Data Support ctDNA as a Prognostic Marker for Recurrence/Benefit in Colorectal Cancer

An expert from Natera indicates that post-operative ctDNA level may have strong prognostic and predictive value in patients with resectable colorectal cancer.

Circulating tumor DNA (ctDNA) appeared to be prognostic of recurrence in stage II to IV resectable colorectal cancer (CRC) and predicted which patients may benefit best from adjuvant chemotherapy, according to findings from the observational GALAXY arm in the CIRCULATE-Japan study.

With a median follow-up of 16.74 months (range, 0.49-24.83) in the GALAXY cohort, 61.4% of patients with ctDNA positivity detected 4 weeks after surgery (n = 187) had disease recurrence; moreover, 9.5% of patients with ctDNA negativity had recurrent disease ( Hazard ratio [HR, 10.0; 95% CI, 7.7-14.0; P <.0001). This translated to an 18-month disease-free survival (DFS) rate of 38.4% (95% CI, 31.4%-45.5%) vs 90.5% (95% CI, 88.3%-92.3%) among the ctDNA-positive and ctDNA-negative groups, respectively.

ctDNA positivity at 4 weeks was also determined to be the most significant prognostic factor linked to disease recurrence risk among patients with stage II or III CRC (HR, 10.82; 95% CI, 7.07-16.6; P <.001). Additionally, ctDNA detection at 4 weeks identified patients with stage II or III disease who significantly achieved benefit from adjuvant chemotherapy (HR, 6.59; 95% CI, 3.53-12.3; P <.001).

“I hope that our colleagues internationally realize that ctDNA…is a strong prognostic factor,” Minetta Liu, MD, chief medical officer of oncology at Natera, said in an interview with CancerNetwork®. “Prognostic factors are helpful in assessing risk of recurrence, but we're really at the cusp of transforming how we want to take care of our patients.”

The prospectively conducted GALAXY study, part of the CIRCULATE-Japan project, was designed to evaluate ctDNA status for patients with stage II to IV CRC who were eligible to undergo complete surgical resection. Investigators extracted cell-free DNA from patient plasma at a given time point to detect ctDNA. Plasma samples with at least 2 of 16 tumor-specific variants detected above a pre-defined threshold were identified as being ctDNA positive.

The primary end point was DFS, defined as the time between surgery and diagnosis of relapse or death. The secondary end point was the ctDNA clearance analysis comparing the cumulative incidence fusion differences between adjuvant chemotherapy and observation groups.

A total of 18.0% (n = 187/1039) of 1039 patients included in the ctDNA analysis were ctDNA-positive 4 weeks following surgery, and 82.0% (n = 852/1039) were ctDNA negative. When evaluating ctDNA dynamics, 45 patients recurred at 12 weeks and 157 had no ctDNA status at 12 weeks and were excluded from the study.

The median patient age in the ctDNA-negative and ctDNA-positive groups, respectively, was 69 years (range, 25-93) and 67 years (range, 39-88). In both respective groups, most patients were 70 years or younger (56.1% vs 61.5%; P = .18), male (50.8% vs 62.6%; P = .003), and had an ECOG performance status of 0 (87.3% vs 88.2%; P = .33).

“This [trial] was performed only in Japan, but the demographics of that patient population are roughly comparable to the U.S population,” Liu explained. “We do have the BESPOKE CRC study [NCT04264702]…an observational registry study in the U.S. Those findings, again, which are following usual care and physician's choice, will help corroborate the observational findings from GALAXY. And then we have parallel, prospective, randomized phase 3 trials for clinical utility in both populations: CIRCULATE-Japan and CIRCULATE-US.”

In 838 patients with available ctDNA status at the 4-week and 12-week mark following surgery, 42.2% (n = 354/838) underwent treatment with adjuvant chemotherapy. Of these patients, 10% retained ctDNA positivity, 78.8% retained ctDNA negativity, 3.8% changed from ctDNA negativity to positivity, and 7.4% converted from positivity to negativity.

Investigators observed a significantly higher risk of disease recurrence among patients who converted to ctDNA positivity with an 18-month DFS rate of 33.8% (95% CI, 18.1%-50.2%; HR, 14.0; 95% CI, 8.5-24.0; P <.001) or in those with persistent ctDNA positivity with an 18-month DFS rate of 22.9% (95% CI, 14.3%-32.7%; HR, 21.0; 95% CI, 14.0-31.0).

Among 187 patients who were ctDNA positive at 4 weeks, 92 received adjuvant chemotherapy and 90 were put into the observation arm with no adjuvant chemotherapy. The clearance analysis indicated that adjuvant chemotherapy was associated with a higher estimated cumulative incidence of ctDNA clearance in 68.48% (n = 63/92) of individuals 24 weeks after surgery compared with 12.2% (n = 11/90) of those in the observation arm (HR, 8.50; 95% CI, 4.2-17.3; P <.0001).

The GALAXY arm of CIRCULATE-Japan also served to screen patients for one of 2 ctDNA-guided trials. Patients with postoperative ctDNA positivity were included in the phase 3 ALTAIR trial (NCT04457297) evaluating the efficacy and safety of treatment with trifluridine/tipiracil compared with standard-of-care chemotherapy to improve outcomes for patients with ctDNA-positive CRC. Patients with postoperative ctDNA negativity were included in the phase 3 VEGA study evaluating the noninferiority of observation treatment vs adjuvant capecitabine/oxaliplatin.

In the United States, investigators are evaluating postoperative ctDNA dynamics in early-stage CRC in the CIRCULATE-US study (NCT05174169).

“In order to support guidelines and true clinical adoption for [ctDNA testing] and hopefully de-escalate chemotherapy in patients, we need to demonstrate this in clinical utility trials. Those are ongoing, and support for the trials both in Japan and the United States where these studies are ongoing; it is really what's going to push the needle forward for patients and our providers,” Liu concluded.

Reference

Kotani D, Oki E, Nakamura Y, et al. Molecular residual disease and efficacy of adjuvant chemotherapy in patients with clinical stage II-IV resectable colorectal cancer. Nature Medicine. Published online January 16, 2023. doi:10.1038/s41591-022-02115-4