Long-term safety, efficacy, and immunogenicity of adalimumab biosimilar BI 695501 and adalimumab reference product in patients with moderately-to-severely active rheumatoid arthritis: results from a phase 3b extension study (VOLTAIRE-RAext)

ABSTRACT

Objective: To evaluate long-term safety, efficacy, and immunogenicity of BI 695501 in patients with moderately-to-severely active rheumatoid arthritis (RA) who have completed VOLTAIRE-RA.

Methods: Eligible patients for this phase 3b open-label extension study (VOLTAIRE-RAext), who had completed 48 weeks’ treatment with BI 695501 (Group A), 24 weeks each of adalimumab RP then BI 695501 (Group B), or 48 weeks of adalimumab RP (Group C) in VOLTAIRE-RA, were enrolled.

Results: Altogether, 430 patients received BI 695501 fortnightly for 48 weeks: Group A, n = 225; Group B, n = 103; Group C, n = 102. The proportion of patients with drug-related adverse events (AEs; overall 20.2%) was similar across Groups A, B, and C: 21.3%, 20.4%, and 17.6%, respectively. The majority of treatment-emergent AEs were non-serious and of mild/moderate intensity. Consistent with adalimumab RP’s safety profile, most drug-related AEs were in the system organ class infections and infestations. BI 695501 and adalimumab RP responses at the end of VOLTAIRE-RA were sustained during VOLTAIRE-RAext and all efficacy and immunogenicity endpoints were similar across groups.

Conclusion: Over 2 years, BI 695501 showed similar safety, efficacy, and immunogenicity to adalimumab RP, independent of initial treatment in VOLTAIRE-RA. No previously unknown adalimumab side effects were identified.

Clinical trial registration: NCT02640612.

KEYWORDS:

• Adalimumab
• BI 695501
• biosimilar
• rheumatoid arthritis
• tumor necrosis factor alpha inhibitor

1. Introduction

Biologic therapies have dramatically changed the management landscape of rheumatoid arthritis (RA) over the past 25 years. Tumor necrosis factor (TNF)-alpha inhibitors are now incorporated into routine clinical practice for treating RA [1–3]. Used in combination with methotrexate, TNF-alpha inhibitors have well-established clinical and safety profiles, with potential to modify radiographic progression [4–9]. Adalimumab reference product (RP; Humira, AbbVie Ltd, Maidenhead, UK), a fully human monoclonal antibody, has received regulatory approval for RA and several other diseases including psoriatic arthritis, inflammatory bowel disease, plaque psoriasis, axial spondyloarthritis, uveitis, and hidradenitis suppurativa [10,11]. The benefits of adalimumab have been well established in long-term trials and large registries such as the Corrona registry [12–14].

The high cost of biologic therapies is a drawback and can limit their use [3,15]. Biosimilars are alternative products with highly similar molecular structure, equivalent efficacy, and comparable safety and immunogenicity. Development programs for biosimilars are designed specifically to demonstrate similarity to the respective reference product [16]. Biosimilars may be economically preferable to their reference products, and cost savings have been realized following the introduction of anti-TNF biosimilars outside of the United States [11]. Several biosimilars to adalimumab have been licensed and others are in development. One example with established structural and biologic similarity to adalimumab is BI 695501 (Cyltezo, adalimumab-adbm; Boehringer Ingelheim, Germany) [12,17]. A phase 1 clinical study (VOLTAIRE-PK; NCT02045979) showed three-way pharmacokinetic similarity (bioequivalence) between BI 695501 and both the European Union (EU)-approved and United States (US)-licensed adalimumab RPs, as well as similar safety, tolerability, and immunogenicity profiles [18].

In 2017, BI 695501 was approved in Europe and the US for the treatment of moderately-to-severely active RA and other chronic inflammatory diseases in adults and children [12,17] based on the findings of the phase 3 VOLTAIRE-RA equivalence trial in RA (NCT02137226 [19]). Furthermore, no differences in efficacy, safety, and immunogenicity were observed when patients were switched from adalimumab RP to BI 695501. At the end of the 48-week treatment period, all qualifying patients had the option to enter the VOLTAIRE-RAext trial where they either continued on BI 695501 or switched from adalimumab RP to BI 695501. Here we describe the findings of the VOLTAIRE-RAext trial, which aimed to provide long-term safety, efficacy, pharmacokinetic, and immunogenicity data on BI 695501, administered via prefilled syringe, for patients with moderately-to-severely active RA.

