Clinical Utility of ctDNA-Guided Adjuvant Therapy in Stage II Colon Cancer

Clinical Summary: 

• Design/Population: The phase 3 CIRCULATE trial evaluated tumor-informed ctDNA testing to guide adjuvant chemotherapy in patients with proficient mismatch repair, microsatellite-stable stage II colon cancer who lacked conventional indications for postoperative chemotherapy.
• Key Outcomes: ctDNA positivity identified a subgroup at substantially higher risk of recurrence, and adjuvant chemotherapy markedly improved disease-free survival compared with observation among treated ctDNA-positive patients. These findings support the prognostic and predictive value of postoperative ctDNA testing.
• Clinical Relevance: The results suggest that ctDNA can identify patients with molecularly high-risk stage II colon cancer who may benefit from treatment escalation despite otherwise low-risk clinicopathologic features.

Gunnar Folprecht, MD, PhD, University Hospital Dresden, Dresden, Germany, discusses results from the CIRCULATE trial evaluating whether circulating tumor DNA (ctDNA) can identify patients with stage II colon cancer who benefit from adjuvant chemotherapy.

The study demonstrated that ctDNA-positive patients had a substantially higher risk of recurrence than ctDNA-negative patients and that adjuvant chemotherapy significantly improved outcomes among patients with molecularly detectable residual disease. Together with findings from DYNAMIC, CIRCULATE-Japan, and BESPOKE, these results support ctDNA-guided treatment decisions in patients with pMMR, microsatellite-stable stage II colon cancer.

Transcript: 

Hello, my name is Gunnar Folprecht, I'm working at Dresden University Hospital, and I'm a medical oncologist specialized in colorectal cancer. We performed the CIRCULATE trial, which is a trial that investigated circulating tumor DNA (ctDNA) in stage II colon cancer.

As you know, decision-making in stage II colon cancer for adjuvant therapy is difficult. There are some risk factors, like T4 or emergency resection, that are regarded as indications towards adjuvant therapy. However, the predictive value is limited.  The mainstay of adjuvant therapy in colon cancer is fluoropyrimidine, and there is only a limited incremental benefit of oxaliplatin.

What we already knew is that ctDNA has a prognostic value when we determine the mutations that are present in the tumor and take a blood sample after resection of the patient. What we wanted to know was whether the patients actually benefit from adjuvant chemotherapy. For that purpose, we investigated patients with stage II colon cancer who had no certain indication for adjuvant therapy. We determined the ctDNA status with an academic tumor-informed test, so we knew which mutations were in the tumor. 

We found approximately 3% of the patients to be ctDNA positive. This meant that, out of 2,100 screened patients and 1,700 analyzed patients, 55 patients were ctDNA positive. We could not randomize all analyzed patients, we randomized nearly 1,400. In total, this meant 41 patients who were ctDNA positive. These patients were randomized 2:1 to receive chemotherapy or to be in an observation arm, and the ctDNA-negative patients were randomized to be observed or to be off study. 

Of the 26 patients who were assigned to receive chemotherapy, not all actually received it. In 2 patients, the investigators immediately performed imaging because we unblinded both the patient and the investigator with the information that the patient was ctDNA positive and should receive chemotherapy. At that time, metastatic disease was found that otherwise would not have been detected. Three patients withdrew consent. 

Therefore, in addition to the intention-to-treat analysis, we performed a per-protocol analysis where we analyzed the patients who actually received chemotherapy and compared them with the ctDNA-positive patients who did not receive chemotherapy. The differences we found were dramatic. First, there was a huge prognostic effect between ctDNA-positive and ctDNA-negative patients. The ctDNA-positive patients had a disease-free survival of only 38%, compared with a disease-free survival of 87% in the ctDNA-negative patients. The hazard ratio was 6.3.

The patients who were simply observed had a recurrence rate of 62%, compared with only 19% of patients who were treated with chemotherapy, and 12% of the ctDNA-negative patients. The hazard ratio was 0.23, with a P value of .009. So, there was a huge reduction in recurrence among the ctDNA-positive patients who actually received chemotherapy—from 62% to 19%. If we look at disease-free survival, it increased from 38% to 77% in the patients who actually received chemotherapy. 

The intention-to-treat analysis also showed a large difference. The hazard ratio was 0.55, but the difference was not statistically significant because of the low patient numbers. That doesn't mean there is no effect– it means that we could not prove the effect in the intention-to-treat analysis.

What we concluded is that, together with the literature we already had—such as the DYNAMIC trial and the observational studies CIRCULATE-Japan and the BESPOKE study in the United States—this evidence together justifies treatment escalation in patients who are proficient mismatch repair, microsatellite stable, and have stage II colon cancer, but who otherwise would not receive chemotherapy.  

Source: 

Folprecht G, Stasik S, Reinacher-Schick A, et al. Chemotherapy for patients with circulating tumour DNA-positive, stage II colon cancer (CIRCULATE)—An AIO/ABCSG trial. Ann Oncol. Published online: May 31, 2026. doi: 10.1016/j.annonc.2026.05.001