This article is mentioned in:

Introduction

Chemotherapy has contributed to improvements in the outcome of colorectal cancer (1,2), yet the control of metastatic lesions is crucial for improving outcomes. The metastatic mechanisms of colon cancer are still not known in detail; however, previous studies by us and others have identified the key molecules related to liver metastasis in colon cancer (3–5), and pathological findings of budding and venous invasion of tumors are known to be important in predicting the outcome of colorectal cancer (6,7). Clinically, 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG-PET/CT) fusion imaging is a useful tool for evaluating the stage, recurrence, outcome, and effectiveness of treatment in human cancers (8–11). Some studies have reported that PET/CT is useful in colon cancer (12–14). However, PET/CT imaging has a limitation in revealing hepatic metastasis due to normal uptake in the liver (15).

Many studies have shown that in vivo studies with subcutaneous xenografts in nude mice and severe combined immunodeficient (SCID) mice are useful for analyzing human cancers (16–19), whereas they are limited in evaluating internal organ metastases associated with human cancers. Immunodeficient mice commonly show poor metastatic lesions in vivo. We reported that distant metastatic lesions were easily reproduced by human cancer cell lines in newly developed NOD/Shi-scid/IL-2Rγnull (NOG) mouse models employing systemic injections (20–23). It was confirmed that the experimental metastasis model of human cancer using NOG mice was more sensitive and easier to achieve than that using SCID mice due to their multiple immunological dysfunctions not only in cytokine production capability, but also in the functional competence of T, B and NK cells (24,25). We previously developed a reliable new model for assaying hepatic metastasis with superimmunodeficient NOG mice (26,27).

Some in vivo studies have evaluated metastatic lesions of internal organs in mice with conventional modalities such as bio-luminescence imaging (BLI), but difficulties were found in revealing the mechanisms of metastasis (28). Therefore, a new in vivo system to detect internal organ metastatic lesions is required to simulate the clinical situation of metastatic lesions using reliable and quantitative approaches.

In this study, we examined whether 18F-FDG-PET/CT scans could semi-quantitatively reveal abnormal FDG uptakes in hepatic metastasis of human colon cancer cell line HCT116 in NOG mice, thereby overcoming the above described limitations encountered when evaluating metastatic lesions in the liver. Moreover, liver metastasis in NOG mice was studied by histopathological analysis. We showed here that the evaluation of metastatic lesions in NOG mice by PET/CT may be an extremely useful in vivo metastatic model.

Materials and methods

Cell culture

The HCT116 cell line was obtained from the American Type Culture Collection (Manassas, VA, USA) and maintained in McCoy’s 5A medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin and 100 μg/ml streptomycin. Cells were incubated in a humidified (37°C, 5% CO2) incubator and passaged on reaching 80% confluence.

In vivo transplantation of human colon cancer cell line, HCT116

NOG mice (9–12 weeks of age, male) were maintained at the specific pathogen-free facilities of the Central Institute for Experimental Animals (CIEA, Kawasaki, Japan). Experimental liver metastases using NOG mice were generated by intrasplenic injection of HCT116 cells (1.0×105 and 1.0×106/mouse, n=2) and splenectomy. All experiments involving laboratory animals were performed in accordance with the care and use guidelines of the CIEA, according to our previous reports (26,27).

Whole-body imaging of NOG mice with 18F-FDG-PET and CT scans

Positron-emitting fluorine-18 (18F) was produced by 18O(p, n) 18F nuclear reaction using a cyclotron (HM-18; Sumitomo Heavy Industry, Osaka, Japan) at Hamamatsu Photonics PET Center. 18F-FDG-PET and CT scans were performed in NOG mice 14 days after transplantation. The production of 18F-FDG was carried out according to a method described elsewhere (29). The kinetics and distribution patterns of the radiolabeled compound were determined with a small animal PET scanner (ClairvivoPET; Shimadzu Corp., Kyoto, Japan). This scanner consists of depth of interaction (DOI) detector modules with an axial field of view (FOV) of 151 mm, a transaxial FOV of 102 mm, and a transaxial spatial resolution of 1.54 mm in the center (30). Anesthetized mice were placed in a prone position on a fixation plate and then placed in the gantry hole of the PET scanner. After transmission measurement with an external 137Cs point source (22 MBq) for attenuation correction, 18F-FDG at a dose of 6 MBq (0.2 ml) was injected intravenously into each mouse via the tail vein. Data were acquired in list-mode format for 60 min; full 3D sinograms with corrected efficiency, scattering, attenuation, count losses and decay were reconstructed using an iterative 3D dynamic raw-action maximum likelihood algorithm (Drama). Summation images from 40 to 60 min after 18F-FDG injection were reconstructed, and average and maximum values of the standardized uptake (SUVmean, SUVmax) were semi-quantitatively calculated in various organs of NOG mice. After PET scanning, a CT scan was performed with ClairvivoCT (Shimadzu Corp.) in each NOG mouse.

