Three-year safety and two-year effectiveness of etanercept in patients with rheumatoid arthritis in Japan: Results of long-term postmarketing surveillance.

Abstract Objectives: To evaluate the real-world safety and effectiveness of etanercept (ETN) in Japanese patients with rheumatoid arthritis. Methods: This postmarketing surveillance study (NCT00503139) assessed the safety and effectiveness of ETN treatment over 3 and 2 years (from June 2007 to September 2011), respectively. Safety was evaluated by occurrence and seriousness of adverse drug reactions (ADRs), and of adverse events (AEs) for malignancies. Effectiveness was assessed using the Disease Activity Score in 28 joints based on the erythrocyte sedimentation rate (ESR) with four variables (swollen and tender joint counts, ESR, and patient global assessment; DAS28-4/ESR). Treatment was considered effective if patients had a good/moderate response by the European League Against Rheumatism response criteria. Results: ADRs occurred in 256/675 (37.9%) patients, the most common being injection site reactions (4.4%) and nasopharyngitis (3.3%). Serious ADRs occurred in 60/675 (8.9%) patients, the most frequent being pneumonia (1.2%). The incident rate of malignancies (AEs) was 1.06 per 100 patient-years. Mean baseline DAS28-4/ESR for the 581 patients included in effectiveness analysis was 5.42, which decreased to 3.32 at 2 years. Eighty-two percent of patients achieved a moderate/good response at 2 years. Conclusion: Long-term ETN treatment safety and effectiveness were sustained over 3 and 2 years, respectively.


Introduction
Rheumatoid arthritis (RA) is a chronic autoimmune disease involving the progressive destruction of the joints that significantly causes pain and functional disability [1]. The prevalence of RA is 0.2% globally [2], and between 0.6% and 1.0% in Japan [3]. Management of RA involves the use of conventional synthetic disease modifying antirheumatic drugs (csDMARDs), such as methotrexate (MTX), as firstline therapy [4,5]. In patients failing to respond to csDMARDs, current recommendations suggest the use of csDMARDs combined with biologic DMARDs (bDMARDs), such as tumor necrosis factor (TNF) inhibitors [4,5]. The efficacy and safety of TNF inhibitors in patients with RA have been demonstrated in several randomized clinical trials [6][7][8][9][10][11][12][13]. However, rare safety issues associated with the treatment may not arise in the controlled conditions of clinical trials due to the strict study inclusion and exclusion criteria. Observational studies, with no strict inclusion/exclusion criteria and with a high number of patients, including those with multiple comorbidities [14], provide essential information for clinical practice, because they reflect reallife scenarios [15]. Postmarketing surveillance (PMS) is a method to monitor and assess real-world effectiveness and safety of a drug following its market release. The long-term nature of these surveillance studies allow the assessment of health risks associated with the use of a medicinal drug over time.
In RA clinical trials with TNF inhibitors, Japanese patients had higher responses than Western patients treated with the same TNF inhibitors [16][17][18], suggesting that Japanese patients with RA may have a unique genetic and environmental background that influences the effectiveness and safety of these agents [19]. Etanercept (Enbrel V R ; ETN; Wyeth, Tokyo, Japan; Pfizer and Takeda Pharmaceutical Company Limited, Osaka, Japan) is a TNF inhibitor used to treat patients with RA who have had an inadequate response to at least one csDMARD. ETN has been marketed in Japan since the end of March 2005 [20], following conditional approval by the Ministry of Health, Labour and Welfare (MHLW) that required both an All Cases Surveillance program and a long-term PMS to confirm the safety and effectiveness of ETN treatment. The All Cases Surveillance program has previously demonstrated the safety and effectiveness of ETN treatment over 6 months in $14,000 Japanese patients with RA [21].
In this independent and nationwide PMS study, we evaluated the long-term safety of ETN, including incidence of malignancies, in Japanese patients with RA over a period of 3 years and effectiveness over 2 years. , and patient information was obtained from case report forms using electronic data capture. Patients with RA who had an inadequate response to csDMARDs, who were never treated with ETN, who had neither history nor concurrency of malignancies were enrolled in this study. The protocol for this PMS was reviewed and approved by the MHLW and the study was conducted in accordance with good postmarketing surveillance practice.

