A form of idiopathic chronic pancreatitis (CP) suspected to be due to an autoimmune process was first described by Sarles et al1 in 1961. In 1991, Kawaguchi et al2 described "an unusual lymphoplasmacytic sclerosing inflammatory disease involving the total pancreas, common bile duct, gallbladder, and, in one patient, the lip" in 2 patients that presented with masslike enlargement of the pancreatic head. Histopathologic characteristics included diffuse lymphoplasmacytic infiltration, marked interstitial fibrosis, acinar atrophy, and obliterative phlebitis of the pancreatic and portal veins. They called the condition lymphoplasmacytic sclerosing pancreatitis with cholangitis.2 In 1995, Yoshida et al3 described a 68-year-old woman with a steroid-responsive disease presenting with obstructive jaundice, diffusely enlarged pancreas, an irregularly narrowed pancreatic duct, hypergammaglobulinemia, and elevated autoantibody titers. Drawing parallels from the literature on autoimmune hepatitis, the authors coined the term "autoimmune pancreatitis"3 (AIP) to describe this clinical entity. In 2002, the Japan Pancreas Society proposed diagnostic criteria for AIP4 based on the classic imaging and serological findings; these were later revised in 2006.5
In 2001, Hamano et al6 reported from Japan that elevated serum IgG4 levels were highly specific and sensitive for the diagnosis of AIP. In 2003, Kamisawa et al7 suggested that AIP is a systemic disease based on the findings that the pancreas and other involved organs have abundant infiltration with IgG4+ve plasma cells. These features were included in the Korean diagnostic criteria for AIP proposed in 2007.8 In 2008, Japanese and Korean societies agreed on Asian consensus criteria for the diagnosis of AIP.9 Recently, a Japanese study based on a survey of 17 centers in Japan identified 563 patients with AIP in Japan.10
Meanwhile, reports from Europe and the United States described unique histological patterns in resected pancreata of patients with mass-forming chronic nonalcoholic pancreatitis, showing overlapping clinical and histopathologic features with those of Japanese patients. In the first European study published in 1997, Ectors et al11described the histological pattern of "non-alcoholic duct destructive pancreatitis" in 12 cases of idiopathic chronic pancreatitis, a histological pattern that was clearly distinguishable from that of alcoholic CP. The authors noted a similarity of the pancreatic histopathologic findings not only with those reported in association with ulcerative colitis by Ball et al12 from the Mayo Clinic in 1951 but also with the pancreatic involvement seen in sclerosing cholangitis reported from Japan noted earlier.2 Italian diagnostic criteria for AIP were reported in 200313, which were based on the histological hallmarks outlined in the article by Ectors et al.11
In 2003, the Mayo Clinic group in the United States reported 35 cases of "idiopathic chronic pancreatitis with lymphoplasmacytic infiltration, sometimes called autoimmune pancreatitis."14 They observed 2 distinct histological patterns in these patients: (a) lymphoplasmacytic sclerosing pancreatitis (LPSP) and (b) idiopathic duct centric pancreatitis (IDCP). Lymphoplasmacytic sclerosing pancreatitis resembled Japanese descriptions of histological features seen in AIP and IDCP resembled European description of duct destructive pancreatitis. The authors noted an overlap in the histological features of the 2 patterns. They did not speculate on the etiology of IDCP but wondered if LPSP was of autoimmune etiology. In 2006, Mayo Clinic investigators outlined diagnostic criteria for AIP using clinical data from patients with histologically confirmed LPSP.15
In 2004, Zamboni et al16 described the histological features of 62 patients with autoimmune pancreatitis from Europe; the unifying histological feature in all patients was a periductal lymphoplasmacytic infiltrate with periductal fibrosis without any of the features seen in alcoholic pancreatitis, viz, ductal dilatation or irregularity, calculi, or pseudocysts. As in the Mayo Clinic series, 2 groups of patients were distinguished on the basis of a histological criterion that was called granulocytic epithelial lesion (GEL). Interestingly, the 2 groups of patients also differed in their clinical features such as sex, mean age, and associated immune-related diseases.
Autoimmune pancreatitis has been extensively described in reports from Japan, Europe, and the United States. Large series of AIP reported from Japan have been based on a distinct clinical phenotype, with little emphasis on or need for histological examination to diagnose the disease. On the other hand, detailed descriptions of at least 2 histopathologic patterns in patients with a nonalcoholic idiopathic chronic pancreatitis, viz, LPSP or AIP without GEL and IDCP or AIP with GEL, have been reported from Europe and the United States. Both histopathologic patterns have been included under the term "autoimmune pancreatitis," based on the presence of features common to both, viz, periductal lymphoplasmacytic infiltration and peculiar periductal fibrosis. The European diagnostic criteria for AIP use the presence of GEL as the hallmark of AIP, whereas criteria from the United States are based on the clinical features of LPSP. Not surprisingly, the clinical phenotypes associated with both of these histopathologic patterns have been called "autoimmune pancreatitis."
