly lower" or "slightly higher," but the exact extent to which the difference is minor remains unclear for most reported QAs and pCQAs. A more appropriate method would be in line with what was reported in the EPAR of MSB11022, in which the ranges of high mannose glycans (ranging from 1.9% to 2.5%) and the reference adalimumab (ranging from 5.3% to 12.0%) were reported. Such information on the test results allows for a better understanding of the regulatory interpretation and scientific justification behind the regulatory approval of biosimilars.

The present study used a classification scheme to investigate in a standardized manner how EU regulators present information on the biosimilarity of QAs and pCQAs in EPARs. The focus on the pCQAs to be considered in biosimilarity assessment, which may affect the clinical profiles of adalimumab products, was a strength of this investigation. The selection of adalimumab pCQAs was based on the literature review, providing an overview concerning which QAs are considered pCQAs with the current knowledge. This study stresses the importance of EPARs as a source of information that provides insight into the scientific evidence underpinning the regulatory approval of biosimilars.

Our study does have some limitations, which are noted as follows. First, these study findings are restricted to adalimumab biosimilars, which may hamper the generalizability to biosimilars of other biological molecules. Nevertheless, even if a biosimilarity assessment of another molecule is conducted with a different set of QAs and pCQAs, the findings, especially the focus on reporting the pCQAs, are expected to be comparable to other types of biosimilars because all EPARs are published by the same regulatory agency (i.e., EMA). Second, the generalizability of our findings to the regulatory reports from various jurisdictions, such as in the US FDA review reports, is unknown and beyond the scope of this study. Third, the QA classification scheme may not have captured all pCQAs of adalimumab because no consensus list is currently available. However, a literature search for publications on comparability and biosimilarity studies of adalimumab products was performed, and no pCQAs were identified that were not included in our classification.

Our observations reveal that minor differences in certain QAs between biosimilars and reference biological can occur at the same level of variability between pre- and postmanufacturing change batches of the reference biological [23,35,63], which reassures the

biosimilar regulation system. Although EU regulators have focused on describing pCQAs, these critical attributes were not explicitly defined in EPARs. Because biosimilar companies have conducted extensive analyses to define pCQAs based on their risk assessments, it would be preferable if regulators clearly define which QAs are identified as pCQAs by the companies. A clear definition of pCQAs in EPARs would enable stakeholders to better understand the links between QAs and the clinical profile and the meaning of the QAs concerning patient safety and product efficacy. The pCQAs may also (d)evolve over the drug life cycle based on the knowledge gained regarding the product and process. Standardized reporting of pCQAs in EPARs would benefit regulatory learning by allowing future researches to track pCQAs over time. Learning of pCQAs over time might result in reducing the need for comparative clinical trials and streamlining biosimilar approvals [9–12].

Although the EMA quality guidance of biosimilars provides high-level information on QAs, the guidance was last updated in 2014 and may not reflect the current state of knowledge and regulatory experience regarding QAs for biosimilars [5]. The lack of information on pCQAs in the guidance is understandable because these were not entirely known in the early years of biosimilar regulation. Nevertheless, the accumulated and long experience of EU biosimilar regulation as reflected in EPARs would fuel regulatory guidance with product-specific pCQAs, making the regulatory standard more visible and predictable.

As EPARs are considered an unbiased information source, there is great value in providing insight into the biosimilarity assessment of QAs for various stakeholders involved in biosimilar development, adoption, and regulation. The pharmaceutical industry can use EPARs to learn from past successes and failures and predict the regulatory process, and EPARs as such may contribute to reducing the time and cost of biosimilar development [64]. Healthcare professionals (HCPs) can use EPARs to understand the QA assessment's crucial role during the regulatory approval of biosimilars [65,66]. Reporting more extensive information about pCQAs in EPARs could help HCPs understand the predominant role of QAs and the reduced weight of evidence from comparative clinical trials in biosimilar approval. Among HCPs, pharmacists are uniquely positioned to take a leading role in informing other HCPs and patients about the scientific evidence underpinning biosimilar approval. Such efforts could increase confidence in and acceptance of using biosimilars in medical practice to fully capture the societal and patients benefits offered by biosimilars. Non-European regulatory authorities can use EPARs to support their own decision-making process, relying on the regulatory assessment undertaken by competent authorities in the world [67–72]. Therefore, EPARs could contribute to accelerating the regulatory review process and patients access to biosimilars in non-European jurisdictions.

For a comprehensive understanding of biosimilarity concepts and the predominant role of QAs in the approval of biosimilars, continued improvement in presenting biosimilarity assessments of QAs in EPARs is recommended. One method could include applying a structured uniform approach to QA reporting in EPARs. Such an approach may enhance the completeness and consistency of QA data and avoid missing crucial regulatory reflection on clinically relevant pCQAs. Greater consistency in QA reporting could make the EPAR a valuable and reliable tool for stakeholders to support evidence-based education to address the lack of knowledge and understanding of the scientific rationale behind biosimilar approval. Biosimilarity assessments of QAs in EPARs could be summarized in a standardized format that includes the type of evaluated QAs, explicit definition of the pCQAs, the test methods used and their results, the biosimilarity interpretation and scientific justification of the differences, if applicable. This summary could be achieved through adopting the International Pharmaceutical Regulators Program's regulatory review templates to optimize the current content with respect to biosimilarity assessment of QAs in EPARs [73]. Alternatively, initiating a project similar to the collaborative study between the EMA and European network for health technology assessment [74], which has resulted in a template to improve the contribution of EPARs in health technology assessments of relative drug effectiveness.

### **4. Methods**

### *4.1. Study Cohort*

In this study, the initial EPARs of all adalimumab biosimilars approved by the EMA before 30 November 2020 were included. The initial EPARs of adalimumab biosimilars were retrieved from the official EMA website (http://www.ema.europa.eu (accessed on 1 June 2020 )) [37]. The EPAR contains a summary of the submitted registration dossier and the scientific assessment undertaken by the CHMP, a body that advises the EC on marketing authorization of medicines for human use. Only the initial EPAR of each adalimumab biosimilar released following the final EC decision was included in this study because biosimilarity assessments of QAs and pCQAs between biosimilar and reference biological are presented only in the initial EPARs.

The initial EPARs were used to extract baseline characteristics for each adalimumab biosimilar, including the company code(s), brand name(s), date of the initial EPAR publication, and member states of the rapporteurs responsible for the assessment. Some adalimumab biosimilars are produced by the same manufacturer but marketed under different brand names (e.g., the company code for Hefiya®, Halimatoz®, and Hyrimoz® is GP2017) for which the registration dossier and initial EPARs are identical. In such cases, only the EPAR of one brand name (e.g., Hefiya® for GP2017) was included in the study for subsequent analysis. The date of the initial EPAR publication was defined as the month and year when the EPAR was published by the EMA, which is generally the same date as the EC decision on marketing authorization. The member state was defined as the rapporteurs' European country of origin. The rapporteurs are the two CHMP members who led the regulatory assessment of a marketing authorization application.

### *4.2. Information on (Potentially Critical) Quality Attributes in EPARs*

The study outcome was the determination of how EU regulators report information on the biosimilarity assessment of QAs and pCQAs in the EPARs. Two aspects were studied: the type and extent of information on the reported QAs and pCQAs.

