Postoperative surveillance after curative-intent colorectal cancer surgery aims to identify recurrence while metastatic disease remains amenable to potentially curative treatment. However, conventional follow-up using scheduled computed tomography imaging and carcinoembryonic antigen testing may detect recurrence only after the disease has become more extensive.
Results from the randomized phase 3 FIND trial showed that a dynamic surveillance strategy guided by circulating tumor DNA methylation detected recurrence earlier and significantly increased the proportion of patients who received curative-intent metastasis-directed therapy compared with standard CT-based surveillance. Overall recurrence rates were similar between the groups, but ctDNA-guided surveillance identified recurrent metastatic disease with a more favorable and potentially resectable anatomical profile.
The article, titled “Dynamic Circulating Tumor DNA Methylation Monitoring Guiding Postoperative Surveillance in Nonmetastatic Colorectal Cancer: A Prospective, Randomized, Phase III FIND Trial,” was published in the Journal of Clinical Oncology on July 29, 2026.
Authors: Shaobo Mo, Chaoqiang Zhou, Maoguang Ma, Wenqin Luo, Yaqun Li, Ping Lu, Peiyuan Tang, Yaqi Li, Xiaoji Ma, Xiang Hu, Chunliang Cheng, Jiajun Yang, Changhua Zhuo, Jinliang Jian, Cheng Yu, Jinhuo Ding, Chaoxia Xiong, Fei Jiang, Rongrong Mu, Zhuo Lu, Jiayi Yu, Shengnan Jin, Ju Luan, Xinxiang Li, Sanjun Cai, Hai Zou, Yiwei Li, Qingguo Li, Fangqi Liu, Chunming Ding, and Junjie Peng.
Study Design
FIND was an investigator-initiated, prospective, multicenter, randomized phase 3 trial conducted across six academic medical centers in China. The trial enrolled adults with stage I–III colorectal cancer who had undergone R0 resection and had no radiologically detectable distant metastases before surgery. Patients were required to be medically fit for curative-intent resection.
A total of 728 patients were randomly assigned to ctDNA-guided surveillance or standard CT-based follow-up. After exclusions, the modified intention-to-treat population included 584 patients: 289 in the ctDNA-guided group and 295 in the control group. The majority of patients had stage II or III disease, and baseline characteristics were generally balanced between the groups.
In the experimental group, plasma ctDNA methylation testing was performed before surgery, within one month after surgery, and every three months for two years.
A positive ctDNA result triggered immediate contrast-enhanced CT imaging of the chest, abdomen, and pelvis. If imaging did not confirm recurrence, CT was repeated every two months alongside quarterly ctDNA testing. Standard imaging intervals resumed after two consecutive negative ctDNA results.
Patients in the control group underwent conventional postoperative surveillance according to established guidelines.
The primary endpoint was the proportion of patients with radiologically confirmed recurrence who received curative-intent metastasis-directed therapy. Secondary endpoints included time to clinical recurrence, overall survival, ctDNA-based disease-free survival, imaging-based disease-free survival, and patient-reported outcomes.
Increased Use of Curative-Intent Treatment
At the February 2026 data cutoff, the median follow-up was 23.3 months. Recurrence occurred in 52 of 289 patients in the ctDNA-guided group and 55 of 295 patients in the control group. The cumulative recurrence rates were similar between the two strategies, at 18.0% and 18.6%, respectively.
Among patients who developed recurrence, however, the proportion receiving curative-intent metastasis-directed therapy was significantly higher with ctDNA-guided surveillance:
- 48.1% in the ctDNA-guided group
- 23.6% in the standard-surveillance group
The relative risk was 2.03, with an absolute difference of 24.4 percentage points in favor of ctDNA-guided surveillance. The benefit remained consistent in the per-protocol analysis. Among patients with recurrence, postrecurrence progression-free survival was also longer in the ctDNA-guided group, with a hazard ratio of 0.51.
Earlier Detection of Recurrence
The median time from surgery to clinical detection of recurrence was 9.5 months with ctDNA-guided surveillance, compared with 13.4 months with standard follow-up.
This represented a median lead time of 3.9 months.
Among patients who experienced recurrence in the ctDNA-guided group, ctDNA detection preceded radiographic confirmation by two months or less in 36% of cases, four months or less in 56%, and eight months or less in 81%. Although recurrence was identified earlier, the similar overall recurrence rates between the groups indicate that ctDNA surveillance did not alter the underlying likelihood of relapse. Instead, it changed the timing and clinical presentation of recurrent disease.
More Favorable Metastatic Features
The distribution of recurrence sites was similar between the groups, with approximately 61% of recurrences confined to the liver and/or lungs. Among all patients with recurrence, curative-intent treatment for disease confined to the liver and/or lungs was administered to 42.3% in the ctDNA-guided group and 18.2% in the control group.
Patients with liver metastases detected through the ctDNA-guided strategy also had more favorable anatomical disease characteristics. Compared with patients undergoing standard surveillance, they were more likely to have three or fewer liver lesions, a maximum lesion diameter of 3 cm or less, and disease confined to a single liver lobe.
Three or fewer liver lesions were reported in 75.0% versus 28.6% of patients, lesions measuring 3 cm or less in 90.0% versus 57.1%, and unilobar disease in 80.0% versus 28.6%, respectively. These findings suggest that ctDNA-triggered imaging identified hepatic metastatic disease while it remained more anatomically suitable for curative-intent intervention.
ctDNA and CEA Performance
Longitudinal methylation-based ctDNA monitoring demonstrated a sensitivity of 84.1% and a specificity of 91.4% for predicting recurrence. By comparison, carcinoembryonic antigen monitoring within the same cohort had a sensitivity of 35.7% and a specificity of 94.0%.
Positive ctDNA status was also a stronger independent predictor of recurrence-free survival than elevated carcinoembryonic antigen levels. Persistent ctDNA positivity after adjuvant chemotherapy was associated with worse recurrence-free survival than conversion from positive to negative status.
Limitations
The investigators emphasized that longer follow-up is required to determine whether earlier recurrence detection and increased use of curative-intent treatment translate into improved overall survival.
The median follow-up of 23.3 months was sufficient to evaluate the primary endpoint but remains too short for mature three-year and five-year survival analyses. The trial also evaluated a specific methylation-based ctDNA assay. The generalizability of the findings to mutation-based assays, next-generation sequencing platforms, or alternative surveillance schedules remains uncertain.
Comprehensive health-economic analyses were pending, while longitudinal patient-reported outcomes, including quality of life and fear of recurrence, remained immature.
Takeaway
The phase 3 FIND trial showed that ctDNA methylation-guided dynamic surveillance significantly increased the proportion of patients with recurrent colorectal cancer who received curative-intent metastasis-directed therapy compared with standard CT-based follow-up.
The strategy detected recurrence a median of 3.9 months earlier and identified liver metastases with fewer lesions, smaller tumor size, and more frequently unilobar involvement.
These results support the potential role of ctDNA methylation as a dynamic marker to guide postoperative surveillance and the timing of intervention. However, mature survival data are needed before determining whether earlier detection and increased use of curative-intent treatment ultimately improve long-term survival. The full article is available in the Journal of Clinical Oncology.
Find more videos on OncoDaily YouTube TV.