Clinical and Genetic Analysis of Chronic Pancreatitis

The aim of the present doctoral thesis is to investigate the clinical and genetic background of chronic pancreatitis in Chinese in Taiwan.

7.1 Introduction and background

7.1.1 Definition of chronic pancreatitis and diagnosis of chronic pancreatitis
Chronic pancreatitis has been defined as a continuing inflammatory disease of the pancreas characterized by irreversible morphologic changes that typically cause pain and/or permanent loss of function. In Western countries, the most common cause of chronic pancreatitis is alcohol abuse. The understanding of the pathogenesis of CP has improved in recent years In recent years, research effort has also focused on the genetic abnormalities that may predispose to CP.
Clinical diagnosis currently depends on identifying defined clinical, functional, morphologic, and histological features .Histopathology remains the golden standard of diagnosis of chronic pancreatitis, especially if the early diagnosis would be made. However, the uneven distribution of fibrosis and the different severity of fibrosis coexist simultaneously, the needle biopsy could not yield good result. Therefore, serrogate criteria for diagnosis of chronic pancreatitis is mandatory.

The present situation descended from the historical limitations in knowledge of etiology, a disproportionate inclusion of patients with decompensated chronic pancreatitis, the complexity of obtaining accurate functional and morphologic data from the pancreas of living patients, and an inability to make an early diagnosis and to follow disease progression. At present, chronic pancreatitis is regarded to an disease of uncertain pathogenesis, unpredictable clinical course, and unclear treatment.

The gold standard
Tissue diagnosis must be the gold standard. Any persistently abnormal inflammatory state or distortion of the normal architecture may serve as strong evidence for chronic pancreatitis, because the correlation with the tissue biopsy “gold standard” is high in the later stages of the disease. Diagnostic pancreatic tissue biopsies are seldom available so that a “gold standard” diagnosis must often be deferred. However, the gold standard must remain a tissue diagnosis. The ideal diagnostic test for chronic pancreatitis should be sensitive, specific, accurate and reliable, widely available, inexpensive, easy to perform, and at no or low risk to the patient. The evaluation of diagnostic methods in current clinical practice must be mindful that a gold standard and definition of chronic pancreatitis remains in question.
For tissue diagnosis,obtaining tissue for evaluation is a primary consideration. Surgical sampling of the pancreas through laparoscopy or laparotomy can safely provide significant amounts of tissue for confirming the diagnosis of chronic pancreatitis. Percutaneous core needle biopsy, guided by either ultrasonography or CT, has been successfully performed for more than 2 decades with a reported complication rate about 1%. However,some etiologies of chronic pancreatitis, including alcohol, hereditary, autoimmune, and “idiopathic” pancreatitis might display patchy abnormalities in pancreatic parenchyma raising the possibility that a single random biopsy may not be diagnostic. Thus, a more sensitive and targeted technique than CT scan or routine transabdominal ultrasound examination is being considered for directed biopsies. EUS-guided biopsies may have similar or better levels of safety as percutaneous biopsy but it has not been perfected owing to possible inadequate specimen and is not widely available.
The key histopathologic features of CP (regardless of etiology) are pancreatic fibrosis, acinar atrophy, chronic inflammation, and distorted and blocked ducts.Additional distinctive histologic features have been described in some forms of CP such as prominent lymphocytic and plasma cell infiltrate in autoimmune pancreatitis.
The diagnosis of chronic pancreatitis can be made by histological or morphologic criteria alone ( calcification in the parenchyma of pancreas) or by a combination of morphologic, functional, and clinical changes. Functional abnormalities alone are not diagnostic of chronic pancreatitis because these tests do not differentiate chronic pancreatitis from pancreatic insufficiency without pancreatitis. However, there are some criteria including abnormal function testing in the diagnostic criteria such as those proposed by Japanese Pancreas Society (JPS).
Furthermore, some distinct type of chronic pancreatitis would not with pancreatic insufficiency . The diagnosis of severe chronic pancreatitis with extensive calcifications (and/or ductal dilation) is simple, even made by a plain abdominal X-rays. The difficulty in diagnosis arises in patients with early, mild, or minimal change pancreatitis, characteristic pancreatic pain alone, patients in whom chronic pancreatitis is being differentiated from a pancreatic malignancy, and in patients with a recent episode of acute pancreatitis.
The correlation between structural and functional impairment of the pancreas in CP is often poor. Patients with severe exocrine insufficiency may have a largely normal pancreatic structure and vice versa. In general, advanced stages of CP may be diagnosed easily by imaging procedures such as ultrasound, computerized tomography (CT), magnetic resonance tomography, and magnetic resonance cholangiopancreatography (MRCP). In contrast, the diagnosis of early disease presents a considerable challenge.The biochemical, structural, and functional parameters used to assist in the diagnosis of CP are described below:
Imaging modalities are commonly used for the evaluation of pancreatic disease( CT, ERP, EUS, and MRI). Chronic pancreatitis with calcifications can also be identified on abdominal x-rays or by transabdominal ultrasound examination, and when present, the diagnosis of chronic pancreatitis can be made.These techniques are used as inexpensive initial screening techniques in some centers.
The abdominal CT scan should be the first test in the evaluation of possible chronic pancreatitis because it is noninvasive, widely available, and has relatively good sensitivity for diagnosing usually late moderate-to-severe chronic pancreatitis. It is useful in identifying most complications of chronic pancreatitis and in identifying pancreatic malignancy. Pancreatitis is diagnosed by CT with the identification of pathognomonic calcifications within the pancreatic ducts or parenchyma, and/or dilated main pancreatic ducts combined with parenchymal atrophy.
Endoscopic retrograde pancreatography(ERP) could detect the morphology of pancreatic duct. However, false positive might be obtained in elder patients who may develop benign pancreatic duct changes without pancreatitis and in patients with recent acute pancreatitis . On the other hand, in patients with chronic pain typically originated from pancreas with a histologic gold standard of chronic pancreatitis may with a normal appearance of pancreatic duct. These conditions decrease the sensitivity and specificity of ERP for diagnosis of chronic pancreatitis. So, most of the scoring system for diagnosis of chronic pancreatitis did not regard the ERP as the essential or independent diagnostic criteria.
The role of endosonography (EUS)for diagnosing early stage CP is not well defined. Several endosonographic features have been noted in patients with chronic pancreatitis such as reduced or increased echogenicity, increased lobulation, and alteration of small and large ducts, have been described. It is accepted that the absence of these criteria reliably rules out CP, whereas the presence of 5 or 6 criteria strongly indicates the diagnosis. The significance of less than 5 criteria, however, is unclear. Beside, a nonhomogeneous echo structure is not a specific sign for CP but also can be seen in the normal pancreas, especially in the elderly.
The pancreas has significant functional reserve, so that it must be damaged significantly before functional loss is clinically recognized. Invasive tests of pancreatic function are the gold standard for determining exocrine pancreatic function. However, pancreatic function testing is not diagnostic of chronic pancreatitis, but rather serves as a sign of chronic pancreatitis and a measure of the severity of injury. Abnormal function test results alone are not diagnostic of chronic pancreatitis in the scoring systems from the Mayo Clinic or Lüneburg Clinic. Although an abnormal secretin test does meet diagnostic criteria for chronic pancreatitis in the JPS criteria, this criteria has been questioned (asia-pacific consensus) .

