A retrospective study published in the American Journal of Roentgenology determined that CT is a useful tool for staging colon cancers that have moved beyond the bowel wall. While performance suffers in certain situations, the largely Dutch authors recommend using thin-slice CT to detect cancers.

Colon and rectal cancers are so similar that they are usually classified together. While they share a common genome and risk factors, they differ in typical diagnosis and treatment methods. Therapy for rectal cancer is based on staging with MRI or ultrasound, and neoadjuvant therapy is recommended for high-risk tumors.

Colon cancer is staged with CT, but there are no official guidelines on neoadjuvant therapy. The FOxTROT trial at the University of Birmingham in the U.K. is investigating the benefits of neoadjuvant therapy in colon cancer, leading Dutch researchers to conduct a review to determine just how accurate CT staging really is.

The team reviewed 13 studies, pooling sensitivity and specificity data and separating the tumors by stage (T1-T4).

CT performed well when the tumor was less than 5 mm beyond the bowel wall, but tumors more than 5 mm past were challenging to detect, decreasing sensitivity and specificity from 90 and 69 percent to 77 and 71 percent, respectively. The low specificity overall was interpreted as radiologists’ tendencies to over-stage tumors.

“Radiologists, to minimize the risk for understaging, interpret minimal pericolonic fat stranding due to benign desmoplastic reaction as tumor invasion. This is a well-known problem in colorectal staging,” the authors wrote.

In addition, using an image slice of less than five millimeters improved sensitivity, but not specificity. The jury was again out on the performance of tumors far from the bowel wall, but thin imaging slices improved detection of fat infiltration. However, the unchanged specificity confirms CT’s issues with fat stranding.

Around 45 percent of colon cancer cases have lymph node involvement, and the spread of cancer to the lymph nodes is used as a staging measure. The authors were careful to note that while thinner slices increased sensitivity to lymph nodes, specificity remained the same, supporting the notion that lymph node diameter is an unreliable indicator for nodal metastasis.

False positives can be diagnosed when lymph nodes are inflamed and enlarged, and false negatives can occur when the metastases are too small to view on CT.  

However, perhaps the most interesting revelation from the review is the potential usefulness of CT colonography. Usually used as a diagnostic tool, it has additional value as a staging modality according to the authors.

“Although no proper subgroup analysis could be performed, CT colonography seems to improve the accuracy of CT in the local staging of colon tumors, which could have a major clinical effect, especially if neoadjuvant treatment is adopted in the treatment guidelines for colon cancer.”