ERCP findings in idiopathic pancreatitis: patients who are cystic fibrosis gene positive and negative

Transcription

1 ORIGINAL ARTICLE ERCP findings in idiopathic pancreatitis: patients who are cystic fibrosis gene positive and negative Waleed M. Alazmi, MD, Evan L. Fogel, MD, Suzette Schmidt, RN, James L. Watkins, MD, Lee McHenry, MD, Stuart Sherman, MD, Glen A. Lehman, MD Indianapolis, Indiana, USA Background: Our series of patients with idiopathic pancreatitis (IP) found a cystic fibrosis (CF) gene abnormality in 19% compared with 3.5% in patients without pancreatitis. Objective: The objective was to determine whether the CF gene predicts more severe ERP findings. Design: This was a retrospective case-control study. Setting and Patients: From July 1998 to August 2004, CF gene analysis was performed in 819 patients with IP via Genzyme Genetics. The panel tests for 70 to 87 alleles and has a detection rate of more than 90% of the cases. Sixty-nine patients (8.4%) who had at least one CF gene positive mutation were the study cohort. A total of 218 patients with IP and negative CF gene mutation were randomly selected from our database to be in the control group. Main Outcome Measurements: Pancreatograms were evaluated for chronic pancreatitis (CP) based on Cambridge criteria. The results of the gene analysis were not available at the time of pancreatogram interpretation. Results: Among patients positive for the CF gene, 42 (61%) were women. The mean age at intervention was 40 years (range years), and 48 patients (70%) had cholecystectomy. Among patients who were negative for the CF gene, 147 (67%) were women. The mean age at intervention was 41 years (range 9-89 years), and 125 patients (57%) had cholecystectomy. Compared with controls, cases had higher incidence of CP (62% vs. 48%, p Z 0.05), grade III CP (35% vs. 18%, p Z 0.004), pseudocysts (12% vs. 4%, p Z 0.036) and pancreatic strictures (20% vs. 8%, p Z 0.008). Limitations: The limitations of the study were (1) retrospective design and (2) the panel used tests only for 70 to 87 alleles (of approximately, 900 CF transmembrane conductance regulator genes known). Conclusions: The mean age at intervention in both groups was similar. CP, grade III CP, pseudocysts, and pancreatic strictures were more common among patients who were CF gene positive. (Gastrointest Endosc 2006; 63:234-9.) Cystic fibrosis (CF) is the most common fatal autosomal recessive disease affecting white populations, with an incidence of 1 in 2000 to 3000 births. 1 A disease incidence of this magnitude implies that approximately 1 in 30 individuals is a carrier of the recessive genetic defect. The CF gene was mapped to chromosome 7q31.2 and was cloned in ,3 The protein encoded by the CF gene is designated as the CF transmembrane conductance regulator (CFTR), which activates a cyclic AMP-regulated See CME section; p Copyright ª 2006 by the American Society for Gastrointestinal Endoscopy /$32.00 doi: /j.gie chloride channel protein of 1480 amino acids. 2 The major mutation that causes CF is a 3-base-pair deletion, resulting in the loss of phenylalanine 508 (DF508). 3 This mutation accounts for about 70% of mutant chromosomes worldwide. More than 900 mutations in the CFTR gene have been described. 4 The leading category of chronic pancreatitis (CP) in nonalcoholic adults is idiopathic CP (ICP). 5 In 1998, two groups reported a strong association between CFTR mutations and ICP. 6,7 Other groups have confirmed this association In pediatric populations, the homozygous CFTR mutations (classic CF) account for 10% to 15% of otherwise categorized idiopathic pancreatitis (IP). In adult populations, homozygous state accounts for less than 1% of new 234 GASTROINTESTINAL ENDOSCOPY Volume 63, No. 2 :

