Abstract
A 67-year-old man with type 1 autoimmune pancreatitis (AIP) and IgG4-related sclerosing cholangitis experienced a relapse, while presenting with pancreatic head swelling and distal bile duct stenosis. Despite an increase in the prednisolone dosage (from 7.5 to 20 mg/day), the disease worsened. A significant dilatation of the main pancreatic duct (8 mm) and new stenosis of the pancreatic tail duct were observed. Malignancy was ruled out by repeated endoscopic ultrasound-guided fine-needle aspiration biopsies. A further escalation of the prednisolone dose to 40 mg/day led to clinical and radiological improvements. This case underscores the difficulty in distinguishing a steroid-resistant AIP relapse from pancreatic cancer, and the importance of ruling out malignancy before intensifying immunosuppressive therapy.
Keywords: autoimmune pancreatitis, steroid resistance, main pancreatic duct dilatation, pancreatic cancer
Introduction
Autoimmune pancreatitis (AIP) is a unique form of pancreatitis that is characterized by lymphoplasmacytic infiltration, storiform fibrosis, and responsiveness to corticosteroids. It is now recognized as a pancreatic manifestation of IgG4-related disease and it is often accompanied by sclerosing cholangitis, retroperitoneal fibrosis, or renal involvement. This condition is classified as type 1 (lymphoplasmacytic sclerosing pancreatitis, IgG4-related) or type 2 (idiopathic duct-centric pancreatitis and granulocytic epithelial lesions, IgG4-independent), which can both be treated with corticosteroids (1). As corticosteroids are highly effective in inducing remission, the response to steroid therapy can sometimes be a decisive clue in differentiating between AIP and pancreatic cancer (2,3). Thus, it is difficult to differentiate steroid-resistant AIP from pancreatic cancer. One of the most challenging scenarios is the marked dilatation of the main pancreatic duct (MPD), which indicates pancreatic cancer or intraductal papillary mucinous neoplasm (IPMN), rather than AIP. We herein present the case of a 67-year-old man with known type 1 AIP and IgG4-related sclerosing cholangitis (IgG4-SC) who developed a steroid-resistant relapse with atypical imaging, requiring careful differentiation from multifocal pancreatic cancer.
Case Report
A 67-year-old man with a history of type 1 AIP and IgG4-SC diagnosed in May 2022 at another hospital was referred to our institution in August 2024. His medical history included diabetes mellitus, which developed after corticosteroid therapy initiation, and cough-variant asthma, which was diagnosed at the age of 65. He had no history of alcohol consumption, but he had smoked 30 cigarettes daily for 20 years.
When he was initially diagnosed in 2022, he presented with pancreatic head enlargement, bile duct stenosis, and elevated serum IgG (2,406 mg/dL) and IgG4 levels (1,361 mg/dL). Treatment with prednisolone (PSL) (30 mg/day) led to the resolution of pancreatic and biliary lesions, and maintenance therapy with PSL at a dose of 7.5 mg/day was continued without relapse. Owing to slightly elevated serum IgG and IgG4 levels, a maintenance dose of 7.5 mg/day of PSL was adopted instead of the usual 5 mg.
During his first visit to our hospital in August 2024, elevated pancreatic enzyme levels were noted (amylase, 149 U/L; lipase, 380 U/L) despite the absence of any abdominal symptoms (Table). There were also no significant changes in weight (height, 152 cm; weight, 56 kg). Computed tomography (CT) and magnetic resonance cholangiopancreatography (MRCP) performed in September 2024 revealed recurrent pancreatic head swelling and biliary stenosis, which was interpreted as a relapse of AIP (Fig. 1). These findings were not evident in a previous CT scan performed in May 2024.
Table.
Trends in Laboratory Data.
| Laboratory studies (Reference ranges) | May 2024 | August 2024 (at relapse) | December 2024 | January 2025 |
|---|---|---|---|---|
| T.Bil (0-1.0 mg/dL) | 0.2 | 0.8 | 0.5 | 1.0 |
| AST (11-33 IU/L) | 23 | 19 | 24 | 240 |
| ALT (6-43 IU/L) | 20 | 19 | 24 | 289 |
| ALP (38-113 IU/L) | 59 | 74 | 63 | 551 |
| γGTP (10-87 IU/L) | 31 | 44 | 38 | 745 |
| Amy (39-115 IU/L) | 51 | 149 | 80 | 60 |
| Lipase (11-53 IU/L) | 11 | 380 | 81 | 53 |
| IgG (mg/dL) | 1,421 | 1,382 | 999 | 1,123 |
| IgG4 (11-121 mg/dL) | not tested | 491 | 315 | 410 |
| CA19-9 (0-37 U/mL) | 15.1 | 56.1 | 39.3 | 142.3 |
Figure 1.

