Biosimilar infliximab (CT-P13) in the treatment of inflammatory bowel disease: A Norwegian observational study

Abstract

Objective: To assess the efficacy, tolerability, and safety of CT-P13 (Remsima^®) in patients with Crohn’s disease (CD) or ulcerative colitis (UC). Methods: This was a prospective observational study performed in a single center in Norway. Patients with CD (n = 46) or UC (n = 32) received CT-P13 (5 mg/kg) by intravenous infusion at weeks 0, 2, and 6. Efficacy end points included remission at week 14, measured by a Harvey-Bradshaw Index score of ≤4 or partial Mayo score of ≤2. Levels of the inflammatory markers C-reactive protein and calprotectin were measured. Adverse events up to week 14 were also recorded. Results: Seventy-nine percent of CD and 56% of UC patients achieved remission at week 14. Significant reductions in C-reactive protein and calprotectin occurred between baseline and week 14. There were no unexpected adverse events reported during the study. Conclusion: CT-P13 is efficacious and well tolerated in patients with CD or UC.

Keywords:

• biosimilar
• Crohn’s disease
• CT-P13
• efficacy
• inflammatory bowel disease
• infliximab
• Norway
• safety
• ulcerative colitis

Crohn’s disease (CD) and ulcerative colitis (UC) are the two main clinicopathologic subtypes of inflammatory bowel disease (IBD), characterized by relapsing periods of remission and acute flares of uncontrolled inflammation in the gut [1,2]. Over the past two decades, the introduction of a number of biologic drugs has dramatically improved patient outcomes in various gastroenterologic and rheumatologic immune-mediated inflammatory diseases, including IBD [3–10].

TNF is a multipotent cytokine and key mediator of inflammation and other processes involved in homeostasis. TNF antagonists, such as the monoclonal antibody infliximab, act by blocking TNF binding to its receptor thus neutralizing its activity and alleviating mucosal inflammation [11,12]. The clinical application of TNF antagonists and other biologic drugs has led to a significant rise in the number of IBD patients receiving treatment. Although this has improved patient outcomes, it has occurred at a much higher financial cost than that for traditional treatment options [13–15]. Consequently, widespread use of biologics has placed substantial financial burden on healthcare systems, in some cases restricting patient access to treatment [16,17]. Indeed in Norway, which provides an annual tender system for the supply of biologics, cost calculations and impact on healthcare budgets are now a key factor for drug selection.

In response to the often high costs of originator biologics, interest has grown in biosimilars; that is cost-effective, ‘follow-on’ biologic drugs with highly similar quality, efficacy, and safety to an already licensed biologic [15,18]. CT-P13 (Remsima^®, Inflectra^®) is a biosimilar of the infliximab ‘reference medicinal product (RMP)’ (Remicade^®) and the first monoclonal antibody biosimilar to be approved by the EMA. CT-P13 and the infliximab RMP are chimeric monoclonal IgG1 antibodies produced in independently developed cell lines derived from the same murine hybridoma cell type [11]. CT-P13 and its RMP share an identical amino acid sequence and have highly comparable higher-order structures [11,19].

Use of CT-P13 in the treatment of rheumatic disease was evaluated in two pivotal clinical trials, PLANETAS and PLANETRA. These randomized studies demonstrated the pharmacokinetic and efficacy equivalence of CT-P13 and RMP in ankylosing spondylitis (AS) and rheumatoid arthritis (RA), respectively [20,21]. The safety profiles of the two infliximab formulations were also comparable. Based on data from these clinical studies, in conjunction with non-clinical investigations which showed equivalent in-vitro activity of CT-P13 and RMP, CT-P13 was approved for all of the same indications as the RMP, including CD and UC, in Europe, Japan, and South Korea [11,22,23]. However, concerns have been expressed by some commentators regarding the ‘extrapolation’ of clinical data from rheumatic to gastrointestinal indications, in part because the pathologic role of TNF may vary in these diseases and as such the mechanism-of-action of CT-P13 and RMP may be different across these indications [24,25]. Indeed, Health Canada approved CT-P13 for the treatment of RA, AS, psoriatic arthritis, and plaque psoriasis but not for CD or UC due to observed differences between CT-P13 and RMP in some in vitro assays of antibody-dependent cellular cytotoxicity, a putative mechanism of infliximab in IBD [26]. In contrast, the EMA noted that no differences between CT-P13 and RMP were observed in more clinically relevant antibody-dependent cellular cytotoxicity assays [11].

