Efficacy and safety of rilonacept for recurrent pericarditis: results from a phase II clinical trial

Objective Recurrent pericarditis (RP) incurs significant morbidity. Rilonacept inhibits both interleukin-1 alpha (IL-1α) and IL-1β; these cytokines are thought to play a major role in RP. This phase II study evaluated rilonacept efficacy and safety in RP. Methods This multicentre, open-label study enrolled adult patients with idiopathic or postpericardiotomy RP, symptomatic (≥2 pericarditis recurrences) or corticosteroid (CS) dependent (≥2 recurrences prior). Patients received rilonacept 320 mg SC load/160 mg SC weekly maintenance in a 6-week base treatment period (TP) followed by an optional 18-week on-treatment extension period (EP) (option to wean background therapy). Results Outcomes: pericarditis pain (numeric rating scale (NRS)) and inflammation (C reactive protein (CRP)) for symptomatic patients; disease activity after CS taper for CS-dependent patients. Secondary outcomes: health-related quality of life (HRQOL), pericarditis manifestations and additional medications. 25 unique patients enrolled, while 23 completed the EP (seven colchicine failures and five CS failures). In symptomatic patients, NRS and CRP decreased; response was observed after first rilonacept dose. NRS decreased from 4.5 at baseline to 0.7, and CRP decreased from 4.62 mg/dL at baseline to 0.38 mg/dL at end of TP. Median time to CRP normalisation: 9 days. Pericarditis manifestations resolved. 13 patients on CS at baseline completed the EP; 11 (84.6%) discontinued CS, and 2 tapered; CRP and NRS remained low without recurrence. Mean HRQOL scores improved in symptomatic patients. One serious adverse event (SAE) resulted in discontinuation of rilonacept. Conclusions Rilonacept led to rapid and sustained improvement in pain, inflammation (CRP and pericarditis manifestations) and HRQOL. CSs were successfully tapered or discontinued; safety was consistent with known rilonacept safety profile. Trial registration number NCT03980522.


Changes in Concomitant Medication Use
Overall, out of 20 patients on concomitant pericarditis medications at baseline and who completed the EP, 75% of patients (n=15) successfully stopped, and 30% of patients (n=6) reduced the dose of at least one concomitant pericarditis medication by the end of the study without experiencing a pericarditis recurrence (see Table 3 in main text).

Tapering and Discontinuation of Corticosteroids and Colchicine
Prednisone (the only CS used in the study for pericarditis) was most frequently discontinued or reduced. A total of 15 patients entered the study with ongoing CS treatment, receiving prednisone for pericarditis at the mean dose 12.7 mg/day (range 1mg-50 mg/day). Of these 15 patients, 13 completed the EP (one patient in Part 1 and one CS-dependent patient in Part 3 did not enter the EP). Of the 13 patients on CS at baseline who completed the study, 11 discontinued prednisone completely (4/5 symptomatic patients, and 7/8 CS-dependent patients) and 2 tapered the dose of CS (1/5 symptomatic patients, and 1/8 CS-dependent patients) without experiencing a recurrence of signs and symptoms of RP while maintaining low average pain and CRP levels (Supplemental Figure 2). One of the patients remaining on CS at study end had reduced from 30 to 2.5 mg/day, and the other remained on 30 mg/day (from 50 mg/day), per investigator discretion for disease management during the finite study period. Of the 4 acute symptomatic patients with elevated CRP (Parts 1 and 4) on CS at baseline, 1 patient did not enter the EP; all 3 patients who completed the study stopped their prednisone without disease recurrence while on rilonacept.
In the subset of 9 CS-dependent patients, who were not experiencing acute pericarditis episodes (Parts 3 and 5), 1 patient did not enter EP, but all 8 patients who completed the study either BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Figure 2) while maintaining low average pain and CRP levels. Of these 9 CS-dependent patients, 2 patients from Part 3 presented with a pericardial effusion on echocardiogram at baseline. Pericardial effusion resolved in one patient and was assessed as trivial/physiologic in another patient at the end of the study.
In the subset of 6 CS-failure patients (patients experiencing an acute pericarditis episode at baseline while receiving CS and colchicine), 5 patients entered the EP; 4 of these patients discontinued CS during the EP.
Overall, the CS-sparing effect of rilonacept was consistent among patients with and without active recurrence at the time of enrollment, and no recurrences were observed (see Table 4 in main text).
Of 23 patients who completed the EP, 7 (30%) were experiencing an active episode while being treated with colchicine and not corticosteroids at baseline (colchicine-failure patients). One of these seven colchicine-failure patients discontinued colchicine use during the study.