7.1.2 Classification of chronic pancreatitis
Classification of the various forms of chronic pancreatitis within this definition is likewise challenging. An ideal disease classification system for chronic pancreatitis would be simple, objective, accurate, and relatively noninvasive, incorporating etiology, pathogenesis, structure, function, and clinical status into one overall schema. Although these criteria have never been met, several systems have been advanced. The most widely used classification systems include the Marseille classification of 1963 with revisions in 1984 and 1988 and the Cambridge classification of 1984.
The Marseille system distinguishes acute from chronic pancreatitis :acute pancreatitis must lead to clinical, histologic, and functional resolution of disease once the instigating factor has been removed. Chronic pancreatitis, on the other hand, requires permanent histologic irregularity, often (but not always) associated with persistent clinical and functional impairment. This system relies heavily on the presence of characteristic pain. The two revisions in the original classification attempted to better categorize histopathologic findings to intimate common etiologic pathways leading to disease. The most recent Marseille–Rome classification of 1988 includes more causal factors. However, this system proves more useful in defining chronic pancreatitis than serving as a classification system. The Cambridge classification uses imaging tests to provide a grading system of pancreatic duct. However, these classifications are not satisfactory to distinguish different forms of chronic pancreatitis , not to distinguish patients or the functional abnormalities associated with etiologies. Thus, the Cambridge system proves more useful as a staging system once the diagnosis is made rather than a system for classifying the etiologies of chronic pancreatitis. There are observations emphasize the reality that different disorders causing similar-appearing injury to the pancreas may follow different clinical courses. Thus, clear definitions of chronic pancreatitis, an etiology-based classification system, and functional, structural, and morphologic staging systems are needed. Therefore, a TIGAR-O system was proposed for etiology/classification/risk assessment of chronic pancreatitis: toxic-metabolic causes, idiopathic, genetic, autoimmune, recurrent severe acute pancreatitis–associated chronic pancreatitis, and obstructive chronic pancreatitis.