Before escalation of prednisolone (PSL) to 20 mg. Contrast-enhanced computed tomography (CT) and magnetic resonance cholangiopancreatography (MRCP) performed in September 2024. a: Arterial phase. Swelling in the pancreatic tail is visible as a low-density area on CT (arrows). b: MRCP revealed stricture of the main pancreatic duct and distal common bile duct (oval). The dilation of the distal pancreatic duct was mild (3 mm). The hilar bile duct appeared to be sclerotic (arrow).
The PSL dose was increased to 20 mg/day; however, the elevation of pancreatic enzymes did not improve. Azathioprine (AZT) was added to the treatment regimen in early December.
By December 2024, on PSL 15 mg/day plus AZT 50 mg/day, the levels of pancreatic enzymes had improved. The CT and MRCP examinations performed in late December to assess steroid efficacy showed slight improvement in his pancreatic head enlargement but revealed new findings: (1) marked dilatation of the MPD in the body (8 mm) and (2) localized stenosis of the pancreatic tail duct, which worsened one month later (Fig. 2). These changes are atypical for an AIP relapse, thus raising the suspicion of multifocal pancreatic cancer. Laboratory evaluation showed a decrease in serum IgG and IgG4 titers from August to December despite the progression of radiological abnormalities.
Figure 2.

Three months after the escalation of PSL to 20 mg (a, b). Four months after escalation of PSL to 20 mg (c). CT and MRCP performed in December 2024. a: Arterial phase. A slight improvement in the pancreatic head swelling was observed on CT (arrows). b: MRCP revealed marked dilatation of the main pancreatic duct in the body (8 mm in diameter) and localized stenosis of the pancreatic tail duct (arrowhead). Distal biliary strictures seemed to persist. MRCP performed in January 2024. c: The length of stenosis in the pancreatic tail duct (arrowhead) appeared to have increased.
In January 2025, an increase in hepatobiliary enzyme levels, seemingly caused by bile duct stricture, was observed. Endoscopic ultrasonography (EUS) demonstrated a hypoechoic lesion measuring 2.3 cm in the pancreatic head, a lesion measuring 1.2 cm in the tail, and irregular wall thickening of the distal pancreatic duct. The common bile duct wall was also diffusely thickened. An EUS-guided fine-needle aspiration biopsy (EUS-FNAB) was performed using a 22-gauge needle (Acquire; Boston Scientific, Marlborough, USA). Two passes were performed for both head and tail lesions. This revealed only fibrotic tissue, with no evidence of malignancy, lymphocytic infiltration, or plasmacytic infiltration (Fig. 3). The second EUS-FNAB procedure yielded the same findings.
Figure 3.

Three months after escalation of PSL to 20 mg. EUS demonstrated a 2.3-cm hypoechoic lesion (arrows) in the pancreatic head (a), a 1.2-cm lesion (arrows) in the tail (b), irregular wall thickening of the distal pancreatic duct (arrowheads) (b), and wall thickening of the common bile duct (c). EUS-FNAB revealed only fibrotic tissue without any evidence of malignancy, lymphocytic infiltration, or plasmacytic infiltration (Hematoxylin and Eosin staining, 100×) (d).
Endoscopic retrograde cholangiopancreatography (ERCP) demonstrated a narrowing of the pancreatic ducts of the head and tail, without either an abrupt or complete obstruction (Fig. 4). The bile duct was diffusely narrowed, but without complete occlusion. A biopsy of the bile duct, brushing cytology of the biliary and pancreatic ducts, and serial pancreatic juice aspiration cytological examination (4) were negative for malignancy.
Figure 4.

Four months after the escalation of PSL to 20 mg (Before escalation of PSL to 40 mg). ERCP demonstrated a narrowing of the head and tail pancreatic ducts (a: head, b: tail) without any abrupt or complete obstruction.
Given the absence of pathological findings suggesting malignancy, the absence of ductal cutoff, and diffuse bile duct wall thickening consistent with IgG4-SC, the patient was diagnosed with steroid-resistant relapse of type 1 AIP with IgG4-SC. Thus, the PSL dose was increased to 40 mg/day in late January 2025.
