Blood and stool DNA colon-cancer tests are easy. They still miss most precancer

The newest at-home colorectal cancer tests — a blood draw and a next-generation stool kit — reliably detect cancers that have already formed, but they are weak at the job screening is really for: finding the advanced precancerous polyps that can be removed before they ever become cancer. In the two pivotal 2024 trials, the tests detected 83% and 94% of established colorectal cancers, but only 13% and 43% of advanced precancerous lesions [s1][s2].

That gap is the whole story, because colorectal cancer is one of the few cancers screening can prevent outright rather than merely catch early. This piece is about the molecular tests; the full menu of screening options and why the starting age dropped to 45 is covered separately.

The blood test

In the ECLIPSE trial, published in the New England Journal of Medicine, researchers evaluated a cell-free DNA blood test (marketed as Shield) in an average-risk screening population, using colonoscopy as the reference [s1]. Among 7,861 evaluable participants, the test correctly identified 83.1% of the people found to have colorectal cancer (95% CI, 72.2 to 90.3), and its specificity for advanced neoplasia was 89.6% [s1].

But sensitivity for advanced precancerous lesions was just 13.2% (95% CI, 11.3 to 15.3) [s1]. In plain terms: if the blood test is your only screen, it will miss roughly seven of every eight advanced polyps that a colonoscopy would have found and removed. The trial was funded by the test's maker, Guardant Health [s1].

The next-generation stool test

The BLUE-C trial tested a next-generation multitarget stool DNA test — the successor to the widely used Cologuard — in 20,176 adults undergoing screening colonoscopy [s2]. It performed better than the blood test on both counts: sensitivity for colorectal cancer was 93.9% (95% CI, 87.1 to 97.7), and sensitivity for advanced precancerous lesions was 43.4% (95% CI, 41.3 to 45.6), with specificity for advanced neoplasia of 90.6% [s2].

The trial also ran a standard faecal immunochemical test (FIT) — the cheap, established stool test that looks only for hidden blood — on the same samples. The stool DNA test beat FIT on sensitivity for cancer (93.9% vs 67.3%) and for advanced precancer (43.4% vs 23.3%), but had lower specificity (90.6% vs 94.8%), meaning more false alarms that send healthy people to colonoscopy [s2]. That trade-off is the recurring one in screening: you buy sensitivity with false positives. The study was funded by Exact Sciences, the test's maker [s2].

The specificity gap is not a footnote. Lower specificity means more people with a positive stool DNA test turn out, on the follow-up colonoscopy, to have nothing that needed finding — a colonoscopy they would not have had if they had used FIT, with its cost, preparation and small procedural risk. Multiplied across a screening population, a few percentage points of specificity translate into a large number of unnecessary procedures. That is the hidden price of the stool DNA test's higher sensitivity, and it is why guideline bodies treat the cheaper FIT as a legitimate first-line option rather than an inferior one — the "best" test depends on what a health system and an individual are willing to trade [s3].

Why the precancer number matters most

A test that finds 94% of cancers sounds excellent, and for catching disease early it is. But the reason colonoscopy anchors colorectal screening is that it does something the stool and blood tests structurally cannot: it finds and removes polyps in the same sitting, interrupting the cancer before it starts [s3]. A molecular test that detects fewer than half of advanced precancerous lesions is catching the disease later in its natural history — better than nothing, far better than no screening at all, but not the prevention that a clean colonoscopy delivers.

This is why the US Preventive Services Task Force frames the choice as a menu, not a ranking, and stresses that the best screening test is the one that gets done — repeated on schedule [s3]. A stool or blood test done every one to three years, with a colonoscopy triggered by any positive result, is a legitimate strategy; a single non-invasive test treated as a one-and-done is not [s3]. Any positive at-home result must be followed by a colonoscopy, or the screening is incomplete.

What this means for a reader

The convenience is real, and for people who will not schedule a colonoscopy, an at-home molecular test that they will actually use is a genuine gain over nothing [s3]. The catch is what a negative result licenses. A negative blood or stool DNA test does not mean the bowel is clear; it means no cancer signal was detected on that test, on that day, and the precancer it is worst at finding is exactly the thing screening exists to remove [s1][s2]. These tests widen access. They do not replace the test that both diagnoses and prevents.

Sources

• [s1] New England Journal of Medicine — A Cell-free DNA Blood-Based Test for Colorectal Cancer Screening (2024-03-13)
• [s2] New England Journal of Medicine — Next-Generation Multitarget Stool DNA Test for Colorectal Cancer Screening (2024-03-13)
• [s3] JAMA — Screening for Colorectal Cancer: US Preventive Services Task Force Recommendation Statement (2021-05-18)