Post-marketing study of biosimilar infliximab (CT-P13) to evaluate its safety and efficacy in Korea

Abstract

Objective: To evaluate the safety and efficacy of CT-P13 (Remsima^®) in patients with inflammatory bowel disease (IBD) in South Korea. Methods: This post-marketing study included patients with active moderate-to-severe Crohn’s disease (CD), fistulizing CD (FCD), or moderate-to-severe ulcerative colitis (UC) treated with CT-P13 and followed for 30 weeks. Assessments included treatment-emergent adverse events (TEAEs) and disease-specific clinical response and remission. Results: No unexpected TEAEs were observed in the 173 patients recruited to date. TEAEs occurred in 18.1, 16.7, and 26.9% of CD, FCD, and UC patients, respectively. Treatment-related TEAEs occurred in 10% of patients and were mostly mild-moderate in severity. There were five serious TEAEs (two infusion-related reactions, two infections, one abdominal pain) and no cases of malignancy, pneumonia, or death. Positive outcomes for response/remission were reported regardless of whether patients had received prior infliximab or not. Conclusion: CT-P13 was well tolerated and efficacious in patients with IBD.

Keywords:

• biosimilar
• Crohn’s disease
• CT-P13
• efficacy
• inflammatory bowel disease
• infliximab
• post-marketing study
• safety
• South Korea
• switching
• ulcerative colitis

Inflammatory bowel disease (IBD) and its two major forms, Crohn’s disease (CD) and ulcerative colitis (UC), are characterized by chronic inflammation of the gastrointestinal tract and periodic episodes of relapse and remission [1,2]. Tumor-necrosis factor (TNF) is involved in the inflammatory processes that underlie IBD [2] and serum levels of this cytokine are increased in patients with IBD [3]. The introduction of biological therapies, such as anti-TNF monoclonal antibodies, has revolutionized IBD management since these treatments can alleviate symptoms and have also been shown to reduce the need for hospitalizations and surgical procedures, induce mucosal healing, prolong periods of remission, and improve quality of life [4–7]. In addition, anti-TNF agents may help reduce the incidence of postoperative endoscopic recurrence after surgery [8].

Infliximab, the first anti-TNF agent to be licensed, is approved for the treatment of CD, and UC, as well as rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis, and psoriasis [9]. Since the launch of infliximab, other anti-TNF agents have become available for the treatment of IBD and there are currently four such products marketed in the USA (infliximab, adalimumab, certolizumab pegol, and golimumab) and three in Europe and Canada (infliximab, adalimumab, and golimumab) [10].

CT-P13 (Remsima^®, Inflectra^®) is a biosimilar of the infliximab reference medicinal product (RMP; Remicade^®) that was approved in Europe for all indications relevant for the RMP based on extensive comparability studies in vitro and in healthy human subjects and patients with RA and AS [11–14]. It is noteworthy that, following their guidelines on the evidence required for such a process, European regulators extrapolated comparability data observed between CT-P13 and RMP in rheumatic diseases to approve use in CD and UC [15]. Nevertheless, scientific debate is continuing around the issue of extrapolation and the potential for clinical differences in efficacy, safety, and/or immunogenicity between indications [16]. Indeed, some of these questions have led to differences in decisions between regulatory authorities around the world. For example, although CT-P13 has been licensed in IBD in the EU, South Korea [17] and Japan [18], use in IBD was not approved in Canada [19]. This was because the Canadian authorities felt that more evidence was needed to fully understand the relationship between the mechanism of action of infliximab and biosimilars in rheumatology, psoriasis, and IBD indications.

