The Science Behind Biosimilars

Abstract

An important challenge currently facing rheumatologists and rheumatology health professionals is to gain an understanding of the landscape surrounding biosimilar agents, including the scientific, clinical, economic, and prescribing issues related to their use. The US Food and Drug Administration (FDA) defines a biosimilar as a biologic product that is "highly similar to" an approved biologic product (the "reference," "originator," or "bio-originator" product) and that has "no clinically meaningful differences" in safety or effectiveness as compared to the reference product 1. The number of biosimilars in development has burgeoned since the Biologics Price Competition and Innovation (BPCI) Act of 2009 was passed by the US Congress with the intent of reducing costs and thereby increasing patients' access to biologics. Another factor driving development of biosimilars is that the exclusivity period, as defined by the FDA, is approaching completion for several biologics. Nomenclature for biosimilars in the US uses the common nonproprietary name followed by a 4-letter suffix, so that each drug, both reference product and biosimilar, will have a unique nonproprietary name. Proprietary names, in contrast, are generated by the manufacturer and may vary from country to country for the same drug. As of December 2017, the FDA had approved 9 biosimilars, 6 of which are tumor necrosis factor (TNF) inhibitors, for rheumatologic and other inflammatory diseases: infliximab-dyyb (Inflectra), infliximab-abda (Renflexis), and infliximab-qbtx (Ixifi) (biosimilars of infliximab [Remicade]), etanercept-szzs (Erelzi) (biosimilar of etanercept [Enbrel]), and adalimumab-adbm (Cyltezo) and adalimumab-atto (Amjevita) (biosimilars of adalimumab [Humira]) 2-5. Many additional biosimilars of reference products with indications for both rheumatologic and nonrheumatologic diseases are under development. An updated list of approved biologics and biosimilars is available online in the "Purple Book" through the FDA Center for Drug Evaluation and Research 6. Biosimilars should not be confused with generic medications, despite the fact that for both, pathways for regulatory approval are abbreviated as compared to the pathways for approval of new drugs (Figure 1). The chemical structure of a generic small-molecule drug must be identical to that of its reference product. Thus, once chemical identity has been demonstrated, approval of a generic drug may rely upon a study demonstrating pharmacokinetic equivalence to the reference drug. Subsequent comparative clinical trials to demonstrate efficacy equivalent to that of the reference drug are not required. In contrast, because biologics are proteins produced in living cells, biosimilars usually are not identical to their reference products. The biosimilar manufacturer reverse-engineers the DNA sequence based on the primary protein sequence of the reference drug (available in the public domain). The synthesized gene is then transfected into host cells, in which transcription and translation mechanisms generate the core protein of the biosimilar. Several factors integral to the manufacturing process influence the degree to which the molecular structure of the biosimilar matches that of its reference product. These include choice of host cell, which influences posttranslational modifications of the protein (e.g., glycosylation 7), and methods used to purify and stabilize the final product. Importantly, to achieve FDA approval, a biosimilar must be highly similar in structure and function, equivalent in efficacy, and comparable in safety and immunogenicity to its reference product, despite potential slight molecular differences between the 2 agents. The BPCI Act established an abbreviated pathway for FDA approval of biosimilars. Subsequently, the FDA has issued guidance documents 8, 9 that detail the requirements for demonstrating biosimilarity (Figure 1). Similar guidance has been issued by the European Medicines Agency (EMA), an analogous regulatory body in Europe 10. The FDA has adopted a "totality of the evidence" approach to the approval of biosimilars, in which multiple lines of evidence are integrated to determine biosimilarity by the reduction of "residual uncertainty." The regulatory pathway for biosimilar development (351[k]) mandates a sequential series of rigorous comparative studies to demonstrate biosimilarity of the biosimilar candidate to its reference product 8-10. It places heavy emphasis on preclinical analytical comparison of structure and function between the proposed biosimilar and its reference product (Figure 2), as well as comparative clinical studies of pharmacokinetic parameters and, if a relevant biomarker is available, of pharmacodynamic parameters. The FDA also requires assessment of the stability and formulation of the biosimilar. In addition, at least one clinical trial comparing the proposed biosimilar to its reference product, in patients with a disease for which the reference product is indicated, is required in order to assess equivalence of efficacy and comparability of safety and immunogenicity. Once a biosimilar has satisfied the requirements of this regulatory pathway, both patients and providers should expect that clinical outcomes with the biosimilar will parallel the accumulated experience with the use of its reference product. Drift and evolution. Batches of originator biologics administered to patients today often differ from those of the same drugs that were dispensed in previous years, particularly