Lower injection-site reactions and long-term safety, immunogenicity, and efficacy of etanercept biosimilar YLB113: Results from a post-hoc analysis of a double-blind, randomized, phase III comparative study and its open-label extension in patients with rheumatoid arthritis.

A part of the data on this manuscript is based on work that has been previously presented. These study data were submitted as abstracts/posters to the European Alliance of Associations for Rheumatology in June 2021 and were accepted for publication.

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work, and have given their approval for this version to be published.

The disposition of the participants is summarized in Figure  1(B) . Of the 203 participants enrolled in the preceding DB study, 201 were randomized to receive YLB113. Of these, two participants discontinued before the randomization. Overall, 184 participants completed the study, with a completion rate of 91.5%. A total of 17 participants discontinued, with the withdrawal of consent (n = 8) and adverse events (n = 4) being the main reasons. Participants' demographic and baseline characteristics are shown in Table  1 . The mean ± SD age was 51.5 ± 12.2 years, 82.6% were younger than 65 years. The male:female sex ratio was 53:148. The majority of participants had Class II ACR global functional status (68.2%), and 25.4% of participants had a Class I ACR global functional status. In this study, the DAS28 (mean ± SD) at baseline was 2.2 ± 1.0. Demographic and baseline disease characteristics Note : Mean (standard deviation) are presented for the following continuous variables: age, sex, weight, height, BMI, and ACR. The values are the mean ± SD for VAS, CRP, and ESR. Abbreviations: ACR, American College of Rheumatology; CRP, C‐reactive protein, DAS28, Disease activity score 28; ESR, erythrocyte sedimentation rate; OLE, open label extension; RA, rheumatoid arthritis. An overview of the AEs that occurred in the OLE study is shown in Table  2 . A proportion of the participants (93.5%; 188/201) who received at least one dose of YLB113 experienced at least one TEAE. A total of 27 serious AEs were reported by 21 (10.4%) participants throughout the study. The serious AE event of interest reported was pneumonia and endometriosis each 1.0% (two events), all other events observed were 0.5% (one event). A total of seven (3.5%) participants experienced serious adverse drug reactions. All incidences were resolved in a short duration, and none reported mortality. The overall incidence of AEs leading to discontinuation was 2.0% (six events). The most common TEAEs of interest by system organ class were infections, experienced by 70.1% of participants. Of these, the most common TEAE infection event reported was nasopharyngitis, which occurred in 49.8% of participants. Other infections include acute sinusitis, conjunctivitis, bronchitis, cellulitis, cystitis, dermatophytosis of nail, empyema, infectious enteritis, bacterial enterocolitis, gastroenteritis, gingivitis, Helicobacter infection, herpes simplex, herpes zoster, hordeolum, influenza, localized infection, oral herpes, otitis, periodontal localized infection, oral herpes, otitis, peritonitis, pharyngitis, pharyngotonsillitis, pneumonia, postoperative wound infection, and pulpitis dental. The TEAEs with high incidence by preferred term included influenza (14.9%), ISR (10.0%), back pain (6.5%), pharyngitis (6.0%), and RA (6.0%). Most TEAEs and adverse drug reactions observed in this study were mild or moderate. There were no clinically significant abnormal findings or changes in laboratory test results, vital signs, electrocardiogram, or physical examination for any of the participants. Summary of safety in OLE study (N = 201) Note : N = all participants assigned to the population set; n = number of patients with at least one adverse event in the category/preferred term. Abbreviation: OLE, open‐label extension. The incidence of general disorders and administration site conditions (by system organ class) at week 24 (stage A) of the phase III study and at 96 weeks of the OLE study is summarized in Table  3 . At week 24, the overall incidence of general disorders and administration site conditions occurring in at least 1% of the participants was lower in the YLB113 arm (11%) compared with the RP arm (31.1%). No serious ISR events were reported, all incidences were mild (Grade 1) to moderate (Grade 2) and resolved subsequently. In the OLE study, the proportion of participants with general disorders and administration site conditions was markedly lower in both the continued treatment group (YLB113 at 24 weeks, 11% vs. YLB113/YLB113 at 96 weeks, 6%) and switched treatment group (RP at 24 weeks, 31.1% vs RP/YLB113 at 96 weeks, 13.9%). The overall incidence of ISRs was 20.0% at week 96. There were no serious ISR events, and all incidences were mild (Grade 1) with only one participant experiencing a moderate (Grade 2) ISR. In addition, it has been observed that the onset of ISR occurrence for YLB113 from week 52 to 96 weeks continued to be lower and less than 5%. Summary of injection‐site reaction: safety analysis set Abbreviations: DB, double blind; OLE, open‐label extension; RP, reference product; SOC, system organ class. Immunogenicity was measured by the emergence of ADAs (rate of). A total of 201 participants were evaluated for ADAs (neutralizing, N and non‐neutralizing, n) at the time of transition to the OLE study. These results are presented in Table  S1 . In the OLE study, one (0.52) participant developed ADAs at week 24 and one (0.53) participant at week 48, but these were transient and non‐neutralizing, and their status was negative (non‐neutralizing [ADA negative]) at week 96 and the end of the study. Overall, the immunogenicity observed in this OLE study was similar to the previous pivotal DB study. Of note, the responses to ADA/neutralizing antibodies were transient. At study transition (week 0), week 72, and at the end of the study, the DAS28 (mean ± SD) scores were 2.22 ± 0.95, 2.10 ± 0.91, and 2.06 ± 0.89 respectively, indicating a decreasing trend against time after the study transition and continuing until the end of the study. The mean DAS28 score remained low with long‐term administration of YLB113, indicating that the effects of the study drug were maintained throughout the study period. The summary of key demographic characteristics (age, gender, and race) is presented in Table  S2 . A total of 874 participants were screened, of whom 528 were randomized (266 [100%] to the YLB113 arm and 262 [100%] to the RP arm); 263 participants in the YLB113 arm and 254 participants in the RP arm were included in the safety analysis set. The result showed a statistically significant difference in the incidences of ISRs (10 [3.8%] vs 35 [13.8%], P  < 0.0001) and ISE (5 [1.9%] vs 25 [9.8%], P  < 0.0001) between participants who received YLB113 and RP. The risk difference between YLB113 and RP arms for ISR was −9.98% (95% CI WALD –14.81% to −5.15%) and for ISE it was −7.94% (95% CI WALD –11.96% to −3.92%) (Table  4 ). This could be explained by the absence of latex in the syringe needle cap of YLB113. Post‐hoc analysis of phase III study: Differences in injection‐site reactions and injection‐site erythema between YLB113 and reference product Note : N = all participants assigned to the population set; n = number of participants. Abbreviation: CI, confidence interval; RP, reference product.

