NEW YORK – Presentations at the American Society of Clinical Oncology's virtual annual meeting this week signal continued growth in commercial liquid biopsy tests for tracking minimal residual disease, as well as clinical applications emerging for these assays.
Companies with competing products in the clinic are amassing more data for their respective offerings. In parallel, they and others continue to debut new technologies and test new use cases.
Among notable presentations at the meeting, both Natera and Guardant Health shared new data on their commercial MRD products. The two firms are currently in a legal dispute after Guardant filed a lawsuit last month against Natera alleging false advertising, unfair competition, and unlawful trade practices related to Guardant's MRD product Guardant Reveal. Natera responded with its own suit against Guardant, accusing its competitor of making false and misleading claims.
Under the shadow of this dispute, Natera shared notable new data at the ASCO meeting from a trial called CIRCULATE-Japan, one of the largest prospective MRD studies that has been initiated thus far.
In an initial analysis from the first 400 early-stage and relapsed colorectal cancer patients recruited to the trial, Natera's tumor-informed Signatera assay had a pre-surgical detection rate of more than 94 percent and a post-surgical relapse detection rate of more than 91 percent with longitudinal sampling. Of 12 patients diagnosed with recurrences by the January data cutoff for the presentation, 10 had positive ctDNA on their first test four weeks after surgery. Another patient was ctDNA negative at four weeks but ctDNA positive at a second 12-week test.
Natera highlighted that these detection rates compare favorably to longitudinal results from tissue-naive approaches. Guardant, for example, has reported a 91 percent detection rate with ongoing sampling. The interim analysis also showed a failure rate for Signatera of less than 3 percent.
In a separate poster, academic collaborators from Aarhus University Hospital in Denmark reported on a study serially monitoring 168 stage III CRC patients for 36 months as they underwent post-surgery adjuvant therapy, surveillance, or both.
According to the authors, by using Signatera to measure ctDNA levels, they could assess the growth rates of metastases in different parts of the body at the same time. In addition, patients could be classified as having either a slow- or fast-growing tumor based on the rate of increase in ctDNA levels. Those with fast-growing tumors had significantly poorer overall survival compared to those with slow-growing tumors.
This prognostic ability represents a potential expanded use case for Signatera, beyond the post-surgical MRD and recurrence monitoring for which it is much more thoroughly validated.
Natera also continues to pursue expanded utility in other, non-CRC tumor types. A third poster at ASCO explored the use of Signatera in a cohort of 113 patients with hepatocellular carcinoma, cholangiocarcinoma, or gallbladder cancer.
Considering the current and emerging adjuvant options for these cancer types, molecular MRD assessment could be of value in identifying patients at risk of recurrence or metastasis, especially in the context of new adjuvant clinical trials, the authors wrote.
Like Natera, Guardant also has ambitions for its Reveal test beyond CRC. In a poster at ASCO, company investigators described the expansion of this assay to detect MRD across multiple tumor types.
The researchers tested Reveal using 175 pre-treatment clinical samples from patients with early-stage, non-small cell lung cancer and bladder cancer, as well as 171 self-declared healthy donors. According to the authors, Reveal's pre-treatment detection sensitivity in NSCLC ranged from about 56 percent in stage I tumors to 83 percent in stage III tumors. Sensitivity for bladder cancer was between 36 percent and 69 percent.
According to Guardant, similar work is ongoing in larger cohorts and for other tumor types.
Lung cancer, in particular, seems to be a promising emerging target for genomic MRD tests. In a poster discussion session during the conference, investigators from Cancer Research UK said that although conclusive evidence for the utility of ctDNA in early-stage non-small cell lung cancer is still lacking, detection of residual disease and recurrence in this population could greatly benefit patients.
The group shared data from their LUCID study, where they analyzed 363 plasma samples from 88 patients with stage I to III NSCLC, using another commercial MRD assay, Inivata's RaDaR, which like Signatera is a tumor-informed personalized assay.
