Postoperative ctDNA Status May Prognosticate DFS Outcomes in Resected CRC

This article was originally published by OncLive. This version has been lightly edited.

Postoperative circulating tumor DNA (ctDNA)–based minimal residual disease (MRD) status demonstrated prognostic value for patients with stage 1 to 4 resected colorectal cancer (CRC), showing that those with positive ctDNA after surgery have significantly lower disease-free survival (DFS) outcomes at 24 months vs those who were ctDNA negative, according to findings from the observational GALAXY study (UMIN000039205) presented during the 2023 ESMO Congress. Results also indicate that this biomarker may be used to identify patients with stage 2 or 3 CRC who would derive benefit from adjuvant chemotherapy.¹

Assessment of ctDNA status during the MRD window of 2 to 10 weeks following surgery revealed that ctDNA-negative patients with stage 1 to 4 disease (n = 1773) experienced a 24-month DFS rate of 89% (95% CI, 86.9-90.8) vs 31.4% (95% CI, 24.7-38.3) in those who were ctDNA positive (n = 247; HR, 12.46; 95% CI, 9.85-15.76).

Large and small intestine | Image credit: Tom - stock.adobe.com

Moreover, early ctDNA dynamics were also associated with DFS outcomes. Examination of ctDNA dynamics from the MRD window of 10 to 16 weeks post-surgery showed patients with persistent ctDNA negativity (n = 1369) experienced a 24-month DFS rate of 90.3% (95%, 88.1-92.1) compared with 53.9% (95% CI, 42.1-64.3) in patients who converted to negative ctDNA status (n = 109; HR, 6.06; 95%, 42.1%-64.3%), 20.7% (95% CI, 4.7-44.5) in patients who converted to positive ctDNA status (n = 28; HR, 12.7; 95% CI, 7.57-21.3), and 9.8% (95% CI, 4.4-17.7) in patients who remained ctDNA positive (n = 90; HR, 38.4; 95% CI, 27.94-52.78).

"Analysis of the ctDNA dynamics showed the best DFS [was observed] for the patients with consistently negative ctDNA, followed by those converting from positive to negative, negative to positive, and persistently positive," Yoshiaki Nakamura, MD, PhD, of National Cancer Center Hospital Japan East details in an interview with OncLive during the Congress. "[Notably], patients converting from ctDNA negative to positive [status] can still have poor prognosis and a lower DFS. [Their DFS rate] was superior compared with the persistently positive patients, but…we should consider some interventions for such high-risk populations."

A multivariate analysis in patients with stage 2 to 3 CRC also highlighted that ctDNA positivity during the MRD window 2 to 10 weeks following surgery had the highest prognostic value for poor DFS outcomes (HR, 10.44 [range, 7.691-14.18]). The 24-month DFS rates were 89.1% (95% CI, 86.9-91.0) for the ctDNA-negative population (n = 1416) vs 33.8% (95% CI, 26.3-41.5) for the ctDNA-positive population (n = 209).

Postoperative recurrence remains a major challenge for patients with CRC who undergo standard surgical resection. Accordingly, the prospective, observational GALAXY study was designed to explore how ctDNA status may inform patient treatment and serve as a potential prognostic marker for patient outcomes.

“ctDNA-[based] MRD [status] was known to be associated with recurrence risk [in CRC],” Nakamura expands. “However, the utility of the MRD analysis as a [prognostic] model was unknown.”

The observational GALAXY study monitored ctDNA status in patients with clinical stage 1 to 4 CRC following curative-intent surgery as part of the ongoing CIRCULATE-Japan platform. Prior results from the study published in Nature Medicine in January 2023 showed that ctDNA positivity at 4 weeks after surgery was associated with a higher risk of recurrence (HR, 10.0) and was the most significant prognostic factor associated with recurrence risk in patients with stage 2 or 3 CRC (HR, 10.82).²

“At ESMO, we updated our results. Updated results [showed that] ctDNA-positivity at the MRD window was significantly associated with shorter DFS. That was consistent with our previous report,” Nakamura said.

In the updated analysis, investigators assessed the relationship between post-operative ctDNA–based MRD positivity and DFS in this population. The data cut off was August 28th, 2023, and the median follow-up was 16.4 (range, 0.1-37.4) months. The study was designed to include patients with clinical stage 2 to 4 CRC. In this study, serial ctDNA assessments were performed using the Signatera tumor-informed assay at specific timepoints following surgery.

“Signatera is a tumor-informed assay. First, a surgically resected tumor was analyzed by sequencing, and then 16 mutations were selected…then ctDNA was assessed. If a mutation was detected, ctDNA was regarded as a positive,” Nakamura said.

