Controversies in clinical pancreatology: autoimmune pancreatitis: does it exist?

Abstract

ArticlePlus Click on the links below to access all the ArticlePlus for this article. Please note that ArticlePlus files may launch a viewer application outside of your web browser. https://links.lww.com/MPA/A1 INTRODUCTION Dr. William M. Steinberg The term “autoimmune pancreatitis” (AIP) has been used to define a relatively new syndrome of clinical and histologic findings. This entity has been described in several articles published in Europe and Japan, yet few cases have been seen or reported in the United States. Therefore, it was thought that this was a timely topic for discussion at our Controversies in Clinical Pancreatology session at the American Pancreatic Association meetings on November 14, 2002. Five discussants participated in this session. Dr. Randall Pearson of the Gastroenterology Section, Mayo Clinic presented a case, and Dr. Dan Longnecker of the Department of Pathology, Dartmouth-Hitchcock Medical Center discussed the pathology. Dr. Suresh Chari of the Gastroenterology Section, Mayo Clinic, Dr. Kazuichi Okazaki of the Gastroenterology Section, Kyoto University, and Dr. Luca Frulloni of the Gastroenterology Section, University of Verona then proceeded with reports of the Mayo Clinic, Japanese, and Italian experiences with this syndrome. Finally, the discussants kindly submitted the following manuscripts for this article. BACKGROUND INDETERMINANT PANCREATIC MASS Dr. Randall K. Pearson A 53-year-old white woman in generally good health (postcholecystectomy) felt vaguely unwell for nearly 4 months with dyspeptic symptoms, two episodes of nausea and diarrhea characterized as “flu,” and a 20-pound weight loss. Two weeks prior, she noted dark urine, pale stools, and scleral icterus. She had no known history of pancreatitis, drank minimal amounts of alcohol, and had no family history of chronic pancreatitis. Biochemical studies confirmed jaundice, and computed tomography (CT) of the abdomen showed dilatation of the biliary tree down to the level of the pancreas. The pancreas was generally enlarged, but the cause of the obstruction was not obvious. Endoscopic retrograde cholangiopancreatography (ERCP) demonstrated a distal bile duct stricture that was stented, and she was referred to the Mayo Clinic. After endoscopic stenting, she felt progressively better and was essentially without complaints. Her physical examination was normal. Laboratory studies were consistent with resolving obstructive jaundice. A CA 19-9 level was mildly elevated at 91.4 U/mL (normal, <40 U/mL). Repeat triple-phase contrast CT with thin cuts through the pancreas failed to reveal a mass or cause for biliary obstruction. Endoscopic ultrasound findings were suspicious for malignancy; a 3-cm mixed echogenicity region in the head was prominently seen in the setting of diffuse alteration in the parenchyma consistent with pancreatitis. Fine-needle aspiration of the mass and peripancreatic lymph nodes was negative. ERCP (Fig. 1) confirmed the distal bile duct stricture; a limited pancreatogram showed a normal duct in the head with slight narrowing at the neck of the pancreas and irregularity in the tail (Fig. 2).FIGURE 1.: Retrograde cholangiogram demonstrating a distal common bile duct stricture with proximal dilatation.FIGURE 2.: Mildly irregular pancreatic duct observed with endoscopic pancreatography.The patient underwent laparotomy, and a pancreaticoduodenectomy (Whipple resection) was performed. PATHOLOGY OF AUTOIMMUNE PANCREATITIS Dr. Daniel Longnecker The pathologic changes found in cases of pancreatitis associated with other autoimmune diseases, such as Sjögren disease and extrahepatic sclerosing cholangitis, 1 can be divided into two groups. The first group includes constant features characteristic and suggestive of AIP, i.e., found in all cases, and the second group comprises variable features seen in a fraction of cases that, when present, provide additional support for the diagnosis. On gross examination, the involved pancreas is firm or hard (a constant feature) and may be enlarged or “mass forming.” These features may lead to resection because of the suspicion that the lesion is a