This open‐label extension study evaluated the long‐term safety, efficacy, and immunogenicity of YLB113 50 mg (once every 2 weeks) in patients with RA who were previously treated with etanercept RP and YLB113 for 52 weeks during a phase III study. In this study, the sustainability of long‐term administration of YLB113 50 mg for up to 96 weeks was assessed. The analysis at week 96 of the OLE study shows that on long‐term administration, the safety profile of YLB113 was maintained with no significant differences when compared with the DB study, confirming the longer‐term safety of YLB113; no unexpected serious adverse reactions occurred, and no new safety concerns were noted during the study. The incidence rates of infection types of AEs reported here are similar to those observed in biologic agents in patients with RA. 13 Nasopharyngitis was one of the most common TEAEs reported in both the DB and OLE studies; however, the common terminology criteria for adverse events grade observed in both the parent and the extension study described here was mild to moderate. Overall, the safety results remained in line with the reported safety profile of the etanercept RP. 14 A good clinical efficacy profile based on DAS28 scores was shown in both the groups in which YLB113 was administered for 3 years and the group in which YLB113 was switched from Enbrel to YLB113 (data not shown). The mean reductions in DAS28 scores showed a decreasing trend. These data demonstrated similar efficacy between the DB study and the OLE study. Published reports suggest that etanercept biosimilars have no impact on safety, efficacy, or immunogenicity after switching. 15 , 16 , 17 The overall result of this study appears to be consistent with previously reported biosimilar studies with etanercept. 15 , 16 , 17 In summary, the incidence of ADAs was low and transient, without neutralizing capacity in the Nepexto group during the parent study and was maintained during the OLE based on pivotal trials of Nepexto. 7 This suggests that antibody production, which was a concern when switching from the RP to a biosimilar, was not observed. Notably, the incidence rates of ADA formation observed in the participants treated with the RP group were high compared with those of the YLB113 group in the parent study and seemingly higher in what has been reported in previous studies. 15 , 18 These results indicate that ADAs may be dependent on product‐specific variables (eg product aggregation or impurities) or other factors. 19 , 20 Of note, the incidence of ISRs reported was markedly lower in both the continued treatment group (YLB113/YLB113) and in the switched treatment group (RP/YLB113) during the open‐label period. This confers some confidence in patients and prescribers to safely switch to YLB113 biosimilar from the RP and/or to continue with YLB113 for long‐term use. Moreover, the post‐hoc analysis of phase III safety data showed significantly lower incidence of ISRs and ISE ( P  < 0.0001) compared with the reference product. This may translate into better acceptability by patients, which can in turn lead to treatment adherence and longer drug persistence. The key strength of this study is its long duration, which provides valuable safety information over 3 years. A potential limitation of this study is that this trial was conducted in Japan and only Japanese participants were included. Despite this limitation, the long‐term safety data assessed parameters that are of clinical relevance and provides information for the long‐term management with YLB113 of RA patients and the incidence of ISRs.