for drugs introduced in the late 1990s and early 2000s 11. Such differences are due to both drift and evolution—the intentional and unintentional alterations in manufacturing that result in lot-to-lot variability of both monoclonal antibodies and fusion proteins 12. The manufacturers of virtually every originator biologic have made intentional, post-approval changes to their manufacturing processes, each of which is reviewed and approved by regulatory agencies 13, 14. Regulatory agencies permit such changes if they fall within predetermined proven acceptable ranges when the products are carefully monitored using sophisticated analytical techniques. Thus, for originator biologics, the effects of drift and evolution accrue over time after their approval. This will continue to occur, not only for reference drugs, but also for approved biosimilars. Despite the occurrence of drift and evolution in the manufacture of virtually all commercially available biologic medications, these originator biologics have continued to be safe and effective. Accordingly, favorable outcomes in comparative analytical, clinical, and immunologic studies, with rigorous regulatory oversight and careful postmarketing pharmacovigilance, should reassure patients and prescribers that biosimilars will be both as safe and as effective as their reference products. Immunogenicity. Another critical component of the biosimilar approval pathway strives to ensure comparable immunogenicity of biosimilars and their reference products. This includes measurement of both binding and neutralizing antidrug antibodies, which have been well characterized in patients treated with originator biologics for rheumatologic conditions 15, 16. Antidrug antibodies have been detected in patients taking originator biologics, such as infliximab or adalimumab, and are associated with lower trough drug concentrations, reduced efficacy, and increased frequency of infusion reactions 17, 18. The prevalence of antidrug antibodies varies widely among studies, depending on the assay used for their detection 16. Protein aggregation or impurities introduced during the manufacturing process of any biologic agent may stimulate antidrug antibody formation. Although the primary structures of biosimilars must be identical to those of their reference products, there may be differences in secondary (e.g., α-helix and β-sheet), tertiary (e.g., disulfide and salt bridges), or quaternary (e.g., protein subunit interactions) structure as well as differences in glycosylation and other posttranslational modifications. Theoretically, these structural differences may also induce the development of antidrug antibodies in patients initially exposed to the reference product 7. Thus, both the FDA and EMA mandate at least 1 clinical trial in which comparative immunogenicity of a biosimilar and its reference product is assessed 8, 10. For some biosimilars, regulatory agencies may require a clinical trial that includes a single crossover from the reference product to the proposed biosimilar, to assess whether this transition induces antidrug antibody formation with a resultant loss of efficacy. If a biosimilar is intended for long-term administration, as is the case for rheumatologic indications, it is recommended that immunogenicity data be acquired over at least 1 year. If immunogenicity findings are to be extrapolated from one disease to additional indications, the subjects being studied should be those most likely to develop antidrug antibodies, such as patients not receiving concomitant immunosuppressive medications. For instance, infliximab-dyyb and adalimumab-atto were studied as monotherapy in ankylosing spondylitis (AS) and psoriasis, respectively 19, 20. The nature of the antidrug antibodies (e.g., neutralizing versus non-neutralizing) and the relative magnitude of the immune responses to the biosimilar and its reference product were assessed in order to characterize the immunogenicity risk and determine the clinical relevance of the antidrug antibodies. Any candidate biosimilar that is found to be more immunogenic than its reference product will fail to meet criteria for biosimilarity 10. Importantly, a biosimilar found to be less immunogenic than its reference product may be approved, provided that a subset analysis of clinical trial subjects who did not develop antidrug antibodies reveals comparable efficacy and safety. For example, among patients with rheumatoid arthritis (RA) treated through 24 weeks, antidrug antibodies were observed in 13.1% of those receiving reference etanercept but in only 0.7% of those receiving the biosimilar etanercept SB4 21. There was a modest effect of antidrug antibodies on reducing trough drug levels, but this had no significant effect on efficacy or safety. This was most likely due to the transient expression of these antibodies 22. SB4 (etanercept) is now marketed as Benepali in the European Union (EU) and as Brenzysin in South Korea, but it is not available in the US. As methods to detect molecular changes associated with drift and evolution have improved, assays to detect antidrug antibodies have evolved over time to become more sensitive. In early studies of therapeutic monoclonal anti-TNF antibodies, antidrug antibodies were identified in only a small proportion of subjects (~8–12%) 23, 24. This low detection rate was due primarily to the inability of assays to measure antidrug antibodies in the presence of the drug. In recent clinical trials, antidrug antibodies have been detected in a larger proportion of patients, using