Two weeks later, the patient's liver function improved, and CT revealed a resolution of MPD dilatation. At the 3-month mark, MRCP showed an improvement in the pancreatic tail duct stricture. Another MRCP performed at 6 months also showed an improvement in the narrowing of the pancreatic head and distal bile ducts, thus indicating that the patient had achieved near-complete remission (Fig. 5). The PSL dose was gradually tapered over four months, after which it was transitioned to maintenance therapy (PSL 10 mg+AZT 75 mg/day). The treatment course for this case is shown in Fig. 6.
Figure 5.

Six months after escalation of PSL to 40 mg. MRCP performed in July 2025. a: The narrowing of the pancreatic head duct and distal bile duct improved. b: MPD dilatation resolved and the pancreatic tail duct stricture improved (arrow).
Figure 6.

The course of treatment for this case is shown. ①The PSL dose was increased to 20 mg/day due to pancreatic head swelling. ②The PSL dose was increased from 15 mg to 20 mg owing to elevated pancreatic enzyme levels. AZT was then added to the treatment regimen. ③The PSL dose was reduced to 7.5 mg/day in early January because the patient may have had pancreatic cancer and thus would have to undergo surgery. ④The PSL dose was increased to 40 mg/day because the patient was diagnosed with a steroid-resistant relapse of type 1 AIP with IgG4-SC.
Discussion
Patients with AIP are generally highly responsive to corticosteroid therapy. Initial treatment with 30-40 mg/day (approximately 0.6 mg/kg/day) PSL induces remission in over 95% of patients (5,6). Despite continued maintenance corticosteroid therapy, relapse occurs in 25.8% of cases within 3 years and in 35.1% of cases within 5 years (7), and is usually managed by increasing the dose of corticosteroids back to the initial level and, in some centers, adding immunosuppressants for long-term prevention (8). Because Buijs et al. reported that even lower doses (≤20 mg/day) are effective for inducing remission (9), four cases of relapse without severe symptoms were treated with the outpatient administration of PSL at a dose of 20 mg or less per day at our institution. All patients showed a good response, which makes us believe that our treatment approach to date for recurrence is appropriate. In contrast, the present case revealed a paradoxical worsening of pancreatic ductal changes despite administering 20 mg/day of PSL, which is a rare clinical course. This rare case does not lead us to advocate treating all recurrent cases with 30 or 40 mg of PSL or performing a pancreatic biopsy on all patients with recurrent cases. While such tactics may reduce misdiagnoses, they may also result in an unnecessary burden on such patients. We believe that the approach to managing relapses should be determined on a case-by-case basis, according to the severity of the condition. It is important to recognize that steroid-resistant cases, such as those described herein, do exist.
Mild MPD dilatation, generally <5 mm, is observed in AIP (10). In contrast, pancreatic cancer often causes a marked upstream ductal dilatation. Our case showed MPD dilatation of up to 8 mm, which is unusual for AIP and thus raised the suspicion of malignancy. Moreover, ductal abnormality progression after corticosteroid dose increment fortified our suspicion. A previous report noted such progression in a case of coexisting IPMN (11), however, there were no findings suggestive of coexisting IPMN in this case. Cases in which pancreatic duct dilatation worsens markedly due to steroid resistance after treatment intensification are considered rare. Therefore, the careful exclusion of malignancies is mandatory.
AIP can closely mimic pancreatic cancer, both clinically and radiographically. It has also been reported that patients with AIP have a higher risk of developing pancreatic cancer (12), thus making the differential diagnosis between the two conditions extremely important. Features favoring AIP include diffuse or segmental enlargement of the pancreas, delayed enhancement, long and smooth ductal strictures without an abrupt cutoff, and extrapancreatic IgG4-related disease (8). In our patient, diffuse bile duct wall thickening and the absence of a ductal cutoff supported the diagnosis of AIP. However, disproportionate ductal dilatation indicated malignancy. EUS-FNAB is highly useful for pathological diagnosis of pancreatic cancer and AIP. Specifically, when distinguishing between AIP and pancreatic cancer is problematic, specimen collection via EUS-FNAB is strongly recommended (8). However, EUS-FNAB has a limited ability to rule out cancer, and there is still a chance of sampling error (13). In this case, repeated negative FNABs, bile duct biopsies, and pancreatic juice cytology attenuated the suspicion of malignancy.
