Highlight
- Endoscopic ultrasound (EUS) with fine needle aspiration (FNA) adds significant diagnostic specificity in patients with pancreatic cystic neoplasms (PCNs) initially evaluated by MRI.
- EUS identifies morphological high-risk features (MHRF) missed by MRI in a subset of patients, which are associated with a substantial upgrade rate for malignancy.
- Patients with MHRF detected on MRI but not confirmed by EUS have a high likelihood of remaining cancer-free on surveillance.
- Decision-curve analysis highlights the greatest clinical benefit of EUS combined with FNA in patients whose MRI does not reveal MHRF.
Study Background
Pancreatic cystic neoplasms (PCNs) are increasingly detected due to widespread imaging use. Differentiating benign lesions from those with malignant potential is critical given the significant morbidity associated with pancreatic surgery. Magnetic resonance imaging (MRI) is the preferred noninvasive modality for initial characterization, providing detailed morphology. However, MRI sometimes fails to detect subtle features that portend malignancy, leading to diagnostic uncertainty.
Endoscopic ultrasound (EUS), particularly when combined with fine needle aspiration (FNA) for cyst fluid analysis, offers enhanced resolution and the ability to obtain cytologic and biochemical markers that improve risk stratification. Although guidelines recommend EUS-FNA in selected patients, the incremental value of EUS and cyst fluid analysis over MRI alone has been incompletely characterized, especially concerning applying surgical thresholds.
This study addresses the unmet clinical need to precisely define the additive diagnostic utility of sequential testing with EUS-FNA following MRI in PCNs, to guide more tailored management decisions.
Study Design
This retrospective cohort study utilized an institutional PCN registry encompassing 3,702 patients. Inclusion criteria required both MRI and EUS evaluation.
Morphologic high-risk features (MHRF) assessed included the presence of mural nodules, pancreatic duct dilation, and thickened or enhanced cyst walls, evaluated independently on MRI and EUS.
Clinical endpoints comprised histopathologic confirmation of high-grade dysplasia or carcinoma at surgery or cancer development during surveillance. Diagnostic performance was benchmarked against surgical pathology when available.
Receiver operating characteristic (ROC) curve analysis quantified diagnostic accuracy. Decision-curve analysis evaluated net clinical benefit across different diagnostic strategies.
Key Findings
Of 1,674 patients meeting inclusion criteria, MRI detected MHRF in 28% (462) while EUS detected them in 24% (400), with significant discordance in 26% (436) cases.
Notably, in lesions where both MRI and EUS were morphologically negative, FNA yielded positive indicators of malignancy in 5% of cases.
Among 187 patients with MHRF identified by EUS but not MRI, 66% were subsequently confirmed to have disease upgrade to high-grade dysplasia or cancer—a strong indication that EUS can unveil high-risk features missed by MRI.
Conversely, 249 patients with MRI-detected MHRF but negative EUS largely remained cancer-free during a median surveillance of 49 months, with 92% showing no malignancy, suggesting possible false-positive MRI findings or overestimation of risk.
In the 215 patients undergoing surgery, adding EUS after an MRI without MHRF improved specificity by 14%, and combined MRI+EUS+FNA elevated specificity further to 38%. The area under the curve (AUC) for MRI+EUS+FNA was significantly superior to MRI alone (P=0.022), indicating enhanced diagnostic accuracy.
Decision-curve analysis demonstrated that the greatest clinical benefit of adding EUS occurs in patients whose MRI scans do not show MHRF, especially when combined with FNA results.
Expert Commentary
This robust comparative analysis clarifies the complementary roles of MRI and EUS in PCN evaluation. The findings underscore that reliance on MRI alone risks both under- and over-treatment due to diagnostic discordance. EUS with FNA adds significant value, particularly by unmasking high-risk features not evident on MRI, and by mitigating false-positive MRI findings that may prompt unnecessary surgery.
The study reinforces current clinical practice guidelines advocating a tailored, multimodal diagnostic approach. It also highlights the importance of integrating imaging morphology with cyst fluid analysis for precise risk stratification.
Limitations include retrospective design and potential referral bias inherent to surgical cohorts, which might affect generalizability. Nonetheless, the large sample size and rigorous decision-analytic methodology strengthen the conclusions.
Future prospective and multicenter studies are warranted to validate these observations and refine selection criteria for EUS-FNA referrals.
Conclusion
The integration of endoscopic ultrasound and fine needle aspiration substantially improves the diagnostic specificity and clinical decision-making in patients with pancreatic cystic neoplasms initially evaluated by MRI. This is especially impactful for patients without high-risk features on MRI, in whom EUS-FNA uncovered significant risk in a substantial subset, thereby informing appropriate management strategies and potentially avoiding unnecessary surgery.
Clinicians managing PCNs should consider adopting a sequential diagnostic algorithm incorporating both MRI and EUS-FNA to optimize patient outcomes.
Funding and ClinicalTrials.gov
The study was conducted as a retrospective registry cohort without external funding reported. No clinical trial registration was indicated.
References
1. Tocci NX, Deverakonda DL, Schmidt EM, et al. Defining the Incremental Value of Endoscopic Ultrasound in Assessing Pancreatic Cystic Neoplasms. Ann Surg. 2026 Jul 1; PMID: 42381052.
2. Tanaka M, Fernández-del Castillo C, Adsay V, et al. Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology. 2017;17(5):738-753.
3. Vege SS, Ziring B, Jain R, Moayyedi P. American Gastroenterological Association Institute guideline on the diagnosis and management of asymptomatic neoplastic pancreatic cysts. Gastroenterology. 2015;148(4):819-822.