Biosimilar infliximab for inflammatory bowel disease: from concepts to clinical practice. Case study illustrated with CT-P13

Abstract

The introduction of biologic drugs represents the most significant advance in the management of immune-mediated inflammatory diseases for a decade. However, complex proteins are expensive to produce and manufacture. Biosimilar versions of established biologics are becoming available as another version of the reference medicinal product and are expected to provide substantial cost savings. However, because of their complexity, the approval of biosimilars requires strict controls to ensure that all therapeutically relevant characteristics are comparable to the reference medicinal product. This review summarizes the scientific principles and data requirements underpinning regulatory approval of biosimilars and the assumptions that enable extrapolation of data between indications. These important concepts are exemplified by CT-P13 (Remsima^®, Inflectra^®), the first biosimilar monoclonal antibody approved in Europe.

Keywords:

• biosimilar
• CT-P13
• extrapolation
• infliximab
• manufacture
• regulatory approval

Biologics or biopharmaceuticals are products for medical use produced in, or extracted from, a living system (i.e., microorganism, plant or animal cell). In contrast to conventional, small-molecule drugs, biologics are frequently large, structurally complex proteins. Current biologics include blood products, vaccines, gene therapies, recombinant proteins and monoclonal antibodies (mAbs) [1]. Biologics have advanced the treatment of many chronic and life-threatening diseases that together represent a growing global public health challenge, from cancer to inflammatory diseases such as Crohn’s disease (CD) and ulcerative colitis [1–4]. Consequently, biologics comprise an increasingly important segment of the pharmaceutical industry.

A biosimilar is another version of an already licensed biologic, with no meaningful differences from the reference medicinal product (RMP) in terms of quality, physicochemical properties, biologic activity, safety or efficacy based on a comprehensive comparability exercise [5,6]. Just as patent expiry for conventional drugs has led to the marketing of generic versions, the expiry of data protection and patents for original biologics has prompted development of biosimilars. In parallel, the lower cost of biosimilars, which is the major advantage compared with the original RMPs, will further accelerate this trend and may also lead to patients gaining greater access to these treatments [7].

However, biosimilars for the therapy of chronic inflammatory or malignant diseases enter the market with more to be understood about their clinical use than is typically the case for generics. In indications for which generics have been introduced, the dose and prescribing guidelines for the therapeutic substance are comprehensively known. This is not the case, for example, for anti-TNF biosimilars in inflammatory bowel disease (IBD), where there are still questions being addressed about the use of RMPs, with the optimum dose, administration frequency and treatment time point for an individual patient still subject to discussion [8].

In this paper, the authors discuss the development and manufacturing process of biosimilars, review the scientific rationale behind the approval of these drugs, and provide an introduction to the topic of extrapolation of indications. Finally, to illustrate these concepts, they discuss data obtained with CT-P13 (Remsima^®, Inflectra^®), the first mAb, anti-TNF biosimilar that was approved in Europe.

Development & manufacturing process of biologics & biosimilars

Small-molecule drugs are synthesized using highly reproducible chemical processes and therefore generic versions can be manufactured using standardized procedures. Furthermore, quality assurance during manufacture can utilize analytic techniques to demonstrate that a generic version of a small molecule is chemically and structurally identical to the original drug. Therefore, the production and approval of a generic version of a small-molecule drug represents an exercise with foreseeable challenges lying only in the chemical production and manufacturing process.

In contrast, similar versions to large biologic drugs are associated with inherent complexity due to the tertiary and quaternary structure of their protein molecules. Their production involves a complicated manufacturing process including a living organism (i.e., a cell line) that cannot be copied and that remains private property of the original manufacturer. These drugs cannot be chemically synthesized and, therefore, must be produced from a similar living system. Consequently, potential sources of variation in the manufacture of biologics are numerous, meaning that biosimilars will be never identical to the RMP. Furthermore, batch-to-batch variation occurs during production of both a biosimilar and the RMP itself, and production may have to be adjusted over time to avoid an uncontrolled drift from the original composition that once led to approval. Therefore, manufacture of biosimilars cannot be considered in the same way as generic versions of small molecules [2,9].

