Circulating tumor DNA detection as a prognostic tool in patients with localized colon cancer: A systematic review and meta-analysis.

Abstract

e15629 Background: Current modalities for postoperative risk stratification in localized colon cancer are still suboptimal in guiding adjuvant chemotherapy (ACT) administration. In recent years, circulating tumor DNA (ctDNA) has emerged as a novel biomarker to identify patients at high-risk for recurrence following R0 tumor resection. In this meta-analysis, we aimed to assess the prognostic value of ctDNA detection on postoperative recurrence risk that in the future, with further study, may help guide ACT treatment. Methods: Electronic databases Pubmed, Google Scholar, and Embase were searched up to 31 January 2023. Patient data were extracted using the following inclusion criteria: stage I to III colon cancers, R0 resection, sufficient ctDNA data following surgery, and no other concurrent malignancy. Pooled relative risk (RR) and 95% confidence interval (CI) were estimated to assess ctDNA positivity on the risk of recurrence in patients having surgery only, patients having surgery followed by ACT, and all patients combined. Random effects model was implemented to account for inter-study variation. Study heterogeneity was detected using χ2 test and I2 coefficient, with I2 more than 50% indicating significant heterogeneity. All analyses were completed using RevMan 5, with p < 0.05 being statistically significance. Results: Fifteen studies with a total of 3411 patients (64.4±10.7 years old) were evaluated. In patients undergoing surgery only without ACT treatment, postoperative ctDNA-positive patients were at higher risk of recurrence compared to those with negative ctDNA (RR = 7.73, 95% CI: 5.73-10.42, p < 0.01, I2= 0%). In patients having surgery followed by ACT, postoperative/pre-ACT ctDNA-positive patients were at higher risk of recurrence compared to ctDNA-negative ones (RR = 5.25, 95% CI: 3.67-7.50, p < 0.01, I2= 19%); post-ACT ctDNA-positive patients were also at higher risk of recurrence (RR = 6.34, 95% CI: 4.12-9.78, p < 0.01, I2= 29%). When combining the two populations (surgery only and surgery followed by ACT), postoperative ctDNA-positive patients again have higher risk of recurrence (RR = 3.86, 95% CI: 2.63-5.67, p < 0.01, I2= 87%); after completing all treatments, ctDNA-positive patients continue to have higher risk of recurrence (RR = 7.02, 95% CI: 3.40-14.51, p < 0.01, I2= 94%). Conclusions: This meta-analysis shows that the presence of ctDNA post-surgery correlates with higher risk of disease recurrence. In localized colon cancers, ctDNA detection can be a valuable tool to identify high-risk patients who, regardless of pathologic stage, may be the most likely to benefit from ACT treatment. Additionally, patients with persistently positive ctDNA after completing ACT are at very high risk for recurrence and should require closer follow-up than current guidelines, along with consideration of enrollment in clinical trials of newer treatment strategies.