Granulomatosis With Polyangiitis (Wegener Granulomatosis) Presenting as a Pancreatic Mass: A Rare Malignancy Mimic

Introduction

Granulomatosis with polyangiitis (GPA), formerly known as Wegener granulomatosis, is a rare autoimmune disorder characterized by necrotizing granulomatous inflammation and vasculitis of small- to medium-sized vessels. GPA typically affects the upper and lower respiratory tract and kidneys, with anti-neutrophil cytoplasmic antibodies (ANCAs), particularly anti-proteinase 3 (PR3), also known as cytoplasmic anti-neutrophil cytoplasmic antibody (C-ANCA), serving as key diagnostic markers. Gastrointestinal involvement is found to be rare among GPA cases (approximately 5–10%); pancreatic manifestations are exceedingly rare and remain sparsely reported in the medical literature [1,2].

When GPA presents as a pancreatic mass, it poses a formidable diagnostic challenge due to its close clinical, radiologic, and even cytological resemblance to pancreatic adenocarcinoma. Reported cases include presentations with acute pancreatitis, pancreatic pseudocysts, and pancreatic masses initially suspected to be neoplastic [3]. Early recognition is essential to prevent unnecessary surgical intervention and to initiate timely immunosuppressive therapy.

The case we present is unique in that the pancreatic lesion was the primary and initial finding in a patient later diagnosed with C-ANCA-positive GPA. The lesion was initially reported as a metastatic well-differentiated adenocarcinoma on fine-needle aspiration (FNA) cytology from pancreas and level IB cervical lymph nodes, leading to the administration of 1 cycle of chemotherapy. Only upon further evaluation – including renal biopsy, immunologic profiling, and repeated cytopathological analyses from the pancreatic mass and submandibular lymph nodes – could the diagnosis of GPA be established. This case underscores the importance of integrating immunologic evaluation – particularly C-ANCA and complement assays—into the diagnostic work-up of pancreatic lesions with systemic features. Future studies should further define when vasculitic disorders should be considered in the differential diagnosis of pancreatic masses.

Case Report

A 62-year-old man with a background of seasonal bronchial asthma presented in April 2025 with intermittent low-grade fever, cough, hemoptysis and progressive abdominal bloating. He was initially evaluated at a tertiary care center elsewhere. Initial blood reports showed mild normocytic normochromic anemia (Hb 11.1 g/dL), normal leukocyte count (7.73×103/μL) with neutrophil predominance (68%), eosinophilia (10%), and normal platelets. The inflammatory markers erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were markedly raised (ESR 103 mm/h, CRP 84.5 mg/L) with low procalcitonin (0.10 ng/mL). Liver function tests were within normal limits – bilirubin 0.29 mg/dL, aspartate aminotransferase 20 U/L, alanine aminotransferase 19 U/L, alkaline phosphatase 79 U/L, lactate dehydrogenase 188 U/L, albumin 3.7 g/dL. Renal function was normal except for mildly raised urea (29 mg/dL); electrolytes were normal. Tumor markers were unremarkable (CA 19-9: 19.59 U/mL; CEA: 1.3 ng/mL). Chromogranin A was mildly elevated (108.8 ng/mL), IgG4 was normal (0.999 g/L), and QuantiFERON-TB Gold was negative. Fasting insulin (18.8 mU/L) and C-peptide (3.10 ng/mL) were within reference ranges. Blood cultures remained sterile. Laboratory investigations showed elevated CRP and renal dysfunction, while tumor markers (CEA and CA19-9) were within normal limits, arguing against malignancy. A contrast-enhanced CT of the thorax was performed, which revealed a right upper lobe pulmonary nodule. A subsequent fluorodeoxyglucose (FDG) positron emission tomography-computed tomography (PET-CT) scan demonstrated a metabolically active lesion in the pancreatic body and tail measuring 49×51 mm with a maximum standardized uptake value (SUVmax) of 8.53; metabolically active right pulmonary hilar lymph node (SUVmax 5.8); right-side, level IB cervical (submandibular) lymphadenopathy (SUVmax 9.7); and multiple low-grade FDG-avid pulmonary nodules. These findings were concerning for pancreatic malignancy with nodal and pulmonary metastases. Later on, endoscopic ultrasound (EUS) revealed a 5×4.8 cm mass with calcifications in the pancreatic body and tail. FNA of the pancreatic lesion was performed and showed well-formed, fused, and angulated ductal structures lying within a remarkably desmoplastic stroma with a prominent infiltrate of eosinophils and neutrophils. Mild to moderate anisonucleosis was seen. No mitotic figure was identified. Normal pancreatic parenchyma was also included in the biopsy, indicative of well-differentiated ductal adenocarcinoma. Additionally, a cervical lymph node core needle biopsy was performed, which showed desmoplastic stroma with neutrophils, eosinophils, and histiocytes along with small glands/acini lined by cells with moderate cytoplasm and round to oval nuclei. CK7 was positive for atypical epithelial cells, CK19 was weakly positive, CK20 was negative, and CA19.9 was luminally positive in atypical epithelial cells. After evaluating all of these parameters, a diagnosis of metastatic pancreatic adenocarcinoma was made. Based on this, the patient received 1 cycle of gemcitabine-based chemotherapy.