carcinoma. The lesion may be limited to one portion of the pancreas, most often the head, or less commonly may involve the body, tail, or whole organ. Thus, location in the pancreas, mass formation, and pancreatic enlargement are variable features. At the histologic level, the characteristic changes include sclerosing pancreatitis (Fig. 3) with prominent infiltrating lymphocytes and plasma cells, and conspicuous involvement of ducts in the inflammatory process (Fig. 4). The main duct, branch ducts, and the common bile duct (CBD) are commonly involved. Localization of the infiltrate in the duct wall is variable–-sometimes subepithelial (Figs. 4 and 5), sometimes focused in the stroma of the duct wall, and sometimes diffusely involving all layers and adjacent connective tissue. Immunostaining has demonstrated that T cells predominate over B cells among the infiltrating lymphocytes. 1 The lumen of inflamed ducts is characteristically narrowed (Fig. 4B), and scarring or edema may thicken their walls.FIGURE 3.: This field shows sclerosing pancreatitis with perilobular fibrosis and some atrophy of acinar tissue in lobules. This pancreas contained typical areas of ductal inflammation as described for autoimmune pancreatitis. Hematoxylin and eosin (H & E).FIGURE 4.: Large pancreatic ducts with a prominent subepithelial lymphoplasmacytic infiltrate, with thickening of the duct wall (A and B) and compression of the lumen (B). This specimen was classified as chronic pancreatitis, etiology unknown, when it was originally seen but in retrospect is regarded as autoimmune pancreatitis; H & E.FIGURE 5.: Lymphocytes and plasma cells are conspicuous in this duct, where they infiltrate the subepithelial layer and extend into the lumen (left). A few granulocytes are also present in the infiltrate. Figures 3–5 are from the same pancreas; H & E.Variable features include the presence of eosinophils or neutrophils in the leukocytic infiltrate, degree of epithelial injury and inflammation in the ducts, presence of vasculitis (veins more than arteries), and degree of acinar cell atrophy, scarring, and inflammation in lobular tissue. The epithelium of involved ducts is well preserved in some patients, whereas the infiltrate extends into the epithelium and lumen in others. Fibrosis may be perilobular but commonly extends into the lobules as acinar cells atrophy. Conspicuous negatives typically include the absence of calculi in ducts, absence of pseudocyst, and absence of more than minimal fat necrosis. Are these changes specific for AIP? There is growing confidence that AIP is a legitimate pathologic diagnosis for some cases of sclerosing pancreatitis that have the constant features of AIP and lack stigmata of alcoholic chronic pancreatitis. 1 However, many pathologists are more comfortable with a descriptive diagnosis, such as lymphoplasmacytic sclerosing pancreatitis (LPSP). In the past, cases of AIP have often been included with idiopathic chronic pancreatitis. Why are the changes so variable? One obvious explanation is that the lesions change with duration or progression of the disease. This seems certain to be at least part of the basis for the spectrum of changes. Another possible factor is that the immune response is triggered by different epitopes with varying cellular targets in patients with differing histologic changes. Thus, one pending issue is whether there are recognizable subtypes of AIP with different etiologies, different pathologic features, and perhaps different clinical courses. The patient presented by Dr. Pearson had a Whipple resection that included a 4.7-cm portion of the head of the pancreas containing a firm 3 × 2 × 2 cm mass. There were no stones or cysts. Histologic sections showed sclerosing chronic pancreatitis with a somewhat irregular distribution of lobular inflammation, fibrosis, and atrophy. A large pancreatic duct contained a conspicuous subepithelial leukocytic infiltrate consisting of lymphocytes and plasma cells. Medium-sized ducts in the same section contained a lymphoplasmacytic infiltrate throughout the fibrous wall (Fig. 6). Perilobular fibrous tissue contained leukocytic aggregates of lymphocytes, fewer plasma cells, and prominent eosinophils. A few