After launch of a new drug, an ever-increasing number of patients with potentially varied co-morbidities and concomitant medications become exposed to the drug during routine clinical practice. In these circumstances, postmarketing studies (PMS) are an important component of safety monitoring, helping to establish how the drug performs in ‘real-life’ settings and generating important data on benefit:risk profiles [20]. Rare safety issues not apparent under the controlled conditions of a clinical trial may arise as the number of exposed patients increases [21]. PMS conducted with the infliximab RMP in patients with RA have shown comparable efficacy and safety to that seen in controlled clinical trials [22,23]. So far, there is relatively limited clinical experience with CT-P13 in IBD, although some relevant data have been reported; for example, from a small case series performed in South Korea, the first country to approve CT-P13 [24]. To collect further data, a postmarketing clinical study of CT-P13 (Remsima^®) is being conducted in South Korea in patients with CD or UC. Interim results from this study are reported here.

Methods

This is an ongoing, open-label, phase 4 PMS conducted in 15 centers in South Korea to evaluate the safety and efficacy of CT-P13 in patients with IBD. The study was initiated on 23 January 2013. The cut-off date for the analysis reported here was 14 November 2014.

Patients

The study enrolled adult patients with moderate-to-severe active CD, fistulizing active CD (FCD), or moderate-to-severe active UC. For CD, mild disease was defined as a Crohn’s Disease Activity Index (CDAI) between 150 and 220, moderate disease as a CDAI between 220 and 450 and severe disease as a CDAI over 450 [25]. For UC, moderate-to-severe disease was defined as a Mayo score between 6 and 12 and an endoscopic sub-score of 2 or higher [26]. Patients with CD were eligible if they had not responded despite a full and adequate course of therapy with corticosteroid and/or an immunosuppressant or were intolerant to or had medical contraindications for these therapies. Patients with FCD were eligible if they did not show any response to general treatments (including antibiotics, drainage, and immunosuppressive therapy). Patients with UC were eligible if they had an inadequate response to conventional therapy including corticosteroids and 6-mercaptopurine or azathioprine, or were intolerant to or had medical contraindications for these therapies. Patients with contraindications to study treatment (in accordance with documented risks in the label) were excluded [11].

Written informed consent was obtained from all patients at the time of their enrolment. The study was conducted in accordance with the Declaration of Helsinki and its amendments and with 21 Code of Federal Regulations (CFR) Parts 50, 56, and 312.

Study design & treatment

All patients were followed for 30 weeks after initiation of CT-P13 treatment under routine clinical care and were classified according to whether they had received (‘switch’) or not received (‘naïve’) any doses of infliximab RMP before enrolment. Patients who were switched from RMP had previously been well during maintenance treatment. No formal criteria were used to determine when patients should be started on treatment with CT-P13 and this was determined by the investigators. The dose regimen administered, including dose escalation from the starting dose of 5 mg/kg if required, was consistent with the prescribing information for CT-P13 (Remsima^®) for an individual patient’s condition [11]. Intervals between doses were determined by the investigator according to clinical need and, due to the retrospective nature of the study, it was not possible to gather data on the actual treatment intervals used in this study. At the investigator’s discretion, patients could receive premedication with analgesics, anti-histamines or corticosteroids prior to, or during CT-P13 infusion. Concomitant medications were permitted, but any increase in dose or new treatments for IBD were considered as rescue medications.

Endpoints & assessments

To assess safety, all AEs, including serious AEs (SAEs) and adverse drug reactions occurring at any time throughout the study period were recorded, regardless of their causality relationship with CT-P13. AEs were coded using the Medical Dictionary for Regulatory Activities Version 16.0 which also provided the definitions of the severity of AEs to allow these events be characterized as mild, moderate, or severe [27].

Efficacy assessments were performed at baseline, and weeks 2, 6, 14, 22 and 30. For patients with moderate-to-severe active CD, efficacy was assessed by the proportion of patients achieving clinical response (defined by a ≥25% and ≥70 points decrease in CDAI score from baseline) or clinical remission (defined by CDAI score of <150) and, in patients who switched, by the proportion with disease control (defined as the absence of disease worsening, with worsening defined as an increase in CDAI of ≥70 points from the qualifying score with a total score of ≥175 and an increase in CDAI of ≥35% or more from baseline, or the introduction of a new treatment for active CD) [28].