This open‐label single‐arm study of a phase III DB trial confirms that long‐term administration of YLB113 was well‐tolerated and that switching from the reference product to YLB113 did not impact safety, efficacy, or immunogenicity. In addition, a post‐hoc analysis of the YLB113‐002 pivotal DB study showed a statistically significantly lower incidence of ISRs and ISE with the administration of YLB113 versus the reference product. In summary, this evidence demonstrates that YLB113 maintained a favorable safety, efficacy, and immunogenicity profile throughout 3 years. The availability of Nepexto 7 can positively impact the lives of patients by offering a safe and effective biosimilar.

Rheumatoid arthritis (RA) is a persistent and progressive immune‐mediated inflammatory disease that results in joint damage of varying severity among individuals. 1 Its management requires long‐term treatment with a significant cost burden. Although the conventional synthetic disease‐modifying anti‐rheumatic drugs (DMARDs) and biologic DMARDs delay or halt the progression of joint destruction and deformity, they come with an increased financial burden. 1 A biosimilar drug is a biologic agent that is highly similar to a licensed biologic product (reference product [RP]) with no clinically meaningful differences in terms of safety, purity, or potency and is intended to be used in the same way as the RP. 2 , 3 , 4 Several biosimilar DMARDs have been approved for use in rheumatic diseases. The European Alliance of Associations for Rheumatology (EULAR) recommendations reinforce in their overarching principles that the availability of biosimilar DMARDs provides significant reductions in healthcare budgets. 5 Therefore, biosimilars can be an alternative treatment option to an approved biologic for increased patient access and lower cost in treating rheumatic diseases. 6 YLB113 (Nepexto®) is a recombinant fusion protein tumor necrosis factor inhibitor produced by recombinant DNA technology in Chinese Hamster ovary cells. YLB113 was developed as a biosimilar to the RP, Enbrel® (etanercept RP, manufactured and marketed by Pfizer, Inc., Basel, Switzerland) by a joint venture made between Lupin Limited‐Biotechnology Division, (Pune, India) and YL Biologics Co., Ltd (Tokyo, Japan). Lupin Biotech was involved in CMC (chemistry, manufacturing, and control), preclinical, toxicity, and phase I studies supporting its biosimilarity. YL Biologics conducted the phase I study in Japan and the global phase III study was conducted by both companies. This open‐label extension (OLE) study was conducted by YL Biologics as a follow up to the Phase III study in Japanese patients who completed the Phase III study and provided consent to continue the clinical investigation. YLB113 (Nepexto) was approved as an etanercept biosimilar by the European Medicines Agency in March 2020 after the Committee for Medicinal Products for Human Use provided a positive opinion for all the indications of its reference product. 7 It was also approved in Japan in March 2019 by Japan's medicines regulatory agency, the Pharmaceuticals and Medical Devices Agency, for the treatment of moderate‐to‐severe RA and juvenile idiopathic arthritis. 8 The physicochemical and analytical comparability of YLB113 to etanercept RP was demonstrated sufficiently through physicochemical and biological state‐of‐the‐art characterization methods. 7 The phase I pharmacokinetics study demonstrated the similarity of YLB113 25 mg to RP etanercept 25 mg with comparable safety in healthy volunteers. 9 The pharmacokinetics study was followed by a phase III study (YLB113‐002) that confirmed the comparable clinical efficacy and safety of YLB113 to etanercept RP in patients with moderate‐to‐severe RA. 10 As the management of RA is lifelong, it is important to assess the effects of anti‐rheumatic drugs under circumstances closer to the real‐life treatment. To this end, a 96 week OLE of a double‐blind (DB) study evaluated the long‐term safety of YLB113 in patients with active RA. The safety and tolerability outcomes included adverse events (AEs), incidence of immunogenicity, and long‐term efficacy as improvement in Disease Activity Score in 28 joints (DAS28). In addition, a post‐hoc analysis of a DB, randomized, phase III study was conducted to determine differences in injection‐site reactions (ISRs) and injection‐site erythema (ISE) between YLB113 and RP.

