Switching to biosimilars: current perspectives in immune-mediated inflammatory diseases

ABSTRACT

Introduction: The expiry of patents for biologics has led to the introduction of biosimilars for the treatment of immune-mediated inflammatory diseases (IMIDs). These treatment alternatives may allow earlier and wider access to appropriate therapy for patients without increasing the economic burden on health-care systems. Prescription of biosimilars to treatment-naïve patients is well accepted; however, additional considerations must be taken into account when switching clinically stable patients from reference products to biosimilars.

Area covered: We discuss the current considerations related to switching from reference products to biosimilars from a physician and patient perspective. We review the clinical data and real-life experience on switching patients with IMIDs, present the position of the relevant medical societies, and discuss the importance of patient–physician communication and need for shared decision-making.

Expert opinion: The introduction of biosimilars provides an opportunity to expand access to treatment for patients with IMIDs across Europe and support the financial sustainability of health-care systems. We anticipate that as the real-world evidence base grows, confirming the results of clinical trials, there will be a corresponding increase in physician and patient acceptance, not only to initiating treatment with a biosimilar, but also to switching medication from a reference product to a biosimilar.

KEYWORDS:

• Biological therapy
• biosimilar
• immune-mediated inflammatory disease
• switching

1. Introduction

In the past two decades, the treatment of immune-mediated inflammatory diseases (IMIDs) has been transformed by the introduction of biological therapies [1]. Current guidelines often reserve these drugs for patients with more severe disease and/or patients who have failed to respond to multiple other treatments [2,3]. In some countries, access to biologics may then be further limited by the implementation of national eligibility and reimbursement criteria that are more stringent than those recommended in international guidelines [4], leading to inequities in the provision of biologics in many countries worldwide [3,5–7]. Despite these restrictions, there is growing evidence supporting the benefits of biologics for patients with early or moderate disease [8–15].

The high cost of reference biologics is, at least in part, linked to the high costs of research, development, and complex manufacturing processes [1]. Furthermore, the economic impact of biologics may be exacerbated by the use of dose optimization in clinical practice to maintain treatment benefits [16,17]. However, as many biologics have reached or are near patent expiry, the introduction of biosimilars may reduce the impact of biologics on state health-care budgets, as has been predicted in many European countries [18–20]. Biosimilars are alternatives to already licensed therapies (referred to as reference products), which are demonstrated to have analytical similarity, and comparable clinical efficacy, immunogenicity, and safety profiles. A summary of currently approved biosimilars with an IMID indication is provided in Table 1. The availability of biosimilars creates price competition, with the reference products also lowering their prices in order to remain competitive in the tendering process. The benefits of biosimilars have already been observed in the United Kingdom, where the introduction of infliximab and etanercept biosimilars was estimated to have saved the National Health Service (NHS) around £38.8 million from March 2014 to February 2017 [21]. Furthermore, it is estimated that NHS England could save £300 million by 2021 following the recent launch of adalimumab biosimilars [22]. Along with reduced health-care costs, this offers the potential for improved access to biologics where needed as well as other benefits such as enhanced access to allied health-care services.

Table 1. Currently approved biosimilars with an IMID indication.

Product name(s)	Manufacturer(s)	Initial marketing authorization date
Infliximab biosimilars
BOW015: Infimab	Reliance Life Sciences	India: 09/2014
CT-P13: Inflectra, Remsima	Celltrion	EMA: 09/2013 (Remsima/Inflectra) FDA: 04/2016 (Inflectra) Korea: 07/2012 (Remsima/Inflectra) Australia: 08/2015 (Remsima/Inflectra) Canada: 01/2014 (Remsima/Inflectra)
SB2: Flixabi, Renflexis	Samsung Bioepis	EMA: 05/2016 (Flixabi) FDA: 05/2017 (Renflexis) Korea: 12/2015 (Renflexis) Australia: 11/2016 (Renflexis) Canada: 02/2018 (Renflexis)
Etanercept biosimilars
SB4: Benepali, Brenzys	Samsung Bioepis	EMA: 01/2016 (Benepali) Korea: 09/2015 (Brenzys) Australia: 07/2016 (Brenzys) Canada: 08/2016 (Benepali/Brenzys)
GP2015: Erelzi	Sandoz	EMA: 06/2017 FDA: 08/2016 Australia: 11/2017 Canada: 08/2017
Adalimumab biosimilars
ABP501: Amgevita, Amjevita	Amgen	EMA: 03/2017 (Amgevita) FDA: 09/2016 (Amjevita) Australia: 09/2017 (Amgevita)
BI 695501: Cyltezo	Boehringer Ingelheim	EMA: 11/2017 (subsequently withdrawn by manufacturer) FDA: 08/2017
FKB327: Hulio	Mylan S.A.S.	EMA: 09/2018
GP2017: Hyrimoz	Sandoz	EMA: 07/2018 FDA: 10/2018 Australia: 03/2019
MSB11022: Idacio	Fresenius Kabi	EMA: 01/2019
SB5: Imraldi, Hadlima	Samsung Bioepis	EMA: 08/2017 (Imraldi) Korea: 09/207 (Hadlima) Australia: 01/2018 (Hadlima) Canada: 05/2018 (Hadlima)
Rituximab biosimilars
CT-P10: Truxima	Celltrion	EMA: 02/2017 FDA: 11/2018 Korea: 11/2016 Australia: 04/2018
GP2013: Riximyo, Rixathon	Sandoz	EMA: 06/2018 (Riximyo, Rixathon) Australia: 11/2017 (Riximyo)

Marketing authorization information taken from publicly available sources (correct as of March 2019)

Regulatory approval of biosimilars by the European Medicines Agency (EMA) and US Food and Drug Administration (FDA) is a rigorous process requiring an extensive comparability exercise to demonstrate similarity between the biosimilar and the reference product [23–25]. Once a biosimilar is approved, it can be prescribed to patients who are naïve to the reference product. In addition, for patients already treated with a biologic, physicians can switch treatment with the reference product to treatment with a biosimilar. Before approval, the FDA expects additional safety, efficacy and immunogenicity data for patients who undergo a switch from a reference product to a biosimilar [26]. In line with this, the vast majority of pivotal biosimilar studies conducted to date have included a switch component in the study design, in order to demonstrate that efficacy, safety, and immunogenicity are maintained after switching [27].