For patients with FCD, efficacy was assessed by the proportion of patients achieving: clinical response (defined as a reduction of ≥50% from baseline in the number of draining fistulas) or clinical remission (defined as the absence of draining fistulas) and, in patients who switched, by the proportion with disease control (defined as no loss of response, with loss of response defined as the recrudescence of draining fistulas, the need for a change in medication for CD, the need for additional therapy for persistent or worsening luminal disease activity, the need for a surgical procedure for CD, or the discontinuation of study medication owing to a perceived lack of efficacy) [29,30].

Efficacy endpoints for patients with moderate-to-severe active UC were the proportion of patients achieving clinical response (defined by a decrease in partial Mayo scores from baseline of ≥2 points and ≥30%, with an accompanying decrease in the subscore for rectal bleeding of ≥1 point, or an absolute subscore for rectal bleeding of 0 or 1), clinical remission (defined as a total partial Mayo score of ≤2 points, with no individual subscore >1 point) or mucosal healing (assessed through endoscopy, and defined by Mayo endoscopic subscore of ≤1 point) and, in patients who switched, by the proportion with disease control (defined as the absence of disease worsening, with worsening defined by an increase in partial Mayo score of ≥3 points from baseline [before switching] and a partial Mayo score ≥5 points) [31,32].

For all patients, the proportion receiving rescue medication (defined as any new or increased dose of concomitant medication needed after the first infusion date to treat new or unresolved symptoms of CD or UC) was also an efficacy endpoint.

Statistical analyses

All statistical analyses were conducted using SAS Version 9.3 (SAS Institute, Inc.). Continuous data were summarized using descriptive statistics and categorical data were summarized using counts and percentages.

The safety population consisted of all patients who received at least one dose (full or partial) of CT-P13 during any dosing period. The analysis populations for each efficacy endpoint consisted of all patients who received at least one dose of CT-P13 and had at least one efficacy assessment following baseline.

Efficacy analyses were performed by subgroup and data are reported for clinical response, remission, and disease control at weeks 14 and 30. The naïve group consists of patients who had not received at least one dose of infliximab RMP before first infusion of CT-P13. The switch group consists of patients who had ever received at least one dose of infliximab RMP before the first infusion of CT-P13. The last observation carried forward (LOCF) method was considered for missing data imputation for clinical response and remission at weeks 14 and 30 for naïve CD, FCD, and UC patients.

Results

Patient disposition & baseline characteristics

One hundred and seventy-three patients were included in the safety population, comprising groups with moderate-to-severe CD (n = 83), FCD (n = 12), and moderate-to-severe UC (n = 78). Patient demographics and baseline characteristics for the safety population (shown according to disease type for naïve and switch subgroups) are summarized in Table 1. It is notable that many patients had CDAI scores below 220 or partial Mayo scores below 6 (both indicative of relatively low-severity disease), because no inclusion or exclusion criteria applied regarding baseline disease activity scores and a considerable number of patients had baseline scores measured after initial treatment (i.e., RMP or other anti-TNF agent). Because there were no formal inclusion or exclusion criteria, the decision to initiate treatment in patients with mild disease was taken by the individual investigator. However, there were no important differences between disease types except that patients with UC were slightly older in both the naïve and switch subgroups (mean age 45.2 and 42.9, respectively) than those with FCD (27.9 and 33.0, respectively) and CD (31.8 and 31.1, respectively). There were more male than female patients across all subgroups.

Table 1. Patient demographics & baseline disease characteristics (safety population).