To date, there is relatively limited published evidence regarding the efficacy and safety of CT-P13 in patients with CD or UC. Here, we present the results of a prospective observational study performed in a single center in Norway in which patients with CD or UC were treated with CT-P13.

Methods

Study design & patients

This prospective observational study recruited all patients with CD or UC who started treatment with CT-P13 at Akershus University Hospital between January 2014 and February 2015. All patients had moderate to severe IBD and had failed treatment with steroids and/or immunosuppressants. Some patients had received intravenous steroids without sufficient response. The decision to start treatment with TNF antagonist therapy was based on clinical symptoms, endoscopy, imaging, markers of inflammation (serum C-reactive protein [CRP] and calprotectin), previous disease course, and/or response to conventional therapy.

The study was approved as a quality assurance study by the Data Protection Official for Research at Akershus University Hospital and good clinical practice guidelines were followed. Written informed consent is not required in quality assurance studies.

Study treatment

For the majority of patients, CT-P13 (Remsima^®) was administered at a dose of 5 mg/kg via 2-h intravenous infusion at weeks 0, 2, and 6. For a small number of patients with severe UC, higher doses and an increased number of infusions of CT-P13 during the induction phase were administered during this time frame, as dictated by clinical need. Three such patients received an extra infusion after 5–7 days and one received a first infusion with 10 mg/kg bodyweight due to low albumin and high CRP. A fourth infusion was given to all patients at week 14 at the same visit where the evaluation of the response to induction therapy took place.

Efficacy assessments & end points

The efficacy end points assessed were the proportion of CD patients who achieved remission, as indicated by a Harvey-Bradshaw Index (HBI) score of ≤4, at week 14; the proportion of UC patients who achieved remission, as indicated by a partial Mayo score (pMayo) of ≤2, at week 14; the mean HBI score in CD patients at baseline and week 14; the mean pMayo and Simple Clinical Colitis Activity Index (SCCAI) scores in UC patients at baseline and week 14; and the mean levels of CRP and fecal calprotectin (Calprotectin ELISA; Buhlmann Laboratories AG, Basel, Switzerland) at baseline and week 14.

Pharmacokinetic (serum drug levels) assessments

Trough serum levels of CT-P13 were measured at week 14 using an automated immunofluorometric assay on the automated dissociation-enhanced lanthanide fluorescent immunoassay platform (AutoDELFIA; PerkinElmer, Turku, Finland). The presence or absence of anti-drug antibodies (ADAs) was also evaluated at week 14 in patients with trough serum levels of CT-P13 of 0 mg/l, using an automated assay designed to measure neutralizing antibodies, on the AutoDELFIA platform.

Safety assessments

Adverse events (AEs) were monitored from the start of the first infusion of CT-P13 until the end of the study at week 14. All AEs are reported, regardless of possible relationship to study drug.

Statistical analyses

All patients treated with at least one dose of CT-P13 were included in the safety analysis population. Efficacy and pharmacokinetics were analyzed in patients who provided relevant data at baseline and/or week 14. Ninety-five percent confidence intervals (95% CIs) were calculated for efficacy end points using Student’s t tests. The statistical significance of differences between HBI, pMayo, and SCCAI scores, and serum CRP and fecal calprotectin levels at baseline and week 14 was assessed using a Student’s t test. All two-sided p-values below 0.05 were considered significant.