7.1.3 Clinical features of CP
Abdominal pain is the most vexing clinical problem. Severe pain decreases appetite, thereby contributing to malnutrition and weight loss. The pain is usually epigastric in location (although more diffuse pain in the upper abdomen can occur) and may radiate to the back. Although recurrent (type A) or continuous (type B) pain is considered to be the hallmark of CP, a subgroup of patients may have no pain at all, presenting instead with symptoms of pancreatic insufficiency. In patients with known CP, pain also may result from an acute attack of pancreatitis, from a pancreatic pseudocyst, portal or splenic vein thrombosis, or bile duct obstruction (associated with jaundice). Associated peptic ulcers also may contribute to pain in patients with chronic pancreatitis.
Steatorrhea and weight loss are further important features of CP. Steatorrhea is a symptom of advanced disease and does not occur until pancreatic lipase secretion is reduced to less than 10% of normal. Maldigestion of lipids occurs earlier than that of other nutrients (proteins and carbohydrates). In addition to exocrine insufficiency, diabetes mellitus may develop in the long-term course of the disease. The diabetes is is characterized by destruction of both insulin- and glucagon-producing cells. Co-existing deficiency of glucagon synthesis usually aggravates hypoglycemic situations in pancreatic diabetes..
7.1.4 Epidemiology of chronic pancreatitis
Generally,epidemiological studies are lack in the world. The reported incidence, prevalence, and manifestations of chronic pancreatitis with overt disease underestimate the true spectrum of this disorder. Earlier series from Denmark(Copenhagen), the United States, reported similar incidence of about 4 per 100,000 inhabitants per year. In a study from Japan, 68% of patients with chronic pancreatitis were diagnosed with the use of computed tomography (CT), endoscopic retrograde cholangiopancreatography (ERCP), or other advanced techniques leading to an estimated overall prevalence rate of 45.4 per 100,000 in males and 12.4 per 100,000 in females. Readdressing the epidemiology of chronic pancreatitis using new criteria and better techniques will be important for determining public health care policy and appropriating resources for clinical care and research.
7.1.5 New advance of chronic pancreaitis
More than 50 years ago, it was recognized for the first time that CP may cluster in selected families, suggesting an inherited disease in these patients. The underlying genetic defect, however, remained obscure for more than 4 decades. In 10%–30% of patients suffering from CP, no apparent underlying cause, including heredity, can be identified. Recent research indicates that a significant percentage of these patients with so-called idiopathic CP may also have a genetic basis for their condition.
Several groups located a gene for hereditary pancreatitis on the long arm of chromosome 7 (7q35) by linkage analysis. mutations in the cationic trypsinogen gene, also referred to as serine protease 1 (PRSS1) (OMIM 276000), was identified as a possible underlying defects. R122H appears to be the most common PRSS1 mutation observed worldwide. Subsequent studies have reported other PRSS1 alterations including p.A16V, p.N29I, p.N29T, p.R116C, and p.R122C, as well as several others, in families with suspected hereditary pancreatitis or in patients without a family history. The functional relevance of PRSS1 mutations has been examined by studies using recombinant cationic trypsinogen subjected to site-directed mutagenesis.Thus, gain-of-function mutations leading to enhanced intrapancreatic trypsinogen activation may be the common initiating step of pancreatitis caused by PRSS1 mutations, whereas stabilization of trypsin may be an accessory mechanism.The serine protease inhibitor, Kazal type 1 (SPINK1) (OMIM 167790), also known as pancreatic secretory trypsin inhibitor, is thought to be a potent inhibitor of intrapancreatic trypsin activity.
Genes regulating trypsinogen activation/inactivation and cystic fibrosis transmembrane conductance regulator (CFTR) function have received particular attention. Mutations in these genes are now increasingly recognized for their potential ‘disease modifier’ role in distinct forms of CP such as tropical, and idiopathic pancreatitis.
7.2 Clinical study of chronic pancreatitis in Chinese in Taiwan
There are few reports in clinical study of chronic pancreatitis in Chinese. Most of the studies in chronic pancreatitis originated from western countries or Japan or India in Asia. With the advance and changes of the concept of pathogenesis and classification in chronic pancreatitis, it is necessary and mandatory to understand the current status of chronic pancreatitis in our Chinese.
