Inappropriate Multi-Target Stool DNA Use for Colorectal Cancer Screening: Risks, Compliance, and Outcomes

Background Inappropriate or "off-label" use of multi-target stool DNA (mt-sDNA) tests refers to their use in patients for whom colonoscopy or no testing at all is warranted. Examples include a positive family history of colorectal cancer, a history of inflammatory bowel disease, or medical issues necessitating diagnostic colonoscopy, among others. Current understanding of off-label mt-sDNA use for colorectal cancer screening, its associated risks, and outcomes is lacking. We examined off-label mt-sDNA prescription and compliance with testing in an outpatient setting in southeast Michigan. Aims The primary aims of the study were determining the extent of off-label mt-sDNA testing and compliance, and results of all testing, as well as demographic factors associated with off-label prescriptions. The secondary aims were to examine explanations for incomplete testing and factors contributing to successful completion. Methods Using a retrospective design, we identified mt-sDNA orders from outpatient internal medicine clinics between January 1, 2018, to July 31, 2019, to evaluate the proportion of off-label mt-sDNA, results of testing, and follow-up colonoscopies up to one year after order placement. Patients were categorized as "off-label" if any inappropriate criteria were met. Statistical analysis was performed for primary and secondary outcomes. Results From 679 mt-sDNA orders within the study period, 81 (12.1%) had at least one off-label criterion for testing. In total, 404/679 (59.5%) patients completed testing. Lack of follow-up comprised the majority of incompletions (216/275; 78.6%). Only 52 (70.3%) out of 74 positive results were followed by diagnostic colonoscopy. Retired employment status (OR = 1.87; 95%CI, 1.17-2.98; P = 0.008) and age of 76 years or older (OR = 2.28; 95%CI, 0.99-5.21; P = 0.044) were significantly associated with increased risk of off-label mt-sDNA prescription. Increasing age range was associated with higher test completion (χ2 (5) = 12.085, p = 0.034). Multinomial logistic regression revealed an increasing age range (OR = 1.29; 95% CI, 1.09-1.54; P = 0.004), predictive of a positive mt-sDNA result for both groups. There was no significant difference between off-label or on-label groups in the mean number of resected polyps or pathology scores on follow-up colonoscopy. Conclusions Off-label mt-sDNA use remains a concern in the outpatient setting. Compliance for test completion and follow-up colonoscopy for positive results require further improvement. Our findings shed new light on the factors associated with off-label testing while reiterating its burden. We also describe common reasons for incomplete tests in an attempt to augment future colorectal cancer (CRC) screening initiatives.