Hisashi Yamanaka has received speaker or consultant fees from Astellas, Bristol‐Meyers‐Squibb, CorEvitas LLC, Eisai, Pfizer, Tanabe‐Mitsubishi, Teijin Pharma, and YLBio. Yoshiya Tanaka is an Editorial Board member of the journal and co‐author of this article and was excluded from the peer‐review process and all editorial decisions related to the acceptance and publication of this article. Peer‐review was handled independently by members of the Editorial Board to minimize bias. Yoshiya Tanaka has received speaking fees and/or honoraria from Daiichi‐Sankyo, Astellas, Chugai, Eli Lilly, Pfizer, AbbVie, YL Biologics, Bristol‐Myers, Takeda, Mitsubishi‐Tanabe, Novartis, Eisai, Janssen, and Teijin, and has received research grants from Asahi Kasei, Mitsubishi‐Tanabe, Chugai, Takeda, Sanofi, Bristol‐Myers, UCB, Daiichi‐Sankyo, Eisai, and Ono. Toshihiko Hibino is an employee of YL Biologics Limited. Chirag Shah is full‐time employee of Lupin Pharmaceuticals Ltd. Dhananjay Bakhle is a full‐time employee and shareholder of Lupin Pharmaceuticals Ltd. Dimitris Stefanidis and Unmesh G are full‐time employees of Viatris, Inc.

