CT perfusion and diffusion-weighted MR imaging of pancreatic adenocarcinoma: can we predict tumor grade using functional parameters?

Abstract

Background Pancreatic adenocarcinoma is a highly lethal disease even in initially resectable patients. Functional imaging procedures such as diffusion-weighted imaging (DWI) and computed tomography (CT)-perfusion might facilitate preoperative prediction of factors influencing prognosis in patients with pancreatic adenocarcinoma. Purpose To evaluate CT-perfusion and DWI quantitative parameters of pancreatic adenocarcinoma and to assess their correlation with clinicopathological features. Material and Methods Forty-four patients with histopathologically proven pancreatic adenocarcinoma underwent CT-perfusion and DWI for estimating blood volume (BV), blood perfusion (BF), mean transit time (MTT), time to peak (TTP), and apparent diffusion coefficient (ADC) values. The statistical analysis was performed using Wilcoxon matched-pairs test, t-test for independent samples, Spearman’s rank correlation coefficient (rs), and receiver operating characteristic analysis. Results The mean CT-perfusion parameters and ADCs were significantly different in pancreatic adenocarcinoma versus healthy parenchyma. BV (2.66 ± 0.98 mL/100g), BF (17.45 ± 4.06 mL/min/100g), and ADCs (0.91 ± 0.15 × 10−3mm/s2) in high-grade tumors were significantly lower in comparison to low-grade tumors (BV = 5.35 ± 1.36 mL/100g, BF = 28.51 ± 7.73 mL/min/100g, ADC = 1.07 ± 0.21 × 10−3mm/s2). For prediction of high-grade tumors, the sensitivity and specificity were 79.2% and 82.4% for BF and 87.5% and 88.2% for BV, respectively. A significant negative correlation was found between BV and tumor size (rs = −0.445, P = 0.029), MTT and tumor size (rs = −0.330, P = 0.043), BV and M-stage (rs = −0.286, P = 0.049), and ADC and M-stage (rs = −0.274, P = 0.038). Moreover, BF and BV values were significantly associated with ADCs. Conclusion CT-perfusion parameters and ADC values could improve preoperative assessment of pancreatic adenocarcinoma with possibility of tumor grade prediction.

Dynamic Perfusion Computed Tomography and Apparent Diffusion Coefficient as Potential Markers for Poorly Differentiated Pancreatic Adenocarcinoma

“…However, most of the studies, including the aforementioned ones, describe only one of the imaging modalities: either perfusion CT or MRI with DWI; most of them even with histologically confirmed PDAC before the radiological examination. Only in one recently published study by Kovać et al [14], perfusion CT and MRI with DWI for evaluation of pancreatic tumor grade were described. The main limitation of this study was the retrospective design, which could have influenced or even misinterpreted the data because all the calculations were done with already known morphological characteristics of the tumor.…”

Quantitative CT perfusion imaging in patients with pancreatic cancer: a systematic review

“…Peak enhancement showed in two studies to be significantly higher in pancreatic parenchyma than in PDAC (Lu: 30.8 vs 57 HU p < 0.016, Tan: 59 vs 101 HU p < 0.001) [31,34]. In two other studies, MTT was significantly higher in PDAC compared to healthy pancreatic parenchyma [Aslan: 11.2 vs. 3.7 s, (p < 0.001), Kovac: 7.4 vs. 4 s (p < 0.001)] [22,24].…”

“…In 15 out of 21 studies, BF was measured in both PDAC and non-tumorous pancreatic parenchyma, including a total of 519 patients. In all these studies, mean blood flow was significantly lower in tumor tissue compared to pancreatic parenchyma outside the tumor or in healthy pancreatic tissue in a control group [22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37]. Mean BF ranged from 17 to 60 mL/100 g/min for PDAC and 71-164 mL/100 g/min for pancreatic parenchyma.…”

“…In two articles, tumors were classified into two groups: low grade (well or moderate differentiated) and high grade (poorly differentiated) [24,40]. Both studies found higher BV in low-grade tumors compared to high-grade tumors (p = 0.001 and p = 0.004), whereas a significant difference in BF was only reported by one article (p = 0.041).…”

“…Three articles evaluate the use of CTP to predict histopathological grading of PDAC compared to histopathological diagnosis obtained using a tumor biopsy or resected specimen [24,39,40]. These three studies included a total of 124 patients [24,39,40].…”