lymphoid follicles were present in these scarred areas. Neutrophils were inconspicuous, and there were no stigmata of acute pancreatitis, such as fat necrosis or pseudocyst. The ducts were free of calculi or proteinaceous “plugs.” The original histopathologic diagnosis was “Chronic pancreatitis involving an area 3 × 2 × 2 cm with slight stricture of the distal common bile duct.” The features of this case fall within the spectrum of those described previously as typical of AIP.FIGURE 6.: Interlobular duct infiltrated by lymphocytes and plasma cells throughout the wall in the resected pancreas from the patient presented by Dr. Pearson. From a slide provided by Dr. Thomas Smyrk, Mayo Clinic, Rochester; H & E.CLINICOPATHOLOGIC FEATURES OF IDIOPATHIC TUMEFACTIVE CHRONIC PANCREATITIS Drs. Suresh T. Chari and Thomas C. Smyrk Some benign conditions simulate malignancy. In a recent review of 40 patients undergoing subtotal pancreatoduodenectomy for “benign but clinically suspicious” disease, the most common postoperative diagnosis was LPSP. 2 This idiopathic inflammatory pancreatitis has also been termed autoimmune pancreatitis, nonalcoholic duct-destructive pancreatitis, and chronic inflammatory sclerosis of the pancreas. The majority of reports of AIP have come from Japan, with few reports from Western countries. It is not clear whether this reflects the rarity of the disease or a failure to recognize it in Western countries. To identify potential cases, we studied patients with tumefactive chronic pancreatitis (TCP), defined as chronic pancreatitis presenting with a pancreatic mass or obstructive jaundice. The results of these studies have been reported in detail elsewhere. 3,4 We report here a summary of our findings. We reviewed a consecutive series of 254 pancreatic resections performed for benign disease at the Mayo Clinic in Rochester between 1985 and 2001. Twenty-seven patients met the three inclusion criteria for our study: dense lymphoplasmacytic periductal inflammation, no known cause for pancreatitis, and “tumefactive” presentation (i.e., presenting with a mass or obstructive jaundice). The mean age of patients with idiopathic TCP was 58 ± 2 years, and the median duration of symptoms was 3 months. Although two thirds of patients presented with jaundice, 70% of them had minimal or no pain. No patient had a personal or family history of autoimmune disorder. One patient had Crohn disease, and one had ulcerative colitis. One third of the patients with idiopathic TCP (9 of 27; 33%) had diffuse enlargement of the pancreas (6 on imaging studies and 3 at surgery). The most frequent findings on ERCP were strictures in the bile duct, pancreatic duct, or both ducts. Pancreaticoduodenectomy was performed in 25 patients: 1 patient had distal pancreatectomy, and 1 had total pancreatectomy. None of the patients were treated with steroids preoperatively or postoperatively, and none has had a recurrence on follow-up evaluation. Pathology review suggested the presence of two histologic subtypes. The first, which we have termed lymphoplasmacytic sclerosing pancreatitis, seemed to be a more generalized fibroinflammatory process. Thus, although there was prominent periductal inflammation, inflammation and fibrosis also extended into pancreatic parenchyma and peripancreatic soft tissue (Fig. 7). Obliterative phlebitis was a striking and consistent feature. Interestingly, duct epithelium tended to remain intact despite the dense periductal inflammation. Fourteen of the 27 patients were classified histologically with LPSP.FIGURE 7.: (A–D) LPSP features dense periductal inflammation without duct destruction (A). Inflammation and fibrosis extend into parenchyma (B) and peripancreatic soft tissue (C). Obliterative phlebitis is always seen (D).The second subtype was characterized by patchy inflammatory infiltrate involving mainly lobules and pancreatic ducts (Fig. 8). The infiltrate was confined to the pancreas. It was predominantly neutrophilic with occasional microabscesses. The inflammatory infiltrate involved the entire wall of pancreatic ducts, and epithelial destruction was commonly seen with obstruction of lumen. Obliterative phlebitis was rarely seen. We applied the descriptive term