NEW YORK – Liquid biopsy firm Guardant Health passed a highly anticipated milestone on Thursday evening with the majority of a US Food and Drug Administration advisory committee backing its Guardant Shield blood-based colon cancer screening test.
During the event, which highlighted the conclusions of an FDA-commissioned group tasked with evaluating Guardant Shield, a majority of panel members voted to recognize the test as safe, likely effective, and having benefits that outweigh its risk when used in its designated indication.
In a press release following the meeting, Guardant Health described the outcome as a strong endorsement, and in early Friday trading on the Nasdaq, shares of the company jumped about six percent to $23.90 after trading was suspended on Thursday in light of the advisory committee meeting.
Despite a majority endorsement, several members of the advisory committee broke ranks, especially on the second question regarding the real-world efficacy of Guardant's test. One member, Stephen Hewitt, registered his disapproval across all three committee questions. Hewitt is head of the experimental pathology laboratory at the National Cancer Institute's Center for Cancer Research.
The bulk of the first half of the FDA's meeting was focused on Guardant's description of its test, the studies it has conducted on it, and the agency's review of the data it received from the company for its assessment.
Following that was a period of public commentary, and eventually, a discussion among the advisory committee members guided by three pre-specified questions about safety, efficacy, and risks versus benefits.
Panelists questioned in particular the possibility that a blood-based test might sway patients away from colonoscopy, or, conversely, that patients testing positive with Shield might add to already long backlog of individuals seeking out colonoscopies as a primary screen.
A variety of modeling studies were cited, with many on the panel noting a need for post-market analyses to assess the test's impact, both on colonoscopy use and overall mortality.
Shield's low sensitivity for advanced adenomas was also prominent in the discussion, with some committee members accepting the test as simply a cancer detection tool, and others more concerned that it does not provide the same preventative benefit as colonoscopy or stool-based tests that allow doctors to remove precancerous adenomas.
Although Hewitt maintained a hard 'no' throughout, eventually calling the Guardant test not "fit for purpose" as a cancer prevention tool, a majority of the panel members concluded that Guardant's CRC screening test was most likely effective, adequately safe, and that its benefits should outweigh its risk when used appropriately for colorectal cancer screening, but not necessarily for the detection (and destruction) of adenomas.
On the first question of whether the data Guardant had presented provided "reasonable assurance" that Guardant Shield was safe for patients who met its use criteria, panelists voted yes, 8 to 1.
"I do think there's a responsibility going forward for education and proper labeling," said Alexander Borowsky, a professor of pathology at the UC Davis School of Medicine, who voted yes on all three questions." And a further expectation that ongoing studies will specifically address eventual mortality impacts and, intermediate to that, biological features of the differences between screen-detected and undetected lesions when they do arise."
For the second question on expected efficacy, six members voted that they agreed that there is a "reasonable assurance that the Shield test is effective for use in patients who meet the criteria specified in the proposed indication," while another three said they did not.
Both Guardant and the FDA reiterated many times during the meeting that the Shield test's proposed use criteria is to detect colorectal cancer, not necessarily to detect advanced adenomas. However, Hewitt and others noted that the language of the proposed indications for use does still mention adenoma detection.
Padma Rajagopal, a physician scientist at the NCI, said that her 'no' vote on the second question could be shifted with better labeling to indicate the test's poor performance in detecting adenomas.
Stanford University professor and gastroenterologist Sean Spencer said that his votes of yes across the board reflect the test's indication for use in asymptomatic individuals for the detection of colon cancer. "I did feel like it met that indication," he said, though he also stressed that the test's labeling should clearly indicate that the assay is not intended to detect adenomas and is not "designed as a preventative strategy."
Finally, seven panel members voted yes, and two voted no on whether the benefits of Shield outweigh its risk in the patients covered by the test's intended use.
Charity Morgan, a professor in the department of biostatistics at the University of Alabama said that she was "on the fence" regarding Shield's lower sensitivity for stage I tumors compared to competitors like Exact Science's Cologuard. But she ultimately voted yes to all three questions.
"I think if the labeling was very clear that this test is strongest for testing stages two, three, and four, that would turn my vote from a weak yes to a strong yes," Morgan said. "But ultimately I concluded that they did show effectiveness as well as safety and a good risk benefit profile."
Vikesh Singh, a professor of medicine and director of pancreatology at Johns Hopkins Medicine, said that he shared other panelists' issues around labeling, and stressed the need to caution patients that a negative result shouldn't assure them that they don't need a colonoscopy.
However, he argued, Shield's shortcomings compared to colonoscopy and stool-based testing are mediated by its potential to resolve a longstanding issue of low compliance.
"I know the stool test is a little bit better, but it is much more cumbersome for the patients and in many instances, I've seen them just not get done," he said.
"I think if a physician sat down and said, 'Look, I'm happy to order this test, but you need to be clearly consented so I can express to you what are the risks and benefits and the performance characteristics of this test,' I think that would add a lot of value. But I don't know if that's something that the FDA has jurisdiction over."
Investor Reaction
Guardant Shield, as a proposed competitor to Exact Sciences' stool-based Cologuard assay, has garnered a great deal of interest from investors, and anticipation was high for the now-completed advisory committee meeting, which had been previously postponed.
In a Friday note to investors, Craig-Hallum senior research analyst Bill Bonello wrote that although the vote didn't determine whether FDA will approve the test as a first-line screen — akin to Cologuard — or second line-screen for patients that decline the stool test, the majority vote in favor of the test suggests the agency will issue a positive decision one way or another.
Meanwhile, UBS analyst Dan Leonard wrote that the investment bank believes Shield could receive FDA approval with a first-line screening label but with some cautionary language included, exceeding some analysts expectations of a second-line outcome.
JP Morgan's Rachel Vatnsdal said much the same, citing the caveats listed by "yes" voters as evidence that the committee seemed to have leaned, on average, toward a first-line approval with stringent labeling, and reiterating that most investors, in JP Morgan's conversations, had been assuming a second-line approval.