CT colonography (CTC) nearly matched the predictive accuracy of colonoscopy for colorectal cancer screening, while also allowing for risk stratification by providing detailed insight into the nature of colorectal findings in a new study, published online April 11 in the American Journal of Roentgenology.
Researchers from the U.S. and Italy evaluated the viability of relying on CTC, as opposed to more widely used stool-based tests such as the fecal immunochemical test, for colorectal cancer screening. Clinicians have the option of offering CTC or a stool-based test as less invasive alternatives to colonoscopy and only referring patients for a colonoscopy after discovering a positive finding on one of the noninvasive screening options.
For the study, the group examined the data of 1,650 patients who underwent CTC colorectal cancer screening exams followed by standard colonoscopy between April 2004 and June 2018. The team also assigned each case a CTC Reporting and Data System (C-RADS) score based on the size and number of lesions detected on the CTC exams. The patients' average age was 59.7 years, and more than 53% were men.
Overall, the positive predictive value (PPV) for lesions at least 6 mm in size was 90.8% per patient and 88.8% per lesion, compared with the reference standard of colonoscopy.
| CT colonography vs. colonoscopy for colon cancer screening (lesions ≥ 6 mm) |
||
| PPV | Colonoscopy | CT colonography |
| Per patient | 100% | 90.8% |
| Per polyp | 100% | 88.8% |
In addition to its high accuracy, a positive CTC exam also pointed to the likelihood of new, abnormal tissue growth. To be precise, a positive CTC exam indicated neoplasia for 72.3% of the patient cases and advanced neoplasia for 38.8% of the cases.
What's more, CTC enabled radiologists to identify critical information about the kind of condition behind each positive finding -- an added benefit that stool-based tests are incapable of offering, noted lead author Dr. Perry Pickhardt, of the University of Wisconsin School of Medicine and Public Health, and colleagues. Categorizing the different types of positive CTC exams with C-RADS scoring could allow clinicians to use such stratified data to triage patients more effectively.
For example, only 5.8% of positive CTC exams classified as C-RADS C2 were associated with advanced neoplasia, but the association jumped to 67.1% for CTC scans classified as C-RADS C3. In the case of cancer diagnosis, none of the C-RADS C2 cases was tied to cancer, whereas more than half of all C-RADS C4 cases were linked to cancer.
"Unlike stool-based tests that provide only a binary positive or negative result, CTC can specify the nature of the positive findings, resulting in much greater specificity and risk stratification for patient management decisions," the authors wrote.
Furthermore, Pickhardt and colleagues found considerably greater consistency among the interpretations of all but one of the radiologists who reviewed the CTC exams (an accuracy ranging from 84.3% to 95.5%) than among the clinicians who evaluated the colonoscopies (between 63.2% and 100%).
"The uniform performance of the radiologists over the gastrointestinal endoscopists in terms of PPV is encouraging because it suggests greater generalizability for CTC over optical colonoscopy," they wrote.