2 Alazmi et al Idiopathic pancreatitis: cystic fibrosis diagnoses in patients with IP. 19,20 Heterozygous CFTR mutations initially were thought to indicate a disease-free carrier state; however, 11% to 39% of patients with ICP carry a single identified CFTR mutation. 6-8,14-18 Heterozygous carriers, therefore, are at greater risk for IP and may be subjected to compound injury from alcohol, drugs, or other risk factors. It is unknown whether a single CFTR mutation is associated with more severe CP. The aim of this study was to determine whether a CFTR mutation would predict more severe findings on endoscopic retrograde pancreatogram. PATIENTS AND METHODS The Institutional Review Board of Indiana University Purdue University of Indianapolis approved this study. A case-control study design was used, in which patients with IP and with a positive CFTR mutation (cases) were compared with a control group (patients with IP and negative CFTR mutation analysis). The control group was selected by computer randomization. Cases and controls were taken from a database of patients referred to the Indiana University Medical Center for evaluation with ERCP from 1998 to Cases were chosen from a subgroup of 819 outpatients with acute, recurrent, or CP that had CFTR gene mutation tested by Genzyme Genetics (Westborough, Mass). In patients with a single episode of pancreatitis, CFTR mutation analysis was performed at the discretion of the managing physician, whereas CFTR mutation analysis was performed as a clinical routine in all patients with recurrent episodes. Criteria for the diagnosis of acute pancreatitis included characteristic abdominal pain and one or more of the following features: serum amylase more than 4 times the upper limit of normal; serum lipase more than 3 times the upper limit of normal; and/or pancreatic inflammation or necrosis demonstrated on abdominal CT. Patients with two or more bouts of pancreatitis were labeled as having acute recurrent pancreatitis. After a thorough diagnostic evaluation to exclude other causes of chronic abdominal pain and/or steatorrhea, a diagnosis of CP was established if pancreatic calcification, typical histologic changes, or characteristic findings on ERCP were present. Patients with a known or discovered cause of pancreatitis (e.g., hypertriglyceridemia, hypercalcemia), either before or after ERCP evaluation, were excluded. The remaining patients for whom pancreatitis was unexplained formed the case group. For the purpose of this study, IP was defined as failure to identify a cause for pancreatitis after a thorough history and physical examination, laboratory data, noninvasive imaging studies (e.g., CT), and ERCP. Three controls per case were randomly selected from the same database of patients who underwent ERCP for either acute or chronic IP during the study period. The exposure variable, which was assessed identically in the case and control groups, was the presence of Capsule Summary What is already known on this topic d d Of patients with ICP, 11% to 39% carry a single cystic fibrosis transmembrane conductance regulator (CFTR) mutation. CF gene abnormalities may be associated with more severe ERP findings. What this study adds to our knowledge d In a case-controlled, retrospective study from a tertiary referral center, CP, pseudocysts, and pancreatic strictures are more common among patients who are CF gene positive. CFTR mutations. For this analysis, deoxyribonucleic acid was extracted from blood samples and tested for 70 to 87 common mutations via Genzyme s expanded mutation panel. The different mutations tested over the past 6 years are listed in Tables 1 and 2. These mutations comprise more than 90% of the CF gene mutations in whites of northern European origin. Because testing for the poly (T) variant helps provide valuable information for counseling patients who carry the R117H mutation, Genzyme reflexively tested for poly (T) on all individuals who were R117H positive. Patients were not routinely tested for serine protease inhibitor Kazal type-1 gene (SPINK1) mutations or for cationic trypsinogen gene (PRSS1) mutations. However, patients who had a positive gene for hereditary