### 4.2.1. Types of Reported (Potentially Critical) Quality Attributes

The types of QAs and pCQAs reported in the biosimilarity assessment were identified from the quality, non-clinical, and clinical sections of the initial EPARs. A general classification scheme of QAs was used to extract information from the EPARs. Information about the development of the classification scheme has been described elsewhere [46]. In short, the first draft was developed by the authors based on information from the EMA and US FDA biosimilar guidelines [5–7] and publicly available information relevant to the molecular characterization of a biological drug. The classification scheme was validated by regulators involved in the quality assessment of biosimilars at the Dutch Medicines Evaluation Board (MEB) to ensure that no critical and relevant QAs were missed. The classification scheme divides the QAs into seven types with additional subtypes of structural (physiochemical properties, primary structure, higher-order structure, PTMs and purity and impurities) and functional attributes (biological and immunochemical activity), resulting in the classification of 77 (53 structural and 24 functional) QAs of biologicals considered in the biosimilarity assessment (Figure 2) [46,47].

Subsequently, a list of pCQAs was defined in a two-step process. First, the pCQAs of adalimumab were identified from scientific publications presenting comparability or biosimilarity studies of adalimumab products, including the reference biological (Humira®) and corresponding biosimilars [23–32]. The publications were selected from an updated search of our previous systematic review [46]. From this search, an initial list of 29 pCQAs of adalimumab was constructed based on the pCQAs proposed by the authors. Second, the initial list was compared with the pCQAs identified for monoclonal antibodies, in general, in the previous literature [17–22] to verify and broaden the initial selection of pCQAs. If a new pCQA was identified in this second step, the authors (A.M.A., T.J.G., and H.G.) discussed its relevancy to adalimumab and reached a consensus on the inclusion of the

attribute. In this way, two pCQAs were added to the initial list, resulting in a final list of 31 (18 structural and 13 functional) pCQAs considered relevant to adalimumab products. These pCQAs were classified according to the previously described scheme (Figure 2). discussed its relevancy to adalimumab and reached a consensus on the inclusion of the attribute. In this way, two pCQAs were added to the initial list, resulting in a final list of 31 (18 structural and 13 functional) pCQAs considered relevant to adalimumab products. These pCQAs were classified according to the previously described scheme (Figure 2).

Subsequently, a list of pCQAs was defined in a two-step process. First, the pCQAs of adalimumab were identified from scientific publications presenting comparability or biosimilarity studies of adalimumab products, including the reference biological (Humira®) and corresponding biosimilars [23–32]. The publications were selected from an updated search of our previous systematic review [46]. From this search, an initial list of 29 pCQAs of adalimumab was constructed based on the pCQAs proposed by the authors. Second, the initial list was compared with the pCQAs identified for monoclonal antibodies, in general, in the previous literature [17–22] to verify and broaden the initial selection of pCQAs. If a new pCQA was identified in this second step, the authors (A.M.A., T.J.G., and H.G.)

The types of QAs and pCQAs reported in the biosimilarity assessment were identified from the quality, non-clinical, and clinical sections of the initial EPARs. A general classification scheme of QAs was used to extract information from the EPARs. Information about the development of the classification scheme has been described elsewhere [46]. In short, the first draft was developed by the authors based on information from the EMA and US FDA biosimilar guidelines [5–7] and publicly available information relevant to the molecular characterization of a biological drug. The classification scheme was validated by regulators involved in the quality assessment of biosimilars at the Dutch Medicines Evaluation Board (MEB) to ensure that no critical and relevant QAs were missed. The classification scheme divides the QAs into seven types with additional subtypes of structural (physiochemical properties, primary structure, higher-order structure, PTMs and purity and impurities) and functional attributes (biological and immunochemical activity), resulting in the classification of 77 (53 structural and 24 functional) QAs of biolog-

*Pharmaceuticals* **2021**, *14*, 189 10 of 15

4.2.1. Types of Reported (Potentially Critical) Quality Attributes

icals considered in the biosimilarity assessment (Figure 2) [46,47].

**Figure 2.** Classification scheme for 77 common quality attributes (QAs) of biologicals including 31 potentially critical quality attributes (pCQAs) relevant to adalimumab. The pCQAs are presented in gray boxes. Definitions: ADCC: antibodydependent cellular cytotoxicity, ADCP: antibody-dependent cellular phagocytosis, CDC: complement-dependent cytotoxicity, C1q: complement component 1q, TNFα: tumor necrosis factor-alpha, s-TNFα: surface tumor necrosis factor-alpha, tm-TNFα: transmembrane tumor necrosis factor-alpha, Fc: fragment crystallizable, FcR: Fc receptor. **Figure 2.** Classification scheme for 77 common quality attributes (QAs) of biologicals including 31 potentially critical quality attributes (pCQAs) relevant to adalimumab. The pCQAs are presented in gray boxes. Definitions: ADCC: antibody-dependent cellular cytotoxicity, ADCP: antibody-dependent cellular phagocytosis, CDC: complement-dependent cytotoxicity, C1q: complement component 1q, TNFα: tumor necrosis factor-alpha, s-TNFα: surface tumor necrosis factoralpha, tm-TNFα: transmembrane tumor necrosis factor-alpha, Fc: fragment crystallizable, FcR: Fc receptor.

### 4.2.2. Extent of Reported Information on (Potentially Critical) Quality Attributes

4.2.2. Extent of Reported Information on (Potentially Critical) Quality Attributes

The extent of the information on QAs and pCQAs provided in the EPARs was categorized by whether a biosimilarity interpretation was reported (yes/no) and whether test results were reported (yes/no) for a given QA or pCQA. The four possible combinations of answers resulted in four categories for each reported QA and pCQA (Table 3) [47].

Biosimilarity interpretation was defined as reported (yes) if the EPAR contained keywords demarcating the regulatory interpretation of the biosimilarity of a QA and pCQA as identical, similar, or having minor differences. The interpretation of *similar* included wording such as "same," "match," "(highly) similar," "comparable," and "consistent".

Test results were defined as reported (yes) if the EPAR included the quantitative or qualitative acceptance criteria of a given QA and pCQA, which included the numerical limits, range, and distribution, as shown in the examples in Table 3, or other suitable visual assessment measures, such as the spectra for higher-order structures and chromatograms for purity and impurities.

### *4.3. Data Analysis*

The frequency of the reported QAs and pCQAs stratified by structural and functional attributes was used to express the consistency in reporting the QAs and pCQAs of adalimumab biosimilars by EU regulators in EPARs. A QA and pCQA was considered to be consistently reported if EU regulators describe it in all included EPARs. The proportion of reported QAs and pCQAs for the four reporting categories (see Table 3) was calculated and stratified by structural and functional attributes to compare the extent of information on reported QAs and pCQAs in EPARs. If the regulatory interpretation of the biosimilarity or test results were presented for a given QA or pCQA in the EPARs, the type of interpretation (identical, similar, or minor differences) and the acceptance biosimilarity criteria were identified.

**Table 3.** Definitions of the four reporting categories for the quality attributes (QAs) and potentially critical quality attributes (pCQAs) reported in biosimilarity assessments in the initial European public assessment reports (EPARs) [47].


ADCC: antibody-dependent cellular cytotoxicity, CDC: complement-dependent cytotoxicity, EC50: half-maximal effective concentration, TNFα: tumor necrosis factor-alpha, s-TNFα: surface tumor necrosis factor-alpha, tm-TNFα: transmembrane tumor necrosis factor-alpha, Fc: fragment crystallizable, FcR: Fc receptor, KD: equilibrium dissociation constant, nM: nanomoles, pM: picomoles.

### **5. Conclusions**

In conclusion, we found variations in the frequency of reported QAs between EPARs of adalimumab biosimilars. The minor differences in the identified QAs did not affect functions and clinical performance and seem to be largely quantitative differences and not essentially meaningful for the overall conclusion of biosimilarity.

