Long-term safety of secukinumab in patients with moderate-to-severe plaque psoriasis, psoriatic arthritis, and ankylosing spondylitis: integrated pooled clinical trial and post-marketing surveillance data

The integrated clinical trial safety dataset included data pooled from 21 randomized controlled clinical trials of secukinumab in PsO (14 Phase 3 trials and 1 Phase 4 trial; N = 5181), PsA (3 Phase 3 trials; N = 1380), and AS (3 Phase 3 trials; N = 794) indications (see Fig. 1 for details), along with post-marketing surveillance data for secukinumab across the PsO, PsA, and AS indications from December 26, 2014 to June 25, 2017. Multiple secukinumab dose regimens were used in these studies and included intravenous (up to 10 mg/kg) or subcutaneous (s.c.; 75, 150, or 300 mg) loading followed by s.c. maintenance dosing (75, 150, or 300 mg). Placebo-treated patients were re-randomized to secukinumab between 12 and 24 weeks’ post-baseline in the various studies. Eligible patients aged ≥ 18 years with moderate-to-severe plaque PsO or active PsA or AS were enrolled based on pre-specified eligibility criteria, which, along with the design of each study, have been reported in detail elsewhere [12, 15, 16, 18, 20, 23‐26]. Of note, patients enrolled in these trials could have active/ongoing cardiovascular disease (unless severe or uncontrolled), previous history of inflammatory bowel disease (IBD), including Crohn’s disease (CD) or uveitis (but not if active and ongoing), history of basal cell carcinoma or actinic keratosis (successfully treated with no evidence of recurrence in the past 3 months), carcinoma in situ of the cervix or non-invasive malignant colon polyps (successfully removed), or latent tuberculosis (TB; prophylactic treatment started prior to enrollment). Patients could continue the following medications at stable doses: sulfasalazine, methotrexate, corticosteroids, and non-steroidal anti-inflammatory drugs. Patients with an inadequate response (if taken ≥ 3 months) or intolerance to tumor necrosis factor (TNF) inhibitors (not more than one in AS studies and not more than 3 in PsO and PsA studies) could also be included. Two studies (ERASURE and FIXTURE) in the PsO pool also allowed patients with a prior history of usage of biologics other than TNF inhibitors (alefacept, briakinumab, efalizumab, and ustekinumab) after an appropriate washout period.

Safety analyses included all patients who had received ≥ 1 dose of study medication, with data pooled at the individual patient level for all secukinumab treatment groups (any secukinumab) separately for the studies by indication. Analysis of the entire secukinumab treatment period was performed on safety data pooled at the patient level from the date of treatment initiation up to June 25, 2017. Analysis of the safety data was also performed by the approved secukinumab 300 and 150 mg doses separately for the studies by indication. Risks for AEs, serious AEs (SAEs), and selected AEs were expressed as exposure-adjusted incidence rates (EAIRs) per 100 patient-years for the entire treatment period. The EAIR was defined as the number of subjects exposed to the drug and experiencing a certain event divided by the total exposure time of all subjects who were at risk for the event.

AEs were coded using the Medical Dictionary for Regulatory Activities (MedDRA) version 20.0 preferred terms (PTs) (https://​www.​meddra.​org/​). ADAs were measured using a Meso-scale Discovery bridging assay [27]. ADA-positive samples were analyzed to week 52 for drug-neutralizing potential, immunogenicity-related AEs, and impact on pharmacokinetics (PK) and efficacy of secukinumab. All clinical studies were conducted in compliance with the Declaration of Helsinki [28], International Council for Harmonization Guidelines for Good Clinical Practice, and local country regulations.

The EAIRs of any AE with secukinumab treatment across the entire safety period were 204.4, 147.0, and 140.1 per 100 patient-years in the PsO, PsA, and AS pools, respectively. The EAIRs of any SAE with secukinumab treatment across the entire safety period were 6.9, 7.9, and 6.3 per 100 patient-years in the PsO, PsA, and AS pools, respectively, showing no discernible pattern across the treatment groups. A total of 9 (0.2%) deaths were reported in the secukinumab-treated PsO cohort (the primary reasons for death being pulmonary embolism, arrhythmia, alcohol poisoning, ruptured aneurysm, and myocardial infarction [2 patients] in the secukinumab 300 mg group and cerebral hemorrhage, cardiorespiratory arrest, and completed suicide in the secukinumab 150 mg group). There were 11 (0.8%) deaths in the PsA pool (150 mg group: acute myocardial infarction, septic shock, sepsis, pneumonia, metastatic small cell lung cancer, pancreatic carcinoma, and cardiac failure [2 patients]; 75 mg group: squamous cell carcinoma of the pharynx, myocardial infarction, and cerebrovascular event) and 5 (0.6%) in the AS pool (150 mg group: 1 unknown; 75 mg group: acute respiratory failure, cerebrovascular accident, acute myocardial infarction, and respiratory arrest). Discontinuations due to AEs numbered 331 (6.4%), 104 (7.5%), and 58 (7.3%) in the PsO, PsA, and AS pools, respectively (Table 2). The EAIRs of selected AEs with secukinumab were comparable across the PsO, PsA, and AS studies (Table 3). The rates were comparable with those reported previously [15, 16, 18, 20], and no new safety signals were identified from those reported previously.