The complexity of the manufacturing process for biologic drugs is illustrated in Figure 1. Inherent variability found in the biologic expression systems used to manufacture biologics will result in some heterogeneity in the final product, even between batches [6,10]. Furthermore, all steps in production, purification, formulation and storage can have an effect on the clinical and biologic properties of the finished product [1]. Any alteration of the manufacturing process can create functional changes in the product that can result from subtle structural alterations or changes in post-translational modifications. For example, a higher degree of fucosylation may lead to decreased receptor binding of Fc-bearing antibodies and possibly decreased cytotoxicity [11]. In addition, environmental conditions such as formulation, light, temperature or humidity can affect the characteristics of proteins, like some aspects of packaging and delivery processes can affect them [1,9]. This is important, since the biologic activity of a recombinant protein is closely related to its structure and physicochemical characteristics. Even small alterations in protein characteristics can lead to changes in the efficacy or toxicity of a biologic [12]. A further issue is that the protection of proprietary information concerning production of the RMP means that biosimilar manufacturers must develop their own processes (and should update them with regard to improving patient safety), which makes differences in production techniques inevitable [6,10,12,13]. Taken together, this means that maintenance of strict controls in manufacturing and quality assurance processes is essential in order to ensure consistency and minimize the differences between biosimilars and RMPs [1,14,15].

Figure 1. Schematic overview of the complexity of the manufacturing and production process for a recombinant biologic. The first steps in the production of a recombinant biologic include cloning of the relevant DNA sequence into an expression vector, its transfer into a suitable host cell, followed by expression screening and selection. Once selected, the biologic-expressing cell line is expanded and frozen down in aliquots to generate master stocks for storage. The initial step in large-scale production is small-scale expansion and quality control of an aliquot of the master stock. The master stock is expanded in a seed bioreactor and then transferred to production bioreactors which maintain the optimum conditions for the expression of the active recombinant protein. Cells are then harvested and separated from the secreted recombinant protein by ultracentrifugation. Recombinant protein is harvested from the supernatants by capture chromatography. The protein then undergoes a virus inactivation step followed by further chromatography and filtration to remove viral particles. Ultrafiltration completes the procedure. The drug then undergoes extensive characterization before formulation and packaging.

Development trends & challenges in the regulatory approval of biosimilars

As a result of the issues discussed above, regulatory authorities clearly recognize that the types of assessments used to evaluate and approve small-molecule generics are not applicable to biosimilars. Consequently, methods to determine comparability of biosimilars and RMPs are based on a combination of analytic testing, biologic assays and, in some cases, both non-clinical and clinical data [6]. The language of calling a biosimilar ‘another version of the RMP’ underscores the willingness of regulatory agencies to accept subtle differences as long as they are within the narrow definition of still being a ‘version’ [6]. Although, in the great majority of cases, detailed pharmacokinetic (PK), efficacy and safety analyses are required to ensure that biosimilars are comparable to their RMP [10,15–17], general guidance for the development of biosimilars issued by the EMA and the US FDA indicate that in certain specific circumstances, a confirmatory clinical trial may not be necessary, depending on the identification of a clinically relevant pharmacodynamic endpoint in an appropriate population for study [9,18]. However, all biosimilar class-specific guidelines issued by the EMA on the development of biosimilars of somatotropin, follicle-stimulating hormone, erythropoietin, filgrastim, insulin and mAbs stipulate that clinical trials should be carried out [19–24]. Following completion of this type of detailed study program, iterative modifications in the manufacturing procedure can occur to increase output in the lead up to full production. Manufacturing site changes may also be necessary and these need to be carefully controlled because of the potential to result in changes in the quality of the final product. After any such changes, extensive analysis of pre- and post-change product is required in the form of a comparability exercise, with submission of data for subsequent approval by regulatory authorities [12]. Analytic methods are currently well developed, allowing detection of even small changes in quality attributes. These can be used to detect batch-to-batch consistency and to sensitively monitor for any variability in the manufacturing process.

It is important to recognize that many of these challenges are neither new nor limited to the manufacture of biosimilars. With any approved biologic or RMP, manufacturing processes will generally have been modified several times over the years [25]. It is important to note in this context that the manufacture of biologics (in contrast to chemical synthesis of small molecules) does not lead to a uniform molecule, but rather a spectrum of molecules. Therefore, currently used biologics may not be exactly the same as the ‘original’, but more accurately each production batch represents ‘a version of the original’ [6]. For example, analysis of three glycosylated recombinant therapeutic proteins – Aranesp^® (darbepoetin alfa), Rituxan^®/Mabthera^® (rituximab) and Enbrel^® (etanercept) – sourced from the market between 2007 and 2010 highlights the sensitivity of biologics to manufacturing process change and informs the debate on acceptable changes in quality [26]. For each of the three drugs, the analysis revealed a step change in the glycosylation profile during that time period. An example is shown in Figure 2. Also, for rituximab, differences in glycosylation were accompanied by a change in the amount of the C-terminal lysine and the N-terminal glutamine variants. However, such changes are common for mAbs and are unlikely to impact upon the biologic characteristics of the molecule [26]. However, all three biologics remained on the market with unaltered labels during the tested time frame, indicating that the changes observed were not predicted to result in an altered clinical profile.