For 62 patients, the investigators collected plasma at a landmark time point, between two weeks and four months after initial treatment. RaDaR detection of ctDNA at that landmark was strongly predictive of clinical disease relapse.
All 11 cases with ctDNA detected at landmark had disease progression at a median of 121 days after detection, including eight patients who relapsed within 300 days of treatment. Across 27 patients whose disease progressed during the study, ctDNA was detected at any time point post-treatment in 17 cases, with a median lead time of 203 days and up to 741 days prior to clinical progression.
According to the investigators, the results demonstrate that liquid biopsies can offer the opportunity for earlier therapeutic intervention in NSCLC by reliably detecting recurrence at a preclinical stage, long before clinical progression.
Guardant also presented prospective observational data from a study using its plasma-only MRD assay, Reveal, in colorectal cancer — not in the more established localized disease setting, but in patients with limited metastasis.
Investigators analyzed blood samples from 46 patients before they underwent a curative-intent procedure, three weeks after their procedure, and at various additional follow-up time points.
As of Jan. 1, the recurrence rate in the cohort was 60 percent, with a median follow-up of 50 weeks. Twenty-four patients had positive ctDNA test results post-treatment, and of those, 23 had recurred by the cutoff point. On average, Reveal picked up emerging tumor DNA 28 weeks before radiographic recurrence, the group wrote.
For the two patients with post-treatment positive results but no recurrence, one received adjuvant chemotherapy and subsequently cleared the ctDNA. Among the 22 patients with no positive Reveal tests, only six recurred during the follow-up period. This represents a negative predictive value of 80 percent, the researchers reported.
Overall, the group calculated a sensitivity of 79 percent and a specificity of 84 percent for the test in this setting.
Assessment of oligometastatic patients could be a new area for Guardant, Natera, and competitors' MRD tests. Independent groups have also expressed excitement about using ctDNA MRD assays in this setting. A recent study led by investigators at Washington University in St. Louis, for example, explored recurrence prediction in this same population using a Roche ctDNA assay.
As Medicare contractors have begun to issue coverage determinations for MRD assays in solid tumors, they have kept the coverage criteria limited to patient populations where oncologists can act on post-treatment MRD results — in other words, where adjuvant therapies or early interventions of some other kind are available for patients who test positive for MRD or for emerging recurrence.
As a result, the current local coverage determination for Signatera is limited to colorectal cancer, though the firm is hopeful an update currently in draft form will expand the use of the test to other areas deemed sufficiently actionable.
In the meantime, the availability of these technologies in the research space is allowing new investigations to progress, potentially establishing novel MRD-based treatment paradigms for a wider group of tumors and time points in the near future.
Another exciting potential MRD test application highlighted at the meeting was in ovarian cancer. Researchers from a group of French institutions presented data from the CIDOC study, a multicenter prospective trial exploring ctDNA as an early marker of disease relapse after first-line treatment for epithelial ovarian cancer. The group has been using a method similar to Natera's, using tumor sequencing to design patient-specific monitoring panels, which are then performed using droplet digital PCR.
Successful surgical resection remains one of the main prognostic factors for ovarian cancer, which is the most deadly gynecological tumor type. Patients undergo mandatory laparoscopic exploration to assess their chances for surgical resection at diagnosis or after neoadjuvant chemotherapy, but for tumors that don't look resectable on first glance, there is no good biomarker to predict if they will become so after chemo.
In their results from the CIDOC study, investigators reported that in a cohort of 47 patients, ctDNA allele frequency was correlated with peritoneal tumor spread. In addition, ctDNA clearance after preoperative chemotherapy was associated with successful surgical outcomes.
Based on these initial results in CIDOC, the researchers wrote that they believe ctDNA could serve as a tool to predict which patients' tumors don't look removable on an initial laparoscopy following neoadjuvant chemo, or whether their disease is responding to treatment and can likely be removed with surgery.