Macroscopic and microscopic examinations

Mice were autopsied after PET/CT scan examinations to evaluate liver metastases. These metastatic lesions were also histologically confirmed. Immunohistochemistry was carried out on 4 μm tissue sections using the Bond Polymer Refine Detection system (Leica Microsystems, Tokyo, Japan) according to the manufacturer’s instructions with minor modifications. In brief, 4 μm sections of formalin-fixed, paraffin-embedded tissues were deparaffinized by Bond Dewax Solution (Leica Microsystems) and an antigen retrieval procedure was carried out using Bond ER solution (Leica Microsystems) for 30 min at 100°C. Endogenous peroxidases were quenched by incubation with hydrogen peroxide for 5 min. Sections were incubated for 30 min at ambient temperate with primary monoclonal antibodies for anti-HLA class 1-A, B, C (Hokudo, Sapporo, Japan) using the biotin-free polymeric horseradish peroxidase (HRP)-linker antibody conjugate system in a Bond-Max automatic slide stainer (Leica Microsystems). Nuclei were counterstained with hematoxylin.

Statistical analysis

Statistical comparisons of data sets were analyzed by a one-way factorial ANOVA. Data are shown as means ± standard error of mean (SEM). These analyses were performed using JMP version 8 software (SAS Institute, Inc., Cary, NC, USA). P-values <0.05 were considered to indicate statistically significant differences.

Results

18F-FDG-PET findings in control NOG mice

18F-FDG- PET/CT showed intense FDG uptakes in the brain (SUVmean 1.519±0.083, SUVmax 2.140±0.127), heart (SUVmean 1.343±0.035, SUVmax 1.868±0.016), kidney (SUVmean 2.163±0.125, SUVmax 3.148±0.150) and bladder (SUVmean 32.986±3.590, SUVmax 49.692±8.152) of control NOG mice (Table I, Fig. 1).

Figure 1 PET and CT findings in control NOG mice. FDG-PET/CT shows intense FDG uptakes in the brain (Br, SUVmax 2.140±0.127), heart (H, SUVmax 1.868±0.016), kidney (K, SUVmax 3.148±0.150) and bladder (Bd, SUVmax 49.692±8.152) of control NOG mice. (A) Coronal view, (B) axial view. Upper panels, CT; middle panels, PET; lower panels, PET/CT fusion.

Table I SUV values in the various organs of control NOG mice by 18F-FDG-PET/CT.

Table I

SUV values in the various organs of control NOG mice by 18F-FDG-PET/CT.

	SUVmean	SUVmax
Brain	1.519±0.083	2.140±0.127
Heart	1.343±0.035	1.868±0.016
Kidney	2.163±0.125	3.148±0.150
Bladder	32.986±3.590	49.692±8.152
Lung	0.577±0.064	1.384±0.104
Liver	0.450±0.033	0.635±0.017
Muscle	0.151±0.011	0.327±0.040
Bone	0.214±0.013	0.476±0.034
Intestine	0.824±0.033	2.061±0.111
Testis	0.510±0.022	1.212±0.335

[i] Standardized uptake values (SUV) were examined by ¹⁸F-FDG.

18F-FDG-PET and CT findings of tumor metastasis in NOG mice

18F-FDG-PET/CT showed higher multiple FDG uptakes (SUVmean 0.853±0.087, SUVmax 1.254±0.237) in the livers of NOG mice transplanted with 1.0×105 cells of the HCT116 cell line (Table II, Fig. 2) than those of control NOG mice (SUVmean 0.450±0.033, SUVmax 0.635±0.017). CT scans also showed marked liver swelling in NOG mice transplanted with 1.0×106 cells of HCT116 (Fig. 3). PET/CT showed diffuse higher FDG uptakes (SUVmean 1.211±0.108, SUVmax 1.701±0.158) in the livers of NOG mice with injection of 1.0×106 HCT116 cells than FDG uptakes with 1.0×105 cells. There were significant differences in FDG uptakes between the three groups (ANOVA, P=0.017 in SUVmean, P=0.044 in SUVmax). No other organs with abnormal FDG uptake were noted besides that of the livers of NOG mice transplanted with 1.0×105 and 1.0×106 cells of HCT116.

Figure 2 PET and CT findings in NOG mice transplanted with 1.0×105 of HCT116 cells. FDG-PET/CT shows significantly increased multiple FDG uptakes (red arrow, SUVmax 1.017) in the livers of NOG mice transplanted with 1.0×105 cells of the HCT116 cell line. (A) Coronal view, (B) axial view. Upper panels, CT; middle panels, PET; lower panels, PET/CT fusion. Br, brain; H, heart; K, kidney; Bd, bladder.

