The Diverticulitis Study

KEY POINTS

• Colonoscopy is recommended after acute diverticulitis to rule out CRC, and yearly more than 2000 colonoscopies are performed.
• CRC incidence is low in the diverticulitis population and most colonoscopies are unnecessary. It takes approximately 100-200 colonoscopies to diagnose just 1 CRC.
• ctDNA is a minimally invasive and extraordinarily powerful biomarker of cancer.

GOALS

● To use ctDNA analysis to triage diverticulitis patients for colonoscopy. Diagnosing the same CRCs, but minimizing the number of patients needing colonoscopy

PERSPECTIVES

• Practice-changing evidence for ctDNA-guided triaging of diverticulitis patients for colonoscopy, an important and well-recognized clinical dilemma in the management of CRC.
• Evaluation of the cost-effectiveness of ctDNA-guided triaging versus standard-of-care.

PROJECT DESCRIPTION For patients who have experienced an episode of acute colonic diverticulitis, Danish and international guidelines recommend opportunistic screening with colonoscopy to rule out CRC as the underlying reason for the diverticulitis symptoms. In Denmark, there are more than 4000 hospitalizations per year due to diverticulitis. Accordingly, a substantial number of post-diverticulitis colonoscopies are performed each year to rule out underlying CRC. Colonoscopy is an invasive procedure associated with discomfort and pain in more than 50% of patients. The procedure also carries the risk of complications such as perforation and hemorrhage, particularly for patients with diseased intestines such as diverticulitis patients. CRC prevalence in suspected diverticulitis patients is low, \<2% overall, and as low as 0.5% for uncomplicated diverticulitis. Thus, an unreasonably high number of colonoscopies, ranging from 50 and up to more than 200, are needed to diagnose 1 CRC. In comparison, a >4% CRC prevalence is considered the minimal acceptable among patients selected for follow-up colonoscopy in the Danish nationwide CRC screening program, corresponding to less than 25 colonoscopies per CRC. Hence, it may be questioned whether the current guidelines on follow-up colonoscopy after diverticulitis are at all acceptable, both from a patient perspective and from a health-economy perspective. However, it has been very difficult to change. Consequently, there is an urgent need for a highly sensitive, low-risk, minimally invasive cancer detection test that can be used to triage newly diagnosed diverticulitis patients and ensure that underlying CRC is still diagnosed while minimizing the number of patients needing colonoscopy. Here, we propose to conduct a case/control study to explore the feasibility of using our recently developed TriMeth ctDNA analysis test for this purpose.

We hypothesize that ctDNA analysis can triage diverticulitis patients for colonoscopy and achieve a diagnosis of prevalent CRC but at a substantially lower colonoscopy/CRC ratio. For this purpose, we will use our tumor-agnostic ctDNA analysis "TriMeth". TriMeth was originally developed as a CRC screening tool for selecting individuals from the average-risk population for colonoscopy. Accordingly, the threshold for calling the test positive was set with the objective of high specificity. At 99% specificity, it has a sensitivity of 85%. Based on data from the original development and validation cohorts (400 CRC and 800 controls), we will tailor TriMeth for this specific purpose, i.e., setting the threshold for test positivity to achieve near 100% sensitivity by accepting a lower specificity. Second, we will conduct a case-control study to validate this new TriMeth-diverticulitis threshold in the target population. We expect a minimum of 20 cases with CRC diagnosed at colonoscopy after suspected diverticulitis to be identified through the IMPROVE study (NCT03637686). For each case, 10 controls will be randomly selected (to a minimum of 200 total) from patients with CT-verified diverticulitis and a follow-up colonoscopy without CRC. Diverticulitis patients may at the time of colonoscopy exhibit symptoms, that would in any case serve as an absolute indication for colonoscopy (patients age >40 with changes in bowel habits >1 month, rectal bleeding, iron-deficiency anemia, or excessive weight-loss). Therefore, in addition to blood sampling, we will at the time of CRC diagnosis or negative colonoscopy collect information on other established indicators for colonoscopy. As part of the study, we will conduct a health-economy assessment comparing the TriMeth-guided approach to the current practice of "colonoscopy for all". The primary outcome is the sensitivity and specificity of TriMeth for CRC detection in this population. The secondary outcome is health economy, considering the other indicators for colonoscopy. If high sensitivity and clinically and economically relevant colonoscopy-sparing potential is demonstrated, the data will be presented to the Danish Colorectal Cancer Group and the Danish Health Technology Council with the aim of obtaining approval to change the national guideline and implement the approach.