Amol Agarwal, MD, Talan Zhang, MS, Nishal Ravindran, MD, Paul Thuluvath, MD, Anurag Maheshwari, MD; Mercy Medical Center, Baltimore, MD
Introduction: Guidelines for colorectal cancer (CRC) screening recommend the use of multi-target stool DNA (mt-sDNA) testing in patients at average risk for CRC. However, mt-sDNA testing is often used outside of recommended guidelines (off-label use). We evaluated the off-label use of mt-sDNA testing in a primary care setting. Methods: Orders for an mt-sDNA test between 7/1/2018-6/30/2019 by primary care providers (PCP) were reviewed. Above-average risk for CRC was defined as either a personal history of adenoma(s), CRC, polyposis syndrome, or inflammatory bowel disease; or a family history of CRC. Primary outcome measured was the number of off-label mt-sDNA orders, defined as an order for a patient who 1) required screening colonoscopy due to increased risk for CRC, 2) required diagnostic colonoscopy, or 3) did not require screening at that time. High-volume PCPs (≥20 mt-sDNA orders) were evaluated to assess variability in off-label orders. Predictors of test compliance and positive mt-sDNA result were analyzed. Results: 902 (62.9% female) patients were advised an mt-sDNA test (Table 1), of which 160 (17.7%) met at least one criterion for off-label order (Table 2). Overall compliance with testing was 73.5%, and median interval to result availability was 29 days. Age and sex did not predict compliance. Analysis of high-volume PCPs showed off-label orders ranged from 0-71% (Figure 1), but was not skewed toward outliers. Increasing age was a predictor of an off-label order (OR 2.32 [95% CI, 1.86-2.89] for every 10-year increase in age, p< .0001). On multivariate analysis, increased age (OR 1.04 [1.02-1.06], p=.001) and having an indication for a diagnostic colonoscopy (OR 2.9 [1.01-8.34], p=.048) were significantly associated with a positive mt-sDNA result. Only 96/125 (76.8%) patients with a positive mt-sDNA test underwent a follow-up colonoscopy, and of these 31.6% did not have any neoplastic findings. Discussion: Off-label mt-sDNA testing was advised in 17.7% of our cohort by their primary care providers and this was more likely with increased age. Our results were not skewed by a small group of PCPs despite a wide variation in rate of off label orders. We found high rates of non-compliance with the test (26.5%) and with follow-up colonoscopy for a positive test (23.2%). Off-label use of mt-sDNA testing may lead to increased costs. Increased education of both patients and primary care providers may help reduce the off-label use of mt-sDNA testing.
Table 1. Baseline Data Stratified by mt-sDNA status (compliant or non-compliant). mt-sDNA, multitarget-stool DNA; NA not applicable
Table 2. Off-label criterion met for each mt-sDNA order. *Some patients met more than one criterion. mt-sDNA, multitarget stool DNA; CRC, colorectal cancer; IBD, inflammatory bowel disease; FOBT, fecal occult blood testing; FIT, fecal immunochemical test
Figure 1. Rate of Off-Label Testing per High-Volume Primary Care Provider. High volume primary care providers are defined as those with at least 20 mt-sDNA orders during study period. The top blue horizontal line at 77.1% represents the maximum allowed frequency of off-label orders above which a high-volume provider would be identified as a statistical outlier, however no high-volume PCP exceeded this threshold in the data set.
Disclosures: Amol Agarwal indicated no relevant financial relationships. Talan Zhang indicated no relevant financial relationships. Nishal Ravindran indicated no relevant financial relationships. Paul Thuluvath indicated no relevant financial relationships. Anurag Maheshwari: Exact Sciences – Speaker's Bureau.