We reviewed the medical records, morphological/imaging studies and clinical course of each patients with diagnosed as chronic pancreatitis in hospital registry system and also in pathological database between 1994-2005 in National Taiwan University Hospital( a tertiary referred medical center in Taiwan). We validated the accuracy of diagnosis of chronic pancreatitis during this period and also re-analyzed each case according the TIGAR-O system. The demographic data, etiology, and course of different etiology of chronic pancreatitis were collected and analyzed.
There were initially 536 patients coded as chronic pancreatitis in the discharge register system. However, only 330 patients met the criteria of chronic pancreatitis based on histopathological, morphological and clinical criteria. Among them, the etiology, age, gender and increased cased number each year were shown in Figure. On the contrast, there were initially 238 patients coded as chronic pancreatitis in the pathological database. The coding error in pathological database mostly resulted from the wrong given clinical diagnosis of chronic pancreatitis on the pathology requesting from. There were only 126 patients considered chronic pancreatitis after excluding 53 cases with periampullary or any malignancy. In both databases, male gender was predominant. The median age of each type of chronic pancreatitis was similar to previous reports from western countries. The distribution of age also clustered in the middle age. In the idiopathic chronic pancreatitis, there were two peak of onset of age. Idiopathic pancreatitis was the dominant type of chronic pancreatitis in the pathological databases and most of them did not have calcification in the pre-pathologcial imaging diagnosis.
7.3 Genetic analysis of chronic pancreatitis in Chinese in Taiwan
7.3.1 Genetic study of CFTR mutation/polymorphism/haplotype in idiopathic chronic pancreatitis
The cystic fibrosis transmembrane conductance regulator (CFTR) is a cAMP regulated Cl– channel that is expressed in many epithelial tissues. Loss of CFTR (MIM# 602421) gene function due to mutations may causes cystic fibrosis (CF). In addition, there are several other phenotypes associated with CFTR gene mutations, such as congenital bilateral absence of the vas deferens (CBAVD) and idiopathic chronic pancreatitis. More than 1000 different mutations and 200 polymorphisms have already been reported in cystic fibrosis patients (http://www.genet.sickkids.on.ca/cftr/app). Classic cystic fibrosis (CF) is caused by two loss-of-function mutations in the cystic fibrosis transmembrane conductance regulator gene, whereas patients with nonclassic CF have at least one copy of a mutant gene that retains partial function of the CFTR protein. The so-called “mild “ mutations that retain higher CFTR function have been reported to be associated with idiopathic pancreatitis. These reports suggest that different types of mutations in the CFTR gene, causing different degrees of CFTR impairment, can be associated with widely different disease phenotypes. It is now well recognized that the spectrum of CFTR related disease is much broader than previously thought.
One obstacle to comprehensive CFTR studies is the large size of its 27 exon spanning gene. So, most studies that have addressed the role of CFTR mutations in chronic pancreatitis used commercially available screening panels of the 20–30 most common cystic fibrosis causing (rather than pancreatitis causing) CFTR mutations. Furthermore, the geographic distribution of CFTR mutations also varies worldwide.
In addition to mutations, functional polymorphisms have also been detected within the CFTR locus in non-CF individuals that could also potentially result in alterations in CFTR gene expressiony. Haplotype-based analysis has gained increasing attention in evaluating candidate gene in various clinical situations. Merits of haplotype study are more evident in the fine mapping of complex diseases and in identifying genetic variations that influence individual’s susceptibility to diseases and responses to therapy. Therefore, it is advisable to use the haplotype-based approach to address the role of CFTR variants in the risk of idiopathic chronic pancreatitis.
In order to determine whether CFTR mutations, functional polymorphisms and haplotypes affect the risk of developing idiopathic chronic pancreatitis in Chinese, we recruited 78 patients with idiopathic chronic pancreatitis (ICP) and 200 geographically and ethnically matched controls. All patients and control subjects were Chinese an from the same, narrowly confined area in Taiwan. We analyzed the entire coding region and intronic regions of the CFTR gene by denaturing high-performance liquid chromatography (DHPLC), a simple, rapid, non-gel–based, non-fluorescence–based method that uses ion-pair reversed-phase liquid chromatography for detection of DNA variations sensitively and specifically. We also genotyped the common functional SNPs of CFTR by direct sequencing in all subjects and controls. By this means, we could firmly determine that compound heterozygous mutation carriers, as well as heterozygotes with mild and uncommon mutations (previously regarded as unaffected cystic fibrosis carriers) have an increased risk of developing chronic pancreatitis. Furthermore, we analyzed the CFTR haplotype patterns and genotype-phenotype correlations in idiopathic chronic pancreatitis.