Figure 2. Example of variation in biologic drugs: comparative analysis (capillary zone electrophoresis) of Aranesp^® batches up to April 2010 (pre-change) and after September 2010 (post-change). (A) Relative content of the individual isoforms (different forms of the same protein) of the pre-change (n = 18) and the post-change (n = 4) batches. (B) Representative electropherograms; peaks are labeled with the isoform number. Batches expiring in April 2010 showed a higher sialylation rate than the batches expiring after 2010.

Extrapolation of indications for biosimilars

Extrapolation of indications for biosimilars involves extending the interpretation of data from clinical studies completed for a primary indication to other indications. Extrapolation of efficacy and safety data between indications may be considered if biosimilarity to the RMP has been demonstrated in a comprehensive comparability program [10]. Also, the molecular mechanism of action of the biosimilar should be the same in the different indications [10]. Around the world, different regulatory authorities have issued guidance on extrapolation for approval of biosimilars. The EMA released final guidance on this matter on October 23, 2014 [18]. This guidance suggests that extrapolation may be possible if clinical similarity can be shown in the key indication during comparability studies.

The FDA has also issued guidance and states: ‘If the proposed product meets the statutory requirements for licensure as a biosimilar product under section 351(k) of the Public Health Service Act based on, among other things, data derived from a clinical study or studies sufficient to demonstrate safety, purity, and potency in an appropriate condition of use, the applicant may seek licensure of the proposed product for one or more additional conditions of use for which the reference product is licensed’ [9].

Therefore, guidance from both EMA and FDA is that the manufacturer needs to provide sufficient scientific justification for extrapolating clinical data from the key indication to support biosimilarity in other indications for which approval is sought. Provided the mechanism of action of the biosimilar and/or the pathophysiologic targets is/are the same in the extrapolated indication, then extrapolation should not be an issue.

If this is not the case, for example, when the mechanism of action is complicated and involves multiple receptors or sites of effect, additional data including in vitro functional tests or in vivo pharmacodynamic studies are necessary to demonstrate the similarity of the biosimilar and the RMP [18].

Biosimilars have now been marketed in Europe for several years and have performed as expected in all licensed indications, including those extrapolated from data generated in other conditions [6]. Several examples of extrapolation with EMA-approved biosimilars are described below.

Biosimilar filgrastim

Filgrastim is a drug used for the treatment of neutropenia of various etiologies and for the mobilization of hematopoietic progenitor cells from the bone marrow to peripheral blood in patients and healthy donors. Several filgrastim biosimilars have been licensed in Europe based on comparable efficacy and safety profiles with their RMP in studies in patients with chemotherapy-induced neutropenia. In each case, all indications of the originator product were approved, even though clinical trials were completed in a single indication [27]. This decision to extrapolate indications to include hematopoietic progenitor cell mobilization was controversial, and expert organizations called for additional data specifically confirming efficacy and safety in that indication [28–30]. Post-marketing studies have since confirmed the efficacy and safety of biosimilar filgrastim in all approved indications, including hematopoietic progenitor cell mobilization, demonstrating the robustness of the biosimilar assessment and approval process in this case [6,28–30]. More recently, in March 2015, a biosimilar of filgrastim (Zarxio^®) became the first biosimilar to be approved by the FDA. Notably, the FDA also approved this biosimilar across all indications for which the originator drug is approved.

Biosimilar epoetin

Epoetin is approved for treatment of renal anemia and chemotherapy-induced anemia. A major concern associated with epoetin is related to the immunogenicity of products that can result in pure red cell aplasia caused by an interaction between the stabilizer polysorbate 80 and leachates from uncoated rubber in the stoppers of prefilled syringes [31–33]. Consequently, when a biosimilar epoetin is being developed, the EMA requires the manufacturer to provide a large amount of functional, physicochemical, biologic and clinical data (including efficacy, safety and immunogenicity) and at least one clinical trial in patients with renal anemia, because this population is most at risk from pure red-cell aplasia [21]. With these requirements in place, the indication for biosimilar epoetin has been extrapolated to include chemotherapy-induced anemia, and no specific efficacy or safety issues have been identified in clinical practice for EMA-approved biosimilar epoetins [21,31,32].