THE HONOLULU CONSENSUS CONFERENCE ON AIP (HCCAIP)
On November 4, 2009, experts from Japan, Korea, Europe (United Kingdom, Germany, Sweden, and Italy), and the United States met in Honolulu, Hawaii to describe the entity of "autoimmune pancreatitis" as they recognized it. The experts included gastroenterologists, pathologists, radiologists, and surgeons. The goals of the meeting were to (a) agree on a clinical and histological definition of "autoimmune pancreatitis," (b) determine if the descriptions of the disease from Japan, Europe, and the United States refer to one or more disease entities, and (c) arrive at a consensus on diagnostic criteria for AIP. In this review, the deliberations of the expert panel, regarding questions (a) and (b) are detailed. During the deliberations, which were in a question-and-answer format, the following questions were discussed. The document was subsequently revised by the participants.
Definition of AIP
Question: Can AIP Be Distinguished From Other Forms of Chronic Pancreatitis (CP) on Histological Grounds in Resected Pancreata?
At the Honolulu Consensus Conference on AIP, preliminary data were presented from an international concordance study of 40 resected cases of CP to determine if AIP can be distinguished from alcoholic and obstructive forms of CP. This study is ongoing. Whereas the interobserver variability was moderate for the group as a whole, data from the 5 most accurate reviewers demonstrated 91.2% sensitivity and 98% specificity. The κ statistic for the 5 readers was 0.89, reflecting excellent interobserver agreement. These findings suggest the need for additional educational efforts to improve the overall performance among pathologists. Full results of this study will be published shortly.
The expert panel agreed that AIP has unique histopathologic features, which allow it to be distinguished from other forms of CP.
Question: Is There More Than One Histopathologic Subtype of AIP?
In the concordance study noted earlier, readers were asked to classify the histological patterns seen in AIP as LPSP and IDCP. The sensitivity and specificity among the top 5 readers was 84% and 76.4%, respectively, with a κ of 0.59, reflecting moderate interobserver agreement. Some readers had not previously diagnosed both histological patterns and chose not to subdivide AIP into LPSP and IDCP, underscoring the need for additional educational efforts to improve the overall performance among pathologists.
Question: What Are the Histological Diagnostic Criteria for LPSP and IDCP?
Expert pathologists agreed on the following (Table 1): LPSP (AIP without GELs) has 3 essential histological features14,17: (1) a lymphoplasmacytic infiltrate surrounding small-sized interlobular pancreatic ducts that does not destroy the pancreatic ductal epithelium (Fig. 1), (2) a swirling fibrosis centered around ducts and veins (storiform fibrosis) but most prominent in the peripancreatic adipose tissue (Fig. 2), and (3) obliterative phlebitis wherein the infiltrate surrounds and obliterates pancreatic veins (Fig. 3). Destructive changes of the ducts and acini caused by infiltrating granulocytes are typically absent. Immunostaining reveals abundant (>10 cells/high-power field [hpf]) IgG4-positive cells (Fig. 4).18,19
Idiopathic duct centric pancreatitis (AIP with GELs) has a histological pattern distinct from that of LPSP,2,14,17 although it also shares some features with LPSP. Periductal lymphoplasmacytic infiltrate is seen in both forms of AIP. Diffuse inflammation and diffuse storiform fibrosis (Fig. 2) as well as obliterative phlebitis (Fig. 3), which are characteristic of LPSP, are less prominent in IDCP. The most distinctive feature of IDCP is the presence of GELs seen in medium-sized and small ducts and also often in the acini, changes that may lead to the destruction and obliteration of the duct lumen.14,17 The other distinctive feature is the scanty presence (<10 cells/hpf) or the complete absence of IgG4-positive plasma cells on immunostaining (Figs. 5 and 6).
Question: Do LPSP and IDCP Have Distinct Clinical Profiles?
Data were presented from Europe and the United States based on histologically confirmed cases of LPSP and IDCP, highlighting differences in demography, clinical presentation, serological findings, other organ involvement, and disease relapse (Table 2).
The participants agreed that patients with a condition currently diagnosed as AIP have 2 distinct histopathologic types that are associated with 2 distinct clinical profiles. Thus, it is possible that LPSP (AIP without GELs) and IDCP (AIP with GELs) are histopathologic correlates of 2 distinct forms of AIP.
