Back

Point Sensitivity of the Multi-target Stool DNA Test for Colorectal Cancer Screening in Individuals Ages 45-49 Years

Domanico, M.; Statz, S.; Weiser, E.; Berger, B. M.; Limburg, P. J.

2020-09-03 gastroenterology
10.1101/2020.09.01.20177568 medRxiv
Show abstract

BackgroundMost colorectal cancer (CRC) screening tests have not been rigorously studied in younger age groups. AimsTo estimate sensitivity of the multi-target stool DNA (mt-sDNA) test in patients ages 45-49 years. MethodsWe identified archived stool samples (Exact Sciences; Madison, WI) from individuals ages 45-49 years who had completed an index colonoscopy and had confirmed diagnoses of CRC or advanced precancerous lesions (APL; defined as high-grade dysplasia, >25% villous morphology, or [≥]1 cm in size [conventional adenoma or serrated lesion]). Data annotation referent to potential CRC risk factors, other than age, was limited. Stool samples were collected at least 7 days after the index colonoscopy, prior to lesion excision or treatment. Stool samples were processed and analyzed per established laboratory protocols for the mt-sDNA assay. Mt-sDNA test sensitivity for CRC, APL, and CRC+APL was estimated from the available sample set. Samples were collected from 2010-2013 (NCT01260168) and 2014-2017 (NCT02503631), with sample testing and analysis in 2019. ResultsStool samples were analyzed from 19 eligible subjects, 13 with CRC and 6 with APL. Estimated mt-sDNA test sensitivity for CRC, APL, and CRC+APL were 92%, 83%, and 89%, respectively. ConclusionsIn this small pilot study using existing archived stool samples from subjects ages 45-49 years, mt-sDNA test sensitivity was similar to previously reported estimates for individuals ages [≥]50 years. These results support the application of mt-sDNA screening to 76 average-risk patients beginning at age 45 years. Larger studies are needed to confirm and extend these findings.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.