Endpoints
The primary endpoints assessed were safety and effectiveness of ETN treatment in Japanese patients with RA over 3 and 2 years, respectively. Safety data were coded with preferred terms according to system organ class from the Medical Dictionary for Regulatory Activities (MedDRA/J ver.15.0). Effectiveness of treatment was assessed using the European League Against Rheumatism (EULAR) response criteria as measured by level of and change in the Disease Activity Score in 28 joints (DAS28) based on the erythrocyte sedimentation rate (ESR) with four variables (swollen and tender joint count, ESR and patient global assessment; DAS28-4/ESR) [22]. Secondary effectiveness endpoints included change in DAS28-4/ESR and DAS28-3/ESR calculated using three variables (swollen and tender joint count and ESR), change in DAS28 based on C-reactive protein

Registered patients, N=684
Patients included in safety analysis, N=675

Patients included in effectiveness analysis, N=582
Case report form collected, N=680 • After 0.5 year, n=680 • After 1 year, n=559 • After 1.5 years, n=473 • After 2 years, n=407 • After 2.5 years, n=374 • After 3 years, n=340 Patients excluded from effectiveness analysis, N=93 • Indeterminate due to a lack of data before or after administration b , n=93 • Patients with less than 2 weeks of administration b , n=4 Patients excluded from safety analysis, N=5 • Patients with a prior history of or comorbid malignant tumors, n=5 Patients with unavailable case report form a , n=4 Figure 1. Patient disposition. a Reasons for unavailability are: tightening of regulations for facility visits (n ¼ 2), patient was found to be ineligible (n ¼ 1), patient was referred to another facility (n ¼ 1). b Four patients of 'Indeterminate due to a lack of data before or after administration' also fall under 'Patients with less than 2 weeks of administration'.

Data analysis
The proportion of patients experiencing adverse drug reactions (ADRs), serious ADRs (SADRs), and infections were reported. Malignancies were evaluated as both ADRs and adverse events (AEs

Baseline characteristics
A total of 675 and 582 patients were included in the safety and effectiveness analysis, respectively. For the effectiveness analysis, data either before or after drug administration were missing from 93 patients. The initial ETN dose was 25 mg twice a week in 83.6% of patients and 25 mg once a week in 16.1%. In two patients, the dosing pattern was a 15 mg dose once every 2 weeks and a 25 mg dose five times every 3 weeks. A total of 684 patients were registered in the study; the case report form was not available for four of them and they were excluded from the analysis (Figure 1). Of the 680 remaining patients, five were excluded as having history of or concomitant malignant tumors ( Figure 1). Baseline characteristics of the 675 patients included in the safety analyses are shown in Table 1. The majority of patients were female (82.7%); the mean (SD) duration of disease was 8.4 (6.6) years (RA duration !10 years in 38.1% of patients). Sixty-one percent of patients were in either Steinbrocker stage III or IV, and 23.1% were in either Steinbrocker functional class 3 or 4. The mean (SD) DAS28-4/ESR before starting drug administration was 5.42 (1.26). At baseline, 60.4% of patients received concomitant MTX (mean [SD] dose of 7.0 [2.1] mg) and 73.8% of patients received concomitant corticosteroids. Nearly all patients (97.5%) received a chest X-ray or computed tomography scan, and 89.5% of patients received a tuberculin skin test.

Safety analysis
Over the 3-year period of the surveillance, 37.9% (256/675) of patients developed ADRs (Table 2). Injection site reaction was the most commonly observed ADR (4.4%) followed by nasopharyngitis (3.3%), bronchitis (3.0%), and upper respiratory tract inflammation (3.0%). Infections and infestations were reported in 106/675 patients (15.7%). Over the  (Table 3), corresponding to an IR of 1.06 (95% CI, 0.58-1.78)/100 patient-years. There was no clear trend in the time of onset of malignant tumors. One 79-year-old female patient with a previous history of interstitial lung disease developed concurrent gastric cancer and lung neoplasm malignant. The administration of ETN was discontinued but the outcome for both conditions was unrecovered/unchanged. The cause of these malignant events could not be determined. Five deaths were reported during the study, two of which (one each of metabolic acidosis and interstitial lung disease) were deemed to be potentially treatment-related. The other three deaths were sudden cardiac death, cerebral hemorrhage, and sudden death. All ADRs and SADRs reported during the 3year surveillance period are shown in Supplementary Table  S1 and Supplementary Table S2, respectively. Of the patients who developed ADRs over the study period, 27.3% (184/675) were reported within 1 year, 18.9% (90/475 patients) between 1 and 2 years, and 11.6% (43/372) between 2 and 3 years. SADRs occurred in 4.3% (29/675) of patients within 1 year, 5.1% (24/475) between 1 and 2 years, and 3.5% (13/372) between 2 and 3 years. The cumulative incidence of ADRs and SADRs are shown in Figure 2. In terms of average dose per week, ADRs occurred in no patients (0/11) at <20 mg ETN dose, 33.3% (3/9)  Several demographic and baseline characteristics were found to be significantly associated with the risk of ADRs, SADRs, infections, and serious infections as shown in Table 4. Concomitant use of MTX was a significant factor that decreased the risk ratio for the incidence of ADRs and serious infections. Increased age was a significant factor that increased the risk of SADRs and serious infections.