pancreatitis (PRSSI, SPINK1) and those who were homozygous for CFTR were excluded. Demographic, clinical, and endoscopic features of patients with IP and CFTR mutations were compared with patients with IP without mutations. The specific features included age of first ERCP; ERCP findings; and complications of pancreatitis, such as pancreatic-duct stricture, pseudocyst, and pancreatic-stone disease. Pancreatograms, obtained during the first ERCP performed at Indiana University Medical Center on the patients, were assessed for the presence and the severity of CP according to the published criteria 24 by 5 experienced endoscopists who were unaware of patients CFTR status. Pancreatography was performed via the major and/or the minor papilla with injection of contrast medium (60% diatrozate meglumine [Renografin 60; Bracco Diagnostics Inc, Princeton, NJ]). A stricture was defined as a main-duct narrowing with partial or complete obstruction to contrast flow with or without associated stone or pseudocyst. Nearly all patients had a CT. Calcifications and/or pseudocysts seen per CT but not definitely seen per ERP were tallied along with ERP findings. Pancreatograms were read with other full clinical data available. Controversial findings were reviewed by two or more endoscopists, and consensus final diagnosis was reached. Volume 63, No. 2 : 2006 GASTROINTESTINAL ENDOSCOPY 235

3 Idiopathic pancreatitis: cystic fibrosis Alazmi et al TABLE 1. Mutations in Genzyme expanded mutation panel ( ) DF508 K710X R352Q Y563D 3120C1GOA 621C1GOT DI507 N1303K R553X 1078delT 3120GOA 711C1GOT A455E P574H R560T 1677delTA 3659delC W1310X* A559T Q1238X S1196X GOA 3662delA W1316X* C524X Q493X S1255X GOA 3791delC Y122X* E60X Q890X S364P 1898C1GOA 3821delT 2909delT* G178R R1158X S549N 2043delG 3849C10kbCOT 3358delAC* G330X R1162X S549R 2183delAAOG 3849C4AOG 3750delAG* G542X R117H V520F 2184delA 3905insT 556delA* G551D R334W W1089X 2307insA 405C1GOA G85E R347H W1282X 2789C5GOA 444delA I148T R347P Y1092X 2869insG 574delA *Mutations not tested in the new panel ( ). TABLE 2. Mutations in Genzyme expanded mutation panel ( ) DF311* I148T R117H S549R 1949del84* 3849C10kbCOT DF508 K710X R1283M* T338I* 2043delG 3849C4AOG DI507 L206W* R334W V520F 2143delT* 3876delA* A455E M1101K* R347H W1089X 2184delA 3905insT A559T N1303K R347P W1282X 2307insA 394delTT* C524X P574H R352Q Y1092X 2789C5GOA 405C1GOA D1152H* Q1238X R553X Y563D 2869insG 405C3AOC* E60X Q493X R560T 1078delT 3120C1GOA 444delA G178R Q552X* S1196X 1161delC* 3120GOA 574delA G330X Q890X S1251N* 1609delCA* 3659delC 621C1GOT G480C* R1066C* S1255X 1677delTA 3662delA 711C1GOT G542X R1158X S364P GOA 3791delC 711C5GOA* G551D R1162X S549I* GOA 3821delT 712-1GOT* G85E R117C* S549N 1898C1GOA Q359K/T360K* G91R* 1898C5GOT* 2183delAAOG *New mutations added to the panel in Patients with equivocal CP were classified as normal. Pancreas divisum was diagnosed only if dorsal and ventral ductography confirmed this congenital anomaly. Patients with head main-duct stricture and possible acquired pancreas divisum were excluded. Pancreatography was successful in all patients. Statistical analysis Summary results of continuous and dichotomous variables were expressed as means. Selected clinical and endoscopic comparisons between groups with vs. without CFTR mutations were made with the Student t test and the chi-square test for continuous and dichotomous 236 GASTROINTESTINAL ENDOSCOPY Volume 63, No. 2 :

