The pivotal study of Exact Sciences' (EXAS - Get Report) Cologuard stool-based DNA screening test for colon cancer were published in the New England Journal of Medicine tonight, one day earlier than expected due to an embargo break by the Mayo Clinic.
There are a lot of numbers to go over so let's get to it.
The sensitivity of Cologuard to detect colon cancer was 92.3% compared to 73.8% for fecal immunochemical testing (FIT), the comparator used in the study.
The sensitivity to detect advanced pre-cancer was 42.4% for Cologuard compared 23.8% for FIT. For polyps with high-grade dysplasia, the comparative sensitivities were 69.2% for Cologuard and 46.2% for FIT.
Exact Sciences' previously disclosed Cologuard's superiority to detect colon cancer and pre-cancerous lesions compared to FIT. The data in the NEJM study bear out this finding, with statistically significant results. The study authors note that Cologuard's superior sensitivity was due to the screening test's DNA marker and algorithm. The performance of Cologuard's hemoglobin component was similar to FIT.
There were more false positives with Cologuard than FIT in the study. Cologuard specificity was 86.6% compared to 94.9% for FIT.
The number of people required to be screened to detect 1 cancer would be 154 with colonoscopy, 166 with Cologuard and 208 with FIT, the study concludes.
To detect a single precancerous polyp would require screening of 13 people by colonoscopy, 31 by Cologuard and 55 by FIT.
Cologuard was associated with a higher rate of technical failure. Of the 11,016 people evaluable in the study, 689, or 6%, had Cologuard results excluded due to stool samples that could not be evaluated or technical failure. By comparison, only 0.3% of FIT results were excluded.
The study authors conclude Cologuard was associated with a 27% relative increase in the rate of detection of stage I to stage III colon cancer and a 78% relative increase in the detection of advanced precancerous lesions compared to FIT.
An accompanying editorial in the NEJM concurs with the author's conclusions about the superior sensitivity of Cologuard over FIT but includes cautious remarks about the former's specificity and the higher technical failure rate, which raises the risk that cancers might be missed.
The higher false positive rate for Cologuard is an "important consideration when determining the appropriate interval for screening," the editorial states.
Cologuard is expected to cost more than FIT, so any recommendation of more frequent testing interval could cause reimbursement headaches and lower adoption.
The FDA is convening an advisory panel to review Cologuard on March 27.