NEW YORK – As the advent of competition from blood-based cancer screening tests looms, Exact Sciences is continuing to project confidence in its Cologuard stool assay.
During a call discussing the firm's first quarter financial results on Tuesday, CEO Kevin Conroy argued that the path forward for blood-based colorectal cancer screening, including its own planned test, is not without hurdles. Moreover, regulatory and reimbursement requirements are likely to silo blood tests in a distinct category, separate from Cologuard and its market, likely hampering clinical adoption.
"We know that blood testing will be less accurate than Cologuard and certainly less accurate than colonoscopy," Conroy said during the call, citing internal calculations and modeling. As such, Exact believes that persuading physicians to order a less accurate test will be a challenge.
"Gaining [US Food and Drug Administration] approval [with] an equivalent claim to Cologuard is probably going to be challenging for all of those who are aspiring to enter the colon cancer screening market with a blood-based test … and the bar for guideline inclusion is high," he added.
The discussion was a timely one, as one of the main competitors expected in the blood-based screening space, Guardant Health is expected to release prospective trial data for its assay in the next month or two, offering highly anticipated measures of sensitivity and specificity.
In a note to investors, William Blair analyst Brian Weinstein wrote that Guardant's looming data release has caused recent volatility in the stocks of both companies, adding that "the pickleball is now in Guardant’s court."
Regarding test sensitivity, Conroy argued that while overall cancer sensitivity is important, detection of early-stage tumors and advanced pre-cancers is where much of this challenge lies.
US Preventive Services Task Force "guidelines are based on a rigorous statistical modeling approach that is largely dependent on advanced adenoma detection," he added.
Assuming overall sensitivity of 80 percent, the company's modeling suggests that a blood-based cancer screening test would have stage I sensitivity around 70 percent, compared to Cologuard's 90 percent. Advanced adenoma detection would be expected to fall around 15 percent versus Cologuard's 42 percent sensitivity.
"FDA historically does not like to see performance degradation with new technologies that are subject to [premarket approval] or 510(k)" clearance, Conroy said.
"We’ve seen examples of this in colon cancer screening in the recent past," he added, citing Septin9 testing, which the FDA approved but only with relatively restrictive labeling. "We think there is a role for blood-based testing. But marketing a test with inferior claim language is certainly going to be a challenge."
Based on Exact's modeling, Conroy said that a CRC blood test performed at the same three-year interval as Cologuard would be expected to detect significantly fewer cancers and pre-cancers, leading to a significant degradation in life years gained. In this light, "the only way a blood test really gets into guidelines is if it's an annual test," which means lower reimbursement rates from Medicare.
"Cologuard at every three years was priced just under one third of the Medicare average for colonoscopy screening, which was indicated for 10 years. So, 10 divided by 3 roughly got you to $500 a test," Conroy said. "$500 divided by 3, we believe, gets you into somewhere between $100 and $200 a test."
In another note to investors this week BTIG analyst Mark Massaro called the modeling "interesting" and said that the investment bank believes Guardant Health's actual data will come in with sensitivity above 80 percent, potentially as high as 90 percent.
Guardant has previously reported data from retrospective analyses showing 88 percent sensitivity in stage I-II patients and 93 percent sensitivity in stage III patients with an overall specificity of 94 percent.
But Exact Sciences Chief Commercial Officer Everett Cunningham argued that prospective trials tend to report performance lower than seen in case-control cohorts.
"There's going to be a day when everybody who has these programs has to flip over a card and show their data, and that day is going to be an interesting day," Cunningham said during the company's call.
"Having been down this road before, [performance] is going to suffer degradation in the prospective setting where you have no idea what patients are involved in, whether that matches your case control data," he argued.
In a separate report, Massaro also shared results of a BTIG survey of 85 US primary care physicians, rebutting to a certain extent Conroy's argument that marketing a blood-based test with lower performance than Cologuard would be challenging.
"While our survey size is small, we believe our data can be used to understand clinician demand for a new blood-based CRC test," Massaro wrote. "If Guardant is able to report out sensitivity of [at least] 85 percent at specificity of 90 percent … we expect the blood-based Guardant Shield assay to gain significant uptake in the United States," he added.
In their survey responses, 88 percent of the physicians in the study said they would order a blood-based CRC test, assuming it's FDA cleared, reimbursed by Medicare, and deemed "accurate."
Among these, 40 percent would order testing for patients that do not comply with colonoscopy, 27 percent for the majority of their patients, and 20 percent for patients that do not comply with any screening modality, including Cologuard.
According to Massaro, 90 percent sensitivity "marks a key inflection point." More than 70 percent of doctors responded that they would order testing at sensitivity of 90 percent or higher, but only about 14 percent would do so at 75 percent sensitivity.
The Centers for Medicare and Medicaid Services already has a coverage policy in place mandating that any CRC blood test that is FDA approved and achieves at least 74 percent sensitivity and 90 percent specificity will be reimbursed.
According to BTIG, 75 percent of doctors surveyed also said they would order testing regardless of whether it was recommended by the USPSTF.
On Exact's call, Conroy painted a different picture based on the company's real-world experience. "We have been actively promoting Cologuard to primary care physicians for eight years. We know this market well, and even with a large commercial organization and a large marketing effort, there is still a bias towards colonoscopy," he said.
He added that Exact does not believe a competitor blood-based screening test will be FDA-cleared and well reimbursed for about two years.