acid dissociation to allow measurement of antidrug antibodies in the presence of the drug, along with a more sensitive bridging enzyme-linked immunosorbent assay 25. In the PLANETRA trial, antidrug antibodies were detected in 48% of patients receiving reference infliximab and in an identical proportion of those treated with the biosimilar (infliximab-dyyb), through 30 weeks of treatment 26. In the clinical trial comparing adalimumab-atto to reference adalimumab, the frequency of antidrug antibodies was similar in both groups of patients though 26 weeks (38%), as was the incidence of neutralizing antidrug antibodies (11.1% with reference adalimumab versus 9.1% with biosimilar) 20. In summary, the frequency of binding and neutralizing antidrug antibodies has been similar between biosimilars approved in the US to date and their reference products, and there have been no signals to suggest a differential effect of antidrug antibodies on efficacy, safety, or patient outcomes between biosimilars and their reference products. An important implication of the comparative immunogenicity studies carried out to date is that a patient who develops antibodies to a reference drug with resultant loss of clinical response should not be switched to its biosimilar. This recommendation is based on the finding that antidrug antibodies to a reference product cross-react with its biosimilar, as shown in studies of infliximab-dyyb conducted in RA, AS, and inflammatory bowel disease (IBD) patient groups 27, 28. Thus, patients in whom loss of efficacy is attributable to antidrug antibodies directed against a reference drug or its biosimilar should subsequently receive an unrelated therapeutic agent. Whether multiple switches between a reference drug and various biosimilars might result in increased immunogenicity has not yet been studied adequately. However, the EGALITY study 29, which compared the biosimilar etanercept-szzs to reference etanercept, included 3 switches between the 2 products (each of 6 weeks' duration), and demonstrated no loss of efficacy or increase in adverse events with repeated switching, compared to continued treatment without switching. Postmarketing pharmacovigilance using observational registry data will be critical to assess the effect of switching on immunogenicity. The FDA and the EMA require both manufacturers of reference biologics and manufacturers of biosimilars to conduct the same postmarketing pharmacovigilance after any manufacturing change 30. Enrollment of biosimilars-treated patients into registries, such as the Rheumatology Informatics System for Effectiveness (the registry of the American College of Rheumatology [ACR] 31) and the pediatric registry sponsored by the Childhood Arthritis and Rheumatology Research Alliance in the US, along with comprehensive spontaneous adverse event reporting in both adults and children, should facilitate detection of any loss of efficacy or toxicity as signals of clinically significant immunogenicity. Extrapolation of indications. When a biosimilar is approved by the FDA for use in one indication, it may be approved simultaneously in any or all of the other indications for which the reference product has been approved, without the requirement for clinical trial data from each disease. This process, which is known as "extrapolation of indications," eliminates the costs and delays introduced by the need to replicate other previously successful phase III randomized controlled trials that had been conducted with the reference product. The FDA and EMA have provided guidelines for justification of extrapolation of indications for biosimilars 10. Once a biosimilar has met criteria for approval, the cumulative experience with its reference product can be applied to justify extrapolation of the approval of the biosimilar to those other indications in which it was not initially studied. For example, the infliximab biosimilar infliximab-dyyb (referred to as CT-P13 during its development) was evaluated extensively in comparison to reference infliximab (Remicade) using functional assays such as complement-dependent cytotoxicity, antibody-dependent cell-mediated cytotoxicity, and Fcγ receptor III binding. The results of these tests showed this biosimilar to be highly similar to infliximab, according to statistical parameters established by the FDA 32. Infliximab-dyyb was then compared to reference infliximab in randomized clinical trials in RA (the PLANETRA trial), in combination with methotrexate (MTX), and in AS without concomitant MTX (the PLANETAS trial). The PLANETAS study demonstrated equivalent pharmacokinetic parameters and similar clinical outcomes, and the PLANETRA study demonstrated equivalent efficacy. In both studies, safety and immunogenicity of the biosimilar were comparable to those of the reference product 19, 33. Data from the preclinical analytical and functional studies, along with the results of the clinical trials, supported FDA approval of infliximab-dyyb with extrapolation to all other diseases for which reference infliximab had already been approved, but which were no longer protected by exclusivity. Similar data on the adalimumab biosimilars BI 695501 (adalimumab-adbm) and ABP 501 (adalimumab-atto), for the etanercept biosimilar GP2015 (etanercept-szzs), and for the infliximab biosimilars SB2 (infliximab-abda) and infliximab-qbtx (PF-06438179/GP1111) 5 resulted in extrapolation to all nonexclusive indications for which the respective reference products had been approved 29, 34-37. Although none of the biosimilars approved to