The new Japanese clinical guidelines for autoimmune pancreatitis (JCGAIP) presented in this issue1 reflect a careful, consensus-building effort to guide and standardize the diagnosis and treatment of patients with autoimmune pancreatitis (AIP) in Japan. These guidelines are based on the extensive and carefully documented experience of Japanese clinicians and pathologists with AIP-a disease that is increasingly recognized worldwide. The guidelines provide an effective review of AIP and merit attention and careful consideration regarding their application internationally. We wish to focus on 2 issues that arise because of differences in experience and practice among clinicians and pathologists in Western nations: (a) recognition of histological and clinical subtypes of AIP and (b) diagnostic criteria for AIP.
The most basic issue is the question of whether there is more than 1 form of AIP in regard to pancreatic pathology, clinical features, and pathogenetic mechanisms. The JCGAIP focus on patients with the histopathologic changes called lymphoplasmacytic sclerosing pancreatitis (LPSP), most of whom have elevations of plasma immunoglobulins, and specifically of IgG4. It has been proposed that AIP is a manifestation of a systemic IgG4-related autoimmune disease.2,3 These patients are typically older than 50 years and male.
Large series reported from Europe and the United States consistently identify a second set of patients with slightly different histopathologic changes that have been called idiopathic duct-centric chronic pancreatitis (IDCP)4 or AIP with granulocytic epithelial lesions (GELs).5 This subgroup of patients is more diverse in age, includes a higher fraction of female patients, and characteristically does not have elevated plasma IgG4.5 These patients may have chronic inflammatory bowel disease such as ulcerative colitis, but typically lack evidence of extrapancreatic involvement of the organs typical of IgG4-related autoimmune disease.4,5
More recently, these 2 patterns have been designated as AIP types 1 and 2.6 In the United States and Europe, type 1 AIP (LPSP) is more common than type 2 (IDCP/AIP with GELs) in series that identify both types.4,5
The JCGAIP specifically deal with type 1 AIP and leave open the question as to whether the type 2 pattern is an autoimmune disease. In turn, this raises the question of what evidence suggests an autoimmune origin for IDCP/AIP with GELs. The following observations provide support. Both types of AIP share a number of histopathologic features-most notably a periductal lymphoplasmacytic infiltrate and a phlebitis. Only the concomitant duct infiltration by neutrophilic granulocytes, the GEL, and a less marked phlebitis distinguish the histological changes of type 2 from type 1 AIP. A fraction of patients with type 2 AIP also has inflammatory bowel disease, and one had multiple sclerosis4,5 (some forms of multiple sclerosis are considered to be of autoimmune origin). Type 2 AIP patients may also show biliary involvement similar to that seen in type 1 AIP (G.K., personal observations). A recent report provides evidence that type 2 AIP is found in association with other autoimmune diseases, such as Evans syndrome and Hashimoto thyroiditis, and responds to steroid therapy.7 Finally, both types 1 and 2 AIP cases reveal a similar increase in the numbers of CD4 and CD8 lymphocytes that infiltrate the pancreas.8 In aggregate, these observations provide a significant support for an autoimmune mechanism in the pathogenesis of type 2 AIP.
The second issue is more clinical-the question of how to establish the diagnosis of AIP. Let us first discuss the diagnosis of type 1 AIP, which is the focus of Japanese (JCGAIP), Asian,9 and American10 criteria. It is clear from the discussion above that type 1 AIP has a distinct pancreatic histology, that is, LPSP. The Japanese diagnostic criteria, however, do not include diagnostic histology as being a valid independent criterion for diagnosis of AIP. If these criteria are strictly followed, one would not be able to diagnose AIP in a patient who undergoes a pancreaticoduodenectomy for suspicion of pancreatic cancer without any clinical workup for AIP and turns out to have LPSP on examination of the resection specimen, a scenario that was common just a few years ago11-13 and may occasionally still occur even in our centers! Now, with availability of core biopsies of the pancreas, LPSP can be diagnosed by histology even without pancreatic resection. Therefore, as suggested in the consensus Asian criteria,9 the presence of LPSP should be sufficient to diagnose AIP regardless of imaging or other findings. However, unlike the Asian consensus criteria, this should not be restricted only to resection specimens. Instead, features necessary to diagnose LPSP, including the role of IgG4 immunostaining, should be elucidated for both resection specimens and core biopsies.