2. Patients and methods

2.1. Study design and treatment

VOLTAIRE-RAext (NCT02640612) was a long-term, open-label, extension of the phase 3 VOLTAIRE-RA trial [19]. Briefly, VOLTAIRE-RA was a randomized, double-blind, parallel-arm, equivalence trial of BI 695501 and US-sourced adalimumab RP conducted in 14 countries, comprising a 48-week treatment period and 10-week safety follow-up. Patients with moderately-to-severely active RA on stable methotrexate were randomized 1:1 to receive either drug subcutaneously (SC) by prefilled syringe once every 2 weeks for 24 weeks (Figure 1). Then, patients in the adalimumab RP arm were re-randomized to either transition to BI 695501 or continue on adalimumab RP. Patients originally randomized to BI 695501 continued to receive it following a dummy re-randomization. At the end of the VOLTAIRE-RA treatment period, all qualifying patients could enter the open-label extension phase (i.e. the VOLTAIRE-RAext study) and receive BI 695501, regardless of the study drug they had received in the original trial. Patients receiving BI 695501 for 48 weeks in VOLTAIRE-RA were assigned to Group A; patients receiving adalimumab RP for 24 weeks followed by BI 695501 for a further 24 weeks during VOLTAIRE-RA were assigned to Group B; patients treated for 48 weeks with adalimumab RP in VOLTAIRE-RA were assigned to Group C (Figure 1). The VOLTAIRE-RAext trial consisted of a screening visit 14 days prior to day 1, a 48-week treatment period, and a 10-week safety follow-up period. The screening visit in VOLTAIRE-RAext was the week 48 visit in VOLTAIRE-RA. For clarity, study weeks will be discussed throughout this article from the start of VOLTAIRE-RA, i.e. week 50 corresponds with day 1 (week 0) of VOLTAIRE-RAext and week 98 (end of treatment) corresponds with week 48 of VOLTAIRE-RAext.

Figure 1. Study design. EOT; end of trial; RP: reference product.

*BI 695501 or adalimumab RP 40 mg/0.8 mL solution for SC injection, every 2 weeks via prefilled syringe. ^†Week 50 on treatment corresponds with day 1 (week 0) of VOLTAIRE-RAext(2 weeks after last injection in VOLTAIRE-RA).

Patients received BI 695501 40 mg/0.8 mL solution for SC injection, every 2 weeks via prefilled syringe. The study dose was administered at the screening visit by suitably qualified trial personnel, and subsequent injections were self-administered by patients either at the study site or at home. For at least 12 weeks, patients continued to receive methotrexate within the dose range for VOLTAIRE-RA (15–25 mg/week, or 10–14 mg/week if the higher dose could not be tolerated), with either folic acid (≥5 mg/week) or folinic acid (≥1 mg/week). After week 12, at the Investigator’s discretion, the methotrexate dose could be adjusted, combined, or replaced with another non-biologic disease-modifying anti-rheumatic drug. Patients were allowed to continue receiving any oral nonsteroidal anti-inflammatory drugs (NSAIDs) or corticosteroids (equivalent to prednisolone ≤10 mg/day) throughout the trial.

Blood samples (2.7–3 mL) were taken at weeks 50 (baseline), 62, 74, 98, and 108 to determine drug plasma concentrations, as part of the pharmacokinetic evaluation, and to determine the presence of anti-drug antibodies (ADA) and neutralizing anti-drug antibodies (nAbs). This could be taken at any time during each visit, except at week 62, when blood samples had to be taken before dosing and a second sample taken 1 to 6 h after dosing.

2.2. Patients

Eligible patients were those with moderately-to-severely active RA who had completed the 48-week treatment period in VOLTAIRE-RA, wished to participate in this extension trial, and could benefit from receiving BI 695501, based on the Investigator’s assessment. Patients were aged 18 to 80 years on entering VOLTAIRE-RA. Those of reproductive age were required to use appropriate methods of contraception throughout the extension trial and for 6 months after study discontinuation or completion.

Exclusion criteria included any drug-related serious adverse events (AEs), acquired disorders, infections, or hypersensitivity events experienced during VOLTAIRE-RA. Patients were also excluded if they had a significant disease other than RA, significant uncontrolled disease, or were taking any treatment (including biologic therapies) that would put them at unacceptable risk during the trial in the Investigator’s opinion. Full exclusion criteria are shown in the Supplementary Information.

