Safety and effectiveness of peficitinib (ASP015K) in patients with rheumatoid arthritis: interim data (22.7 months mean peficitinib treatment) from a long-term, open-label extension study in Japan, Korea, and Taiwan

Background Peficitinib (ASP015K), a novel oral Janus kinase inhibitor, has demonstrated efficacy and safety for the treatment of rheumatoid arthritis (RA) in randomized, controlled trials of up to 52 weeks’ duration. However, safety and effectiveness after long-term treatment have not been assessed. Methods This was an interim analysis of an ongoing open-label, multicenter extension study in RA patients who completed phase 2b (RAJ1; 12 weeks) and phase 3 (RAJ3 and RAJ4; 52 weeks) peficitinib studies in Asia (mainly Japan). Eligible patients (n = 843) received oral peficitinib once daily (100 mg, or 50 mg for patients transferring from RAJ1). The peficitinib dose could be increased (up to 150 mg) or reduced (to 50 mg) at the discretion of the investigator. Efficacy variables assessed included American College of Rheumatology (ACR) response rates, ACR components, and disease activity score in 28 joints based on C-reactive protein (DAS28-CRP). Results Results up to May 2018 are summarized. Mean peficitinib duration of exposure was 22.7 months and the maximum dose was 100 mg in most (66.5%) patients. ACR responses were maintained during the extension study, with ACR20/50/70 response rates of 71.6%, 52.1%, and 34.7% at week 0 and 78.9%, 61.4%, and 42.7% at end of treatment, respectively. ACR components and DAS28-CRP showed improvements from baselines of the preceding studies and continued to show improvements during the extension study. Treatment-emergent adverse events (TEAEs) were reported in 757/843 (89.8%) patients, the most common being nasopharyngitis (39.7%) and herpes zoster (11.7%). The majority of TEAEs were severity grade 1/2. Drug-related TEAEs leading to permanent study drug discontinuation occurred in 55/843 (6.5%) patients. Regarding AEs of special interest, the incidence per 100 patient-years of serious infections was 2.3 (95% CI 1.6 – 3.1), herpes zoster-related disease 6.8 (95% CI, 5.6 – 8.3), and malignancies 1.1 (95% CI, 0.7 – 1.8). One death from diffuse large B cell lymphoma during the study and one death from uterine sarcoma after the study were considered probably and possibly related to study drug, respectively. Conclusions The effectiveness of peficitinib was maintained or improved during long-term administration and treatment up to 6 years was well tolerated in Asian patients with RA. Trial registration ClinicalTrials.gov, NCT01638013, registered retrospectively 11 July 2012.


Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by persistent joint inflammation. Without adequate treatment, it can progress to joint deformity and functional impairment [1]. Current treatment of RA is based on disease-modifying antirheumatic drugs (DMARDs), usually starting with methotrexate (MTX), and biologic agents such as tumor necrosis factor-alpha (TNF-α) inhibitors in patients not responding to conventional DMARDs [2,3]. The Janus kinase (JAK) family of non-receptor protein tyrosine kinases (JAK1, JAK2, JAK3, and tyrosine kinase 2 [TYK2]) plays a crucial role in multiple cytokine receptor signaling pathways [4] and is a promising target for RA treatment in patients with insufficient response to DMARDs and biologic agents. JAK inhibitors available for the treatment of patients with RA include tofacitinib (a pan-JAK inhibitor approved in the USA, Europe, and Asia [5][6][7][8]), baricitinib (a JAK1 and JAK2 selective inhibitor approved in the USA, Europe and Asia [9][10][11][12]), and more recently peficitinib (ASP015K; approved in Japan).
Peficitinib is an orally bioavailable, once-daily JAK inhibitor, developed for the treatment of patients with RA (including the prevention of structural injury of joints) who have had an inadequate response to DMARDs, including MTX. Peficitinib is a pan-JAK inhibitor that inhibits JAK1, JAK2, JAK3, and TYK2 [13]. Compared with other JAK inhibitors, peficitinib is moderately selective for JAK3 (over JAK1, JAK2, and TYK2) [4,14]. Less potent inhibition of JAK2 may explain why reduced hemoglobin levels, potentially attributable to JAK2 inhibition, have not been observed with peficitinib [15,16].
In a phase 2b, randomized, double-blind, placebocontrolled trial (RAJ1; ClinicalTrials.gov identifier, NCT01649999), peficitinib monotherapy showed efficacy and an acceptable safety profile in Japanese patients with moderate-to-severe RA after 12 weeks [15]. In two phase 3, randomized, double-blind, placebo-controlled trials, peficitinib treatment for 52 weeks showed significantly improved efficacy compared with placebo and an acceptable safety profile in patients with RA who had an inadequate response to DMARDs (RAJ3 study in Japan, Korea, and Taiwan; ClinicalTrials.gov identifier, NCT02308163 [17]) or MTX (RAJ4 study in Japan; ClinicalTrials.gov identifier, NCT02305849 [18]).
As peficitinib is a relatively novel compound, it is important to assess safety and effectiveness after long-term treatment. We present safety and efficacy data from a long-term, open-label extension study of patients who completed the phase 2b (RAJ1) or phase 3 (RAJ3 or RAJ4) peficitinib clinical trials.