Despite the potential and realized benefits, physicians and patients may have questions about the benefits versus the risks of switching. In this review, we discuss the potential considerations associated with switching from a reference product to a biosimilar and review the strong evidence supporting switching as well as the methods by which patient and physician concerns may be addressed to ensure both an efficacious and safe treatment for all patients.

2. Literature search

Randomized controlled trials and real-world studies that include a switching design were identified from three previously published systematic reviews [28–30]. Additional studies were identified through a narrative literature search conducted in March 2019. The PubMed and European Union Clinical Trials Register databases were searched using the following search terms: infliximab, etanercept, adalimumab, rituximab, CT-P13, SB2, BOW015, GP2015, SB4, ABP 501, BI 695501, CHS-140, FKB327, GP2017, MSB11022, SB5, CT-P10, GP2013, biosimilar, and switch. Only articles in English were included. We also searched abstracts presented at the EULAR, ECCO and ACR congresses using the search terms biosimilar and switch.

3. Considerations regarding switching

3.1. Immunogenicity

The possibility of increased immunogenicity following switching from a reference product to a biosimilar is a consideration for many physicians and patients [31]. Although a biosimilar and its reference product have an identical amino acid sequence, subtle differences in post-translational modifications that occur during the manufacturing process in living organisms can be expected and may be sufficient to trigger an immune response [32]. During the development of a biosimilar, it is essential to assess the potential for immunogenicity through an extensive analytical characterization exercise [25,33]. Furthermore, a direct evaluation of immunogenicity through the measurement of anti-drug antibodies and neutralizing antibodies in a clinical study should be performed [33]. This assessment of immunogenicity via analytical characterization and clinical evaluation remains an important part of the development program for biosimilars. As discussed later in this article, it is reassuring that the majority of clinical trials published to date have not reported a change in immunogenicity following a switch from a reference product to a biosimilar [29].

3.2. Multiple switching

Another consideration relates to the possibility of multiple switching between the reference product and the biosimilar or between different biosimilars. While elective multiple switching between biosimilars is not expected, patients may be switched back to the reference product due to adverse events or because of patient preference or perceptions of treatment [34–37]. Additionally, the authorization of increasing numbers of biosimilars could produce treatment scenarios with multiple switches between biosimilars [28]. This may have implications for pharmacovigilance with regard to traceability and tracking safety signals for specific biosimilars. To ensure patient safety, the EMA requires the trade name for a biologic to be used for all prescriptions, patient administration, and for reporting suspected adverse drug reactions [38].

3.3. Physician perception of biosimilars

Inadequate knowledge and education on biosimilar development and manufacturing may be an important factor contributing to hesitancy in prescribing and/or switching patients to biosimilars. A study in the United States (US) showed that some physicians considered the abbreviated approval process for biosimilars to pose a safety risk, despite the regulatory requirements for an extensive data package showing similarity in quality, efficacy, and safety [39]. Physicians in Europe also reported concerns around safety and extrapolation of data to different indications, which may again reflect a gap in understanding of the approval process [40]. In terms of resource utilization, it is of concern to some health-care providers (HCPs) that switching patients to a biosimilar might require longer and/or more consultations, which would off-set the cost savings associated with the switch. However, a registry study in Denmark found no evidence of an increase in the rate of outpatient visits following a switch from reference infliximab to the biosimilar CT-P13 in a population of patients with inflammatory arthritis [41]. In addition, gain share agreements have been established with payers in some European countries to ensure that part of the cost savings achieved through the use of biosimilars are used to support any additional resources required to facilitate the switch [42]. Finally, physicians may have concerns regarding switching patients on stable, long-term therapy from a reference biologic to a biosimilar if the patients themselves are reluctant to switch. Increasing levels of data and experience are likely to raise the confidence in biosimilars both for HCPs and patients.

3.4. Patient perception of biosimilars

Patient perception with regard to biosimilars and their potential preference for the reference product are important factors to consider with regard to switching medications. Patients receiving treatment with a reference product may naturally question whether the biosimilar is as effective and safe as their current therapy. If their concerns are not addressed, patients may be reluctant to switch, or the success of the switch may potentially be compromised by the occurrence of a phenomenon known as the ‘nocebo’ effect. Similar to the placebo effect but in the opposite direction, the nocebo effect is a change that cannot be explained on the basis of the known pharmacology of the drug [43,44]. In the case of switching from a reference product to a biosimilar, the nocebo effect can result in the patient experiencing a perceived loss of efficacy and/or occurrence of adverse events following the switch. This effect can lead to non-adherence or to switching back to the reference product according to patient preference [35,45]. The impact of the nocebo effect has been described in observational studies, where patient withdrawals from biosimilar treatment have been reported despite a lack of change in disease activity [34,36,37,46–48]. Further support for the existence of a nocebo effect was provided in a recent systematic review conducted by Odinet et al., which demonstrated that discontinuation rates (both overall and due to adverse drug events) were higher in open-label switch studies (i.e. where patients were aware they were switched to a biosimilar) compared with double-blind studies [49].

Another potential concern for patients is that they may be switched from their existing treatment to a biosimilar without their knowledge or appropriate information and awareness. However, medical societies in Europe remain strongly opposed to automatic substitution of biologics and have emphasized the need for open communication between patients and physicians and a joint decision-making approach [27,50,51].