Results

Patients

A total of 78 patients with IBD (46 with CD; 32 with UC) were treated with CT-P13 and were therefore included in the safety population. All patients received four or more infusions of CT-P13 with the exception of three patients. Two patients with UC each received two infusions of CT-P13 (one stopped treatment due to infusion reaction and the other was colectomized due to lack of response). One patient with CD had three infusions (treatment stopped due to infusion reaction). Patient demographics and disease characteristics at baseline are shown for the safety population in Table 1. For both CD and UC groups, there were similar numbers of male and female patients. Median age at baseline was 39 years in the CD group and 40 years in the UC group (range: 16–80 and 18–70, respectively). Thirteen patients (28%) in the CD group were previously treated with TNF antagonists (one of these patients had received two such therapies; the remainder had received one). All other CD patients (n = 33 [72%]) were naive to TNF antagonist therapy. In the UC group, five patients (16%) switched from TNF antagonists (or had been treated with TNF antagonists previously); of these patients, one had received three such therapies, another had received two, and the remainder had received one. Twenty-seven UC patients (84%) were TNF antagonist therapy-naive. In three CD patients, HBI calculation could not be done because of either ileostomy or colostomy. Endoscopy was performed in 30 UC patients at the start of treatment with CT-P13 and a Mayo endoscopic subscore of 2 and 3 was present in 7 and 23 patients, respectively.

Table 1. Baseline patient demographics and clinical characteristics (safety population).

Characteristic	Crohn’s disease	Ulcerative colitis
n	46	32
Gender, n, males/females	22/24	17/15
Age, years
Median (range)	37.5 (16–80)	43 (18–70)
Mean	39	40
Disease duration, years
Median (range)	3.5 (0–38)	3 (0–44)
Mean	8	6.5
Disease activity
HBI score, mean (95% CI)	7.67 (6.07–9.28)^†	–
Mayo score, mean (95% CI)	–	9.90 (9.03–10.77)^‡
Previous TNF antagonist treatment, n (%)
Adalimumab (Humira^®)	10 (21.7)	4 (12.5)
Infliximab (Remicade^®)	4 (8.7)	2 (6.3)
Golimumab (Simponi^®)	0	2 (6.3)
Concomitant treatment, n (%)
Azathioprine/6-mercaptopurine	25 (54.4)	18 (56.3)
Steroids	5 (10.9)	18 (56.3)
Methotrexate	6 (13.0)	1 (3.1)
Site of disease, n (%)
Ileocolon	21 (45.7)	–
Terminal ileum	15 (32.6)	–
Colon	8 (17.4)	–
Upper gastrointestinal tract	2 (4.3)	–
Extensive colitis	–	19 (59.3)
Left-sided colitis	–	7 (21.9)
Proctosigmoiditis	–	5 (15.6)
Proctitis	–	1 (3.1)
Disease behavior, n (%)
Non-stricturing/non-penetrating	19 (41.3)	–
Penetrating	15 (32.6)	–
Stricturing	12 (26.1)	–
Previous bowel resection surgery, n (%)	17 (37.0)	–

^†n = 43; ^‡n = 30.

CI: Confidence interval; HBI: Harvey-Bradshaw Index.

Efficacy

Of the 72 patients for whom disease activity scores (HBI or pMayo) at week 14 were available, 52 (72%) were in remission. All patients in remission at week 14 were free of steroids. All patients except one who responded at week 6 stayed well to week 14. An HBI score of ≤4 was achieved in 34/43 (79%) CD patients. In addition, the mean (95% CI) HBI score was significantly lower at week 14 than at baseline (2.41 [1.72, 3.09] vs 7.67 [6.07, 9.28], p = 0.0001; Figure 1A). Eighteen of 32 (56%) UC patients achieved a pMayo score of ≤2 and the mean (95% CI) pMayo score was significantly reduced at week 14 compared with baseline (2.10 [1.35–2.86] vs 7.19 [6.46–7.92], p = 0.0001; Figure 1B). In UC patients, a reduction in mean (95% CI) SCCAI score was also observed at week 14 compared with baseline (3.27 [2.30–4.23] vs 9.88 [8.83–10.92], p = 0.0001; Figure 1C).

