Comparing American Gastroenterological Association Pancreatic Cyst Management Guidelines with Fukuoka Consensus Guidelines as Predictors of Advanced Neoplasia in Patients with Suspected Pancreatic Cystic Neoplasms

Background Society consensus guidelines are commonly used to guide management of pancreatic cystic neoplasms (PCNs). However, downsides of these guidelines include unnecessary surgery and missed malignancy. The aim of this study was to use computed tomography (CT)-guided deep learning techniques to predict malignancy of PCNs. Materials and Methods Patients with PCNs who underwent resection were retrospectively reviewed. Axial images of the mucinous cystic neoplasms were collected and based on final pathology were assigned a binary outcome of advanced neoplasia or benign. Advanced neoplasia was defined as adenocarcinoma or intraductal papillary mucinous neoplasm with high-grade dysplasia. A convolutional neural network (CNN) deep learning model was trained on 66% of images, and this trained model was used to test 33% of images. Predictions from the deep learning model were compared to Fukuoka guidelines. Results Twenty-seven patients met the inclusion criteria, with 18 used for training and 9 for model testing. The trained deep learning model correctly predicted 3 of 3 malignant lesions and 5 of 6 benign lesions. Fukuoka guidelines correctly classified 2 of 3 malignant lesions as high risk and 4 of 6 benign lesions as worrisome. Following deep learning model predictions would have avoided 1 missed malignancy and 1 unnecessary operation. Discussion In this pilot study, a deep learning model correctly classified 8 of 9 PCNs and performed better than consensus guidelines. Deep learning can be used to predict malignancy of PCNs; however, further model improvements are necessary before clinical use.

Download Full-text

Prediction of Advanced Neoplasia in Pancreatic Cystic Neoplasms: A Validated Risk Score

Gastroenterology ◽

10.1016/s0016-5085(17)32367-3 ◽

2017 ◽

Vol 152 (5) ◽

pp. S672

Author(s):

Gene K. Ma ◽

Yvonne N. Lee ◽

David Goldberg ◽

Nikhil Thiruvengadam ◽

Vinay Chandrasekhara ◽

...

Keyword(s):

Risk Score ◽

Cystic Neoplasms ◽

Advanced Neoplasia ◽

Pancreatic Cystic Neoplasms

Download Full-text

Sendai and Fukuoka Consensus Guidelines Identify Advanced Neoplasia in Patients With Suspected Mucinous Cystic Neoplasms of the Pancreas

Clinical Gastroenterology and Hepatology ◽

10.1016/j.cgh.2015.03.017 ◽

2015 ◽

Vol 13 (10) ◽

pp. 1808-1815 ◽

Cited By ~ 42

Author(s):

Pavlos Kaimakliotis ◽

Brian Riff ◽

Kamron Pourmand ◽

Vinay Chandrasekhara ◽

Emma E. Furth ◽

...

Keyword(s):

Consensus Guidelines ◽

Cystic Neoplasms ◽

Advanced Neoplasia ◽

Mucinous Cystic Neoplasms

Download Full-text

Survival in Cystic Neoplasms of the Pancreas

Canadian Journal of Gastroenterology ◽

10.1155/2009/139780 ◽

2009 ◽

Vol 23 (8) ◽

pp. 537-542 ◽

Cited By ~ 7

Author(s):

Ali Cadili ◽

Amy Bazzerelli ◽

Sipi Garg ◽

Robert Bailey

Keyword(s):

Overall Survival ◽

Pancreatic Cyst ◽

Patient Survival ◽

Survival Rates ◽

Pancreatic Cysts ◽

Cystic Neoplasms ◽

Patient Age ◽

Patient Âgé ◽

History Of ◽

Pancreatic Cystic Neoplasms

BACKGROUND: The natural history of pancreatic cystic neoplasms remains poorly understood despite growing evidence on the subject. Pancreatic cysts display a wide spectrum of pathological phenotypes, each associated with a different prognostic implication. Many pancreatic cysts are of undetermined malignant potential at presentation and remain so until surgically resected. While the survival rates of patients with malignant cysts are known to be poor, survival rates in patients with undetermined pancreatic cysts are unknown.OBJECTIVE: To identify the factors associated with survival in a group of patients diagnosed with a pancreatic cyst(s).METHODS: The present study was a retrospective multicentre review of pancreatic cystic neoplasms. All patients with a diagnosis of a neoplastic pancreatic cyst from 1994 to 2003 were identified at five different institutions in Edmonton, Alberta. The data collected included patient age, sex, imaging modality, cyst location, cyst size, number of cysts, comorbid illnesses, history of upper abdominal surgery, previous cancer, previous or concurrent metastases, symptoms (pain, upper gastrointestinal bleeding, signs of biliary obstruction, nausea/vomiting), remarkable radiological features, elevated amylase or lipase, type of pancreatic surgery, final pathology (benign or malignant) and overall survival. Survival models were used to assess whether any covariates were predictors of the survival time. Patient data were plotted using the Kaplan-Meier method. The resulting plot was used to calculate survival in the cohort.RESULTS: In total, 64 patients were identified as having neoplastic pancreatic cysts from 1994 to 2003 at the five institutions. The median overall patient survival time was 86 months. The median age at diagnosis for the patient population was 73 years, with 40 patients being women. Univariate analysis revealed that the risk of death was associated with patient age, sex and history of major comorbid illness. Multivariate models identified increased patient age and male sex as the factors that correlated most strongly with decreased overall survival.CONCLUSION: Overall survival in patients with neoplastic pancreatic cysts is determined by patient factors (ie, age and sex) rather than factors descriptive of the cyst such as size and morphology. No conclusions could be made regarding the relationship between cyst pathology and patient survival.

Download Full-text