3.5. Delivery device

For biologics that are self-administered via subcutaneous injection, any changes to the delivery device that affects comfort or ease of administration may impact a patient’s decision to switch. However, it should be noted that patient surveys have shown that in rheumatoid arthritis, where dexterity is a particular issue for patients who are self-administering medication, biosimilar manufacturers have introduced innovations in delivery that have been positively received by patients and HCPs [52–54].

4. Growing evidence to support switching

4.1. Development and approval process for biosimilars

As biosimilar manufacturers can build on the safety and efficacy experience gained with the reference product, an abbreviated regulatory pathway for approval of biosimilars has been established in many countries, including in the European Union and US, which does not require extensive clinical trials [23–25]. Despite this abbreviated pathway, the regulatory requirements for biosimilars are stringent and are based on a demonstration of the biochemical and clinical equivalence between the biosimilar and reference product, with a key focus on analytical similarity [25,38,55].

Similarity between a reference product and its biosimilar is established in a three-tier process. The first tier covers comparative CMC (chemistry, manufacturing, and controls) and preclinical studies. These include extensive in vitro analytical physicochemical and functional studies using state-of-the-art and fit-for-purpose analytical methods designed to comprehensively compare the physicochemical structure and biological function of the two drugs. These studies use sensitive techniques that are capable of detecting minor differences in key product attributes between the two drugs that can be of clinical relevance. In the second tier, to further support the assessment of similarity and extrapolation of indications, pharmacodynamic (PD) studies compare the binding and activation of downstream physiological targets in biological cell systems that are relevant to the mode of action and indications approved for the molecule. In the third tier, comparative clinical studies are performed. Establishing pharmacokinetic (PK) and PD similarity is a key part in the development process and similarity assessment exercise of biosimilars. PK studies are usually performed in healthy volunteers and assess exposure to the reference product and the biosimilar through various parameters including peak concentration and area under the concentration–time curve. The safety profile, including immunogenicity, of a biosimilar will also be evaluated in these studies. Finally, clinical trials are performed in sensitive patient populations with appropriate endpoints to demonstrate that there are no clinically meaningful differences in terms of efficacy, safety, and immunogenicity after repeated exposure between the biosimilar and the reference product [23–25,38]. Specifically, it should be noted that after assessment of the primary endpoint, switching a clinical meaningful number of subjects from the reference product to the biosimilar and follow-up until end of the study, usually up to 52 weeks, is frequently performed [27]. It should also be noted that the scientific principles underlying the comparability exercise for biosimilars are the same as those for manufacturing changes for a reference biological product. For various reasons, there have been many changes over time in the production of reference products; these changes had to be demonstrated to result in comparable quality with regard to safety and efficacy [56,57].

4.2. Switching studies – randomized controlled trials

In order to evaluate the impact of switching from a reference product to a biosimilar, a switch element is frequently included in the pivotal randomized controlled trials (RCTs). These RCTs are performed as part of the development and approval process for a biosimilar, and the evidence obtained has been evaluated in a recent systematic review by Cohen et al. [29]. In this review, the authors concluded that the data were reassuring and that the potential risk of immunogenicity-related safety concerns or diminished efficacy is unchanged after switching from a reference product to a biosimilar [29]. A similar literature review was conducted by Inotai et al. in 2017, which evaluated 58 articles, including clinical studies, systematic reviews, and non-empirical studies [58]. The authors of this review concurred with Cohen et al. that, although clinical trials for biosimilars are rarely adequately powered to detect the consequences of switching, empirical studies and systematic reviews performed to date do not show any concerning clinical outcomes or additional risk for patients switching to biosimilars. A more recent systematic review, presented at the International Society for Pharmacoeconomics and Outcomes Research meeting in 2018, identified 178 switch studies, including 25 RCTs. The authors of this review concluded that there was no indication that switching from a reference product to a biosimilar was associated with major safety issues or a loss of response. However, it was noted that most of the trials were not adequately powered to detect rare adverse events or changes in efficacy due to a small sample size and short follow-up period [59].

In the setting of IMIDs, we have identified 25 RCTs that have been completed to date that include a switch arm (either as part of a double-blind trial or during an open-label extension [OLE]). These studies are summarized in Table 2 and include biosimilars for etanercept, adalimumab, infliximab, and rituximab. A number of studies have examined switching from reference infliximab to CT-P13, including the pivotal NOR-SWITCH study, which demonstrated across a number of indications that switching to CT-P13 was not inferior to continued treatment with the infliximab reference product [60]; although it should be noted that NOR-SWITCH was not powered to show non-inferiority in individual diseases. There is also a growing body of RCT evidence for adalimumab biosimilars (ABP 501, BI 695501, CHS-1420, FKB 327, GP2017, MSB11022, and SB5) that have, to date, found no impact of switching on efficacy, safety, or immunogenicity [61–69]. Switch studies also provide supporting evidence for switching to the infliximab biosimilars SB2 [70], NI-071 [71], and BOW015 [72], along with the etanercept biosimilars GP2015 [73,74], SB4 [75], and LBEC0101 [76], and the rituximab biosimilars CT-P10 [77] and GP2013 [78]. In agreement with previously published systematic reviews, a limitation common to many of the studies included in Table 2 is the lack of adequate statistical power to detect differences between the switched vs continuously treated groups. In addition, many studies lack a comparator arm that includes patients receiving continuous reference product treatment.

Table 2. Summary of randomized controlled trial switching studies (including open-label extensions).