4 Alazmi et al Idiopathic pancreatitis: cystic fibrosis variables, respectively. All significant testing was two sided, with a threshold for statistical significance of RESULTS TABLE 3. ERCP findings in cases and controls CF gene positive (n Z 69) CF gene negative (n Z 218) p Value* Normal 26 (38%) 113 (52%) 0.05 Chronic pancreatitis 43 (62%) 105 (48%) 0.05 Grade Iy 11 (16%) 33 (15%) NSz Grade IIy 8 (12%) 14 (6%) NSz Grade IIIy 24 (35%) 39 (18%) Pseudocysts 8 (12%) 9 (4%) Pancreatic strictures 14 (20%) 18 (8%) Pancreatic stones 10 (14.5%) 17 (8%) NSz Biliary strictures 4 (6%) 8 (3.7%) NSz Pancreas divisum 21 (30%) 43 (20%) NSz CF, Cystic fibrosis; NS, not significant. *Based on chi-square test. ycambridge criteria for classification of chronic pancreatitis. znot statistically significant (p O 0.05). From July 1998 to August 2004, CF gene analysis was performed in 819 patients with IP via Genzyme s expanded mutation panel. Sixty-nine patients (8.4%), who had a single (n Z 64) or who were compound heterozygous (n Z 5) for CF gene mutation, were the study cohort. A total of 218 patients undergoing ERCP in association with a diagnosis of IP and negative CF gene mutation analysis were randomly selected from our database to be the control group. ERCP findings The findings on ERCP in patients with CFTR mutations and in those without mutations among patients with IP are shown in Table 3. Compared with the controls, the cases had a higher incidence of CP (62% vs. 48%, p Z 0.05), grade III pancreatitis (35% vs. 18%, p Z 0.004), pseudocysts (12% vs. 4%, p Z 0.036), and pancreatic strictures (20% vs. 8%, p Z 0.008). Demographic data Among the 69 patients positive for the CF gene, 42 (61%) were women, the mean age at intervention was 40 years (range years), and 48 (70%) had cholecystectomy. Among the 218 patients negative for the CF gene, 147 (67%) were women, the mean age at intervention was 41 years (range 9-89 years), and 125 (57%) had cholecystectomy. CFTR gene mutation The CFTR genotypes for the 69 patients with IP are summarized in Table 4. Twelve different mutations were detected: DF508 in 41 patients; R117H in 8 patients; D1270N in 4; and I148T in 4; 3120 C 1(G-A) in two; and 1898C1(G-A), 3659DEL C, D1152H, P574, G542X, N1303K, DI507 each in one patient. Five of the 69 patients found to have a compound heterozygous mutation. None of the 12 patients (8 single mutation and 4 compound heterozygous) who were positive for R117H, tested positive for the 5T allele. The ERCP findings were not statistically different in any subgroup according to their specific CFTR gene mutation. Pancreas divisum Pancreas divisum was seen in 21 patients (30%) with CF gene positive compared with 43 patients (20%) in the CF gene negative group (p Z not significant). DISCUSSION Several studies have addressed the association between CFTR gene mutations and pancreatitis. 6-8,14-18 Selected published series, including the present study, reporting CFTR mutation frequencies in patients with IP are summarized in Table 5. A total of 1239 patients with IP were described. The number of CFTR mutations tested varied among these series and ranged from 13 to 135. The frequency of CFTR mutations ranged from 8.4% to 38.7%, with a mean of 12.8%. Some series 7,14,18 included alcoholic CP patients, with reported frequencies of CFTR mutation that varied from 8.5% to 40.5%. The disparity of the results in these series is likely to be because of the inclusion of heterogeneous groups of patients, as well as the various genetic analyses used in each series. Choudari et al 15 found a trend toward more frequent pancreatic-stone disease among the 19 patients with CFTR mutations. Our series involves 69 patients with CFTR mutations that resulted in the demonstration of the statistical differences in the pancreatogram findings. Compared with controls, cases had higher incidence of CP (62% vs. 48%, p Z 0.05), grade III CP (35% vs. 18%, p Z 0.004), pseudocysts (12% vs. 4%, p Z 0.036), and pancreatic strictures (20% vs. 8%, p Z 0.008). Studies that compared the genotypes and phenotypes of CF patients suggest a relationship between specific gene mutations and the extent of pancreatic disease In this study, 12 different mutations were detected in the 69 patients, and 5 patients were found to have a compound heterozygous mutation. The pancreatogram findings were not statistically different in any subgroup according to their specific CFTR gene mutation. There Volume 63, No. 2 : 2006 GASTROINTESTINAL ENDOSCOPY 237

