Comparable long-term efficacy, as assessed by patient-reported outcomes, safety and pharmacokinetics, of CT-P13 and reference infliximab in patients with ankylosing spondylitis: 54-week results from the randomized, parallel-group PLANETAS study

Background CT-P13 (Remsima®, Inflectra®) is a biosimilar of the infliximab reference product (RP; Remicade®) and is approved in Europe and elsewhere, mostly for the same indications as RP. The aim of this study was to compare the 54-week efficacy, immunogenicity, pharmacokinetics (PK) and safety of CT-P13 with RP in patients with ankylosing spondylitis (AS), with a focus on patient-reported outcomes (PROs). Methods This was a multinational, double-blind, parallel-group study in patients with active AS. Participants were randomized (1:1) to receive CT-P13 (5 mg/kg) or RP (5 mg/kg) at weeks 0, 2, 6 and then every 8 weeks up to week 54. To assess responses, standardized assessment tools were used with an intention-to-treat analysis of observed data. Anti-drug antibodies (ADAs), PK parameters, and safety outcomes were also assessed. Results Of 250 randomized patients (n = 125 per group), 210 (84.0 %) completed 54 weeks of treatment, with similar completion rates between groups. At week 54, Assessment of Spondylo Arthritis international Society (ASAS)20 response, ASAS40 response and ASAS partial remission were comparable between treatment groups. Changes from baseline in PROs such as mean Bath Ankylosing Spondylitis Disease Activity Index (BASDAI; CT-P13 −3.1 versus RP −2.8), Bath Ankylosing Spondylitis Functional Index (BASFI; −2.9 versus –2.7), and Short Form Health Survey (SF-36) scores (9.26 versus 10.13 for physical component summary; 7.30 versus 6.54 for mental component summary) were similar between treatment groups. At 54 weeks, 19.5 % and 23.0 % of patients receiving CT-P13 and RP, respectively, had ADAs. All observed PK parameters of CT-P13 and RP, including maximum and minimum serum concentrations, were similar through 54 weeks. The influence of ADAs on PK was similar in the two treatment groups. Most adverse events were mild or moderate in severity. There was no notable difference between treatment groups in the incidence of adverse events, serious adverse events, infections and infusion-related reactions. Conclusions CT-P13 and RP have highly comparable efficacy (including PROs) and PK up to week 54. Over a 1-year period, CT-P13 was well tolerated and displayed a safety profile comparable to RP; no differences in immunogenicity were observed. Trial registration ClinicalTrials.gov identifier: NCT01220518. Registered 4 October 2010. Electronic supplementary material The online version of this article (doi:10.1186/s13075-016-0930-4) contains supplementary material, which is available to authorized users.


Background
Infliximab is a human-murine chimeric monoclonal antibody directed against tumor necrosis factor (TNF). Originator infliximab (Remicade®; hereafter referred to as the reference product (RP)) was the first TNF antagonist proven to be efficacious in patients with ankylosing spondylitis (AS) [1,2], and is now regarded as an important component of AS care [3,4]. However, the costs of infliximab RP and other originator biologics are often high, placing considerable financial burden on healthcare systems and, in many countries, restricting patient access [5,6]. As a result, the development of biosimilar drugs has garnered considerable interest as many originator biologics have reached or are close to patent expiry. A biosimilar may be defined as 'a biotherapeutic product that is similar in terms of quality, safety and efficacy to an already licensed reference biotherapeutic product' [7]. Due to their assumed lower price, biosimilars have the potential to reduce costs and increase patient access and drug therapy adherence.
CT-P13 (Remsima®, Inflectra®) is a biosimilar of RP produced in the same murine hybridoma cell line as RP and approved in Europe and elsewhere for the same indications. CT-P13 and RP are identical in amino acid sequence, pharmaceutical form, strength, composition, and route of administration [8,9]. Secondary and tertiary structures of CT-P13 and RP are highly comparable to each other, although not identical. Both drugs bind to known infliximab ligands and receptors (e.g., TNF, Fcγ and C1q) with similar affinity, while neither bind to lymphotoxin. In vitro studies have shown that CT-P13 and RP possess equivalent TNF neutralization activity, apoptotic activity, complement-dependent cytotoxicity and antibody-dependent cellular toxicity [8,9]. Biosimilar guidelines in Europe and the US state that demonstration of clinical comparability between a biosimilar and its innovator requires completion of comparator clinical trials assessing pharmacokinetics (PK), efficacy and safety [10,11]. Efficacy equivalence of CT-P13 and RP was established in the PLANETRA (Programme evaLuating the Autoimmune disease iNvEstigational drug cT-p13 in Rheumatoid Arthritis) study [12]. Another study called PLANETAS (Programme evaLuating the Autoimmune disease iNvEstigational drug cT-p13 in AS patients) was conducted to assess PK equivalence and comparability of efficacy and safety in patients with AS. The primary analysis of PLANETAS was performed at week 30 of the study [13]. That analysis showed that the primary endpoint of the study-PK equivalence of CT-P13 and RP as assessed by steady state area under the serum concentration curve (AUC) and maximum serum concentration (C max ) between weeks 22 and 30-was met. Efficacy and safety findings up to week 30 were also similar in the patients with AS.
In order to compare the extended efficacy, immunogenicity, PK and safety of CT-P13 and RP in patients with AS, PLANETAS continued up to 54 weeks (plus a subsequent off-dose period of 8 weeks). Here, we report the results of the study at week 54.