The strict scrutiny of the properties of biosimilars applied by biopharmaceutical companies and regulatory authorities is exemplified by the case of a biosimilar of epoetin that did not receive approval for subcutaneous use because of increased immunogenicity; clinical trials investigating its safety were also halted [34]. A review of post-marketing surveillance data and clinical trials encompassing 12,039 patients, switched from reference biologics including epoetin-alfa, filgrastim and somatotropin to biosimilars, highlighted no safety concerns [35]. Clinically relevant increases of immunogenicity-associated events have not been reported in any marketed biosimilar and no product-specific label changes have been required for safety reasons [36].

Biosimilar infliximab

Infliximab is a chimeric, human–mouse, anti-TNF mAb. In a landmark decision in September 2013, CT-P13 became the first mAb approved through the EMA’s biosimilars regulatory pathway [37]. Approved indications of the RMP (Remicade^®) by the EMA are rheumatoid arthritis (RA), ankylosing spondylitis (AS), adult and pediatric CD, adult and pediatric ulcerative colitis, psoriatic arthritis and psoriasis. Biosimilar infliximab was licensed for all these same indications in the EU, South Korea [38] and Japan [39]. However, extrapolation from rheumatic diseases to IBD was not approved in Canada [40]. This was because the Canadian authorities felt that the correlation between the mechanism of action of infliximab in rheumatology and psoriasis indications and in IBD has not been fully and conclusively demonstrated [41]. Further information on these issues and the basis for the approval of CT-P13 for IBD by the majority of regulatory authorities is described in the following section.

CT-P13: the first mAb biosimilar approved by the EMA

CT-P13 and infliximab RMP are produced in independent cell lines developed from the same murine hybridoma cell type (Sp2/0-Ag14) and share many downstream processing steps [42]. The approval of CT-P13 was based on the results of a comprehensive and rigorous non-clinical and clinical evaluation program described here [37].

Physiochemical properties, biologic activity & non-clinical evaluations of CT-P13

The formulation of CT-P13 has been shown to be highly similar to that of RMP and both products share comparable physicochemical characteristics [43,44]. CT-P13 has been shown to be similar to RMP in terms of primary, secondary and higher order structure, including protein folding, and post-translational glycosylation. Also, a series of qualitative and quantitative formulation studies demonstrated that the CT-P13 formulation compares closely to RMP in terms of product stability and quality [37,44].

Studies have shown that the binding of both CT-P13 and RMP to soluble and transmembrane TNF was comparable, as were the functions mediated by transmembrane binding, including reverse signaling and apoptosis [37]. Supplementary tests showed similar inhibition of CT-P13 and RMP on the direct effects of TNF on epithelial cells, which is an important consideration in CD [37]. Also, in these studies, CT-P13 and RMP demonstrated similar binding affinities for the Fcγ receptors (FcγRI, FcγRIIa and FcRn) [37]. A difference in the relative binding affinities for FcγRIIIa was identified, translating into lower antibody-dependent cell-mediated cytotoxicity activity of the biosimilar when NK cells were used as effector cells. However, antibody-dependent cell-mediated cytotoxicity activity was highly similar when peripheral blood mononuclear cells or whole blood was employed as effector cells, instead of NK cells [6,26,37]. The comparability of CT-P13 and infliximab RMP was investigated in mixed lymphocyte reaction assays using peripheral blood mononuclear cells from healthy and CD donors. No differences were detected between CT-P13 and RMP in the proportion of regulatory macrophages induced or in the inhibition of T-cell proliferation when genotype-matched PMBCs were used [37].

PK and toxicology aspects of CT-P13 and RMP were also highly similar in preclinical studies. In terms of PK, when single intravenous doses of CT-P13 or RMP were administered to rats, the geometric mean area under the curve (AUC) of serum concentration–time from time zero to the last measurable concentration (AUC_0–t) ratio of CT-P13 and RMP was 96.7% for a dose of 10 mg/kg (90% CI: 79.69–117.23) and 112.7% (90% CI: 87.30–145.49) for a dose of 50 mg/kg [37]. Two pivotal 2-week repeat dose toxicity studies were performed to compare the toxicity profiles in rats. In an initial study, 10 male and 10 female rats, aged 10–11 weeks, were dosed intravenously with 0, 10 or 40 mg/kg of CT-P13 or RMP at day 1 and 8. In the second study, 10 male and 10 female rats, aged 6–7 weeks, were dosed intravenously with 0, 10 or 50 mg/kg of CT-P13 or RMP at day 1 and 8. Toxicities were observed until day 15 in both studies. The treatment-related findings were similar in magnitude and frequency between CT-P13 and RMP [37].

