Long-term Safety of Risankizumab From 17 Clinical Trials in Patients With Moderate-to-severe Plaque Psoriasis

K.B. Gordon; M. Lebwohl; K.A. Papp; H. Bachelez; J.J. Wu; R.G. Langley; A. Blauvelt; B. Kaplan; M. Shah; Y. Zhao; R. Sinvhal; K. Reich

Disclosures

The British Journal of Dermatology. 2022;186(3):466-475. 

In This Article

Results

Baseline Characteristics and Risankizumab Exposure

In total 3072 risankizumab-treated patients were included in the long-term analysis set, representing 7927 cumulative PY of risankizumab exposure. The median risankizumab treatment duration was 2·9 years. The treatment duration (i.e. the time between the first and last doses plus 12 weeks) was ≥ 3, ≥ 6, ≥ 12, ≥ 24, ≥ 36 and ≥ 48 months in 2942 (96%), 2831 (92%), 2447 (80%), 2019 (66%), 1455 (47%) and 172 (6%) patients, respectively. The short-term analysis set included 1306 patients receiving risankizumab 150 mg (402 PY of exposure). The baseline characteristics were similar between the two analysis sets (Table 1).

General Safety Overview

The EAER for AEs in the long-term analysis set was 171 events per 100 PY (Table 2). The majority of AEs were mild to moderate in severity, did not require study discontinuation and were considered by the investigators to be unrelated to risankizumab treatment. In the short-term analysis set, EAERs were comparable between risankizumab- and placebo-treated patients (Table 2 and Table S2; see Supporting Information) and were consistent with those in the long-term analysis set. In a post hoc subanalysis of patients ≥ 65 years of age (11·7% of the analysis population), EAERs were consistent with or lower than those in the entire population, with the exception of malignancies excluding NMSC, and deaths – trends typical for this older subgroup of patients (Table S3; see Supporting Information).

Adverse Events in Areas of Safety Interest

Areas of safety interest were identified based on their higher prevalence in the population with moderate-to-severe psoriasis, customary concerns with injected immunoglobulin products, or the immunomodulatory activity of risankizumab. EAERs for TEAEs of safety interest were low for both analysis sets and generally remained consistent or decreased over time (Table 2 and Figure 1).

Figure 1.

Plaque psoriasis. Rates of adverse events (AEs), serious AEs and AEs leading to discontinuation over 6-month to 1-year intervals, calculated by events in each interval divided by total patient-years (PY) of risankizumab exposure. The error bars show the 95% confidence interval.

Infections. EAERs for infection were 90·8 and 56·4 events per 100 PY in the short- and long-term analysis sets, respectively – most commonly nasopharyngitis and upper respiratory tract infection (Table 2 and Table S2). Rates of serious infection were 1·7 and 1·2 events per 100 PY in the short- and long-term analysis sets, respectively, and for long-term analysis were comparable with placebo at 16 weeks (1·1 events per 100 PY; Table 2). Summaries of EAERs over time are provided in Figure 2. Treatment-emergent serious infections with an occurrence of at least two events in the long-term analysis set are summarized in Table S4 (see Supporting Information). Those reported most frequently (sepsis, pneumonia, cellulitis, diverticulitis and appendicitis) are those commonly reported in the general population. Rates of serious infections were consistent with those in the Psoriasis Longitudinal Assessment Registry (PSOLAR; 0·93–2·91 events per 100 PY; Figure 3).[14]

Figure 2.

Plaque psoriasis. Rate of serious infections, malignant tumours excluding nonmelanoma skin cancer (NMSC), NMSC and major adverse cardiovascular events (MACE) over 6-month to 1-year intervals, calculated by events in each interval divided by total patient-years (PY) of risankizumab exposure. The error bars show the 95% confidence interval.

Figure 3.

Plaque psoriasis. Treatment-emergent adverse events of safety interest per 100 patient-years (PY) in the 16-week and long-term risankizumab analyses. The week 16 (n = 1306; 402 PY) and long-term (n = 3072; 7927 PY) risankizumab analysis sets represent different pools of patients with varying lengths of treatment exposure included in the long-term set. Events counted in the week 16 data are also included in the long-term data. Reference data are from the Psoriasis Longitudinal Assessment Registry (PSOLAR)14, 18 (ranges shown for ustekinumab, infliximab, other biologics and nonbiologics) and the MarketScan® claims database cohort study.17 CI, confidence interval; MACE, major adverse cardiovascular event; NB, nonbiologics arm; NMSC, nonmelanoma skin cancer; PsO, psoriasis cohort.