Figure 1. Mean (95% CI) disease activity scores at baseline and after 14 weeks of treatment with CT-P13 in patients with IBD. (A) Harvey-Bradshaw Index (HBI) score in CD patients. (B) Partial Mayo (pMayo) score in UC patients. (C) Simple clinical colitis activity index (SCCAI) score in UC patients.

Improvements in disease activity scores were accompanied by reductions in inflammatory markers (Figure 2). Mean (95% CI) serum CRP levels were significantly reduced at week 14 compared with baseline in CD patients (4.9 mg/l [2.4–7.4] vs 22.5 mg/l [11.3–33.7], p = 0.006; Figure 2A) and UC patients (9.6 mg/l [4.0–15.2] vs 36.8 mg/l [18.0–55.7], p = 0.012; Figure 2B). Significant reductions in mean (95% CI) fecal calprotectin levels were also observed (CD patients: 214 mg/kg [53–374] vs 1148 mg/kg [756–1540], p = 0.0002; Figure 2C; UC patients: 857 mg/kg [458–1255] vs 2582 mg/kg [2016–3147], p = 0.0001; Figure 2D). The majority of IBD patients had normal or only slightly elevated calprotectin levels following 14 weeks of treatment (Table 2). At week 14, 32 of 36 (89%) CD patients and 9 of 22 (41%) UC patients had normal or slightly elevated calprotectin levels.

Figure 2. Levels of inflammatory markers at baseline and after 14 weeks of treatment with CT-P13 in patients with IBD. (A) Mean (95% CI) serum C-reactive protein in CD patients. (B) Mean (95% CI) serum C-reactive protein in UC patients. (C) Mean (95% CI) fecal calprotectin in CD patients. (D) Mean (95% CI) fecal calprotectin in UC patients.

Table 2. Categorization of fecal calprotectin levels at baseline and after 14 weeks of treatment with CT-P13 in patients with IBD.

Calprotectin level (mg/kg)	CD, % (n/N)		UC, % (n/N)
	Baseline	Week 14	Baseline	Week 14
Normal (<200)	23% (9/39)	82% (30/36)	0% (0/26)	27% (6/22)
Slightly elevated (≥200–≤500)	15% (6/39)	6% (2/36)	4% (1/26)	14% (3/22)
Moderately elevated (>500–≤1000)	15% (6/39)	6% (2/36)	8% (2/26)	23% (5/22)
High (>1000)	47% (18/39)	6% (2/36)	88% (23/26)	36% (8/22)

CD: Crohn’s disease; IBD: Inflammatory bowel disease; UC: Ulcerative colitis.

Only 13 CD patients had previously received treatment with a TNF antagonist. In these patients, there was a significant improvement in HBI, mean (95% CI), 7.08 (3.94–10.21) versus 2.25 (0.87–3.64), p = 0.0103. No significant differences were shown for calprotectin and CRP (due to small numbers). Only five UC patients had previously received TNF antagonist therapies and the number is too small to make any statistical calculations.

Pharmacokinetics (serum drug levels)

Trough serum levels of CT-P13 were measured at week 14. Median (range) trough CT-P13 concentration was 6.8 mg/l (0.0–26.1) in CD patients and 6.2 mg/l (0.0–10.9) in UC patients. Four CD patients and four UC patients had trough levels of 0 mg/l. Two of these patients had high ADA levels (≥80 AU/l), five had medium/high ADA levels (<80 AU/l), and one had low ADA levels (<10 AU/l). All eight patients received monotherapy with CT-P13 and three had previously been treated with TNF antagonists. One UC patient was treated with infliximab RMP in 2008 and again in 2013, but stopped 18 months before starting treatment with CT-P13. A severe infusion occurred at the second infusion and the treatment with CT-P13 was stopped. Two CD patients had been treated with adalimumab, but switched to CT-P13 due to loss of response. No information regarding ADAs to adalimumab in these two patients is available. One of them had low levels of ADAs to CT-P13, and after week 14 the treatment interval was shortened and satisfactory trough levels were achieved during further treatment.

