“This supports the guidelines and recent research suggesting that when an MRI is performed before a pancreatectomy in search of liver secondary lesions, it may provide significantly higher sensitivity. Considering that the hepatic parenchyma is the most common location for PDAC metastases, the recent trends to include DWI in the MRI protocols further increased the detection rate of liver metastases, as this sequence has proved to be highly sensitive for small liver lesions,” noted lead meta-analysis author Ioana-Irina Rezus, MD, who is affiliated with the Center for Translational Medicine at Semmelweis University in Budapest, Hungary, and colleagues.
Where Does CT Exceed MRI in Pancreatic Cancer Imaging?
However, in comparison to MRI, the study authors pointed out that CT demonstrated 12 percent higher sensitivity (82 percent vs. 70 percent) for assessing vascular invasion in this patient population. The researchers also noted a 25 percent higher specificity for CT in contrast to MRI for the evaluation of resectability (88 percent vs. 63 percent).
“The lower specificity of MRI … is clinically relevant and could be explained by the difficulty of differentiating from PDAC mimickers, such as focal acute, groove or autoimmune pancreatitis or other types of pancreatic neoplasia,” acknowledged Rezus and colleagues.
Three Key Takeaways
• MRI is the stronger choice for detecting liver metastases. MRI's pooled sensitivity for liver metastases was significantly higher than CT's (93 percent vs. 73 percent). With the liver being the most common site of PDAC spread, preoperative MRI, ideally with DWI, may help catch small hepatic lesions that would change surgical candidacy.
• CT remains the workhorse for vascular invasion and resectability. CT outperformed MRI in sensitivity for vascular invasion (82 percent vs. 70 percent) and in specificity for resectability (88 percent vs. 63 percent). MRI's lower specificity may come from difficulty distinguishing PDAC from mimickers such as focal acute, groove, or autoimmune pancreatitis. Together, these findings support a complementary approach: CT for local staging and surgical planning, and MRI to rule out liver disease.
• EUS is a problem-solving tool, not a primary staging modality. EUS showed high sensitivity for lymph node involvement (87 percent vs. 58 percent for CT and 52 percent for PET/CT), but this result comes from only three studies. The authors recommend reserving EUS for equivocal cases or when tissue sampling is needed rather than using it for overall resectability assessment.
Is Endoscopic Ultrasound the Best Modality for Evaluating Lymph Node Involvement in Patients with PDAC?
While cautioning that only three included studies in the meta-analysis examined the use of EUS in lymph node evaluation, the researchers noted that EUS offered 87 percent sensitivity versus 58 percent for CT and 52 percent for positron emission tomography/computed tomography (PET/CT).
“The role of EUS remains a clinically critical question. Despite our good sensitivity results for lymph node assessment from only three available studies, this method should be used as a problem-solving tool or when a biopsy is needed, since it is generally considered not appropriate for overall resectability assessment,” added Rezus and colleagues.
(Editor’s note: “Can a Radiomic AI Model Facilitate Earlier CT Detection of Pancreatic Ductal Adenocarcinoma?,” “MRI and CT Data Show Improved Progression-Free Survival with Emerging Agent for Gastroenteropancreatic Neuroendocrine Tumors” and “Can a New MRI-Based Risk Stratification Model Bolster Survival Prediction with Pancreatic Ductal Adenocarcinoma?”)
