This story has been updated to clarify study results that relate to the test's false-positive rate.
--
NEW YORK – Marking what could be a significant shift in the scope of precision oncology, cancer early detection firm Grail launched on Friday the field's first broadly available multi-cancer early detection blood test, Galleri.
The announcement was timed to a presentation of the first data from an ongoing prospective interventional study of Galleri, called PATHFINDER, at the American Society of Clinical Oncology's virtual annual meeting.
As of Friday morning, Galleri can now be used clinically, by prescription, to screen individuals for incipient cancer. According to Grail, the test is intended for use in those with an elevated risk of cancer, such as adults aged 50 or older, and as a complement to existing single cancer screening tests.
The launch is, at least initially, focused on what Grail has called "partner health systems, medical practices, and self-insured employers."
Thus far, only one partner health system has been announced — Renton, Washington-based Providence, which is initially providing the test in California, Washington, and Oregon but could eventually make it available across its seven-state footprint that includes more than 50 hospitals, nearly 1,100 health clinics, and 5 million patients.
The company has said though, that individual medical practices can also order the test for clients if desired, presumably with patients paying out of pocket.
To help facilitate access and provide support for doctors in seeking follow-up for patients with positive results, the company announced earlier this year that it will be working with Quest to provide blood collection, as well as with a company called AccessHope to help facilitate positive-test follow-up.
Based on modeling data, Grail has previously estimated that broad implementation of multi-cancer early detection could reduce late-stage (stage III and IV) cancer diagnoses by more than half in the age-50 and older US population. Such a shift could translate to an overall reduction of five-year cancer deaths by 26 percent.
But this would require that the test's real-world performance match what the company has seen in its development and validation studies, something that, until today, the field could only speculate on.
Presenting the PATHFINDER interim results in a poster discussion session at the ASCO meeting, Tomasz Beer, deputy director at the Oregon Health and Science University Knight Cancer Institute, said that the new data do confirm some of Grail's earlier retrospective findings — namely that Galleri detects a wide variety of tumors in patients who don't yet have any symptoms or other signs of cancer.
Most importantly, he added, the results confirm that Galleri detects not only cancers that might be picked up by other screening programs but also tumors that currently go unscreened because of a lack of effective tests.
The trial was designed to assess Galleri's performance in a real-world clinical care setting — testing an asymptomatic cohort undergoing typical medical care and following individuals who received a positive or "signal detected" test result to track their subsequent care pathways, outcomes, and the time required to achieve an ultimate diagnostic resolution.
Overall, the study successfully tested 6,629 individuals aged 50 years or older. Although this is a group at elevated risk for cancer, none of the recruited participants had signs or suspicion of having a cancerous tumor. The group was split between individuals with additional risk factors other than age — including a history of smoking, prior cancer treatment, or genetic cancer predisposition — and individuals without any known risk factors.
Beer reported at ASCO that the version of Galleri used during the trial (the clinically launched test is an updated version) yielded a positive result in 92 cases, or 1.4 percent of the cohort. The positive result rate was a little higher for the subset of participants with known elevated risk (1.5 percent) than it was for those without (1.2 percent).
As of the interim analysis, which reflects a data lock as of March 2021, 27 positive results hadn't yet been confirmed as either true positives or false positives based on clinical follow-up. Among the 65 that were "diagnostically resolved," 29 were deemed true positives. Most were de novo, but there were also some diagnosed as recurrences of previously diagnosed cancers.
Based on an initial diagnostic workup, 36 percent of the positive results were deemed false positives, but longer-term tracking through the trial's full 12-month follow-up period could convert some of those to true positives, researchers cautioned. "With longer follow‑up, additional participants may reach diagnostic resolution and some currently categorized as not having cancer (false positives) may be found to have cancer," Beer and his coauthors wrote.
Encouragingly for Grail, the positive predictive value for Galleri in the PATHFINDER analysis came out to about 45 percent for those with a diagnostic resolution, which it said is consistent with findings from the earlier Circulating Cell-free Genome Atlas, or CCGA, study, which assessed the test in a much larger cohort of known cancer cases and controls.
Among the true positives in PATHFINDER confirmed by further diagnostic workup, 39 percent were localized stage I and II tumors, and more than half were stage III or below.
In addition to reading out a positive or negative signal of a person having an unknown cancer, Galleri also predicts where in the body this cancer originated, giving clinicians a hint of where to focus their attention as they then act on the results.
In the study, Galleri predicted a correct origin as one of its top two choices more than 96 percent of the time.
"Finding cancer early, when treatment is more likely to be successful, is one of the most significant opportunities we have to reduce the burden of cancer,” Grail's chief medical officer Joshua Ofman said in a statement announcing the launch this morning.
"These data suggest that, if used at scale alongside existing screening tests, the Galleri test could have a profound impact on how cancer is detected and, ultimately, on public health," he added.
As Galleri and its expected competitors have emerged over the last few years, questions and concerns raised by clinicians have focused on the potential for negative impact on patients or on the healthcare system due to false positive results.
Although the false-positive calls in the PATHFINDER interim analysis outnumbered the true positives, the rate of false positives in the overall cohort remained below one percent. In addition, researchers wrote that any practical negative impact appeared to be limited, based on their surveys of participants.
Among the 6,118 participants who completed a questionnaire either following a "cancer signal not detected" result or upon diagnostic resolution for those with a "cancer signal detected" result, the mean satisfaction score with Galleri testing was about 83 out of 100. In other words, 83 percent of the study participants were "extremely" or "very confident" that the test was a good thing for them to have done. In addition, 90 percent of the cohort were extremely or very certain that "the good things about the test outweigh the bad things."
Galleri testing also didn't appear to reduce individuals' intention to pursue future recommended cancer screening. Before and after testing, the same 96 percent of the cohort said they were likely or very likely to undergo future cancer screening recommended by their physician.
In terms of healthcare utilization, most of the 63 individuals who had a diagnostic resolution had at least one imaging test with the average being about the same for both true- and false-positives.
Most invasive procedures pursued were minimally invasive. And invasive procedures were more common in true positive participants than the false positive participants. Overall, 70 percent of those who underwent invasive procedures were confirmed to have cancer, the study authors reported.
In a separate presentation at ASCO, Beer also shared results of a study that compared the performance of the initial Galleri assay to an updated version of the test optimized to better minimize false positives. In the PATHFINDER cohort, the new version did result in fewer false positive calls, while maintaining a conservative estimate of at least 40 percent positive predictive value.
As Grail continues to build data for Galleri in the clinic, physicians will be looking for even better resolution of the test's true sensitivity and specificity based on longer follow-up. According to the company, PATHFINDER participants will continue to be followed for 12 months, with final results from the trial expected in the first half of 2022.
The firm also announced that it intends to conduct an additional real-world evidence study, called REFLECTION, tracking outcomes for at least 35,000 of the individuals prescribed Galleri by a healthcare provider now that it has been launched for clinical use in the US.
As oncology organizations like ASCO have increased their focus in recent years on racial and socioeconomic cancer disparities, another area of concern for tests like Galleri has become their consistency of performance across diverse populations. Although it was not broken down in PATHFINDER, Grail has begun to research this question, presenting a separate poster at the meeting this week analyzing data from its retrospective CCGA study.
According to the CCGA investigators, Galleri showed consistent specificity in the case-control cohort regardless of race/ethnicity, with the sensitivity of cancer signal detection similar across white, black, Hispanic, and otherwise undefined populations.
The authors wrote that this bodes well for efficacious and equitable use of the test on a population-wide scale.