Study	Indications	N*	Study/switch design^†	Duration after switch (weeks)	Efficacy outcomes	Safety/immunogenicity/PK/other outcomes	Limitations
CT-P13 (infliximab)
PLANETRA-OLE Yoo 2017 [79]	RA	302	OLE study Single switch after a 62w DB study CT-P13 maintenance pts vs RP/CT-P13 switch pts	40	ACR20/50/70 similar between groups	Proportion of pts with ADAs and AEs similar for maintenance vs switch groups 16% of maintenance and 11% of switch groups discontinued treatment	OL study No continuous RP control arm Not powered to evaluate switching
Tanaka 2017 [80]	RA	71	OLE study Single switch after a 54w DB study CT-P13 maintenance pts vs RP/CT-P13 switch pts	105	Mean DAS28-ESR similar between groups	Proportion of pts with AEs similar for maintenance vs switch groups 16% of maintenance and 33% of switch groups discontinued treatment	OL study No continuous RP control arm Not powered to evaluate switching Small number of pts in each group
PLANETAS-OLE Park 2017 [81]	AS	174	OLE study Single switch after a 54w DB trial: CT-P13 maintenance pts vs RP/CT-P13 switch pts	40	ASAS20/40 and ASAS partial remission were similar between groups	Proportion of pts with ADAs and AEs similar between groups 8% of maintenance and 10% of switch groups discontinued treatment	OL study No continuous RP control arm Not powered to evaluate switching
NOR-SWITCH Jorgensen 2017 [60]	IBD, SpA, RA, PsA, PsO	481	DB RCT Single switch after ≥6 months of treatment with RP RP maintenance vs RP/CT-P13 switch	52	RP/CT-P13 switch was not inferior to RP maintenance (measured by disease worsening)	Frequency of AEs was similar between groups 10% of maintenance and 8% of switch groups discontinued treatment	Not powered to show non-inferiority in individual diseases
NOR-SWITCH OLE Goll 2019 [82]	IBD, SpA, RA, PsA, PsO	380	OLE study Single switch after a 52w DB trial CT-P13 maintenance pts vs RP/CT-P13 switch pts	78	Disease worsening occurred in 16.8% maintenance pts vs 11.6% switch pts	Frequency of AEs and ADAs was similar between groups	OL study No continuous RP control arm Not powered to show non-inferiority in individual diseases
NCT02096861 Kim 2017 [83] Ye 2018 [84]	IBD	220	DB RCT Single switch at w30 RP maintenance vs CT-P13 maintenance vs RP/CT-P13 switch vs CT-P13/RP switch	24	Clinical worsening and CDAI-70 responses were similar between the 4 groups	Frequency of AEs, SAEs and infections was similar between groups No clinically meaningful differences in immunogenicity between groups	No statistical analyses present Short follow up after switch
SIMILAR Volkers 2017 [85]	IBD	47	DB RCT Single switch in patients receiving RP prior to randomization RP maintenance vs RP/CT-P13 switch	30	To date 1 pt experienced a relapse (switch group)	2 pts experienced an SAE	Preliminary results only Only 21 pts have completed the 30w follow up to date
NI-071 (infliximab)
NCT01927263 Matsuno 2018 [71]	RA	242	OLE study Single switch after a DB 30w trial Continued NI-071 vs RP/NI-071 switch	24	No significant differences in DAS28-ESR, DAS28-CRP, or ACR20/50/70 responses between treatment groups	Incidence of related TEAEs was comparable between groups	OL study No continuous RP control arm Not powered to show differences after switching
SB2 (infliximab)
NCT01936181 Smolen 2018 [70]	RA	584	DB RCT Single switch at w54 RP maintenance vs SB2 maintenance vs RP/SB2 switch	24	ACR20 was sustained and comparable across treatment groups	TEAEs occurred in 35.6%, 40.3%, and 36.2% of RP, SB2 and RP/SB2 groups ADAs occurred in 14.6%, 14.9%, and 14.1% of RP, SB2 and RP/SB2 groups 5.0%, 7.5%, and 6.4% of RP, SB2, and RP/SB2 pts discontinued before w78	Not powered to show differences in efficacy after switching Short follow up after switching
BOW015 (infliximab)
UNIFORM OLE Kay 2015 [72]	RA	157	OLE study Single switch after a 16w DB study BOW015 maintenance pts vs RP/BOW015 switch pts	38	Disease activity was similar between the 2 groups throughout the OLE	NR	OL study Not powered to show differences after switching No continuous RP control arm
GP2015 (etanercept)
EGALITY Gerdes et al 2018 [73]	PsO	531	DB RCT Multiple switches: 3 alternating 6-week periods between w12 to 30 Pooled RP and GP2015 maintenance groups vs pooled switch groups	40	PASI 50/75/90 response rates and all other efficacy parameters were similar between the pooled switched and pooled continued treatment groups	Incidence of TEAEs was similar between the switched and continued groups Discontinuations due to TEAEs occurred in 3.1% of switched pts and 1.0% of continued pts	Not powered to show differences after switching Groups were pooled rather than analyzed separately
EQUIRA Matucci-Cerinic et al 2018 [74]	RA	376	DB RCT Single switch after a 24w DB study GP2015 maintenance pts vs RP/GP2015 switch pts	24	Mean change in DAS28-CRP and EULAR and ARC 20/50/70 response rates were comparable between the two groups	Frequency of AEs and SAEs was similar between the 2 groups Non-neutralizing, low titer ADAs were detected in 4 GP2015 maintenance pts	No statistical analyses described or presented Short follow up after switching No continuous RP control arm
SB4 (etanercept)
