)
FINAL DIAGNOSIS
Type 1 Autoimmune Pancreatitis (IgG4 sclerosing disease)
BACKGROUND
Autoimmune pancreatitis (AIP) is a form of chronic pancreatitis that can diffusely involve the pancreas or can be mass-forming with localization to the pancreatic head in approximately 80% of cases [1]. The mass-forming type may present with obstructive symptoms suggestive of adenocarcinoma, prompting unnecessary surgery. AIP has also been called chronic sclerosing pancreatitis, lymphoplasmacytic sclerosing pancreatitis (LPSP), idiopathic chronic pancreatitis with periductal lymphoplasmacytic infiltration, idiopathic duct-centric chronic pancreatitis (IDCP), idiopathic tumefactive chronic pancreatitis, and duct-destructive pancreatitis [2].
"Classic" AIP is associated with LPSP histology and elevated IgG4, but it is now known that there are at least two distinct forms of AIP. Hence, today, AIP can be classified into two clinicopathologic subtypes: LPSP (Type 1 AIP) and IDCP (Type 2 AIP) [2, 3]. LPSP (Type 1 AIP) is associated with a dense periductal lymphoplasmacytic infiltrate, obliterative venulitis, a swirling or storiform fibrosis, elevated serum IgG4 levels with IgG4-positive immunohistochemical staining of plasma cells (>10/HPF) in the pancreatic tissue, and frequent systemic involvement. IDCP (Type 2 AIP), on the other hand, tends to have normal IgG4 levels in the serum with histologically insignificant IgG4 staining of the pancreas. In addition, IDCP is more likely to be associated with a lobule-centric inflammatory infiltrate that includes neutrophils, granulocytic epithelial lesions (GEL) which are lesions characterized by focal detachment and destruction of the duct epithelium due to infiltrating neutrophils and eosinophils [1, 3], patchy venulitis, and inflammatory bowel disease (ulcerative colitis and Crohn's disease). Due to these distinct histologic variants, AIP can be considered a heterogeneous disease process and the possibility of histologic variants should be considered when making the diagnosis.
Since our case is an example of "classic" or Type 1 AIP, the remainder of this discussion will be focused on this sybtype. AIP can be associated with other autoimmune diseases such as Sjogren's syndrome, systemic lupus erythematosus, diabetes mellitus, primary sclerosing cholangitis, primary biliary sclerosis, and inflammatory bowel disease. More recent studies suggest the existence of a systemic IgG4-related autoimmune syndrome [4] that may or may not have multiple organ system involvement including sclerosing sialadenitis, retroperitoneal fibrosis, hepatic inflammatory pseudotumor, lymphadenopathy, interstitial nephritis, sclerosing cholangitis, and pulmonary fibrosis[2, 5, 6]. Due to these observations and continued research of the disease process, autoimmune pancreatitis is currently believed to be one manifestation of a larger, systemic IgG4-related sclerosing disease [7].
IMMUNOPATHOLOGY
The inflammatory infiltrate is composed of T and B lymphocytes, with a predominance of T cells. Many patients with AIP have various autoantibodies. Antinuclear antibody and antibodies to rheumatoid factor, lactoferrin, carbonic anhydrase-II and smooth muscle have been described in cases of AIP [2]. Some patient populations with autoimmune pancreatitis have been shown to have an HLA-DRB10405/DQB10401 haplotype [4, 8]. Finally, patients with Type 1 AIP usually have hypergammaglobulinemia with elevated serum IgG4 levels. The exact mechanism of IgG4 in the disease process is unclear and the underlying immunological stimulus for AIP remains unknown.
CLINICAL PRESENTATION AND IMAGING
AIP occurs more commonly in men, presenting in the sixth and seventh decade of life. The most common clinical presentation is that of painless, obstructive jaundice associated with biliary stricture and focal mass or diffuse enlargement of the pancreas as seen on imaging [2, 8]. General findings of AIP on imaging studies include diffusely or focally enlarged pancreas and stricturing of the main pancreatic duct with dilatation of the upstream pancreatic or common bile ducts. Characteristic description on computed tomography (CT) imaging includes a "capsule-like" border, or low-density rim surrounding the pancreas. Calcifications, pancreatic duct stones, and pseudocysts are typically absent [2]. Endoscopic ultrasound (EUS) may visualize the pancreas however this modality is used primarily for biopsy or fine-needle aspiration (FNA) of the lesion.