The study was conducted in accordance with the Declaration of Helsinki and the protocol was approved by the appropriate independent ethics committees and institutional review boards. All patients provided written informed consent, consistent with the International Conference on Harmonization guidelines, Good Clinical Practice, and local legislation, prior to the week 48 assessment (i.e. the screening visit).

2.3. Objectives and endpoints

The objectives of this trial were to provide long-term safety, efficacy, pharmacokinetic, and immunogenicity data on BI 695501, administered via prefilled syringe, in patients with RA who had completed VOLTAIRE-RA. The primary endpoint was the proportion of patients with Investigator-assessed drug-related AEs during the treatment phase. The treatment phase began at the start of the extension trial (week 50) and finished at the end of the 10-week follow-up (week 108). AEs – defined as any unfavorable and unintended sign, symptom, or disease – included infections, hypersensitivity reactions, drug-induced liver injury, injection site reactions, and abnormal laboratory findings.

Secondary endpoints included assessments of efficacy at week 98, such as the change from baseline in Disease Activity Score 28 (DAS28)-erythrocyte sedimentation rate (ESR); the proportion of patients meeting the American College of Rheumatology 20% response criteria (ACR20); the proportion of patients meeting the ACR/European League Against Rheumatism (EULAR) definition of remission; and the proportion of patients with EULAR response (good, moderate, or no response). Further efficacy endpoints included the change from baseline in DAS28-ESR at weeks 62 and 74; at weeks 62 and 74, the proportion of patients meeting ACR20 response, ACR/EULAR definition of remission, EULAR response (good, moderate, or no response); the proportion of patients meeting the ACR 50% response (ACR50) and ACR 70% response (ACR70) at week 62, 74, and 98 using the VOLTAIRE-RA baseline data as reference; change from VOLTAIRE-RA baseline in DAS28-C-reactive protein (CRP) and 36-item Short-Form Health survey version 2 (SF-36 v2) at weeks 62, 74, and 98; individual parameters of the ACR improvement criteria at weeks 62, 74, and 98.

Other safety, pharmacokinetic, and immunogenicity endpoints were assessed, including the proportion of patients with infections/serious infections, hypersensitivity reactions, drug-induced liver injury, injection site reactions, protocol-specified adverse events of special interest (AESIs; hepatic injury defined as an elevation of aspartate aminotransferase [AST] and/or alanine aminotransferase [ALT] >threefold upper limit of normal [ULN] combined with an elevation of total bilirubin >twofold ULN measured in the same blood draw sample, and/or ALT and/or AST elevations ≥10-fold ULN; anaphylactic reactions, serious infections [i.e. requiring intravenous (IV) antibiotics or meeting the regulatory definition of a serious adverse event], hypersensitivity reactions), ADA, or nAbs. Safety evaluations also included physical and vital signs, 12-lead electrocardiograms (ECGs), laboratory tests, and continuous adverse event monitoring.

2.4. Data analyses

Not all patients involved in the VOLTAIRE-RA study continued in this open-label extension trial; instead, inclusion was based on eligibility criteria, and patient and physician decision. Therefore, data analyses were performed descriptively according to the drugs patients had received in the original VOLTAIRE-RA study: patients who had received BI 695501 for 48 weeks (Group A), patients who had received adalimumab RP for 24 weeks followed by BI 695501 for 24 weeks (Group B), and patients who had received adalimumab RP for 48 weeks (Group C). Baseline values were derived from the VOLTAIRE-RA study, unless specified differently. No formal hypothesis testing was performed; analysis of the data was performed descriptively. Where confidence intervals or p-values are presented, they are interpreted in an exploratory fashion only.

The primary endpoint was analyzed descriptively using the safety analysis set (SAF). This included all patients who received at least one dose of trial drug during VOLTAIRE-RAext. Statistical analysis focused on treatment-emergent AEs (TEAEs), i.e. all AEs occurring between the start of treatment (baseline in the VOLTAIRE-RAext study) and the end of the residual effect period (i.e. 10 weeks after the last dose of trial drug). AEs were classified by system organ class and term using the MedDRA coding dictionary, version 20.1. In case of missing AE relationship status, the AE was considered to be related to trial drug. No other rules for handling of missing safety data were specified.