	Moderate–severe CD		FCD		Moderate–severe UC		Total
	Naïve (n = 43)	Switch (n = 40)	Naïve (n = 8)	Switch (n = 4)	Naïve (n = 62)	Switch (n = 16)	Naïve (n = 113)	Switch (n = 60)
Age (years)
Mean	31.8	31.1	27.9	33.0	45.2	42.9	38.9	34.3
(SD)	(10.88)	(10.41)	(10.13)	(13.34)	(14.57)	(13.54)	(14.71)	(12.43)
(Range)	(18–67)	(19–60)	(18–48)	(20–45)	(19–74)	(21–64)	(18–74)	(19–64)
Sex, n (%)
Male	34 (79.1)	24 (60.0)	6 (75.0)	3 (75.0)	40 (64.5)	9 (56.3)	80 (70.8)	36 (60.0)
Female	9 (20.9)	16 (40.0)	2 (25.0)	1 (25.0)	22 (35.5)	7 (43.8)	33 (29.2)	24 (40.0)
Weight (kg)
Mean	57.1	58.9	71.2	64.4	58.5	58.0	58.8	59.0
(SD)	(12.13)	(13.45)	(11.77)	(12.36)	(9.67)	(9.46)	(11.26)	(12.34)
(Range)	(35–80)	(37–97)	(57–93)	(50–79)	(40–81)	(40–73)	(35–93)	(37–97)
Baseline disease activity^†
Mean	340.0	168.9	1.6	0.5	6.8	4.8	NA	NA
(SD)	(90.35)	(112.65)	(0.92)	(0.58)	(1.13)	(2.81)
(Range)	(181–564)	(24–412)	(1–3)	(0–1)	(3–9)	(0–9)
Time between last treatment with RMP and first treatment with CT-P13
<6 weeks, n (%)		3 (7.5)		0 (0)		2 (12.5)		5 (8.3)
≥6 weeks to <10 weeks, n (%)		22 (55.0)		3 (75.0)		7 (43.8)		32 (53.3)
≥10 weeks, n (%)		15 (37.5)		1 (25.0)		7 (43.8))		23 (38.3)
Median (weeks)		9.0		8.6		8.6		NA

^†Crohn’s Disease Activity Index for moderate to severe active CD, number of fistula for fistulizing active CD and partial Mayo scores in moderate to severe active UC patients were used to assess disease activity.

CD: Crohn’s disease; FCD: Fistulizing active Crohn’s Disease; NA: Not available; RMP: Reference medicinal product; SD: Standard deviation; UC: Ulcerative colitis.

One hundred and forty-five patients were included in the efficacy population (moderate-to-severe CD [n = 70], FCD [n = 10], and moderate-to-severe UC [n = 65]).

Exposure to CT-P13 & dose escalation

The majority of infliximab-naïve patients (102/113 [90.3%]) were exposed to at least three doses of treatment which spanned a 6-week induction period. Doses four, five and six were administered to 87 (77.0%), 77 (68.1%) and 59 (52.2%) infliximab-naïve patients, respectively. Overall, 52/60 (86.7%) patients in the switch group received three doses of CT-P13. Doses four, five and six were administered to 44 (73.3%), 39 (65.0%), and seven (11.7%) of these patients, respectively.

More than half of patients (59.0%) received a maximum dose of CT-P13 of 5 mg/kg (i.e., did not undergo dose escalation). Forty-one percent of patients were treated with maximum doses of >5 to ≤10 mg/kg. There were no notable differences in dose escalation between infliximab-naïve and switched patients (Table 2).

Table 2. Dose escalation by indication, n (%) (safety population).

Maximum dose (mg/kg)	Moderate–severe CD		FCD		Moderate–severe UC		Total
	Naïve (n = 43)	Switch (n = 40)	Naïve (n = 8)	Switch (n = 4)	Naïve (n = 62)	Switch (n = 16)	Naïve + switch (n = 173)
= 5	23 (53.5)	23 (57.5)	7 (87.5)	1 (25.0)	38 (61.3)	10 (62.5)	102 (59.0)
>5–≤7	16 (37.2)	10 (25.0)	1 (12.5)	1 (25.0)	22 (35.5)	6 (37.5)	56 (32.4)
>7–<10	1 (2.3)	2 (5.0)	0 (0.0)	0 (0.0)	2 (3.2)	0 (0.0)	5 (2.9)
= 10	3 (7.0)	5 (12.5)	0 (0.0)	2 (50.0)	0 (0.0)	0 (0.0)	10 (5.8)

CD: Crohn’s disease; FCD: Fistulizing active Crohn’s disease; UC: Ulcerative colitis.

