NEW YORK – Geneoscopy, which is poised to launch its ColoSense colorectal cancer screening assay as the first major competitor to Exact Sciences' Cologuard, has reached a significant milestone with the recent presentation and simultaneous publication of its CRC-PREVENT trial, a core component of the premarket approval application the firm filed with the US Food and Drug Administration this January.
Major findings of the prospective study, published in the Journal of the American Medical Association, included acute sensitivity for detecting early-stage tumors, notable sensitivity for advanced precancers, and a striking 100 percent detection of colorectal cancer among individuals ages 45 to 49 — a group recently added to the screening-eligible population, and one in which Geneoscopy Cofounder and CSO Erica Barnell argued that colonoscopy alternatives like Cologuard and the new ColoSense may play an especially important role.
The trial enrolled a total of 8,920 individuals aged 45 and older between June 2021 and June 2022, with a mean age of 55 in the final cohort. Overall, the sensitivity of ColoSense for detecting colorectal cancer was 94 percent, and its sensitivity for advanced adenomas was 46 percent.
Specificity, or the test's ability to correctly yield a negative result for patients with no lesions on a subsequent colonoscopy, was 88 percent, investigators reported in the JAMA paper and in a presentation at the American College of Gastroenterology annual meeting.
Compared to the performance numbers for the currently marketed version of Exact Sciences' Cologuard, established by its own pivotal prospective trial, DeeP-C, Colosense had better sensitivity for both cancer and advanced precancers (92 percent and 42 percent, respectively for Cologuard. Specificity, however, was lower compared to Cologuard's 90 percent.
Confounding comparisons, Exact Sciences is readying a second-generation assay for an anticipated 2025 launch. At the same meeting where Geneoscopy presented CRC-PREVENT, Exact shared top-line results from its DeeP-C study of the new version of Cologuard, citing a boost in cancer detection sensitivity to 94 percent and advanced adenoma detection raised to 43 percent, as well as improved specificity of up to 93 percent.
Assuming these numbers, the two tests would offer equivalent cancer detection sensitivity, with Cologuard beating out ColoSense in specificity, and vice versa for advanced precancer detection.
According to Barnell, a particularly exciting finding for the team was that ColoSense maintained its accuracy in individuals below 50 years old, something that Exact has yet to demonstrate.
"About 18 percent of patients currently eligible for colorectal cancer screening are in that age bracket," she said.
Although Exact Sciences has not released data on its sensitivity in 45- to 49-year-olds, it did publish a study of a small cohort, in which it showed a specificity of 95 percent. Because the cohort had no detected cancers, calculating cancer detection sensitivity was impossible. However, the data did show a sensitivity of 33 percent for precancerous lesions.
In CRC-PREVENT, meanwhile, 23 percent of the overall cohort were aged 45 to 49, and more than 10 percent the cancers detected occurred in this group. Overall, ColoSense maintained its performance, demonstrating 100 percent colorectal cancer sensitivity (there were 17 cancers detected) and 45 percent advanced adenoma sensitivity in these younger individuals. Specificity remained 88 percent.
Although there has not been a true head-to-head performance comparison between the two tests, the CRC-PREVENT investigators did query a subset of their cohort who reported previous negative colorectal cancer tests. A total of 323 participants reported having either a negative FIT test in the last year, or a negative Cologuard test within the last three years. Among this group, ColoSense demonstrated 100 percent sensitivity for cancer and 63 percent sensitivity for advanced adenomas.
Study authors also contrasted the population recruited for CRC-PREVENT versus that of Exact Sciences' DeeP-C trial.
In Deep-C, they wrote, 10 percent of all participants and 25 percent of participants with colorectal cancer were in age categories (over 75) for which clinical guidelines do not recommend screening. In contrast, 99 percent of CRC-PREVENT's cohort was within age ranges that would be endorsed by the US Preventive Services Task Force with an A or B rating.
According to Barnell, the commercial success of Cologuard benefits new entrants in an expanded market with continuing unmet need.
"I applaud all that they've done to make headway in terms of demonstrating the utility of noninvasive screening, [but] as we move towards a world where we're introducing the 45- to 49-year-old patient population to the eligible screening cohort, we now have 150 million individuals that require screening and we're still at a noncompliance rate of about 40 percent. That's a lot of people," she said.
"For me, one of the things I'm most excited about is that we're providing another option that has a completely novel modality using RNA and we're providing data to demonstrate very high sensitivity in that young patient population, which is currently the cohort that is least compliant with, colonoscopy and other existing modalities," she added. "As a clinician, I just want patients to have options."
Geneoscopy's test differentiates itself in algorithmically incorporating gene expression — assessed by droplet digital PCR — with fecal immunohistochemical testing and clinical assessment of smoking status to generate a final risk result.
Barnell maintained a similar position in regard to blood-based colorectal cancer screening assays that have recently entered the scene. Although she said emerging data on precancer detection sensitivity raises some concerns, doctors ultimately want patients to get screened one way or another.
"I think the focus for so long has been to try to change the way that the patients view prevention and preventative medicine. And I think whatever we can do to make sure that happens would be beneficial," she said.
Even CRC-PREVENT's own recruitment data illustrates these challenges. About 85 percent of patients were compliant with returning their stool collection kit to Geneoscopy's lab, but it was still a struggle to ensure that these individuals followed up with a colonoscopy.
This was due to what Barnell believes was a particular strength of the study, in that it recruited patients in a different way than other studies like DeeP-C. Instead of funneling patients presenting to endoscopy clinics for a routine colonoscopy, investigators sought out participants in the designated age range for screening using social media, offering monetary compensation to drive compliance. Specifically, participants were offered $200 to participate with $50 awarded to those who sent in a stool sample, and the remaining $150 was provided after completion of a colonoscopy.
This meant that "about 68 percent of patients at the time of enrollment did not have a colonoscopy scheduled," Barnell said. "These are the patients that need these noninvasive tests … so these data are truly reflective of our target population."
Overall, the trial initially recruited over 14,000 individuals, which winnowed down to 8,920. A few thousand never submitted stool samples. Of the 12,000 who did, 91 percent, or 11,034 individuals, passed other inclusion criteria. Barnell and her colleagues managed to get 85 percent of this group to complete a colonoscopy, even though, as stated, 68 percent hadn't entered the study with plans for one.
As the company moves toward commercialization, Geneoscopy is hoping to leverage the strategies they developed in the trial to drive compliance for patients who receive positive ColoSense test results.
"We are hoping that helps us make sure that patients that have a positive test are ultimately seen at the endoscopy center to have those lesions removed," she said.
And just as Exact Sciences has made the move to build out an improved version of its test, Barnell said that Geneoscopy is already working on its own enhancements.
"We built this amazing biobank, this amazing registry that we can now leverage … and while we spent a lot of time and effort on going into our pivotal study to ensure that our current test was accurate, we have not yet reached our limit on how RNA biomarkers can be used to detect both colorectal cancer and precancerous lesions," she said. "I really hope that we can dramatically improve our sensitivity for those precancerous adenomas with our second-generation test and we're actively working on that right now."