For further evaluation and management, the patient was referred to our institute. On referral to our institute, after evaluation of the previous history and reports, despite cytologic interpretation suggestive of adenocarcinoma, several discordant clinical features prompted reconsideration of the diagnosis, including normal CA19-9 and CEA levels, prominent systemic inflammatory symptoms, eosinophilia, pulmonary involvement, and evolving renal dysfunction. These findings were atypical for metastatic pancreatic adenocarcinoma, raising concern for a systemic inflammatory or vasculitic process rather than primary malignancy. Further evaluation was therefore initiated involving a multidisciplinary team. On evaluation, hemoglobin was 10.2 g/dL, total leukocyte count was 3750/mm3, platelet count was 327 000/mm3, serum creatinine was 1.30 mg/dL, and a urine routine test showed plenty of red blood cells. In view of multiorgan involvement and atypical findings, an autoimmune and vasculitis work-up was undertaken, keeping small vessel vasculitis as a differential diagnosis. Serologic testing revealed strongly positive PR3/C-ANCA antibodies (1034 cu) with negative anti-myeloperoxidase (P-ANCA). Complement levels were evaluated and were found to be borderline high (C3: 2181 mg/L, C4: 463 mg/L). IgG4 was 60 mg/dL (10–120 mg/dL).

A renal biopsy was performed, and demonstrated pauci-immune focal crescentic glomerulonephritis in view of the supportive C-ANCA-positive status (Figure 1A, 1B). The presence of PR3-ANCA positivity, along with pauci-immune crescentic glomerulonephritis on renal biopsy, was pivotal in distinguishing GPA from a neoplastic pathology.

Given the significant clinical–pathologic discordance, the potential morbidity of pancreatic resection, and the emergence of multisystem involvement suggestive of vasculitis, a decision was made by the multidisciplinary team to pursue repeat tissue sampling rather than proceeding directly to oncologic therapy or surgery. Repeat EUS-FNA of the pancreatic mass performed at our center revealed mainly benign pancreatic acini, few duct fragments, and focal areas of acute inflammation. No atypical cell was identified nor any features suggestive of IgG4-related disease, including absence of storiform fibrosis, obliterative phlebitis, or IgG4-positive plasma cell infiltrates (Figure 2A, 2B). Together, this all was suggestive of a benign pancreatic aspirate with no evidence of malignancy or autoimmune pancreatitis. The immunologic profile, in conjunction with benign cytology on repeat EUS-FNA, led to a multidisciplinary decision to forgo surgical intervention in favor of medical therapy.

The patient had a persistent submandibular swelling, from which FNA cytology cell blocks had been previously taken. These were reviewed in view of the new FNA cytology findings from the pancreatic space-occupying lesion, and found to be inconclusive. In light of this, the patient underwent excisional biopsy of the neck swelling. However, during the procedure no lymph node was found; in contrast, the submandibular gland was found to be swollen and it was excised for diagnostic confirmation. The biopsy report was suggestive of inflammatory changes without any evidence of malignancy (Figure 3).

