NEW YORK (GenomeWeb) – Investigators from MD Anderson plan to present a study at the upcoming annual meeting of the American Association for Cancer Research that adds to growing evidence for the value of comprehensive liquid biopsy tests — in this case Guardant Health's Guardant360 assay — in the care of certain cancer patients.
Investigators reported in an abstract to be highlighted at the meeting that using Guardant's test prospectively enabled them to pick up more cases of actionable mutations, more rapidly, than did tissue biomarker analyses that were ordered for the same patients with non-small cell lung cancer.
The researchers are also following participants to track outcomes, something that is central to assessing things like medical necessity and cost effectiveness. But the results thus far do not yet speak directly to the impact of the test in that sense, investigators said this week.
Guardant and MD Anderson first announced in 2017 that they had partnered to prospectively study the clinical impact of the company's test. The report being shared at AACR is from the partners' NILE (Noninvasive vs. Invasive Lung Evaluation) study, which Guardant Health President AmirAli Talasaz named in the firm's Q4 earnings call as as a critical piece of its strategy to establish a "blood-first" paradigm in NSCLC and to seek US Food and Drug Administration approval and reimbursement for the assay across tumor types.
MD Anderson investigators partnered in the study with several other institutions, using Guardant360 to profile (as of the abstract release) a total of 282 patients with NSCLC — a tumor type that has become a proving ground for blood-based biomarker testing much like it has for precision medicine in general.
"When choosing therapy for patients with NSCLC, it is vital that we know which patients have gene mutations that often respond to molecular therapies. The response rates can be up to 30 percent higher than with chemotherapy or immunotherapy," MD Anderson professor of thoracic oncology Vassiliki Papadimitrakopoulou, the lead author of the study, said in a statement.
Amongst other findings, Papadimitrakopoulou and colleagues are reporting at AACR that while standard tissue testing detected one or more of seven defined "predictive biomarkers" in 60 patients, Guardant360 found biomarkers in 89 patients.
The results don't represent a head-to-head comparison of tissue and blood in each patient. Some of the discrepancy between mutation detection in blood versus tissue is a result of patients in the study either not having any tissue for genomic analysis or having only limited genotyping compared to the Guardant panel. This might seem like a caveat to the findings, but during a media call discussing the abstract on Wednesday, Papadimitrakopoulou said that the results are powerful precisely because they reflect that clinical reality.
Much of the promise of liquid biopsy, in lung cancer at least, is that it offers a way to analyze cancer genetics in patients for whom it otherwise wouldn't be possible, so proving its utility is more a matter of demonstrating this, than it is of proving a perfect concordance.
Importantly, the group's analysis revealed that none of the most actionable mutations detected in blood were false positives. For each patient in whom Guardant360 identified a target of an FDA-approved therapy (EGFR, ALK, ROS1, and BRAF), tissue also showed the same alteration.
While the team plans to describe additional analyses of negative predictive value at the AACR meeting, those calculations aren't ready yet, Papadimitrakopoulou added.
Although Guardant believes these mutation detection "non-inferiority" results will be important as it works to increase the use of its test in guiding care for NSCLC patients, some of the early anxiety around the blood-tissue discordance has already relaxed in the face of mounting evidence that druggable mutations are readily detected by liquid biopsy tests, and by regulatory recognition that test users must understand that negative blood results don't preclude an actionable alteration being detected in tissue.
Because of that, focus has shifted to whether tests like Guardant 360 can prove that they have clinical utility — that they truly improve patient outcomes in a way that merits their cost. One aspect of this is turnaround time, which test-makers have begun to focus more on.
The company is not the first to try to make this argument to the field. Biocept, a company that has been trying to build a business providing single-gene liquid biopsy tests (as opposed to a larger NGS panel) has also pursued marketing campaigns encouraging clinicians to order a liquid test alongside or ahead of tissue testing as a standard practice.
In the MD Anderson study being shared at AACR, investigators reported that the Guardant test had a median turnaround time from order to final results of just nine days, compared to 15 days on average for tissue analysis.
Papadimitrakopoulou mentioned in reporting on the results how scantily existing recommendations for tissue-based testing in NSCLC have actually been implemented thus far. "Guideline-recommended testing is normally completed in only 8 percent of patients with NSCLC," she said.
Such challenges to successful implementation of genomics-guided, or precision cancer care, were also recently illustrated in a grim report from a retrospective analysis of health records in Flatiron Health's database, which concluded that despite growing enthusiasm for genomic testing, individuals whose tumors are analyzed with comprehensive tests don't seem to be surviving longer or better compared to those who get older, gene-specific tests — at least not in a community setting.
Guardant and others are hoping to make a case that by increasing the number of patients analyzed successfully, and doing it more quickly, liquid biopsy can help solve the bottlenecks that prevent the promise of precision medicine from being realized.
The new MD Anderson results join other data Guardant has begun to amass that speak to clinical impact. A study published by UPenn researchers in JAMA Oncology last fall, for example, also reported that Guardant360 identified targeted mutations in more patients than did tissue sequencing, as well as that patients treated on the basis of Guardant360 results responded comparably to those treated based on tissue mutations.
Other companies have been working in the same vein. Resolution Biosciences saw a prospective analysis led by US and Australia researchers published in November, which showed that its own test was able to rapidly detect actionable mutations in a majority of NSCLC cases, including individuals whose tissue couldn't be successfully tested.
One thing the newest MD Anderson results do not speak to, but which Guardant recognizes it does need to demonstrate, is that the improvements in mutation detection (and the potential for more rapid initiation of targeted treatment that are suggested by a reduced turnaround time) actually improve patient outcomes.
Authors of the abstract did say that outcome analysis is a part of the current NILE study, but is not something they can report from the cohort yet.
Hints from other analyses, including the UPenn study of Guardant's test and data from others like Inivata, have supported that patients treated on the basis of liquid biopsy sequencing seem to benefit from molecularly targeted treatment at the same rates as seen in initial, tissue-based pivotal clinical trials for these drugs.