PHILADELPHIA — A newly available, noninvasive screening test for colorectal cancer based on detecting a cancer-specific form of DNA had a sensitivity and specificity of greater than 80% in studies with a total of 363 people.
The indication for this screening test, known as ColoSure, is “people who refuse to have a colonoscopy,” Dr. Sanford Markowitz said at a conference sponsored by the American Association for Cancer Research.
“If we use this test, it has the potential to substantially lower the morbidity and mortality from colon cancer,” said Dr. Markowitz, a professor of cancer genetics at Case Western Reserve University in Cleveland.
The new genetic test is “lousy, compared with colonoscopy” as a screen for colon cancer, Dr. Markowitz noted, but many Americans older than age 50 years avoid colonoscopy screening despite the many guidelines that promote it. The new test is a better alternative than fecal occult blood testing (FOBT), he said.
Although a head-to-head comparison between the ColoSure test, which also involves testing a stool specimen, and FOBT has not yet been done, ColoSure was significantly more sensitive than a first-generation genetic-based stool test, PreGen-Plus; PreGen-Plus was previously shown to be significantly better than FOBT (N. Engl. J. Med. 2004;351:2704-14). This pair of findings is highly suggestive that when a head-to-head study is eventually done, ColoSure will prove to be more sensitive than FOBT, Dr. Markowitz said.
Marketing of the ColoSure test, made by Exact Sciences Corp., began last July. The test is offered by two companies: Laboratory Corporation of America (LabCorp), which requires a physician's prescription and charges about $240 as the retail price; and by DNA Direct, which charges $399 for the test but will accept specimens directly from patients without a physician's involvement. The test is licensed by Case Western Reserve University, and Dr. Markowitz and his associates who developed the test receive royalty payments through Case Western.
Although Medicare coverage for the screen was still pending in late September, colorectal cancer-screening guidelines published in May by the American Cancer Society, the U.S. Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology said that “there now are sufficient data to include sDNA [stool DNA] as an acceptable option for CRC [colorectal cancer] screening” (CA Cancer J. Clin. 2008;58:130-60).
The ColoSure test is based on finding a hypermethylated form of the gene that codes for vimentin, a filament protein that helps form cell structure. This hypermethylated form of the gene that's been found in roughly 80% of colorectal cancers probably has no direct relevance to the pathogenic process that results in colon cancer. “It's a marker that is probably downstream from cancer-causing changes,” Dr. Markowitz said in an interview.
The screening test requires that patients send the testing laboratory a complete bowel movement with at least 36 g of stool. Immediately after the specimen is collected it is treated and stored in a preservative solution that aids in maintaining the DNA content of the specimen. Once in the preservative solution, the specimen can be stored and shipped at room temperature.
Evidence for the efficacy of the stool test based on the vimentin gene was published online by Dr. Markowitz and his associates in August. They reported results from a two-phase study that involved 82 people aged 50 years or older who were known to have colorectal cancer based on a recent coloscopy examination, and 281 people who were free of colorectal cancer based on a recent screening colonoscopy.
The sensitivity of the ColoSure test in combined results from both phases of the study was 83%, and the specificity was 82% (Am. J. Gastroenterol. 2008 Aug. 27 [doi:10.1111/j.1572-0241.2008.02088.x]). Sensitivity levels were similar regardless of tumor stage or location in the colon.
In the paper, Dr. Markowitz and his associates also said that the false positives they found in the study may actually represent early detection of neoplasia before it becomes visible on colonoscopy. Although this notion will require confirmation, they suggested that if an apparently false-positive result from stool-DNA testing is encountered in clinical practice, then the patient may need repeat screening by the stool-based test or by colonoscopy sooner than the generally recommended screening interval.
Work is also underway to find one or more additional DNA-based screening markers that could be added to the hypermethylated vimentin gene to boost the sensitivity of the DNA-based screen closer to 100%. Dr. Markowitz said he was optimistic that useful, additional markers will be found.
The new test is 'lousy, compared with colonoscopy,' but it is a better alternative than fecal occult blood testing. DR. MARKOWITZ