Methods
Study methods have been reported in detail previously [13].

Patients
AS patients aged 18-75 years meeting the 1984 modified New York classification criteria [14] for ≥3 months prior to screening, with Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score of ≥4 (range 0-10) and a visual analogue scale score for spinal pain of ≥4 (range 0-10) were eligible for the PLANETAS study. Patients were permitted to receive both oral glucocorticoids (equivalent to ≤10 mg daily prednisolone) and nonsteroidal anti-inflammatory drugs, if they had received a stable dose for ≥4 weeks prior to screening.

Study design and treatment
This randomized, double-blind, parallel-group study (ClinicalTrials.gov identifier: NCT01220518) was conducted in 46 centers in 10 countries. Patients were randomly assigned in a 1:1 ratio to receive CT-P13 (CELLTRION Inc, Incheon, Republic of Korea) or RP (Janssen Biotech Inc, Horsham, Pennsylvania, USA), at a dose of 5 mg/kg via 2hour intravenous infusion at weeks 0, 2 and 6, and then every 8 weeks thereafter up to week 54. Antihistamines were provided prior to infusion at the investigator's discretion. The overall randomization code was broken for regulatory reporting purposes once the database up to week 30 had been finalized for analysis [13]. However, the study remained blinded to the investigators and patients up to week 54.
The study was carried out according to the principles of the Declaration of Helsinki and International Conference on Harmonisation good clinical practice guidelines. The study was approved by the relevant independent ethics committees (see Acknowledgements). All patients provided written informed consent.

Study endpoints and assessments
Efficacy was assessed at baseline and weeks 14, 30 and 54. Efficacy endpoints included: patient-reported outcomes (PROs), including BASDAI, the Bath Ankylosing Spondylitis Functional Index (BASFI), and the Short Form (36) Health Survey (SF-36); the proportion of patients achieving clinical response according to Assessment of Spondyloarthritis international Society (ASAS)20 and ASAS40 criteria; the proportion of patients with ASAS partial remission; changes from baseline in Ankylosing Spondylitis Disease Activity Score (ASDAS)-C-reactive protein (CRP), Bath Ankylosing Spondylitis Metrology Index (BASMI) and chest expansion.
The proportion of patients who tested positive for anti-drug antibodies (ADAs) was assessed at baseline and weeks 14, 30 and 54 using an electrochemiluminescent immunoassay method, as previously reported [13]. The neutralizing activity of ADAs was also assessed by a flow-through immunoassay method utilizing the Gyros Immunoassay platform (Gyros AB, Sweden). Primary PK endpoints included the area under the concentration time curve over the dosing interval, at steady state between week 22 and week 30 (AUC τ ) and the observed maximum serum concentration at steady state between week 22 and week 30 (C max,ss ). Secondary assessments included C max and minimum concentration (C min ) immediately before the next infusion up to week 54. Blood samples were collected immediately before each treatment, at the end of each infusion, and 60 minutes post-infusion. PK analyses were performed using a flow-through immunoassay platform (Gyros AB, Sweden). The influence of ADA titer level on primary PK endpoints was assessed with a low, medium and high ADA titer based on tertile grouping of the data.
Regarding safety, adverse events (AEs) were assessed from enrollment and up to 8 weeks after last study treatment. A treatment-emergent AE (TEAE) was defined as any event not present before exposure to study treatment or an event that worsened in intensity or frequency after exposure to study treatment. Other safety assessments included vital signs monitoring, physical examination, clinical laboratory analyses, and tuberculosis (TB) monitoring. An AE of latent TB was reported if a patient had a positive result on interferon-γ release assay after a negative result at baseline together with a negative result on chest X-ray examination. Patients were monitored for infusion-related reactions, including hypersensitivity and anaphylactic reaction.
In a series of subgroup analyses, the impact of ADA status on efficacy, PK and safety was determined. For subanalysis of ASAS20 response, ADA status was determined at week 54. For C max subanalysis, a non-visit-based approach was adopted to evaluate the incidence of patients who developed ADAs up to and including week 54. In subgroup analysis of infusion-related reactions, patients were grouped according to their seroconversion status, which was defined as positive if the patient had a negative ADA test followed by a positive ADA test at a subsequent visit; all other patients who provided at least one ADA test result were included in the non-seroconverted group.