Over the entire safety period, upper respiratory tract infection (URTI) was the most common type of infection across all indications. In an overall dataset of 7355 patients, the EAIRs for serious infections were 1.4, 1.9, and 1.2 per 100 patient-years in the PsO, PsA, and AS pools, respectively (Table 3). A total of 14 cases (EAIR of 0.13 per 100 patient-years) of opportunistic infections were reported in the PsO pool, which included 10 cases of esophageal candidiasis, 3 cases of gastrointestinal candidiasis, and 1 case of herpes zoster infection. Eight cases (EAIR of 0.21 per 100 patient-years) of opportunistic infections were reported in the PsA pool, which included 4 cases of esophageal candidiasis and 1 case each of herpes zoster infection, toxoplasmosis, Mycobacterium avium complex infection, and pneumonia. The 2 cases (EAIR of 0.1 per 100 patient-years) of opportunistic infections reported in the AS pool were herpes zoster infection and esophageal candidiasis. A total of 5 cases (EAIR of 0.05 per 100 patient-years) of sepsis (PT) were reported in the PsO pool, 3 (EAIR of 0.08 per 100 patient-years) in the PsA pool, and none in the AS pool. Most infections were mild to moderate.

Cutaneous or mucosal Candida infection (MedDRA high-level term) was reported in 221 (EAIR of 2.2 per 100 patient-years) patients across the PsO studies, 57 (EAIR of 1.5 per 100 patient-years) patients across the PsA studies, and 13 (EAIR of 0.7 per 100 patient-years) patients across the AS studies (Table 3 and Additional file 1: Table S1). All cases of Candida infection were localized, most were mild or moderate in severity (except for 4 cases in the PsO pool that were considered severe), were self-limited or responsive to standard treatment, and did not lead to discontinuation of study treatment. No cases of systemic candidiasis were reported in patients treated with secukinumab across all studies.

Over the entire treatment period, the EAIRs per 100 patient-years for neutropenia with secukinumab treatment were 0.3, 0.2, and 0.5 in the PsO, PsA, and AS studies, respectively (Table 3). In the PsO pool, grade 3 neutropenia (defined as an absolute neutrophil count between 1.0 and 0.5 × 10⁹/L) was reported in 33 (0.6%) patients and grade 4 neutropenia (defined as an absolute neutrophil count of less than 0.5 × 10⁹/L) was reported in 2 (0.04%) patients. Of the cases of neutropenia, 7 were reported as severe, while the remaining were mild or moderate in severity (viral URTI was the most frequently co-reported AE [17 cases]); there were 5 cases in which study treatment was withdrawn. In the PsA pool, 12 (0.9%) patients reported grade 3, and 3 (0.2%) patients reported grade 4 neutropenia. All cases were either mild or moderate in severity (anemia, diarrhea, fatigue, leukopenia, URTI, and urinary tract infection were the most frequently co-reported AEs [3 cases each]), with 1 instance in which study treatment was withdrawn. In the AS pool, 9 (1.1%) patients reported grade 3, and 5 (0.6%) patients reported grade 4 neutropenia in the AS pool. All cases of neutropenia reported in the AS pool were either mild or moderate in severity (diarrhea was the most frequently co-reported AE [4 cases]), with none leading to study treatment discontinuation.

The EAIRs per 100 patient-year exposure for CD, UC, or unspecified IBD combined ranged from 0.01 to 0.1 in PsO, 0.05 to 0.08 in PsA, and 0.1 to 0.4 in the AS pools (Table 3). There were 41 (0.6%) reported cases of IBD (including CD and UC). Of these, 30 (0.4%) were new-onset cases. Across indications, 14 (0.2%) patients who reported IBD discontinued the study.

The EAIR per 100 patient-years for major adverse cardiovascular events (MACE) with secukinumab treatment over the entire treatment period was 0.3, 0.4, and 0.6 in the PsO, PsA, and AS pools, respectively (Table 3).

In the PsO pool, the EAIR for uveitis was 0.02 per 100 patient-years over the entire treatment period; all cases of uveitis (n = 2) were de novo. In the PsA pool, the EAIR for uveitis was 0.1 per 100 patient-years over the entire treatment period; all cases of uveitis (n = 3) were de novo. The EAIR for uveitis in the AS pool was 1.4 per 100 patient-years over the entire treatment period; a total of 12 (46%) cases were de novo (Table 3). Among all cases of uveitis (n = 26) in the AS pool, one was reported as severe and the remainder were reported as mild or moderate. A total of 135 (17%) patients in the AS pool reported pre-existing (but not active or ongoing) uveitis at baseline, and 589 (74.2%) were HLA-B27 positive, which has a known association with the development of uveitis.