NCT01895309 Emery et al 2017 [75]	RA	245	OLE study Single switch after 52w DB study SB4 maintenance pts vs RP/SB4 switch pts	48	ACR20 response rates were comparable between the two groups	Rates of TEAEs were comparable between the groups 1 pt in each group developed ADAs Approx. 5% of both groups discontinued before the end of the trial	OL study Not powered to show differences after switching No continuous RP control arm
LBEC0101 (etanercept)
Song et al. 2018 [76]	RA	148	OLE study Single switch after a 52w DB study Continued LBEC0101 pts vs RP/LBEC0101 switch pts	48	DAS28-ESR scores were maintained in both groups to end of study	Incidence of AEs and new ADAs was comparable between groups	Not powered to show difference after switching No continuous RP control arm
ABP501 (adalimumab)
NCT01970488 Papp et al 2017 [61]	PsO	326	DB RCT Single switch at w16 Continued ABP501 pts vs continued RP pts vs RP/ABP501 switch pts	36	PASI % improvements were similar across all 3 groups No significant differences in % of PASI 50, 75, 90, 100 responders	AEs were balanced between groups % of pts with ADAs and NABs were similar between groups	Not powered to show differences after switching
NCT02114931 Cohen et al. 2016 [62]	RA	467	OLE study Single switch after a 26w DB study Continued ABP501 pts vs RP/ABP501 switch pts	72	Efficacy results were similar between ABP501 maintenance and RP/ABP501 switch groups	Rates of TEAEs and ADAs were similar between groups	OL study Not powered to show differences after switching No continuous RP control arm Only interim analysis results available (at this time only 1 pt had completed the study)
BI 695501 (adalimumab)
VOLTAIRE-RA Cohen et al 2018 [63]	RA	593	DB RCT Single switch at w24 Continued BI 695501 pts vs continued RP pts vs RP/BI 695501 switch pts	34	Mean change in DAS28-ESR and ACR20/50/70 response rates were similar across all 3 groups	Immunogenicity and safety and tolerability profiles were similar between all three groups	Not powered to show differences after switching
CHS-1420 (adalimumab)
NCT02489227 Hodge et al 2017 [64]	PsO, PsA	545	DB RCT Single switch at w16 Continued CHS-1420 pts vs RP/CHS-1420 switch pts	8	No statistically significant differences in PASI75 response rates, HAQ-DI, or hsCRP between groups at w24	Similar incidences of TEAEs and ADAs between groups	Not powered to show differences after switching Short follow up after switching No continuous RP control arm
FKB327 (adalimumab)
ARABESC-OLE Genovese et al 2017 [65]	RA	645	OLE study Multiple switches: 1/3 switched to alternate treatment at week 24 of parent trial, all switched to FKB327 at week 52	76	ACR20 was comparable after continuous or switched treatment	Safety profiles were comparable for all treatment sequences	Not powered to show differences after switching Group sizes were reduced after switching Only interim OLE results published to date
GP2017 (adalimumab)
ADACCESS Blauvelt et al 2018 [66]	PsO	379	DB RCT Multiple switches: 3 alternating 6w periods between w17 and w35 Pooled continuous treatment pts vs pooled switched pts	34	PASI improved over time and was similar between treatment groups	No relevant safety or immunogenicity differences between treatment groups	Not powered to show differences after switching
NCT02744755 ADMYRA [69]	RA	284	OL RCT Single switch at w24 Continued GP2017 pts vs RP/GP2017 switch pts	22	No impact of switch on efficacy	No impact of switch on immunogenicity and safety	OL after switching Not powered to show differences after switching No continued RP control arm Results not yet published (EU-CTR data only)
MSB11022 (adalimumab)
NCT02660580 AURIEL-PsO [68]	PsO	443	Double blind RCT Single switch at w16 Continued MSB11022 pts vs continued RP pts vs RP/MSB11022 switch pts	36	Switching from RP to MSB11022 did not impact efficacy	Switching from RP to MSB11022 did not impact safety or immunogenicity	Not powered to show differences after switching Results not yet published (EU-CTR data only)
SB5 (adalimumab)
NCT02167139 Weinblatt et al. 2018 [67]	RA	542	DB RCT Single switch at w24 Continued SB5 pts vs continued RP pts vs RP/SB5 switch pts	28	ACR20/50/70 rates were comparable across treatment groups	Safety profile and incidence of ADAs were comparable across treatments 2.4%, 3.9%, and 6.4% of continued SB5, continued RP, and RP/SB5 pts discontinued treatment before end of study	Not powered to show differences after switching
CT-P10 (rituximab)
NCT01873443 Park et al. 2017 [77]	RA	87	OLE study Single switch after completion of a 48w or 72w DB study Continued CT-P10 pts vs RP/CT-P10 switch pts	56	Mean change in DAS28-ESR and proportion of pts with good/moderate EULAR responses was comparable between groups	ADAs were detected in 13.2% and 15.0% of pts in the continued and switch groups, respectively CT-P10 treatment was well-tolerated 8% and 15% continued CT-P10 and switch pts discontinued before end of study	Not powered to show differences after switching Small group sizes No continued RP control arm
GP2013 (rituximab)
ASSIST-RT Tony et al. 2019 [78]	RA	107	DB RCT Single switch in patients receiving the RP prior to randomization Continued RP pts vs RP/GP2013 switch pts	24	NR	Incidence of hypersensitivity and infusion-related reactions was similarly low in both treatment groups No NABs were reported No clinically meaningful differences in AEs between treatment groups	Unpowered study Small group sizes

