A panel of experts advising the U.S. Food and Drug Administration overwhelmingly decided Thursday that a new blood test to screen for colon and rectal cancer is safe and effective and that its risks outweigh the benefits.
But the charity warned that the blood test has limitations, adding that it is promoting it in the hope of improving the very low numbers of people who are routinely screened for the cancer.
The FDA typically follows the advice of expert committees.
About 150,000 people are diagnosed with colon and rectal cancer each year in the United States, and about 53,000 are expected to die from it this year. Most people who get tested for the disease get a colonoscopy or a stool test, methods that the FDA approved long ago and that studies have shown to be more accurate than a new blood test called SHIELD, developed by Guardant Health in Palo Alto, California.
But for people at average risk for the disease, a blood test is convenient — it doesn't require the complex preparation, fasting, or anesthesia required for a colonoscopy, nor the discomfort of a self-sample test — but if cancer or precancerous changes are detected, a colonoscopy is still required.
The biggest problem with the blood test is that, unlike a colonoscopy, it misses most precancerous changes in the colon that, if found and removed, could prevent cancer from developing. That “really undermines the concept of cancer prevention,” says Dr. Stephen M. Hewitt, commissioner of the National Cancer Institute.
The test is “better than nothing for patients who haven't had anything, but it's not better than a colonoscopy,” said Charity J. Morgan, a committee member and professor of biostatistics at the University of Alabama at Birmingham.
And many people get nothing.
The FDA noted that one-third of people who should be screened for colorectal cancer were not screened, and more than 75% of those who died had not been screened up to date.
If regulators approve Guardant Health's test, it's hoped it could dramatically change the gloomy statistics about colorectal cancer by giving average-risk patients who decline colonoscopy a convenient option to get screened.
Colon cancer is one of the few cancers that can actually be prevented with screening. That's because the disease develops slowly as polyps, tiny, harmless growths in the wall of the large intestine. Most polyps don't cause problems, but some eventually become cancer. If the polyps are found and removed, cancer can be avoided.
If polyps are missed and turn into cancer, they can usually be effectively treated if they are caught before they spread. The five-year survival rate for early-stage colorectal cancer is 91 percent, compared with 14 percent if the cancer has spread.
The Guardant test detected 83% of colon cancers but only 13% of dangerous polyps.
Colonoscopies find 95 percent of dangerous polyps, while the most advanced fecal tests find 42 percent. The Guardant test is less accurate because the work it does is so difficult: It has to find tiny fragments of DNA from dead colon cells that have become mixed in with the blood.
The company claims that its blood test is easy to administer and can be done frequently, increasing the chances that dangerous polyps will eventually be found years before they develop into cancer, but that hasn't been proven.
The committee then considered a remaining question: Does the risk that a blood test could miss dangerous polyps justify the potential for a significant increase in the number of people being screened?
For some committee members, the answer is clearly yes: any screening is better than no screening at all.
“The key is to get more people tested somehow,” said Dr. Alexander D. Borowski, a professor in the department of pathology and laboratory medicine at the University of California, Davis School of Medicine.