Malignant or unspecified tumors (Standardized MedDRA Queries [SMQ]) were reported in 81 patients (EAIR of 0.8 per 100 patient-years) in the PsO pool, which included 76 patients with non-hematological malignant tumors, 2 patients each with a hematological malignant tumor or a non-hematological tumor of unspecified malignancy, and 1 patient with a hematological tumor of unspecified malignancy. In the PsA pool, malignant or unspecified tumors were reported in 43 patients (EAIR of 1.1 per 100 patient-years), which included 37 patients with non-hematological malignant tumors, 5 patients with a non-hematological tumor of unspecified malignancy, and 1 patient with a hematological malignant tumor. In the AS pool, malignant or unspecified tumors were reported in 9 patients (EAIR of 0.5 per 100 patient-years), which included 6 patients with non-hematological malignant tumors, 2 patients with a non-hematological tumor of unspecified malignancy, and 1 patient with a hematological malignant tumor.

Over the entire treatment period, in the PsO studies, 8 patients (0.2%) reported some form of suicidality-related AEs, including 4 attempted suicides, 1 completed suicide, 2 cases of suicidal ideation, and 1 case of suicidal depression. Three cases (0.2%) of suicidal ideation were reported in the PsA studies. All suicidality cases presented at least one of the following: depression, anxiety, insomnia, bipolar disorder, co-medication with psychoactive drugs, alcoholism, and/or ongoing psychological or socio-economic issues. No cases of suicidality-related AEs were reported in the AS studies on secukinumab treatment. These suicidality safety results with secukinumab were consistent with an earlier report from a pooled analysis of data from 10 clinical studies in moderate-to-severe plaque PsO [29].

Treatment-emergent ADAs were reported with secukinumab in < 1% of patients across all studies at week 52. All treatment-emergent ADAs were associated with normal PK [30], and none were associated with loss of secukinumab efficacy or immunogenicity-related AEs.

The EAIR of injection site reactions (high-level term) was 1.2, 1.3, and 0.8 in the PsO, PsA, and AS pools, respectively.

Safety data for any AE, any SAE, serious infections, Candida infection, IBD, and MACE for the approved 300 and 150 mg secukinumab dose are shown in Additional file 1: Table S2. The EAIRs were generally comparable, and there was no dose-response relationship observed in terms of safety events with secukinumab treatment within each indication.

Data on a year-by-year basis for any AE, any SAE, serious infections, Candida infection, IBD, and MACE showed no increase with secukinumab treatment over time across studies within each indication (Fig. 2). The EAIR per 100 patient-years for uveitis on a year-by-year basis was as follows: 0.02 (N = 5181; year 1), 0.0 (N = 3268; year 2), 0.0 (N = 2246; year 3), 0.07 (N = 1627; year 4), and 0.0 (N = 1210; year 5) in PsO clinical trials; 0.2 (N = 1380; year 1), 0.0 (N = 1183; year 2), 0.0 (N = 948; year 3), 0.2 (N = 587; year 4), and 0.0 (N = 290; year 5) in PsA clinical trials; and 1.1 (N = 794; year 1), 1.9 (N = 700; year 2), 2.2 (N = 557; year 3), and 2.5 (N = 332; year 4) in AS clinical trials.

The cumulative post-marketing exposure to secukinumab was estimated to be ~ 96,054 patient-years across the approved indications (Table 4). The exposure-adjusted reporting rates (EARRs) for infections and serious infections were 4.7 and 1.8 per 100 patient-years, respectively. Neutropenia was reported at a rate of 0.07 per 100 patient-years. The reporting rate of hypersensitivity was 2.4 per 100 patient-years. The EARRs for malignancies and MACE were both 0.2 per 100 patient-years, with most assessable cases having multiple confounders, risk factors, or alternative explanations for the events. IBD was reported at a rate of 0.2 per 100 patient-years. The reporting rate for suicidal ideation and behavior was 0.04 per 100 patient-years. Twenty-nine cases of opportunistic infections were reported in the post-marketing data, including 5 cases of TB infection (no reactivation cases), 2 cases of herpes infection, and 17 cases of esophageal candidiasis. There was 1 case of immunogenicity, and no cases of either hepatitis B reactivation or interactions with live vaccines were reported. The safety profile from the PSUR was consistent to that reported in randomized clinical trials (RCTs) with secukinumab.

Secukinumab demonstrated a favorable safety profile over long-term treatment in patients with PsO, PsA, and AS. The safety profile of secukinumab was consistent in these pooled patient populations and with what has been reported earlier for individual studies of secukinumab [12, 13, 15‐21, 23, 24, 26, 49]. This long-term (up to 5 years) safety assessment provides a broader understanding of the safety of secukinumab and supports its long-term use in these chronic systemic inflammatory conditions.