CT angiography of the abdomen conducted 4 days later showed an ill-defined hypodense lesion in the pancreatic tail (13×18 mm) with preserved fat planes and no evidence of any sausage-shaped pancreas. The lesion was found to be reduced in size in comparison with the previous scan by the referring hospital that initially diagnosed the metastatic cancer. A thin rim of hyperdense collection was found in the left perinephric space, suggestive of inflammation or vasculitis rather than malignancy (Figure 4). Microbiological studies, including bronchoalveolar lavage, sputum, and blood cultures, were negative for tuberculosis, fungi, and other infectious agents. Tumor markers including CA19-9 were within normal limits (10.6 U/mL).

With histological confirmation of vasculitis and serological positivity for C-ANCA, a diagnosis of GPA was established, with rare pancreatic involvement mimicking adenocarcinoma. The unusual presentation of GPA as a pancreatic mass – typically suggestive of adenocarcinoma – highlights a critical diagnostic pitfall with significant therapeutic implications. The patient was started on intravenous rituximab and systemic corticosteroids (prednisolone 30 mg daily). Following treatment, the patient showed significant clinical improvement, resolution of fever, stabilization of renal parameters with a significant decrease in red blood cells in urine, and was discharged in stable condition after 7 days with outpatient follow-up.

The patient had bleeding per rectum after a month and on evaluation was found to have cytomegalovirus colitis and was started on ganciclovir for 3 weeks, which resulted in complete resolution of infection. After 6 months of follow-up, the patient has been doing well. A further follow-up CT angiography of the abdomen (Figure 5) showed loss of lobulations along the pancreatic tail region. Compared with the previous scan, the ill-defined hypodense lesion is no longer seen. No evidence of peripancreatic fat stranding was noted in the follow-up scan, suggestive of near-complete resolution of the pancreatic lesion, accompanied by clinical improvement. Prior to evaluation at our center, the patient had received only a single cycle of chemotherapy, making a sustained chemotherapeutic response unlikely, as demonstrated on the CT scan performed at our institution. Subsequent regression of the pancreatic lesion following initiation of corticosteroid therapy further supported an inflammatory rather than a neoplastic etiology.

Discussion

RADIOLOGICAL OVERLAP BETWEEN PANCREATIC ADENOCARCINOMA AND GPA:

Radiologic imaging remains central to evaluating pancreatic masses, with contrast-enhanced CT and FDG PET-CT serving key diagnostic roles. However, these modalities cannot reliably differentiate pancreatic malignancy from inflammatory or vasculitic lesions such as those due to GPA. Both may appear as hypodense pancreatic lesions with irregular margins, regional lymphadenopathy, or vascular encasement – features typical of pancreatic ductal adenocarcinoma (PDAC) [12,13].

FDG PET-CT, though valuable for detecting hypermetabolic malignancies, can yield false-positive results in cases of active vasculitis due to glucose-avid inflammatory infiltrates of macrophages and neutrophils [14,15]. In our case, the pancreatic tail lesion showed an SUVmax of 8.53, within the usual range for PDAC. Comparable uptake has been reported in GPA-related pancreatic lesions described by Valerieva et al and Castillo et al, where FDG-avid foci proved inflammatory rather than malignant [1,3].

Metabolically active hilar and right cervical lymphadenopathy further mimicked metastatic spread. Isolated right cervical lymph node involvement is exceptionally rare in pancreatic adenocarcinoma [16,17] but may occur in GPA as reactive or granulomatous disease. Moreover, PET patterns in GPA often demonstrate multisystem uptake – including pulmonary, renal, or salivary involvement – which, when correlated with systemic features, should alert clinicians to a vasculitic rather than malignant process [15,18].

These observations underscore the limitations of relying solely on radiologic criteria in distinguishing malignancy from mimickers like GPA. Integration of imaging findings with clinical symptoms, immunologic assays (especially ANCA), and histopathologic confirmation from multiple sites is essential to avoid misdiagnosis and unnecessary oncologic intervention.

