Feasibility and Reliability of Pancreatic Cancer Staging Using Fiberoptic Confocal Fluorescence Microscopy in Rats

Background & Aims

Surgical management of pancreatic cancer depends on tumor resectability and staging. This study evaluated a new in vivo technique, fiberoptic confocal fluorescence microscopy (FCFM), for detection and staging of pancreatic tumors in rats.

Methods

FCFM was used with a protease-activated fluorescent marker (ProSense; VisEn Medical Inc, Woburn, MA) for in vivo imaging of solid organs (1.8-μm resolution) in a rat model of pancreatic ductal adenocarcinoma. A preliminary study described the FCFM rendering of normal and pathologic tissues. Subsequently, 2 double-blind studies compared FCFM to standard histology in (1) detection of tumors in rat models of cancer and controls and (2) detection of nodal involvement (splenic, celiac, mesenteric, and colic) 4, 5, and 6 weeks after tumor induction vs controls.

Results

Tumor cells displayed a fluorescent ductal pattern compared with non-fluorescent normal pancreas or normal follicular pattern of lymph nodes (LNs). FCFM detected all the pancreatic tumors (1.7-mm mean diameter) and identified 23 LNs that contained metastases of 99 LNs examined. Standard histologic analyses resulted in 1 false-negative result in tumor detection and 2 false negatives in LN detection, whereas FCFM produced no false-negative results. Additional serial sectioning confirmed all tumors and 16 metastatic LNs; FCFM had a negative predictive value of 100% and a positive predictive value of 69.6%.

Conclusions

Real-time “virtual biopsy” using FCFM detects tumors and LN metastases with 100% sensitivity and 92.2% specificity in rats, making it a reliable technique for detection and staging of pancreatic cancer.

Abbreviations used in this paper: ADU, arbitrary digital units, FCFM, fiberoptic confocal fluorescence microscopy, LN, lymph node