Fluorescence-guided surgery allows for more complete resection of pancreatic cancer, resulting in longer disease-free survival compared with standard surgery in orthotopic mouse models

Abstract

Background: Negative surgical margins are vital to achieve cure and prolong survival in patients with pancreatic cancer. We inquired if fluorescence-guided surgery (FGS) could improve surgical outcomes and reduce recurrence rates in orthotopic mouse models of human pancreatic cancer.

Study design: A randomized active-control preclinical trial comparing bright light surgery (BLS) to FGS was used. Orthotopic mouse models of human pancreatic cancer were established using the BxPC-3 pancreatic cancer cell line expressing red fluorescent protein (RFP). Two weeks after orthotopic implantation, tumors were resected with BLS or FGS. Pre- and postoperative images were obtained with the OV-100 Small Animal Imaging System to assess completeness of surgical resection in real time. Postoperatively, noninvasive whole body imaging was done to assess recurrence and follow tumor progression. Six weeks postoperatively, mice were sacrificed to evaluate primary pancreatic and metastatic tumor burden at autopsy.

Results: A more complete resection of pancreatic cancer was achieved using FGS compared with BLS: 98.9% vs 77.1%, p = 0.005. The majority of mice undergoing BLS (63.2%) had evidence of gross disease with no complete resections; 20% of mice undergoing FGS had complete resection and an additional 75% had only minimal residual disease (p = 0.0001). The mean postoperative tumor burden was significantly less with FGS compared with BLS: 0.08 ± 0.06 mm(2) vs 2.64 ± 0.63 mm(2), p = 0.001. The primary tumor burden at termination was significantly less with FGS compared with BLS: 19.3 ± 5.3 mm(2) vs 6.2 ± 3.6 mm(2), p = 0.048. FGS resulted in significantly longer disease-free survival than BLS (p = 0.02, hazard ratio = 0.39, 95% CI 0.17, 0.88).

Conclusions: Surgical outcomes were improved in pancreatic cancer using fluorescence-guidance. This novel approach has significant potential to improve surgical treatment of cancer.

Figure 1

Schematic diagram of study timeline. Two weeks post orthotopic tumor implantation, the mice were randomized and tumors were resected using the MVX 10 dissecting microscope. Starting on POD 1, half of the mice from each surgical arm underwent 4 weeks of gemcitabine treatment. At week 9 (or 7 weeks postoperatively) the mice were sacrificed for intravital imaging to evaluate extent of tumor burden in the pancreas. During the entire postoperative period, all mice were imaged weekly using the OV-100.

Figure 2

(A) Pre- and postoperative tumor burden by surgical method. The mean postoperative tumor burden under BLS was 2.6 ± 0.6 mm², a significantly higher value than the mean postoperative tumor burden achieved under FGS (0.08 ± 0.06 mm²). Asterisk indicates statistical significance with p=0.001. On average, 77% of the tumor burden was resected under BLS. More than 98% tumor burden reduction was achieved with FGS (p=0.005). (B) Preoperative and postoperative images of the pancreas in two different mouse specimens under different surgical methods. The top row contains representative pre- and postoperative images of a specimen from the BLS group. A tumor reduction of only 77% was achieved in the BLS group. The bottom row images are representative pre- and postoperative images of a specimen from the FGS group. A significant improvement in tumor reduction was achieved in this group (98.9%, p=0.005). A complete surgical resection of pancreatic tumor with negative surgical margins was achieved in this mouse without requiring significant resection of the pancreas. All images were analyzed for tumor burden using ImageJ. Blue bar, preoperative tumor burden; red bar, postoperative tumor burden.

Figure 2

(A) Pre- and postoperative tumor burden by surgical method. The mean postoperative tumor burden under BLS was 2.6 ± 0.6 mm², a significantly higher value than the mean postoperative tumor burden achieved under FGS (0.08 ± 0.06 mm²). Asterisk indicates statistical significance with p=0.001. On average, 77% of the tumor burden was resected under BLS. More than 98% tumor burden reduction was achieved with FGS (p=0.005). (B) Preoperative and postoperative images of the pancreas in two different mouse specimens under different surgical methods. The top row contains representative pre- and postoperative images of a specimen from the BLS group. A tumor reduction of only 77% was achieved in the BLS group. The bottom row images are representative pre- and postoperative images of a specimen from the FGS group. A significant improvement in tumor reduction was achieved in this group (98.9%, p=0.005). A complete surgical resection of pancreatic tumor with negative surgical margins was achieved in this mouse without requiring significant resection of the pancreas. All images were analyzed for tumor burden using ImageJ. Blue bar, preoperative tumor burden; red bar, postoperative tumor burden.

Figure 3

Representative OV-100 images of resection outcomes categorized by the remaining disease analyzed by ImageJ from postoperative images. (a) F0 resection was defined as a complete absence of fluorescent tumor in the postsurgical bed. (b) F1 resection was defined as postoperative tumor burden >0 mm² but less than 1 mm². The specimen in this image had a postoperative tumor burden of 0.110 mm². (c) F2 resection was defined as postoperative tumor burden >1 mm². The mouse specimen in this image had a postoperative tumor burden of 2.778 mm².

Figure 4

Extent of tumor burden by surgical method in the pancreas at time of termination. The BLS group had significantly greater tumor burden in the pancreas at termination compared to the FGS group (p=0.048). The improved resection under FGS at the initial operation had a significant effect on recurrence, disease-free survival and overall extent of disease in the pancreas. Error bars: +/− 1 SE.

Figure 5

Kaplan-Meier disease-free survival curve. FGS lengthened disease-free survival from 1 week (95% CI, ≥1 week) to 7 weeks (95% CI, ≥3 weeks). By termination date, 70% of the mice in the BLS group had evidence of tumor recurrence based on images obtained with the OV-100 Small Animal Imaging System, while 50% had recurred in the FGS group.

Figure 6

Correlation of imaging with histology for documentation of tumor resection using either bright light or fluorescence-guided surgical techniques. Brightfield (lower) and fluorescence (upper) images obtained at the time of surgery for either (A) a bright light or (B) fluorescence-guided surgical approach. Histologic cross-sections (top image in each 6A and 6B, H&E) taken through the surgical margins of the unresected tissues. The arrows in Figure 6A show the correlation of the small focus of unresected (fluorescent) tumor cells retained in the post-operative tumor bed and the histologic section through this focus in a mouse resected for tumor under BLS.

Figure 6

Correlation of imaging with histology for documentation of tumor resection using either bright light or fluorescence-guided surgical techniques. Brightfield (lower) and fluorescence (upper) images obtained at the time of surgery for either (A) a bright light or (B) fluorescence-guided surgical approach. Histologic cross-sections (top image in each 6A and 6B, H&E) taken through the surgical margins of the unresected tissues. The arrows in Figure 6A show the correlation of the small focus of unresected (fluorescent) tumor cells retained in the post-operative tumor bed and the histologic section through this focus in a mouse resected for tumor under BLS.