Postoperative ctDNA Provides Prognostic Value for DFS Outcomes in CRC

This article was originally published by OncLive^®.

Circulating tumor DNA (ctDNA) status at the time of postoperative minimal residual disease (MRD) assessment was found to be a prognostic factor for disease-free survival (DFS) in patients with radically resected stage II to IV colorectal cancer (CRC), according to updated data from the observational GALAXY study, which is part of the CIRCULATE-Japan trial (UMIN000039205). Data were presented during the 2023 ASCO Breakthrough Meeting.¹

Jun Watanabe, MD, PhD

At a median follow-up of 16.3 months, patients who had negative postoperative ctDNA at 4 weeks (n = 1797) experienced an 18-month DFS rate of 93.9% (95% CI, 92.5%-95.0%) vs 51.6% (95% CI, 45.2%-57.6%) among patients with positive ctDNA at the same time point (n = 286; HR, 12.0; 95% CI, 9.1-15.0; P < .001). Ninety-six patients who were ctDNA negative experienced a DFS event compared with 130 who were ctDNA positive.¹

Additionally, among patients who underwent dynamic ctDNA assessment from weeks 4 to 12, 16 (14.3%) converted to negative (n = 112; HR, 3.5; 95% CI, 1.9-5.8; P < .001), 20 (46.5%) converted to positive (n = 43; HR, 14.5; 95% CI, 8.8-23.8; P < .001), and 78 (62.9%) remained positive (n = 124; HR, 25.4; 95% CI, 18.3-35.3; P < .001). All 3 groups of patients were more likely to have disease recurrence compared with those who had persistently negative ctDNA (n = 69/1529; 4.5%) in the same time span. The 18-month DFS rates in these subgroups were 82.2% (95% CI, 72.3%-88.9%), 47.4% (95% CI, 30.4%-62.7%), 33.8% (95% CI, 25.0%-42.8%), and 94.9% (95% CI, 93.5%-96.0%), respectively.¹

“Our study builds on existing evidence from the recently published, prospective GALAXY study, demonstrating the prognostic value of ctDNA analyzed in more than 2000 patients,” Jun Watanabe, MD, PhD, of the Gastroenterological Center at Yokohama City University Medical Center in Japan, said during the presentation. “Combining BRAF and microsatellite instability status with 4-week postoperative ctDNA status increased the predictive value of BRAF and microsatellite instability status.”

GALAXY is the observational arm of the CIRCULATE-Japan trial, monitoring the ctDNA status of patients with stage II to IV or recurrent CRC who are eligible for complete surgical resection. CIRCULATE-Japan also includes the phase 3 VEGA trial (jRCT1031200006), which is comparing postoperative surgery with capecitabine (Xeloda) plus oxaliplatin in patients with high‐risk stage II or low‐risk stage III CRC with negative ctDNA status 4 weeks after curative surgery in GALAXY, and the phase 3 ALTAIR trial (NCT04457297), comparingtrifluridine/tipiracil (Lonsurf) with placebo in patients with resected, ctDNA-positive CRC in GALAXY.²

In GALAXY, serial ctDNA status was assessed before surgery and at 1, 3, 6, 9, 12, 18, and 24 months after surgery. The primary end point was DFS. The data cutoff was November 10, 2022.¹

The median age in the overall population was 69 years and most patients were male (54%). Most patients had a left primary tumor site (75%), pathologic stage III disease (51%), and RAS wild-type status (57%); 34% of patients underwent adjuvant chemotherapy. Baseline patient characteristics were generally well balanced between the BRAF wild-type/MSS (n = 1707), BRAF V600E–mutant/MSS (n = 59), BRAF wild-type/MSI-H (n = 97), and BRAF V600E–mutant/MSI-H (n = 110) subgroups.¹

Additional results indicated that patients who had negative postoperative ctDNA at 4 weeks with BRAF wild-type/microsatellite instability–high (MSI-H; n = 88), BRAF V600E–mutant/MSI-H (n = 107), BRAF wild-type/microsatellite stable (MSS; n = 1450), and BRAF V600E–mutant/MSS disease (n = 50) experienced 18-month DFS rates of 100%, 100%, 93.3% (95% CI, 91.7%-94.6%), and 86.3% (95% CI, 71.8%-93.7%), respectively. Using patients with BRAF wild-type/MSS disease as the reference group, the hazard ratios were 0.09 (95% CI, 0.001-0.61; P = .006), 0.07 (95% CI, 0.001-0.50; P = .002), and 2.06 (95% CI, 0.84-4.23; P = .107), in the BRAF wild-type/MSI-H, BRAF V600E–mutant/MSI-H, and BRAF V600E–mutant/MSS groups, respectively.¹

Comparatively, those who had positive postoperative ctDNA at 4 weeks with BRAF wild-type/MSI-H (n = 9), BRAF V600E–mutant/MSI-H (n = 3), BRAF wild-type/MSS (n = 257), and BRAF V600E–mutant/MSS disease (n = 9) experienced 18-month DFS rates of 66.7% (95% CI, 28.2%-87.8%), 0%, 50.9% (95% CI, 44.1%-57.3%), and 33.3% (95% CI, 7.8%-62.3%), respectively. Using patients with BRAF wild-type/MSS disease as the reference group, the hazard ratios were 0.67 (95% CI, 0.21-2.20; P = .488), 7.54 (95% CI, 2.37-24.0; P < .001), and 2.33 (95% CI, 1.03-5.30; P = .043), in the BRAF wild-type/MSI-H, BRAF V600E–mutant/MSI-H, and BRAF V600E–mutant/MSS groups, respectively.¹

“[Patients with] BRAF V600E-mutant/MSS tumors had a significantly worse prognosis compared with patients with BRAF wild-type/MSS tumors,” Watanabe said. “Of note, patients with BRAF V600E–mutant/MSI-H tumors had a significantly worse prognosis compared with those with BRAF wild-type/MSI-H tumors. This is a different trend from the ctDNA-negative patients.”

“Our multivariate analysis for DFS indicated that postoperative ctDNA status at 4 weeks, BRAF mutational status, and MSI status were significantly associated with DFS,” Wantanabe stated in conclusion. The hazard ratios for DFS for patients who were ctDNA positive at 4 weeks, BRAF V600E–mutant, and MSI-H were 11.68 (95% CI, 8.61-15.85; P < .001), 2.05 (95% CI, 1.05-3.99; P = .035), and 0.26 (95% CI, 0.10-0.62; P = .003), respectively.¹

References

1. Watanabe J, Oki E, Kotani D, et al. Postoperative circulating tumor DNA-based molecular residual disease in patients with BRAF V600E and MSI-H colorectal cancer: updated results from GALAXY study in the CIRCULATE-Japan. JCO Global Oncology. 2023;9(suppl 1):32. doi:10.1200/GO.2023.9.Supplement_1.32
2. Taniguchi H, Nakamura Y, Kotani D, et al. CIRCULATE-Japan: circulating tumor DNA-guided adaptive platform trials to refine adjuvant therapy for colorectal cancer. Cancer Sci. 2021;112(7):2915-2920. doi:10.1111/cas.14926