The design of the pivotal phase III YLB113‐002 study has been described in detail previously. 10 Briefly, the parent study was a multicenter, DB, randomized, parallel‐group comparative study with participants receiving a once‐weekly, subcutaneous dose of 50 mg YLB113 or the etanercept RP, alongside stable maintenance therapy with methotrexate. Following the parent study, this open‐label, phase III, multicenter, single‐arm, extension study (YLB113‐003) was conducted between August 2016 and May 2019 to assess the sustainability of long‐term administration of YLB113. The study was conducted on Japanese patients with RA at 48 sites in Japan. Eligible participants were given the option to enter an extension study (within 4 weeks of completion of the randomized DB study). Participants were given 50 mg of YLB113, subcutaneously in a prefilled syringe once a week and the administration interval could be once every 2 weeks according to the judgment of the principal/sub‐investigator through 96 weeks (Figure  1A ). A biweekly or every other week dosage regimen of etanercept is the recommended dose in the treatment of RA, which is described in other etanercept studies like Bertrand et al. 11 All participants provided written informed consent to participate in the extension study. The study was conducted in accordance with the approved protocol, and principles of the Declaration of Helsinki, Council for International Organizations of Medical Sciences International Ethical Guidelines, applicable International Council for Harmonization Good Clinical Practice Guidelines, and applicable laws and regulations at the study site. A, Study design. B, Participant flow The inclusion criteria of the parent study have been described previously. 10 To be eligible for this present study, participants who had completed the main 52 week DB study and met the following criteria were included: (a) participants who did not experience a severe AE related to the study drug; (b) participants in the parent study who were confirmed as tolerating long‐term administration well in week 24 or week 44; (c) participants with DAS28 scores that were decreased by 0.6 or more from baseline. Key exclusion criteria were: (a) having experienced emergency treatment due to the exacerbation of RA symptoms; (b) having experienced AEs (including progressing complications), or safety reasons. Participants were required to receive concomitant treatment with methotrexate in combination with the study drug at doses of 6‐16 mg/week. In the parent study, participants received stable doses of methotrexate with minimal changes in dosage and administration, considering the possibility of affecting the evaluation of safety and efficacy. While in this OLE study, participants received methotrexate as needed, which is relatively similar to the real‐world treatment setting of RA. Safety was assessed by the incidence of treatment‐emergent adverse events (TEAEs) and severe AEs throughout the parent and extension studies. Other safety variables assessed included monitoring of ISRs, vital signs, electrocardiogram, physical examination findings, and immunogenicity. All AE records contained AE term/AE diagnosis, date of onset, date of recovery or outcome, severity, seriousness of relation to the study drug, action taken with the study drug, outcome, action taken with AE, and AE leading to discontinuation of the patient from the study. The incidences of TEAEs and adverse drug reactions were summarized by system organ class and preferred term and by common terminology criteria for adverse events grade. As a confirmation of long‐term efficacy, the DAS28 score was assessed at the time of transition (week 0), and at weeks 12, 24, 48, 72, and 96, and at the end of the study. This score was calculated using a composite score (range 0‐9.4) from the results of the 28 Swollen Joint Count/tender Joint subset of the 66/68 SJC/TJC, C‐reactive protein levels (mg/L), or erythrocyte sedimentation rate levels (mm/h), and the participant's global assessment of disease activity (0‐100 scale). 12 No American College of Rheumatology 20/50/70 criteria (ACR 20/50/70) were assessed in this OLE study. Although ACR20 was used as the main end point in the previous study, DAS28 was used as the end point of efficacy in this study because DAS28 assesses fewer joints and is as such more commonly used in general practice. Immunogenicity, measured by the presence or absence of anti‐drug antibodies (ADAs) and the antibody titer, was assessed at weeks 24, 48, 72, and 96 of the extension study. In this study, the measurement of anti‐etanercept and neutralizing antibodies was performed in the same manner as in the parent study. 10 First, a screening test for the presence of an anti‐etanercept antibody was performed. A confirmatory test was performed to see if the screening test was a false positive. Antibody titer was measured if a positive result was obtained in the confirmatory test. Neutralizing antibodies were measured if the results were positive for the presence of anti‐etanercept antibodies. The assessment of anti‐etanercept and neutralizing antibodies was performed using a validated and highly sensitive meso‐scale discovery electrochemiluminescence method with acid dissociation and affinity capture as sample pretreatment techniques to enhance drug tolerance. The most common side effect of etanercept therapy is ISR. The objective of a post‐hoc analysis of this safety data was to evaluate any differences in the incidence of ISRs and ISE during the first 24 weeks of DB treatment. Participants of this DB study who were on a stable methotrexate concomitant dose were treated with YLB113 or RP (50 mg) once‐weekly subcutaneously. The safety analysis set was considered for this analysis. Local reactions at the site of injection were assessed at each of the eight study visits. The proportion of participants who experienced ISRs and ISE was compared for YLB113 and etanercept RP. The risk difference and 95% confidence interval (CI) were calculated between arms to understand the magnitude of the difference in the incidence of events. The statistical significance between group difference was tested using the χ 2 test. The sample size was determined based on the assumption that Japanese participants who had participated in the previous study would transition to this single arm OLE study. In this study, efficacy, safety, and immunogenicity were analyzed using the full analysis set (population of participants who received at least one study drug during the study). Statistical data analysis was performed using SAS® Version 9.3 or a higher version (SAS Institute Inc., Cary, NC, USA). For continuous data, summary statistics included the number, arithmetic mean, standard deviation (SD), median, minimum, and maximum. For categorical data, frequency and percentage were presented. All statistical tests were two‐sided and evaluated at a 5% level of significance. A DAS28 count was calculated as confirmation of efficacy (tolerability). Hence detailed analysis of DAS28 was not performed. If data necessary for assessment were missing, DAS28 was calculated by complementation of the average of data before and after. Efficacy assessment in dropout patients was conducted before dropout and was not involved after dropout, because the major evaluation in this study was safety and immunogenicity, including switchability of YLB113 from Enbrel, and efficacy was tried as a supplemental assessment. Safety results were presented using descriptive statistics by visit/treatment. Safety outcomes included the incidence of all TEAEs occurring after study drug administration. The anti‐etanercept antibody test was summarized at each time point.

Tables S1‐S2 Click here for additional data file.