HISTOLOGIC FEATURES
The most common histologic features of AIP are dense periductal lymphoplasmacytic infiltrate, periductal fibrosis and venulitis [2, 4, 8]. The inflammatory infiltrate is predominantly lymphocytes and plasma cells with eosinophils and macrophages admixed. Periductal fibrosis is accompanied by perilobular and intralobular fibrosis. The fibrotic process can exhibit a storiform pattern and contains the same lymphoplasmacytic infiltrate that involves the ducts (Figure 3), a feature that distinguishes Type 1 AIP fibrosis from that seen in other forms of chronic pancreatitis. Venulitis of small to medium-sized veins has been reported in up to 90% of cases in one study by Zamboni and colleagues [9] often in areas of dense fibrosis. The inflammatory cells infiltrate the walls and the endothelium, leading to an obliterative venulitis (Figure 5). Due to the dense inflammatory infiltrate it may be difficult to detect veins involved, but elastic stains such as Verhoeff Van Gieson can help (Figure 7). Because serum levels of IgG4 are usually increased in patients with AIP, the detection of IgG4 plasma cells via immunohistochemistry is another histopathologic finding. Based on studies of IgG4 staining in AIP, the presence of >10 IgG4-positive cells per high power field is an accepted cut-off for making the diagnosis (Figure 6) [2].
DIFFERENTIAL DIAGNOSIS
The differential diagnosis of AIP includes chronic pancreatitis, pancreatic adenocarcinoma and inflammatory myofibroblastic tumor (IMT) in cases with prominent fibrosis and myofibroblasts arranged in a storiform growth pattern. Chronic pancreatitis (especially alcohol-related) may show calcification, pseudocyst formation, fat necrosis, and protein plugging of ducts, all features that are usually absent in AIP. The amount of chronic inflammation is variable, but is typically scant to mild and plasma cells are not typically present. IMT may express ALK1 and p53 with Ki-67 showing a high proliferative index, differentiating it from AIP [10].
AIP can mimic pancreatic (ductal) adenocarcinoma from the clinical, radiological and pathological standpoint. In addition, AIP may be associated with lymphadenopathy, confusing the picture even further. However, histology along with high serum levels of IgG4 may help guide the diagnosis. A cut off >135 mg/dL IgG4 has been suggested by Hamano et al [11] to distinguish AIP from pancreatic adenocarcinoma. Their study found an IgG4 level >135 mg/dL to be 97% specific and 95% sensitive for differentiating AIP from pancreatic ductal adenocarcinoma in cases with identical radiologic findings. That being said, IgG4 levels between 135 - 200 mg/dL should be interpreted with caution as elevations can be found in up to 7% of patients with pancreatic cancer [12]. Immunohistochemical expression of IgG4 plasma cells, >10 IgG4-positive cells/HPF in pancreatic tissue, is a useful diagnostic criteria, however the presence of IgG4 positive plasma cells must be used in adjunct with histological and clinical information. The Mayo Clinic "HISORt Criteria" has been proposed to aid in the diagnosis of AIP. It combines histology, imaging, serology, organs of involvement, and response to corticosteroid therapy [13, 14].
TREATMENT AND PROGNOSIS
Corticosteroid therapy is the mainstay of treatment for autoimmune pancreatitis. Other extrapancreatic lesions usually improve with immunosuppression as well. The patient should respond to steroid therapy within 2 to 4 weeks. If imaging and serum IgG4 values do not improve, then a diagnosis other than AIP should be entertained [15]. The recurrence rate ranges from 6 to 26% in some studies [8]. Prognosis is generally believed to be better than non-autoimmune pancreatitis however data on long-term follow-up is lacking from the literature.
REFERENCES
Contributed by Sarah K Harm, MD and Alyssa M Krasinskas, MD
  |
){kind=link}
){kind=link}
){kind=link}