idiopathic duct-destructive pancreatitis (IDCP) and assigned 13 cases to this category.FIGURE 8.: (A–D) IDCP has periductal inflammation but includes neutrophils, and there is damage to epithelium (A and B). Inflammation is less in fibrous areas (C and D).Although histology suggests the existence of subtypes, the only clinical difference between patients with LPSP and IDCP was the prevalence of jaundice, which developed in 13 of 14 (93%) patients with LPSP but in only 5 of 13 of those with IDCP (p = 0.005). DISCUSSION A number of case reports and small series have described patients presenting with a pancreatic mass who prove to have a unique form of chronic inflammatory pancreatitis. The presence of increased gamma globulins, a nonspecific increase in various antibody titers, and a response to steroids has prompted the use of the term AIP. 5–11 Other nomenclatures, based on histology, include LPSP, nonalcoholic duct-destructive pancreatitis, and chronic inflammatory sclerosis of the pancreas. 12–14 Though the terms AIP and sclerosing pancreatitis have been used interchangeably, 10 the absence of specific serologic markers makes it unclear if all authors are describing the same disease. Do the patients with idiopathic TCP described in our study have the same disease as that described by Japanese authors as sclerosing pancreatitis or AIP? This question is difficult to answer with certainty because reports of sclerosing pancreatitis or AIP have used serologic criteria and response to steroids to make the diagnosis, and histology is generally not available. 10,15,16 Conversely, all of our patients have well-described pathology, but due to lack of clinical suspicion of AIP and the absence of associated autoimmune disorders, none of our patients had serologic parameters measured. Still, the histology of idiopathic TCP described in our study resembles that described as AIP or sclerosing pancreatitis. It may be a disservice to add a new term (IDCP) to a literature already overcrowded with terminology, particularly because LPSP and IDCP were clinically indistinguishable. It may be that our categories LPSP and IDCP are simply part of the spectrum of the same disease or perhaps different stages of the same disease. Still, we think it worth highlighting these two histologic patterns as a possible prelude to discovering different etiologies for this poorly understood condition. Interestingly, the features seen in our patients with IDCP are similar to those described by Ectors et al. 12 as nonalcoholic duct-destructive pancreatitis. Obliterative phlebitis was described only in 1 of their 10 patients, and that patient had Sjögren syndrome. Thus, it may be that some variation in description and nomenclature in the literature may derive from the fact that there is more than one type of chronic inflammatory pancreatitis. Chronic inflammatory pancreatitis is a distinctive pattern of histologic injury observed in patients with idiopathic TCP. We have described two histologic subtypes and designated them LPSP and IDCP, but it is not clear whether LPSP and IDCP are the same disease or represent two or more different entities. The condition(s) remains idiopathic. Resection of the mass prevents recurrence of symptoms. AUTOIMMUNE PANCREATITIS: THE JAPANESE EXPERIENCE Dr. Kazuichi Okazaki Definition and concept of AIP Since Sarles et al. observed a particular case of pancreatitis with hypergammaglobulinemia, occasional coexistence of pancreatitis with other autoimmune diseases, such as Sjögren syndrome, has been reported. These findings led us to the concept of autoimmune-related pancreatitis, called autoimmune pancreatitis. 