Intra-cystic glucose concentration may be used to accurately discriminate between mucinous and non-mucinous pancreatic cysts, according to study data published in Digestive and Liver Disease.
Current guidelines advise that all patients with pancreatic cysts should undergo radiological examination to identify malignancy. However, the ability of these imaging systems to differentiate benign from malignant cysts is limited. Instead, analysis of tumor markers in the intra-cystic fluid is often used to differentiate mucinous from non-mucinous cysts.
Investigators conducted a prospective, observational study of patients with pancreatic cysts to assess the discriminatory capacity of intra-cystic glucose concentration. Patients who underwent endoscopic ultrasound and fine needle aspiration of a pancreatic cyst at a hospital in Turin, Italy, between 2015 and 2019 were enrolled. Fluid taken from the cysts during fine needle aspiration was analyzed for Carcinoembryonic Antigen (CEA) and glucose concentrations. The final diagnosis of cyst subtype was made based on surgical pathology (if resected), cytological examination of intra-cystic fluid, or needle-based confocal laser endomicroscopy. The diagnostic accuracy of intra-cystic glucose was compared to that of CEA, the most frequently used biomarker of mucinous cysts.
Data from 56 patients were used in analyses. The majority of participants were women (64.3%), with a mean age of 62.6±14.4 years. Twenty-five cysts were classified as non-mucinous, while 31 were mucinous. Median intracystic CEA concentration was significantly higher in mucinous vs non-mucinous cysts (247 ng/mL vs 1.6 ng/mL; P <.0001). A cut-off point of 192 ng/mL for the diagnosis of mucinous cysts had a sensitivity, specificity, and accuracy of 54.8%, 100%, and 75%, respectively.
For the diagnosis of non-mucinous cysts, CEA level <5 ng/mL had a sensitivity, specificity, and accuracy of 72%, 87.1%, and 80.4%, respectively. Mean intracystic glucose concentration was significantly lower in mucinous vs non-mucinous cysts (11 mg/dL±19.6 vs 91.1±22 mg/dL; P <.0001). An intracystic glucose concentration <50 mg/dL showed a sensitivity, specificity, and accuracy of 93.6%, 96%, and 94.6%, respectively, for the identification of mucinous cysts. For the diagnosis of non-mucinous cysts, a glucose concentration ≥ 50 mg/mL had a sensitivity, specificity, and accuracy of 93.6%, 96%, and 94.6%.
Compared with CEA >192 ng/mL, a glucose concentration <50 mg was significantly more sensitive for mucinous cysts (93.6% vs 54.8%; P =.003), though not more specific (96% vs 100%; P =1). CEA <5 ng/mL was also less sensitive for non-mucinous cysts than glucose concentration ≥50 mg/mL, though the difference was not significant (96% vs 72%; P =.07).
Study limitations include the inability of intracystic glucose to detect dysplasia or carcinoma, or to discriminate between IPMN, MCA and distinguish SCA from PC, solid pseudopapillary neoplasm, and cystic neuroendocrine tumors. An additional limitation is the way in which the “definitive diagnosis” of pancreatic cysts was obtained. Finally, in 32 patients it was not possible to achieve a diagnosis of certainty based on either surgical specimen or cytology. The small cohort size may also limit generalizability.
In a cohort of patients with pancreatic cysts, intracystic glucose concentration outperformed intracystic CEA as a diagnostic marker of mucinous cysts. Additional study is necessary to further explore the diagnostic capacity of intra-cystic glucose. “Given its simplicity and cost-effectiveness, we confirm that intracystic glucose determination should be introduced into clinical practice as an additional test in association with those already used,” investigators wrote.
Ribaldone DG, Bruno M, Gaia S, et al. Differential diagnosis of pancreatic cysts: a prospective study on the role of intra-cystic glucose concentration [published online July 13, 2020]. Dig Liver Dis. doi: 10.1016/j.dld.2020.06.038