PK in healthy volunteers

The PK of CT-P13 in healthy volunteers was characterized in a double-blind Phase I study that compared the biosimilar with two RMP products from different manufacturing sources [45]. A total of 213 volunteers were randomized 1:1:1 (i.e., 71 in each group) to receive a single 5 mg/kg dose of CT-P13, RMP sourced from Europe, or RMP sourced from USA. Geometric mean values of three primary endpoints (maximum serum concentration [C_max], AUC from time zero to last quantifiable concentration [AUC_last] and AUC from time zero to infinity [AUC_inf]) of the treatment groups are shown in Table 1. For each treatment comparison, the 90% CIs of the ratio of geometric means were entirely contained within the predefined equivalence limits of 80–125%.

Table 1. Pharmacokinetic exposure estimates for CT-P13 and two different versions of reference medicinal product (mean ± standard deviation) [45].

Parameters (units)	CT-P13 (n = 71)	EU-RMP (n = 71)	US-RMP (n = 71)
Cmax (μg/ml)	127.6 ± 21.6	121.3 ± 19.7	119.9 ± 19.5
AUClast (h*μg/ml)	31,343.2 ± 7107.3	30,669.0 ± 6020.5	31,629.6 ± 5886.8
AUCinf (h*μg/ml)	33,212.3 ± 8326.1	32,986.6 ± 7806.8	34,363.3 ± 8004.2

AUC_inf: Area under the concentration–time curve from time zero to infinity; AUC_last: Area under the concentration–time curve from time zero to last quantifiable concentration; C_max: Maximum serum concentration; EU-RMP: Europe-approved reference medicinal product; US-RMP: US-licensed reference medicinal product.

Clinical data

Clinical data that contributed to approval of CT-P13 were generated in PLANETAS and PLANETRA, the two randomized clinical trials which directly evaluated CT-P13 and the RMP infliximab in AS and RA [46,47]. The highly similar efficacy and safety profiles between CT-P13 and RMP in these studies are illustrated in Tables 2 & 3, respectively. Data up to 54 weeks have also been presented at scientific meetings [48–50]. Results of subsequent open-label, 48-week extension phases, in which patients receiving RMP were switched to CT-P13, have also been presented at scientific congresses [51,52].

Table 2. Highly comparable efficacy between CT-P13 and reference medicinal product infliximab in patients with active ankylosing spondylitis (PLANETAS study) or rheumatoid arthritis (PLANETRA study) [46,49].

Efficacy in patients with AS (PLANETAS study)
Endpoint	CT-P13 (n = 125)	RMP (n = 125)	p-value
ASAS20 response (percentage of patients)
Week 30	70.5	72.4	Not significant
Week 54	67.0	69.4	Not significant
ASAS40 response (percentage of patients)
Week 30	51.8	47.4	Not significant
Week 54	54.7	49.1	Not significant
Efficacy in patients with RA (PLANETRA study)
Endpoint	CT-P13 (n = 248)	RMP (n = 251)	p-value
ACR20 response (percentage of patients)
Week 30	60.9	58.6	Not significant
Week 54	57.0	52.0	Not significant
ACR50 response (percentage of patients)
Week 30	35.1	34.2	Not significant
Week 54	33.1	31.6	Not significant
ACR70 response (percentage of patients)
Week 30	16.6	15.5	Not significant
Week 54	16.2	15.1	Not significant

ACR20: American College of Rheumatology 20% response; ACR50: American College of Rheumatology 50% response; ACR70: American College of Rheumatology 70% response; AS: Ankylosing spondylitits; ASAS20: Ankylosing Spondylitis Response Criteria 20% response; ASAS40: Ankylosing Spondylitis Response Criteria 40% response; RA: Rheumatoid arthritis; RMP: Reference medicinal product.

Table 3. Treatment-related serious adverse events up to week 30 in patients with active ankylosing spondylitis (PLANETAS study) or rheumatoid arthritis (PLANETRA study) treated with CT-P13 or reference medicinal product [46,47].

