FormalPara Key Summary Points

Why carry out this study?

The Phase 3 clinical trials of satralizumab proved the efficacy and safety of satralizumab in a limited sample size of a well-selected patient population.

However, the effectiveness and safety of satralizumab in a broader group of patients with neuromyelitis optica spectrum disorder (NMOSD) have not been assessed.

The hypothesis of the study was that the safety and efficacy results of satralizumab in clinical trials can be translated to patients with NMOSD in a routine clinical setting.

What was learned from the study?

Satralizumab was found to be safe, well tolerated, and effective in patients with NMOSD in routine clinical practice at 6-month interim analysis, and these results are consistent with those of previous Phase 3 trials.

The study will be continued to assess the safety and effectiveness of satralizumab in Japanese patients with NMOSD over the 6-year surveillance period.

Introduction

Neuromyelitis optica spectrum disorder (NMOSD) is a chronic autoimmune disease, primarily affecting the optic nerve and spinal cord in the central nervous system through the aquaporin-4 (AQP4) water channel of astrocytes, ultimately resulting in paralysis and blindness [1]. The disease is characterized by a variety of symptoms including visual impairment and loss of motor, sensory, and bowel–bladder functions along with occasional incidences of nausea, emesis, and hiccups [2]. It preferentially affects adult females with a female:male ratio of 9:1 worldwide. Around 5–10% of patients with NMOSD have been diagnosed in adolescence [3]. In Japan, a nation-wide survey estimated the prevalence of NMOSD to be around 3.42 per 100,000 population in 2011 [4]. Northern Japan has a slightly higher prevalence of 4.1 per 100,000 population as per a 2016 local survey [5]. Relapses are commonly associated with NMOSD, leading to a high risk of permanent disability over a period of time. Hence, disease management is aimed to reduce the number of relapses with simultaneous reduction in disease severity [6].

Approaches to treatment include acute treatment of attack and long-term preventive or maintenance therapy. Acute relapse therapy comprises intravenous methylprednisolone, plasma exchange/immunoadsorption therapy, and immunoglobulins, while long-term maintenance therapy includes non-selective immunosuppressants and specific immune targeting agents [7]. Interleukin-6 (IL-6) levels are generally increased in the cerebrospinal fluid of patients with frequent NMOSD relapses [7]. Targeting IL-6 with suitable inhibitors has shown to decrease relapse rate in patients with NMOSD [8].

Satralizumab is a humanized immunoglobin G2 (IgG2) anti-IL-6 receptor monoclonal recycling antibody engineered to have a longer plasma half-life [7]. Previous Phase 3, randomized, double-blind, placebo-controlled clinical trials, SAkuraSky (NCT02028884) and SAkuraStar (NCT02073279), have used satralizumab as combination therapy with immunosuppressants or as monotherapy in adults and children aged ≥ 12 years with NMOSD who are AQP4 water channel auto-antibody (AQP4-IgG) seropositive [9, 10]. These trials have shown satralizumab therapy to be efficacious in reducing the risk of protocol-defined relapse (PDR) and severe PDR when compared with placebo, and provide relevant evidence on the mechanism of action, efficacy, and safety of satralizumab. The median treatment duration with satralizumab in the double-blind periods were 107.4 and 92.3 weeks, respectively. In these trials, the eligible participants had experienced at least 1 relapse in the past 12 months and had an Expanded Disability Status Scale (EDSS) score of 0–6.5 at screening.

Moreover, in the open-label extension following the double-blind periods of SAkuraSky and SAkuraStar, the long-term efficacy of satralizumab in patients with AQP4-IgG seropositive NMOSD has shown that 71% and 73% of satralizumab-treated patients were free from annualized investigator-determined PDR (iPDR), 91% and 90% patients were free from severe iPDR, 90% and 86% had no sustained EDSS worsening, and the overall adjusted iPDR rates (ARRs) were 0.12 and 0.08, respectively [6]. The median treatment durations with satralizumab in the double-blind and extension periods were 4.4 and 4.0 years, respectively [6].

Interestingly, SAkuraMoon (NCT04660539), a single-arm, open-label study in adult participants with AQP4-IgG seropositive NMOSD who had completed the SAkuraSky and SAkuraStar trials, has shown that satralizumab was effective long term at preventing relapses, and 72% of satralizumab-treated patients were free from ARRs, 91% patients were free from severe iPDR, and 85% had no sustained EDSS worsening [11].