Study design
This was an open-label, long-term extension study ("RAJ2") conducted at 165 sites in Japan, nine sites in Korea, and nine sites in Taiwan (see Additional file 1: Study sites). The extension study commenced in June 2012 and, after marketing approval of peficitinib in Japan in March 2019, continues as a post-marketing clinical study in Japan, and a clinical study in Korea and Taiwan. Safety and efficacy data are presented at the cut-off date of 31 May 2018. The duration of study treatment, therefore, varied between patients.
Enrolled patients with RA had completed one of three peficitinib clinical trials ( Fig. 1): (1) "RAJ1," a phase 2b trial in which patients received peficitinib monotherapy or placebo for 12 weeks with a 4-week follow up (without peficitinib treatment) [15]; (2) "RAJ3," a phase 3 trial in which patients with an inadequate response to DMARDs (DMARD-IR) received peficitinib for 52 weeks (patients in the etanercept reference group of RAJ3 were not included in the extension study) [17]; and (3) "RAJ4," a phase 3 trial in which patients with an inadequate response to MTX (MTX-IR) received peficitinib for 52 weeks [18].
Patients eligible for the extension study received oral peficitinib (50 mg, 100 mg, or 150 mg) once daily after breakfast. The starting dose was 100 mg (patients from RAJ3 and RAJ4) or 50 mg (patients from RAJ1). Following a protocol amendment in June 2014, all patients from RAJ1 were permitted to increase their peficitinib dose to 100 mg, unless deemed by the investigator to be inappropriate from a safety perspective. The dose could be increased following a lack of clinical response (DAS28-ESR ≥ 3.2 after 4 weeks of peficitinib treatment) and no potential safety concerns, as pre-specified in the protocol; the dose could also be reduced in line with safety assessments, as judged by the investigator. Treatment suspension (temporary discontinuation < 7 days), interruption (temporary discontinuation > 7 days and < 4 consecutive weeks, up to two per year separated by ≥ 16 weeks), or discontinuation during the extension study were at the discretion of the investigator and based on pre-specified criteria. In cases of suspension or interruption, peficitinib administration could be resumed if a retest confirmed that the suspension/interruption criteria were no longer met and the investigator deemed resumption of treatment to be in the patient's best interests. Patients who discontinued early from the study were given the necessary tests (scheduled for the end of study) within 2 days after taking the final dose of the study drug, where possible. The reason for the discontinuation was recorded.

Patients
Eligible patients with RA had completed study treatment in the preceding trials, as specified in the protocol, and completed assessments at week 16 (RAJ1 study) or week 52 (RAJ3 and RAJ4 studies). The baseline of this extension study (week 0) coincided with RAJ1 week 16 or RAJ3 and RAJ4 week 52.