*Number of patients randomized to a whole study with a switch design; for OLEs, patients randomized to the OLE. †Switching could occur in a single arm or both arms. Studies included in this table were identified from three systematic reviews of the literature [28–30] combined with a PubMed search performed in March 2019. ACR: American College of Rheumatology; ADA: antidrug antibody; AE: adverse event; AS: ankylosing spondylitis; ASAS: Assessment of Spondyloarthritis international society; CRP: C-reactive protein; DAS28-ESR: Disease Activity Score in 28 joints using erythrocyte sedimentation rate; DB: double-blind; HAQ-DI: Health Assessment Questionnaire Disability Index; hsCRP: high sensitivity C-reactive protein; IBD: inflammatory bowel disease; OL: open-label; OLE: open-label extension; NAb: neutralizing antibody; PASI: Psoriasis Area and Severity Index; PK: pharmacokinetics; PsA: psoriatic arthritis; PsO: psoriasis; pts: patients; RA: rheumatoid arthritis; RCT: randomised controlled trial; RP: reference product; SAE: serious adverse event; SpA: spondyloarthritis; TEAE: treatment-emergent adverse event; w: weeks.

In order to obtain an interchangeable designation from the FDA, studies which include at least three switches are required according to draft guidance [86]. However, the EMA does not make interchangeability decisions and therefore, from a regulatory perspective, multiple switching studies are not necessary. To date, three studies have been published that evaluated multiple switching in IMIDs. Two of the studies were multicenter, randomized, double‐blind, phase 3 studies in chronic plaque psoriasis, which evaluated reference etanercept or adalimumab versus their respective biosimilars, GP2015 and GP2017. In both trials, a proportion of patients receiving either treatment underwent three 6-week switches between the reference product and the biosimilar. Both studies pooled the analyses of the switching and non-switching patients and found that multiple switches had no effect on treatment efficacy. Furthermore, the safety and immunogenicity profiles of GP2015 and GP2017 were unaffected by multiple switches [66,73]. The third study was an OLE of a multicenter, randomized, double‐blind, phase 3 study in rheumatoid arthritis, which evaluated the adalimumab biosimilar FKB327 compared with the reference product. At the end of the 24-week double-blind treatment period, patients were re-randomized so that two-thirds of patients continued on the same treatment as in the double-blind study and one-third switched to the alternate treatment. After 28 weeks in the OLE, all patients received FKB327 for another 48 weeks. Comparable efficacy, safety, and immunogenicity were reported in the controlled trial period. Interim analyses of the OLE indicated that none of the treatment sequences studied had an effect on efficacy, safety, and immunogenicity [65].

4.3. Switching studies – real-world evidence

As a number of biosimilars for IMIDs are approved for use in Europe, real-world studies, national registries, and large patient cohorts are now providing data on switching from reference products to biosimilars in large and diverse populations. These studies (Supplementary Table 1) provide additional data to complement the robust evidence provided by RCTs performed prior to approval. Many of these real-world studies evaluated infliximab biosimilars in inflammatory bowel diseases (IBD), chiefly CT-P13. This is unsurprising given that infliximab biosimilars were the earliest biosimilars for autoimmune diseases to receive approval in Europe (CT-P13 was approved in 2013) [87]. A systematic review of 53 observational studies evaluating a single switch from infliximab reference product to a biosimilar (predominantly CT-P13) has recently been published by Feagan et al. [30]. This review did not identify any significant risks associated with a single switch between reference and biosimilar infliximab; however, the authors noted that there were no studies investigating multiple switches or switching between biosimilars. Longer real-world studies are also now emerging for infliximab biosimilars and are providing reassuring evidence that switching from reference infliximab to a biosimilar does not impact clinical outcomes up to 2 years [37,46,88].

Beyond the infliximab studies, an etanercept biosimilar, SB4, was approved in 2015 [89] and results from real-world studies with this biosimilar are beginning to emerge. For example, a registry study evaluating the use of SB4 in a broad population of patients with rheumatic diseases has recently been published [90]. This study from the DANBIO Registry in Denmark demonstrated that disease activity remained stable after switching and that no major safety events occurred. It was noted that a small proportion (7%) of SB4-treated patients switched back to the reference product; however, reasons for switching back appeared to be subjective rather than relating to clinical outcomes. Additionally, retention rates were higher in patients in remission at switching than in those not in remission, possibly indicating that a different therapeutic strategy would have been more suitable for such patients.

Although this large and rapidly growing body of real-world evidence is generally supportive of switching, it is important to highlight that none of these studies were individually powered to detect differences between treatment arms and the majority do not have a comparator arm that includes patients receiving continuous treatment with the reference product. In addition, real-world studies have demonstrated variable discontinuation rates, with many reporting a much higher frequency of discontinuations compared with RCTs. The reasons for this remain to be fully elucidated; however, subjective disease worsening has been reported as a frequent cause of discontinuations, potentially as a result of the nocebo effect discussed in section 3.4 [91].

4.4. Safety

Safety is an important topic when considering switching a patient from a reference product to a biosimilar. In all RCTs and real-world studies conducted to date, there is little evidence of any significant differences in adverse event frequency or type following a switch from a reference product to a biosimilar. In addition, no concerns have been raised with regard to changes in immunogenicity after switching. However, it should be noted that studies are rarely powered to detect differences in safety or immunogenicity and therefore ongoing pharmacovigilance is essential to ensure patient safety.

5. Medical society perspectives on switching

Multiple medical societies in Europe have issued recommendations on the use of biosimilars, including switching between reference products and biosimilars. The European Crohn’s and Colitis Organisation, various national rheumatology societies, and the British Association of Dermatology have all expressed support for prescriber-initiated switching [27,50,51]. However, all emphasized the need for clear communication between patients and physicians and recommended that decisions on switching must be made jointly by patients and their HCPs. Furthermore, given the concerns over immunogenicity, the need for pharmacovigilance was reiterated. For example, the British Society for Rheumatology emphasized the need for close monitoring of patients who switch from a reference product to a biosimilar and suggested that the option of reverting back to the reference medicine should be retained in case of any changes in the patient’s condition or in case any adverse events occur [92].

These conclusions were supported by the Task Force on the Use of Biosimilars to Treat Rheumatological Diseases, an international multidisciplinary panel of 25 physicians specializing in IMIDs from the US, Japan, and eight European countries. Based on the consensus-based recommendations of this panel as well as patient representatives and pharmacologists, the Task Force concluded that adequate evidence exists to support the decision to switch from a biologic to its biosimilar, but that the decision to switch must be made jointly by patients and their HCPs [93].

Given the importance of nurses in implementing the switch between medications in some countries, the guidance of the European Specialist Nurses Organisation (ESNO) is also worth noting. In the 2017 Communication and Information Guide for Nurses, ESNO stressed the importance of communication with patients, including information on the nocebo effect. ESNO also recommended the use of real-world data to support communication with patients and colleagues [94].