Question: Are There Geographic Differences in the Proportion of AIP Identified as LPSP or IDCP?
The diagnosis of LPSP and IDCP requires histopathologic examination, which is not frequently available. Based on data presented at the meeting, there are only a very small number of histologically confirmed IDCP cases in Japan. More than 500 cases of AIP with a clinical profile resembling that seen in subjects with LPSP have been reported from Japan. In contrast, in the United States and in most centers in Europe (Germany, Sweden, and Italy), both types of AIP, ie, LPSP (AIP without GELs) and IDCP (AIP with GELs), are observed. Reports from London suggest predominantly LPSP-like profile of AIP. Mayo Clinic series had 78 patients with LPSP-like clinical profile and 19 with IDCP-like clinical profile. A series of 88 patients from Germany, Italy, and Belgium who were treated by pancreatic resection showed a predominance of LPSP (60%) over IDCP (40%).
The proportion of LPSP versus IDCP among patients labeled as having AIP varies substantially among centers across the world. Whether this reflects a true geographic difference in the incidence of these 2 forms of AIP is not yet clear.
Question: Should the Clinical Phenotypes Associated With LPSP and IDCP Be Referred to as Type 1 and Type 2 AIP?
This question induced a vigorous debate. The center of the controversy related to the use of the term "autoimmune" for IDCP. The Japanese experts contended that there is a strong evidence to support LPSP as an autoimmune disorder (hypergammaglobulinemia, prevalence of autoantibodies, albeit non-specific, and a steroid-responsive lymphoplasmacytic infiltrate), whereas IDCP lacks such evidence. They believed that because of the relative paucity of data concerning IDCP, it was premature to label it as an autoimmune disorder. They further noted that unlike LPSP, which is commonly associated with extrapancreatic manifestations, IDCP seems to occur in isolation, except for a potential association with inflammatory bowel disease, particularly ulcerative colitis.
Others contended that the 2 diseases had many similarities. The most common clinical presentation of both diseases is with obstructive jaundice with a pancreatic enlargement/mass. However, in the AIP patients with LPSP (unlike in IDCP), obstructive jaundice is caused by the specific pathological change, ie, sclerosing cholangitis with a similar pathological feature to LPSP. The few pancreatograms in IDCP that were shown during the meeting were not distinguishable from those seen in LPSP. Also, IDCP is associated with a periductal lymphoplasmacytic infiltrate and storiform fibrosis, although to a lesser extent than LPSP. In fact, many pathologists in the concordance study could not distinguish LPSP from IDCP. Finally, IDCP is also steroid responsive. Many experts believed therefore that IDCP may be an organ (pancreas)-specific autoimmune disorder. However, the occurrence of neutrophils infiltrating some ducts and acini (Fig. 6) remains so far unexplained by an autoimmune mechanism.
At least some experts believe that the overlap between the 2 forms of AIP is further confounded by the fact that IDCP and LPSP may not always be distinguishable using current diagnostic criteria. For example, both can fulfill Japanese and Asian diagnostic criteria for AIP based on imaging criteria andbiopsy showing lymphoplasmacytic infiltrate with fibrosis, histological features common to both LPSP and IDCP. Similarly, in the Italian criteria, such patients would fulfill the criteria for AIP if they responded to steroids, as both forms do. Similarly, physicians have used steroid trial for diagnosis of AIP using the HISORt criteria, although collateral evidence of AIP in the form of raised serum IgG4 or other organ involvement (features of LPSP) is necessary before steroids are given. A case of AIP from Japan was presented, which resembled IDCP in clinical profile (young, seronegative, and associated with inflammatory bowel disease) but met the Japanese/Asian diagnostic criteria noted earlier. A histologically confirmed case of IDCP from Korea was presented whose imaging features resembled that seen in LPSP and which responded dramatically to steroids with normalization of imaging abnormalities.
The expert panel acknowledged that the current practice is to refer to both disease entities as AIP owing to the inability to differentiate LPSP from IDCP without histological findings and review by an experienced pathologist. There is, therefore, a clear need to nosologically distinguish these entities to avoid continued confusion between them, to help provide prognostic information, and to guide patient care. This would also provide the framework for future research in the field, including identification of specific biomarkers of both entities.
The European and American experts favored the continued inclusion of IDCP as a unique type of AIP owing to the similar clinical presentations, overlapping diagnostic criteria (including histological characteristics) and similar response to steroid administration. The terms type 1 and type 2 AIP have recently been introduced into the literature to refer to the clinical profiles associated with LPSP and IDCP, respectively.21,22 Most, but not all present, agreed that this terminology best reflected our current state of knowledge. All agreed that as we learn more about both entities, the terminology would surely change.