Concomitant medications

The proportions of patients who used specific concomitant medications (steroids, azathioprine/6-mercaptopurine, and/or 5-aminosalicylate) before and during the study are summarized in Table 3. Across almost all indications and naïve/switch subgroups, the most commonly used concomitant medication of interest was 5-aminosalicylate.

Table 3. Concomitant medication by indication, n (%) (safety population).

	Moderate–severe CD		FCD		Moderate–severe UC
	Naïve (n = 43)	Switch (n = 40)	Naïve (n = 8)	Switch (n = 4)	Naïve (n = 62)	Switch (n = 16)
Steroids	6 (14.0)	12 (30.0)	0	0	34 (54.8)	5 (31.3)
Azathioprine/6-mercaptopurine	26 (60.5)	18 (45.0)	4 (50.0)	4 (100.0)	37 (59.7)	8 (50.0)
5-aminosalicylate	30 (69.8)	19 (47.5)	6 (75.0)	2 (50.0)	49 (79.0)	11 (68.8)

CD: Crohn’s disease; FCD: Fistulizing active Crohn’s disease; UC: Ulcerative colitis.

Safety

At least one treatment-emergent adverse event (TEAE) was recorded in 15 patients (18.1%) with moderate to severe active CD, two patients (16.7%) with FCD, and 21 patients (26.9%) with moderate to severe active UC. In total, 51 TEAEs were reported in 38 patients (22.0%). Of these, 22 events in 18 (10.4%) patients were considered to be related to treatment. Table 4 summarizes the treatment-related TEAEs observed in the study in the different patient subgroups. One severe TEAE was rated as treatment related (an infusion-related reaction in a patient in the switch group with moderate to severe active CD). Five SAEs were reported in the study, including the severe treatment-related infusion-related reaction (resolved), moderate treatment-related tuberculosis (improved), moderate treatment-related lung abscess (resolved), moderate treatment-related anaphylactic reaction (resolved), and severe treatment-unrelated abdominal pain (resolved). Of these, three led to patients discontinuing the study (infusion-related reaction, lung abscess, and anaphylactic reaction).

Table 4. Treatment-related^† treatment-emergent adverse events (safety population). Data are the number (percentage) of patients who experienced each event.

Event	Moderate–severe CD		FCD		Moderate–severe UC		Total
	Naïve (n = 43)	Switch (n = 40)	Naïve (n = 8)	Switch (n = 4)	Naïve (n = 62)	Switch (n = 16)	Naïve (n = 113)	Switch (n = 60)
Leukopenia	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Neutropenia	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Abdominal pain	1 (2.3)	-	-	-	-	-	1 (0.9)	-
Chest pain	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Pyrexia	-	-	-	-	2 (3.2)	-	2 (1.8)	-
Anaphylactic reaction	-	-	-	-	-	1 (6.3)	-	1 (1.7)
Lung abscess	-	-	-	-	-	1 (6.3)	-	1 (1.7)
Rhinitis	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Tuberculosis	-	-	1 (12.5)	-	-	-	1 (0.9)	-
Infusion-related reaction	-	2 (5.0)	-	-	-	1 (6.3)	-	3 (5.0)
WBC decreased	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Dizziness	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Hypoesthesia	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Photosensitivity	-	-	-	-	1 (1.6)	-	1 (0.9)	-
Pruritus	1 (2.3)	-	-	-	1 (1.6)	-	2 (1.8)	-
Rash	-	-	-	-	1 (1.6)	1 (6.3)	1 (0.9)	1 (1.7)
Skin lesion	-	-	-	-	1 (1.6)	-	1 (0.9)	-

^†An event was considered to be related to treatment if the relationship was defined as “definite,” “probable,” or “possible”.

CD: Crohn’s disease; FCD: Fistulizing active Crohn’s disease; GI: Gastrointestinal; UC: Ulcerative colitis; WBC: White blood cell; -: Not reported.