6. Automatic substitution at the pharmacy level

Switching to a biosimilar should be at the discretion of the treating physician [38,55]. However, to increase uptake, automatic substitution of biosimilars at the pharmacy level is of interest to payers. Substitution refers to the replacement of a reference product with a biosimilar at the pharmacy level, without consulting the prescriber. Automatic substitution is already in use for generic drugs, but its role in the biosimilar market remains a highly controversial issue and is regulated differently across health-care systems [38,55]. According to draft FDA guidance, in the US manufacturers can apply for an interchangeable designation for their biosimilars. Once an interchangeable designation is granted, individual states can permit substitution of the biosimilar for the reference product once they have agreed and enacted substitution laws [86]. In Europe, however, the EMA only provides marketing authorization for a biosimilar. Decisions regarding switching, interchangeability, and substitution are taken at a national level [38], and are often guided by medical society-specific positioning papers and recommendations [95]. Medical societies in Europe are currently opposed to substitution at the pharmacy level for multiple reasons, including the risk to traceability, the difficulty of maintaining pharmacovigilance, and the lack of data on the effects of multiple switching [27,50,93,95]. In contrast to the medical societies, the European Association of Hospital Pharmacists (EAHP) considers biosimilars approved by the EMA to be interchangeable with the reference products, but not with other biosimilars. However, the EAHP reiterated the need for physician and patient involvement in decision-making [96].

7. Payer perspectives on switching

As demand for important but expensive biologics continues to grow, there is a clear need to reduce the burden on already constrained health-care budgets. From a payer perspective, the introduction of biosimilars is an attractive proposition due to the potential for significant cost savings and expanded patient access. In addition, treating patients with biologics at an earlier disease stage may reduce long-term costs as a result of delaying or avoiding costly surgical interventions [97,98]. While the prescribing of biosimilars to treatment-naïve patients will contribute to long-term savings, switching existing patients will be essential to realize the immediate financial benefits of biosimilars. The importance of switching to payers is clearly illustrated in the commissioning framework published by NHS England, which sets out an aim to switch at least 80% of existing patients to the best-value biological medicine within 12 months from the launch of a biosimilar product [99]. Similarly, in 2015, nationwide guidelines in Denmark dictated a mandatory switch from the infliximab reference product to a biosimilar (CT-P13) due to the fact that the cost of CT-P13 was 36% of the cost of reference infliximab [41]. In order to achieve sustained savings, payers may utilize various levers. These include tendering procedures to foster competition [100,101], and financial incentives such as gain share agreements to drive uptake of biosimilars [101]. Longer-term strategies include supporting real-world studies in order to address evidence gaps and increase physician confidence in the use of biosimilars [100]. An important example of this is the NOR-SWITCH study, which was funded by the Norwegian Ministry of Health and Care Service and concluded that switching to the infliximab biosimilar CT-P13 was not inferior to continued treatment with the reference product in terms of efficacy, safety, and immunogenicity [60].

8. Addressing physician and patient concerns

Although switching from a reference biologic to a biosimilar seems to be supported by the evidence available to date, and by medical societies in Europe and many national regulatory authorities, there remains a need to address the physician and patient reservations discussed earlier.

8.1. Addressing physician concerns

For physicians, real-world studies can provide additional effectiveness and safety data to supplement the strong evidence obtained from RCTs and to increase confidence in prescribing biosimilars [57]. Data from post-marketing studies and real-world experience can also be used to inform best practice. However, there may still be a need for educational initiatives to ensure that data are effectively disseminated to physicians in clinical practice [39]. Finally, biosimilar manufacturers must engage with clinicians in order to build credibility and trust and provide information and education.

8.2. Addressing patient concerns

Education will play a crucial role in gaining acceptance of biosimilars from patients. This necessitates patient education programs through HCPs, patient groups, and regulators that explain that biosimilars are alternative treatments to reference biologics and highlights the nature of the development and approval processes. For example, the EMA provides an online animated video for patients that explains key facts on biosimilars [102]. In addition, patients could also be informed of the societal benefits of switching, i.e. expanding access to expensive treatments and supporting the development and use of new and innovative products. Patient empowerment is also thought to be vital to ensure successful switching; specifically, no switching should occur without patient consent or knowledge, and policies need to be implemented to ensure open communication and shared decision-making between patients and physicians. Close monitoring and pharmacovigilance are also necessary, and the option to switch back to the reference product should always remain available. These proposals are supported by the European League Against Rheumatism standing committee of People with Arthritis and Rheumatism (PARE). In their position paper published in 2015, PARE recommended that the scientific community make lay summaries available for all relevant and important studies involving biosimilars [103]. The group also emphasized the need for reliable codes of practice and recommendations in order to build confidence and widen understanding of the use of biosimilars among patients. The benefits of enhanced patient education and shared decision-making strategies were highlighted in a recent real-world study in the Netherlands, which reported higher acceptance and persistence rates after switching from infliximab to CT-P13 after the introduction of an enhanced communication strategy [104].

8.3. A role for patient stratification?

To address these complex patient needs, a one-size-fits-all approach may not be appropriate. Therefore, it may be beneficial to stratify patients using the Beliefs about Medicine Questionnaire (BMQ), which assesses representations of medication prescribed for personal use and beliefs about medicines in general. This questionnaire categorizes patients according to specific needs and concerns into four groups: skeptical, indifferent, accepting, and ambivalent [105]. The BMQ has previously been used in the assessment of medications for IBD [106,107], and it has recently been used in the assessment of patient acceptance of the infliximab biosimilar CT-P13 for the treatment of IBD [108]. Using an established tool like the BMQ to stratify patients according to their underlying perspective before switching would allow information to be provided according to individual need.

9. Conclusions

Switching patients from reference biologics to more cost-effective biosimilars offers potentially significant cost savings, which can be used to expand access to biologics and support sustainability of health-care systems. Biosimilars are approved following a rigorous development process and the evidence from RCTs, performed predominantly prior to approval, appears supportive of switching from a reference product to a biosimilar. Furthermore, real-world, post-marketing studies and clinical experience have provided results that are consistent with the evidence demonstrated by RCTs. Based on these data, many medical societies have now expressed their support for switching, conditional upon initiation by the physician and patient consent. Addressing residual patient and physician concerns through education and ongoing real-world experience with biosimilars will play an important role in ensuring the acceptance and success of switching to a biosimilar, and ultimately to realizing the full benefits of biosimilars to society.