Statistical analyses
Statistical analyses were performed using Version 9.1.3 of SAS software (SAS Institute, Inc, Cary, North Carolina, USA). All efficacy analyses were performed by intention-totreat, using the 'missing equals excluded' (MEX) approach. The proportion of patients achieving clinical response (ASAS20/ASAS40) was analyzed by a logistic regression model, with treatment group as a fixed effect and the stratification factors (region and baseline BASDAI score) as covariates. Treatment effect was estimated by calculating the odds ratio (OR) and 95 % confidence interval (CI). Other efficacy endpoints were analyzed using descriptive statistics.
For PK analyses, serum drug concentrations were summarized using quantitative descriptive statistics by actual treatment group, visit and time point. The PK population included patients who received at least the first five doses of study treatment, had sufficient blood concentration data to facilitate the calculation of PK parameters, and had no major protocol deviations. The safety population included all patients who received at least one (full or partial) dose of CT-P13 or RP during any dosing period.
Sensitivity analyses of ASAS response data were conducted using a last observation carried forward (LOCF) approach to compare the full-set analysis (LOCF) with the complete-case analysis (MEX).

Patient disposition and baseline characteristics
Overall, 250 patients were randomly assigned to study treatment (n = 125 in each group; Fig. 1). A total of 210 patients completed 54 weeks of treatment (n = 106 (84.8 %) and 104 (83.2 %) in the CT-P13 and RP groups, respectively). The most common reason for discontinuation was AEs (n = 13 (10.4 %) and 10 (8.0 %), respectively) in each group.
Baseline patient demographics and disease characteristics are shown in Table 1. In both the CT-P13 and RP groups the majority of patients were male (n = 99 (79.2 %) and n = 103 (82.4 %), respectively), and median age was 38 years in both groups (range 18-69 years). Baseline scores for efficacy assessments, including PROs, were similar between groups.

Efficacy
Efficacy was similar between both treatment groups, as measured by all efficacy endpoints. The proportion of patients achieving clinical response according to ASAS20 and ASAS40 criteria at weeks 14, 30 and 54 was similar between the CT-P13 and RP groups, as were ASAS partial remission rates (Fig. 2a). Logistic regression indicated no difference in ASAS20 responses between CT-P13 and RP at week 54 (0.89 (0.50, 1.59)). Similarly, there was no difference between CT-P13 and RP for ASAS40 responses at week 54 (1.26 (0.73, 2.15)). Sensitivity analyses of ASAS response rates using the full-set (LOCF) population also showed no differences between CT-P13 and RP (Fig. 2b). In a subgroup analysis performed according to ADA status, the proportion of ADA-negative patients achieving ASAS20 at week 54 was 72.3 % in the CT-P13 group and 73.1 % in the RP group. In comparison, 47.8 % and 60.0 % of ADApositive patients in the CT-P13 and RP groups, respectively, achieved ASAS20 at week 54.