Cases and Controls
Patients and controls were recruited at a tertiary referral center (National Taiwan University Hospital) from July 2000 through Dec 2005. All patients and control subjects were Chinese and from the same, narrowly confined area in Taiwan. The control subjects were from the same areas as the cases, were recruited during the same time period from the hospital, and had no history of acute pancreatitis, chronic pancreatitis, diabetes mellitus, pancreatic adenocarcinoma and or apparent biliary or pancreatic diseases. They were also no family history of pancreatitis and pancreatic adenocarcinoma. Chronic pancreatitis was defined as the presence of pancreatic calcifications or histological evidence of chronic pancreatitis. The etiology of chronic pancreatitis was classified with TIGAR-O system. All of Non–insulin-dependent diabetes mellitus was defined as fasting blood glucose >126 mg/dL and/or being administered at least one oral hypoglycemic agent. Patients who reported any known risk factor for pancreatitis, such as a history of nutritive (including alcohol consumption of more than two drinks a day (20 g), biliary, metabolic (such as elevated serum triglycerides, calcium), autoimmune, medication with known pancreatic toxicity, pancreas or periampullary tumors, pancreatic duct anomaly or endocrine disorders, as well as subjects with any evidence of pulmonary diseases or cystic fibrosis, were excluded from the study. Patients identified as mutation carriers in the cationic trypsinogen (PRSS1) gene and serine protease inhibitor Kazal type (SPINK1) gene evaluated by complete sequencing of the coding region were also excluded. There were seventy-eight cases with idiopathic chronic pancreatitis (48 men and 30 women) and 200 control subjects (125 men and 75 women) recruited for analysis. The study was approved by the local institution committee, and the subjects gave their informed consent. The demographic and laboratory data were collected from the medical chart records.
We conducted a comprehensive study involving a cohort of 78 patients with idiopathic chronic pancreatitis as well as a geographically and ethnically matched 200 control subjects. All 27 exons, including flanking intronic regions, were analyzed by DHPLC analysis. The distribution of gender between patients with ICP and controls were not significant.
CFTR mutations
We identified a total number of 22 abnormal CFTR alleles (including T5allele), which amounts to thrice the allele frequency in healthy controls (19 abnormal CFTR alleles in 200 controls; 14.1% vs. 4.8 %, p<0.0001). Among 78 ICP patients, there were 19 patients (19/78, 24.4%) with single abnormal allele and three patients (3/78, 3.8%) with two abnormal alleles. All of our patients were heterozygous for a single CFTR mild/uncommon mutation. All of the identified mutations were mild mutation of CFTR, which might produce a small amount of functional CFTR and was regarded as to be associated with pancreatic phenotype which was concordant with previous report. A total of 20 IVS8-5T (1 homozygote in control group and the others all heterozygotes) and 9 other mutations were identified. The IVS8-5T mutation accounts for 27.2% (6 out of 22) in ICP and 73.7% (14 out of 19) in controls of all identified CFTR mutant alleles. In ICP patients carrying 5T, the mutant allele was all associated with 12 or 13 TG repeats. If we did not include the 5 allele, the mutation rate of CFTR were 10.3 %( 16/156) in ICP and 1.3% 9 5/400) in controls (p<0.0001). The other mutations were identified in heterozygous patients who were all homozygous for 7T in intron 8. These mutations include I556V, G to A 3849+45, N287Y, I125T, E217G, S895N, G1O69R, and Q1352H which have been found in patients with chronic pancreatitis or CBAVD (http://www.genet.sickkids.on.ca).
Genotype and phenotype correlations
The age of onset was younger in patients with ICP who carrying T5 allele (27.5+12.9, p= 0.009). Furthermore, these patients carrying T5 allele were all with TG12 repeat in intron 8. A trend of earlier onset was noted with the increase of number of CFTR mutations but did not reach statistical significance. In subgroup analyses for patients with idiopathic chronic pancreatitis, the age, gender, diabetes mellitus, existence of calcification or pancreatic stones and pseudocyst formation were not significantly associated with the existence or numbers of CFTR mutations. The presence of DM, calcification and pseudocyst were also not associated with the existence of CFTR mutations.
Polymorphisms
Poly T
The 7T was the most common (94.9% in ICP and 96% in controls) and hence the 7T/7T was a dominant genotype in Chinese, similar to Japanese(Fujiki et al., 2004) and Vietnamese(Nam et al., 2005). The T5 (3.6%, 20/556) and T9 alleles (0.7%, 4/556) were much less frequent than the T7 allele. The T6 allele was found with a frequency about 0.1% (1/556). One alleles of 6T was found in normal subjects.
TG repeats