5 Idiopathic pancreatitis: cystic fibrosis Alazmi et al TABLE 4. CFTR gene mutation analysis and ERCP findings* CFTR gene mutation No. (% of study population) Chronic pancreatitis Pseudocyst Pancreatic stones Pancreatic stricture DF (59.4%) 25 (61%) 6 (14.6%) 5 (12%) 11 (27%) R117H 8 (11.6%) 5 (62.5%) 0 1 (12.5%) 1 (12.5%) D1270N 4 (6%) 2 (50%) 0 1 (25%) 0 I148T 4 (6%) 2 (50%) 1 (25%) C1(G-A) 2 (3%) DF508/R117Hy 3 (4.4%) 3 (100%) 1 (33%) 1 (33%) 1 (33%) DF508/P574y R117H/N1303Ky C1(G-A) DEL C D1152H G542X DI CFTR, Cystic fibrosis transmembrane conductance regulator. *No statistically significant differences were found comparing groups of patients with different CFTR mutation. ycompound heterozygous. TABLE 5. Selected published series reporting CFTR mutation frequency in patients with IP Authors, y No. CFTR mutations tested No. patients included No. patients with IP Frequency of CFTR mutation in IP patients Cohn et al, % Sharer et al, % Ockenga et al, % Truninger et al, % Maire et al, % Frulloni et al, % Choudari et al, % Casals et al, % Present study % Total ,396 1, % CFTR, Cystic fibrosis transmembrane conductance regulator; IP, idiopathic pancreatitis (acute, recurrent acute, or chronic pancreatitis). was a trend toward more pancreas divisum in cases compared with controls (30% vs. 20%). This finding may suggest that other predisposing factors, such as CFTR mutations, may be necessary for pancreatitis to occur in patients with pancreas divisum. Consistent with the previous studies, our study lends further support to the association between CFTR mutation and pancreatitis by extending the findings to another population. Our study contains an internal control group for direct comparison with the case group. The control 238 GASTROINTESTINAL ENDOSCOPY Volume 63, No. 2 :

6 Alazmi et al Idiopathic pancreatitis: cystic fibrosis group was obtained from an ERCP procedure registry. Using the same sampling frame for both case and control groups provides assurance against referral bias as a possible reason for differences between the groups and ensures valid and accurate classification of cases and controls with respect to diagnosis. However, the increase in the number of CFTR mutations tested in the Genzyme panel in 2001 would underestimate the cases, as well as the controls. Thus, it is possible that some of the controls tested before 2001 were false negative. Future research should be directed toward the evaluation of the clinical course and long-term prognosis of patients with CFTR mutations. Another observation that warrants further studies is the fact that only a minority of cases had detectable CFTR mutations in patients with IP. Casals et al 18 tested for 175 alleles and detected the highest frequency (38.7%) of CFTR mutations. This suggests that current genetic panels still miss a major portion of the mutations. More thorough genetic analysis in a large patient population is needed. Our results indicate that patients with IP and with CFTR mutations have more severe CP (especially grade III), pseudocysts, and pancreatic strictures on their pancreatograms. Whether this translates to a worse clinical course, long-term prognosis, and response to endoscopic or surgical intervention needs to be evaluated. Specific therapies for these genetic diseases are awaited. REFERENCES 1. Welsh MJ, Tsui L-C, Boat TF, et al. Cystic fibrosis. In: Scriver CR, Beaudet Al, Sly WS, et al, editors. Metabolic basis of inherited disease. 