Patients with AS (PLANETAS study)
Event, n (%)	CT-P13 (n = 128)	RMP (n = 122)	Total (N = 250)
Esophageal perforation	1 (0.8)	0	1 (0.4)
Infusion-related reaction	0	2 (1.6)	2 (0.8)
Cellulitis	0	1 (0.8)	1 (0.4)
Disseminated tuberculosis	1 (0.8)	0	1 (0.4)
Pulmonary tuberculosis	1 (0.8)	1 (0.8)	2 (0.8)
Wound infection	0	1 (0.8)	1 (0.4)
Dyspnea	1 (0.8)	0	1 (0.4)
Patients with RA (PLANETRA study)
Event, n (%)	CT-P13 (n = 301)	RMP (n = 301)	Total (N = 602)
Neutropenia	1 (0.3)	0	1 (0.2)
Vestibular disorder	1 (0.3)	0	1 (0.2)
Infusion-related reaction	4 (1.3)	4 (1.3)	8 (1.3)
Hepatitis toxic	1 (0.3)	0	1 (0.2)
Appendicitis	0	1 (0.3)	1 (0.2)
Infective arthritis	0	1 (0.3)	1 (0.2)
Disseminated tuberculosis	2 (0.7)	0	2 (0.3)
Pulmonary tuberculosis	1 (0.3)	0	1 (0.2)
Herpes zoster	0	1 (0.3)	1 (0.2)
Lobar pneumonia	1 (0.3)	0	1 (0.2)
Pneumonia	1 (0.3)	0	1 (0.2)
Staphylococcal wound infection	1 (0.3)	0	1 (0.2)
Musculoskeletal chest pain	1 (0.3)	0	1 (0.2)
Flare in RA activity	1 (0.3)	0	1 (0.2)
Breast cancer	0	1 (0.3)	1 (0.2)
Metastatic ovarian cancer	0	1 (0.3)	1 (0.2)
Renal neoplasm	1 (0.3)	0	1 (0.2)
Cerebrovascular disorder	1 (0.3)	0	1 (0.2)
Endometrial hyperplasia	1 (0.3)	0	1 (0.2)
Metrorrhagia	0	1 (0.3)	1 (0.2)
Thrombophlebitis	1 (0.3)	0	1 (0.2)

AS: Ankylosing spondylitis; RA: Rheumatoid arthritis; RMP: Reference medicinal product.

The AS population was selected for PK and immunogenicity comparisons in PLANETAS because of its representativeness in terms of sensitivity, the use of higher treatment doses (5 mg/kg) and the fact that concomitant methotrexate is not used in these patients [37,46]. The study confirmed PK equivalence between CT-P13 and RMP, as measured by steady-state AUC and observed maximum steady-state serum concentration (C_max,ss) between weeks 22 and 30 (dose 5 and 6) [46].

Highly similar efficacy was also observed between CT-P13 and RMP based on various disease indexes, including Ankylosing Spondylitis Response Criteria (ASAS) 20% response, ASAS 40% response and AS disease activity score (ASDAS)-C reactive protein [46]. Immunogenicity was also comparable between CT-P13 and RMP. In the CT-P13 group, anti-drug antibodies (ADAs) were detected in 9.1, 27.4 and 22.9% of patients at weeks 14, 30 and 54, respectively. In the RMP group, ADAs were detected in 11.0, 22.5 and 26.7% of the patients at the same respective time points [46,48].

The pivotal efficacy and safety study (PLANETRA) [47] was conducted in patients with RA because the EMA considered RA to be a sufficiently sensitive model to detect differences between CT-P13 and its RMP, with the primary endpoint (American College of Rheumatology 20% (ACR20) response at week 30) and the equivalence margin (±15%) in line with their scientific guidance [37]. The ACR20 responses at 30 weeks were 60.9 and 58.6% in the CT-P13 and RMP groups, respectively. In terms of PK, the geometric means of C_max were similar with CT-P13 and RMP [47]. Regarding immunogenicity, ADAs were detected in comparable proportions of RA patients receiving CT-P13 (25.4, 48.4 and 52.3%) and RMP (25.8, 48.2 and 49.5%) at weeks 14, 30 and 54, respectively [47,49]. With regards to cross-reactivity, a study in IBD patients with and without measurable anti-RMP antibodies showed that these ADAs recognized CT-P13 with similar reactivity [53].