Concomitant medications
The following concomitant medications (with dose adjustments) were permitted: NSAIDs, oral morphine (≤ 30 mg/day or equivalent of other opioid analgesics), and acetaminophen. If concurrent MTX was administered, the concomitant use of folic acid (maximum of 10 mg/week) was considered, where possible.
The daily dose (prednisolone equivalent) of any oral corticosteroid was prohibited from exceeding the amount used at the initiation of studies RAJ1, RAJ3, or RAJ4. Outcomes

Efficacy endpoints
Efficacy endpoints were assessed in the overall population and in patients grouped according to their preceding study (RAJ1, RAJ3, or RAJ4): American College of Rheumatology (ACR)20/50/70 response rates, changes from the baselines of the preceding studies in ACR components, tender joint count at 68 joints (TJC68), swollen joint count at 66 joints (SJC66), Physician's Global Assessment of Disease Activity (PGA), Subject's Global Assessment of Disease Activity (SGA), Subject's Global Assessment of Pain (SGAP), Health Assessment Questionnaire -Disability Index (HAQ-DI), change from the baselines of the preceding studies in 28-joint Disease Activity Score based on C-reactive protein (DAS28-CRP), proportion of patients achieving DAS28-CRP score < 2.6 (disease remission), and ACR20 response rate according to maximum peficitinib dose (50 mg, 100 mg, 150 mg).

Safety
The following safety assessments were reported during the overall period (from start of initial dosing in preceding studies): treatment-emergent adverse events (TEAEs), defined as any adverse event (AE) that started or worsened in severity after initial dose of study drug in the extension study until the end of the final observation, and TEAEs or serious AEs (SAEs) leading to permanent discontinuation. AEs of special interest-serious infections, herpes zoster-related disease, malignancies, and thromboembolic events (ad hoc analysis: first deep vein thrombosis [DVT] and pulmonary embolism [PE] identified using the MedDRA embolic and thrombotic SMQ preferred terms restricted to the respiratory, thoracic, mediastinal, and vascular disorder System Organ Classes)-were reported as incidence per 100 patientyears (95% confidence interval [CI]). Mean changes in laboratory measures, including hematology, lipoprotein, creatinine, creatine kinase, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were measured.

Patient populations
The safety analysis set (SAF) included all patients who received at least one dose of study treatment. The full analysis set (FAS), used for efficacy analyses, included all patients who received at least one dose of study treatment and had measurements for any of the efficacy endpoints.

Statistical analysis
Efficacy results were summarized by sample size, mean, standard deviation, minimum, median and maximum by time point for continuous variables, and frequency and percentage for categorical variables.
In relation to TEAEs, any unrecovered event that first occurred during the preceding studies (RAJ1, RAJ3, and RAJ4) was considered to be a medical condition and was not considered to be a TEAE in this extension study.

Missing data
The last observation carried forward (LOCF) method was used for ACR components, DAS28-CRP, and safety variables at the end of treatment (EOT). All outliers were included in the analysis.

Ethics
This study was conducted in accordance with Good Clinical Practice, the International Council on Harmonization (ICH) of Technical Requirements for Registration of Pharmaceuticals for Human Use guidelines, and local laws and regulations. The protocol and amendments were approved by an Institutional Review Board (IRB) at each study site, and safety data were reviewed by the independent Data and Safety Monitoring Board. Each patient provided written informed consent prior to treatment initiation.