Based on the encouraging efficacy and tolerable safety results, satralizumab received approval on June 1, 2020, in Canada, and subsequently on June 29, 2020 in Japan, for the management and prevention of NMOSD relapse. This was followed by approval for the drug in other countries [12].

The longer plasma half-life of satralizumab has enabled an effective dose schedule of once in 4 weeks of satralizumab administered subcutaneously. In Japan, the approved dosage of 120 mg given at weeks 0, 2, and 4 as loading doses, followed by a maintenance dose of 120 mg every 4 weeks [12].

However, the Phase 3 clinical trials involved a limited sample size with a well-selected patient population. Thus, it is necessary to examine if the safety and efficacy results of satralizumab can be translated to patients with NMOSD in a routine clinical setting. This large-scale, ongoing postmarketing surveillance (PMS) study will evaluate the real-world safety and effectiveness of satralizumab over 6 years in patients diagnosed with NMOSD in Japan. In this study, we aimed to present the results of 6-month interim analysis from this ongoing study, to describe PMS safety of satralizumab in the early post-marketing period in Japan.

Methods

Study Design

This study is an ongoing observational, general drug use surveillance (All Patient Surveillance) investigating the safety and effectiveness of satralizumab in patients with NMOSD conducted across 234 centers in Japan (UMIN Clinical Trials Registry, UMIN000041047). The total surveillance period for this study is from August 26, 2020, to February 28, 2027.

The case enrollment period was from August 26, 2020 to July 31, 2021, during which patients with NMOSD scheduled to receive satralizumab were registered using a fax-based central registration system. The planned observation period is from the start date of treatment (index date) with satralizumab to August 31, 2026, (6 years after the launch of satralizumab in Japan) for each patient or until the time of treatment discontinuation in patients who discontinued satralizumab. This interim analysis was conducted when the data lock of the 6-month case report form (CRF) was completed for all patients (the end of observation period). For the patients who discontinued satralizumab during the 6-month follow-up, the date of last observation was defined as the end date.

Thus, the results reported here are of the interim analysis of all patients whose 6-month CRFs and the data clarification of the 6-month CRFs were collected (Fig. 1).

Fig. 1
figure 1

Study design

Full size image

Study Participants

The study included all patients who used satralizumab during the registration period and whose survey forms were collected prospectively. However, data from all patients who had started using satralizumab before initiation of this surveillance were also included retrospectively. Enrolled patients received satralizumab subcutaneously in routine clinical practice. CRFs were collected from all patients who received satralizumab. The data on the patient demographics and characteristics, treatment with satralizumab, treatment with concomitant therapy (at the index date and at 6 months), relapse, and adverse events were collected.

Outcomes

The primary outcomes assessed at each time point were the proportion of patients with adverse drug reactions (ADRs) which were classified according to the International Conference on Harmonisation Medical Dictionary for Regulatory Activities Japanese edition (ICH MedDRA/J) Version 25.1; the incidence of adverse reactions (person-years method; by safety specification); and the concomitant oral glucocorticoids given as maintenance therapy. The daily oral glucocorticoid doses were recorded as categorical values and mean doses at the start of treatment, 6 months, the end of the observation period, and the end date. The secondary outcomes were time to relapse and relapse rate while being on continuous satralizumab treatment, wherein relapse was defined clinically based on the judgement of the physicians without setting any predetermined criteria. The presence or absence of relapse was recorded. For those who had a relapse, the date of the diagnosis of relapse and the treatment status for relapse were recorded.

Statistical Analysis

All analyses were descriptive. Continuous variables were summarized descriptively using mean (standard deviation [SD]), median (minimum, maximum), or median (interquartile range [IQR]) while categorical data were summarized using frequencies and percentages. No imputation of missing values was performed. Since the data after the discontinuation of satralizumab were not available, the data on oral glucocorticoid reduction were analyzed only for patients who continued satralizumab treatment. Analyses were conducted using SAS, version 9.4 (SAS Institute Inc., Cary, NC, USA).