In line with our hypothesis, the pCQAs, specifically functional pCQAs, were reported most frequently and consistently in EPARs, as these reflect the MoA and can potentially affect the clinical profile. Greater consistency could be applied in reporting of QAs with more emphasis on pCQAs in EPARs, which could improve the understanding of the relationship between QAs and the clinical profile, which may positively contribute to adopting biosimilars in clinical practice.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/1424-824 7/14/3/189/s1, Figure S1: The types of and extent of information on quality attributes (QAs) and potentially critical QAs (pCQAs, in bold and gray boxes) as part of biosimilarity assessment reported by regulators in the initial European public assessment reports (EPARs) of seven adalimumab biosimilars; Table S1: Types of biosimilarity interpretation of reported quality attributes (QAs) stratified by the company code of adalimumab biosimilars in the European public assessment reports

(EPARs); Table S2: Comparison of potentially critical quality attributes (pCQAs) where test results and interpretation were reported for ABP501 and MSB11022 biosimilar.

**Author Contributions:** Conceptualization, A.M.A., T.J.G., T.C.E., H.G.L., and H.G.; methodology, A.M.A., T.J.G., T.C.E., H.G.L., and H.G.; validation, A.M.A., T.J.G., and H.G.; formal analysis, A.M.A.; investigation, A.M.A., T.J.G., T.C.E., H.G.L., and H.G.; resources, T.J.G., T.C.E., H.G.L., and H.G.; data curation, A.M.A.; writing—original draft preparation, A.M.A.; writing—review and editing, A.M.A., T.J.G., T.C.E., H.G.L., and H.G.; visualization, A.M.A.; supervision, T.J.G., T.C.E., H.G.L., and H.G.; project administration, T.J.G., T.C.E., H.G.L., and H.G.; funding acquisition, A.M.A.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was funded by the Saudi Food and Drug Authority (SFDA) as a part of a Doctor of Philosophy (Ph.D.) project for A.M.A. The SFDA has no role in any aspect of the study, including the preparation, review, the approval of the manuscript, nor the decision to publish the manuscript.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The datasets during and/or analyzed during the current study are available from the corresponding author on reasonable request.

**Acknowledgments:** The authors acknowledge the contribution of Lotte A. Minnema for constructive suggestion on the search strategy and Magdalena A. Gamba for kind advice on the process of data collection.

**Conflicts of Interest:** A.M.A., T.J.G., T.C.E., H.G.L., and H.G. declare that they have no conflict of interest.

### **References**


### *Review* **Informing Patients about Biosimilar Medicines: The Role of European Patient Associations**

**Yannick Vandenplas 1,\* , Steven Simoens <sup>1</sup> , Philippe Van Wilder <sup>2</sup> , Arnold G. Vulto 1,3 and Isabelle Huys <sup>1</sup>**


**Abstract:** Biosimilar medicines support the sustainability of national healthcare systems, by reducing costs of biological therapies through increased competition. However, their adoption into clinical practice largely depends on the acceptance of healthcare providers and patients. Patients are different from health care professionals (HCPs), who are informing themselves professionally. For patients, the biosimilar debate only becomes actual when they are confronted with disease and drug choices. This paper provides a literature review on how patients are and should be informed about biosimilars, searching in scientific databases (i.e., Medline, Embase). Several large surveys have shown a lack of knowledge and trust in biosimilars among European patients in recent years. This review identified five main strategies to inform patients about biosimilars: (1) provide understandable information, (2) in a positive and transparent way, (3) tailored to the individual's needs, (4) with one voice, and (5) supported by audiovisual material. Moreover, the importance of a multistakeholder approach was underlined by describing the role of each stakeholder. Patients are a large and diffuse target group to be reached by educational programs. Therefore, patient associations have become increasingly important in correctly informing patients about biosimilar medicines. This has led to widespread biosimilar information for patients among European patient associations. Therefore, a web-based screening of European Patients' Forum (EPF) and International Alliance of Patients' Organizations (IAPO) member organizations on publicly available information about biosimilars was performed. We found that the level of detail, correctness, and the tone of the provided information varied. In conclusion, it is paramount to set up a close collaboration between all stakeholders to communicate, develop, and disseminate factual information about biosimilars for patients.

**Keywords:** biosimilar; biological; information; education; communication; patient; Europe

### **1. Introduction**

Since their introduction to the European market in 2006, biosimilar medicines have contributed to a more sustainable healthcare system in several European markets [1]. Biosimilars are biological medicines that contain a version of the active substance of an already authorized biological medicine in the European Economic Area (EEA) [2]. They are allowed to enter the market when market exclusivities of the original biological product have expired, and market authorization has been granted by the European Commission (EC). Market authorization is achieved after a rigorous regulatory evaluation process by the European Medicines Agency (EMA) and subsequent approval of the EC. This guarantees that biosimilars are as effective and safe as their reference product, making them equal treatment options for patients [2,3]. Several benefits have been identified following the increased competition induced by biosimilar market entry [4]. Due to the decreased costs of biological medicines, generated savings could be allocated to providing patients with

**Citation:** Vandenplas, Y.; Simoens, S.; Van Wilder, P.; Vulto, A.G.; Huys, I. Informing Patients about Biosimilar Medicines: The Role of European Patient Associations. *Pharmaceuticals* **2021**, *14*, 117. https://doi.org/ 10.3390/ph14020117

Academic Editor: Jean Jacques Vanden Eynde Received: 31 December 2020 Accepted: 1 February 2021 Published: 4 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

more access to biological therapies. In addition, these savings can be utilized to finance high-cost innovative treatments [1,5,6].

However, the extent to which these benefits are being captured in Europe largely depends on the adoption of biosimilars by European Union (EU) member states. Adoption into clinical practice might be hampered by limited healthcare provider (HCP) and patient acceptance of biosimilars. Often, besides other factors such as the absence of tangible incentives, a lack of acceptance among HCPs and patients comes down to shortcomings in knowledge and understanding about biosimilars [7–9]. Patients' access to information and education about biosimilar medicines is therefore considered as one of the key elements for a sustainable market [10]. Hence, policy initiatives aiming to increase understanding among clinicians and patients have been implemented in most European countries in past years [11,12].

Several studies have brought an inadequate understanding and acceptance among European patients about biosimilars to light, underlining the need for information and education of patients [13–19]. Especially when transitioning or *switching* current original (or innovator) biological therapy to its biosimilar, the value of adequate patients' understanding about biosimilars cannot be underestimated [20]. Clinical studies have proven the positive effect on patient outcomes when patients with rheumatological disorders were properly informed before transitioning to a biosimilar [21,22]. The authors attributed the improvement in patient outcomes after a structured communication strategy to a reduction in the risk of *nocebo effects.* The nocebo effect is described as the worsening of symptoms associated or an increase in side effects with a negative attitude towards a given therapy, in this case the biosimilar medicine. A lack of patient knowledge is the main underlying reason for negative attitudes towards biosimilars, contributing to nocebo effects and possible treatment failure [20,23].

Educating patients about biosimilars is crucial to provide clarity and prevent misinformation [9,20,24]. Patients need access to understandable and evidence-based information that allows them to make informed decisions about their treatment. Regulatory authorities, medical scientific associations, and patient organizations have therefore been active in developing and disseminating educational material on biosimilars for European patients during past years. However, information and educational material are widespread, requiring a mapping of the available material [8,9]. Mapping the available information or material for patients makes it possible to have an overview of what material exists, and to verify the information found for its scientific correctness. In addition, a proper inventory will facilitate the dissemination of information through collaboration between stakeholders.