treat rheumatologic and other inflammatory diseases has been studied in children, several have been approved for pediatric indications by extrapolation. In the US, infliximab-dyyb, infliximab-abda, and infliximab-qbtx are all approved for treatment of pediatric Crohn's disease. In the EU, infliximab-dyyb and infliximab-abda are both approved for treatment of pediatric Crohn's disease and pediatric ulcerative colitis. The adalimumab biosimilars adalimumab-adbm and adalimumab-atto and the etanercept biosimilar etanercept-szzs are both approved for treatment of polyarticular juvenile inflammatory arthritis. The benefits and risks of using biosimilars to treat children likely are similar to those in adults. However, since children frequently metabolize drugs, including biologics, more rapidly than adults, it may be important to conduct pharmacodynamic and pharmacokinetic studies in children. The FDA encourages manufacturers to study biosimilars in the pediatric population during product development 38. After approval, postmarketing surveillance of biosimilars should be conducted in children and adolescents as well as in adults, since potential immunogenicity may be of particular importance in these younger patients with chronic diseases, who might be exposed to several biologic agents during their lifetime. Extrapolation continues to be an area of uneasiness among clinicians, who are surprised to find that a biosimilar may be approved for an indication, such as IBD, in the absence of clinical trials conducted in those patient populations 39, 40. Indeed, lack of confidence among clinicians in extrapolation of indications may limit acceptance of biosimilars in geographic areas where utilization is not mandatory. To date, however, "real-world" controlled trials such as NOR-SWITCH 41, registry studies such as DANBIO 42, and preliminary results of a postmarketing 54-week randomized clinical trial in Crohn's disease 43 have demonstrated the efficacy and safety of biosimilar infliximab-dyyb to be comparable to those of reference infliximab in a range of diseases, providing reassuring evidence to support regulated extrapolation of indications for biosimilars. Substitution and interchangeability. Changing from one biologic to another due to inadequate efficacy or side effects, as directed by a provider in partnership with the patient, occurs commonly and requires that the provider write a new prescription. "Substitution" is the FDA-preferred term to describe a change made by someone other than the prescriber (such as a pharmacist) and is regulated by state law. Non-medical substitution (also known as payor substitution) refers to a change imposed on a medically stable patient for reasons unrelated to a drug's efficacy or in that and benefits of drug for and important in Such changes in are made in response to or that determine which biologics are due to and by the is a that may be to a biosimilar under the pathway that allow for substitution 1. To be as a biosimilar must be to the same clinical result as the reference product in any In addition, administered more than once to an the risk of an adverse event or efficacy with between the reference product and the biosimilar must not be than the risk of using the reference product without such or 1. In 2017, the FDA guidance of of an already approved biosimilar with its reference product The FDA on postmarketing pharmacovigilance as well as data from at least 1 controlled switching comparing treatment with the reference product with the biosimilar to treatment with the reference product. Such a clinical trial should of an during which all subjects are treated with the reference product, followed by a during which subjects are randomized to the switching or to receive treatment with the reference product. In the switching subjects be over at least 3 to the product (e.g., reference product to biosimilar to reference product to such as drug and area under the from the the of the period, should as the primary for such should include other pharmacokinetic drug and time to and of efficacy, safety, and immunogenicity. The FDA that the used in such studies be a reference product that is in the US. It should be that the biosimilar product has already preclinical (Figure To date, the pathway has not been no manufacturer has and no biosimilar has been the However, in of most and the of have passed the substitution of biosimilars and several more are in various of similar vary by the the to substitution by as or medically on the prescription. In some the prescriber must be of any substitution made at a vary and from to to prescribers of a In provider that the after a date they are in the already has require that the patient be if a biosimilar has been for the drug, that the patient must to the or that the must to the patient the in between the reference biologic and the biosimilar. in some must of biologic medications, and the state must a public list of for which is and Changing to a biosimilar is an intentional therapeutic that is by a health provider in partnership with the Such a change often is made for since a biosimilar usually costs less than its reference product. However, this change may be made for reasons when a patient is not to a drug other than the reference product and the biosimilar is as the treatment because of its lower The term is used in the BPCI Act when to to or from a biosimilar which has been as trials comparing biosimilars to their reference products include an during which subjects who were treated initially with the reference drug are