The issue of core biopsies to diagnose AIP has been a point of contention. Whereas the Mayo group has published on the technical aspects and usefulness of endoscopic ultrasound-guided pancreatic core biopsies14 and routinely uses it to establish the diagnosis of AIP, others have not found it as helpful in its ability to get sufficient tissue to make the diagnosis of either AIP15 or nonfocal chronic pancreatitis.16 Although the first liver biopsy was performed in 1893, it took decades for it to be accepted as a diagnostic tool for liver diseases. We believe, as experience with the technique of pancreatic core biopsies and its interpretation increases, it will be accepted as an invaluable tool to diagnose pancreatic disease because there are already very promising results available.7,17
The Japanese (JCGAIP), Asian,9 and American10 criteria differ in the use of imaging to diagnose AIP. In AIP, the pancreas has peculiar parenchymal and ductal features that can be very helpful in its diagnosis. These are well described in the JCGAIP. However, to diagnose AIP, the Japanese and Asian criteria mandate the performance of an endoscopic retrograde pancreatogram (ERP). Diagnostic pancreatograms in patients with obstructive jaundice are not routinely performed in North America. In fact, endoscopists are careful to avoid cannulating the pancreatic duct in patients with suspected pancreatic cancer to reduce the risk of pancreatitis. Therefore, in a practice setting where ERPs are not routinely performed, mandating its use will be hard to implement. More importantly, the Mayo Clinic experience shows that an ERP is not essential to diagnose AIP. In a recent study, it was shown that ERP has very poor sensitivity and very poor interobserver agreement when read by experts familiar with AIP in centers outside Japan.18
So, how can we come to a consensus on the use of imaging to diagnose AIP? One must acknowledge the differences in practices across different centers and continents. Those centers that routinely use ERP to diagnose AIP and that have the expertise to recognize the pancreatographic features of AIP may continue to use it as suggested in the JCGAIP. For those not routinely using diagnostic ERP, an alternative approach to diagnosis of AIP should be available. An approach used at the Mayo Clinic that does not use ERP has recently been published.19
The American criteria10 also differ significantly from the Japanese and Asian criteria in the use of serological markers. Whereas the American criteria use only serum IgG4, the Japanese and Asian criteria use any one of immunoglobulin G, immunoglobulin G4, and autoantibodies such as antinuclear antibody and rheumatoid factor. The JCGAIP document provides important data to understand the utility of serological markers to diagnose AIP. In 100 patients with AIP and 80 patients with pancreatic cancer, the Asian panel had a sensitivity of 97% and a specificity of only 61% for AIP, while IgG4 alone was 86% sensitive and 96% specific for AIP. One must keep in mind that AIP is a rare disease. In a collective experience of 3 studies,11,12,20 only 43 (2.4%) of 1808 resections for presumed pancreatic malignancy had LPSP on histological examination of the resected specimen. Therefore, if the Asian serological panel were used in 100 consecutive patients with obstructive jaundice suspected to be due to pancreatic cancer, we can predict that ∼41 would have a positive serological panel (39 false-positive results in pancreatic cancer and 2 true-positive results in AIP), of which only two would have AIP. In this setting, a test with low specificity is unlikely to be helpful in clinical decision making.
In understanding the limitations of serological markers to diagnose AIP, it is best to tailor their use to findings on cross-sectional imaging. In a patient presenting with obstructive jaundice who has classic parenchymal features on computed tomography or magnetic resonance imaging (diffuse pancreatic enlargement (with or without rim enhancement) and delayed enhancement without a dilated pancreatic duct or low-density mass), the diagnosis of AIP is likely as these findings are highly specific for AIP.19,21 In such patients, minimal collateral serological evidence of AIP (elevated autoantibodies, IgG, or IgG4) may be sufficient to clinch the diagnosis. In patients without typical imaging features, the pretest probability of cancer is high.19 Such patients should have a thorough workup for cancer. If negative, additional strong collateral evidence for AIP should be sought. In such situations, elevation of autoantibodies and mild elevations of serum IgG4 (1-2 times the upper limit of normal) do not have sufficient positive predictive value.22 Therefore, we suggest using serum IgG4 greater than 2 times the upper limit of normal, which has much higher specificity than even serum IgG4 1 to 2 times the upper limit of normal.9,22
Other organ involvement is characteristic of type 1 AIP, consistent with the concept that it is a manifestation of a systemic disease. The presence of other organ involvement is also diagnostically helpful. In a patient with obstructive jaundice, the presence of hilar biliary strictures, retroperitoneal fibrosis, renal lesions, or salivary gland enlargement strongly suggests AIP. This is acknowledged in the JCGAIP that states that "…recognition of these extra-pancreatic lesions should also aid in correct diagnosis of AIP." However, unlike the American criteria, in the Japanese or Asian criteria, other organ involvement cannot be used as collateral evidence to diagnose AIP. We believe that the presence of other organ involvement can be as helpful as, if not more helpful than, serological abnormalities in the diagnosis of AIP and should be included in the diagnostic armamentarium as suggested by the JCGAIP.
Whether response to a trial of steroid treatment should be used as a criterion to diagnose AIP is hotly debated. Whereas the American criteria include steroid trial in select situations, the Japanese criteria forbid the use of steroid trial. In the Asian criteria, the option of steroid trial is available, again in select situations. Steroids are used to treat AIP and when a response is seen, it is both reassuring and supportive of the diagnosis. Although there are legitimate concerns with the facile use of steroids as a diagnostic tool for AIP,23 in select situations its use may be appropriate.9 Rather than forbidding the use of steroids until the diagnosis is secure, we suggest that strict criteria be established to define when use of steroids is appropriate. The Mayo group has recently proposed such criteria.9
The diagnosis of type 2 AIP is currently much more difficult. It can be suspected in subjects with obstructive jaundice who are seronegative and have no other organ involvement typically seen in type 1. However, definitive diagnosis still requires histological confirmation.7 Hopefully, with greater awareness and recognition of the condition, biomarkers will be identified that can help diagnose type 2 AIP without need for histology.