Treatment duration and reasons for treatment discontinuation
Among the patients included in the safety population, 53.5% (361/675) of patients dropped out or were discontinued. The mean time period from administration to discontinuation was 382.8 days (Figure 2(c)). The most common reason for discontinuation was lack of effectiveness, which occurred in 13.3% of patients (90/675), followed by AEs and hospital transfer in 12.4% (84/675) and 9.0% (61/675) of patients, respectively.     tuberculosis. a HR > 1 indicates that the risk was higher in 'female' than in 'male'; HR <1 indicates that the risk was lower in 'female' than in 'male'. b Continuous unit ¼1. c HR > 1 indicates that the risk was higher in 'yes' than in 'no'; HR <1 indicates that the risk was lower in 'yes' than in 'no'. d HR > 1 indicates that the risk was higher in patients in 'Steinbroker stage I and II' than in patients in 'Steinbroker stage III and IV'; HR < 1 indicates that the risk was lower in patients in 'Steinbroker stage I and II' than in patients in 'Steinbroker stage III and IV'. e HR > 1 indicates that the risk was higher in patients in 'Steinbroker functional class 1 and 2' than in patients in 'Steinbroker functional class 3 and 4'; HR < 1 indicates that the risk was lower in patients in 'Steinbroker functional class 1 and 2' than in patients in 'Steinbroker functional class 3 and 4'.
For factors affecting effectiveness, a significant difference in the EULAR response was observed depending on DAS28-4/ESR activity at the start of administration (    were not associated with a significant difference in the effectiveness rate (all p > .05).

Discussion
Use of biologics alone or in combination with MTX greatly improved health outcomes in patients with RA; however, a substantial proportion of patients do not achieve remission or low disease activity. In addition, these treatments come with notable side effects that challenge their long-term use.
In this PMS, we evaluated the long-term safety and effectiveness of ETN administration in a population of Japanese patients with RA under different dose regimens. ADRs were reported in 37.9% of patients, the majority of them were not serious; the most frequent being injection site reactions (4.4%) and nasopharyngitis (3.3%). Study discontinuation due to ADRs accounted for 12.4% of the study population. Over the whole surveillance period, SADRs occurred in 8.9% of patients and pneumonia was the most common (1.2%), followed by interstitial lung disease (0.9%). In terms of infectious events, use of ETN and other TNF inhibitors has been associated with an increased risk of reactivating latent TB [25,26]. In this study, only two patients with one event each developed TB; thus, incidence of TB in this population of patients was not deemed to be clinically significant. The incidence rate of herpes zoster in this study   (IR 1.07 per 100 patient-years) was similar to that reported in a large observational study conducted between 2005 and 2010 in Japan (IR 9.1 per 1000 patient-years), and lower than the incidence rate reported in Japanese patients with RA treated with Janus kinase (JAK) inhibitors (IR 4.4 per 100 patient-years) [27,28]. Patients with RA are also at higher risk of developing malignancies than the general population is [29,30]; however, no trend toward increased cancer incidence was noted in this study. Concomitant use of MTX was a factor associated with a decreased risk of developing ADRs and serious infections, which is consistent with previous studies [21,31] whereas concomitant diabetes and advanced age were associated with an increased risk of developing infections and SADRs, respectively. As lurking confounders might affect the results of this analysis, tight monitoring is advisable in the presence of concomitant use of MTX. For instance, concomitant use of MTX in Japanese patients with RA treated with adalimumab was reported to be a risk factor for infections [32]. In conclusion, this PMS did not reveal new safety signals and continuous ETN treatment in Japanese patients with RA remained favorable. With respect to effectiveness, our data demonstrate that treatment with ETN improved DAS28-4/ESR scores over time in the majority of patients. Reduced disease activity was also demonstrated by increments in the proportion of patients in remission over the 2-year study period. VAS fatigues and mHAQ scores were also significantly improved at every time point compared to baseline. Of note, given that 45.9% of patients were aged !60 years, and a combined 61.0% of patients were in either Steinbrocker stage III or IV, this PMS showed good effectiveness of ETN treatment in elderly patients and in patients with advanced RA. Among factors affecting effectiveness, patients with low DAS28-4/ESR scores at baseline showed the least improvements by the end of the observational period compared with patients with a high DAS28-4/ESR score at the beginning of the study. However, this outcome was expected as disease activity is already low in patients with low DAS28-4/ ESR scores at baseline. Discontinuation of treatment due to lack of effectiveness was reported in 13.3% of patients. In conclusion, the findings of this study support long-term effectiveness of ETN treatment in Japanese patients with RA.
The limitations of this study include the restriction to a Japanese population with RA and its relatively small size, thus the findings and conclusions may not be generalizable to other patients with RA in Japan and in the rest of the world. Another limitation was the high number of patients who dropped out due to unknown reasons or with missing data. As this was a PMS study, there was no comparator arm; thus it cannot provide insights into the relative safety or effectiveness of ETN versus other bDMARDs.
In conclusion, this long-term study of ETN treatment in Japanese patients with RA showed no evidence of novel safety concerns, and demonstrated that effectiveness was sustained over a 2-year period.