7th ed. New York: McGraw-Hill; p Riordan JR, Rommens JM, Kerem B, et al. Identification of the cystic fibrosis gene: cloning and characterization of complimentary DNA. Science 1989;245: Kerem B, Rommens JM, Buchanan JA, et al. Identification of the cystic fibrosis gene: genetic analysis. Science 1989;245: CF Genetic Analysis Consortium, 1997 [cited 2005 Apr 1]. Available from URL: 5. Etemad B, Whitcomb DC. Chronic pancreatitis: diagnosis, classification, and new genetic developments. Gastroenterology 2001;120: Cohn JA, Friedman KJ, Noone PG, et al. Relation between mutations of the cystic fibrosis gene and idiopathic pancreatitis. N Engl J Med 1998;39: Sharer N, Schwarz M, Malone G, et al. Mutations of the cystic fibrosis gene in patients with chronic pancreatitis. N Engl J Med 1998;339: Ockenga J, Stuhrmann M, Ballmann M, et al. Mutations of the cystic fibrosis gene, but not cationic trypsinogen gene, are associated with recurrent or chronic idiopathic pancreatitis. Am J Gastroenterol 2000; 95: Arduino C, Gallo M, Brusco A, et al. Polyvariant mutant CFTR genes in patients with chronic pancreatitis. Clin Genet 1999;56: Malats N, Casals T, Porta M, et al. Cystic fibrosis transmembrane regulator (CFTR) deltaf508 mutation and 5T allele in patients with chronic pancreatitis and exocrine pancreatic cancer. PANKRAS II Study Group. Gut 2001;48: Gomez Lira M, Benetazzo MG, Marzari MG, et al. High frequency of cystic fibrosis transmembrane regulator mutation L997F in patients with recurrent idiopathic pancreatitis and in newborns with hypertrypsinemia. Am J Hum Genet 2000;66: Noone PG, Zhou Z, Silverman LM, et al. Cystic fibrosis gene mutations and pancreatitis risk: relation to epithelial ion transport and trypsin inhibitor gene mutations. Gastroenterology 2001;121: Castellani C, Bonizzato A, Rolfini R, et al. Increased prevalence of mutations of the cystic fibrosis gene in idiopathic chronic and recurrent pancreatitis. Am J Gastroenterol 1999;94: Truninger K, Malik N, Ammann RW, et al. Mutations of the cystic fibrosis gene in patients with chronic pancreatitis. Am J Gastroenterol 2001;96: Choudari CP, Imperiale TF, Sherman S, et al. Risk of pancreatitis with mutation of the cystic fibrosis gene. Am J Gastroenterol 2004;99: Maire F, Bienvenu T, Ngukam A, et al. Frequency of CFTR gene mutations in idiopathic pancreatitis [French]. Gastroenterol Clin Biol 2003; 27: Frulloni L, Castellani C, Bovo P, et al. Natural history of pancreatitis associated with cystic fibrosis gene mutations. Dig Liver Dis 2003;35: Casals T, Aparisi L, Martinez-Costa C, et al. Different CFTR mutational spectrum in alcoholic and idiopathic chronic pancreatitis? Pancreas 2004;28: Choudari CP, Lehman GA, Sherman S. Pancreatitis and cystic fibrosis gene mutations. Gastroenterol Clin North Am 1999;28: Cohn JA, Jowell PS. Are mutations in the cystic fibrosis gene important in chronic pancreatitis. Surg Clin North Am 1999;79: Kiesewetter S, Macek M Jr, Davis C, et al. A mutation in CFTR produces different phenotypes depending on chromosomal background. Nat Genet 1993;5: Chu CS, Trapnell BC, Curristin S. Genetic basis of variable exon 9 skipping pancreatitis in cystic fibrosis transmembrane conductance regulator mrna. Nat Genet 1993;3: Massie RJ, Poplawski N, Wilcken B, et al. Intron-8 polythymidine sequence in Australasian individuals with CF mutations R117H and R117C. Eur Respir J 2001;17: Axon AT, Classen M, Cotton PB, et al. Pancreatography in chronic pancreatitis: international definitions. Gut 1984;25: The Cystic Fibrosis Genotype Phenotype Consortium. Correlation between genotype and phenotype in patients with cystic fibrosis. N Engl J Med 1993;329: Santis G, Osborne L, Knight RA, et al. Independent genetic determinants of pancreatic and pulmonary status in cystic fibrosis. Lancet 1990;336: Kristidis P, Bozon D, Corey M, et al. Genetic determination of exocrine pancreatic function in cystic fibrosis. Am J Hum Genet 1992;50: Kerem E, Corey M, Kerem BS, et al. The relation between genotype and phenotype in cystic fibrosis: analysis of the most common mutation (DF508). N Engl J Med 1990;323: Received February 17, Accepted June 29, Current affiliations: Division of Gastroenterology and Hepatology, Indiana University Medical Center, Indianapolis, Indiana, USA. Reprint requests: Glen A. Lehman, MD, Indiana University Medical Center, 550 N. University Blvd, Suite 4100, Indianapolis, IN Volume 63, No. 2 : 2006 GASTROINTESTINAL ENDOSCOPY 239