Optimized FDG-PET/MRI protocol reveals metabolic predictors of long-term survival in pancreatic cancer patients

Abstract

Read online

PurposeTo optimize and assess an abbreviated dual time-point 18-Fluor-Deoxyglucose (FDG)-Positron Emission Tomography (PET)/Magnetic Resonance Imaging (MRI) protocol for predicting patient outcomes in pancreatic cancer.Methods70 patients (47 pancreatic cancer, 23 chronic pancreatitis) underwent hybrid PET/MRI with dual time-point PET/CT at 60 and 84 minutes post-injection. Metabolic indices (MI) were calculated from Standardized Uptake Value (SUV) changes (SUVmin, SUVmean and SUVmax). Multivariate analysis was performed on PET, MRI, laboratory, and histologic data. Top predictors were used for survival analysis.ResultsMI SUVmax, thresholded at 11%, was the best outcome predictor, distinguishing high-risk (2year (2y)-Overall Survival (OAS) 32%, 5y-OAS 14%, 10y-OAS 8%) and low-risk groups (2y-OAS 76%, 5y-OAS 32%, 10y-OAS 23%). Tumor size, CBD obstruction, and infiltrative disease had lower predictive value.ConclusionsMetabolic indices from abbreviated dual time-point FDG-PET/MRI can differentiate pancreatic malignancy from pancreatitis and predict outcomes, outperforming other indices. This protocol offers a valuable diagnostic tool for characterizing pancreatic lesions and predicting outcomes based on imaging criteria.

Keywords

• pancreatic neoplasms
• positron emission tomography computed tomography
• magnetic resonance imaging
• fluorodeoxyglucose F18
• prognosis
• survival analysis