FDA Approves First DNA-Based CRC Screening Tool

Patrick Conway, MD

The FDA has approved Cologuard as the first noninvasive stool-based DNA test for the detection of colorectal cancer (CRC) in asymptomatic individuals at average risk, based on clinical trial results demonstrating superiority to the fecal immunochemical test (FIT).

In the pivotal study, Cologuard detected CRC in 92% of patients compared with 74% with FIT screening. Additionally, advanced adenomas were detected in 42% of patients with Cologuard compared with 24% with FIT. The novel DNA screening test was reviewed through a joint FDA-CMS program that resulted in an already established coverage plan. Under this plan, CMS will cover the Cologuard test when administered once every 3 years for Medicare beneficiaries between age 50 and 85 who do not demonstrate signs of CRC and are at average risk.

“This is the first time in history that FDA has approved a technology and CMS has proposed national coverage on the same day,” Patrick Conway, MD, chief medical officer and deputy administrator for innovation and quality for CMS, said in a statement. “This parallel review represents unprecedented collaboration between the two agencies and industry and most importantly will provide timely access for Medicare beneficiaries to an innovative screening test to help in the early detection of colorectal cancer.”

The Cologuard test, which is manufactured by Exact Sciences, uses quantitative molecular assays for KRAS mutations, abnormalities in methylated BMP3, NDRG4, and β-actin along with a hemoglobin immunoassay. The approval for the test follows a unanimous recommendation from the FDA's Molecular and Clinical Genetics advisory committee in March 2014.

The study that was the basis for the approval initially enrolled 12,776 participants at average risk for CRC, with 9989 meeting the criteria for evaluation. In the FDA's analysis of the study, they selected data from 10,023 participants for assessment. All individuals in the study underwent a colonoscopy and screening with FIT and Cologuard. By colonoscopy, 65 participants had CRC and 757 had precancerous lesions (advanced adenomas or sessile serrated polyps). The mean age in the study was 64.2 years. The majority of participants (84%) were Caucasian.

According to results published in The New England Journal of Medicine, the sensitivity for detecting CRC in the 9989 eligible patients was 92.3% with Cologuard compared with 73.8% with FIT (P = .002). Cologuard detected 42.4% of advanced precancerous lesions compared with 23.8% with FIT. For high-grade dysplasia (n = 39) and sessile serrate polyps (n = 99), Cologuard detected 69.2% and 42.4% compared with 46.2% and 5.1%, respectively.

Cologuard detected 13 of 60 screening-relevant (stage I-III) cancers that were undetected by FIT compared with 1 cancer detected by FIT that was not found with DNA testing (P <.001). For patients who tested negative for CRC and advanced precancerous lesions (n = 9167), Cologuard detected 86.6%, representing a false positive rate of 13.4%. FIT detected 94.9% of negative findings, for a false positive rate of 5.1%.

“This approval offers patients and physicians another option to screen for colorectal cancer,” Alberto Gutierrez, PhD, director of the Office of In Vitro Diagnostics and Radiological Health at the FDA’s Center for Devices and Radiological Health, said in a press release. “Fecal blood testing is a well-established screening tool and the clinical data showed that the test detected more cancers than a commonly used fecal occult test.”

At this point in time, fecal DNA testing is not currently recommended as a method to screen for colorectal cancer by the USPSTF. The current guidelines for screening in adults between 50 and 75 at average risk for CRC include fecal occult blood testing, sigmoidoscopy, or colonoscopy. The parallel CMS and FDA decision for Cologuard arrived under a pilot program for concurrently reviewing medical devises. The program is open to devises and technologies that fall within the Part A and Part B Medicare benefit categories.

“Parallel review allows the last part of the FDA process to run at the same time as the CMS process, cutting as many as six months from the time from study initiation to coverage,” Nancy Stade, CDRH’s deputy director for policy, said in a statement. “The pilot program is ongoing, but we will apply what we have learned to improve the efficiency of the medical device approval pathway for devices that address an important public health need.”