This review aimed to provide an overview of existing scientific literature on how to inform patients about biosimilars and compile available information about biosimilars for patients, developed or disseminated by European patient associations. Based on this review, an overview of the important aspects when talking to patients about biosimilars is provided for policymakers, healthcare providers, patient organizations, and other relevant stakeholders, in support of a sustainable market for off-patent biological and biosimilar medicines in Europe.

### **2. Methods**

### *2.1. Literature Review*

This comprehensive structured literature review identified articles on what information patients need about biosimilars and how this information can be communicated, by looking into scientific databases (Embase, Medline) using a structured search strategy (Cfr. Supplementary Table S1). Relevant English-language scientific publications published between 2006 and 2020 were included. This period was chosen since biosimilars have been introduced in Europe in 2006, thereby encompassing the whole period of time when biosimilars were available on the European market. Search terms were related to patient communication about biosimilars and included the following terms: 'biosimilar', 'information', 'education', 'communication', 'knowledge', and 'patient'. All terms were modified

according to the respective scientific database. Both abstracts and full texts were included in the analysis. Only articles relevant to the European landscape were within the scope of this analysis. Articles were searched up to the 21st of October 2020.

All identified records were imported from Embase or Pubmed into Mendeley software to remove duplicates. Next, all articles were screened on title and abstract for relevance in the Rayyan (Qatar Computing Research Institute, Doha, Qatar) software. In a third step, articles were carefully reviewed based on their full text. Lastly, reference lists of included articles were searched for additional relevant articles. The articles included in the final analysis were analyzed qualitatively according to the thematic framework method [25]. A combination of inductive and deductive coding was used, since some aspects were already identified as relevant for this research question. During the initial coding step, general themes were identified prior to the literature review. Similar codes were grouped together to form the coding tree. Second, the identified literature was coded deductively. Meanwhile, additional codes were created inductively and added to the coding tree.

### *2.2. Mapping of Patient Information*

A web-based screening on relevant patient information (i.e., general information not intended for educational purposes) or educational material (i.e., brochures, toolboxes, position papers, audiovisual material, etc.) was performed to provide an overview of the educational material disseminated by European patient organizations. This screening included all public websites of European Patients' Forum (EPF) and European International Alliance of Patients' Organizations (IAPO) members. EPF and IAPO are two major umbrella associations, uniting a large number of European patient organizations in a variety of disease areas. Websites were screened on available information about biosimilars by searching for 'biosimilar' or related terms in the search bar. In addition, the name of the respective patient association was combined with the term 'biosimilar' via Google to make sure no information was missed.

After all identified information was analyzed and mapped together, the tone in which each association reports about biosimilars was evaluated on a five-point Likert scale. This was done by scoring the overall attitude towards biosimilar medicines on the following scale: "− −" (negative), "−" (somewhat negative), "0" (neutral), "+" (somewhat positive), "+ +" (positive). Neutral information was taken as a starting point. Neutral information refers to factually correct information about biosimilars, without any additional positive or negative undertone. The initial scoring was done by one researcher (Y.V.), and afterwards reviewed by four other researchers (S.S., A.G.V., P.V.W., I.H.).

The purpose of the web-based screening was (1) to examine to what extent patient information about biosimilars is provided on their public websites, (2) to have a closer look at the actual content of these materials, and (3) to evaluate the tone in which they report about biosimilars. All different types of information found was schematically listed per patient association (Table 1).





 The evaluation of the overall attitude towards biosimilars for each patient organization is done on a five-point Likert scale. The scale is as follows: "− −" (negative), "−" (somewhat negative), 0 (neutral), "+" (somewhat positive), and "+ +" (positive).

### **3. Results**

### *3.1. Literature Review*

After a screening of 1319 records, a total of 51 articles were included in this literature review. Although conference abstracts (*n* = 6) were also eligible for inclusion, most identified records were full-text articles (*n* = 45). Most articles were identified through the structured literature search after title and abstract screening (*n* = 38). Nonetheless, the screening of reference lists resulted in 13 additional records. A complete overview of the literature search process is included in Supplementary Table S1 and Figure S1.

### 3.1.1. Points to Consider When Talking to Patients about Biosimilars

In the vast body of literature, we can conclude that several specific aspects are essential when informing patients about biosimilars. An overview of these aspects is provided below.

### Provide Understandable and Up-to-Date Information

Biosimilars are a relatively new and difficult concept for patients. It is therefore important that the given information to patients is easy to understand and not overly complicated. The message must be concise, using simple language, avoiding redundant medical jargon [50–52]. When informing the patient face-to-face, make sure they understand all information by asking questions and involving them in the discussion [52]. In this way, the patient will feel more involved and can participate in the discussion as well. In addition, the information must be up-to-date and adapted to the most recent insights [53]. It should not contain outdated concepts or outdated data.

### Communicate Positively

Several studies have already shown that it is crucial to positively formulate the message about biosimilars towards patients. An empathic and positive communication (including positive framing) or attitude increase the acceptance to switch and reduce the development of nocebo effects after transitioning to a biosimilar [54–58]. An open and positive communication, emphasizing the equalities and not the differences between the reference product and its biosimilar, should be the norm when talking to patients. Information or communication should avoid messages such as: "biosimilars have no meaningful differences with their reference product". Instead, the similarities must be underlined in any communication to reassure patients that biosimilars are equal treatment alternatives [20,58,59]. When transitioning to a biosimilar, it is unnecessary to mention all possible side effects. It is rather recommended to provide patients with the opportunity to contact their physicians or nurse when any unexpected side effect would occur [60]. Moreover, a positive communication about biosimilars should be adopted for information towards HCPs as well, thereby supporting overall acceptance of biosimilars in clinical practice [9,50].

Patients generally feel that their physician's opinion and attitude on biosimilars strongly influences their decision to use a biosimilar [61]. Yet, an open and positive attitude should be adopted by all healthcare providers (i.e., physicians, nurses, and pharmacists) who communicate with patients. This involves empathy, reassurance, and nonverbal elements in their communication towards patients when discussing medicines in general [51,62]. It will be essential to educate HCPs using these communication techniques or 'soft skills' in the future.

### Provide Information Tailored to the Individual Patients' Needs

A one-size-fits-all approach to communicate or inform patients about biosimilars does not exist, nor would it be appropriate [8,63]. Some patients will naturally be more concerned about their treatment and ask for more information. While other patients trust their physician completely and will express no further concerns about biosimilars [55,60,64]. However, many patients will be somewhere between these two extremes of the spectrum, highlighting the importance of tailored communication. Providing too much information

could lead to unnecessary concerns of patients, while giving too little information could leave patients with remaining concerns [65]. It is the task of all HCPs to assess the individual patients' needs and find the right balance. Specific tools or questionnaires exist to assess prior beliefs or concerns of patients about their medicine, such as the Beliefs about Medicine Questionnaire (BMQ) [63]. The BMQ might help HCPs stratify patients based on their prior thoughts about biosimilars before transitioning.

In addition, information should be tailored to the individual patient's demographics and health literacy as well [66]. For example, patients affiliated to a patient association or previously treated with a biological medicine generally have a better knowledge about biosimilars [16]. Some patients might have already looked for information about biosimilars elsewhere, given the broad access to information on the internet [64,67]. It is therefore advised to account for this and assess whether their prior knowledge is factual. Furthermore, in order to make sure that the information is accessible for all patients, educational material should be translated into local languages.

### Communicate with One Voice

As already touched upon in the above, communication towards patients must be consistent across resources, so confusion among patients is avoided. Homogenous information leads to higher acceptance and better treatment outcomes after transitioning to a biosimilar [54,57]. Stakeholders should therefore deliver the same message or speak with *one voice* to patients about biosimilars [7,20]. Such an approach means that all healthcare providers are involved and educated about biosimilars, ensuring a coherent and unified message to patients. Not only the information itself, but also the way it is explained to patients should be coherent (i.e., positive and open communication, tailored information) [68].