According to the joint clinical practice guidelines of the European Society for Medical Oncology and the European Society of Digestive Oncology (ESMO-ESDO), preoperative histologic confirmation of malignancy is not mandatory in patients undergoing surgery with curative intent when imaging and clinical findings strongly suggest pancreatic cancer [19]. The National Comprehensive Cancer Network similarly recommends that surgery should not be delayed for a negative biopsy if the clinical suspicion remains high, though histologic diagnosis is mandatory before neoadjuvant or palliative chemotherapy [20]. Consequently, some patients undergo pancreatic resection for lesions that ultimately prove benign or inflammatory, particularly when vasculitic or autoimmune processes such as GPA closely mimic malignancy. This approach, while guideline-concordant, raises ethical and clinical concerns given the substantial morbidity and mortality associated with pancreatic surgery. In the UK, postoperative mortality following pancreaticoduodenectomy ranges from 3–5%, with morbidity rates reaching 35–50%, while distal pancreatectomy carries a mortality rate of 2% and morbidity of approximately 35% [21]. In this context, our case highlights the potential consequences of following guidelines without factoring in systemic inflammatory conditions.

DIAGNOSTIC REASONING:

In our case, diagnostic uncertainty was compounded by cytopathological overlap. The initial EUS-FNA showed well-differentiated ductal structures in a desmoplastic background, suggesting adenocarcinoma. GPA may mimic adenocarcinoma cytologically due to prominent desmoplastic stromal reaction, ductal hyperplasia, inflammatory atypia, and eosinophil-rich infiltrates, which can be misinterpreted as malignant glandular architecture. Awareness of these pitfalls is essential for cytopathologists when evaluating pancreatic aspirates in patients with systemic inflammatory features. Similar pitfalls have been reported, where patients underwent major pancreatic resections such as pancreaticoduodenectomy or distal pancreatectomy before histopathology confirmed GPA [1,4,5]. Valerieva et al described a distal pancreatectomy performed for a presumed tail malignancy that was ultimately diagnosed as GPA [1], while Reddy et al reported a young woman who underwent Whipple’s surgery for suspected carcinoma, later reclassified as GPA despite ANCA-negative status [4].

The clinical implications of a presumed pancreatic malignancy are profound, often influencing both medical decision-making and patient morale. Our case was distinctive in that the diagnosis of C-ANCA (PR3)-positive GPA was established without surgical resection or further chemotherapy. The presence of systemic features – renal involvement (pauci-immune crescentic glomerulonephritis), low-grade fever, and weight loss – along with serologic and imaging findings supported a vasculitic process. Repeat EUS-FNA of the pancreatic lesion and excisional biopsy from the submandibular gland revealed only inflammatory changes, further steering suspicion away from malignancy. The patient’s marked clinical and radiologic improvement with immunosuppressive therapy confirmed the diagnosis of vasculitic GPA.

ROLE OF IMMUNOLOGIC PARAMETERS: C-ANCA (ANTI-PR3):

C-ANCA, directed against PR3, is a hallmark serologic marker of GPA and can be detected in nearly 90% of patients with active systemic disease. In our patient, a strongly positive anti-PR3 antibody (1034 CU) was pivotal in establishing the diagnosis of GPA when repeated pancreatic and nodal biopsies failed to confirm malignancy. Similar diagnostic redirection through ANCA positivity has been documented by Valerieva et al and Castillo et al, despite misleading radiologic and cytologic findings [1,3]. Although ANCA testing is not routinely included in the evaluation of pancreatic masses, its consideration in cases with systemic features or incongruent pathology can prevent misdiagnosis and unwarranted chemotherapy or surgery, as reported in multiple cases [1,3,4,6].

ROLE OF IMMUNOLOGIC PARAMETERS: COMPLEMENT LEVELS (C3, C4):

Complement assessment can aid in differentiating systemic vasculitis. In GPA, complement levels are typically normal or slightly elevated, contrasting with the hypocomplementemia of immune complex–mediated vasculitis such as lupus. Our patient’s borderline-high C3 (2181 mg/L) and C4 (463 mg/L) levels supported a pauci-immune mechanism consistent with GPA rather than complement-mediated disease.