Immunosuppressants are frequently used to prevent relapse, particularly in patients requiring repeated or prolonged corticosteroid therapy. AZT is widely used for its maintenance (14). Given the wild-type NUDT15 genotype, continued AZT treatment was deemed safe.
In conclusion, we herein described a rare case of steroid-resistant relapse of type 1 AIP with IgG4-SC, complicated by marked MPD dilatation mimicking multi focal pancreatic cancer. The careful exclusion of malignancy allowed for safe PSL dose escalation to 40 mg/day, culminating in remission. This case highlights the importance of considering atypical steroid resistance in AIP and underscores the need for individualized relapse management strategies.
The authors state that they have no Conflict of Interest (COI).
References
- 1.Shimosegawa T, Chari ST, Frulloni L, et al. International consensus diagnostic criteria for autoimmune pancreatitis: guidelines of the International Association of Pancreatology. Pancreas 40: 352-358, 2011. [DOI] [PubMed] [Google Scholar]
- 2.Moon SH, Kim MH, Park DH, et al. Is a 2-week steroid trial after initial negative investigation for malignancy useful in differentiating autoimmune pancreatitis from pancreatic cancer? A prospective outcome study. Gut 57: 1704-1712, 2008. [DOI] [PubMed] [Google Scholar]
- 3.Kojima H, Kitago M, Iwasaki E, et al. Pancreatic cancer mimicking relapse of autoimmune pancreatitis: case reports. Surg Case Rep 11: cr-25-0076, 2025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hanada K, Shimizu A, Kurihara K, et al. Endoscopic approach in the diagnosis of high-grade pancreatic intraepithelial neoplasia. Dig Endosc 34: 927-937, 2022. [DOI] [PubMed] [Google Scholar]
- 5.Kamisawa T, Shimosegawa T, Okazaki K, et al. Standard steroid treatment for autoimmune pancreatitis. Gut 58: 1504-1507, 2009. [DOI] [PubMed] [Google Scholar]
- 6.Overbeek KA, Poulsen JL, Lanzillotta M, et al. Type 1 autoimmune pancreatitis in Europe: clinical profile and response to treatment. Clin Gastroenterol Hepatol 22: 994-1004.e10, 2024. [DOI] [PubMed] [Google Scholar]
- 7.Kubota K, Kamisawa T, Okazaki K, et al. Low-dose maintenance steroid treatment could reduce the relapse rate in patients with type 1 autoimmune pancreatitis: a long-term Japanese multicenter analysis of 510 patients. J Gastroenterol 52: 955-964, 2017. [DOI] [PubMed] [Google Scholar]
- 8.Okazaki K, Kawa S, Kamisawa T, et al. Amendment of the Japanese consensus guidelines for autoimmune pancreatitis, 2020. J Gastroenterol 57: 225-245, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Buijs J, van Heerde MJ, Rauws EA, et al. Comparable efficacy of low- versus high-dose induction corticosteroid treatment in autoimmune pancreatitis. Pancreas 43: 261-267, 2014. [DOI] [PubMed] [Google Scholar]
- 10.Takuma K, Kamisawa T, Tabata T, et al. Utility of pancreatic duct diameter in the diagnosis of autoimmune pancreatitis. World J Gastroenterol 17: 2332-2337, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hirano K, Itoi T, Sasahira N, Koyama H, Ihara F. Coexisting type 1 autoimmune pancreatitis and mixed-type intraductal papillary mucinous neoplasm. Intern Med 60: 2793-2797, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Takikawa T, Kikuta K, Sano T, et al. Maintenance steroid therapy is associated with decreased risk of malignancy and better prognosis of patients with autoimmune pancreatitis: a multicenter cohort study in Japan. Pancreatology 24: 335-342, 2024. [DOI] [PubMed] [Google Scholar]
- 13.Kanno A, Miwata T, Nagai H, et al. Endoscopic ultrasound-guided pancreatic sampling for the histopathological diagnosis of autoimmune pancreatitis. Dig Endosc 34: 420-427, 2022. [DOI] [PubMed] [Google Scholar]
- 14.Masaki Y, Nakase H, Tsuji Y, et al. The clinical efficacy of azathioprine as maintenance treatment for autoimmune pancreatitis: a systematic review and meta-analysis. J Gastroenterol 56: 869-880, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
