Abstract

e16145 Background: Colorectal cancer (CRC) is the second leading cancer worldwide in terms of incidence, 5-year prevalence and mortality for both women and men ages 45 years old and up. The current screening method for many countries with organized screening programs is the FIT test for fecal occult blood; however, this test can result in false positive rates as high as 65%. A FIT reflex test could reduce unnecessary colonoscopies while reducing wait times for those patients that need confirmatory colonoscopies the most. Methods: Danish FIT positive colonoscopy confirmed serum samples (n = 1,499) were divided into training and validation sets maintaining approximately equivalent percentages of clean colonoscopy (40%), low risk adenomas (16%), medium risk adenomas (19%), high risk adenomas (13%), stage I CRC (5%), stage II CRC (2%), stage III CRC (4%), and stage IV CRC (0.5%). Proteins were quantified by custom 16-plex immunoassays utilizing the Luminex xMAP platform. A support vector machine supervised machine learning algorithm was trained with the 16 biomarkers plus age and FIT concentration using 1,291 samples for the outcome medium risk adenoma, high risk adenoma, and CRC. Then this algorithm was tested on a blind 208 sample validation set. Results: The training set was 90% sensitive and 27% specific (AUC = 0.68) and the validation set was 93% sensitive and 21% specific (AUC = 0.63). The sensitivities of the validation by risk/stage was as follows: medium risk adenoma 91%, high risk adenomas 92%, stage I CRC 100%, stage II CRC 100%, stage III CRC 100%, stage IV CRC 93%. Conclusions: This study demonstrates feasibility of a novel blood-based multiplex protein immunoassay for use as a reflex to FIT positive results in population wide screening. It detected nearly all adenomas and carcinomas while reducing FIT false positives and thus unnecessary colonoscopies by more than 20%. A FIT reflex test could alleviate endoscopy burden experienced in countries with organized cancer screening programs, while providing better patient outcomes by detecting polyps and early-stage CRC with high sensitivity.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.