Search

The joint linearity of the categorical covariates on the logit scale

Background: Barrett's esophagus is often asymptomatic and only a small portion of Barrett's esophagus patients are currently diagnosed and under surveillance. Therefore, it is important to develop risk prediction models to identify high-risk individuals with Barrett's esophagus. Familial aggregation of Barrett's esophagus and esophageal adenocarcinoma, and the increased risk of esophageal adenocarcinoma for individuals with a family history, raise the necessity of including genetic factors in the prediction model. Methods to determine risk prediction models using both risk covariates and ascertained family data are not well developed.Methods: We developed a Barrett's Esophagus Translational Research Network (BETRNet) risk prediction model from 787 singly ascertained Barrett's esophagus pedigrees and 92 multiplex Barrett's esophagus pedigrees, fitting a multivariate logistic model that incorporates family history and clinical risk factors. The eight risk factors, age, sex, education level, parental status, smoking, heartburn frequency, regurgitation frequency, and use of acid suppressant, were included in the model. The prediction accuracy was evaluated on the training dataset and an independent validation dataset of 643 multiplex Barrett's esophagus pedigrees.Results: Our results indicate family information helps to predict Barrett's esophagus risk, and predicting in families improves both prediction calibration and discrimination accuracy.Conclusions: Our model can predict Barrett's esophagus risk for anyone with family members known to have, or not have, had Barrett's esophagus. It can predict risk for unrelated individuals without knowing any relatives' information.Impact: Our prediction model will shed light on effectively identifying high-risk individuals for Barrett's esophagus screening and surveillance, consequently allowing intervention at an early stage, and reducing mortality from esophageal adenocarcinoma. Cancer Epidemiol Biomarkers Prev; 25(5); 727–35. ©2016 AACR.

Supplementary Methods of imputation of age and estimating the parameters of the prediction model. Supplementary Table S1 Missing rates on the 8 covariates in the training and validation datasets. Supplementary Table S2 Stepwise removal of covariates in the MLM model on the basis of LRT tests and AIC. Supplementary Table S3. Description of the eight predictor covariates. Supplementary Table S4 Number of informative pedigrees without missing covariate data.

Purpose: Methylation of CpG islands contributes to gene silencing during cancer development, and although some methylation alterations are promising diagnostic markers of cancer, some CpG islands are also methylated in normal tissues. We have previously observed that some normally unmethylated CpG islands that undergo methylation in pancreatic cancers are normally methylated in the adjacent duodenum. Because duodenal methylation patterns are an important consideration when sampling pancreatic tissues for pancreatic cancer methylation alterations, we determined the DNA methylation patterns of 24 genes in the normal duodenum of patients with pancreatic disease and related these patterns to demographic factors. Experimental Design: The nonneoplastic duodenal mucosa of 158 patients with pancreatic carcinoma and 41 patients with chronic pancreatitis was analyzed using methylation-specific PCR and combined bisulfite restriction analysis. Secretin-stimulated pancreatic/duodenal juice from 15 individuals undergoing endoscopic investigation for upper gastrointestinal disease was also analyzed. Results: Low-level methylation was detectable by methylation-specific PCR in the nonneoplastic duodenum of many patients with pancreatic cancer and chronic pancreatitis as well as in the pancreaticoduodenal secretions of patients without pancreaticobiliary disease. For many genes, the prevalence of methylation increased with age and was more prevalent in patients with pancreatic cancer than in age-matched patients with chronic pancreatitis. Our results indicate that strategies to detect pancreatic cancer using aberrantly methylated genes should rely on analysis of pure pancreatic juice rather than on pancreatic juice collected within the duodenal lumen. Patients who develop pancreatic cancer may have a greater propensity to methylate CpG islands than age-matched controls.

Long standing chronic pancreatitis is a risk factor for developing pancreatic cancer. Inheritance of polymorphisms in SPINK1 and CFTR are associated with an increased risk of developing pancreatitis. The aim of this study was to determine if patients who carry polymorphisms in SPINK1 and CFTR are at increased risk of developing pancreatic cancer through the development of chronic pancreatitis. DNA from patients with histologically-confirmed surgically-treated chronic pancreatitis, familial and sporadic pancreatic adenocarcinoma and controls were analyzed for the N34S polymorphism of SPINK1 and the two commonest polymorphisms of the CFTR gene, the DF508 mutation and the 5T polymorphism. These polymorphisms were determined using restriction fragment length polymorphism, PCR and cycle sequencing methods. The SPINK1 N34S polymorphism was detected in 5 of 172 (2.9%) patients with chronic pancreatitis, in 4 of 200 (2.0%) patients with sporadic pancreatic adenocarcinoma, in 0 of 36 (0%) of patients with familial pancreatic cancer and in 3 of 177 (1.7%) controls of chronic cholecystitis. The CFTR 5T polymorphism was identified in 31 of 334 (9.3%) patients of sporadic pancreatic cancer, in 5 of 43 (11.6%) patients with familial pancreatic cancer and in 10 of 112 (8.9%) controls with colorectal cancer. The CFTR DF508 mutation was recognized in 6 of the 240 (2.5%) patients with pancreatic adenocarcinoma, a prevalence similar to that of control populations. We conclude that the N34S polymorphism of SPINK1 and the 5T and DF508 CFTR polymorphisms do not predispose to the development of pancreatic adenocarcinoma. Furthermore, the N34S polymorphism is rarely found in patients with severe idiopathic chronic pancreatitis.

