Relapsing Autoimmune Pancreatitis: Another Steroid Taper or a Switch to Rituximab?
Two clean steroid responses, two relapses once the taper finished — the question is whether a third round of the therapy that has already stopped holding is still the right first move.
Gordon L. retired from coaching high-school football two years ago, not long before the jaundice and painless weight loss that first brought him to a gastroenterologist and, eventually, to a diagnosis of Type 1 autoimmune pancreatitis — the "sausage-shaped" pancreas on CT, an elevated serum IgG4, and biopsy-confirmed lymphoplasmacytic infiltrate with storiform fibrosis all pointing the same direction. He responded completely to his first steroid taper eighteen months ago, only to relapse four months after stopping it, with the same imaging pattern and a repeat IgG4 elevation. A second taper, six months ago, produced the same clean response and the same story: symptoms and labs normalized, and within ten weeks of finishing the taper, both relapsed again.
He is back in clinic today with the same picture a third time — mild jaundice returning, IgG4 climbing again, imaging showing the pancreas re-thickening in the same distribution as before. Two relapses is a real pattern, not two isolated bad breaks: each time steroids have worked completely while he's on them and failed to hold once he's off them, which is a different clinical question than whether steroids work at all. The team is deciding not whether he needs treatment — his labs and imaging both say he does — but whether the honest lesson of two identical relapses is to treat this one the same way a third time, or to change what "induction" actually means for him going forward.
His most recent DEXA scan, obtained after his second taper specifically because his gastroenterologist wanted a baseline before considering a third course, already shows osteopenia in his lumbar spine — not severe, but a real, measured finding rather than a theoretical worry about steroid exposure. He is otherwise active, still refereeing youth football games most weekends, and has no other symptoms suggesting broader IgG4-related disease has developed since his diagnosis: no salivary gland swelling, no retroperitoneal findings on this admission's imaging, no new kidney involvement on his labs. His relapse pattern, in other words, has stayed confined to the pancreas and bile duct each time, which is itself a piece of information the team is weighing alongside the relapse count.
Clinic, facing a second relapse
I'd treat this with a third steroid taper. He's had a complete response both previous times — steroids are the best-characterized induction therapy for Type 1 AIP, and reaching for a biologic on a second relapse is a bigger step than the evidence for his specific case yet requires.
He's right that steroids have worked cleanly every time he's on them — that was never in question. What's changed is that this is now a documented relapsing pattern, not a single episode, and Hart and colleagues' Mayo Clinic cohort of relapsing Type 1 AIP patients (Gut, 2013) found rituximab induced remission effectively and, unlike immunomodulator therapy alone, kept most patients relapse-free without needing repeated steroid courses in exactly this population. His DEXA already shows osteopenia partly attributed to cumulative steroid exposure — cycling him through a third taper treats this relapse as if it were his first, when his own history says it isn't.
I'd name a real option neither of you has raised: adding azathioprine as a steroid-sparing agent at this induction, rather than choosing between another steroid-alone taper and a B-cell-depleting biologic. It's a genuinely smaller step than rituximab, and it's one he hasn't actually tried — going straight to a biologic on the strength of two relapses skips over an intermediate option that might change his trajectory with less immunosuppressive weight than either alternative.
The honest gap in that argument is that we don't have head-to-head relapse-rate data for azathioprine-plus-steroid against rituximab in a relapsing population the way the rheumatologist just cited for rituximab specifically — this is a reasonable middle step, not a better-evidenced one.
Agreed: rituximab induction with a short bridging steroid taper, given his now-documented relapsing pattern and the bone-density cost of a third full steroid course.
Not agreed: whether azathioprine should have been tried before escalating to a biologic. The pharmacologist would have preferred exhausting that intermediate step first; the rheumatologist's relapse-rate data carried the room, but the disagreement about sequencing wasn't resolved, only outvoted for this decision.