Ethical Consideration

The study was conducted in accordance with relevant regulations in Japan (Ministerial Ordinance on Good Post-Marketing Study Practice [GPSP], Ministry of Health, Labour and Welfare Ordinance Number 38, March 23, 2005).

The study protocol was reviewed and approved by The Japanese Pharmaceuticals and Medical Devices Agency (PMDA) [13] prior to study initiation.

The study did not undergo review by the ethics committee of the participating medical institutions or follow procedures for informed consent as this was not required for PMS studies according to Japanese regulations of the Act on Securing Quality, Efficacy and Safety of Products Including Pharmaceuticals and Medical Devices [14], to ensure the registration of all patients with NMOSD and all patients set out as approval conditions in Japan. The patient data were collected after de-identification.

Results

Study Population and Baseline Characteristics

Overall, 575 patients were registered in the PMS study (Fig. 1). Of the 571 patients whose survey forms were collected, 1 patient was excluded from the safety analysis as no consent was obtained for including the data for publication. Hence, 570 patients were included in the safety analysis set. Of these, 10 patients were excluded from the effectiveness analysis set (7 patients previously treated with satralizumab and 3 patients for whom effectiveness data were not analyzed as the details of presence or absence of relapse were unknown). Thus, 560 patients were included in the effectiveness analysis set.

Baseline characteristics for the safety analysis set are shown in Table 1 and Table S1 in the electronic supplementary material. The mean age was 52.4 years, and majority of patients were aged between 50 and 75 years (53.68%). Most patients were female (91.75%); 2 female patients were pregnant during the treatment period. About half (48.42%) the patients had comorbidities. Most (98.24%) patients were AQP4-IgG positive. Prior treatment with immunosuppressants was given to 61.22% patients and with other biologics to 7.71% patients. Most (99.29%) patients reported clinical signs and symptoms of NMOSD including acute myelitis (73.50%), optic neuritis (62.10%), symptomatic cerebral syndrome with magnetic resonance imaging (MRI) lesions of the brain typical of NMOSD (7.19%), and symptomatic narcolepsy with MRI lesions of the diencephalon (1.75%). NMOSD EDSS score (mean ± SD) was 4.19 ± 2.19. The mean ± SD time from diagnosis to index date was 7.3 ± 6.4 years. There were no relapses in 44.91% patients and ≥ 3 relapses in 8.95% patients over a period of 2 years before the index date. Most (85.96%) patients were receiving concomitant oral glucocorticoids and about half of the patients (48.59%) were receiving immunosuppressants concomitantly (Table 1 and Table S1).

Table 1 Patient characteristics (analysis set, N = 570)
Full size table

Treatment Using Satralizumab

The mean ± SD cumulative dose of satralizumab was 918.5 ± 275.5 mg, and median (minimum–maximum) satralizumab treatment duration was 171.0 (1–540) days. Of 570 patients, 84.56% patients were receiving satralizumab treatment, dose interruption occurred in 0.35% patients with 14.91% discontinuations, and status was unknown for 0.17% patients at 6-month follow-up. Reasons for discontinuation were adverse events (6.14%), hospital transfer (5.08%), relapse (1.40%), missed visit (0.52%), and others (1.75%).

Safety

Occurrence of ADRs

In the safety analysis set, for the overall ADRs, 76.22 (66.07–87.48) events/100 person-years occurred in 118 (20.70%) patients (Table 2). Infections were the most common ADRs with 13.58 (9.51–18.80) events/100 person-years in 28 (4.91%) patients (Table 2). The occurrence of serious infections in 570 patients of the safety analysis set by percentage and person-years method based on patient background is shown in Table 3.

Table 2 Summary of adverse drug reactions by seriousness (safety analysis set, N = 570)
Full size table
Table 3 Summary of serious adverse drug reactions (serious infection)
Full size table

Serious infections occurred in 3.15% patients, with an event rate of 9.05 (5.80–13.47) events/100 person-years (Table 3). There was a difference in the event rate of serious infections between those aged 12–50 years (4.63 events/100 person-years) and those aged 50–75 years of age (11.11 events/100 person-years), but the rate was considerably higher for those over 75 years of age (24.35 events/100 person-years). Furthermore, the event rate of serious infections was higher in those with longer disease duration (≥ 10 years; 18.74 events/100 person-years) and having ≥ 3 relapses in the last 2 years (21.11 events/100 person-years; Table 3).