The secondary efficacy endpoints were analyzed descriptively using the full analysis set (FAS). This included all patients who received at least one dose of trial drug during VOLTAIRE-RAext and had at least one DAS28 or ACR20 measured during the trial. For the efficacy analyses, last observation carried forward (LOCF) or a combination of LOCF and non-responder imputation was used. Other missing efficacy data were not imputed.

Immunogenicity evaluations have been described previously [19]. For the pharmacokinetic analysis, drug plasma concentration data were based on the pharmacokinetic full analysis set (PKFS), using a population-based approach, and were summarized descriptively. This included all patients who had received at least one dose of trial drug and had at least one post-treatment valid pharmacokinetic concentration measurement.

3. Results

3.1. Patient demographics and baseline characteristics

A total of 479 patients were screened at 94 clinical sites and 430 entered into the VOLTAIRE-RAext trial (Figure 2). The three highest recruiting countries were Poland, Ukraine, and the United States. All 430 patients received at least one dose of study drug and were included in the safety analysis: 225 patients in Group A, 102 in Group B, and 103 in Group C. Almost all patients (96.5%) were exposed to the study drug for ≥24 weeks. One patient in Group B was excluded from the FAS for not having at least one DAS28 or ACR20 assessment. A total of 406 patients were included in the PKFS: 210 patients in Group A, and 98 each in Groups B and C.

Figure 2. Patient disposition. AE: adverse event.

*49 of the 479 patients failed screening and were not assigned to treatment.

Overall, 388 (90.2%) patients completed the trial and 403 (93.7%) completed the safety follow-up visit (Figure 2). A total of 42 (9.8%) patients discontinued the trial prematurely (Group A, 9.8%; Group B, 5.9%; Group C, 13.6%). The most common reasons for trial discontinuation were patient withdrawal (n = 16/42; 38.1%) and TEAEs (n = 15/42; 35.7%). Furthermore, the number of patients discontinuing the trial due to TEAEs was relatively low across all three groups: n = 6 in Group A, n = 2 in Group B, and n = 7 in Group C. Interstitial lung disease, reported in two patients in Group C, was the only TEAE leading to trial drug discontinuation reported in more than one patient. The number of patients discontinuing the trial drug due to TEAE was also low (overall, n = 16 [3.7%]; Group A, 3.1%; Group B, 2.0%; Group C, 6.8%).

Baseline demographics and clinical characteristics were balanced between treatment groups (Table 1).

Table 1. Baseline demographics and clinical characteristics (SAF).

Baseline characteristic	Group A (N = 225)	Group B (N = 102)	Group C (N = 103)
Mean age, years (SD)	53.8 (11.87)	54.6 (9.9)	51.7 (11.21)
Age category, (%)
<65 years	180 (80)	86 (84.3)	92 (89.3)
≥65 years	45 (20.0)	16 (15.7)	11 (10.7)
Female, n (%)	188 (83.6)	84 (82.4)	88 (85.4)
Mean weight, kg (SD)	73.8 (15.9)	79.4 (17.4)	72.8 (15.6)
Mean BMI, kg/m^2 (SD)	27.2 (5.3)*	29.2 (6.5)	27.1 (5.4)*
Duration of RA, mean years (SD)	8.2 (6.7)	8.1 (8.0)^†	8.4 (6.3)
Physicians’ global assessment of disease, mean (SD)	22.2 (16.1)	23.4 (15.9)^†	21.7 (15.3)
Patients’ global assessment of disease activity, mean (SD)	32.7 (21.1)	37.3 (22.6)^†	31.5 (19.8)
DAS28-ESR	3.8 (1.34)	3.8 (1.42)^†	3.8 (1.41)
DAS28-CRP	3.22 (1.21)	3.33 (1.30)^†	3.24 (1.30)

*Height and BMI missing for one patient in Group A and one patient in Group C. ^†One patient was missing all baseline disease data at baseline. BMI: body mass index; CRP: C-reactive protein; DAS: Disease Activity Score; ESR: erythrocyte sedimentation rate; N: number of patients in the analysis set; n: number of patients; SAF: safety analysis set; SD: standard deviation.

3.2. Safety

Mean duration of exposure to trial drug was 325.6 days for patients in Group A, 331.3 days for Group B, and 313.1 days for Group C.

