CT features predictive of nodal positivity at surgery in pancreatic cancer patients following neoadjuvant therapy in the setting of dual energy CT
Abdom Radiol (NY) . 2021 Jan 20. doi: 10.1007/s00261-020-02917-5. Online ahead of print.
Ott Le, Sanaz Javadi, Priya R Bhosale, Eugene J Koay, Matthew H Katz, Jia Sun, Wei Yang, Eric P Tamm
Purpose: Evaluate utility of dual energy CT iodine material density images to identify preoperatively nodal positivity in pancreatic cancer patients who underwent neoadjuvant therapy.
Methods: This IRB approved retrospective study evaluated 62 patients between 2012 and 2016 with proven pancreatic ductal adenocarcinoma, who underwent neoadjuvant therapy, tumor resection and both baseline and preoperative assessment with pancreatic multiphasic rapid switching dual energy CT. Three radiologists in consensus identified on imaging nodes > 0.5 cm in short axis, evaluated nodal morphology, size and on each phase density in HU, and concentrations on iodine material density images normalized to the aorta.
Results: Of 62 patients, 33 were N0, 20 N1, and 9 N2. Total of 145 lymph nodes were evaluated, with average number of nodes per anatomic site ranging from 1.3 (body tumors) to 5 (uncinate) versus average of 24 and 30 nodes recovered respectively at surgery. Most (N = 44) were pancreatic head tumors. For all patients, regardless of site of primary tumor, the minimum measured iodine value of all of a patient's measured nodes taken as a group on preoperative studies, as normalized to the aorta, was significant at P = 0.041 value in differentiating N0 from N1/2 and ROC analysis showed an AUC of 0.67. With a cutoff of 0.2857, sensitivity was 0.78 and specificity was 0.58, with values < 0.2857 indicative of N1/2. Node morphology and changes in nodal size weren't statistically significant.
Conclusion: The dual energy based minimum normalized iodine value of all nodes in the surgical field on preoperative studies has modest utility in differentiating N0 from N1/2, and generally outperformed conventional features for identifying nodal metastases.
Read Full Article Here: https://doi.org/10.1007/s00261-020-02917-5