Abstract 1933: Improving detection of colorectal cancer recurrence using serial ctDNA measurements: A systematic literature review

Abstract Introduction: Detecting molecular residual disease (MRD) in solid tumors via circulating tumor DNA (ctDNA) is an emerging practice to effectively assess curative treatment success and inform risk for future recurrence. Methods: A systematic literature review (SLR) search was executed in PubMed and Embase from 01/01/2018 to 11/01/2023 to identify studies detecting MRD via ctDNA in adult patients with solid tumors after surgical resection. Studies providing diagnostic sensitivity and specificity for recurrence, and clinical lead times for colorectal cancer (CRC) are reported. Results: A total of 4, 308 non-duplicate records were screened by title and abstract; 224 were screened as full texts; 19 of the included studies reported the outcomes of interest in CRC. There was notable heterogeneity related to sample timing and frequency, length of follow-up time, and ctDNA assay methodology. Most patients were stage II/III; median follow-up time ranged from 12.5-49 months. Overall sensitivity for detecting recurrence ranged from 27.0%-100%; specificity from 80.1%-100%. Sensitivity for recurrence for single post-surgical or post-adjuvant samples ranged from 27.0%-67.0% (n=5). Sensitivity was higher for studies employing serial sampling (e.g., every 3-6 months), ranging from 66.7%-100%; with most studies (n=4) reporting values ranging from 82.6%-100%. Importantly, ctDNA detected recurrence several months earlier than radiological imaging for all studies reporting lead time (i.e., difference between ctDNA and radiological detection) (Table Table: Diagnostic performance and lead time for included CRC studies Sensitivity for Recurrence (Range %) Specificity for Recurrence (Range %) Median Lead Time Mean Lead Time Overall (All Studies) 27.0%-100% (13 studies, n=1, 917) 80.1%-100% (10 studies, n=1, 688) 1-11.5 months (6 studies, n=88) 2.6-8.7 months (3 studies, n=36) Single Sample 27.0%-67.0% (5 studies, n=746) 86.5%-100% (5 studies, n=746) 3 months (1 study, n=9) 2.6 months (1 study, n=3) Serial Sampling (Predetermined Intervals) 66.7%-100% (5 studies, n=452) 84.0%-100% (4 studies, n=394) 5.3-11.5 months (2 studies, n=23) 5.01-8.7 months (2 studies, n=33) ). Conclusion: Variability of ctDNA sensitivity in detecting recurrence may be due to factors related to the heterogeneity of sample timing, follow-up duration, population, and assays. Regardless of the variation in sensitivity, ctDNA appears to be a valuable tool for assessing MRD and detecting recurrence earlier than radiological imaging, especially with repeated sampling. Citation Format: Ning Yan Gu, Christine Waters-Banker, Gebra Cuyun Carter, Danielle Badgley, Aleksandra (Sasha) Vidisheva, Melanie Palomares, Heather Johnson, Nathaniel J. Smith. Improving detection of colorectal cancer recurrence using serial ctDNA measurements: A systematic literature review [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1933.