Safety

All AEs observed during the study, regardless of their possible relationship to study treatment, are listed in Table 3. All AEs occurred in one patient only, with the exception of infusion reaction (one CD and UC patient) and colectomy (one UC patient). Two UC patients were referred to surgical therapy after week 14 because of continuous intestinal symptoms. One CD patient had muscle and joint pain due to antibody formation and undetectable trough serum levels so treatment was therefore stopped.

Table 3. Adverse events in IBD patients treated with CT-P13 for up to 14 weeks.

Patients with CD (n = 46)	n (%)
Infusion reaction	1 (2.2)
Skin rash (right shoulder)	1 (2.2)
Infectious enterocolitis (campylobacter)	1 (2.2)
Herpes zoster	1 (2.2)
Herpes simplex	1 (2.2)
Arthralgia	1 (2.2)
Fatigue	1 (2.2)
Recurrent upper airway infection	1 (2.2)
Muscle and joint pain	1 (2.2)
Patients with UC (n = 32)	n (%)
Pneumonia	1 (3.1)
Erysipelas (right knee)	1 (3.1)
Elevated transaminases	1 (3.1)
Infusion reaction	1 (3.1)
Palpitations	1 (3.1)
Colectomy	1 (3.1)

CD: Crohn’s disease; IBD: Inflammatory bowel disease; UC: Ulcerative colitis.

Discussion

This prospective observational study reported on clinical experience with CT-P13, a biosimilar of infliximab, in patients with IBD attending a single, tertiary care center in Norway. In both CD and UC patients, a significant reduction in disease activity was observed at week 14 (i.e., after three infusions of CT-P13) compared with baseline. Disease activity was assessed via HBI score in CD patients and pMayo and SCCAI scores in UC patients. Of those who provided efficacy data at week 14, 79% of CD patients and 56% of UC patients were in remission as assessed by HBI and pMayo score, respectively. Significant reductions in the inflammatory markers CRP and calprotectin were also observed at week 14 relative to baseline. No unexpected AEs were observed during the study.

Multi-center, randomized, double-blind clinical trials have established the efficacy and tolerability of the infliximab RMP in the treatment of IBD [3–10]. To date, similarly large clinical studies evaluating CT-P13 in patients with CD or UC have not been reported, although a number of such studies have been initiated. A randomized parallel-group Phase III study will assess safety and noninferiority of efficacy of CT-P13 and RMP in adults with CD [27]. A global registry study is currently recruiting adults or children aged ≥6 years with active CD and adults with fistulizing CD or UC [28]. The study will assess various safety, efficacy, and health economic end points. The Norwegian government is funding the NOR-SWITCH study, a randomized, double-blind comparison of CT-P13 and RMP in approximately 500 patients across all indications for which these drugs are approved in Europe, including CD and UC [29]. While data from these large studies of CT-P13 in IBD are awaited, currently available evidence suggests that this biosimilar is as effective as the infliximab RMP in CD and UC.

In the current study, CT-P13 was efficacious in both CD and UC patients, achieving similar clinical responses to those observed with the RMP in large randomized studies [3–10]. In line with these findings, a Korean study of 110 IBD patients (CD, n = 59 and UC, n = 51) found that clinical remission rates of 77.3 and 47.8% were achieved in TNF antagonist naive CD and UC patients, respectively, treated with CT-P13 [30]. Reports from other observational studies also support use of CT-P13 in IBD patients [31,32].

The finding that the efficacy and safety profiles of CT-P13 and RMP appear to be similar in IBD is consistent with the findings of randomized controlled trials comparing this biosimilar with its RMP in rheumatic disease. At week 30 of the PLANETAS study, Assessment in Ankylosing Spondylitis 20 and 40% responses, respectively, were observed in 70.5 and 51.8% of CT-P13-treated patients and 72.4 and 47.4% of RMP-treated patients [20]. At the same time point in PLANETRA, the proportion of RA patients with American College of Rheumatology 20 and 50% responses, respectively, was 60.9 and 35.1% for CT-P13 compared with 58.6 and 34.2% for RMP [21].