17 Although it has not yet been widely accepted as a new clinical entity, the present section reports recent experiences of AIP in Japan. Although the pathogenesis and pathophysiology of AIP are still unclear, the characteristic findings in most cases of AIP can be summarized as follows 18: (i) increased levels of serum gamma globulin, IgG, or IgG4; (ii) presence of autoantibodies; (iii) diffuse enlargement of the pancreas; (iv) diffusely irregular narrowing of the main pancreatic duct and occasionally stenosis of intrapancreatic bile duct on ERCP imaging; (v) fibrotic changes with lymphocyte infiltration; (vi) no symptoms or only mild symptoms, usually without acute attacks of pancreatitis; (vii) rare pancreatic calcification or cysts; (viii) occasional association with other autoimmune diseases, and (ix) effective steroid therapy. AIP is a rare disorder, although the exact prevalence is still unknown. More than 309 cases have been reported as AIP or pancreatitis with narrowing of the pancreatic duct (PNPD) in the Japanese literature. We encountered 21 cases of AIP in a review of 451 cases of chronic pancreatitis, and incidence predominated in older men versus women in Japan. Although the correct morbidity of primary or secondary AIP is unclear, more than half of the cases are primary. It is still unclear whether the pathogenetic mechanism of secondary (or syndromic) AIP with other autoimmune diseases differs from primary AIP. Recently, “Diagnostic Criteria for Autoimmune Pancreatitis, 2002” was proposed by the Japan Pancreas Society and contained three criteria: pancreatic imaging, laboratory data, and histopathologic findings (Table 1) and associated diseases (Table 2). 19 Patients with AIP often show narrowing of the CBD mainly in the intrapancreatic area, which may result in dilatation of the upper biliary tract. Sclerosing changes of the extrapancreatic bile duct similar to primary sclerosing cholangitis (PSC) are reported. Unlike with PSC, steroid administration usually shows therapeutic effects on biliary lesions in patients with AIP, suggesting that the developmental mechanism of biliary lesions in AIP may be different from typical PSC.TABLE 1: Diagnostic criteria for autoimmune pancreatitis, 2002, proposed by the Japan Pancreas SocietyTABLE 2: Autoimmune pancreatitis and associated diseasesDiabetes mellitus (DM) is often observed in patients with AIP (in approximately 43–68%), and the majority has type 2 DM, with some patients improving after steroid therapy. 20 Although the mechanism is obscure, cytokines from T cells and macrophages suppressing the function of islet β-cells may be downregulated by steroid treatment. Clinical symptoms Patients with AIP usually have no or only slight discomfort in the epigastrium or back, in addition to symptoms related to other associated diseases. Thus, clinical symptoms are different from acute or severe pancreatitis. Obstructive jaundice due to the stenosis of the intrapancreatic CBD is characteristic for AIP and is rare in other types of pancreatitis. Laboratory data Patients with AIP usually show increased levels of serum pancreatic enzymes, hypergammaglobulinemia, and presence of several autoantibodies, such as antinuclear, antilactoferrin (LF), and anticarbonic anhydrase (CA-II) antibody and rheumatoid factor. 21 However, the antimitochondrial (M2) antibody specific for PBS is rarely observed (Fig. 9). CA-II and LF are distributed in the ductal cells of several exocrine organs, including the pancreas, salivary glands, biliary duct, and distal renal tubules. Serum levels of IgG4, immune complexes, and the IgG4 subclass of immune complexes are often increased in patients with AIP. 22 Patients with CBD stenosis show increased serum IgG4 and abnormalities in the serum bilirubin and hepatobiliary enzymes. For these patients, other liver diseases such as viral hepatitis, autoimmune hepatitis, or primary biliary cirrhosis (PBC) should be ruled out. After steroid therapy, many abnormal laboratory findings are reversible.FIGURE 9.: Prevalence of autoantibodies in AIP. Antinuclear antibodies, antilactoferrin antibodies, and anti-CAII antibodies were identified in 75%, 75%, and 55% of 20 cases, respectively. Ninety percent of the cases showed either antilactoferrin or anti-CA-II antibody, and 35% showed both antibodies, suggesting that immune responses against these proteins are heterogeneously activated.Pancreatic and biliaryimaging Computed tomography (CT), magnetic resonance imaging (MRI), and ultrasonography (US) demonstrate the diffusely enlarged pancreas, with a “sausage-like” appearance (Fig. 10) and a capsule-like rim showing low density on CT, low intensity on T2-weighted MRI, and delayed enhancement on dynamic MRI. Pancreatic calcification or pseudocyst is seldom observed. ERCP images