We developed a pancreatic tumor xenograft model in a nude mice strain in order to implement Magnetic Resonance Imaging (MRI) protocols which help us to make a descriptive picture of the tumor and nearby lymph nodes, as well as the evolution of the disease. We utilized the human PANC-1 cell line for the induction of tumors, as it has been described to have lymphatic metastasis capacity (1).

IN VIVO MODEL

Male immunodeficient mice (CD-1 nu/nu, 6-7 old week) were injected s.c in the upper left flank with 2 million PANC-1 cells. We monitorized the animals by weighting them and following the tumor growth with the use of a caliper. The tumor volume was calculated using the formula (L x W²)/2 (L = Length, W= Width). Along with control mice without tumor, animals were subjected to MRI longitudinal study which included weeks 8, 14, 16 and 20 after PANC-1 injection.

MRI ACQUISITION PROTOCOL

A gradient echo MRI protocol was used on an 11.7 Tesla horizontal bore Bruker Biospec 117/16 scanner: Respiration synchronized sequence, TR=1 respiratory cycle, TE = 3.084 ms, Flip Angle = 60 degree, Field of View = 30 mm x 30 mm. Matrix = 256 x 256, Slice Thickness = 0.6 mm, no of continuous slices = 20 and NA = 2. The spatial resolution of the proton density weighted scan is 117 x 117 x 600 μm³.

IMAGING ANALYTICS

The volumes of the tumors and lymph nodes were obtained manually. Paravision 5.1 was used to draw the regions of interest in each of the slices that covered the individual lymph nodes or tumor. The sum of the individual areas and subsequent multiplication by the slice thickness gave rise to the total volume.

IMMUNOHISTOCHEMISTRY

After the last imaging session, animals were sacrificed and tumor and nearby lymph nodes were extracted for immunohistochemical analysis. The tissues were fixed in 10% neutral buffer formalin and subsequently embedded in paraffin for their sectioning in 5 mm slices. Samples were stained using Hematoxilin&Eosin (H&E). For the identification of human origin cells, a specific immunostaining using an antibody to humans cells (MHC-1) was performed.

The growth rate of the tumors showed an exponential curve, as measured by caliper or by MRI (Figure 2A). Caliper values were significantly higher due to non-tumor tissue included in the measurement (e.g. skinlayers, fat). MRI images showed well defined tumors (Figure 2B) and lymph node structures (Figure 3). In several animals, lymph nodes appeared to be enlarged. Some showing a large hypertense region (Figure 3C,D) which has been previously described in athymic nude mice strains (2) and it has been defined as cysts formation. The average size of all type of lymph nodes was between 15-40% higher in tumor-bearing mice than in controls (Figure 4), for all measured time-points. Size of lymph nodes did not correlate with tumor sizes, according to the coefficients of determination obtained (p > 0.05). Histology showed the lack of lymph node metastases in tumor bearing animals. It also revealed the presence of an enlarged number of histiocytes within the lymph nodes (histiocytosis) that was associated with apoptotic bodies only in tumor bearing animals (Figure 5B). Specific immunohistochemistry (MHC-1) confirmed that apoptotic bodies were of human origin (Figure 5C). The presence of intact human single cells was also detected (Figure 5D).The MRI longitudinal study showed an increased size of nearby lymph nodes in the tumor bearing mice. Furthermore, we demonstrated the presence of apoptotic and migrating cancer cells and histiocytosis within the lymph nodes. The lack of metastasis establishes parallels findings in cancer patients with sinus histiocytosis that is correlated with improved survival (3).