The incidence of AEs of special interest (infusion-related reaction, infection, tuberculosis, pneumonia, and malignancy) expressed as rate per 100 patient years (100PY; [95% CI]) is reported in Table 5. Across all patients and regardless of prior therapy, infusion-related reaction (i.e., any AE occurring at the time of infusion, including hypersensitivity and anaphylaxis) was reported in nine patients (10.06; [4.60–19.10]). Infections (both treatment-related and unrelated) were observed in nine patients (10.06; [4.60–19.10]). One patient (1.11; [0.03–6.23]) experienced active tuberculosis after CT-P13 exposure. No cases of malignancy, pneumonia, or death were reported during the study.

Table 5. Incidence of adverse events of special interest (all patients in the safety population).

Adverse event	Incidence n (%)	Incidence per 100PY^† (95% CI)
Tuberculosis	1 (0.6)	1.11 (0.03–6.23)
Infection	9 (5.2)	10.06 (4.60–19.10)
Pneumonia	0 (0.0)	NA
Malignancy	0 (0.0)	NA
Infusion-related reaction	9 (5.2)	10.06 (4.60–19.10)

^†Total exposure in the study was 89.45 patient years.

100PY: 100 patient-years; CI: Confidence interval; NA: Not applicable.

There were no notable dose-dependent differences in the distribution of TEAEs between IBD patients who received 5 mg/kg or >5 mg/kg maximum doses of CT-P13 (data not shown). Neither were meaningful differences observed for the proportions of IBD patients experiencing TEAEs at different doses in the infliximab-naïve group compared with the switch group. In the naïve group, 17 (15.0%) and 10 (8.8%) patients receiving 5 mg/kg and >5 mg/kg had a TEAE, respectively. In the switch group, seven (11.7%) and four (6.7%) patients receiving 5 and >5 mg/kg had a TEAE, respectively.

Efficacy results

Infliximab-naïve patients

Rates of clinical response and remission in infliximab-naïve patients are shown in Figure 1. Using LOCF missing data imputation, 87.2% (34/39) of patients with moderate-to-severe CD were in response at week 14 and 79.5% (31/39) at week 30, while 69.2% (27/39) and 59.0% (23/39) achieved remission at the same time points, respectively. In the case of FCD, the response rate was 66.7% (4/6) at week 14 and 66.7% (4/6) at week 30. The remission rate for patients with FCD was 33.3% (2/6) at week 14 and 50% (3/6) at week 30. For patients with moderate-to-severe UC, 75.5% (40/53) achieved a response at week 14 and 72.2% (39/54) at week 30, with 49.1% (26/53) at week 14 and 37.0% (20/54) at week 30 achieving remission.

Figure 1. Proportion of evaluable infliximab-naïve patients achieving (A) a response or (B) remission at weeks 14 and 30 (efficacy population). The LOCF method was used for missing data imputation.

Switched patients

Twenty-five of the 31 patients (80.6%) with moderate-to-severe CD, who switched from infliximab RMP to CT-P13 and who were evaluable, achieved or maintained remission through visits two to six. Eighteen of these 25 patients were in remission at the time of switching to CT-P13. Overall, 27 (87.1%) patients with moderate-to-severe CD who switched from infliximab RMP did not experience disease worsening and therefore were considered to have disease control.

Two of four (50.0%) evaluable patients with FCD who switched from RMP maintained remission. A majority of these patients (3/4 [75.0%]) maintained disease control (n = 1–4 at visits 2–6).

Of the 11 patients with moderate-to-severe UC who switched from RMP and were evaluable, five patients (45.5%) achieved or maintained remission throughout visits two to five; no patients experienced disease worsening and therefore all 11 patients were considered to have disease control. At the switching point, three of these five patients with moderate-to-severe UC were already in remission.

Mucosal healing in patients with moderate-to-severe UC

In the infliximab-naïve group, 28 out of 39 evaluable patients (71.8%) had evidence of mucosal healing at week 14. Similarly at week 30, nine out of 13 patients (69.2%) had experienced mucosal healing (Figure 2). Overall, six out of nine patients (66.7%) who switched from RMP to CT-P13 experienced mucosal healing throughout visits two to five.