As reviewed elsewhere [33], evidence from non-clinical mechanism-of-action investigations and clinical pharmacokinetic studies also support the use of CT-P13 in IBD. Indeed, such data were a key part of the evidence upon which CT-P13 was approved for use in CD and UC by the EMA and other authorities. Because these authorities considered that the clinically relevant mechanisms of action of CT-P13 and RMP were equivalent, they allowed extrapolation of the clinical data collected in AS and RA to all indications, including IBD [33].

No unexpected immunogenicity or safety findings arose during the current study. AEs were consistent with those previously observed with CT-P13 and infliximab RMP [20,21]. Seven patients had undetectable trough serum levels of CT-P13 at week 14 and one patient at week 6; four CD patients and four UC patients. One CD patient had clinical effect and a decrease in CRP levels. Another CD patient had some initial clinical effect but CRP levels had increased at week 14. A third CD patient with undetectable trough serum levels of CT-P13 at week 14 was operated on after the second infusion because of problems with an internal fistula and widespread severe inflammation. A colectomy with ileostomy was performed. Treatment with CT-P13 continued after the operation and an infusion reaction occurred. One of the UC patients with undetectable trough serum levels at week 14 had an initial clinical effect and endoscopic improvement, but deteriorated and was referred for colectomy. Another UC patient had an infusion reaction at the second infusion and undetectable trough serum levels of CT-P13 at week six. This patient had previously been treated with infliximab in two periods, most recently in 2013. All patients with undetectable trough serum levels of CT-P13 had ADAs. In both PLANETAS and PLANETRA, the proportion of patients with ADAs was similar in the CT-P13 and RMP groups [20,21].

This case study currently has limitations; data were collated from only a single center and there was no control group or RMP treatment group. Despite these limitations, our results show positive clinical outcomes following administration of CT-P13 to CD and UC patients. As of August 2015, patients are still being followed in the study after prospective registration of disease activity, markers of inflammation, immunogenicity and other information, and some patients have been followed for >18 months. In the current paper, however, we have focused on the induction phase of treatment.

A key outcome of TNF antagonist therapy is attenuation of flare-ups by their use in maintenance therapy. Thus, further studies are required to compare the performance of CT-P13 in both induction and maintenance treatment of CD and UC. The current induction study cannot provide sufficient data on long-term maintenance and remission; however, data from the acute phase of dosing are positive and support long-term therapy studies.

Conclusions

In conclusion, this study demonstrated that CT-P13 was efficacious and well tolerated in the early treatment of CD and UC. Although data from larger, randomized controlled studies would be needed to prove comparability with the RMP in these indications, current evidence suggests no noticeable differences between these two infliximab formulations.

Key issues

• The TNF antagonist infliximab has improved outcomes in patients with inflammatory bowel disease; however, the high cost of biologic drugs in comparison to traditional therapy options may restrict access to treatment for some patients.

• A biosimilar to infliximab, CT-P13 has been approved by the EMA for indications including Crohn’s disease and ulcerative colitis, based on evidence of efficacy and safety in ankylosing spondylitis and rheumatoid arthritis and extrapolation of clinical data to other, non-rheumatic indications. However, little direct evidence exists regarding the efficacy and safety of CT-P13 in patients with inflammatory bowel disease.

• This prospective observational study was performed in a single center in Norway and demonstrated the efficacy and safety of CT-P13 in patients with Crohn’s disease or ulcerative colitis.

• Current evidence does not suggest a difference between infliximab reference medicinal product and CT-P13 in these indications.

Acknowledgement

The authors would like to thank the Department of Medical Biochemistry at Oslo University Hospital, Radiumhospitalet, for analysis of infliximab/CT-P13 trough levels and anti-drug antibodies.

Financial & competing interests disclosure

The authors have no 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. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. Editorial support (writing assistance, assembling tables and figures, collating author comments, grammatical editing and referencing) was provided by Rick Flemming (Aspire Scientific Limited, Bollington, UK) and was funded by Celltrion Healthcare Co., Ltd (Incheon, Republic of Korea).

Notes