CONTACT US
GRANTEE: Amber Simpson, PhD
Memorial Sloan-Kettering Cancer Center
Research Project: CT Texture Analysis: A Radiomics Approach to Predicting Malignancy in IPMN
Award: 2016 Pancreatic Cancer Action Network – AACR Career Development Award (Grant funded by an anonymous foundation)
Award Period: July 1, 2016 – June 30, 2018
Amount: $200,000
Biographical Highlights
Dr. Simpson is an assistant attending computational biologist in the hepatopancreatobiliary service in the department of surgery at Memorial Sloan-Kettering (MSK) Cancer Center in New York. Dr. Simpson specializes in medical image processing; her research group is focused on developing novel computing strategies for precision oncology. She joined the MSK faculty in January 2015 after three years as a research assistant professor in biomedical engineering at Vanderbilt University. She received her PhD in computer science from Queen’s University in Kingston, Ontario, Canada.
Project Overview
A type of pancreatic cyst, called intraductal papillary mucinous neoplasm (IPMN), is known to sometimes progress to pancreatic adenocarcinoma, the most common and aggressive form of pancreatic cancer. However, some IPMNs do not progress to a malignant state. Currently, radiologists are depended on to interpret patients’ scans and predict the likelihood of disease progression. Misdiagnoses can lead to either missing a window of opportunity to treat the disease in its earliest stages, or too aggressively treating a cyst that wouldn’t have turned into cancer if left unchecked.
To improve the ability to differentiate between early-stage cancer and benign abnormalities, Dr. Simpson proposes to utilize a quantitative (mathematically measurable) analysis of patients’ computed tomography (CT) scans to define features predictive of the cyst’s aggressive potential. Dr. Simpson and her colleagues hypothesize that these quantitative features derived from CT images can better predict malignancy in IPMNs than visual assessment by radiologists. They will also assess fluid extractable from the cyst itself for markers, or clues, that could assist in its characterization.
At the conclusion of this study, the research team will have established a set of measurable characteristics for prediction of malignancy in IPMN. The next step would be further validation using CT scans from patients previously diagnosed with pancreatic cysts. Since CT imaging is routinely used in the preoperative setting throughout the United States, the dissemination of their methodology to other institutions can be performed quickly and inexpensively. Validation of Dr. Simpson and her team’s findings with multiple groups of patients would then lead to forward-looking multi-center imaging trials, with the long-term goal of predicting earlier stage pancreatic cancer, leading to longer survival.