In conclusion, the new Japanese guidelines reflect progress in standardizing the diagnosis and treatment AIP. The recommendations also reveal differences in the experience of Japanese and Western clinicians and pathologists that may require additional consideration for their application internationally, or slight revision to create guidelines that are applicable worldwide. The questions and discussion in the JCGAIP will guide further research regarding the diagnosis and treatment of autoimmune disease in the pancreas. The issues raised here will be discussed by an Autoimmune Pancreatitis International Co-operative Study Group to arrive at an internationally acceptable consensus.
1. Okazaki K, Kawa S, Kamisawa T, et al. Japanese clinical guidelines for autoimmune pancreatitis. Pancreas
2. Kamisawa T, Okamoto A. Autoimmune pancreatitis: proposal of IgG4-related sclerosing disease. J of Gastroenterology
3. Deshpande V, Chicano S, Finkelberg D, et al. Autoimmune pancreatitis: a systemic immune complex mediated disease. Am J Surg Pathol
4. Notohara K, Burgart LJ, Yadov D, et al. Idiopathic chronic pancreatitis with periductal lymphoplasmacytic infiltration: clinicopathologic features of 35 cases. Am J Surg Pathol
5. Zamboni Z, Lüttges 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
6. Park DH, Kim MH, Chari ST. Recent advances in autoimmune pancreatitis. Gut
. 2009 Feb 24. [E-pub ahead of print].
7. Detlefsen S, Drewes A, Vyberg M, et al. Diagnosis of autoimmune pancreatitis by core needle biopsy: application of six microscopic criteria. Virchows Arch
8. Ectors N, Maillet B, Aerts R, et al. Non-alcoholic duct destructive chronic pancreatitis. Gut
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. Chari ST, Smyrk TC, Levy MJ, et al. Diagnosis of autoimmune pancreatitis: the Mayo Clinic experience. Clin Gastroenterol Hepatol
11. Abraham SC, Wilentz RE, Yeo CJ, et al. Pancreaticoduodenectomy (Whipple resections) in patients without malignancy: are they all chronic pancreatitis? Am J Surg Pathol
12. Weber SM, Cubukcu-Dimopulo O, Palesty JA, et al. Lymphoplasmacytic sclerosing pancreatitis: inflammatory mimic of pancreatic carcinoma. J Gastrointest Surg
13. Yadav D, Notahara K, Smyrk TC, et al. Idiopathic tumefactive chronic pancreatitis: clinical profile, histology, and natural history after resection. Clin Gastroenterol Hepatol
14. Levy MJ, Reddy RP, Wiersema MJ, et al. EUS-guided Trucut biopsy in establishing autoimmune pancreatitis as the cause of obstructive jaundice. Gastrointest Endosc
15. Bang SJ, Kim MH, Kim Do H, et al. Is pancreatic core biopsy sufficient to diagnose autoimmune chronic pancreatitis? Pancreas
16. DeWitt J, McGreevy K, LeBlanc J, et al. EUS-guided Trucut biopsy of suspected nonfocal chronic pancreatitis. Gastrointest Endosc
17. Thomas T, Kaye PV, Ragunath K, et al. Efficacy, safety, and predictive factors for a positive yield of EUS-guided Trucut biopsy: a large tertiary referral center experience. Am J Gastroenterol
18. Sugumar A, Levy MJ, Kamisawa T, et al. Utility of endoscopic retrograde pancreatogram (ERP) to diagnose autoimmune pancreatitis (AIP): an international, double blind, randomized, multicenter study. Gastrointest Endosc
19. Chari ST, Takahashi N, Levy MJ, et al. A diagnostic strategy to distinguish autoimmune pancreatitis from pancreatic cancer. Clin Gastroenterol Hepatol
. 2009 May 4. [E-pub ahead of print].
20. Farnell MB, Pearson RK, Sarr MG, et al. Pancreas Cancer Working Group. A prospective randomized trial comparing standard pancreatoduodenectomy with pancreatoduodenectomy with extended lymphadenectomy in resectable pancreatic head adenocarcinoma. Surgery
21. Kamisawa T, Imai M, Yui Chen P, et al. Strategy for differentiating autoimmune pancreatitis from pancreatic cancer. Pancreas
22. Ghazale A, Chari ST, Smyrk TC, et al. Value of serum IgG4 in the diagnosis of autoimmune pancreatitis and in distinguishing it from pancreatic cancer. Am J Gastroenterol
23. Gardner TB, Levy MJ, Takahashi N, et al. Misdiagnosis of autoimmune pancreatitis: a caution to clinicians. Am J Gastroenterol