Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive malignancies, with a poor five-year survival rate. Therefore, optimizing diagnostic strategies is a critical priority. Endoscopic ultrasound (EUS) plays a key role in characterizing focal pancreatic masses (FPM). The development of ancillary EUS techniques, including contrast-enhanced EUS (CH-EUS), strain ratio elastography (SRE), and Detective Flow Imaging (DFI), has enhanced the assessment of vascularization and tissue stiffness, potentially improving the differentiation between benign and malignant lesions. Emerging evidence suggests that these parameters may provide valuable insights into the malignant potential of FPM. The aim of this study is to evaluate whether the combined use of CH-EUS, SRE and DFI improves diagnostic performance in predicting PDAC.

This single-center retrospective study included consecutive patients who underwent biliopancreatic EUS for FPM between October 2023 and October 2025 at the Digestive Diseases Unit, Sant’Andrea University Hospital of Rome. Examinations were performed using a Fujifilm linear-array echoendoscope and an ultrasound processor (Fujifilm Arietta 850, Fujifilm Healthcare, Tokyo, Japan). During EUS assessment, CH-EUS, SRE and DFI were performed to characterize lesions. The final diagnosis was confirmed by histopathology (EUS fine-needle biopsy) and/or radiological workup (CT, MRI, PET), with a follow-up of at least 6 months. For CH-EUS, lesions were categorized as hypoenhanced, isoenhanced, or hyperenhanced during the early arterial phase after administration of 8 mL of sulphur hexafluoride microbubbles; hypoenhancement was considered predictive of PDAC. SRE was evaluated using the system’s embedded software, comparing a region of interest (ROI) within the FPM to the intestinal wall; a strain ratio>10 was deemed predictive of PDAC. DFI was employed to assess microvascularization, classifying lesions as hypovascular or hypervascular, with hypovascularity considered a predictive feature for PDAC.

Overall, 95 patients were included (50 males, 52.6%; median age 68 ± 11.4 SD years). Lesions were predominantly located in the pancreatic head (64; 67.4%). Among all lesions, 64 (67.4%) were PDAC, 8 (8.4%) inflammatory lesions, 14 (14.7%) pancreatic neuroendocrine tumor (pNET) G1, 5 (5.3%) pNET G2, 3 (3.2%) other malignant tumors and 1 (1.1%) solid pseudopapillary tumor. CH-EUS was applied in 58 (61.1%) cases, elastography in 60 (63.2%) and DFI in 69 (72.6%). All three techniques were performed in 39 (41%) cases; CH-EUS+DFI in 47 (49.5%); SRE+DFI in 53 (55.8%); SRE+CH-EUS in 26 (27.4%). Across the individual modalities, CH-EUS showed a sensitivity of 71.0%, a specificity of 74.1%, with an accuracy of 72.4%, LR+ 2.74, LR– 0.39, PPV 75.9%, and NPV 69.0%. SRE achieved a sensitivity of 88.6% and a specificity of 64.0%, corresponding to an accuracy of 78.3%, with LR+ 2.46, LR– 0.18, PPV 77.5%, and NPV 80.0%. DFI demonstrated the strongest diagnostic performance, with a sensitivity of 90.7%, a specificity of 73.1% and an accuracy of 84.1%. The positive and negative likelihood ratios were 3.37 and 0.13, respectively, with a PPV of 84.8% and an NPV of 82.6%. Regarding combined approaches, CH-EUS+SRE+DFI provided a specificity of 90.5% and a sensitivity of 66.7%, yielding an accuracy of 79.5%, with LR+ 7.00, LR– 0.37, PPV 85.7%, and NPV 76.0%. The combination CH-EUS+DFI resulted in a sensitivity of 70.8% and a specificity of 82.6%, with an accuracy of 76.6%, LR+ 4.07, LR– 0.35, PPV 81.0% and NPV 73.1%. Similarly, SRE+DFI showed a sensitivity of 77.4% and a specificity of 81.8%, with an accuracy of 79.3%, LR+ 4.26, LR– 0.28, PPV 85.7% and NPV 72.0%. Finally, SRE+CH-EUS demonstrated a sensitivity of 82.4% and a specificity of 77.8%, an accuracy of 80.8%, LR+ 3.71, LR– 0.23, PPV 87.5%, and NPV 70.0%.

Among the single techniques, DFI achieved the highest sensitivity. Among the combined approaches, the SRE+DFI association provided good accuracy and specificity without requiring intravenous contrast. Nevertheless, the best overall performance was achieved when all three techniques were applied together, underscoring the value of a multimodal EUS strategy integrating ancillary technologies to enhance the predictive capability for PDAC.