### Make Use of Supportive Material

Several ways exist to inform patients in addition to oral communication of the HCP with the patient. In the context of transitioning or switching to biosimilars, written informed consent before transitioning could be considered. Such information must be in the patient's native language, include only key information on biosimilars, the reasons why transitioning is considered, and who to contact if they have any issues or concerns [50,54,57,69].

For general information accessible to patients, a variety of audiovisual aids can be used, such as videos, infographics, podcasts, and pictures [50,52]. All these ways may contribute to the understandability and confidence in the key biosimilar concepts. Moreover, for subcutaneous biosimilars, instructional leaflets or videos about the injection device might be useful as well. Since patients are increasingly seeking health-related information on the internet, such audiovisual material can be made broadly accessible online [67]. For example, the European Medicines Agency (EMA) and European Commission (EC) developed an animated video explaining the general concepts of biosimilar medicines [70].

### 3.1.2. Information Needs of Patients about Biosimilar Medicines

A multitude of studies has been performed in past years assessing the level of knowledge about or attitudes towards biosimilar medicines among European patients. In general, most of these studies concluded that the level of knowledge of patients is limited, as well as that confidence in biosimilars is rather low. In particular, limited knowledge about the general concepts of biological and biosimilar medicines is reported [13–17,19,54,56,71–73]. Doubts around efficacy, safety, and extrapolation of indications were revealed among most patient populations (i.e., oncology, psoriasis, rheumatology, IBD). It goes without saying that correct information and education can resolve these concerns and lack of knowledge.

A tailored approach was already pointed out earlier in this review in the context of direct communication of HCPs towards patients. The specific biosimilar concepts that should be explained by HCPs will therefore vary from patient to patient, depending on the individual needs and level of understanding. It used to be common practice that the basic concepts about biological medicines, and biosimilars in particular (e.g., definitions,

safety, efficacy, regulatory approval, etc.), have to be clearly explained to patients when transitioning to a biosimilar [20,60,74]. However, nowadays current practice has evolved towards providing the message that another brand of the same medicine will be used, with the same efficacy and safety outcomes at a lower cost. finding incorrect information. The purpose of providing information is to counter such negative reports as well [9,74,75]. Therefore, publicly available information or educational material about biosimilars for patients should address the general definitions of biological

A tailored approach was already pointed out earlier in this review in the context of

There is still a lack of clarity about which aspects of biosimilars should be included

when developing educational material for patients [8]. It should be borne in mind that patients themselves look for information about biosimilars on the internet, potentially

direct communication of HCPs towards patients. The specific biosimilar concepts that should be explained by HCPs will therefore vary from patient to patient, depending on the individual needs and level of understanding. It used to be common practice that the basic concepts about biological medicines, and biosimilars in particular (e.g., definitions, safety, efficacy, regulatory approval, etc.), have to be clearly explained to patients when transitioning to a biosimilar [20,60,74]. However, nowadays current practice has evolved towards providing the message that another brand of the same medicine will be used,

*Pharmaceuticals* **2021**, *14*, x FOR PEER REVIEW 9 of 22

with the same efficacy and safety outcomes at a lower cost.

There is still a lack of clarity about which aspects of biosimilars should be included when developing educational material for patients [8]. It should be borne in mind that patients themselves look for information about biosimilars on the internet, potentially finding incorrect information. The purpose of providing information is to counter such negative reports as well [9,74,75]. Therefore, publicly available information or educational material about biosimilars for patients should address the general definitions of biological and biosimilar medicines in an understandable way. This should include the thorough regulatory evaluation process of EMA that assures the same clinical efficacy and safety between the original and biosimilar product. The potential benefits of biosimilars can also be considered, albeit in understandable language and as direct benefits (i.e., increase in access to necessary medicines or access to treatments at an earlier disease stage) [8,54,76]. However, it should be avoided that the impression is created that the patient is treated with biosimilars only for the sake of cost savings. Other essential concepts such as extrapolation of indication may be explained as well, although overly detailed information should always be avoided [8,20]. and biosimilar medicines in an understandable way. This should include the thorough regulatory evaluation process of EMA that assures the same clinical efficacy and safety between the original and biosimilar product. The potential benefits of biosimilars can also be considered, albeit in understandable language and as direct benefits (i.e., increase in access to necessary medicines or access to treatments at an earlier disease stage) [8,54,76]. However, it should be avoided that the impression is created that the patient is treated with biosimilars only for the sake of cost savings. Other essential concepts such as extrapolation of indication may be explained as well, although overly detailed information should always be avoided [8,20]. 3.1.3. Reaching the Patient All stakeholders, particularly healthcare providers, play a role in informing patients

### 3.1.3. Reaching the Patient about biosimilars. It must be stressed that communicating with patients should be a mul-

All stakeholders, particularly healthcare providers, play a role in informing patients about biosimilars. It must be stressed that communicating with patients should be a multistakeholder effort [8,20,77]. This includes physicians, nurses, pharmacists, scientific associations, regulatory bodies, and patient associations. In the following, we summarize the role of each stakeholder in informing patients about biosimilars (Figure 1). tistakeholder effort [8,20,77]. This includes physicians, nurses, pharmacists, scientific associations, regulatory bodies, and patient associations. In the following, we summarize the role of each stakeholder in informing patients about biosimilars (Figure 1).

**Figure 1.** The multistakeholder approach using five main strategies for informing patients about **Figure 1.** The multistakeholder approach using five main strategies for informing patients about biosimilars.

biosimilars.

### Role of Physicians

Treatment decisions must be based on shared decision-making between patients and their physician. In most European countries, physicians have the ultimate responsibility in making treatment choices. Physicians will often be the first point of contact for patients when treatment decisions are being made, and they should therefore ensure a trusted relationship with the patient. Good communication, based on informed discussions and shared decision-making with the physician, is known to benefit adherence to a prescribed medicine, and thus the adoption of biosimilars [5,20,21,74]. However, shared-decision making about medical therapy in general is not yet established to the same extent in every European country [78,79].

Previous research involving patient surveys has shown that physicians are the most trusted source of information about biosimilars [17,19,80]. However, several surveys among European physicians have concluded that physicians' knowledge on biosimilar medicines could be improved [50]. As a result, it is clear that physicians should be properly trained about biosimilars and be able to communicate adequately about them to the patient. As mentioned earlier, physicians must therefore be trained in communication techniques as well [65].

### Role of Nurses

Nurses play a key role in the daily care for patients and are ideally placed to inform patients by addressing questions or concerns about their medicine. Usually, nurses administer the medication and spend the most time with patients, which allows them to have a closer relationship with the patient [81]. When transitioning from a reference product to its biosimilar, the important role of nurses has been pointed out in several publications during past years [7,8,52,81,82]. Building further on their profound experience with educating patients, nurses can guide patients in the process when transitioning to a biosimilar and manage nocebo effects. Additionally, following the transition or initiation with a biosimilar, patients may have further questions or concerns at home. To prevent any additional concerns or even discontinuation of their treatment, nurses should serve as a contact point to patients [58,83]. For subcutaneously administered biologicals, where injection devices may differ, nurses provide the necessary explanation and guidance to use the new injection device [52].

The above reasons make it clear that nurses are a critical link in the multidisciplinary team, particularly when making the transition to a biosimilar. This has been recognized by the European Specialist Nurses Organization (ESNO), by developing an elaborate communication guide for nurses when transitioning to a biosimilar in 2017 [84]. This document has been translated into eight languages and can serve as a reference document for nurses.