The timepoints of ctDNA assessment included 1 month, 3 months, 6 months, 9 months, 12 months, 18 months, and 24 months after surgery, or until the time of recurrence. The study’s primary endpoint was DFS.

A total of 3034 patients were enrolled onto the study between May 2022 and November 2022. Of these, 2625 had ctDNA available after surgery and were included in the current analysis. Patients were excluded from the analysis for having incomplete staging information (n = 25), confirmed pathological stage 0 (n = 11), incomplete resection (n = 78), incomplete clinical follow-up data (n = 282), enrolled in interventional phase 3 trials (n = 12), or withdrew consent (n = 1).

The median age of the 2625 patients enrolled onto the study was 69 (range, 23-95) years, and 53% were male. Most patients had an ECOG performance status of 0 (90%), and just under half had tumors were left side (49%). Other tumor locations included right side (30%) and rectum (18%). Half of patients had stage N1-N2 disease vs stage N0 disease, and 80% had stage T3-T4 disease. Pathological stage 3 disease was present in 40% of patients, followed by stage 2 (30%), stage 4 (17%) and stage 1 (14%).

Neoadjuvant chemotherapy was administered to 11% of patients, while 89% received up-front surgery. Adjuvant chemotherapy was administered to 39% of patients following surgery vs 61% who underwent observation alone. Regarding mutational status, 92% and 56% of patients were BRAF and RAS wild-type, respectively. The majority were microsatellite stable (MSS) or microsatellite instability (MSI)–low (90%) and had not experienced radiological or clinical recurrence (81%).

The potential benefit of adjuvant chemotherapy was further investigated in patients with stage 2 to 3 CRC. In patients who were ctDNA negative (n = 1418), the 24-month DFS rate of 88.3% (95% CI, 84.4-91.2) for those who received adjuvant chemotherapy (n = 626) vs 89.9% (95% CI, 87.0-92.1) for those who underwent observation (n = 792; adjusted HR, 1.39; 95% CI, 0.88-2.20). Conversely, ctDNA-positive patients with stage 2 or 3 disease who received adjuvant chemotherapy (n = 156) achieved a 24-month DFS rate of 38.6% (95% CI, 29.6-47.5) vs 16.1% (95% CI, 6.0-30.6) in patients who underwent observation (n = 59; adjusted HR, 3.29; 95% CI, 2.13-5.07).

“In our study, [we demonstrated] significant benefit with adjuvant chemotherapy in MRD-positive vs -negative patients,” Nakamura explained. “The MRD-positive patient [population] has a higher risk of disease recurrence; such patients may [therefore experience] more benefit from adjuvant chemotherapy.”

Additionally, patients with stage 2 or 3 CRC who received adjuvant chemotherapy for either 3 months or 6 months within the entire cohort revealed that ctDNA MRD-negative patients did not exhibit a significant benefit in DFS. However, MRD-positive patients benefited significantly from 6 months of adjuvant chemotherapy at the 1- and 2-month landmark time points (HR, 0.53; 95% CI, 0.29-0.94).

“We showed the utility of ctDNA for the prediction of benefit with adjuvant chemotherapy. [Therefore,] we directly propose use of ctDNA at the window defined as 2 to 10 weeks post-surgery [to inform] the decision of the adjuvant chemotherapy,” Nakamura advised.

Further analyses of the predictive ability of ctDNA status to inform the duration of adjuvant chemotherapy within this study are supported by these results and are ongoing. Moreover, the use of ctDNA-guided adjuvant approaches will be further evaluated in 2 randomized, phase 3 trials under the CIRCULATE-Japan platform.

“We are currently conducting 2 randomized phase 3 studies, VEGA and ALTAIR, for ctDNA- negative and -positive patients, respectively. With the results obtained from these randomized trials, we can further validate the utility of ctDNA-[based] MRD analysis,” Nakamura concluded.

Editor’s Note: Dr Nakamura reports honoraria from Chugai and Guardant Health AMEA; and research grants from Taiho, Chugai, Guardant Health, Genomedia, Daiichi Sankyo, Seagen, and Roche Diagnostics.

References

1. Nakamura Y, Kotani D, Saori M, et al. Circulating tumor (ct)DNA as a prognostic biomarker in patients (pts) with resected colorectal cancer (CRC): An updated 24 months (mos) disease free survival (DFS) analysis from GALAXY study (CIRCULATE-Japan). Ann Oncol. 2023;34(suppl 2):558MO. doi:10.1016/j.annonc.2023.09.1749
2. Kotani, D., Oki, E., Nakamura, Y. et al. Molecular residual disease and efficacy of adjuvant chemotherapy in patients with colorectal cancer. Nat Med. 2023;29(1):127-134. doi:10.1038/s41591-022-02115-4