Pharmacokinetics
A total of 223 (89.2 %) patients were included in the PK population (n = 113 and n = 110 in the CT-P13 and RP groups, respectively). C max and C min values were similar between the two treatment groups up to week 54 (Fig. 4).
The influence of ADAs on primary PK endpoints (C max,ss and AUC τ ) was also investigated. Figure 5 shows these endpoints by ADA titer level at week 30 in the CT-P13 and RP groups. There was a trend for both C max,ss and AUC τ to be lower with higher ADA titer levels. This impact of ADAs on drug exposure was similar in the two treatment groups. In a related subgroup analysis, mean (coefficient of variation) C max was comparable in the CT-P13 and RP groups at week 54 in both ADA-negative (142.98 (27.6) versus 135.27 (22.6) μg/mL, respectively) and ADA-positive (122.53 (31.1) versus 117.16 (25.6) μg/mL) patients.
Two cases of active TB were reported in the CT-P13 group and one case was reported in the RP group. A single case of malignancy, which was considered to be unrelated to treatment, was reported in the CT-P13 group (basal cell carcinoma). The onset date of this event was unknown, but was believed to have started 2 years previously, based on the medical history of the patient (melanocyte nevus). Treatment for the event was excision of a skin nodule of the nose. The event was considered to have recovered/resolved according to the investigator.
The proportion of patients who experienced at least one serious AE (SAE) was similar between the CT-P13 and RP groups (n = 10 (7.8 %) and 8 (6.6 %), respectively), as was the proportion who had an SAE considered related to treatment (n = 4 (3.1 %) and 5 (4.1 %); see Additional file 1 for details of treatment-related SAEs). The number of patients with at least one TEAE leading to discontinuation was comparable between the CT-P13 Fig. 2 Proportion of patients with an ASAS20 response, ASAS40 response and ASAS PR* up to week 54 by treatment in the a ITT population (MEX approach) and b ITT population (LOCF approach). "n = numbers" in the bar represent the denominator of patients assessed at each time point. *PR was defined as a value of <20 on a 0-100 scale in each of the following four domains: patient global assessment, pain, function, and inflammation. ASAS Assessment of Spondylo Arthritis international Society (ASAS20 and ASAS40, 20 % and 40 % response according to the ASAS International Working Group criteria for improvement), ITT intention-to-treat, LOCF last observation carried forward, MEX missing equals excluded, PR partial remission, RP reference product (i.e. reference infliximab) and RP groups (n = 11 (8.6 %) and 9 (7.4 %), respectively). Of these, 9 (7.0 %) and 7 (5.7 %) were considered to be related to CT-P13 and RP, respectively. Overall, treatment-related events leading to discontinuation of more than one patient were TB in two patients from the CT-P13 group and one in the RP group, two abnormal liver function tests (one in each group), and infusionrelated reactions in one patient from the CT-P13 group and five from the RP group. Two deaths occurred during the study, one in each treatment group. Both were due to car accidents and were considered not related to study treatment.

