Abstract
Intraductal papillary mucinous neoplasms (IPMN) are pancreatic lesions with uncertain biologic behavior. This study sought objective, accurate prediction tools, through the use of quantitative histopathological signatures of nuclear images, for classifying lesions as chronic pancreatitis (CP), IPMN, or pancreatic carcinoma (PC). Forty‐four pancreatic resection patients were retrospectively identified for this study (12 CP; 16 IPMN; 16 PC). Regularized multinomial regression quantitatively classified each specimen as CP, IPMN, or PC in an automated, blinded fashion. Classification certainty was determined by subtracting the smallest classification probability from the largest probability (of the three groups). The certainty function varied from 1.0 (perfectly classified) to 0.0 (random). From each lesion, 180 ± 22 nuclei were imaged. Overall classification accuracy was 89.6% with six unique nuclear features. No CP cases were misclassified, 1/16 IPMN cases were misclassified, and 4/16 PC cases were misclassified. Certainty function was 0.75 ± 0.16 for correctly classified lesions and 0.47 ± 0.10 for incorrectly classified lesions (P = 0.0005). Uncertainty was identified in four of the five misclassified lesions. Quantitative histopathology provides a robust, novel method to distinguish among CP, IPMN, and PC with a quantitative measure of uncertainty. This may be useful when there is uncertainty in diagnosis.
Highlights
Intraductal papillary mucinous neoplasms (IPMN) are a heterogeneous group of pancreatic lesions with uncertain biologic behavior [1]
The role of quantitative histopathology is yet to be fully utilized, but we have demonstrated that it may be a robust and efficient means to distinguish between IPMN and pancreatic carcinoma (PC), even in the context of chronic pancreatitis (CP)
We demonstrated that quantitative histopathology is approximately 89% accurate in classifying lesions and 100% accurate in identifying CP
Summary
Intraductal papillary mucinous neoplasms (IPMN) are a heterogeneous group of pancreatic lesions with uncertain biologic behavior [1]. One out of three lesions will develop into a malignancy, and identifying which cancers have a higher probability to metastasize is of great debate and difficulty [2]. Lesions of indeterminate or high risk based on imaging are often sent for biopsy or even directly to surgical resection. The overall goal of this research is to develop an objective method to risk stratify IPMN. It can be difficult to distinguish between malignant. IPMN and nonmalignant IPMN on tissue biopsy or even resection unless there is clear invasion demonstrating pancreatic carcinoma (PC) [3, 4]. Quantitative histopathology has a well-d escribed role in classifying premalignant distinguishing aggressive malignancies from nonaggressive malignancies [6,7,8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
