NEW YORK (GenomeWeb) – Investigators published new data this week from a clinical study of Inivata's lung cancer liquid biopsy assay demonstrating high concordance with tissue profiling, when possible, plus an improved ability to detect alterations compared to tissue.
The report, which appeared late yesterday in JCO Precision Oncology, includes an analysis of test results from two prospective, multicenter studies, which involved a total of 264 non-small cell lung cancer patients and provided the backbone for the company's Medicare coverage win earlier this year.
The study's authors wrote that their analyses demonstrate the potential of the company's test to improve the delivery of comprehensive tumor genomic profiling and resulting personalized treatment.
Making good on the promise of precision medicine has been more of a challenge than some may have hoped, as studies have struggled to translate broad profiling for genomic alterations into actual benefit or improved outcomes.
Liquid biopsy firms have hoped to show that their blood-based tests can improve this, by increasing the number of individuals who are successfully sequenced and reducing the time it takes to obtain genetic results, identify targets, and get patients onto a personalized drug. As such, data in that vein has recently been highlighted not just by Inivata, but also competitors like Guardant Health and Resolution Bioscience.
The Inivata study, which expands on a presentation at the World Conference on Lung Cancer last September, speaks both to that question of efficiency and to lingering uncertainty about the potential for liquid biopsy tests to yield false positives.
Various companies are hoping that by providing evidence that blood-borne mutations reliably reflect the truth of the tumor they can persuade clinicians to embrace liquid biopsy not just as an adjunct to tissue testing, but potentially in a "blood-first" paradigm: where blood tests are ordered first and tissue sequencing becomes more of a backup or reflex for those with negative blood-based results.
"Clinicians quite rightly want the data and evidence," Inivata CEO Clive Morris said on Thursday. "They were hoping these tests could serve these patients that can't have tissue or don’t have tissue … and maybe even in the future could replace tissue tests. But you don't do that in the absence of data."
In the study published this week, the authors wrote that Inivata's InVisionFirst-Lung demonstrated "excellent concordance" with tissue profiling, with a high level of sensitivity and specificity. Notably, the liquid biopsy test was able to detect 26 percent more actionable alterations than standard-of-care tissue testing, at least in part because it worked more often than tissue testing.
When all the data came in, at least some tissue sequencing was available for 178 of the patients in the two cohorts. But only 95 of these cases had comprehensive tissue sequencing — covering all of what the study defined as crucial actionable genes. The other 86 patients had results only from InVisionFirst-Lung.
Across the total enrolled population, investigators found that InVisionFirst-Lung detected "actionable alterations" in 48 individuals, compared to only 38 with tissue testing, representing a 26 percent improvement.
In addition, about half of the blood-borne actionable alterations were in patients who had no tissue sequencing, either because they had insufficient material available or no tissue at all.
"The detection of an increased percentage of patients with actionable mutations using liquid biopsy demonstrates how this technology can … [ensure] more NSCLC patients can be treated with the appropriate therapy, improving their chances of durable response and survival," Ramaswamy Govindan, professor of medicine at Washington University School of Medicine and the study's senior author, said in a statement.
Considering tissue as the reference, or gold standard — and looking just at what the investigators defined as the eight most clinically relevant genes for patient management in advanced NSCLC — the investigators calculated the sensitivity of Inivata's test as about 74 percent with 99.8 percent specificity, 97.8 percent positive predictive value, and 97.1 percent negative predictive value.
Importantly, the results matched what others, like Guardant Health, have recently seen regarding false positives. For patients with tissue results available, the tissue recapitulated every actionable driver mutation detected by InVision-First Lung.
Across the full panel, positive predictive value compared to tissue was lower, at 83 percent. The study authors wrote that this difference — representing 32 nonactionable variants detected in plasma but not in tissue — included six patients for whom it looked like the variant was there in tissue, but below thresholds required for calling. Sixteen of the remaining 26 blood-only calls were TP53 variants that "may be subclonal events that only occur at low levels or may be completely absent from the biopsy site," the group wrote.
Possible false-negatives are another story. Inivata's analysis, like other reports from competing tests, showed that there are mutations in tumor tissue, including "actionable alterations" that do not appear in blood-based test results. These included ALK and ROS rearrangements, and BRAF and EGFR alterations, among others.
Although many questions remain, some research has begun to try to clarify when and why tissue-positive, blood-negatives occur. A comparison by UK researchers published last week, for example, found that tumor-only mismatches can be due to mutations being present in blood, but at levels below a variant allele cutoff point, or to biological factors like tumor evolution over time.
Morris said that having more studies of this type in the literature will be important as clinicians work to weigh competing technologies. However, direct comparisons of tests from companies like Inivata, Guardant Health, Resolution Bioscience, Personal Genome Diagnostics, Foundation Medicine and others remain limited, or anonymous.
He declined to detail what kind of projections Inivata may have for its testing volume or market share, now that it has received Medicare coverage based on the results published. The company announced at the end of March that it had closed a Series B financing for £39.8 million ($52.6 million), which is supporting its commercial push for InVision, and fueling R&D in the disease monitoring and early detection applications that are drawing increasing attention in the liquid biopsy field.
The design of Inivata's clinical validation did not allow for physicians to act upon the liquid biopsy results, so the current study does not speak to the question of clinical utility — whether testing improves patient outcomes.
But Morris highlighted that Inivata has completed separate research that does speak to utility, including a 2017 report on 48 patients treated with the EGFR drug osimertinib (AstraZeneca's Tagrisso) based on liquid biopsy detection of T790m mutations. Authors in that study concluded that both the prevalence of T790m in the cohort, and the pattern of patient responses matched with what would be expected based on prior studies of tissue.
"We are just starting our commercial launch," Morris said. "But as there is more and more volume of testing … we would expect to publish more on outcomes."
However, he added, "there is no reason to believe that those are going to be anything other than what we expect based on the drug label … because based on the data [in this study] if you look at the actionable alterations … our specificity was 100 percent. If we said it was true, wherever there was tissue to corroborate it, it was true in tissue."