At the time of approval, pharmacovigilance plans for CT-P13 up to 2026 were submitted to the EMA and accepted by the Pharmacovigilance Risk Assessment Committee [37]. This included 12 post-marketing clinical studies; five of which have been completed and seven are ongoing. Of these ongoing studies, one interventional and two non-interventional studies involve patients with IBD. As required, the marketing authorization holder is also performing ongoing pharmacovigilance activities, including submission of periodic safety updates which have uncovered no safety concerns to date.

Conclusions

Biologic therapies are important for the effective treatment of a wide range of immune-inflammatory diseases, including IBD. However, they are also expensive and their cost can lead to restricted access for many patients [54,55]. Recent expiration of the patents for frequently used biologics, together with the economic burden of these agents, has led to increasing interest from both payers and the biopharmaceutical industry in the development of biosimilars. However, because any biologic drug may show inherent variations when manufacturing processes are modified, regulatory authorities have developed guidelines specifically for the approval of biosimilars. Once a biosimilar has demonstrated comparable PK, efficacy and safety to the RMP in a key indication, then extrapolation of efficacy and safety data to other indications may be possible. However, in order to agree to this, the regulators require a large amount of data from functional and clinical studies.

CT-P13, the first mAb biosimilar approved by the EMA, was developed through a rigorous comparability program, which included comprehensive physicochemical and biologic characterization, non-clinical studies and clinical trials. The comparability program demonstrated that CT-P13 and a version of the RMP have highly similar efficacy, PK, immunogenicity and safety. Physicochemical properties and activity related to the mechanisms of action important for immune-inflammatory diseases were also comparable. These data provided the basis for extrapolation of clinical data with CT-P13 in RA and AS to IBD. Although there have been concerns relating to the extrapolation of indications for this infliximab biosimilar, the totality of the evidence suggests that this regulatory decision was justified.

In addition to CT-P13, the first infliximab biosimilar [37], there are now a number of other anti-TNF biosimilars in the development pipeline (Table 4). In order to gain approval and acceptance, these biosimilars will need to follow an equally robust development program as CT-P13. Because the manufacturing processes for all biologic drugs, including the originators, are subject to some changes over time, continuous quality control becomes an important issue in the production process. This is why the EMA no longer speaks of biosimilars, but considers protein products that have a high degree of similarity to be ‘versions’ of the same molecule. Consequently, it is necessary to introduce product-specific, long-term national and international registries to track the efficacy and side effects in regular clinical use and to use data from these registries to supplement the insights generated from randomized clinical trials.

Table 4. Current status of biosimilar development for anti-tumor necrosis factor biologics (from publicly available data only).

Biosimilar	Brand/molecule	Company	Status
Infliximab	Remsima^®/CT-P13	Celltrion	Approved
	Inflectra^®/CT-P13	Hospira	Approved
	BOW-015	Epirus Biopharmaceuticals/Ranbaxy Laboratories Limited	Phase III
	SB2	Samsung Bioepis	Phase III
	NI-071	Nichi-Iko Pharmaceutical	Phase III
	PF-06438179	Pfizer	Phase III
Etanercept	SB4	Samsung Bioepis	Phase III
	GP2015	Sandoz	Phase III
	CHS-0214	Coherus	Phase III
Adalimumab	ABP 501	Amgen	Phase III
	GP2017	Sandoz	Phase III
	BI 695501	Boehringer Ingelheim	Phase III
	SB5	Samsung Bioepis	Phase I
	FKB327	Fujifilm Kyowa Kirin Biologics/Kyowa Hakko Kirin	Phase I
	PF-06410293	Pfizer	Phase I

Expert commentary

The approval of biosimilars may improve patient access to effective treatments for many immune-mediated inflammatory diseases, including IBD. Variability of biologic expression systems is an important consideration with all biologic drugs, which leads to a degree of heterogeneity, even between batches of a single product. Consequently, a biosimilar will be another version of the RMP. As a result, prior to approving a biosimilar, regulatory authorities take rigorous steps to control the quality of these drugs and require robust programs to be conducted to validate their comparability to the RMP. However, it is important for prescribing physicians to fully understand the scientific principles used in the approval of biosimilars and, in particular, the rationale for extrapolation of data from one indication (such as rheumatic diseases) to additional indications (such as IBD). This article provides an overview of the important lessons from the introduction of biosimilars in a range of different therapy areas and summarizes the key data available for CT-P13, the first infliximab (and first mAb) biosimilar. Ongoing studies are generating additional data on the product, and no concerns about differences in efficacy or safety between CT-P13 and the infliximab RMP have been identified in clinical practice.