The aim of this study was to determine the utility of detecting methylated ppENK and pi6 in pancreatic juice by methylation specific PCR as a marker of pancreatic adeno-carcinoma. Pancreatic juice samples were collected either intraoperatively, from 92 patients undergoing pancreaticoduodenectomy for benign (n=20) and malignant periampullary disease (n = 72) or endoscopically (by duodenal aspiration after secretin infusion), from 13 patients undergoing investigation for pancreatic disease. Methylated ppENK was detected in the pancreatic juice of 30 (66.7%) of 45 patients with pancreatic ductal adenocarcinoma, in 4 (44.4%) of 9 patients with intraductal papillary-mucinous adenocarcinoma, and in 7 (41.2%) of 17 patients with other periampullary carcinomas, using methylation specific PCR. Methylated pi6 was detected in a lower percentage of these patients (11.1%, 11.1% and 23.5%, respectively). In contrast, methylated ppENK and pi6 were not detected in 20 patients with non-malignant periampullary disease including 12 patients with chronic pancreatitis. Methylated ppENK was detected in 30 of 33 (90.9%) primary pancreatic adenocarcinoma and methylated pi6 was in 6/33 (1 8.2%). Despite the absence of ppENKand pi6 methylation in normal pancreas, methylated ppENK and pi6 was present in the duodenum of 90.5% and 28.6%, respectively of patients without cancer. Further, methylated ppENK and pi6 was seen in 88.9% and 11.1%, respectively of pancreatic juice samples obtained by duodenal aspiration from patients without cancer. We conclude that since ppENK and pi6 are not normally methylated in pancreatic secretions, detection of methylated ppENK and pi6 in pure pancreatic juice obtained by direct cannulation of the pancreatic duct to avoid duodenal secretions may suggest the presence of pancreatic adenocarcinoma

New biomarkers of pancreatic adenocarcinoma are needed to improve the early detection of this deadly disease. We performed surface enhanced laser desorption ionization (SELDI) mass spectrometry using ProteinChip technology (Ciphergen Biosystems, Fremont, CA) to screen for differentially expressed proteins in pancreatic juice. Pancreatic juice samples obtained from patients undergoing pancreatectomy for pancreatic adenocarcinoma were compared with juice samples from patients with other pancreatic diseases. We identified a peak approximately 16,570 daltons present in the pancreatic juice from 10/15 (67%) of the patients with pancreatic adenocarcinoma and in the pancreatic juice from 1/7 (17%) of the patients with other pancreatic diseases. Using a ProteinChip immunoassay, we identified this differentially expressed protein as hepatocarcinoma-intestine-pancreas/pancreatitis-associated-protein I (HIP/PAP-I), a protein released from pancreatic acini during acute pancreatitis and overexpressed in hepatocellular carcinoma. We then quantified by ELISA the pancreatic juice HIP/PAP-I levels in 43 patients (28 with pancreatic adenocarcinoma, 15 with other pancreatic diseases) and the serum HIP/PAP-I levels in 98 patients (53 with pancreatic adenocarcinoma, 45 with other pancreatic diseases or healthy individuals). HIP/PAP-I levels were significantly higher in both the pancreatic juice (P0.001) and in the serum (P0.001) of patients with pancreatic adenocarcinoma compared with the control group. HIP/PAP-I levels were approximately 1000-fold higher in pancreatic juice compared with serum and the magnitude of the difference between the pancreatic adenocarcinoma group and the control group was greater in the pancreatic juice samples (143.75 +/- 235.52 microg/ml versus 6.04 +/- 7.59 microg/ml) than in the serum samples (99.96 +/- 140.66 ng/ml versus 35.25 +/- 28.44 ng/ml). In our study, patients with pancreatic juice HIP/PAP-I levelsor= 20 microg/ml were 21.9 times (95% confidence interval, 3.5-136.5; P0.001) more likely to have pancreatic adenocarcinoma than patients with levels20 microg/ml. Immunolabeling of tissue sections revealed that the HIP/PAP-I protein was strongly expressed in acini adjacent to the invasive adenocarcinoma, but it was only rarely (1/30; 3%) expressed in the neoplastic epithelium, which suggests that the main source of HIP/PAP-I release in the pancreatic juice is acini. This low level of HIP/PAP-I expression in pancreatic adenocarcinoma was confirmed by reverse transcription-PCR: only 1 (5%) of 19 pancreatic cancer cell lines expressed HIP/PAP-I transcripts. Taken together, these data suggest that pancreatic juice measurement of HIP/PAP-I may help to identify patients with pancreatic adenocarcinoma.