Use of immunohistochemistry for IgG4 in the distinction of autoimmune pancreatitis from peritumoral pancreatitis

• Published in: Human pathology 2010-05, Vol.41 (5), p.643-652
• Main Authors: Dhall, Deepti, MD, Suriawinata, Arief A., MD, Tang, Laura H., MD, PhD, Shia, Jinru, MD, Klimstra, David S., MD, PhD
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Antigens, CD - immunology; Autoimmune Diseases - blood; Autoimmune Diseases - diagnosis; Autoimmune Diseases - immunology; Autoimmune pancreatitis; Biological and medical sciences; Carcinoma, Pancreatic Ductal - blood; Carcinoma, Pancreatic Ductal - diagnosis; Carcinoma, Pancreatic Ductal - immunology; Diagnosis, Differential; Female; Gastroenterology. Liver. Pancreas. Abdomen; Humans; IgG4; Immunoglobulin G - blood; Immunoglobulin G - immunology; Immunohistochemistry; Investigative techniques, diagnostic techniques (general aspects); Liver. Biliary tract. Portal circulation. Exocrine pancreas; Lymphoplasmacytic sclerosing pancreatitis; Male; Medical sciences; Middle Aged; Other diseases. Semiology; Pancreas - immunology; Pancreatic Neoplasms - blood; Pancreatic Neoplasms - diagnosis; Pancreatic Neoplasms - immunology; Pancreatitis - blood; Pancreatitis - diagnosis; Pancreatitis - immunology; Pathology; Pathology. Cytology. Biochemistry. Spectrometry. Miscellaneous investigative techniques; Plasma Cells - immunology
• ISSN: 0046-8177; 1532-8392
• DOI: 10.1016/j.humpath.2009.10.019

Summary The patients with autoimmune pancreatitis usually present with jaundice and a pancreatic head mass, presumed to have pancreatic cancer, and they often undergo pancreatic resection. Elevated serum IgG4 levels (>135 mg/dL) help to distinguish autoimmune pancreatitis from pancreatic cancer. However, when the biopsy from a pancreatic mass shows dense chronic inflammation and fibrosis and the serum IgG4 level is not available, it presents a diagnostic dilemma whether it represents autoimmune pancreatitis or peritumoral pancreatitis. We performed IgG4 immunohistochemistry on 25 cases of autoimmune pancreatitis-lymphoplasmacytic sclerosing pancreatitis, 7 cases of autoimmune pancreatitis with granulocytic epithelial lesions, 8 cases of nonspecific pancreatitis, 15 cases of pancreatitis associated with pancreatic ductal adenocarcinoma, and 5 biopsies of pancreatic adenocarcinoma with variable inflammation. The distribution of IgG4-positive cells was noted in each case. Eighty-four percent (21/25) of autoimmune pancreatitis-LPSP cases showed diffuse and dense staining for IgG4, with more than 50 positive plasma cells per high-power field (range, 50-150 cells/hpf) in the highest density area. Most (5/7) cases of autoimmune pancreatitis-granulocytic epithelial lesions were negative for IgG4. Thirty-nine percent of nonspecific pancreatitis and peritumoral pancreatitis cases stained positive for IgG4, but the distribution was focal and none of the cases showed more than 50 IgG4-positive cells/hpf in the highest density area of IgG4 staining. IgG4-positive cells in peritumoral pancreatitis and nonspecific pancreatitis cases were closely associated with malignant glands and areas of acute inflammation in some cases. Using a cutoff of 50 IgG4-positive cells/hpf, the sensitivity of IgG4 staining for classical autoimmune pancreatitis-LPSP versus other types of pancreatitis was 84%, the specificity was 100%, and the P value was significant (50 positive cells/hpf) for IgG4 is specifically seen in autoimmune pancreatitis-LPSP, and IgG4 staining along with the histologic features and serum IgG4 levels may be very helpful in diagnosing autoimmune pancreatitis.