ROLE OF IMMUNOLOGIC PARAMETERS: SERUM IGG4:

IgG4-related disease represents an important differential diagnosis for pancreatic masses, often mimicking carcinoma or GPA. It is characterized by lymphoplasmacytic infiltration, storiform fibrosis, and elevated serum IgG4 (>135 mg/dL), although up to 30% of histologically proven cases show normal levels. In our case, serum IgG4 was normal (60 mg/dL), effectively ruling out autoimmune pancreatitis. Distinguishing autoimmune pancreatitis from GPA is crucial, as both may present with pancreatic enlargement and systemic involvement but require distinct therapeutic strategies. Unlike autoimmune pancreatitis, which typically presents with diffuse ‘sausage-shaped’ pancreatic enlargement, elevated serum IgG4, and characteristic storiform fibrosis, obliterative phlebitis, or IgG4-positive plasma cell infiltrates, GPA-related pancreatic involvement more often manifests as focal mass-forming lesions, ANCA positivity, multisystem involvement, and pauci-immune histology. Differentiating these entities is critical, as management strategies differ substantially.

IMPORTANCE OF A SYSTEMIC APPROACH:

Our case exemplifies the need for a multidisciplinary approach when evaluating a pancreatic mass with discordant findings. Repeated EUS-FNA, immunohistochemical stains, systemic inflammatory markers, and focused autoimmune panels – particularly C-ANCA – proved essential in avoiding surgery and its associated potential morbidity. This case highlights that pancreatic tissue alone may be insufficient for diagnosis of systemic vasculitic disease. Definitive diagnosis required histopathologic confirmation from extra-pancreatic tissue, particularly renal and salivary gland biopsies, reinforcing the principle that systemic diseases require systemic tissue evaluation.

This aligns with broader experience in the literature, where early immune work-up has changed the clinical direction in multiple cases. Kontis et al described 2 patients with isolated pancreatic GPA who were spared surgery through early biopsy of lung nodules and renal tissue [5]. In contrast, Reddy et al and Valerieva et al described patients undergoing Whipple or distal pancreatectomy due to initial misdiagnosis [1,4].

Our case also supports the role of non-operative management once GPA is identified. Initiation of rituximab and corticosteroids led to rapid resolution of systemic symptoms and improvement in imaging findings. Similar favorable outcomes with immunosuppressive therapy have been reported across several pancreatic GPA cases, confirming the effectiveness of medical management once a correct diagnosis is made [3,5,6].

Few limitations of the case report need to be mentioned. Sampling error in the initial pancreatic FNA cannot be definitively excluded. Vasculitic changes were not directly demonstrated within pancreatic tissue, limiting organ-specific histologic confirmation. As a single-case report, generalizability is limited, and no standardized diagnostic algorithm exists for suspected pancreatic vasculitis. Nonetheless, the case provides important practice-informing insights.

Conclusions

GPA presenting as a pancreatic mass is an exceedingly rare and diagnostically challenging entity that can closely mimic pancreatic adenocarcinoma both radiologically and cytologically. This case highlights the substantial risk of misdiagnosis and unnecessary chemotherapy or surgical intervention in the absence of a thorough systemic evaluation. The presence of atypical systemic features, strongly positive C-ANCA, and renal biopsy–proven pauci-immune vasculitis were pivotal in revising the initial diagnosis. Our patient represents a distinctive example in which an initial false-positive cytological interpretation of adenocarcinoma was ultimately overturned through repeat tissue sampling, comprehensive immunologic assessment, and multidisciplinary evaluation. This case reinforces that GPA should be considered in the differential diagnosis of atypical pancreatic lesions, particularly when systemic or renal abnormalities are present. Incorporating immunologic assays such as ANCA and complement testing into the diagnostic algorithm may prevent unnecessary oncologic therapy or pancreatic resection. Clinicians should maintain a high index of suspicion for vasculitic mimics of malignancy in cases with discordant clinical, laboratory, or cytologic findings and pursue a comprehensive immunologic work-up – including ANCA, complement, and IgG4 levels – before committing to definitive oncologic management. Early recognition and immunosuppressive therapy, as demonstrated in this case, can lead to complete remission while avoiding significant treatment-related morbidity.