to the biosimilar for the of the The efficacy and safety of the biosimilar among subjects who from the reference product are compared to the efficacy and safety among those who the biosimilar the In of the PLANETAS study and the PLANETRA to the biosimilar infliximab-dyyb after weeks of treatment with reference infliximab did not result in a loss of efficacy or an increase in adverse events or immunogenicity results have been observed in the of other studies as well as in postmarketing observational studies from a reference drug to its biosimilar was studied in a phase clinical trial comparing the effect of from reference infliximab to the biosimilar infliximab-dyyb versus of treatment with reference infliximab in various approved indications This study included patients with RA, psoriasis, Crohn's or ulcerative each of whom had been treated with a stable of reference infliximab for at least 6 The primary was in between the and the patient that increased disease a change in treatment by The NOR-SWITCH trial demonstrated a of of from the originator biologic to the biosimilar for the of patients with the various and immunogenicity were comparable between the 2 treatment However, the study did not have statistical to the treatment among patients with any disease. expect that and substitution will become as common in the US as it has in Europe and the of the on clinical trial and "real-world" observational data on between reference products and biosimilars, and on understanding of product not expect that there will be issues efficacy and safety. However, postmarketing surveillance of both biosimilars and their reference products as the in which patients may multiple changes as a result of and final in the use of biosimilars is issues patient access to these agents. The only of a biosimilar over its reference product is lower since the 2 drugs should be The degree to which the of biosimilars in the US will the of biologic and who will from any to be There are to measure differences of a biosimilar compared to its reference a of these methods is the of this However, multiple the of biosimilars. The US biosimilar relevant to infliximab-dyyb, at a compared to reference infliximab, a reduction that was with a 2009 Subsequent be if from the and infliximab biosimilars, infliximab-abda and and will continue to in the of biosimilars. with on a that is by to the by the manufacturer according to (Figure this significant in the of biologics in the US, or due to that manufacturers have the to increase in order to a more favorable on the and to based on drug If the of a reference product were to the of its biosimilar by and provided to by then the be and biosimilar might be biosimilars to from the there no longer be on reference product manufacturers to lower and the of patients Thus, biosimilars an important in a to drug under under it will likely several for to the US health from the use of biosimilars to be due to multiple the of originator biologics increased in the 6 between of the biosimilar approval pathway and approval of the biosimilars (e.g., the of reference infliximab increased by This is similar to the increase in the of that over the 6 to its the biosimilars in the US and were at only less than their reference products, will be for This when multiple biosimilars for a reference product become available, or if a biosimilar is approved as being and is by for its reference product or other biosimilars. To the use of biosimilars, and drug with patients' costs and provider however, patients with are likely to have similar for both biosimilars and originator biologics, due to or patient patients' costs for biosimilars in the likely will be than for originator biologics due to a that than patient in the (also known as the for biosimilar on 2017, the for and a change of biosimilars under biosimilars of the same originator biologic will be the established in biosimilars of a common originator biologic were under a single System and based upon the of those biosimilars. The change an by to of more biosimilar products, and long-term for biosimilars in the US are than those in with For example, the public in and have biosimilar infliximab-dyyb at a relative to the of reference infliximab This results from the of these to through use of the biosimilar. In the US may be reduced by modest of the biosimilar and the of compared to that of a larger single public in to drug is by more by and the mechanisms that in the This public by groups of these drugs through and the of to drug to the on patients receiving these and other medications. and are to biosimilars will be updated as the relevant and regulatory and as they become integrated into clinical The of the on biosimilars supported the biosimilars, but as they were being and that biosimilars have been used in with acquired data their and as the US is on the of a similar the is to its this of it is reassuring to the with which the FDA and other regulatory agencies the have evaluated biosimilars. providers should now biosimilars, where into to treat patients with rheumatologic It is important to a of approved biosimilars and to and guidelines the development and use of new biosimilars. 1 available for health providers and patients, including a to the on of biosimilars. between and patients will be critical to and the use of these available agents. that the use of biosimilars will patient access to biologic agents, continued of health to be at a lower to the The the of the on and who reviewed the and provided the it for important and approved the final to be