### Role of Pharmacists

The main task of a pharmacist is often simplified to the delivery of medicines. However, pharmacists also have an important task of providing information to patients, although regional differences exist among European countries in their role in direct patient counseling. Especially community pharmacists serve as a first-line contact for patients for any questions about their medicine, including biosimilar medicines [58,66]. Pharmacists thereby contribute to medication adherence by increasing confidence in biosimilar medicines among patients. They may also have to explain differences in injection devices, since subcutaneously administered biosimilars are often dispensed in community or outpatient pharmacies.

For biosimilars delivered in the hospital setting, pharmacists have an increasing role in educating the medical staff about biosimilars [85,86]. The Dutch association of hospital pharmacists (NVZA) has developed a practical guidance document (i.e., toolbox) on how to implement biosimilars in the hospital setting, thereby emphasizing the role of hospital pharmacists in this process [87]. As medicine experts, clinical pharmacists can serve as

a coordinator of the medical team to address patients' concerns about biosimilars when preparing the switch to a biosimilar. Their role should be further explored in the future, particularly in the context of transitioning to biosimilars in the hospital setting.

### Role of Scientific or Medical Associations

Several European scientific associations have developed educational material for patients in past years about biosimilars. Due to their extensive scientific expertise and background, they are an important source of unbiased information about the use of biosimilars [62,66,71]. The European Society for Medical Oncology (ESMO) developed educational leaflets about biosimilars for patients [88]. ESMO uses infographics to explain the key concepts and potential advantages of biosimilars in understandable language. The European League Against Rheumatism (EULAR) has developed a document with general information about biosimilars as well. The main questions or concerns patients may have are addressed in this question and answer brochure [89,90]. Additionally, the need for more patient educational material is highlighted in this document.

### Role of Regulatory Authorities

European regulatory agencies and national competent authorities have a supporting role in disseminating unbiased information about biosimilars in general [55,66,91]. However, room for improvement was recently pointed out for European national competent authorities to disseminate biosimilar information to the public [92]. The widespread patient brochure developed by the European Commission (EC) and European Medicines Agency (EMA) has become a reference document for patients, and is being referred to by many national authorities [93]. This brochure was developed in cooperation with the European Patients' Forum (EPF) in 2016, explaining the key concepts about biological and biosimilar medicines in lay language. It is also publicly available in a more concise video format [70]. In recent years, this material has been translated into all European languages [93]. National authorities should continue facilitating the dissemination of this document, as it provides coherent and factual information about biosimilars in understandable language and graphical format [92].

### Role of Patient Associations

Patient organizations are a trusted source of information for patients about biosimilars. Patients rely on their respective associations or advocacy groups to clarify complex concepts such as biosimilars [16,19]. Patient associations can also serve as a discussion board to discuss complex matters such as biosimilars and share experiences among patients [17]. If patient associations are committed to developing educational material themselves, they should join forces with medical and scientific associations. In this way, it can be ensured that the information is evidence-based and up-to-date [13,71].

A schematic overview of the multistakeholder approach, using the five identified strategies, is provided in Figure 1. The section below takes a closer look at the role of patient associations in developing and disseminating information about biosimilars to patients.

### *3.2. Information Provided by European Patient Organizations*

In total, public websites of 75 European Patients' Forum (EPF) members and 95 members of the International Alliance of Patients' Organizations (IAPO) were consulted. As some organizations were part of both EPF and IAPO, 159 unique members were screened. Of these 159 patient organizations, 16 were actively disseminating information on biosimilars via their website. An overview summarizing all patient organizations, along with the type of information, is provided in Table 1.

Patient associations active in providing information about biosimilars are representing patients with a variety of diseases or regions. The main disease areas are those where biosimilars are marketed today, such as rheumatology, diabetes, oncology, inflammatory

bowel diseases, and psoriasis. The majority of these associations only provide brief information on biosimilars, by explaining key concepts or merely providing a link to the patient brochure developed by the European Commission (EC) [93]. Nonetheless, some patient organizations have developed their own educational material or even produced position statements on the use of biosimilar medicines within their specific disease area. All identified information or educational material on biosimilars intended for patients is summarized below (Table 1).

The Slovakian Association for the Protection of Patients' Rights (AOPP) provides a short article briefly explaining the main characteristics of originator biological and biosimilar medicines. For more information, they refer patients to the EC brochure [26].

One of the larger patient associations discussed in this review is Digestive Cancers Europe (DiCE). It is the umbrella organization of a larger group or national associations representing patients with colorectal, gastric, and pancreatic cancers. DiCE has committed itself in recent years to several educational initiatives. In 2019, they developed a position paper on biosimilars for the treatment of colorectal cancer [27]. In this well-structured paper, they touch on the definitions of biologicals, with specific information about biosimilar medicines. They also draw attention to the benefits of biosimilar usage, in particular the increase in access to biological medicines. Problems regarding unequal access to biologicals among European countries are mentioned, including the possible role of biosimilars to overcome these to a certain extent. More recently, in the context of the licensing of bevacizumab biosimilars in Europe, DiCE started a larger project to provide educational material about biosimilars for patients and HCPs [28].

Similar to DiCE, the European Federation of Crohn's and Ulcerative Colitis Associations (EFCCA) is the umbrella organization representing national Crohn's and ulcerative colitis patient associations. Like most patient associations discussed in this review, they mention the EC brochure about biosimilars for patients. EFCCA also wrote an article about biosimilars, mentioning specific information on biosimilars and their benefits for healthcare systems in their monthly magazine. In this article, EFCCA emphasizes the importance of generics and biosimilars for a competitive market and a more sustainable healthcare system. In addition, they state that physicians should not be obliged to prescribe a biosimilar purely on the grounds of cost, but should be allowed to exercise appropriate clinical judgment and always involve patients in the decision making process [31].

Even though no biosimilars have been marketed yet for the treatment of Parkinson's Disease, the European Parkinson's Disease Association (EPDA) provides a brief explanation of biologicals in general, as well as of biosimilar medicines. They emphasize that all biological medicines are prone to structural variability and the possible consequences on clinical outcomes. Due to the varying composition of biologicals, patient safety may be a concern. The only specific information given on biosimilars is that they aim for the same mechanism of action as the original, even though different cells are used during the production process [33].

The International Diabetes Federation Europe or IDF Europe is the umbrella organization of European national associations for patients with diabetes. IDF Europe has developed a position paper on the use of biosimilars among patients with diabetes in 2017 [34]. In this extensive position document, several topics are highlighted such as the difference between biosimilars and generics, the European legal framework, and the potential impact of biosimilars on healthcare systems. The paper ends with a set of recommendations for the use of biosimilars in clinical practice. Under the list of recommendations, IDF Europe states that stable patients on insulin treatment should not be switched to a biosimilar without good clinical reasons and evidence of interchangeability. Furthermore, patients should always be informed and involved in the decision-making process, based on an informed discussion with their physician. They demand more information for patients from national regulatory authorities, specifically about biosimilar medicines. Routine education for patients with diabetes, facilitated by national authorities, should include a section on biosimilars. However, the position paper emphasizes possible clinical dif-

ferences between insulin biosimilars and reference products. According to the authors, not enough clinical evidence exists to ensure biosimilars are equally safe and effective as their reference product. Possible immunogenicity risks are pointed out, especially when switching the reference biological with its biosimilar. To support this statement, they refer to an epoetin biosimilar (HX-575) that showed an increased occurrence of adverse events linked to a higher immunogenicity of the biosimilar. However, the article they refer to does not mention a possible difference between the biosimilar and originator of epoetin due to increased immunogenicity [94]. Instead, the article describes several cases of pure red cell aplasia (PRCA) with epoetin treatment, among which a trial with a biosimilar of epoetin. The particular clinical study being referred to reported two cases of neutralizing antibodies with the epoetin biosimilar [95]. An extensive analysis revealed that contamination during primary packaging of the prefilled syringes explained the increase in neutralizing antibodies [96]. The manufacturing process was therefore improved, followed by the completion of new open-label study without any patients developing neutralizing antibodies. Subsequently, the respective biosimilar HX-575 was authorized on the European market in 2016. This type of information is an example of a false narrative, by supporting incorrect conclusions with references from published scientific articles. Such kind of incorrect and negatively framed information should be avoided, since it may harm the trust in biosimilar medicines among patients with diabetes and potentially lead to a slower adoption of biosimilars [9].

