Biosimilars in rheumatology: out of the laboratory and into practice

KEYWORDS:

• Biologic
• biosimilar
• rheumatoid arthritis
• drug therapies
• anti-TNF

In 2015, the rheumatology community saw the emergence of the first biosimilar products onto the market. Although biosimilars are not new to the medical community at large, their arrival into the field of rheumatology is new and comes with great anticipation.

Infliximab (Remicade) first received a European license in August 1999. The patent on Remicade expired in February 2015. In March 2015, two infliximab biosimilars were launched in the UK simultaneously: Remsima (NAPP Pharmaceuticals) and Inflectra (Hospira). Both contain the same biological product manufactured by Celltrion, the only difference being the marketing company. The National Health Service (NHS) list price for 100 mg vial of these compounds is £377.66 pence compared with Remicade, which is marketed at £419.62 pence [1]. Local agreements further reduced the cost, with discounts approaching 50% of the cost of the original Remicade price. In February 2016, the US Food and Drug Association (FDA) voted to approve Celltrion’s infliximab biosimilar in the US across all seven indications that Remicade holds a license for [2].

In the UK, the etanercept (Enbrel) patent expired in November 2015 and the first etanercept biosimilar was licensed in January 2016. The situation in the US is more complex. The original Enbrel patents were filed in 1995 but due to several delays were not approved until 2011. Previous legislation granted patent protection from the date of approval rather than the date of application. While this legislation has now changed, the Enbrel patents were filed under the old legislation. Roche, who own the patents, are claiming protection until 2028 but this is being contested by Novartis who have their own etanercept biosimilar currently awaiting FDA approval.

1. Biosimilars versus generics

There are important differences between generic and biosimilar products. Consider ibuprofen: a small molecule with a mass of 206 g/mol. Contrast this to infliximab that has a molecular mass of 144,190 g/mol [3]. Considering the striking difference in size, it is easy to appreciate the complexity of the two molecules is vastly different. The nature of monoclonal antibodies is such that even subtle alterations in manufacturing, storage, and transport can change the properties of the drug.

Acknowledging these differences, the European Medicines Agency (EMA) produced a document that defines a biosimilar as follows:

A biological medicine that is developed to be similar to an existing biological medicine (the ‘reference’ medicine). Biosimilars are not the same as generics which have simpler chemical structures and are considered to be identical to their reference medicines. [4]

In light of the differences between biologics and small molecules, the regulations regarding licensing of copies are more stringent. In addition to the requirement to demonstrate pharmacokinetic equivalence (which is where regulation stops for small molecular chemical entities), biosimilar molecules must also provide data supporting comparable efficacy and safety.

2. Study design for biosimilar trials

Biosimilar trials are equivalence studies. They differ from conventional randomized control trials that are designed to demonstrate superiority of one intervention over another. Equivalence studies introduce methodological challenges. Readers will be familiar with the concept of the ‘intention-to-treat’ analyses where all patients randomized to a specific arm will be analyzed irrespective of whether they complete follow-up or not. This approach is based upon the concept that differential dropout (e.g. more people leaving an intervention arm because of side effects) would lead to bias in favor of the treatment if the analyses were limited to only subjects that completed the protocol; in contrast, an intention-to-treat analysis would bias toward the null hypothesis. However, if you consider that a bioequivalence study is looking to prove the null hypothesis, then in the presence of dropout, an intention-to-treat approach would bias toward concluding that the biosimilar is equivalent (effectively reversing type 1 and type 2 errors). For this reason, equivalence studies should adopt ‘per protocol’ analyses, and present findings with 95% confidence intervals for a difference rather than providing a single p-value. The generally accepted standard to declare equivalence would be if the comparator compound confidence intervals lie within ±15% of the originator estimate.

3. Infliximab biosimilar trials

The launch of Celltrion was based upon two published randomized control trials. The first, PlanetRA, included 606 adult patients with rheumatoid arthritis (RA) who were biologic naïve and receiving Methotrexate [5]. The primary end point was American College of Rheumatology 20% improvement criteria (ACR20) response at 30 weeks. The 95% confidence interval for the biosimilar compound (then named CT-P13) compared to Remicade was −4% to +12%, achieving the equivalence standard of within the −15 to +15 range. Safety findings were reported on over the 30 weeks with no apparent imbalance. Documentation submitted to the EMA reported serious adverse events at 54 weeks in 14% of the CT-P13 arm compared to 10% for Remicade [6]. No differences in anti-drug antibody (ADA) levels were seen between the two arms at any time point. It is notable that ADA levels in PlanetRA study with both the biosimilar and reference products were higher than previously reported in infliximab studies though this may just be a reflection of improved assay sensitivity [7]. An open label extension of the PlanetRA study has recently been published demonstrating efficacy of CT-P13 to 2 years [8]. It also showed that switching from Remicade to biosimilar infliximab did not appear to affect the safety or efficacy of treatment.