The overall percentage of patients reporting at least one TEAE was 44.9%, with similar rates across the treatment groups (Group A, 51.6%; Group B, 33.3%; Group C, 41.7%; Table 2). Investigator-assessed drug-related TEAEs were reported by 20.2% of patients and, as with overall TEAEs, similar rates were observed across treatment groups: 21.3% in Group A, 17.6% in Group B, and 20.4% in Group C (primary endpoint). Consistent with the known safety profile of adalimumab RP and BI 695501, the most frequently reported Investigator-assessed drug-related TEAEs were infections and infestations (overall, 10.2%; Group A, 12%; Group B, 8.8%; Group C, 7.8%), and general conditions and administration site conditions (overall, 5.8%; Group A, 7.1%; Group B, 3.9%; Group C, 4.9%), with injection site bruising (overall, 2.8%) and injection site erythema (overall, 2.3%) being the most commonly reported AEs by preferred term (Table 3). Details of all TEAEs (irrespective of relationship to study drug) are presented in Supplementary Table S1. As with the drug-related TEAEs, the most common system organ class reported was infections and infestations (overall, 24.4%), with upper respiratory tract infection (overall, 4.0%) being the most reported preferred term. General conditions and administration site conditions were reported by 8.6% of all patients, with injection site bruising and injection site pain reported by 4.2% and 1.6% of all patients, respectively. The majority of all TEAEs were mild (overall, 82/430 patients; 19.1%) or moderate (overall, 98/430 patients; 22.8%) in severity. A low incidence of severe TEAEs was reported during the trial (Group A: n = 7, 3.1%; Group B: n = 1, 1.0%; Group C: n = 5, 4.9%), of which six were considered by the Investigators to be drug related: three in Group A (two pneumonia and one endometrial hyperplasia), one in Group B (Henoch–Schönlein purpura), and two in Group C (one each of staphylococcal sepsis and pulmonary tuberculosis). The incidences of all AEs were similar between treatment groups; numerical differences were considered too low to be clinically relevant.

Table 2. Proportion of patients with TEAEs (SAF).

Patients with TEAEs, n (%)	Group A (N = 225)	Group B (N = 102)	Group C (N = 103)
≥1 AE	116 (51.6)	34 (33.3)	43 (41.7)
≥1 drug-related AE*	48 (21.3)	18 (17.6)	21 (20.4)
≥1 serious AE	14 (6.2)	4 (3.9)	8 (7.8)
≥1 serious drug-related AE	5 (2.2)	1 (1.0)	3 (2.9)
AE leading to trial drug discontinuation	7 (3.1)	2 (2.0)	7 (6.8)
AE leading to death	1 (0.4)	0	0
≥1 AESI^†	7 (3.1)	1 (1.0)	3 (2.9)
≥1 AESI that is Investigator-reported and protocol-specified	6 (2.7)	1 (1.0)	2 (1.9)
≥1 serious infection AESI	5 (2.2)	1 (1.0)	2 (1.9)
≥1 hypersensitivity reaction AESI	1 (0.4)	0	0
≥1 ISR	20 (8.9)	4 (3.9)	6 (5.8)

*Relationship to study drug was reported according to Investigator judgment. ^†AESI: hepatic injury, anaphylactic reactions, serious infections, hypersensitivity reactions. AE: adverse event; AESI: adverse event of special interest; ISR: injection site reaction; N: number of patients in the analysis set; n: number of patients; TEAE: treatment-emergent adverse event; SAF: safety analysis set.

Table 3. Frequency of Investigator-assessed drug-related TEAEs with an incidence of ≥2% in any treatment group (SAF).

n (%)	Group A (N = 225)	Group B (N = 102)	Group C (N = 103)
≥1 related TEAE	48 (21.3)	18 (17.6)	21 (20.4)
Infections and infestations	27 (12.0)	9 (8.8)	8 (7.8)
Nasopharyngitis	5 (2.2)	1 (1.0)	0 (0)
Bronchitis	4 (1.8)	2 (2.0)	0 (0)
Pharyngitis	2 (0.9)	3 (2.9)	1 (1.0)
Metabolism and nutrition disorders	2 (0.9)	0	3 (2.9)
Hypercholesterolemia	1 (0.4)	0	3 (2.9)
General conditions and administration site conditions	16 (7.1)	4 (3.9)	5 (4.9)
Infection site bruising	8 (3.6)	3 (2.9)	1 (1.0)
Infection site erythema	5 (2.2)	2 (2.0)	3 (2.9)
Investigations	7 (3.1)	2 (2.0)	1 (1.0)
Increased AST	3 (1.3)	2 (2.0)	0 (0)

Percentage of patients was calculated relative to the number of patients in the SAF.

