Combining radiomic features with a miRNA classifier may improve prediction of malignant pathology for pancreatic intraductal papillary mucinous neoplasms.

Jennifer B Permuth,Kujtim Latifi,Jongphil Kim,Matthew P Doepker,Domenico Coppola,Yoganand Balarunathan,Barbara A Centeno,Geoffrey Zhang,Kun Jiang,Qian Li,Lu Chen,Dung‐Tsa Chen ,Sonia T Orcutt ,Sarah E Hoffe ,Robert J Gillies ,José G Treviño ,Anthony M Magliocco ,Kenneth L Gage ,Nipun B Merchant ,Jung Woo Choi ,Mokenge P Malafa

doi:10.18632/oncotarget.11768

Abstract

Intraductal papillary mucinous neoplasms (IPMNs) are pancreatic cancer precursors incidentally discovered by cross-sectional imaging. Consensus guidelines for IPMN management rely on standard radiologic features to predict pathology, but they lack accuracy. Using a retrospective cohort of 38 surgically-resected, pathologically-confirmed IPMNs (20 benign; 18 malignant) with preoperative computed tomography (CT) images and matched plasma-based ‘miRNA genomic classifier (MGC)’ data, we determined whether quantitative ‘radiomic’ CT features (+/- the MGC) can more accurately predict IPMN pathology than standard radiologic features ‘high-risk’ or ‘worrisome’ for malignancy. Logistic regression, principal component analyses, and cross-validation were used to examine associations. Sensitivity, specificity, positive and negative predictive value (PPV, NPV) were estimated. The MGC, ‘high-risk,’ and ‘worrisome’ radiologic features had area under the receiver operating characteristic curve (AUC) values of 0.83, 0.84, and 0.54, respectively. Fourteen radiomic features differentiated malignant from benign IPMNs (p<0.05) and collectively had an AUC=0.77. Combining radiomic features with the MGC revealed an AUC=0.92 and superior sensitivity (83%), specificity (89%), PPV (88%), and NPV (85%) than other models. Evaluation of uncertainty by 10-fold cross-validation retained an AUC>0.80 (0.87 (95% CI:0.84-0.89)). This proof-of-concept study suggests a noninvasive radiogenomic approach may more accurately predict IPMN pathology than ‘worrisome’ radiologic features considered in consensus guidelines.

Highlights

To revolutionize the early detection of cancer, there is a need to replace invasive and risky tissue biopsies not representative of the entire tumor with noninvasive tests reflecting the tumor and its environment
The value of developing a noninvasive, cost-effective approach to accurately distinguish malignant from benign Intraductal papillary mucinous neoplasms (IPMNs) cannot be overstated because it would enable individuals with malignant lesions to undergo lifesaving surgery while sparing those with benign lesions the inconvenience, morbidity, and cost of major surgery
We conducted the first proof-of-concept radiogenomic study to noninvasively evaluate the clinical utility of radiomic features as predictors of malignant IPMNs alone and in combination with a blood-based miRNA genomic classifier discovered by our team [14]

Summary

Introduction

To revolutionize the early detection of cancer, there is a need to replace invasive and risky tissue biopsies not representative of the entire tumor with noninvasive tests reflecting the tumor and its environment. Consensus guidelines for IPMN management known as ‘Sendai guidelines’ exist [5] and rely on standard radiographic and clinical features These guidelines [5] suggest that patients with ‘high risk stigmata’ (MD involvement/ dilatation > 10 mm, obstructive jaundice with a cystic lesion in the pancreatic head, or an enhanced solid component/ nodule within the cyst) undergo resection, as most harbor high-grade or invasive disease [5]. Novel markers of malignant pathology are needed, especially for cases that do not appear to present with high-risk stigmata

Objectives

Methods

Results

Conclusion