Opportunistic Infections. The EAER for Candida infections in the long-term analysis set was 0·6 events per 100 PY (44 events; Table S5; see Supporting Information) and there were no reports of deep or systemic candidiasis. Among the 37 occurrences of herpes zoster, three were classified as serious; all of these involved a single dermatome and none required treatment discontinuation or recurred (Table S4). The EAER for herpes zoster in the long-term analysis set (0·5 events per 100 PY) was below the PSOLAR reference benchmarks of 1·0 per 100 PY (patients with psoriasis receiving systemic therapy or biologics) and 0·6 per 100 PY (patients with moderate-to-severe psoriasis).[15,16]

There were no cases of active tuberculosis in this analysis. An earlier data cut (1 September 2017) across the four pivotal phase III psoriasis clinical studies showed no development of active tuberculosis among 72 patients who had positive QuantiFERON Gold tests at baseline and concurrent risankizumab and tuberculosis prophylaxis during a mean follow-up of 61 weeks. In IMMhance, none of the 31 patients with positive QuantiFERON Gold tests at baseline who did not receive prophylaxis developed active tuberculosis during a mean follow-up of 55 weeks.

Malignancies. Rates of NMSC and malignant tumours excluding NMSC over time are summarized in Figure 2. In the long-term analysis set, there were 42 non-NMSCs (0·5 per 100 PY), most commonly those reported in the general population (colorectal, < 0·1 per 100 PY; prostate, < 0·1 per 100 PY; breast, 0·1 per 100 PY; Table 2 and Table S4). Rates were within the reference range for moderate-to-severe psoriasis reported in PSOLAR (0·48–0·84 events per 100 PY)[14] and were lower than those in the MarketScan® claims database cohort study (Figure 3).[17] There were no concerning trends in pathological type among malignancies, and no reports of lymphoma or haematological malignancy. Rates of NMSC (54 events, 0·7 per 100 PY) were lower than reported in the MarketScan® study (overall psoriasis population, 1·80 events per 100 PY; non-biologic arm, 1·99 events per 100 PY).[17] Basal cell carcinoma (BCC) was reported in 23 patients and cutaneous squamous cell carcinoma (SCC) in 14 patients in the long-term dataset (BCC : SCC ratio of 1·6 : 1).

Two cases of recurrent breast cancer were reported in the long-term analysis set: one patient continued in the study, the other withdrew. No other recurrence of malignancy was reported. A summary of serious AEs in the 10 patients with a history of malignancy excluding NMSC is provided in Table S6 (see Supporting Information).

Depression and Suicidality. Rates of depression were 1·0 and 0·7 per 100 PY in the short- and long-term analysis sets, respectively. Most were not serious, and none resulted in study drug discontinuation (Table 2). Rates of suicidal ideation and behaviour were 0·5 and < 0·1 per 100 PY in the short- and long-term analysis sets, respectively (Table 2). All patients had multiple confounding factors (prior history of depression, bipolar disorder, suicidal ideation). There were no cases of completed suicide.

Serious Hypersensitivity. Serious hypersensitivity was reported by four patients in the long-term analysis set (< 0·1 events per 100 PY; Table 2). All were considered unrelated to risankizumab and none led to study drug discontinuation. There were no reports of anaphylaxis or serum-sickness-like reactions.

Injection-site Reactions. All injection-site reactions in the long-term analysis set (256 events; 3·2 events per 100 PY) were of mild to moderate severity; none led to study drug discontinuation.

Inflammatory Bowel Disease. One patient reported worsening (moderate severity) of pre-existing ulcerative colitis on day 468. This resolved within 7 days with treatment and was not considered to be related to risankizumab.

Major Adverse Cardiovascular Events. Rates of adjudicated major adverse cardiovascular events (MACE) were 0·2 and 0·3 events per 100 PY in the short- and long-term analysis sets, respectively, and were consistent with reference rates in PSOLAR (0·51–0·64 events per 100 PY; Figure 3).[14,18] Summaries of MACE over time intervals are shown in Figure 2; further details are provided in Tables S4 and S7 (see Supporting Information).

Mortality. In total, 17 deaths, including 16 treatment-emergent deaths, were reported: two each due to sudden cardiac death, myocardial infarction and cardiac arrest; and one each due to intestinal adenocarcinoma/hepatic metastatic cancer, pancreatitis, epilepsy, seizures (subsequently clarified as death due to drug overdose), congestive heart failure, hepatic cirrhosis, and accident; the causes of the remaining four deaths were not determined (Table S7). One, death due to congestive heart failure in a 64-year-old man with multiple cardiovascular risk factors, was assessed by the investigator as possibly related to treatment. The standardized mortality ratio for treatment-emergent deaths was 0·32 (95% CI 0·18–0·53).

processing....