The (TG) 11 and (TG) 12 were dominant TG repeats in Chinese and the ratio was roughly 1:1. Their frequency in chronic pancreatitis was similar to that in normal subjects. The frequency distributions of genotypes of (TG) 11/11, (TG)11/12, and (TG)12/12 were not significantly (p = 0.181) different among normal subjects and patients with idiopathic pancreatitis.

The M470V polymorphism

Our results did not show a significant difference (P=0.316) in genotype and allele frequency distributions of M470V in Taiwanese ICP patients and matched controls (Table ). The frequency distribution of M/M, M/V, and V/V genotypes in chronic pancreatitis was not statistically significant.
5T and its adjacent polymorphic TG repeatsSince the disease penetrance of 5T is affected by its adjacent polymorphic TG repeats, the number of TG dinucleotide repeats in intron 8 of each patient was determined by direct sequencing whatever the DHPLC displayed abnormal profiles. The distribution of 5T/6T/7T/9T, 10/11/12/13TG and M470V alleles were shown in Table. The 5T alleles were not significantly increased in ICP patients (6/156 vs. 14/400; P=0.581). The distribution of TG repeat in each poly T was shown in Table 4. It was found that in all patients carrying 5T, the mutant allele was mostly associated with 12 TG repeats. Allelic variations of 5T–12TG were similarly frequent in the patient group.
Haplotype analysis and association with ICP
Haplotypes were assembled using the genotype data obtained from the 278 tested samples and the haplotype program based on the permutation test as our previous studies. Six loci consisting of five diallelic variants and the Tn(TG)m repeats in intron 8 were analyzed. With 6 loci, there should be 26=64 haplotypes. However, because there were linkage disequilibriums (LDs) in this small region, some haplotypes did not exist or had very low frequencies, and we only listed haplotypes with frequency >0.001. The omnibus haplotype profile test was significant ( 2=26.28184, P=0.008), which indicated the overall haplotype frequency profile difference between cases and controls was significant, and thus there might be some disease-predisposing haplotypes in patients with idiopathic chronic pancreatitis.
In the individual haplotype analyses, we identified haplotypes with significantly higher haplotype frequency in the cases than in the controls at the significance level P<0.05 by permutation tests (the results would not be significant if the stringent Bonferroni correction for probability value were used [ P<0.05/64 for 64 individual haplotype analyses were performed]). All of these significant haplotypes had high haplotype frequencies (>0.01). The 125C /1001+11C/ T7TG11/470V/2694T /4521G haplotype was a dominant haplotype in the Chinese. The 125G /1001+11C/ T7TG12/470M/2694T /4521G haplotype was associated with risk of chronic pancreatitis with a OR 11.3(95% CI: 2.3-54.6) that indicated a large association effect (p=0.008).
The present study is the first comprehensive report on in Chinese with phenotype genotype correlations. The mutations in the CFTR gene ( including 5T allele) account for 14.1% (22/156) of the total alleles and 24.4% (19/78) of patients in our ICP patients, compared to 4.8% (19/400) of the alleles and 9.5%.(19/200) in controls. The CFTR mutation spectrum in the Chinese population was found to be quite different from those observed in the Caucasian populations.
This is the first CFTR study including patients with idiopathic chronic pancreatitis and controls in Chinese by comprehensive analysis. We found that there were marked differences in the mutation spectrum between different ethnic populations in both ICP and controls. Our study has demonstrated that carriers of mild mutations of CFTR and/or a distinct haplotype carry an increased risk of developing idiopathic chronic pancreatitis in Chinese. T5 allele is associated with early onset of idiopathic chronic pancreatitis in our population. To our knowledge, the inheritance pattern of ICP is complex and partially understood. This indicates that the contribution of other factors (such as environmental factors) and interaction with other genetic determinants might affect the disease severity. It remains to be studied and elucidated what the cellular mechanism of mild mutation / functional polymorphisms of CFTR is and how it interact with to other genes to induce chronic pancreatitis. Because of a specific profile of CFTR mutation/haploptype, a population -specific panel should be recommended for targeted populations including idiopathic chronic pancreatitis in Chinese.
7.3.2TNF-alpha promoter polymorphism/haplotype in chronic pancreatitis
Genetic risk factors are attributed to an important role in the pathogenesis of chronic pancreatitis. The genetic basis of chronic pancreatitis is complex. Chronic pancreatitis is a progressive chronic inflammatory disease characterized by irreversible destruction of exocrine pancreatic tissue and extensive fibrosis. Tumor necrosis factor (TNF) , a prototype proinflammatory cytokine, has been implicated as an important pathogenic mediator in a variety of inflammatory diseases. Several biallelic polymorphisms have been described within the TNF-α promoter region upstream of the transcriptional start site. In the past, limited and conflicting data on the associations between TNF- promoter polymorphisms and the pathogenesis of chronic pancreatitis have been reported in western countries
In this study, cases with chronic pancreatitis and controls were recruited consecutively from the National Taiwan University Hospital from July 2000 through June 2003. They were the so-called "Taiwanese" or "Taiwan Chinese." Most of their ancestors moved to Taiwan from southeastern China about 500 years ago. They were not Taiwanese aborigines. Chronic pancreatitis was defined histopathologically or by occurrence of pancreatic parenchymal calcifications demonstrated in imaging studies. All of the patients were negative in trypsinogen gene (PRSS1) and serine protease inhibitor Kazal type 1 (SPINK1) mutations. Patients who had pancreatic adenocarcinoma or any malignancies were also excluded. The etiology of chronic pancreatitis was classified with TIGAR-O system.We genotyped seventy cases (48men and 22women) and 200 control subjects (151 men and 135 women) for five TNF- promoter polymorphisms (-1031, -863, -857, -308 and -238) using direct sequencing. The study was approved by the local institution committee, and the subjects gave their informed consent. The age and gender were not statistically different. All of the study subjects were followed up for at least 3 years and no any malignancy were diagnosed during this period.
The allele frequencies of TNF- promoter -1031C, -863A, -857T, -308A, -238A were as 19%, 19.25%, 26%, 8.25%, and 1.75% respectively and they were consistent with previous reports in the Chinese population. The -863A allele of the TNF- promoter conferred increased risk for chronic pancreatitis (OR=4.949 ; 95% CI 2.678-9.035). In multivariate analysis, the -863A and -1031C were independently associated higher susceptibility of chronic pancreatitis (P<0.0001).
