Grading of pancreatic neuroendocrine neoplasms using pharmacokinetic parameters derived from dynamic contrast-enhanced MRI

The present study aimed to evaluate the diagnostic efficacy of pharmacokinetic parameters derived from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in prospective evaluation of pancreatic neuroendocrine neoplasms (pNENs) grading. A total of 25 histologically proven patients with pNENs (30 lesions in total) who underwent DCE-MRI were enrolled. Lesions were divided into G1, G2 neuroendocrine tumor (NET) and G3 NET/neuroendocrine carcinoma (NEC) groups based on their histological findings according to 2017 World Health Organization Neuroendocrine Tumor Classification Guideline. In addition, the same numbers of tumor-free regions were selected using as normal control group. For each group, pharmacokinetic DCE parameters: volume transfer constant (K(trans)); contrast transfer rate constant (k(ep)); extravascular extracellular space volume fraction (v(e)); and plasma volume fraction (v(p)) were calculated with Extended Tofts Linear model. Receiver operator characteristics analysis was conducted to assess the diagnostic efficacy of these parameters in pNENs grading. There were significant differences of K(trans), k(ep), v(e) and v(p) between tumor-free areas and G1, G2 NET (P<0.001). The K(trans) and k(ep) of G1 NET were significantly lower compared with those of G2 ones (P<0.005). The area under the curve of K(trans) and k(ep) in differentiating G2 from G1 NET were 0.767 and 0.846, respectively. When K(trans) was >0.667 and k(ep) >1.644, the sensitivity of diagnosing G2 NET was the lowest (53.85%), but the specificity was the highest (93.75%). When K(trans) was >0.667 or k(ep) >1.644, the sensitivity of diagnosing G2 NET was 92.31%, but the specificity was 75.00%. Pharmacokinetic parameters of DCE-MRI, particularly the quantitative values of K(trans) and k(ep), are helpful for differentiating G2 NET from G1 ones.