Stool DNA Testing: Is there a Future?

This commentary describes the current status of the multi-target stool DNA test (mt-sDNA) (Cologuard®, Exact Sciences, Madison, WI) and of stool-based markers for GI diseases.

Approved by both the U.S. Food and Drug Administration and the Center for Medicare and Medicaid Services in August 2014, the mt-sDNA test is a non-invasive screening test for colorectal cancer (CRC) for persons whose preference is to avoid screening colonoscopy. Since approval, clinical use of mt-sDNA has had a steady trajectory among both patients and providers (Figure 1). Relevant to the “80 percent by 2018” goal of CRC screening, 42 percent of mt-sDNA users indicated that it was their first CRC screening test.

Scientific and regulatory progress of mt-sDNA includes the following:
• A cross-sectional study of mt-sDNA’s test characteristics among 661 Alaskan Natives scheduled for average-risk screening or surveillance colonoscopy,1 showed findings that are nearly superimposable on those of the pivotal (DeeP-C) study. For the 10 CRCs discovered, sensitivity was 100 percent for mt-sDNA and 80 percent for the fecal immunochemical test (FIT). For the nine large sessile serrated polyps, respective sensitivities were 67 percent and 11 percent. Specificity was 93 percent for mt-sDNA and 96 percent for FIT.1

• Mt-sDNA was included as an option in screening guidelines by the American Cancer Society, U.S. Preventive Services Task Force and National Comprehensive Cancer Network.

• In a retrospective cohort study of 160 subjects with a false-positive mt-sDNA test who were followed for a median duration of four years, eight aerodigestive cancers occurred, all of which were diagnosed more than three years following the false-positive test result, with an occurrence rate lower than expected when compared to SEER data.2 While the results may alleviate concerns about missed cancers, more data are needed; these studies are ongoing.

• Mt-sDNA was included into HEDIS measures by the National Committee for Quality Assurance, ensuring that primary care providers receive “credit” for using mt-sDNA for CRC screening.

The potential of non-invasive testing for CRC is great.

Residual clinical concerns with mt-sDNA include its specificity relative to FIT, cost and cost-effectiveness. While mt-sDNA’s specificity in the pivotal study was 86.6 percent (versus 94.9 percent for FIT), the study population was skewed toward older persons within the screening age range. Older age is associated with higher prevalence of background methylation of DNA, one reason for false positivity. In persons younger than 65 years who had non-advanced neoplasia and “clean” colons, respective specificities were 91.5 and 94.0 percent. Last, the 5 percent false-positive rate for FIT, when considered programmatically over a three-year interval, may be considered comparable to that of mt-sDNA every three years. These data should assuage concerns about mt-sDNA’s “low” specificity, particularly in younger persons within the screening age range.

While mt-sDNA every three years is cost-effective when compared to no screening, it is dominated (i.e., more costly, less effective) by colonoscopy every 10 years and annual FIT when adherence is perfect.3 In the real world, however, adherence for any CRC screening test is usually far less than perfect. For mt-sDNA to be comparably cost-effective to FIT requires substantially higher participation rates to those of FIT in organized programs and in the opportunistic screening that characterizes most of the U.S. The accompanying navigation system for mt-sDNA has resulted in an adherence rate of 67 percent — twice the rate of adherence to three rounds of FIT4 — and in a population with a high proportion of persons resistant to screening. However, it should be noted that there is no comparison of programmatic FIT and mt-sDNA for clinically important outcomes of CRC incidence or mortality.

Beyond CRC screening, stool DNA markers may be useful for surveillance of dysplasia in inflammatory bowel disease. A subgroup of mt-sDNA markers and a panel of methylated markers have demonstrated sensitivity for CRC plus high-grade dysplasia of 92 to 100 percent at specificity of 89 to 94 percent.5 These preliminary findings require clinical validation, with particular attention to specificity and the meaning and management of the scenario in which the stool test is positive, but biopsies for (high-grade) dysplasia are negative. And beyond the colon, there is ongoing investigation of a stool DNA-based test to function as a cancer screening test for the GI tract, although this particular indication is nascent and very preliminary.

Identification of discriminating markers in stool (as well as blood) is an active area of research that includes a myriad of “discovery phase” studies of both individual and panels of markers. Nearly all are case-control studies that compare one or markers in a case group with CRC with a CRC-free control group, and may or may not include a group with advanced or non-advanced adenomas. This discovery phase has been driven by technological advances including the digital melt curve method, digital PCR, quantitative real-time target and signal amplification (QuARTS), and beaming, emulsion, amplification and magnetics (BEAMing), all of which have improved the detection threshold in stool from 1 percent of mutated copies of DNA to less than 0.1 percent.

Based on the stagnant 65 percent adherence rate for CRC screening in the U.S., there is room for mt-sDNA as well as FIT and other accurate, non-invasive tests to improve adherence. The potential of non-invasive testing for CRC is great, with opportunity to identify biomarkers that are discriminating for CRC and advanced, precancerous polyps as well as for other prevalent gastrointestinal cancers. But much time, effort and cost will be required for meticulous marker selection and careful prospective validation of candidate markers in the target population for which the test is intended.

Dr. Imperiale’s institution receives grant support from Exact Sciences.

1. Redwood, D.G., Asay, E.D., Blake, I.D. et al, Stool DNA testing for screening detection of colorectal neoplasia in Alaska native people. Mayo Clin Proc.. 2016;91(1):61-70.
2. Cotter, T.G., Burger, K.N., Devens, M.E. et al, Long-term follow-up of patients having false-positive multitarget stool DNA tests after negative screening colonoscopy: The LONG-HAUL cohort study. Cancer Epidemiol Biomarkers Prev. 2017; 26(4):614-621.
3. Ladabaum, U., Mannalithara, A. Comparative effectiveness and cost-effectiveness of a multitarget stool DNA test for colorectal neoplasia. Crohn’s disease management after intestinal resection: a randomized trial. Gastroenterology. 2016;151(3):427-439.e6.
4. Jensen, C.D., Corley, D.A., Quinn, V.P. et al, Fecal immunochemical test program performance over 4 rounds of annual screening: a retrospective cohort study. Ann Intern Med. 2016;164(7):456-463.
5. Kisiel, J.B., Yab, T.C., Nazer Hussain, F.T. et al,Stool DNA testing for the detection of colorectal neoplasia in patients with inflammatory bowel disease. Aliment Pharmacol Ther. 2013;37(5):546–554.

Join the discussion

Your email address will not be published. Required fields are marked *