We also determine the haplotypes for chronic pancreatitis risk by the EM-based haplotype frequency estimations and permutation-based hypothesis testing procedures based on previous work in our institution(Tsai et al., 2004). Table1 displays the results of 5-locus estimated haplotype frequency analyses for the TNF- promoter. The omnibus haplotype profile test was highly significant ( 2=58.28461, P=0.001). The TACAG, CACAG, and TACGG haplotypes were associated with a ORs (37.27, 10.97, and 8.50) that indicated a large association effect (p<0.05).
Here we report for the first time associations between TNF- promoter polymorphisms and TNF- promoter haplotype in non-hereditary chronic pancreatitis. Our findings provide the possibility that TNF- promoters are candidate genes for non-hereditary chronic pancreatitis in Chinese. In Taiwanese Chinese, the -863 and -1031 allele of the TNF- promoter were also reported to determine the severity of benign ulceration after Helicobacter pylori infection. Moreover, the -1031/-863/-857 three locus haplotype was associated with higher risk of Alzheimer’s disease in Chinese in Hong Kong. These findings support that the TNF- promoter polymorphism/haplotype is truly associated with some disease entities and phenotypes in our population.
7.4 Differentiating chronic pancreatitis and pancreatic cancer
Chronic pancreatitis is one of the reported risk factor for pancreatic cancer.Pancreatic cancer (ductal adenocarcinoma) is one of the most lethal malignant tumors with poor survival. Most of the patients with pancreatic cancers are diagnosed with advanced diseases. Since chemotherapy and radiotherapy still could not significantly improve the survival of patients with pancreatic cancer, early diagnosis of pancreatic cancer is the key to improve the overall survival in this dismal disease. Even with the tremendous improvement of diagnostic imaging modalities, it is still difficult to distinguish malignant pancreatic tumor from chronic pancreatitis, especially in cases involving inflammatory enlargement of the pancreatic head. Therefore, more and more molecular markers are disclosed to try to distinguish them. The serum marker, sialylated Lewis blood group antigen CA19-9, is widely used in monitoring responses to therapy in pancreatic cancer. However, the role of CA 19-9 as a diagnostic marker of pancreatic cancer is still undetermined(Mann et al., 2000; Piantino et al., 1986). Besides, elevated CA19-9 could also be noted in chronic pancreatitis, cholestasis .On the other hand, false negative results of CA19-9 could be noted in patients do not express Lewis antigens even though in whom with heavy tumor burden of pancreatic cancer(DiMagno et al., 1999; Lamerz, 1999). CA19-9 is also unable to differentiate patients with pancreatic cancer from those with chronic pancreatitis accurately and it is reported that up to 40% of patients with chronic pancreatitis will have increased level of CA19-9. Other tumor markers have also been evaluated, including carcinoembryonic antigen (CEA), CA 242, CA 50, and CA 72-4 in the diagnosis of pancreatic cancer. However, the sensitivity and specificity of these markers appeared to be insufficient for the differentiation of pancreatic carcinoma and chronic pancreatitis.
Up to now, there is no good marker available for screening of high risk population to detect early pancreatic cancer or to differentiate chronic pancreatitis from pancreatic cancer. It is mandatory to search markers to facilitate the diagnosis of pancreatic cancer, especially in risk populations of pancreatic cancer.
Adipose tissue is a now considered to be a genuine endocrine organ producing various adipocytokines, including adiponectin. Plasma adiponectin levels have been shown to decrease in patients with cardiovascular diseases, hypertension, or obesity, type II DM and metabolic syndrome. Interestingly, previous studies have shown serum adiponectin levels are inversely related to BMI and decreased adiponectin level will increase the risk in breast , endometrial cancer and gastric cancer. It is hypothesized that low serum adiponectin levels may underline the association between breast cancer and obesity/insulin resistance. Furthermore, the level of adiponectin was reported to be associated with the tumor aggressiveness in prostate cancer. In recent report, adiponectin was found to be related to cancer cachaxia in breast and colon cancer patients. To our knowledge whether lower adiponectin is associated with more aggressive pancreatic cancer is unknown. In the present study, we examined the association of the serum adiponectin levels in patients with pancreatic cancer and chronic pancreatitis.
A total of 72 patients with pancreatic cancer who were histological proved ductal adenocarcinoma were consecutively collected at the Department of Internal medicine and Surgery, National Taiwan University Hospital, Taiwan between Jan. 1999 and Dec. 2001 were enrolled. None of the pancreatic carcinomas had been treated by chemotherapy or radiation therapy prior to resection. The study was approved by the ethics committee, and informed consent was obtained from each patients. The clinicopathological features of the patients were recorded. Of the 72 patients with pancreatic cancer, 3 were classified as stage I, 8 as stage II, 27 as stage III, and 34 as stage IV, according to the Union International Cancer classification. In addition, we examined several discrete histological parameters, including lymphatic invasion, venous invasion, lymph node metastasis, and tumor markers including CEA and CA19-9 antigens. A fasting morning blood sample was obtained for adiponectin assay after admission. Serum adiponectin concentration was also examined in 39 patients with chronic pancreatitis collected during the same period.
The diagnosis of chronic pancreatitis was confirmed by the presence of at least one of the following criteria: significant changes in ERCP, calcification of the pancreas, or histopathologic confirmation (surgical or biopsy samples). None of the patients with chronic pancreatitis had pancreatic cancer in the follow up period (at least 3 years). A total of 290 control subjects participating in one cardiovascular examination for a health promotion program were included and were confirmed as free from pancreatic cancer and chronic pancreatitis from July to December 2002. These control subjects also showed no abnormality in routine blood tests, urine and stool occult blood tests, chest X-ray examination, and abdominal ultrasonography, and thus were considered to be free of malignancy and chronic pancreatitis. For each patient, written informed consent was obtained.
Anthropometrics and Biomedical Measurements
All blood samples were obtained at fasting early in the morning, and the sera were immediately separated by centrifugation and stored at –80°C until use. Serum adiponectin levels were determined using a commercial ELISA assay with recombinant human Adiponectin as standard (Human Adiponectin ELISA Kit, R&D Systems Inc., Minneapolis MN55413 USA) according to the manufacturer’s instructions.