The second infliximab trial from Celltrion was PlanetAS [9]. This was a phase 1 trial in axial spondyloarthritis (AS) that recruited 250 biologic naïve patients. The study was not powered as an equivalence study, and therefore, 95% confidence intervals were not presented. However, the findings were similarly reassuring with comparable pharmacokinetic and clinical responses seen between the two infliximab products. There was no imbalance in safety signals reported at 30 weeks. ADAs were comparable across the compounds although notably lower (approximately 10% in both arms) at 14 weeks in the PlanetAS compared to PlanetRA. The low rate of ADAs reflects differences between RA and AS in terms of immunogenicity risk.

4. Extrapolation across indications

There is a subtle difference between the biosimilar and bio-originator infliximab. The committee for medicinal products in human use published a report on the comparison between Remsima and Inflectra and noted a reduced ability to induce cytotoxicity via a specific constant region (Fc) gamma pathway (FCγR 111A) for CT-P13 compared to Remicade [6]. Neutralization of ligand only requires the variable region of an antibody to be present. Cytotoxicity, however, requires activation via the Fc of an antibody, and therefore, changes in Fc structure could result in diminished ability to result in cytotoxicity.

This may be relevant in context of infliximab as the Fc effector component of the monoclonal antibody may explain why the drug is effective in both inflammatory arthritis and inflammatory bowel disease. In contrast, etanercept is ineffective in inflammatory bowel disease and lacks an Fc component. It was partly based on this consideration that regulatory authorities in Canada did not extrapolate licensing indications for biosimilar infliximab from RA and ankylosing spondylitis. The European regulators and more recently the FDA felt that the changes were too subtle to be relevant and offered licensing for the biosimilar infliximab across all bio-original indications.

5. Etanercept biosimilar

Samsung in conjunction with Biogen Idec has brought the first etanercept biosimilar (Benepali) to the European market. Phase 1 data were published for pharmacokinetic equivalence in healthy volunteers at the European league against rheumatism conference in abstract in 2015 [10]. The phase 3 equivalence trial was published in July 2015 [11]. About 596 adult patients with RA who were biologic naïve were randomized to biosimilar or bio-originator etanercept (Enbrel). The primary end point was an ACR20 response at 24 weeks. The 95% confidence interval was −9.4% to +5.0% confirming equivalence. There were no differences in serious adverse event rates. During the study, 0.7% of Benepali patients and 13.1% of Enbrel patients had detectable ADAs at any time point although none had persisting ADAs by week 24. The high level of ADAs seen with Enbrel has been questioned though all antibodies were non-neutralizing and transient in nature [12,13]. The presence of ADAs did not correlate with differential levels of clinical response.

It is worth mentioning that the evidence base for the etanercept biosimilar exceeds the evidence base that was available for Enbrel when it received its license 15 years ago!

6. Summary recommendations

In 2015, the British Society for Rheumatology published a position statement on biosimilars which highlights nine key considerations [14].

• 1. Prescribing should now always be done by brand name (to avoid unintentional switching between biosimilar and bio-original).

• 2. Prescribing should be for clinical reasons (not simply cost).

• 3. Substitution should only be with direct clinician consent.

• 4. Decisions to switch should be made in partnership with patients.

• 5. All patients should be registered with the British Society for Rheumatology Biologics Register.

• 6. There should be dedicated education to increase awareness around biosimilars.

• 7. Moving forward, robust technology appraisals (via The National Institute for Health and Care Excellence) must be conducted.

• 8. Better information sharing across clinical pathways will be vital for success.

• 9. Local tendering should aim to enhance rather than restrict choice for patients.

These may be seen as cautious recommendations. In contrast, Norway has adopted wholesale switching of patients on Remicade to biosimilar infliximab. Crucially, Norway has embarked upon the NOR-Switch trial, which will provide vital evidence regarding switching.

One thing is undoubtedly clear: increasing the competition within the biologics market will drive down costs. As clinicians we must use this opportunity to optimize access for our patients to this class of drug that has revolutionized outcomes.

Declaration of interest

AI Rutherford is funded by a clinical fellowship from the NIHR Biomedical Research Centre at Guy’s and St Thomas’ NHS Foundation Trust. JB Galloway has received honoraria for speaking from Pfizer, support for attending meetings from Abbvie, advisory board payments from Napp, Pfizer, UCB, Abbvie, MSD. 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.