Health-related quality of life in patients with recurrent pericarditis: results from RHAPSODY, a phase 3 study of rilonacept

Background

Recurrent pericarditis (RP) patients report that painful, debilitating flares negatively impact their health-related quality of life (HRQoL). RHAPSODY, the Phase 3 trial of rilonacept (IL-1α/IL-1β cytokine trap), included a daily pain diary and patient-reported outcome SF-36v2 to measure HRQoL throughout the trial.

Purpose

The purpose of this research is to evaluate the effect of rilonacept on HRQoL in relation to changes in pain for RP patients who have a recurrence.

Methods

RHAPSODY enrolled 86 patients with acute symptomatic RP to receive weekly rilonacept for a 12-week run-in (RI) period and randomized 61 patients (1:1) to receive placebo (n=31) or continue rilonacept (n=30) for the event-driven randomized-withdrawal (RW) period. Patients on placebo who experienced a qualifying recurrence during RW (return of pericarditis pain and increase in C-reactive protein) were rescued with bailout rilonacept. Patients reported daily pericarditis pain electronically, using a 0–10 numeric rating scale (NRS), and completed the SF-36v2 at study visits prior to clinician interaction. Scores from RI Baseline (BL), RI Week 12 (RW BL), Recurrence visit, and RW up to Week 24 (or end of study; EOS) were evaluated for patients who experienced recurrence in RW. Analyses exclude one patient randomized to placebo who had a recurrence after Week 24 of the RW period.

Results

Analyses focused on the 22 of 30 patients (73%) in the placebo group who experienced a recurrence before Week 24 of RW (median time from RW BL to recurrence: 8.6 weeks). During RI, daily pain scores decreased while on rilonacept (Cohen's effect size [ES] d=−2.0), and SF-36v2 scores improved, with scores at RI BL (Fig. 1 red line) below the general population average of 50 and near or above average at RI Week 12 (Fig. 1 blue line); ES were all large (d>0.8), ranging from 0.917 (Mental Component Summary) to 2.021 (Bodily Pain). At recurrence, pain scores increased (d=6.5; Fig. 2) and SF-36v2 scores were below the population average (Fig. 1 orange line), with largest reductions between RI Week 12 (RW BL) and recurrence for Bodily Pain (−13.4) and Physical Component Summary (−10.6). Following rilonacept bailout, average pain decreased (d=−2.1; Fig. 2), and by RW Week 24/EOS, SF-36v2 scores returned to similar levels as at the end of the RI period (Fig. 1 green line).

Conclusion

Impaired RI BL SF-36v2 scores indicate negative impact of RP on HRQOL in RP patients. While receiving rilonacept, HRQoL scores improved to near or above population averages, in conjunction with patient-reported pain. After discontinuing rilonacept during RW, HRQoL scores worsened at recurrence and improved upon receipt of bail-out rilonacept, similar to pain. These results provide support for the broader benefit of rilonacept treatment beyond pain, when administered on top of conventional therapies and as mono-therapy, providing evidence of its potential to improve HRQoL in this patient population.

Funding Acknowledgement

Type of funding sources: Other. Main funding source(s): Kiniksa Pharmaceuticals, Ltd.