Discussion
PLANETAS was a multinational, randomized, doubleblind biosimilar study that evaluated the equivalence of PK and comparability of efficacy and safety between CT-P13 and RP up to 54 weeks in patients with active AS.
Similar clinical efficacy was observed between CT-P13 and RP up to week 54. ASAS responses at weeks 30 and 54, respectively, were comparable in the two treatment groups. PROs-including mean BASDAI, BASFI and SF-36 scores-and other efficacy endpoints such as BASMI score and chest expansion were also similar in both treatment groups up to week 54. The similarity in PROs is notable because, unlike the other efficacy endpoints employed in the study, these are patient-led evaluations.
Thirty-week efficacy data in PLANETAS were comparable to 24-week data from a pivotal placebo-controlled clinical study of RP in AS (ASSERT) [2,13]. Analogously, the efficacy data seen here for CT-P13 at week 54 are in line with findings previously reported for RP in different clinical trials with a similar duration [15][16][17]. For example, in the current analysis, the ASAS20 responses and improvement of BASDAI and BASFI scores with CT-P13 at 54 weeks were 67 %, 3.1 and 2.9, respectively. Similar improvements in these measurements were also observed at weeks 48-58 in historical studies with RP (ASAS20 at week 58 = 75 %; BASDAI from 6.6 at week 0 to 2.4 at week 54; BASFI from 6.1 at baseline to 4.4 at week 48) [15][16][17]. While there are inherent limitations associated with comparing across trials due to differences in patient populations and methods, it is reassuring that the data for CT-P13 are generally in line with historical observations for RP. Long-term efficacy equivalence of CT-P13 and RP has also been established in a 54-week analysis of the PLANETRA study in rheumatoid arthritis (RA). An American College of Rheumatology (ACR)20 response was achieved in a highly similar proportion of patients treated with CT-P13 and RP [18].
The long-term PK profiles (C max and C min ) of CT-P13 and RP were similar to each other in the current study through to week 54. The C max data reported here are consistent with PLANETAS data reported at 30 weeks and those of studies adopting a similar RP dosing pattern in Crohn's disease [13,19]. Similarity of PK was also observed between CT-P13 and RP in patients with RA in the PLANETRA study [12,18].
CT-P13 was well tolerated up to week 54, with a safety profile comparable to that of RP. As previously reported at week 30 for PLANETAS [13], there was no notable difference between study arms in the incidence of AEs, SAEs, infections and infusion-related reactions; this trend continued up to week 54. The safety profile observed with CT-P13 in the current study is also generally aligned with the safety profile of RP in historical studies involving patients with AS and RA [1,2,[15][16][17][20][21][22][23][24][25][26][27][28][29]. One treatment-unrelated case of basal cell carcinoma was reported with CT-P13. It is believed that the malignancy started 2 years prior to study participation. Following appropriate treatment (excision), the event was considered to have resolved.
It is well documented that development of ADAs against infliximab is associated with decreased clinical response and drug serum concentration [13,[30][31][32]. We also observed lower ASAS20 response rates and lower C max values at week 54, as well as a higher incidence of infusion-related reactions in ADA-positive compared with ADA-negative patients. These ADA effects were comparable in both the CT-P13 and RP groups. Further analysis of PK parameters with respect to ADA levels found a trend for both AUC τ , and C max,ss to be lower with increasing ADA titer level, although a formal statistical inference was not made due to a limited statistical power.
Further extensions of the current study and the PLA-NETRA study were conducted to examine the efficacy and safety of switching treatment from RP to CT-P13 in patients with AS and RA, respectively. Additionally, a formal randomized double-blind 'switching' clinical trial is currently progressing in Norway. This trial ('NOR-SWITCH'; ClinicalTrials.gov identifier: NCT02148640) is comparing the safety and efficacy of switching from RP to CT-P13 versus continued treatment with RP in several indications, including RA. Data from the PLA-NETAS/PLANETRA extensions and NOR-SWITCH will provide further evidence on the ability to switch from RP to CT-P13.
The main limitation of this analysis is that PLANE-TAS was primarily designed to compare the PK profiles of CT-P13 and RP at 30 weeks. While prospectively planned, the longer-term efficacy, safety and PK data presented here are from a secondary analysis of the study.

Conclusions
This multinational, randomized, biosimilar study in patients with active AS demonstrated that CT-P13 and RP have highly comparable efficacy, safety, immunogenicity and PK profiles up to week 54. Together with the findings from the PLANETRA study in patients with RA, these data further demonstrate the similarity between CT-P13 and RP in diverse clinical settings. Competing interests DHY and WP: Consultation for CELLTRION. FB reports consultancy fees from Gebro Pharma, and grants from Bioiberica, MSD, GSK, UCB, Roche, Pfizer, Celgene, Amgen, Grunenthal, Sanofi, Tedec-Meiji, and lecture fees from Bioiberica. SSH and SHL are full-time employees of CELLTRION. JBra has received honoraria for talks, advisory boards, paid consultancies or grants for studies from Abbvie (Abbott), Amgen, Boehringer Ingelheim, Bristol-Myers-Squibb, Celgene, CELLTRION, Centocor, Chugai, EBEWE Pharma, Janssen, Medac, MSD (Schering-Plough), Mundipharma, Novartis, Pfizer (Wyeth), Roche, Sanofi-Aventis and UCB. All other authors declare that they have no competing interests.
Authors' contributions WP, DHY, JBra, SSH and SHL were involved in the conception and design of the study, and/or analysis and interpretation of data; drafting of the manuscript and revising it critically for important intellectual content; and final approval of the version to be published. JJ, JBrz, AG, VK, IGS, CA-M, WJOE, SWK, DA and FB were involved in the acquisition of data; drafting of the manuscript and revising it critically for important intellectual content; and final approval of the version to be published.