Serum adiponectin level increased in patients with pancreatic cancer and chronic pancreatitis

The BMI was lower in both the pancreatic cancer and chronic pancreatitis groups than in the control subjects (P<0.0001). The BMI in our study patients with pancreatic cancer and pancreatic cancer were similar.Multivariate analysis was performed to evaluate the role of age, gender, fasting glucose, transaminases, bilirubin, total cholesterol, triglyceride and BMI to predict adiponectin levels in pancreatic cancer patients. Results of the model indicated that only bilirubin and BMI were significant independent predictors of adiponectin levels.

Diagnostic role of adiponectin in pancreatic cancer and chronic pancreatitis

The ROC analysis of ADP and CA19-9 demonstrated the area under the curve of ADP and CA19-9 were performed. The sensitivity, specificity positive predictive value, and negative predictive value of each marker determined at difference cutoff levels were calculated. Increasing the CA 19-9 cut-off value from 37 to 240 lead to increase of the specificity and decrease its sensitivity from 86.5% to 4.1% as the specificity increased from 71.8% to 89.7%. As to the adiponectin, if we use the mean of ADP in chronic pancreatitis (13.7 ng/ml) as the reference cut-off, the sensitivity and the specificity are 70.3% and 74.4% respectively. When the reference limit set at 21.14 ng/ml (mean of ADP in pancreatic cancer patients) was used for ADP, the specificity and sensitivity for discrimination between pancreatic carcinoma and chronic pancreatitis were 29.7% and 92.3%, respectively. Raising the discrimination level for ADP to 28 U/ L (the mean of ADP in chronic pancreatitis plus 2 standard deviations) significantly increased the specificity of ADP to 97.4%.Using the higher cut-off value for ADP allowed ruling out the false positive results for chronic pancreatitis facilitated discrimination of pancreatic adenocarcinoma and chronic pancreatitis.
Adiponectin and clinicopathologic characteristics, staging and prognosis in pancreatic cancers
In our study, there was no correlation between adiponectin level and tumor TNM stage and survival in the pancreatic cancer group. Besides, the CA 19-9 levels did not correlate with TNM stage nor survival, either.

It is the first study to disclose the adiponectin level in pancreatic cancer and chronic pancreatitis. Previous reports in endometrial cancer, and breast cancer both showed an inverse association of cancer risk and the levels of adiponectin. The proposed mechanisms mostly rely on the relationship of obese, insulin resistance or sex hormone status in the cancer risk. The adiponectin is reported to have anti-angiogenic properties both in vivo and in vitro. In our study, the elevated adiponectin is noted in patients with pancreatic cancer after adjust possible factors to influence the adiponectin level including age, gender, BMI , abnormal liver function and jaundice. It is different from the role adiponectin playing in gastric cancer (Goggins, 2005) , prostate cancer and breast cancer. The effects of adiponectin on these cancer cells are currently under investigation.
In our study, the mean of adiponectin level of control subjects, chronic pancreatitis and pancreatic cancer patients were different and an increasing trend among these three groups were noted (p<0.0001).
CA19-9 is the most used traditional serum tumor marker of pancreatic cancer. The sensitivity and specificity to diagnose PC depend on the cut off value we set(Kim et al., 1999). However, the CA19-9 may also increased in patients with CP. In our study, both CA19-9 and adiponectin were elevated in both pancreatic cancer and chronic pancreatitis patients groups than control subjects. Adiponectin outperformed CA19-9 in distinguishing patients with pancreatic cancer from those without pancreatic cancer. Similarly, in the comparison between patients with chronic pancreatitis and those with pancreatic cancer, the adiponectin is still better than CA19-9.These findings suggest that the adiponectin have the potential role as a marker to differentiate the chronic pancreatitis from pancreatic cancer, especially those with elevated CA19-9.
The diagnostic value of ADP for patients with pancreatic carcinoma has never been studied before. In the current study, the results of ADP and CA 19-9 concentrations were measured in two selected patient groups with confirmed pancreatic carcinoma and chronic pancreatitis. Different cut-off levels of ADP and CA 19-9 were considered, and, using ROC curve analysis, the sensitivities for both markers were calculated at predetermined specificity values. Using 28 ng/ml as the reference limit value for ADP and 37 U/mL for CA 19-9 demonstrated an elevation in specificity of the markers reached to 97.4% and 71.8% of pancreatic carcinoma patients, respectively. Based on these results, the upper reference limit for CA 19-9 was kept at 37 U/mL, because increasing its cut-off value did not improve the differential diagnosis between the chronic pancreatitis and pancreatic cancer. On the other hand, using higher ADP levels facilitating discrimination of pancreatic adenocarcinoma and chronic pancreatitis.
The result suggested that ADP has a potential advantage over CA 19-9 because of its significantly higher specificity in pancreatic disease patients, especially in those with elevated CA19-9. The specificity of the high cut-off point for ADP (28ng/ml) seems to offer a better opportunity to distinguish between pancreatic cancer and chronic pancreatitis.