Status of Occurrence of Safety Specifications

Safety specifications examined in the survey were infection, neutropenia/leukopenia/agranulocytosis, thrombocytopenia, hypersensitivity, hepatic function disorder, and cardiac disorder. Table S2 in the electronic supplementary material presents data on these ADRs and the time of onset of the first ADR from the index date. The mean ± SD number of days from the index date to onset of the first adverse reaction were 54.6 ± 64.9 (Table S2a). Adverse reactions were more common within first 30 days (Table S2b). Among serious infection events, the most common infections reported were respiratory infections (7/24, 29.17%) and urinary tract infections (6/24, 25.00%; Table 4). Among total infection events, the most common infections reported were skin infections (10/36, 27.78%), followed by urinary tract infections and respiratory infections (both 9/36, 25.00%; Table 4). Table S3 in the electronic supplementary material shows details of all infections reported by percentage and event rate (per 100 person-years); pneumonia and urinary tract infection were the most common infections/serious infections each reported in 3/570 (0.52%) individuals with an event rate of 1.13 per 100 person-years. One drug-related death due to septic shock was reported. Although the invasive pneumococcal infection and concomitant medications (oral glucocorticoid, tacrolimus) were considered as possible causes of septic shock, a causal relationship with satralizumab could not be ruled out.

Table 4 Breakdown by sites of infectiona
Full size table

Change in Use of Concomitant Oral Glucocorticoids

The time profiles of daily oral glucocorticoid dose (distribution rate by category) from the index date to end date in the safety analysis set (570 patients) are shown in Table 5 and Figure S1 in the electronic supplementary material. The proportion of patients who received a low daily dose of oral glucocorticoid of < 10 mg/day increased substantially from 206/570 (36.14%) at treatment initiation to 294/496 (59.27%) at 6 months (Table 5 and Figure S1). During the same duration, the mean ± SD oral glucocorticoid dose (mg/day) reduced from 12.28 ± 10.17 at the index date to 8.11 ± 7.30 at 6 months for those who continued satralizumab (Table 5 and Figure S2).

Table 5 Oral glucocorticoid dosage from the start of treatment to the end date
Full size table

Effectiveness

The effectiveness was evaluated in terms of the relapse rate (person-years method) and time to relapse in 560 patients in the effectiveness analysis set.

Relapse Rate

The cumulative number of patients with relapse was 28 patients (relapse rate of 5.00%).

Time to Relapse

The cumulative incidence of freedom from relapse for 555 patients in the effectiveness analysis set is shown in Table 6 and Fig. 2. The Kaplan–Meier cumulative event-free rate (95% confidence interval [CI]) was 94.59% (92.25–96.23) at 6 months (sample size at 6 months; at risk: 464 patients).

Table 6 Time to relapse
Full size table
Fig. 2
figure 2

Time to relapse using the Kaplan–Meier method

Full size image

Discussion

The results of this 6-month interim PMS analysis of Japanese patients provided relevant real-world evidence on the safety and effectiveness of satralizumab in Japanese patients with NMOSD. The results suggested that satralizumab treatment was well tolerated and effective in routine clinical practice.

Compared with previous Phase 3 trials of satralizumab monotherapy and combination therapy, the patient population in the present PMS was older (mean 52.4 years vs. 45.3 and 40.8 years in SAkuraStar and SAkuraSky, respectively) [15]. Our results are from an entirely Japanese population providing real-world Japanese data. In the previous pivotal clinical trials, SAkuraStar included no Japanese patients [10] while SAkuraSky included > 70% non-Japanese population [9]. Thus, these data add value to the available evidence demonstrating that the clinical trial data translate well to a real-world Japanese scenario.

ADRs

Infections were the most common events for both total ADRs. as well as serious ADRs in the present PMS. This real-life scenario is consistent with the results of a meta-analysis of 7 randomized-controlled trials and 776 patients with NMOSD using biologics (eculizumab, inebilizumab, rituximab, satralizumab, tocilizumab), wherein severe infections were the most common serious adverse events [16]. The factors that can increase the risk for serious infections in patients with NMOSD during treatment with satralizumab were being male, advanced age (≥ 75 years), longer disease duration (≥ 10 years), and a higher number of relapses in the last 2 years.

