Epigenomics and Exact Sciences are facing their big X Factor moment. The two developers of DNA tests for colorectal cancer – and bitter rivals – face US adcoms for their diagnostics on consecutive days, and the audience is waiting to see if either manages to impress the judges.
That they must face a panel is almost an advantage in disguise. Almost all colorectal cancer tests currently sold in the US got on the market without oversight by the FDA, but this could be set to change as the agency moves to close regulatory loopholes. If either one of the companies can convince the FDA to grant premarket approval (PMA) for their test they will be ahead of the game – but that is a pretty big if.
Neither test has distinguished itself in the clinic. Both benchmark themselves against faecal immunochemical testing (FIT), a moderately accurate test currently used as a preliminary screen. A positive result with FIT must be confirmed with colonoscopy.
Epigenomics is up before the Molecular and Clinical Genetics panel first, with a meeting scheduled for March 26. Epi proColon detects methylated Septin 9 DNA, a biomarker for colorectal cancer, in a patient’s blood, and the company is seeking its approval as a way to screen people who are at average risk of colorectal cancer.
The panel will scrutinise the pivotal US trial, a head-to-head study pitting Epi proColon against Eiken’s OC-Sensor FIT test. In the study, Epi proColon detected 71% of colorectal cancer cases versus OC-Sensor's 67%, showing it to be non-inferior on sensitivity. Specificity, however, was a bust: 81% with Epi proColon, compared with 98% with FIT (EP Vantage Interview – Epigenomics’ confidence in approval could be false positive, December 12, 2012).
Epigenomics contends that this does not matter, saying that the specificity of other screening tests, for example those for prostate specific antigen that are used to screen for prostate cancer, are between 35% and 50%. These are not only approved but successful on the market, it argues.
The following day it is Exact Sciences’ turn. Cologuard analyses stool samples for haemoglobin, multiple DNA methylation and mutational markers, and the total amount of human DNA. This method would be a tougher sell to patients than a blood test, but how much this consideration will weigh with the panel members is hard to say.
Both sensitivity and specificity with Cologuard appear to be higher than with Epi proColon, although comparing two different trials is rarely more than a vague guide. The US approval trial, Deep-C, put Cologuard’s sensitivity and specificity at 92% and 87% respectively; as a result, Exact only claims non-inferiority to FIT when it comes to sensitivity.
Exact had hoped for a secret weapon by proving that Cologuard could detect lesions before they actually became cancerous, thereby permitting earlier treatment. Deep-C kyboshed this, showing a precancer sensitivity of just 42% (Exact Sciences braces for fallout from Cologuard’s precancer miss, April 18, 2013). Full Deep-C data are expected to be published some time this month in advance of the panel. It is possible that they will show greater promise than the topline results.
Whether Exact Sciences’ test is better than Epigenomics’ or vice versa is an academic question: it is FIT that they have to beat. Colonoscopy remains the gold standard for detecting colorectal cancer. But it is painful, time-consuming and expensive, which is why there is a gap in the market for a broad-brush screening test. The trouble is, FIT already occupies this niche, and it is cheap. Neither Epi proColon nor Cologuard will have an easy time displacing it.
|Epigenomics' Epi proColon vs FIT||NCT01580540|
|Exact Sciences' Cologuard vs FIT (Deep-C)||NCT01397747|