Five-year view

The biosimilar infliximab, CT-P13, is already available in many countries for IBD, with other launches being planned. The lower cost of CT-P13 and other biosimilars is anticipated to increase access for many more patients to effective biologic therapies. Given this growth in the use of biosimilars, it is important to fully understand the efficacy and safety in clinical use. Clinical trials, registries and pharmacovigilance studies with CT-P13 are ongoing around the globe and will provide more evidence to inform clinical decision making and close the gaps in knowledge of how to make the best use of infliximab to treat immune-mediated inflammatory disorders. In the meantime, prescribing physicians need to remain informed of the data on important questions such as extrapolation, efficacy and safety in order to elucidate the optimum role of CT-P13 and other biosimilars in effective patient management.

Key issues

• Potential sources of variation in the manufacture of biologics are numerous, leading to a degree of heterogeneity even between batches of a specific product.

• With any biologic, the manufacturing processes will generally have been modified several times over the years, but sophisticated quality assurance systems are designed to avoid changes that might affect its efficacy and safety.

• Recognizing this complexity in manufacturing biologics, there are well-established, specific regulatory procedures for approval of biosimilars, requiring that comparability with the reference medicinal product (RMP) is assessed using a combination of analytic testing, biologic assays, and appropriate non-clinical and clinical data.

• These principles have led to the introduction of biosimilars for treatment of a range of chronic and life-threatening diseases, and post-marketing studies have established their efficacy and safety in clinical use.

• During regulatory review of biosimilars, extrapolation of efficacy and safety data from one indication to another may be considered, if biosimilarity to the RMP has been shown by the comprehensive comparability program.

• CT-P13, a biosimilar infliximab, was approved in September 2013 in Europe for the same indications as the RMP (Remicade^®), including rheumatoid arthritis, ankylosing spondylitis, Crohn’s disease, ulcerative colitis, psoriatic arthritis and psoriasis, based on a large comparability program.

• The comparability program demonstrated that CT-P13 and RMP are highly similar in terms of the therapeutically relevant biologic and physiochemical properties, as well as efficacy, pharmacokinetic, immunogenicity and safety in rheumatoid arthritis and ankylosing spondylitis. These data also led to the extrapolation of the approved indications for CT-P13 to inflammatory bowel disease by many regulatory authorities.

• Although there are important questions relating to the extrapolation of indications for the infliximab biosimilar, the totality of the evidence suggests that this regulatory decision was justified.

Financial & competing interests disclosure

S Schreiber has served as a consultant or advisory member for Abbvie, Boehringer Ingelheim, Celltrion, Ferring, Genentech, Hospira, Janssen, MSD, Mundipharma, Pfizer, Roche, Takeda and UCB; and has served as a speaker for Abbvie, Celltrion, Falk Pharma, Ferring, Janssen, Hospira, MSD, Mundipharma, Shire Pharmaceuticals and UCB. J Panés has served as a consultant or advisory member for Abbvie, Arena Pharmaceuticals, Boehringer Ingelheim, BMS, Celltrion, Ferring, Galapagos, Genentech, Hospira, Janssen, MSD, Nutrition Science Partners, Pfizer, Robarts, Roche, Takeda, Tigenix, Topivert and TxCell. J Panés has also served as a speaker for Abbvie, Ferring, Janssen, MSD, Shire Pharmaceuticals and Tillots; and has received research funding from Abbvie and MSD. L Peyrin-Biroulet has received consulting fees from Merck, Abbott, Janssen, Genentech, Mitsubishi, Ferring, Norgine, Tillots, Vifor, Shire, Therakos, Pharmacosmos, Pilège, BMS, UCB-pharma, Hospira, Celltrion, Takeda, Biogaran, Boerhinger-Ingelheim, Lilly, Pfizer and HAC-Pharma. He has also received lecture fees from Merck, Abbott, Takeda, Janssen, Ferring, Norgine, Tillots, Vifor, Therakos and HAC-Pharma. B Kwon and S Hong are full-time employees of Celltrion, the manufacturer of CT-P13. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. Editorial support (writing assistance, assembling tables and figures, collating author comments, grammatical editing and referencing) was provided by Mark O’Connor (Aspire Scientific Limited, Bollington, UK) and was funded by Celltrion Healthcare Co., Ltd (Incheon, Republic of Korea).

Notes