of the AIP patients show segmental or diffuse narrowing of the main pancreatic duct (Figs. 11–13). Although magnetic resonance cholangiopancreatography (MRCP) poorly shows pancreatic duct stenosis, it can well demonstrate stenosis of the bile ducts mainly in the intrapancreatic area, resulting in dilatation of the upper biliary tract. Sclerosing changes of the extrapancreatic bile ducts similar to PSC are sometimes observed. Steroid therapy is usually effective for changes of the biliary and pancreatic ducts (Fig. 14).FIGURE 10.: CT of AIP. Diffusely enlarged pancreas similar to sausage appearance is typical of autoimmune pancreatitis. Contrast-enhanced CT often demonstrates delayed enhancement. After steroid therapy, the size of the pancreas becomes normal, often atrophic.FIGURE 11.: ERCP of AIP. In the lower left panel, diffusely irregular narrowing of the main pancreatic duct is observed. In the upper left panel, many patients show narrowing of intrapancreatic common bile duct and pancreatic duct. Narrowing of the terminal bile duct is supposed to be induced mainly by compression of the swollen pancreas. After steroid therapy, most patients remarkably improve as shown in the right picture.FIGURE 12.: ERCP of local type AIP. ERCP images show the locally irregular narrowing of the main pancreatic duct in the pancreatic tail (the upper left panel) and local narrowing in the pancreatic head (the lower left panel). Irregular narrowing is observed in more than one third of the length but not in the entire pancreas. After steroid therapy, most of these abnormal findings improve.FIGURE 13.: Pancreatogram of 20 patients with AIP. Of 20 cases, 11 showed diffuse narrowing in the pancreatic duct, 3 from the head to body, 2 in the head, and 4 in the tail.FIGURE 14.: PSC-like cholangiogram associated with AIP. PSC-like cholangiogram showing narrowing of upper or middle biliary tract is observed in 6 of 20 patients with AIP. Different from typical PSC, these abnormal findings responded to steroid therapy very well.Histopathology Microscopic findings, if obtained, show fibrotic changes with infiltration of lymphocytes and plasmacytes mainly around the pancreatic duct (Fig. 15). HLA-DR antigens are often expressed on pancreatic duct or acinar cells. HLA-DR+ T cells predominantly infiltrate over B cells in the pancreas, while plasma cells and lymph follicles are observed. HLA-DR antigens are expressed on the pancreatic duct cells and on CD4+ T cells, suggesting that an autoimmune mechanism may be involved in inflammation. In some patients with AIP, CD4+ Th1 cells predominate over Th2-type cells. Therefore, similar to SjS, Th1 cytokines may be essential in the induction or maintenance of AIP, whereas Th2 cytokines may be involved in disease process progression, especially local B-cell activation.FIGURE 15.: Histopathology and microscopic findings of the pancreas in AIP. Massive infiltration of lymphocytes and plasmacytes (A) and fibrotic changes (B) were observed mainly around the pancreatic duct. In most cases, T lymphocytes (C) predominantly infiltrate over B lymphocytes (D), with occasional follicle formation.Treatment and prognosis For most AIP patients, intensive treatment for acute pancreatitis is not required. For patients with jaundice, percutaneous transhepatic or endoscopic biliary drainage is often required, especially in cases complicated by bacterial infection. Steroid therapy is usually effective for narrowing of the bile ducts and pancreatic ducts. Notable is that some patients may spontaneously improve without treatment. The long-term prognosis of AIP is unknown. Conclusions In conclusion, our experience in Japan supports the concept of autoimmune-related pancreatitis, which appears to be a unique clinical entity. AUTOIMMUNE PANCREATITIS: THE ITALIAN EXPERIENCE Drs. Luca Frulloni and Giorgio Cavallini Definition: “The tower ofBabele” Since 1959, autoimmunity has been proposed as a pathogenetic mechanism in patients suffering from pancreatitis. 23 In 1993, we introduced the term “primary chronic pancreatitis” to define a disease characterized by an autoimmune reaction against the pancreatic ducts, resulting in an obliterating inflammatory fibrosis of the pancreatic duct system. 