NEW YORK – Circulating tumor DNA (ctDNA) status accurately predicted the risk of colorectal cancer (CRC) recurrence and identified patients most likely to benefit from adjuvant therapy in an interim analysis of data from the GALAXY arm of Natera's ongoing CIRCULATE-Japan study.
Additionally, the data showed that ctDNA could be successfully used to direct appropriate chemotherapy to patients, as well as accurately monitor therapy response.
The data was published in Nature Medicine and presented as a poster at European Society for Medical Oncology (ESMO) Congress taking place this week in Barcelona, Spain.
The interim data came from 2,240 individuals with stage II to IV or relapsed CRC following curative intent surgery. Natera's Signatera ctDNA-based minimal residual disease (MRD) test was used to evaluate ctDNA over a median period of 23 months.
Data from 2,109 patients was analyzed during the MRD window, defined as a period of two to 10 weeks after surgery and before the start of any adjuvant therapy. The potential clinical applications of analysis in this window include prognostication and adjuvant treatment decision-making.
Nearly 16 percent of these patients were Signatera-positive, of whom approximately 78 percent experienced disease recurrence, demonstrating an overall 24-month disease-free survival (DFS) of roughly 20 percent and 36-month DFS of nearly 17 percent. In contrast, only approximately 13 percent of the remaining Signatera-negative patients experienced recurrence. These patients showed 24- and 36-month DFS rates of around 85 percent and 84 percent, respectively.
This association between ctDNA positivity and a significantly higher risk of recurrence was seen across all cancer stages.
Similarly, testing Signatera-positive was also associated with lower overall survival (OS) rates, compared to those who tested negative.
Signatera-positive participants showed 24- and 36-month OS rates of approximately 84 percent and 72 percent, respectively, while Signatera-negative patients showed OS rates of nearly 99 percent and 96 percent for the same time points.
The researchers observed similar results in participants analyzed within the surveillance window, defined as the time from 4 weeks after ACT, or the end of the MRD window if the patient had no ACT, until the last follow-up or clinical recurrence.
Here, 1,791 and 1,794 patients were included in the DFS and OS analyses, respectively. Compared to patients who were serially Signatera negative, patients with Signatera positivity at any time point were approximately 34 times more likely to experience disease recurrence.
Approximately 9 percent of Signatera-negative patients in the surveillance window showed DFS at 24 months versus 93 percent of the Signatera-positive patients. OS rates were approximately 83 percent versus 99 percent, respectively, for that same time point.
While other factors such as the presence of BRAF V600E mutations, RAS mutations, and lymph node positivity were also prognostic of poor DFS and OS, a multivariate analysis determined that ctDNA status was the strongest factor.
These results suggest that observation alone may be sufficient for positive outcomes among MRD-negative CRC patients, whereas adjuvant chemotherapy may benefit MRD-positive individuals.
"This is the first time that a prospective study has read out an overall survival signal in colorectal cancer associated with ctDNA, especially a study of this size," said Alexey Aleshin, Natera's chief medical officer.
Jeanne Tie, a medical oncologist with Australia's Peter MacCallum Cancer Centre, largely agreed.
"This is the largest observational study in ctDNA, and the authors should be commended for this impressive effort," she said in an email.
The study's findings, Tie said, validate observations from previous smaller studies and offers reassurance, Tie said, "that spontaneous ctDNA clearance is a rare phenomenon, providing clinicians with more confidence that patients with ctDNA-positive disease should receive further chemotherapy in an attempt to eradicate MRD."
Tie also noted that the results shown in the GALAXY study also agree with those of the DYNAMIC clinical study, of which Tie is the primary investigator. Past published results of that study had demonstrated the non-inferiority of ctDNA-guided treatment, with a clear de-escalation benefit. The use of adjuvant chemotherapy in the trial's ctDNA-guided arm was nearly halved compared to the standard treatment arm, without compromising average recurrence-free survival.
Aparna Parikh, associate professor of medicine at Harvard Medical School, also praised the study, saying in an email, that it demonstrates the "tremendous prognostic ability" of ctDNA across all stages of CRC.
"[It] also continues to provide confidence that ctDNA may be predictive of chemotherapy benefit, showing benefit of adjuvant chemotherapy in the ctDNA positive patients without a clear benefit in the negative patients," she said.
Aleshin expressed hope that the body of evidence supporting the uses of ctDNA would eventually also extend to drug approvals. Overall survival is a key metric in assessing a drug's value, he said, but it can take very long to gather that kind of data.
Given a sufficiently strong and reliable correlation with OS, he said, clinicians might be able to evaluate a patient's ctDNA in the span of a few months and potentially use that to improve existing immunotherapies and find new therapies for early-stage colon cancer.
"How amazing would it be if you're developing a new drug [and] you don't have to wait even three years for DFS to mature," Aleshin said.
Guidance already exists for the use of ctDNA as a surrogate endpoint in clinical trials, Aleshin explained, but none yet covers its use as an endpoint in drug approval trials.
"The hope is to strengthen the surrogate endpoint [so] that it could be used to actually approve new therapies," he said.
While optimistic of the possible uses of ctDNA, Tie cautioned that many questions regarding the proper use and limits of this biomarker remain open, such as the optimal timing for assessing ctDNA clearance after chemotherapy and whether multiple negative tests during surveillance are required.
"I would be keen to know the impact of adjuvant chemotherapy on ctDNA clearance across the various disease stages and molecular subgroups, e.g. KRAS mutant, HER-2 amplified, [and] BRAF mutant," Tie said.
Although reassuring to see that ctDNA-negative patients appear not to benefit from adjuvant chemotherapy, she continued, observational studies such as GALAXY can be limited by inherent selection biases.
"Ultimately," she said, "randomized trials will be required to demonstrate that de-escalation or omitting chemotherapy is not detrimental to recurrence outcome."
Natera is exploring that question to some degree in a randomized protocol that is part of the CIRCULATE-Japan study. That protocol, called Vega, involves randomizing ctDNA-negative patients one-to-one to receive either adjuvant chemotherapy or observation only.
"This data makes us more confident that the [Vega] study will read out in the fashion that we were hoping it would," Aleshin said, "which is to really definitively show, in a randomized controlled fashion, that adjuvant therapy has no substantial benefit in Signatera-negative patients."
Natera expects to be able to read out data from that study in the next year or two, Aleshin said.
In the meantime, the company expects the data from this study to be of considerable use to the numerous academics and clinicians who are also researching the diagnostic and prognostic potential of ctDNA in both research and clinical settings.
Particularly with respect to clinicians, Aleshin said, "the big question we've had is, is there an OS signal associated with Signatera? I think really for the first time, in a peer-reviewed, large prospective study, we can say the answer is yes."