- a. Diagnostic criteria for AIP should recognize that there are 2 forms of the disease. This will allow further study of and identify specific markers of both forms of AIP.
- b. Currently, the disease associated with IDCP can be definitively diagnosed only by histological examination. The use of steroid trial does not distinguish disease associated with LPSP from IDCP.
- c. Although uniform consensus was not achieved, most experts agreed that the clinical phenotypes associated with the histopathologic patterns of LPSP (AIP without GELs) and IDCP (AIP with GELs) should be referred to as type 1 and type 2 AIP, respectively.
1. Sarles H, Sarles JC, Muratore R, et al. Chronic inflammatory sclerosis of the pancreas-an autonomous pancreatic disease? Am J Dig Dis
2. Kawaguchi K, Koike M, Tsuruta K, et al. Lymphoplasmacytic sclerosing pancreatitis with cholangitis: a variant of primary sclerosing cholangitis extensively involving pancreas. Hum Pathol
3. Yoshida K, Toki F, Takeuchi T, et al. Chronic pancreatitis
caused by an autoimmune abnormality. Proposal of the concept of autoimmune pancreatitis
. Dig Dis Sci
4. Society MotCCfAPotJP. Diagnostic criteria for autoimmune pancreatitis
by the Japan Pancreas Society. J Jpn Pancreas (Suizou)
5. Okazaki K, Kawa S, Kamisawa T, et al. Clinical diagnostic criteria of autoimmune pancreatitis
: revised proposal. J Gastroenterol
6. Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4
concentrations in patients with sclerosing pancreatitis. New Engl J Med
7. Kamisawa T, Egawa N, Nakajima H. Autoimmune pancreatitis
is a systemic autoimmune disease. Am J Gastroenterol
8. Kwon S, Kim MH, Choi EK. The diagnostic criteria for autoimmune chronic pancreatitis
: it is time to make a consensus. Pancreas
9. Otsuki M, Chung JB, Okazaki K, et al. Asian diagnostic criteria for autoimmune pancreatitis
: consensus of the Japan-Korea Symposium on Autoimmune Pancreatitis
. J Gastroenterol
10. Kamisawa T, Shimosegawa T, Okazaki K, et al. Standard steroid treatment for autoimmune pancreatitis
11. Ectors N, Maillet B, Aerts R, et al. Non-alcoholic duct destructive chronic pancreatitis
12. Ball WP, Baggenstoss AH, Bargen JA. Pancreatic lesions associated with chronic ulcerative colitis. Arch Pathol (Chic)
13. Pearson RK, Longnecker DS, Chari ST, et al. Controversies in clinical pancreatology: autoimmune pancreatitis
: does it exist? Pancreas
14. Notohara K, Burgart LJ, Yadav D, et al. Idiopathic chronic pancreatitis
with periductal lymphoplasmacytic infiltration: clinicopathologic features of 35 cases. Am J Surg Pathol
15. Chari ST, Smyrk TC, Levy MJ, et al. Diagnosis of autoimmune pancreatitis
: the Mayo Clinic experience. Clin Gastroenterol Hepatol
16. Zamboni G, Luttges J, Capelli P, et al. Histopathological features of diagnostic and clinical relevance in autoimmune pancreatitis
: a study on 53 resection specimens and 9 biopsy specimens. Virchows Arch
17. Kloppel G. Chronic pancreatitis
, pseudotumors and other tumor-like lesions. Mod Pathol
. 2007;20(suppl 1):S113-S131.
18. Deshpande V, Chicano S, Finkelberg D, et al. Autoimmune pancreatitis
: a systemic immune complex mediated disease. Am J Surg Pathol
19. Zhang L, Notohara K, Levy MJ, et al. IgG4
-positive plasma cell infiltration in the diagnosis of autoimmune pancreatitis
. Mod Pathol
20. Sah RP, Chari ST, Pannala R, et al. Comparison of clinical profile and natural history of type 1 and type 2 autoimmune pancreatitis
. (in press).
21. Park DH, Kim MH, Chari ST. Recent advances in autoimmune pancreatitis
22. Sugumar A, Kloppel G, Chari ST. Autoimmune pancreatitis
: pathologic subtypes and their implications for its diagnosis. Am J Gastroenterol
. 2009;104:2308-2310; quiz 2311.
Keywords:© 2010 Lippincott Williams & Wilkins, Inc.
autoimmune pancreatitis; IgG4; chronic pancreatitis