24 We suggested that the key step in the pathogenesis of the disease was an immune reaction against a target antigen in the epithelium of the pancreatic ducts, with secondary periductal inflammatory infiltration (mainly activated T lymphocytes), followed by an obliterating periductal fibrosis. 24,25 Many terms have been used later for this particular form of pancreatitis, such as “nonalcoholic duct-destructive chronic pancreatitis”, 26 “lymphoplasmacytic sclerosing pancreatitis”, 27,28 “autoimmune pancreatitis”, 29,30 “granulomatous pancreatitis”, 31 and “sclerosing pancreatocholangitis”. 32 The terminology was mainly used to identify the pathologic characteristics of the inflammatory process (duct-destructive, sclerosing, lymphoplasmacytic, granulomatous), but some definitions describe the postulated immune-mediated pathogenesis (primary, autoimmune) rather than the histologic pictures. In our opinion, perhaps an easier term we can use for this disease is autoimmune, to stress pathogenesis and address the physician in the use of steroids for management of this particular form of chronic pancreatitis. Epidemiology: “The tip of theiceberg” We do not know the current real incidence of AIP in Italy or in the world at large. In a recent Italian multicenter study coordinated by the University of Verona on the epidemiology of chronic pancreatitis in Italy, called “PanCroInf-AISP,” that involved 21 centers and enrolled 383 patients suffering from chronic pancreatitis during 2 years (2001 and 2002), autoimmunity was recognized as an associated factor in 23 patients (6%). Probably this is only “the tip of the iceberg” because there are many problems associated with the disease. The first problem is awareness of the disease, particularly by surgeons who probably first observe patients suffering from this form of pancreatitis and who frequently misdiagnose it as pancreatic tumor. In all surgical series, we probably find inflammatory pancreatitis in more than 5% of patients who undergo Whipple procedure for pancreatic cancer. 33 Furthermore, we not only need skilled pathologists but also skilled gastroenterologists, surgeons, radiologists, and endoscopists to recognize AIP. Finally, there are not codified criteria yet for the diagnosis of AIP. Typology of the disease:“Heterogeneity” Autoimmune pancreatitis is a heterogeneous disease with different clinical aspects. In our experience (33 patients with a definitive diagnosis of AIP), it may occur as acute pancreatitis (25% of patients) involving all or a part of the pancreatic gland. In some patients (25%), the clinical history, instrumental findings, and functional pancreatic tests are very similar to those observed in chronic pancreatitis. More frequently (50% of patients), it may present as a so-called “mass-forming pancreatitis,” that is, a mass generally localized in the head of the pancreas mimicking a pancreatic cancer. Half of these patients are free of pain at the onset of the disease. Jaundice may be the main clinical sign when the common bile duct is compressed by a pancreatic mass (40% of patients). In some patients, especially in those with a long history of pancreatic-type pain, we may observe diabetes or steatorrhorrea. Calcifications are rare (fewer than 20% of all patients), but their presence does not exclude the diagnosis of AIP. Most patients neither drink alcohol (70%) nor smoke (50%), but some patients (10%) drink more than 80 grams of alcohol a day, and one third smoke more than 10 cigarettes a day. In the past, these patients were probably classified as having alcoholic chronic pancreatitis. As in other cases, alcohol is a confounding factor. The sex ratio (male to female) is approximately one to one. The average age at onset is 42 years, as in chronic pancreatitis. 34 However, we may observe patients at any age, and the distribution of patients is more similar to that observed with idiopathic pancreatitis 34,35 than with alcoholic chronic pancreatitis. 34–36 How to diagnose the disease: “Thechallenge” The diagnosis of AIP is difficult. All patients with suspected AIP should have their clinical, biochemical, and instrumental results carefully evaluated. Japanese authors recently proposed som