Patient populations
In total, 873 patients were screened for inclusion in the extension study. Of these, 843 patients received peficitinib treatment (Fig. 2): 201 patients from RAJ1, 225 from RAJ3, and 417 patients from RAJ4 (Additional file 1: Fig. S1). A total of 234 (27.8%) patients discontinued study treatment, mainly due to AEs (8.5%), other (6.3%), and lack of efficacy (5.9%) (Fig. 2). At the cut-off date for the present analysis (31 May 2018), 609 (72.2%) patients were still receiving study treatment (Fig. 2). All 843 patients who received treatment were included in the SAF, and 837 patients were included in the FAS.  Duration of exposure for overall period (days) was calculated as: date of the last dose of study drugdate of initial dose of study drug + 1 2 Duration from first peficitinib taken (50 mg for patients from RAJ1, 100 mg for patients from RAJ3 and RAJ4) up to first dose change was calculated 3 Treatment compliance for overall period (%) was calculated as: 100 × (total number of tablets actually received in the overall period/total number of tablets planned to receive in the overall period) SAF safety analysis set, SD standard deviation Demographics and characteristics at the start of the extension study Most patients were female (619/843, 73.4%) and Japanese (806, 95.6%), with a mean age of 55.7 years ( Table 1). The mean duration of RA was 6.2 years (SD 5.6). Concomitant MTX treatment was used by 65.1% of patients (Table 1).
There were some differences in characteristics at baseline of the extension study according to the preceding study (Additional file 1: Table S1): all patients from RAJ1 and RAJ4 were Japanese, whereas the RAJ3 study included patients from Japan (83.6%), Korea (8.4%), and Taiwan (8.0%); the mean duration of RA was higher in patients from RAJ1 (7.3 years, SD 6.1) and RAJ3 (8.7 years, SD 7.2) than RAJ4 (4.3 years, SD 3.0); concomitant MTX treatment was used by 0%, 60.4%, and 99.0% of patients in the RAJ1, RAJ3, and RAJ4 studies, respectively.
During the overall period, the peficitinib dose was increased from the initial administered dose, or after a dose reduction, in 42.0% of all patients (Table 2). A higher proportion of patients from RAJ1 (80.6%) than RAJ3 (30.7%) and RAJ4 (29.5%) had an increase in peficitinib dose. This is because the initial administered dose for patients from RAJ1 entering RAJ2 was 50 mg but, following a protocol amendment in June 2014, all patients from RAJ1 were permitted to have a dose increase to 100 mg, unless safety findings deemed it inappropriate (Additional file 1: Table S2). The maximum peficitinib dose was 100 mg for most patients: 66.5% of patients overall, and 71.1% and 71.2% of patients from RAJ3 and RAJ4, respectively (Table 2; Additional file 1: Table S2). In RAJ1, the maximum dose was 50 mg in 19.4% of patients, 100 mg in 51.7% of patients, and 150 mg in 28.9% of patients (Additional file 1: Table S2). Mean treatment compliance was 97.0% overall (range, 96.3-98.8% by preceding study) ( Table 2; Additional file 1: Table S2).

ACR response
In the overall population, the ACR20, ACR50, and ACR70 response rates were 71.6%, 52.1%, and 34.7%, respectively, at the baseline of this extension study (week 0). These rates were maintained during the extension study, and at the end of treatment were 78.9%, 61.4%, and 42.7%, respectively (Fig. 3). When patients were grouped according to their preceding study, the response rates were maintained in those patients from RAJ3 and RAJ4. In patients from the RAJ1 study, the response rates improved for the first 4 weeks and were then maintained during the extension study (Additional file 1: Fig. S2).
ACR components (TJC68, SJC66, SGAP, SGA, PGA, HAQ-DI) improved from the baselines of the preceding studies and continued to improve during the extension study (Additional file 1: Fig. S3).

ACR20
The ACR20 response rate was maintained during the extension study in patients with a maximum peficitinib dose of 100 or 150 mg. In patients with a maximum dose of 50 mg, the ACR20 response rate improved at the beginning of the extension study and was then maintained (Fig. 4).