AST: aspartate aminotransferase; N: number of patients in the analysis set; n: number of patients; SAF: safety analysis set; TEAE: treatment-emergent adverse event.

Overall, 26 (6.0%) patients reported serious AEs during the trial (Table 2). Consistent with the known safety profile for adalimumab RP and BI 695501, the most common serious AEs were infections and infestations (Group A, 3.6%; Group B, 1.0%; Group C, 1.9%). Of these, pneumonia was reported in four patients (1.8%) in Group A and no patients in Group B or C.

Eleven (2.6%) patients had Investigator-reported AESIs and nine (2.1%) had AESIs that were both Investigator-reported and clinical trial protocol-specified (Table 2). There were no AESIs of drug-induced liver injury or anaphylactic reactions. One AESI of hypersensitivity reaction (preferred term hypersensitivity) was reported, for one (0.4%) patient in Group A. The remaining eight AESIs were serious infections. The incidence of further selected AEs of interest is provided in Supplementary Table S2.

One patient from Group A had adverse events (abdominal hernia, adenocarcinoma of colon, ileus paralytic, cardiopulmonary failure) leading to death; this was deemed unrelated to the study drug.

Changes in serum chemistry, hematology, and urinalysis were infrequent and generally similar between the three treatment groups. Cholesterol increased from normal at baseline to above the ULN for 13.3% to 17.6% of patients across the treatment groups during the trial, as expected according to the current adalimumab RP and BI 695501 labels. There were no clinically meaningful changes or differences between the VOLTAIRE-RAext treatment groups with respect to vital signs, ECG, body weight, or physical examination.

3.3. Efficacy

Overall, similar improvements in efficacy were shown across all three treatment groups in this extension study (Figure 3). Improvements in mean DAS28-ESR scores at week 98 from baseline were similar across the three treatment groups: −3.01 in Group A, −2.98 in Group B, and −2.91 in Group C (Figure 3(a)). The 95% CIs for change did not include zero. Furthermore, the proportion of patients meeting the ACR20 response criteria at week 98 from baseline was comparably high across the groups: 76.9% in Group A, 73.3% in Group B, and 76.7% in Group C. For all three treatment groups, ACR20/50/70 responses were sustained over the 2 years of the VOLTAIRE-RA and VOLTAIRE-RAext studies (Figure 3(b)). Furthermore, response rates observed at the end of the VOLTAIRE-RA study (week 48) were sustained throughout 48 weeks of treatment with BI 695501 during VOLTAIRE-RAext.

Figure 3. Efficacy outcomes (FAS). ACR: American College of Rheumatology; EULAR: European League Against Rheumatism; FAS: full analysis set; LOCF: last observation carried forward; MI: multiple imputation; SF: Short-Form.

(a) Mean DAS28-ESR scores from baseline at weeks 62, 74, and 98. (b) ACR20/50/70 response rates through 98 weeks *Week 50 on treatment corresponds to day 1 (week 0) of VOLTAIRE-RAext (2 weeks after last injection in VOLTAIRE-RA). Data presented are MI, rather than LOCF. (c) ACR/EULAR remission at week 98 *Percentage of patients calculated relative to the total number of patients in the analysis set. (d) EULAR ‘good response’ at week 98 *Percentage of patients calculated relative to the total number of patients in the analysis set. (e) Mean change in SF-36 v2 physical component scores from baseline at weeks 62, 74, and 98.

Similar proportions of patients met the ACR/EULAR definition of remission at week 98 across treatment groups: Group A, 8.4%; Group B, 6.9%; Group C, 9.7% (Figure 3(c)). The proportion of patients with a EULAR ‘good response’ at week 98 was also similar across the three treatment groups: Group A, 37.8%; Group B, 41.6%; Group C, 37.9% (Figure 3(d)). Overall, approximately half of the patients (51.3%) had a ‘moderate’ response.

Similar improvements in the mean DAS28-CRP scores at week 98 from the VOLTAIRE-RA baseline were also shown: Group A, −2.70; Group B, −2.59; Group C, −2.64 (Supplementary Figure S1). The 95% CI for change from baseline from week 98 did not include zero.

