Two Autoimmune Pancreatitis Diagnoses, Two Different Management Questions
Type 1 and Type 2 autoimmune pancreatitis are both steroid-responsive by definition — the pivot here isn't which one, it's how differently that shared label plays out once organ involvement, relapse risk, and a patient's own other diagnoses enter the picture.
Efrain M. spent thirty-one years sorting mail before retiring, and it was his daughter, noticing his eyes had turned yellow at a family dinner, who pushed him to see someone rather than wait it out. He'd lost nearly fifteen pounds over two months without trying, and painless jaundice with dark urine brought him in for the workup that followed. CT showed a diffusely enlarged, featureless "sausage" pancreas, and MRCP showed a stricture of the proximal extrahepatic bile duct, above the pancreas rather than running through it — IgG4-related sclerosing cholangitis, a biliary complication of the same disease process rather than simple compression by a swollen pancreatic head. His serum IgG4 came back at four times the upper limit of normal, and an EUS-guided core biopsy confirmed dense lymphoplasmacytic infiltrate with storiform fibrosis and obliterative phlebitis: Type 1 autoimmune pancreatitis, the IgG4-related form.
Type 1 AIP is reliably steroid-responsive, and nobody on the team doubts he'll improve on a taper. What's actually in front of them is a real, documented risk factor sitting in his own imaging, and its location is the whole of the argument. Proximal — extrapancreatic — biliary stricture is one of the few reproducible predictors of relapse in Type 1 disease: Sah et al. (Gastroenterology, 2010) put the hazard ratio at 2.12, and Hart et al.'s 1,064-patient international analysis (Gut, 2013) found relapse concentrated in exactly this group, against a background Type 1 relapse rate somewhere between a third and a half after steroids alone. Roughly half of Type 1 patients have instead an intrapancreatic narrowing, which carries no comparable signal. His is the other kind. He is 68, with no other autoimmune diagnoses and no prior pancreatitis episodes — a first presentation, but not a low-risk one by the registry's own criteria.
His bilirubin, drawn on admission, came back at 4.8 mg/dL, high enough that the team briefly discussed biliary stenting before deciding steroids alone were likely to resolve the stricture without it, given how classically his imaging fits active Type 1 disease rather than a fixed, fibrotic narrowing — though with a proximal stricture that call was closer than it would have been for an intrapancreatic one, and stenting stays on the table if his bilirubin doesn't fall on steroids. His daughter has stayed at every appointment since the diagnosis, and it was her account of his gradually duller appetite over the two months before diagnosis — a detail he'd mentioned only in passing to her, not to any physician — that helped his gastroenterologist date the actual onset more precisely than his own recollection alone would have.
I'd start prednisone alone and hold azathioprine in reserve. Not every patient with a biliary stricture goes on to relapse, and starting a steroid-sparing agent on everyone who fits that profile means immunosuppressing a real number of patients who would have done fine on steroids alone.
I'm not proposing azathioprine for every Type 1 patient — I'm proposing it for this one, because he already has the specific feature the international registry data (Hart et al., Gut, 2013; Sah et al., Gastroenterology, 2010) identified as a major relapse predictor: proximal, extrapancreatic biliary involvement, not the intrapancreatic narrowing most Type 1 patients present with. That's not a policy applied blindly to a category, it's a targeted decision based on a risk factor already sitting in his own imaging.
Agreed: prednisone taper with azathioprine started upfront given his documented relapse risk factor, not withheld until a first relapse.
Not agreed: whether this approach should extend to future Type 1 patients without his specific biliary finding. The gastroenterologist maintains steroid-alone induction as the right default absent a named risk factor; today's decision was scoped to Efrain specifically, not adopted as a new standing rule.
Efua N. has managed ulcerative colitis for five years, currently well-controlled on scheduled infliximab infusions that let her keep working full time as a public-health nurse — a job she took, in part, because her own diagnosis made population-level chronic disease management feel personal. The epigastric pain that brought her to the emergency department three days ago was new, sharp, and unrelated to any flare pattern she recognized from her colitis, and her lipase came back at five times the upper limit of normal. Unlike Efrain, there is no jaundice, no biliary obstruction, and her CT shows focal, not diffuse, pancreatic enlargement.
Her serum IgG4 is normal, and an EUS-guided biopsy shows a duct-centric neutrophilic infiltrate with granulocytic epithelial lesions — the defining histology of Type 2 AIP, a distinct disease from Efrain's, despite sharing the same broad diagnostic label. Type 2 has no serologic marker of its own, carries none of Type 1's extrapancreatic organ involvement, and roughly a third of Type 2 patients, like Efua, have inflammatory bowel disease — most often ulcerative colitis. She is already on a systemic immunosuppressant for that same underlying immune susceptibility, which is exactly what makes her management question genuinely different from Efrain's: not whether steroids work, but whether they're even the right next step in a patient already immunosuppressed for a related condition.
Her infliximab levels, checked as part of the acute workup, are therapeutic — the drug is doing its job for her colitis, which was already quiet on interview with no recent bowel-symptom change to suggest a concurrent flare masking itself as this new pain. That detail matters to how the team reads the pancreatitis: it argues against her presentation being a colitis flare with incidental pancreatic involvement, and toward a genuinely separate, second autoimmune process arising independently of whether her IBD is currently active or controlled.
She needs her own dedicated steroid course for the pancreatitis. Type 2 AIP is defined by its steroid response, and there's no trial evidence that infliximab — indicated and dosed for her ulcerative colitis — reliably treats active Type 2 AIP. Assuming it does because both conditions involve the same underlying immune susceptibility would be extrapolating past what's actually been studied.
I agree there's no trial showing infliximab treats active AIP directly — that's not in dispute. But she's already systemically immunosuppressed, and adding steroids carries real, added infection risk in that setting. Unlike Efrain, she has no jaundice and no biliary obstruction — a lower-acuity presentation that gives us real room to optimize her existing infliximab dosing and watch closely before committing to a second immunosuppressive course.
Steroids remain the only actually-studied induction therapy for Type 2 AIP, so I don't think watchful waiting on an unstudied assumption is the safer choice here — untreated active pancreatitis has its own real cost. But her elevated infection-risk baseline is a legitimate reason to calibrate the course, not skip it: a shorter taper than we'd use in a non-immunosuppressed patient.
Type 2's own relapse rate is meaningfully lower than Type 1's — roughly 10 to 20 percent against Type 1's near 50 — so unlike Efrain, she doesn't need a steroid-sparing agent layered on for long-term relapse prevention. This is a bounded course, not an open-ended addition to a regimen that's already carrying real immunosuppressive weight.
Agreed: a shortened prednisone taper for the pancreatitis itself, infliximab continued unchanged for her colitis, no additional long-term immunomodulator added.
Not agreed: how closely to monitor for infection during the overlap period. The rheumatologist wants scheduled infection surveillance labs through the taper; the gastroenterologist considers her risk adequately covered by routine clinical follow-up alone, given the taper's short intended duration.