Figure 2. Proportion of evaluable infliximab-naïve patients with UC experiencing mucosal healing over time (efficacy population).

Need for rescue medication

Rescue medication to control IBD symptoms was required by 48 of 173 (27.7%) patients. Overall, the requirement for rescue medication was higher in infliximab-naïve groups compared with switched groups, although this trend was solely driven by the moderate-to-severe UC subgroup (Table 6). No rescue medication was required by patients with FCD. Twice as many patients with moderate-to-severe UC required at least one rescue medication (32/78 [41.0%]) than patients with moderate-to-severe CD (16/83 [19.3%]).

Table 6. Rescue medication used by patients during the study, n (%) (safety population).

Medication	Moderate–severe CD		Moderate–severe UC		Subtotal		Total Naïve + switch^† (n = 173)
	Naïve (n = 43)	Switch (n = 40)	Naïve (n = 62)	Switch (n = 16)	Naïve^† (n = 113)	Switch^† (n = 60)
Total number of rescue medications	10	14	74	2	84	16	100
Patients with ≥1 rescue medication	7 (16.3)	9 (22.5)	30 (48.4)	2 (12.5)	37 (32.7)	11 (18.3)	48 (27.7)
Anti-diarrheals, intestinal anti-inflammatory/anti-infective agents	4 (9.3)	3 (7.5)	17 (27.4)	1 (6.3)	21 (18.6)	4 (6.7)	25 (14.5)
Corticosteroids for systemic use	1 (2.3)	5 (12.5)	15 (24.2)	1 (6.3)	16 (14.2)	6 (10.0)	22 (12.7)
Immunosuppressants	3 (7.0)	3 (7.5)	12 (19.4)	0	15 (13.3)	3 (5.0)	18 (10.4)

^†FCD patients included; no rescue medication use was reported in FCD patients.

CD: Crohn’s disease; FCD: Fistulizing active Crohn’s disease; UC: Ulcerative colitis.

Discussion

The biosimilar of the infliximab RMP, CT-P13, was approved for use in the EU in 2013 based on the recognition that all major physicochemical characteristics and biological activities of CT-P13 and RMP infliximab are comparable [12]. This approval by the EMA was received for the same indications as the infliximab RMP and permits use of CT-P13 in CD, UC, RA, AS, psoriatic arthritis and psoriasis. The approval of CT-P13 in IBD was based on extrapolation of data observed with CT-P13 in other indications (RA and AS), which has led to some debate. Because use of biosimilars in the treatment of IBD is a focus of scientific attention, additional clinical and PMS such as that reported here play a crucial part in facilitating appropriate use of such new therapies based on better understanding of benefits and risks [16,20].

The demographics and baseline disease characteristics of the patients in this study were consistent with approved target populations (moderate-to-severe active IBD) in the CT-P13 and infliximab RMP labels [9,11]. Overall, CT-P13 was well tolerated, and no unexpected AEs were observed. Ten percent of patients experienced a treatment-related TEAE, which were mostly mild or moderate in severity. As described, there were five SAEs; however, no cases of malignancy, pneumonia, or death were reported. There were no notable dose-dependent differences in the distribution of TEAEs between patients who received 5 or >5 mg/kg doses of CT-P13. The tolerability profile observed with CT-P13 in IBD appears to be at least in line with what has previously been reported with RMP. Although the number of patients involved in this study is small and direct comparisons cannot be made, the incidence of infusion-related reactions and infection appears to be lower than that previously published for RMP. In this study, the incidence per 100PY (95% CI) of both infusion-related reaction and infection was 10.06 (4.60–19.10) compared with 13.40 (11.07–16.09) for infusion-related reaction [33–38] and 20.12 (11.26–33.18) for infection [33,38] in historical studies with RMP.