Improvements from VOLTAIRE-RA baseline were shown in SF-36 v2 (mental and physical components) at weeks 62, 74, and 98 across all treatment groups (Figure 3(e); physical component scores are shown).

3.4. Pharmacokinetics

Drug plasma concentrations had reached a steady state in the VOLTAIRE-RA study and geometric mean plasma concentrations were maintained in this extension study until week 98, declining after the last drug administration as expected to week 108, which was 10 weeks after the last administration of BI 695501 (Supplementary Figure S2). Mean drug plasma concentrations were comparable across the VOLTAIRE-RA and VOLTAIRE-RAext studies, leading to the conclusion that switching treatment from adalimumab RP to BI 695501 does not negatively impact plasma concentrations.

3.5. Immunogenicity

Similar proportions of patients developed ADAs and nAbs across the three treatment groups, demonstrating similar immunogenicity (Figure 4(a)). Mean titers of ADAs and nAbs were also similar throughout the 2 years across the three treatment groups (Figure 4(b)). As reported for adalimumab RP and similarly observed at week 50 (end of VOLTAIRE-RA) in the group exclusively treated with adalimumab RP [18,19], an impact of ADA development on plasma concentrations was observed (Supplementary Figure S3).

Figure 4. Immunogenic response. ADA: antidrug antibody; FAS: full analysis set; nAbs: neutralizing antibody; SAF: safety analysis set; SFU: safety follow-up.

(a) Development of ADA/nAbs through 98 weeks (FAS) *Week 50 on treatment corresponds to day 1 (week 0) of VOLTAIRE-RAext (2 weeks after last injection in VOLTAIRE-RA). (b) Box and whisker plot of titer within ADA-positive subjects at each ADA time point (SAF).

ADA-negative patients showed an increase in trough plasma concentrations until around week 62 and remained at essentially stable concentration levels thereafter, with limited variability among patients. ADA-positive patients showed decreased plasma concentrations compared with ADA-negative patients and increased variability depending on the titer of the ADA response. ADA-positive patients were grouped into quartiles according to their titer value (Q1, Q2–3, and Q4) at week 48. With increasing quartiles, a more pronounced effect on the PK was observed (Supplementary Figure S3), albeit with increased variability, leading to the conclusion that higher ADA titers reduce exposure. No meaningful differences with regards to this effect was observed among the treatment groups. Nevertheless, it should be considered that fewer patients were included in the ADA titer groups (n = 59, 62, and 64, respectively, for Q1, Q2–3, and Q4) compared with the negative group (n = 243), and therefore conclusions should be drawn with caution. An expected decrease in PK plasma concentration after week 98, i.e. when treatment was ended, was noted for all patient subgroups.

4. Discussion

The safety, efficacy, and immunogenicity results of this phase 3b extension study confirm the clinical equivalence of BI 695501 (administered via prefilled syringe) and adalimumab RP for patients with moderately-to-severely active RA over 2 years.

In this extension study, the primary endpoint was met, as the proportion of patients with Investigator-assessed drug-related treatment-emergent AEs was similar across all three treatment groups (~20%). Furthermore, the type and frequency of AEs observed in this long-term clinical trial were as expected from the current adalimumab RP and BI 695501 labels [10–12,17]. Consistent with the known safety profile for adalimumab RP and BI 695501, infections and infestations accounted for the greatest proportion of drug-related AEs. Furthermore, the majority of AEs were non-serious and of mild or moderate intensity across the treatment arms.

In this extension study, the incidence of AEs was generally similar to those reported in the VOLTAIRE-RA study, both for AEs occurring from day 1 to week 58 in VOLTAIRE-RA, and after re-randomization at week 24 through to completion at week 58 of VOLTAIRE-RA [19]. As for the VOLTAIRE-RA study, the proportion of patients who discontinued treatment due to AEs remained low in the extension study (3.7% in VOLTAIRE-RAext and 4.8% in VOLTAIRE-RA). Overall, the frequency of serious AEs in the extension study was also low (6%) and similar to the rate reported for VOLTAIRE-RA (7%). Numeric differences between treatment groups in the extension study may reflect the exclusion of patients who had experienced Investigator-assessed drug-related serious AEs in the VOLTAIRE-RA study (per the VOLTAIRE-RAext exclusion criteria).

