Clinical Cases in Pharmacology Clinical Cases  ·  Cardiovascular  ·  Pericardial Disease  ·  Recurrent Pericarditis After Colchicine Failure: When to Start an IL-1 Blocker
Cardiovascular, Case 0133 — Pericardial Disease

Recurrent Pericarditis After Colchicine Failure: When to Start an IL-1 Blocker

Colchicine has now failed to prevent recurrence twice, and the steroid courses used to control each flare are showing up in her glucose and bone density — not just her chest pain history.

Abbreviations, terms, and other agents mentioned in this case CRP — C-reactive protein  ·  DEXA — dual-energy X-ray absorptiometry (bone density scan)  ·  IL-1 — interleukin-1  ·  TPMT — thiopurine methyltransferase  ·  NUDT15 — nudix hydrolase 15
Presentation

J.T., a 34-year-old woman, teaches seventh-grade science, a job that does not tolerate a chest-pain flare well — this is her fourth episode of pericarditis in fourteen months, and each one has meant days of missed classes and a substitute scrambling with lesson plans she didn't write. She was previously healthy, without any cardiac or rheumatologic history, when her first episode began after a viral illness just over a year ago. Colchicine controlled that episode and the one after it, but the third flare broke through colchicine prophylaxis entirely, and this fourth one has too — pleuritic chest pain, a friction rub, and CRP rising to 48 while she was still taking her colchicine as prescribed.

What has changed isn't just the recurrence count. Since the third flare broke through, she has been through two separate courses of prednisone, neither of which she could taper off without relapsing, and she remains on a low maintenance dose today. Her most recent labs show what that has cost her: a fasting glucose of 118, newly abnormal, and a DEXA scan flagging early osteopenia in a woman not yet 35. She is, by any reasonable reading, steroid-dependent — unable to complete a taper without another flare — and the metabolic and bone findings are exactly what continued steroid exposure predicts if that pattern continues. The question in front of the team is not whether her current approach is working; a fourth breakthrough episode says it isn't. It's whether to move to a biologic that trial evidence suggests could actually change that pattern, and how much steroid exposure she has to accept while waiting for it to become available.

J.T. · 34 4th Episode (3rd Recurrence), 14 Months
History
Previously healthy; first pericarditis episode 14 months ago, three recurrences since
Current therapy
Colchicine (adherent, confirmed); low-dose maintenance prednisone after two failed tapers
Labs
CRP 48 mg/L; fasting glucose 118 (new); DEXA: early osteopenia
Prior response
Colchicine alone controlled episodes 1–2; broke through starting with episode 3
Exam
Pleuritic chest pain, three-component friction rub
Coverage
Employer plan; prior authorization typically required for biologic therapy

In clinic, after the fourth recurrence

Cardiologist Opening

Four episodes in fourteen months — three recurrences, breaking through colchicine twice now — is the definition of the population rilonacept was studied in. RHAPSODY enrolled patients with at least two prior recurrences who were colchicine-resistant or steroid-dependent, which is exactly where she sits, and it randomized patients with this pattern and found a dramatic reduction in further recurrence compared to placebo, not a marginal one. I think we should be initiating a prior authorization today, not treating it as a next-step-if-this-fails option.

If this were her second episode rather than her fourth, or if colchicine were still holding, I wouldn't be reaching for a biologic at all — colchicine remains genuinely first-line and effective for most patients. The case for rilonacept here is built entirely on the fact that the conventional pathway has now failed twice in a row.

Endocrinologist Response

I agree with where you're heading, but I want the glucose and DEXA findings treated as their own urgent problem, not just supporting evidence for the rilonacept argument. A fasting glucose of 118 in a 34-year-old with no prior metabolic history, appearing after two steroid courses, is early impaired fasting glucose that will likely progress if she goes through a third taper. Whatever the timeline on rilonacept access turns out to be, minimizing further steroid exposure during that wait has to be an explicit part of the plan.

I don't think the metabolic findings change the sequencing decision on their own — if rilonacept access takes months, as it realistically might, she may still need some bridging anti-inflammatory therapy for an active flare regardless of what it costs her. The argument is for minimizing dose and duration during that bridge, not for refusing to use steroids at all.

Clinical Pharmacologist Final

The RHAPSODY data support moving to rilonacept now, and I'd frame the access question directly rather than let it happen by default: prior authorization for IL-1 blockade in recurrent pericarditis is often approved once colchicine failure and steroid-dependence are documented as clearly as they are here, but it isn't instant, and screening for latent infection needs to happen before the first dose regardless of timeline. Worth being precise about what rilonacept is, since it gets loosely grouped with anakinra: it is a dimeric fusion protein that acts as a soluble decoy receptor, trapping circulating IL-1α and IL-1β before they reach the receptor, rather than an IL-1 receptor antagonist that blocks the receptor itself. For the interim, azathioprine as a steroid-sparing agent has real, if less dramatic, evidence in recurrent pericarditis and lets us reduce her maintenance prednisone now, rather than waiting for rilonacept approval to start addressing the glucose and bone findings. Azathioprine needs TPMT and NUDT15 status checked before the first dose, though — deficient activity in either enzyme means severe, potentially fatal myelosuppression at ordinary doses, and that is a pre-treatment test, not a monitoring parameter.

Regimen selected
Rilonacept
IL-1α/IL-1β Trap (soluble decoy receptor fusion protein) · Prior authorization initiated
Directly targets the pattern RHAPSODY was designed to test; the clearest evidence-based option for colchicine-refractory recurrent pericarditis.
Azathioprine
Steroid-Sparing Immunosuppressant · Started as bridge, after TPMT/NUDT15 testing
Allows prednisone tapering to begin now instead of waiting for biologic approval, addressing the glucose and bone findings without stopping anti-inflammatory coverage.
Prednisone (tapering)
Corticosteroid · Dose reduced, tapering
Continues only as a bridge at the lowest dose that controls symptoms while azathioprine takes effect.
Colchicine (continued)
Anti-Inflammatory · Unchanged
Kept in the regimen; broken through at recurrence but still contributes background coverage and is not implicated in her metabolic findings.
Prednisone at Prior Maintenance Dose — Ruled Out
Corticosteroid, unchanged dose · Considered, not adopted
Would avoid a taper-related flare in the short term but extends exactly the exposure her glucose and DEXA findings argue against.
Where this was left

Agreed: rilonacept prior authorization submitted today, azathioprine started now as a steroid-sparing bridge, and prednisone tapering to the lowest dose that holds her stable rather than staying at its current maintenance level.

Not fully settled, because nobody can promise how long the biologic approval will take:

If rilonacept is approved within several weeks

Azathioprine and the low-dose steroid bridge were a short, defensible cost for reaching a therapy with real trial evidence behind it.

If approval is delayed months

The team will need to revisit whether azathioprine alone is holding her, and whether a fifth episode changes the urgency case for prior authorization.

Endocrinology will recheck her fasting glucose and repeat DEXA in six months regardless of which branch this falls into.

Educational content only — a composite teaching case, not a real patient encounter or a substitute for clinical guidance. About These Cases →