DAS28-CRP
The improvement in DAS28-CRP was maintained during the extension study (Fig. 5a): this was seen overall (Fig. 5a) and in patients from RAJ3 and RAJ4 (Additional file 1: Fig. S4a). In patients from the RAJ1 study, DAS28-CRP improved early in the extension study and was then maintained (Additional file 1: Fig.  S4a). The proportion of patients achieving DAS28-CRP remission (score of < 2.6) was maintained during the extension study: this was seen overall (Fig. 5b) and in patients from RAJ3 and RAJ4 (Additional file 1: Fig.  S4b). In patients from RAJ1, DAS28-CRP improved early in the extension study and was then maintained (Additional file 1: Fig. S4b).
There was one death (0.1%) during the study, due to diffuse large B cell lymphoma, which was considered probably related to study drug, and there was one death after the end of the study, due to uterine sarcoma, which was considered possibly related to study drug (see Additional file 1: Case histories).
Regarding AEs of special interest, the incidence per 100 patient-years was 2.3 (95% CI, 1.6-3.1) for serious infections, 6.8 (95% CI, 5.6-8.3) for herpes zoster-related disease, and 1.1 (95% CI, 0.7-1.8) for malignancies. Figure 6 summarizes the incidence of these events per  100 patient-years during the overall period. Peficitinib exposure was 1585.8, 1484.4, and 1612.0 patient-years for serious infections, herpes zoster, and malignancy (overall period in the total peficitinib population). In an ad hoc analysis of thromboembolic AEs, total peficitinib exposure was 1615.2 patient-years and incidence per 100 patient-years was 0.1 (95% CI, 0.0 -0.5). There was one case of deep vein thrombosis (0.1%) and one case of pulmonary artery thrombosis (0.1%). In both cases, the study physician regarded the event as not related to peficitinib.

Laboratory measures
An increase from baseline (week 0) in creatine kinase and a decrease in lymphocytes was observed during the extension study (Table 5). There were no notable changes from baseline (week 0) in other laboratory parameters, including neutrophils, hemoglobin, platelets, low-density lipoprotein, high-density lipoprotein, creatinine, ALT, or AST (Table 5).