Sustained and improved efficacy with long-term treatment were comparable across all treatment groups, confirming the equivalence of BI 695501 and adalimumab RP in patients with RA and proving the absence of negative effects of switching from RP to BI 695501 with regards to efficacy. Similar numerical improvements in all secondary efficacy endpoints across treatment groups were shown at week 98 (from the VOLTAIRE-RA baseline), including DAS28-ESR, ACR20 response, ACR/EULAR definition of remission, and EULAR ‘good response’ rate. Similar patterns of improvement were also observed in these efficacy endpoints at weeks 62 and 74.

Improvements in the proportion of patients meeting ACR50 and ACR70 response criteria, as well as changes in DAS28-CRP score and SF-36 v2, were observed across all treatment groups in the extension study, compared with VOLTAIRE-RA. Thus, the long-term use of BI 695501 further improves efficacy outcomes after 48 weeks initial treatment, as expected from adalimumab RP studies.

The immunogenicity data confirm the similarity of immunogenic response to BI 695501 and adalimumab RP over the long term. Supporting the findings of the VOLTAIRE-RA study, these immunogenicity data also confirm the absence of any relevant impact when switching from adalimumab RP to treatment with BI 695501. The rate of ADA- and nAbs-positive patients changed only minimally during the treatment phase in all treatment groups, and these observed changes were within the variability limits of the assay. This highly sensitive ADA assay was the same as used previously in the VOLTAIRE-RA study [19].

Titers of ADAs and nAbs throughout the 2 years were comparable across the three treatment groups. From the pharmacokinetic analysis, the drug plasma concentrations were also comparable across the VOLTAIRE-RA and VOLTAIRE-RAext studies for the three treatment groups. These findings also support the findings of the VOLTAIRE-PK study, which showed a comparable impact of ADA on key pharmacokinetic parameters for BI 695501 and adalimumab RP (both US- and EU-licensed) [18].

Limitations of the VOLTAIRE-RAext study are that it is an open-label, non-randomized trial, with eligible patients entering into the trial from the phase 3 VOLTAIRE-RA trial via self-selection. Any inferences in differences between treatment groups must, therefore, be made with caution. However, patient demographics and baseline disease characteristics were generally well balanced across treatment groups and representative of a patient population with moderately-to-severely active RA.

Confidence in the robustness of the biosimilar development program can be drawn from the successful development of numerous anti-TNF-alpha biosimilars. In the European Union, 54 biosimilars were approved by December 2018, of which 14 were anti-TNF-alpha drugs and 8 were biosimilars to adalimumab [20]. Clinical trial data and real-world evidence supporting biosimilars are continuing to increase, and the adoption of these agents in clinical practice has been advocated based on their interchangeability with the reference products [21,22]. Our data contribute to the evidence for anti-TNF biosimilars.

5. Conclusions

Over 2 years of treatment, BI 695501, an adalimumab biosimilar, showed similar efficacy, safety, and immunogenicity to the adalimumab RP, independent of initial treatment assignment in VOLTAIRE-RA. No previously unknown adalimumab side effects were identified and the safety, efficacy, PK, and immunogenicity profile in this trial were consistent with historical BI 695501 and adalimumab RP data. This study confirms the findings of the VOLTAIRE-RA study and further supports the acceptability of switching patients from adalimumab RP to BI 695501.

Author Contributions

All authors conceived the study, collected the data, analyzed the data, prepared the manuscript and agree to be accountable for all aspects of the work.

Declaration of interest

S Cohen has received research grants from Amgen, Boehringer Ingelheim, Coherus, Pfizer, and consulting fees from Amgen, Boehringer Ingelheim, Celltrion, Merck, Pfizer, Sandoz. N Czeloth and G Jayadeva are employees of Boehringer Ingelheim. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer Disclosures

One of the reviewers of this paper has served as a consultant to AbbVie, Amgen, Biogen, and Fresenius. One of the reviewers on this paper has received research Support (paid to university from Pfizer Inc.; and UCB, Inc, and has acted as a consultant to AbbVie Inc.; Boehringer Ingelheim GmbH; Celltrion Healthcare Co. Ltd. Merck Sharp & Dohme Corp.; Pfizer Inc.; Roche Laboratories, Inc.; Samsung Bioepis; Sandoz Inc.; and UCB, Inc.)

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

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

The authors thank Tamara Bailey PhD of GeoMed, an Ashfield company, part of UDG Healthcare plc, for medical writing support, which was funded by Boehringer Ingelheim.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

Boehringer Ingelheim provided the funding and was responsible for the design and conduct of this study.