Purpose: A sporadic colon cancer mouse model with conditional mutations in adenomatous polyposis coli (Apc) is biologically relevant for human colorectal cancer (CRC). This study aimed to determine the utility and limitations of colonoscopy for evaluating colon tumors in this mouse model.
Methods: We compared the estimates of location, size, and miss rate of tumors detected during colonoscopy with those determined by necropsy. Sixty-six CPC-Apc mice originating from Apc (F/wt) mice harbor a Cdx2-Cre transgene in which colorectal tumorigenesis was driven by Apc allelic loss. The sensitivity and specificity of colonoscopy for detecting tumors in a mouse CRC model were investigated.
Results: A strong positive correlation was found between tumor location as measured by colonoscopy and the location determined by necropsy (p < 0.001). A total of 120 tumors were graded during colonoscopy (grades 1-5: 0, 8, 20, 27, and 65 lesions, respectively), and a strong positive correlation was found between the tumor grade determined by colonoscopy and size measured by necropsy (grades 2-5: 2.08, 2.98, 4.02, and 5.09 mm, respectively; p < 0.005). Although the miss rate was 47.1 %, most of the missed tumors (96 %) were in close proximity (within 5 mm) of another tumor.
Conclusions: A colonoscopic method for the reliable measurement of colorectal tumors in vivo has been established. The application of this technique to mouse models of colon carcinogenesis will provide a better understanding of the dynamics of tumor growth.