10. Expert opinion

Since their introduction in the 1990s, biological therapies have significantly improved outcomes for patients with IMIDs. However, due to their high costs, the growing use of these therapies has placed a significant financial burden on health-care systems. To ensure costs are contained, strict eligibility criteria are in place in many countries, limiting patient access and often producing inequities in the provision of these drugs. Increasing access to biologics within the same budget constraints is now a key challenge facing health-care providers, and the introduction of biosimilars is expected to be an important step forward in addressing this issue. Biosimilars are generally less expensive alternatives to already-licensed biological therapies and are only approved following a rigorous developmental process that includes an extensive analytical comparability exercise as well as comparative clinical studies. These studies are now complemented by the real-world clinical experience gained since the introduction of the first biosimilar for IMIDs, CT-P13 (infliximab), in Europe in 2013.

Initiating treatment-naïve patients on a biosimilar is well accepted by both physicians and patients. However, switching existing patients from a reference product to a biosimilar will also be required in order to achieve the full financial benefits of biosimilars. As shown in Table 2 and Supplementary Table 1, there are a growing number of RCTs and real-world studies that have evaluated the impact of switching from a reference product to a biosimilar. Studies performed to date have consistently found no impact of a single switch on efficacy, safety, or immunogenicity. These data are reassuring; however, it should be noted that most RCTs were not powered to determine equivalence after switching and data on multiple switching remain scarce. Additional studies will be required to address remaining evidence gaps and to gain insights into the impact of multiple switching, including switching between biosimilars. The data obtained from real-world studies have indicated that treatment discontinuations can be higher with biosimilars compared with reference products. However, most studies concluded that these discontinuations were predominantly for subjective reasons rather than as a result of changes in objective clinical outcomes. Although the reasons for this are uncertain, it has been postulated that this observation may be due to a nocebo effect as a result of patients’ preconceived perceptions of biosimilars.

As a result of the growing body of evidence supporting switching, medical societies in Europe are now generally supportive of prescriber-initiated switching. However, they have emphasized the need for joint decision-making between physicians and patients, and the importance of ongoing pharmacovigilance. Substitution at the pharmacy level remains unpopular with all medical societies, as well as with physicians and patients, and is therefore unlikely to be implemented in Europe in the near future.

As increasing numbers of biosimilars are approved as treatments for IMIDS, it is expected that the accumulating real-world evidence will address residual physician concerns and questions with regard to switching, leading to increased confidence in prescribing. Furthermore, patient education and empowerment will play a vital role in improving patient acceptance of biosimilars and ensuring that patients are confident that the benefits they have achieved with the reference product will not be compromised following a switch to a biosimilar. In particular, it will be important to mitigate the impact of any nocebo effect through improved patient education and optimized communication between patients and HCPs.

Health economists have predicted that considerable cost savings will be achieved following the introduction of biosimilars as a treatment for IMIDs. Increasing patient and physician confidence with switching will be essential to ensure that the full financial benefits of biosimilars can be realized in the coming years.

Article highlights

• Switching patients from reference biologics to biosimilars may lead to significant cost savings, which can be used to increase access to biologics within the same budget constraints.

• Biosimilar development and approval are rigorous processes requiring demonstration of the biochemical and clinical equivalence of the biosimilar and reference product.

• Randomized controlled trials performed to date have demonstrated a lack of impact of switching from a reference product to a biosimilar on immunogenicity, safety, and efficacy.

• Emerging real-world evidence from registries, observational studies, and clinical experience is also supportive of switching to biosimilars and is complementary to clinical trials.

• Medical societies have expressed support for switching, conditional upon joint decision-making by patients and physicians.

• Ongoing real-world experience and patient education programs will play a role in ensuring successful switching from reference products to biosimilars.

This box summarizes key points contained in the article.

Declaration of interest

C Edwards has received honoraria for attendance at advisory boards for AbbVie, Biogen, Bristol-Myers Squibb, Celgene, Fresenius Kabi, GlaxoSmithKline, Janssen, Lilly, Mundipharma, Roche, and Sanofi; and as a consultant for Anthera, Merck, and Samsung Bioepis; and has received grants as an investigator for AbbVie, Biogen, and Pfizer. J Hercogová has received honoraria for attendance at advisory boards for Novartis, Eli Lilly, Leo Pharma, Nordic Pharma, UCB, Sanofi Genzyme, and Fresenius Kabi; as an investigator for AbbVie, Merck, Amgen, Novartis, Eli Lilly, and Pfizer; and as a speaker for AbbVie, Biogen, Eli Lilly, Janssen-Cilag, Leo Pharma, L’Oréal, Nordic Pharma, Novartis, Pfizer, Sanofi Aventis, and Pierre Fabre. H Albrand is an employee of Fresenius Kabi SwissBioSim. A Amiot has received consulting fees from AbbVie, Hospira, Takeda, Gilead, and Biocodex; lecture fees and travel and accommodations from AbbVie, Janssen, Biocodex, Hospira, Ferring, Takeda, and Merck Sharp & Dohme; and advisory board fees from Gilead, Takeda, and AbbVie. 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.

Reviewer Disclosures

Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.

Acknowledgments

Fresenius Kabi SwissBioSim GmbH participated in the writing, review, and approval of the manuscript. All authors contributed to the development of the content; all authors and Fresenius Kabi reviewed and approved the manuscript; and the authors maintained control over the final content. Editorial and medical writing support was provided by Stephanie Carter, PhD, and Dan Hami, PhD, of Arc Medical Communications, Manchester, UK.

Supplemental Material

Supplemental data for this article can be accessed here.

Additional information

Funding

This paper was sponsored by Fresenius Kabi SwissBioSim GmbH.