We sought to identify clinicopathological characteristics of high-grade pancreatic intraepithelial neoplasia (PanIN)/carcinoma in situ to facilitate screening for pancreatic ductal adenocarcinoma. We evaluated PanIN lesions in 173 consecutive autopsy cases with no evidence of pancreatic ductal adenocarcinoma and/or intraductal papillary mucinous neoplasm (mean age, 80.5 years) by submitting the entire pancreas for microscopic examination. PanIN-3 was found in 4% of examined cases, whereas PanIN-1 and PanIN-2 were present in 77% and 28%, respectively. PanIN-3 was more frequently identified in patients with diabetes mellitus and/or older age. PanIN-3 lesions were always multifocal, and the number of PanIN-3 foci was positively associated with those of PanIN-1 or PanIN-2. PanIN-3 was located more frequently in the pancreatic body and tail than in the head and predominantly involved small interlobular/intralobular ducts rather than the main duct. Notably, 71% of pancreata with PanIN-3 showed cystic changes in PanIN-3 and lower grade PanIN lesions. PanIN-3 was also accompanied by higher grade extralobular fibrosis. We found that 4% of the examined pancreata harbored PanIN-3 lesions that were associated with several unique clinicopathological features. The cystic change along with fibrotic pancreatic parenchyma may be detected by imaging studies such as endoscopic ultrasound.
1 August 2021 in Clinical Gastroenterology and Hepatology
Identifying colorectal cancer patient populations responsive to chemotherapy or chemoradiation therapy before surgery remains a challenge. Recently validated mouse protocols for organoid irradiation employ the single hit multi-target (SHMT) algorithm, which yields a single value, the D0, as a measure of inherent tissue radiosensitivity. Here, we translate these protocols to human tissue to evaluate radioresponsiveness of patient-derived organoids (PDO) generated from normal human intestines and rectal tumors of patients undergoing neoadjuvant therapy. While PDOs from adenomas with a logarithmically expanded Lgr5+ intestinal stem cell population retain the radioresistant phenotype of normal colorectal PDOs, malignant transformation yields PDOs from a large patient subpopulation displaying marked radiosensitivity due to reduced homologous recombination–mediated DNA repair. A proof-of-principle pilot clinical trial demonstrated that rectal cancer patient responses to neoadjuvant chemoradiation, including complete response, correlate closely with their PDO D0 values. Overall, upon transformation to colorectal adenocarcinoma, broad radiation sensitivity occurs in a large subset of patients that can be identified using SHMT analysis of PDO radiation responses. Significance: Analysis of inherent tissue radiosensitivity of patient-derived organoids may provide a readout predictive of neoadjuvant therapy response to radiation in rectal cancer, potentially allowing pretreatment stratification of patients likely to benefit from this approach.
Diagnostic criteria for Barrett esophagus in the United States and elsewhere, except the United Kingdom and Japan, include the presence of intestinal metaplasia with goblet cells. We analyzed high-grade dysplasia and adenocarcinoma in a complete series of sections of all interfaces between carcinoma in columnar-lined esophagus (CLE) and the adjacent mucosa, assessed them for the mucosal types present, and calculated the proportion of each mucosal type in each case. The materials were a consecutive series of 100 endoscopic resection cases, comprising a total of 1251 microscopic sections. After excluding cases in which intestinal-type or cardiac-type mucosa was not adjacent to CLE carcinoma, 70 cases remained, for which 885 sections were available. Whether the adjacent mucosa was primarily cardiac or intestinal in type was assessed, as were the patterns of immunostaining for HGM, MUC-2, and MUC-6. Twenty-four (34.3%) of the 70 CLE carcinomas were completely surrounded by cardiac-type mucosa, and 10 (14.3%) were completely surrounded by intestinal-type mucosa. In 13 (18.6%), the entire non-neoplastic mucosa of the endoscopic resection specimen comprised the cardiac type alone without any intestinal type, and only 3 (4.3%) comprised the intestinal type only, without any cardiac type. From these findings, we consider it unproven histogenetically that the background mucosa of CLE carcinoma is predominantly of the intestinal type. CLE carcinoma was not demonstrated to have a close relationship with CLE mucosa with intestinal metaplasia.