Although the current study was not powered for efficacy, positive outcomes for response and remission were also reported in patients with CD, FCD and UC, regardless of whether patients had received any doses of infliximab RMP before enrolment. The proportion of CT-P13 treated patients achieving response and remission for both CD and UC were in line with outcomes reported in published studies with RMP [39–41]. Using the LOCF method for missing data imputation in this study, 79.5 and 59.0% of CT-P13 treated patients with CD achieved response and remission, respectively, at week 30. In SONIC, 56.2% of patients with CD treated with infliximab RMP achieved response and 47.9% remission at week 26 [39]. In ACCENT I, the 5 mg/kg dose of RMP produced 51.3% response and 38.9% remission at week 30 [40]. For patients with UC, in this study, 72.2% of patients achieved a response and 37.0% achieved remission at week 30. In ACT I, 52.1% patients with UC treated with 5 mg/kg of infliximab RMP achieved response at week 30 and 33.9% achieved remission. In ACT II, the equivalent outcomes were 47.1% achieving response and 25.6% achieving remission [41]. Less than 30% of patients in this study required rescue medication. The proportion of patients requiring rescue medication was higher in infliximab-naïve patients and patients with UC compared with those switched from RMP and with those suffering from CD.

The result of this study provides valuable data concerning the use of CT-P13 in clinical practice for patients with IBD. However, because it is a postmarketing study and is open-label and nonrandomized, it should be recognized that this may lead to an overestimate of efficacy when compared with randomized controlled trials. Furthermore, it was conducted in a single country, and was not powered to evaluate efficacy. Another limitation of the study was that analysis of anti-drug antibody incidence was not performed. To confirm the findings of this study and extend the clinical database on use of CT-P13 in IBD, the manufacturer of CT-P13 is currently conducting additional clinical studies, including a phase 3 study with a target recruitment of 214 patients with CD worldwide, which will examine the noninferiority in efficacy of CT-P13 compared with RMP, and also assess safety (ClinicalTrials.gov Identifier: NCT02096861 [42]). An IBD global registry is also planned that will follow at least 500 patients for 5 years. The primary objective of the registry is to assess the safety of CT-P13 in active CD, FCD or UC (ClinicalTrials.gov Identifier: NCT02326155 [43]). In addition, a study is ongoing in Norway to compare the safety and efficacy of switching from RMP to CT-P13 versus continued treatment with RMP in several indications, including IBD (“NOR-SWITCH”; ClinicalTrials.gov identifier: NCT02148640 [44]).

Conclusions

The current PMS demonstrated that CT-P13 is well tolerated in patients with CD, FCD, and UC. The results also suggest that CT-P13 is efficacious in this population of patients. Comparisons with historical data published for the infliximab RMP suggest that clinical outcomes such as safety and efficacy are comparable for CT-P13 and RMP. Overall, these findings emphasize the clinical similarity between CT-P13 and RMP in patients with IBD.

Key issues

• CT-P13 is a biosimilar of infliximab approved in Europe for the same indications as infliximab RMP, including IBD following extrapolation of comparability data from rheumatic diseases.

• Because use of biosimilars in the treatment of IBD is a focus of scientific attention, additional clinical studies and PMS can play a crucial part in facilitating appropriate use of such new therapies based on better understanding of benefits and risks.

• To provide further data on the efficacy and safety of CT-P13 in routine care, a postmarketing clinical study of CT-P13 is being conducted in Korea in patients with IBD; safety data for 173 patients are reported.

• The current PMS demonstrated that CT-P13 is well tolerated and efficacious in patients with IBD. Overall, these findings reinforce the clinical similarity between CT-P13 and RMP in patients with IBD.

Financial & competing interests disclosure

The study was sponsored by Celltrion. Y Kim, JH Lee, HJ Kwon, S Lee, D Park, HK Kim, JH Cheon, JP Im, and YS Kim have all received financial support for research from Celltrion, Inc., SY Lee and SJ Lee are employees of Celltrion Inc. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. Editorial support (writing assistance, assembling tables and figures, collating author comments, grammatical editing, and referencing) was provided by Mark O’Connor (Aspire Scientific Limited, Bollington, UK) and was funded by Celltrion Healthcare Co., Ltd (Incheon, Republic of Korea).

Notes