Malta Health Network (MHN), the national association for Maltese patients, provides a link to the EUPATI toolbox on biosimilar medicines [35]. Although this information is rather difficult to find on the MHN website, the EUPATI toolbox provides understandable information on biological medicines, including biosimilars, for patients [97].

The Spanish Platform for Patient Organizations published an article for patients about biological and biosimilar medicines in 2017 [36]. Definitions about biologicals, biosimilars, and the difference with generic medicines are highlighted. They underline the importance of therapeutic freedom of physicians when prescribing biosimilars, and switching must always be in close dialogue with the patient. While they are not opposed to switching the original product with the biosimilar, they state that there is insufficient evidence to support switching.

Another national patient association is the National Coalition of Dutch Patients, which provides simple and understandable information about biosimilars on its website [39]. Like many other patient associations discussed in this review, the importance of involving the patient in the decision to prescribe a biosimilar is mentioned. For further information, they refer patients to a brochure developed by the Dutch competent authority [40]. This is a structured document providing information about biological and biosimilar medicines in understandable language for patients. Moreover, the question and answer structure of the document might increase the understandability of the brochure. In contrast to other informational material discussed in this article, the same effect of the biosimilar and the reference product is emphasized instead of no expected differences. In general, a more favorable position towards switching to a biosimilar is noted. Yet, switching must remain the physician's responsibility, and the necessary consultation with the patient is required.

The Flemish Patient Platform (FPP) unites Dutch-speaking Belgian patient associations. FPP mentions very limited information on biosimilars, merely explaining that they are biological medicines. FPP refers to biosimilars as copies, similar to generic medicines, which is too simplistic and incorrect. Furthermore, they advise patients to look at the Belgian national competent authority's information on biosimilars [42]. Here, general information about biosimilars is provided, including definitions, approved biosimilars, and guidance when switching to biosimilars. However, this web page is not up to date and only contains information of approved biosimilars until 2016.

The European Multiple Sclerosis Platform (EMSP), European Federation of Neurological Associations (EFNA), European Institute of Women's Health (EIWH), and European

Patients' Forum (EPF) only posted the link to the EC brochure for patients on their website [32,37,38,43].

Three additional patient organization members of IAPO were identified that provide educational material for patients on their website. The International Federation of Psoriasis Associations (IFPA) is the overarching organization of national patient associations representing patients with psoriasis. IFPA recently developed a position paper about the use of biosimilars for the treatment of psoriasis [45]. They acknowledge that biosimilars do not lead to different clinical outcomes compared with their reference product. Again, the patient–physician dialogue is underlined when making treatment decisions in general, which includes decisions to switch to a biosimilar. However, they mention transitioning to a biosimilar should not be done for patients with stable disease control. This shows some hesitance to use biosimilars among patients with psoriasis already treated by biological medicines.

Psoriasis Action, the Spanish patient association for patients with psoriasis, provides several sources of biosimilar information. They share a short video in which a professor explains what biosimilars are to Spanish patients [47]. A video with a more extensive explanation is also provided, intended for patients who prefer more detailed information on biosimilars [48]. Psoriasis Action also published a more general piece of information for patients, where generalities of biological and biosimilar medicines are explained in a specific article [46].

Last but not least, IAPO also developed educational material for patients about biosimilars in collaboration with the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA). On their website, extensive documentation on biological and biosimilar medicines can be found in their toolkit for patients, from which they developed a second version in 2017 [49]. The toolkit includes fact sheets, infographics, frequently asked questions, and a decision guide for patients when choosing between an original biological or biosimilar product. Their educational material includes general information on biological and biosimilar medicines, regulatory requirements, pharmacovigilance, and a communication guide for HCPs. The toolkit is intended for patient organizations worldwide for distribution to their patients. In contrast to the initial version of this toolkit of 2013, which was made available in English, Spanish, and Portuguese, the second version is only available in English [98].

### **4. Discussion**

This article looked at the relevant elements to consider when informing patients about biosimilars. In addition, an overview of the information and educational material by the major European patient associations was provided. Based on this overview, all available material was evaluated on its tone and correctness.

### *4.1. Communication Strategies to Inform Patients about Biosimilars*

Five main points of attention were identified when informing patients about biosimilars. **First** of all, information has to be provided in an understandable way. Patients generally have no scientific background, so one must make sure not to overly complicate the given message [50–53]. **Second**, a positive attitude when talking to patients about medicines in general is paramount [20,51,58,62]. Emphasis must be put on the similarities between biosimilars and their reference product, rather than the possible differences. This can be done by conveying the message that the biosimilar has similar clinical outcomes, instead of no expected differences [59]. An open and positive way of communicating has shown to generate trust, and subsequently improve treatment outcomes and adherence [56,57]. HCPs should therefore be trained on the proper use of such communication strategies with patients. **Third**, a one-size-fits-all approach is not desirable when communicating directly to patients since each patient's individual needs and level of understanding might differ [8,60,63,64,66]. A tailored approach is therefore preferred. It is the task of each member of the multidisciplinary team to assess these needs and to adapt their com-

munication strategy accordingly. This brings us to the **fourth** point of attention, the *one voice* principle. In essence, this means that everyone informing patients about biosimilars has to provide a coherent message. Communication towards patients must be consistent across channels, thereby avoiding suspicion by generating trust between healthcare providers and patients [7,20,54,58,68]. **Fifth**, the use of supportive audiovisual material (i.e., videos, infographics, brochures) may help bringing the information across in a clear and understandable way [9,50,52,67]. Such supportive material closes the gap between the complexity of the biosimilar concepts and the need for understandable information.

A series of studies pointed to a lack of knowledge and trust in biosimilars in various relevant patient populations, making clear the necessity of education [13–17,19,54,56,71–73]. However, the purpose of informing patients should not be to create a high level of knowledge among the whole patient population. This would not be feasible, nor desirable. After all, it is not intended to inform all patients about a treatment the vast majority will not need. Instead, information about biosimilars should be reaching those patients who require such information. In other words, patients who may or will be treated with biosimilars in the near future. This approach differs from informing HCPs about biosimilars, as they all need to have a good understanding of biosimilars.

Educating patients about medicines in general, but in particular biosimilars, should always be a multistakeholder effort [8,20,60,77]. Each stakeholder has its own role to fulfill in order to provide correct, unbiased, understandable, and coherent information. Physicians, nurses, and pharmacists have a coordinating role and are key partners to remove doubts and generate trust in biosimilars, as for any kind of medicine [52,60,85,86]. In addition, other parties such as regulatory authorities, medical societies, and patient associations have a supporting role in informing patients. They are all regarded by patients as reliable sources of information. However, the identified list of stakeholders is not exhaustive, since other stakeholders that were not mentioned in the literature may also play a role. For example, academia might support the development of evidence-based information as a trusted and unbiased source of information. Other national authorities, such as payers and health technology assessment (HTA) bodies, could also disseminate information about biosimilars to patients. Some stakeholders may be of particular importance in the creation of information or educational material (e.g., scientific associations, professional associations, academia), whereas others (e.g., healthcare providers, patient associations, regulatory authorities) in the dissemination of information to patients. Moreover, pharmaceutical companies also play a role in informing the wider public about biosimilar medicines. One must acknowledge that many informational campaigns are supported by pharmaceutical industry, thereby facilitating the development of factual information as well.