Of the 24 events of serious infections, respiratory tract infections (29.17%; 7) and urinary tract infections (25.00%; 6) were the most common. These findings from the PMS data corroborate those observed in the clinical trial setting [10, 15]. Indeed, NMOSD is characterized by neurogenic bladder due to myelitis, resulting in a higher proportion of urinary tract infections [17]. In the present PMS, all 6 patients with serious urinary tract infections had acute myelitis. Contrastingly, the PMS results of tocilizumab, an IL-6 receptor inhibitor for rheumatoid arthritis (RA), show that respiratory infections (41.69%) were the most common type of serious infection [18]. We speculate that this difference in the occurrence of type of infections may result from differences in the disease background (wherein respiratory infections are prevalent in RA) and not due to the drug [17, 19]. The mean EDSS score of 5.0 and above is associated with a higher expression of NMOSD. Patients with such EDSS scores have severe disabilities that impair full daily activities and ability to work a full day without special provisions [20].

Concomitant Oral Glucocorticoid Status

The proportion of patients receiving 10 mg/day or more oral glucocorticoid reduced from ~ 65% at the start of satralizumab treatment to ~ 40% at the end of the 6-month observation period. Even for the 85/570 patients who discontinued satralizumab, the proportion of patients receiving 10 mg/day or more oral glucocorticoid reduced to ~ 48% at the point of discontinuation. The average oral glucocorticoid dosage at the end of the observation period was 8.11 mg/day compared to 12.28 mg/day at the beginning of treatment.

Notably, research suggests that chronic use of glucocorticoids is associated with hypertension, bone fracture, cataract, nausea, vomiting, and metabolic problems like weight gain, hyperglycemia, and type 2 diabetes [21]. Specifically, a daily dose of > 10 mg or a cumulative dose > 700 mg of prednisone can increase the rate of infections [22]. Reducing the maintenance dose of glucocorticoids is often associated with increased risk of relapse. Under these circumstances, a consistent reduction in oral glucocorticoid dosage evident in this real-world PMS is impressive.

Relapse-Free Period

This study reported relapses as a physician-judged clinical relapse, whereas previous clinical trials SAkuraStar and SAkuraSky had protocol-defined relapses adjudicated by the independent clinical end-point committee [9, 10]. Thus, it might be expected that more relapses would be reported here than the protocol-defined relapses in the Phase 3 trials. The relapse rate of 5.0% over 6-month follow-up in the present PMS was comparable or lower than relapse rates previously reported in satralizumab clinical trials (relapse rates in AQP4-IgG seropositive patients in satralizumab groups of 11% and 22% in the SAkuraSky and SAkuraStar trials, respectively, over a median treatment duration of around 48 weeks [9, 10]). Furthermore, in a real-world study using clinical practice data from a Japanese claims database, the relapse rate of NMOSD (defined as prescription of acute-phase treatment) in patients using satralizumab was 4.6%, over a median duration of satralizumab exposure of 197 days [23]. As the relapse rate data in this previous real-world study are almost identical to the relapse rate in the present PMS study, the relapse-free rate of satralizumab in real-world clinical practice may be considered to be approximately 95% over a 6-month follow-up duration.

Limitations

This PMS study is an observational study and, therefore, has certain limitations including potential reporting bias and lack of a control group. The evaluations of risk factors for infections and other ADRs have not been adjusted for other variables. The study may be limited by not using EDSS and MRI as effectiveness indicators. Thus, the relapse data collected in this PMS study are not clearly defined in contrast to a clinical trial. However, it may in fact lead to overestimation of the incidence of relapse. Thus, our PMS may result in underestimation of the effectiveness. Although the duration of observation of this interim analysis of PMS data was relatively short, this PMS will continue data collection over a period of 6 years.

Conclusion

This interim analysis of satralizumab PMS study in Japan offers the first demonstration in a real-world setting that satralizumab is safe, effective, and well tolerated in preventing relapses in 570 patients with NMOSD. No new safety concerns were reported, with ADRs being consistent with the known safety profile of satralizumab. Treatment with satralizumab is possibly associated with a reduction in the concomitant use of immunosuppressive therapies. To understand the long-term effect of satralizumab in Japanese patients with NMOSD, further data will become available over the course of the 6-year surveillance period.