Efficacy in Corticosteroid-Failure and Colchicine-Failure Patients
Of 15 patients who entered the study with ongoing CS treatment, 9 were CS-dependent, and 6 were experiencing an active episode at baseline while also receiving colchicine (CS-failure patients). CS-failure patients experienced rapid and sustained reductions in pericarditis pain and CRP with rilonacept treatment (Supplemental Figure 3). BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) The 7 patients who were experiencing an active episode at baseline while being treated with colchicine and not corticosteroids (colchicine-failure patients) experienced rapid and sustained reductions in pericarditis pain and CRP with initiation of rilonacept (Supplemental Figure 4).

Health-Related Quality of Life
A consistent pattern of increased PROMIS scores reflected improvement in HRQOL with rilonacept treatment (Supplemental Table 2). At baseline, mean Physical and Mental Global Health scores across all patients were below 50, which is the mean score for the general US population, indicating impaired QOL in symptomatic RP patients as well as CS-dependent patients without active pericarditis. In symptomatic patients of idiopathic (Parts 1 and 2) or PPS etiology (Part 4), the mean baseline Physical and Mental Global Health scores were 39.9 and 44.5, respectively, and improved to 51.3 and 50.5 at the Final Visit. In CS-dependent patients with RP (Parts 3 and 5), the mean baseline Physical and Mental Global Health scores were 43.3 and 46.5, respectively, and improved to 46.8 and 50.7, respectively, at the Final Visit. In addition, improvements in HRQOL were observed after the first 6 weeks of rilonacept treatment (Supplemental Table 2).

Exploratory Cardiac Magnetic Resonance Imaging Outcomes
Of 25 study patients, 11 had cardiac MRI at baseline and the Final Visit, including 6 patients with active idiopathic RP and 5 CS-dependent non-active RP patients (4 idiopathic, 1 PPS).
Among 8 patients with baseline pericardial DHE (mild, moderate, or severe) and follow-up MRI, DHE improved or resolved in 6 patients. For the 2 patients in which pericardial DHE was not changed, 1 patient (Part 2) had moderate pericardial DHE at baseline and the Final Visit, and the BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) in DHE were associated with decreases or maintenance of low pain and CRP levels despite discontinuation of CS or reduction in dose.

Patient Enrolled Twice
The case study of the patient who was re-treated with rilonacept for RP provides an example of the persistence and severity of RP as well as the efficacy and tolerability of rilonacept upon retreatment. This patient with severe RP participated in the study twice, having been enrolled in the study a second time with a recurrence of pericarditis approximately 4.5 months after successfully completing 6 months of rilonacept in her first participation in the trial. Retreatment with rilonacept resulted in a similar clinically meaningful response with similar tolerability, suggesting that the disease improvements represent a true response to treatment rather than a spontaneous improvement due to natural history of the disease. In addition, although limited to one patient, this example provides a framework for evaluating the efficacy and safety of repetitive use of rilonacept in RP.

PK Assessments
A total of 25 unique patients provided 211 samples for pharmacokinetic analysis. The mean rilonacept concentration time profile demonstrated moderate to high variability (30.1-69.2%CV) between patients across the duration of the study. Following the loading dose (320 mg), no BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) additional accumulation of rilonacept was observed, and the median trough concentration over the base TP and EP was approximately 30,000 ng/mL (Supplemental Figure 5).

Immunogenicity
In total, among 25 unique patients, 14 (56%) were classified as positive for ADA (anti-drug