### *4.2. The Role of European Patient Organizations*

A variety of information and educational material for patients about biosimilar medicines is made public by European patient organizations. Yet, the quality and level of detail vary among different associations, and it is not clear whether the identified information is effectively reaching the patient. This overview of information was based on a web-based screening. However, one should be aware that information made accessible via the internet will not reach every patient who needs such information. After all, not every citizen across Europe has the opportunity to consult the internet. That is why it remains important that healthcare providers fulfill their role to reach patients, and that patient associations themselves do not limit themselves to disseminating information via their websites.

Patient associations often refer to the biosimilar brochure of the European Commission, which was translated in all European languages in recent years. Some patient organizations have developed educational brochures or position statements about the use of biosimilars by themselves. They generally all agree on the fact that biosimilars are equal treatment options ensuring a sustainable healthcare system and underline that the decision to prescribe a biosimilar should be a shared decision involving the patient. Nonetheless, some patient

associations should be cautious not to fall prey to negatively framed, incorrect, or outdated information about biosimilars. Several patient associations provide detailed information on biosimilars, but express a rather negative attitude in particular towards transitioning from the reference product to a biosimilar (e.g., IDF Europe, Spanish Platform for Patient Organizations, and IFPA). Others provide or refer to incorrect or outdated information, such as EPDA, IDF Europe, and Flemish Patient Platform. The most pronounced example of this is IDF Europe, where they support their concerns about switching to biosimilar insulins by information that was incorrectly interpreted and taken out of context. Generally, national patient associations adopt the position on biosimilars of their European umbrella organization. However, this does not prevent national associations from formulating their own positions that differ from incorrect European ones. For example, the recommendations of the Dutch Diabetes Association about insulin biosimilars are in line with current scientific evidence and do therefore not correspond to those from IDF Europe [99]. A clear contrast was observed when looking at biosimilar information or educational material of DiCE and National Coalition of Dutch Patients. In particular, DiCE puts emphasis on the fact that if biosimilars are implemented on a wider scale, they could help closing the gap in gaining access to the highest standards of care for the treatment of colorectal cancer. The National Coalition of Dutch Patients repeatedly states that biosimilar medicines have the same efficacy, safety, and quality as their reference products. This is an example of positive framing since most information on biosimilars mentions that no meaningful differences are expected with originator biologicals, which is correct, yet framed more neutrally.

Information should always be evidence-based and therefore in line with the most recent scientific developments. As for all stakeholders, patient associations should distance themselves from positions or opinions about biosimilars that are not scientifically or incorrectly substantiated. Clear collaboration with independent and knowledgeable experts to develop such material is necessary to avoid incorrect information. With this overview, we have taken a critical look at the available information about biosimilars for patients developed by major European patient associations.

### *4.3. Future Perspectives*

During past years, the way that most treatment decisions are made has evolved towards shared decision-making [100]. The choice for an originator biological or a biosimilar must therefore be based on a coherent information stream to the patient. Several communication strategies have been identified in this review, guaranteeing correct information is provided adequately to patients. However, not all communication strategies have proven effective in actually increasing patient knowledge and confidence in biosimilars. Moreover, they have not proven to meet the appropriate behavioral objectives among patients. Future research assessing the actual impact of communication strategies based on a behavioral model could help clarify these unmet needs.

Most recommendations identified during this literature review are based on empirical grounds. Communication strategies emerging from theoretical concepts could be explored as well in the future. This would contribute to the overall picture on how to inform patients about biosimilar medicines and increase the robustness of the conclusions.

### *4.4. Strengths and Limitations of the Study*

The main conclusions of this study are based on a structured literature review and a web-based mapping of available information by European patient organizations. This study provides an overview of existing scientific literature on how to effectively inform patients about biosimilar medicines. The structured approach allows for reliable conclusions regarding information strategies for patients about biosimilars. This article is the first of its kind to compile the provided information of the major European patient organizations (i.e., EPF and IAPO members), with the purpose to have an overview of available information or educational material.

Although the literature review was conducted in a structured way, no systematic review was conducted and thus the selection of articles was not based on an agreement between two independent researchers. As a consequence, selection bias might have occurred during the title and abstract screening phase. Furthermore, the web-based mapping only allows for the collection of information that is publicly available on the websites of the patient associations of interest. Educational efforts that were not made available on their websites were therefore not included in this review. The researchers chose to include members of EPF and IAPO in the mapping of information, hence some available information on biosimilars by other European patient associations that are not members of these umbrella organizations might have been missed. Although the assessment of the tone in which patient associations report about biosimilars can be seen as subjective, it does provide an interesting picture of the overall attitude of each individual organization and the differences between them.

### **5. Conclusions**

It is important to set up a close collaboration between all stakeholders to develop and effectively disseminate correct information about biosimilars to patients, bringing together scientific associations, professional associations (including physicians, nurses, and pharmacists), regulatory authorities, and patient associations. Informing and educating patients on biosimilars should be part of a wider approach to support the adoption of biosimilars in Europe. European member states should consider informing patients on biosimilars in their policy frameworks more actively. It is imperative that European national authorities support biosimilar medicines to safeguard an affordable and sustainable healthcare system within their country.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/1424-824 7/14/2/117/s1, Table S1: Structured literature search methodology, and Figure S1: PRISMA flow diagram of the literature review.

**Author Contributions:** I.H., A.G.V., and Y.V. developed the idea for and were involved in the design of the study. Y.V. was involved in the data collection and drafted the initial version of the manuscript. I.H., A.G.V., S.S., and P.V.W. critically reviewed the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This manuscript is supported by KU Leuven and the Belgian National Institute for Health and Disability Insurance (NIHDI).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** This article is based on the preparatory work of master students' projects by Julie Brown, Ben Janssens, Seppe Lenaerts, Marie Pardon, and Olivia Wagman. The authors would like to thank all five master students for their efforts. The authors also express their appreciation towards the Pharmaceutical Policy Department of NIHDI for their support in this research project.

**Conflicts of Interest:** This research project is funded by the Belgian National Institute for Health and Disability Insurance (NIHDI). S.S., I.H. and A.G.V. have founded the KU Leuven Fund on Market Analysis of Biologics and Biosimilars following Loss of Exclusivity (MABEL). S.S. was involved in a stakeholder roundtable on biologics and biosimilars sponsored by Amgen, Pfizer and MSD; he has participated in advisory board meetings for Sandoz, Pfizer and Amgen; he has contributed to studies on biologics and biosimilars for Hospira (together with A.G.V. and I.H.), Celltrion, Mundipharma and Pfizer, and he has had speaking engagements for Amgen, Celltrion and Sandoz. A.G.V. is involved in consulting, advisory work and speaking engagements for a number of companies, a.o. AbbVie, Accord, Amgen, Biogen, EGA, Pfizer/Hospira, Mundipharma, Roche, Sandoz. P.V.W. acted as health care consultant to public and private organizations, including pharmaceutical companies and their

professional associations. All other authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Pharmaceuticals* Editorial Office E-mail: pharmaceuticals@mdpi.com www.mdpi.com/journal/pharmaceuticals

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34

www.mdpi.com ISBN 978-3-0365-6574-3