Objective: The natural history of pancreas cancer recurrence has been difficult to study without a valid model. We present initial data describing the development of a mouse model of pancreatic cancer recurrence following “curative” resection. Our studies use a novel technique of implanting tumor cells within a hyaluronan based synthetic extracellular matrix (sECM) into the distal pancreas of nude mice. This coalesced group of cells represents an immature tumor, lacking established angiogenesis and tumor implantation factors, and therefore may more accurately mimic the changes necessary for malignant cells to develop into an invasive tumor. In this study, we demonstrate post-operative disease recurrence in mice following resection of the primary tumor.
Methods: Twenty athymic nude mice were injected with RFPt MiaPaCa-2 cells (10 mice) and RFPt AsPc-1 cells (10 mice) suspended in Extracel, a commercially available sECM, into the distal pancreas via a left subcostal incision. The mice then underwent resection of the primary tumor at either 2 or 5 weeks after injection. Fluorescence imaging was performed at resection to confirm that the mice had no residual disease. The mice underwent necropsy, were examined and fluorescence imaging was done 8 weeks after resection to evaluate for recurrent disease.
Results: At pancreatectomy, all mice had developed primary tumors. However, 2 mice with evidence of distant metastatic disease were not considered to be eligible for a “curative” resection and were excluded from final analysis. At necropsy, 6 of 18 mice developed local recurrence only, 2 of 18 had both local recurrence and metastases, and 2 of 18 had distant metastases only. Thus, 10 of 18 mice had evidence of pancreatic cancer 8 weeks following resection (Table 1).
Conclusions: Upon orthotopic injection of pancreatic tumor cells embedded with a sECM, we were able to develop primary tumors in nude mice, resect them for cure, and show a 56% recurrence rate 8 weeks after resection. The sECM allows implantation of pancreatic tumor cells without the need for donor animals or human tumor tissue. The use of RFPt cells allows sequential in vivo imaging and provides radiographic evidence that the mice are without residual disease at the time of resection. This model may prove useful for preclinical adjuvant therapeutic trials.
Table 1: RESULTS

RESECTION GROUP	LOCAL RECURRENCE ONLY	METASTASES ONLY	LOCAL RECURRENCE AND METASTASES	TOTAL WITH RECURRENCE
AsPc-1 2 Week	1/5	1/5	0/5	2/5
MiaPaCa-2 2 Week	2/4	0/4	1/4	3/4
AsPc-1 5 Week	1/4	1/4	0/4	2/4
MiaPaCa-2 5 Week	2/5	0/5	1/5	3/5
TOTAL	6/18	2/18	2/18	10/18