Figure 3 PET and CT findings in NOG mice transplanted with 1.0×106 of HCT116 cells. CT scans show marked liver swelling in NOG mice transplanted with 1.0×106 of HCT116 cells. PET/CT show diffuse higher FDG uptakes (red arrow, SUVmax 1.859) in livers. (A) Coronal view, (B) axial view. Upper panels, CT; middle panels, PET; lower panels, PET/CT fusion. Br, brain; H, heart; K, kidney; Bd, bladder.

Table II SUV values in the livers of NOG mice by 18F-FDG- PET/CT.

Table II

SUV values in the livers of NOG mice by 18F-FDG- PET/CT.

	SUVmean	SUVmax
Control	0.450±0.033a	0.635±0.017b
1.0×105 cells of HCT116	0.853±0.087a	1.254±0.237b
1.0×106 cells of HCT116	1.211±0.108a	1.701±0.158b

[i] Standardized uptake values (SUV) were examined by ¹⁸F-FDG-PET/CT. ANOVA, ^aP=0.017, ^bP=0.044; n=2.

In vivo liver metastasis

Experimental liver metastases, which consisted of solid white masses, were detected in all livers of NOG mice transplanted with 1.0×105 and 1.0×106 cells of the HCT116 cell line. Metastatic hepatomegaly transplanted with 1.0×106 cells was more severe than that with 1.0×105 cells (Fig. 4, H&E). It was also immunohistochemically confirmed that there were obviously more metastatic foci in the livers of NOG mice transplanted with 1.0×106 than in mice transplanted with 1.0×105 HTC116 cells (Fig. 4, HLA).

Figure 4 Histopathological findings of liver metastases. Experimental liver metastases in mice transplanted with both 1.0×105 and 1.0×106 of HCT116 cells showed metastatic hepatomegaly. Anti-HLA immunohistochemical examinations revealed that there were many more metastatic foci of human cancer in the livers of NOG mice transplanted with 1.0×106 than foci in mice transplanted with 1.0×105 of HCT116 cells. (A) 1.0×105 HCT116 cells, (B) 1.0×106 HCT116 cells; H&E, hematoxylin and eosin; HLA, anti-HLA immunohistochemistry; Scale bar, 5 mm.

Discussion

In this study, we clearly and quantitatively demonstrated increased multiple 18F-FDG uptakes in hepatic metastases of human colonic cancer cell line HCT116 in NOG mouse models using a small animal PET/CT system. We previously reported that 18F-FDG-PET/CT clinically is a very useful imaging modality to evaluate human cancers, including rare carcinoma cases (31–34). Many reports have shown that animal PET imaging is useful for evaluating human cancer xenografts in vivo. It is usually difficult to reveal metastases as conventional in vivo models are limited in their evaluation of metastases associated with human cancers (28,35). It is easy to produce experimental liver metastases in immunocompromised NOG mice because of their multiple immunological dysfunctions in cytokine production capability as well as functional competence of T, B and NK cells (24–27). However, quantitative evaluation of experimental liver metastases is still being developed. We previously reported the use of liver weight as an index of metastatic hepatomegaly or microvessel counts to reflect tumor cell activity (23,36). We present herein that the model system with NOG mice and a small animal PET/CT system has the feasibility of a quantitative metastasis model to simulate clinical situations. Several studies have reported the development of new anticancer therapies and novel drugs for human cancers using animal PET (37,38). This model system presented here with NOG mice and a small animal PET/CT system may be useful for developing new anticancer therapies and novel drugs for hepatic metastases of human cancers.

Although PET imaging with 18F-FDG is routinely used for tumor detection and therapy monitoring, several limitations have been reported with 18F-FDG, such as high uptake in normal tissues of the brain and inflammatory tissues (15,39). In contrast, positron-labeled amino acids and their analogs are considered to be useful because of the low rate of amino acid use in normal control tissues. Natural and unnatural artificial labeled amino acids have been reported (40,41). This hepatic metastasis model, using intrasplenic injection of small numbers of cancer cells into NOG mice, is reliable and quantitative, and more closely mimics in vivo conditions (26). New compounds, which are more sensitive and specific for the detection of human cancers, are expected to replace 18F-FDG (42,43). The hepatic metastasis model system in NOG mice combined with small animal PET/CT analysis will be useful in developing new labeling compounds for PET.

In conclusion, the hepatic metastasis model system in NOG mice combined with PET/CT has the feasibility to simulate clinical situations, and this model system is useful for analyzing mechanisms of metastasis and developing new therapeutic approaches for metastatic lesions of human cancers.

Acknowledgements

We are grateful to Hiroshi Iwata, Takeharu Kaiuchi, Norihiro Harada and Dai Fukumoto for their technical assistance and helpful discussions.

References