Discussion
Peficitinib, a novel oral JAK inhibitor, previously demonstrated clinical safety and efficacy in patients with RA in a 12-week phase 2b (RAJ1) trial and two 52-week phase 3 (RAJ3 and RAJ4) trials. In this open-label extension study of patients with RA who completed these three trials, we found that peficitinib treatment was associated with ACR20/50/70 responses that were maintained during long-term treatment. In patients with a maximum peficitinib dose of 100 mg or 150 mg, the ACR20 response rate was maintained and in patients with a maximum peficitinib dose of 50 mg, the ACR20 response rate improved and was then maintained during the extension study. Of note, patients from the RAJ1 study had a lower ACR20 response rate at baseline (week 0) of the extension study than patients from RAJ3 and RAJ4 (Additional file 1: Fig. S2). This was not surprising as some patients enrolled in RAJ1 had received a low dose (25 or 50 mg) of peficitinib, or no peficitinib in the case of the placebo arm, in addition to others who had received peficitinib 100 mg or 150 mg; furthermore, the Fig. 6 Adverse events of special interest per 100 patient-years during the overall period: a serious infections, b herpes zosterrelated disease, c malignancies (SAF). Patient-years was calculated from initial dose up to first incidence of the event for patients who had at least one event, and from initial dose through follow up for patients who had no events; IR was calculated as (100 × number of patients with ≥ 1 incidence/total patient-years) CI confidence interval, IR incidence rate Table 5 Mean changes from baseline (week 0) in laboratory measurements (SAF) Mean (SD) at week 0 Mean (SD) change from baseline Week 24 Week 48 Week 72 Week 96 Week 120 Week 144 Week 192 Week 240 Week 288 EOT Absolute neutrophil count,  EOT assessments and tests were to be performed promptly after the end of peficitinib administration; if peficitinib administration was terminated early, these assessments and tests were to be performed within 2 days of the last dose of peficitinib, if possible. As the timing of rollover from the previous study, and consequently the completion of the present trial, varied between individual patients, laboratory data at EOT were averaged regardless of the exact date of the EOT visit.
ALT alanine aminotransferase, AST aspartate aminotransferase, EOT end of treatment, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, SAF safety analysis set, SD standard deviation, Wk week treatment period was 12 weeks, followed by a 4-week follow-up period (without peficitinib treatment) before enrollment in the extension study. In contrast, patients from RAJ3 and RAJ4 had all received at least 24 weeks' treatment with peficitinib 100 mg or 150 mg, after which they were immediately enrolled in the extension study.
Further evidence of favorable clinical responses was provided by measurements of ACR components and DAS28-CRP. ACR components (TJC68, SJC66, SGAP, SGA, PGA, HAQ-DI) and DAS28-CRP improved from the baselines of the preceding studies and continued to improve during the extension study.
The safety analysis indicated that peficitinib treatment was well tolerated over a period of up to 6 years. Rates of AEs of special interest (serious infections, herpes zoster-related disease and malignancies) were consistent with studies of other JAK inhibitors. For tofacitinib and baricitinib, respectively, the reported rates of serious infections per 100 patient-years were 2.7 (including any event requiring hospitalization or parenteral antimicrobial therapy, or otherwise meeting SAE criteria) and 2.9 (including any event meeting ICH E2A criteria), respectively [19,20]. Rates of herpes zoster-related disease for tofacitinib and baricitinib were 9.2 (in patients from Japan and Korea [21]) and 6.5 (in patients from Japan [22]), and 0.9 and 0.8 for malignancies, respectively [19,20]. It should be noted, however, that drug exposure was higher in the tofacitinib (19,406 patient-years [20]) and baricitinib (6637 patient-years [19]) studies. Further follow up is required to assess the significance of the observed increases in creatine kinase and decreases in lymphocytes.
A concern has been raised regarding the use of high doses of tofacitinib to treat patients with RA, due to a safety signal for pulmonary embolism and increased mortality that has emerged during a post-marketing trial [23]. As of the cut-off date, two thromboembolic AEs were reported during the peficitinib extension study, one pulmonary artery thrombosis (0.1%) and one deep vein thrombosis (0.1%), but these AEs were considered by the study physician to be unrelated to peficitinib.
In the preceding phase 3 clinical trials, peficitinib (100 mg/day and 150 mg/day) provided superior efficacy to placebo, with clinical improvements observed by week 12 in patients with DMARD-IR (RAJ3) or MTX-IR (RAJ4). Overall, the interim analysis of this extension study also suggests that peficitinib may be a welltolerated, effective treatment option for patients with RA in the longer term.
The long treatment period (mean duration of peficitinib exposure 22.7 months; maximum: 70.7 months) and the large number of treated patients (n = 843) are notable strengths of this study. This complements the longterm safety and efficacy data already available for other JAK inhibitors, including up to 8.5 years with tofacitinib [20,24] and up to 2.5 years with baricitinib [25]. In addition, the inclusion of patients with varying treatment regimens (such as DMARD combinations, including MTX, as well as monotherapy) and the flexible study design, allowing individualized treatment of RA patients, is likely to be representative of routine medical practice. For example, the study allowed a dose increase in cases of DAS28-CRP > 3.2 and a dose reduction to 50 mg according to a patient's condition or safety issues, which is similar to the way in which MTX is currently administered.
However, there are also a number of limitations associated with this extension study: patients' experience of study treatment in the preceding studies may have influenced their decision to participate; the sample size was entirely dependent on the number of patients who completed the preceding studies; and there was no comparator group (placebo or active) during the extension. Furthermore, comparisons between peficitinib doses were limited because (1) patients were not randomized to different dose groups; (2) dose changes could be made (at the discretion of the investigator); and (3) patient disease states may have changed during the extension study. The patient population was drawn mainly from Japan (95.6%) and the data therefore lack global diversity. The Japanese population has a unique genetic, environmental, and medical background, which is known to influence the effectiveness and safety of biologic agents for RA [26]. This limits the generalizability of findings from this study to other populations.
At the cut-off date, 609 (72.2%) patients were still receiving peficitinib treatment and no new safety signals had been identified. Another phase 3 clinical trial is being conducted at multiple sites in China, Taiwan, and Korea to assess the safety and efficacy of 52 weeks' peficitinib treatment in patients with RA who had an inadequate response to MTX (ClinicalTrials. gov identifier, NCT03660059, estimated completion date June 2020). This will further add to the evidence base for the efficacy and safety of peficitinib for RA in Asian patients.

Conclusion
Administration of peficitinib for a mean duration of 22.7 months provided sustained improvements in ACR response rates, ACR core components, and DAS28-CRP. Peficitinib administration for up to 6 years was well tolerated and no new safety signals were identified. These results suggest that peficitinib may be an effective long-term treatment option for Asian patients with RA.