ctDNA Shows High Recurrence Detection Sensitivity in Stage II/III CRC

Post-operative ctDNA testing led to a change in adjuvant management in 1 of 6 patients with stage II/III colorectal cancer treated in the BESPOKE trial.

Circulating tumor DNA (ctDNA) testing demonstrated high sensitivity in detecting recurrence in multiple sites, including low ctDNA-shedding sites, in patients with stage II/III colorectal cancer (CRC), according to results from a sub cohort final analysis of the observational BESPOKE trial (NCT04264702) presented at the 2025 ASCO Gastrointestinal Cancer Symposium.¹

Data from the trial reveal that 16.3% of treating oncologists stated that ctDNA testing use influenced their treatment decision for adjuvant therapy. Among those who affirmed that ctDNA testing use impacted adjuvant therapy use, 59.9% de-escalated adjuvant chemotherapy (ACT), and 35.7% escalated adjuvant therapy. Of note, 13.2% of those who affirmed ctDNA impact escalated adjuvant therapy for metastatic disease. Among the 83.7% who did not report that ctDNA use influenced their treatment decision-making for adjuvant therapy, 83.8% reported that they felt more comfortable with their previously planned treatment based on test results.

Among 188 patients who recurred, 162 (86%) were ctDNA-positive within week 24. Among those patients, 48 (30%) received oligometastasis–directed therapy (MDT). Furthermore, the most common sites for metastases were the liver (28.7%), abdomino-pelvic mass (11.7%), peritoneum and lymph nodes (both 10.6%), and lung (9.0%). The sensitivity rates in each site were 96% for liver, 89% for lymph nodes, 86% for abdomino-pelvic mass, 79% for peritoneum, and 76% for lung.

Additional data from the trial revealed that patients with ctDNA-positive stage II and III CRC experienced inferior disease-free survival (DFS). For those with ctDNA-positive stage II disease, the median DFS was 12.7 months (95% CI, 8.3-not estimable [NE]) vs not estimable (95% CI, NE-NE) in patients with ctDNA negativity, with respective 2-year DFS rates of 45.9% (95% CI, 32.1%-65.8%) and 91.8% (95% CI, NE-NE). For those with ctDNA-positive stage III disease, the median DFS was 16.2 months (95% CI, 13.6-18.9) vs not estimable (95% CI, NE-NE) in patients with ctDNA-negative disease, with respective 2-year DFS rates of 35.5% (95% CI, 28.6%-44.2%) and 87.4% (95% CI, 83.8%-91.1%).

Furthermore, patients who had ctDNA positivity at the first surveillance timepoint had an inferior DFS than those with ctDNA–negative disease (HR, 20.63; 95% CI, 14.37-29.61; P <.0001). A time-dependent DFS analysis found that the hazard ratio was 26.4 (95% CI, 21.6-32.4) for patients with stage II/III disease and ctDNA positivity (P <.0001).

Additionally, ctDNA clearance, defined as conversion of a positive ctDNA test after surgery into a negative test without subsequent conversion, was associated with superior DFS. For ctDNA clearance by month 3, the hazard ratio was 0.43 (95% CI, 0.29-0.64; P <.0001), and by month 6, it was 0.31 (95% CI, 0.19-0.52; P <.0001). The ctDNA clearance rates in those with minimal residual disease (MRD)–positive status (n = 180) were 49.44% (95% CI, 42.2%-56.7%) by week 12, 67.22% (95% CI, 60.1%-73.7%) by week 20, 73.33% (95% CI, 68.4%-81.3%) by month 6, and 75.56% (95% CI, 68.8%-81.3%) overall.

ctDNA testing was also predictive of the benefit of ACT in patients with stage II/III CRC. Among patients with MRD–positive status, ACT elicited a median DFS of 17.7 months (95% CI, 14.6-21.4) after surgery vs 7.1 months (95% CI, 4.6-21.4) with observation (HR, 0.48; 95% CI, 0.31-0.77; P = .0008). Additionally, the DFS rates were 40.3% (95% CI, 33.3%-48.9%) and 24.7% (95% CI, 13.2%-46.3%) in the respective arms.

“[The] prospective observational study showed that post-operative ctDNA testing led to a change in adjuvant management for 1 of 6 patients, and validated the adjuvant plan for most patients,” Purvi Shah, MD, medical oncologist and hematologist at the Virginia Cancer Institute, stated in the presentation.¹ “In the surveillance setting, ctDNA demonstrated high sensitivity to detecting recurrence across various sites, including low ctDNA–shedding sites such as the peritoneum and the lung. Serial ctDNA testing was associated with a high rate of potentially curative [MDT].”

A total of 1166 patients with stage II/III CRC were enrolled in the final sub cohort analysis. Patients were divided into either the ACT cohort (n = 694) or observation cohort (n = 472).

Among patients in the final sub cohort analysis, the median age was 61.8 years (range, 22-94), 56.7% were male, 72.4% were White, and the median follow-up was 23.9 months (range, 0.5-36.5). A total of 44.3% had stage II disease, 55.7% had stage III disease, 57.8% were younger than 65, and 63.7% had an ECOG status score of 0 followed by 30.4% with a score of 1. Furthermore, 59.5% received ACT, and 83.9% did not experience disease recurrence.

Of note, a total of 7.54% (95% CI, 5.57%-10.15%) of patients with stage II disease had MRD positivity vs 28.35% (95% CI, 25.02%-31.94%) of those with stage III disease.

The coprimary end points were assessment of ctDNA testing use on treatment decisions and recurrence rate of patients diagnosed with CRC while asymptomatic using ctDNA testing.² Secondary end points included MRD clearance, percent of those undergoing surgery for oligometastases, and overall survival.

References

1. Shah PV, Aushev VN, Ensor J, et al. Circulating tumor DNA for detection of molecular residual disease (MRD) in patients (pts) with stage II/III colorectal cancer (CRC): final analysis of the BESPOKE CRC sub-cohort. J Clin Oncol. 2025;43(suppl 4):15. doi:10.1200/JCO.2025.43.4_suppl.15
2. BESPOKE study of ctDNA guided therapy in colorectal cancer. ClinicalTrials.gov. Updated December 12, 2023. Accessed January 25, 2025. https://tinyurl.com/6as8shw9