Fusobacterium

There is a bacterium that belongs in your mouth and generally does not survive well in a healthy colon. When it shows up in stool in large numbers, it is usually because something in the colon has changed to let it stay. In most published studies, that something is a tumor.

That makes this test useful and also easy to over-read. It is a research-grade signal rather than a cancer test: run alongside a standard stool blood test, it improves the odds of catching a colorectal cancer that the blood test alone would miss.

What This Test Actually Measures

The assay counts copies of bacterial DNA in your stool. It uses PCR, which multiplies tiny amounts of genetic material until there is enough to measure. The usual target is Fusobacterium nucleatum, though some panels also pick up related species such as F. mortiferum and F. necrophorum. What comes back is how much of this organism's DNA was in the sample. It is not a diagnosis.

The organism grows where there is no oxygen, which is why it lives in the plaque between your teeth and under your gumline. It gets to the colon by being swallowed, and probably also through the bloodstream. Once there, it uses two surface proteins, FadA and Fap2, to grip molecules that abnormal colon cells display and healthy cells largely do not. That selectivity is the whole basis of the test. The bacterium sticks where the lining has already changed.

So a high result reflects that something has made the colon hospitable to an organism that normally just passes through. A low or undetectable result is the ordinary state of a healthy gut.

Colorectal Cancer

Nearly all the human evidence concerns colorectal cancer. Pooling studies that compared cancer cases against healthy controls, this bacterium is roughly ten times more likely to be found in cancer tissue than in healthy tissue. That is a large association by any standard, and the enrichment has been reproduced across many independent cohorts in both stool and tumor samples.

On its own as a screening test, the performance is only moderate. One meta-analysis of thirteen colorectal cancer cohorts found stool testing caught about 71 out of 100 cancers and correctly cleared about 76 out of 100 people without cancer. A separate meta-analysis of ten studies covering 1,198 people put it slightly higher, at 81 out of 100 cancers caught and 77 out of 100 cleared. A test that misses a fifth to a quarter of cancers is not a test you would use alone.

What changes the picture is pairing it with FIT. FIT is the fecal immunochemical test, the standard non-invasive colorectal screen, and it looks for hidden blood in stool. In a study of 490 people, FIT alone caught 73.1% of cancers. Adding stool Fusobacterium quantification pushed that to 92.3%, at some cost to specificity. It picked up about seven in ten of the cancers FIT had missed.

FIT misses cancers that are not bleeding at the moment you collect the sample. This test does not depend on bleeding, so it catches a partly different set of tumors. That is the entire argument for running it, and it only works if you run both.

Why It Performs Poorly on Polyps

Detection collapses for precancerous lesions. Pooled across studies, stool testing caught only about 36 out of 100 adenomas. Adenomas are the polyps that precede most colorectal cancers. In a screening colonoscopy study of 500 people, stool Fusobacterium was not enriched at all in people with non-advanced or advanced adenomas. It was enriched only in invasive cancer. And among 392 primary care patients followed over time, stool Fusobacterium DNA did not predict whether colorectal lesions progressed.

That pattern is why many researchers now describe this bacterium as a passenger rather than a driver. It does not appear to start the process. It moves in after the tumor has already remodeled the local environment into somewhere it can live. This resolves what would otherwise look like a contradiction: the association with established cancer is strong, and the association with the earliest stages is close to absent, because those are two different biological moments.

The consequence is blunt. A low result tells you very little about whether you have polyps. It is not reassurance that your colon is clean, and it is not a reason to delay a colonoscopy you are otherwise due for.

Prognosis and Recurrence After Cancer Treatment

If you have already been treated for colorectal cancer, the marker carries different weight. Pooling survival studies, people with high levels had roughly 1.9 times the risk of dying over follow-up compared with those with low levels. Most of that evidence comes from measuring the organism inside the tumor rather than in stool, which is a related but separate measurement. A multicenter cohort of 740 people with resectable colorectal cancer found that tumors positive for Fusobacterium were linked to shorter survival and to less benefit from chemotherapy given after surgery.

Stool persistence after surgery is the most striking signal. In one cohort, people whose stool still carried the bacterium after their operation were far more likely to relapse, and the signal appeared as much as three years before symptoms. The reported effect was enormous, about sixty-fold, but the range of uncertainty around it ran from barely elevated to more than a thousand-fold, which is what a very small study looks like. Treat it as a direction, not a magnitude.

Two other findings belong here. Elevated stool levels before polypectomy predicted a new polyp appearing 1 to 3 years after the first was removed, with about six times the odds in a cohort of 605 people. And in 87 people followed through colorectal cancer surgery, high pre-surgical stool levels carried roughly five times the odds of cachexia by six months after the operation. Cachexia is the severe weight and muscle loss that derails recovery.

Where the Tumor Is in the Bowel

Higher stool levels associate with tumor location. In a study of 105 people with untreated colorectal cancer, high stool abundance carried about three times the odds of the cancer being rectal rather than colonic, and roughly five times the odds of rectal versus right-sided colon cancer. That is one modest-sized cohort, so read it as a hint about where a signal might be coming from, not as a way to locate a tumor.

Inflammatory Bowel Disease

The bacterium is not exclusive to cancer. It is also elevated in people with active Crohn's disease and chronic bowel inflammation, which is a direct problem for specificity. In an unselected population, an inflamed bowel can produce an elevated result with no tumor anywhere.

In a prospective study of 51 people with Crohn's disease starting the biologic drug ustekinumab, higher stool abundance at baseline predicted a worse response to treatment. That is a small, single-cohort finding, but it points at the same theme. This organism tracks a disrupted gut lining, whatever disrupted it.

Why a Single Reading Can Fool You

Start with the biggest problem. Variation between stool samples from the same person is large. A single spot sample frequently misclassifies whether someone is carrying this organism at all, which is why repeat sampling on a standardized schedule is the only way to tell a real change from noise.

• Bowel inflammation: Crohn's disease and other chronic inflammation raise stool levels independently of any tumor. Fecal calprotectin measures gut inflammation, and once Fusobacterium levels are adjusted for it, the independent diagnostic value often shrinks.
• Transit time and body weight: how fast stool moves through your gut and your body mass index both independently shift how much bacterial DNA ends up in the sample. In one large quantitative profiling study, these confounders explained more of the variation than disease status did.
• Acid-suppressing medication: in a randomized trial in 49 healthy adults, seven days of a proton pump inhibitor caused this organism to appear newly in the stool of about 9% of participants. Omeprazole and its relatives are the common examples. If you take one daily, your result may reflect the drug pushing mouth bacteria downstream rather than anything in your colon.
• Collection and handling: results depend heavily on how the sample is stabilized and frozen. Kits differ, and so do the numbers they produce, which is part of why published accuracy for this marker varies so widely between cohorts.

There is one more wrinkle specific to this organism. Stool levels do not always match what is happening in the tumor itself. Several studies comparing stool and tissue from the same person found the two did not correlate reliably. Stool captures what is shed into the bowel; tissue captures what is embedded in the tumor. They are related measurements, not the same one.

Tracking It Over Time

Given the variability, a single number is close to uninterpretable on its own. What is worth having is a trajectory: a baseline, then repeat measurements collected the same way, at the same point in your routine, so you can see whether the level is stable, rising, or falling.

A workable cadence: get a baseline alongside your FIT, repeat in 3 to 6 months if you are changing something that plausibly affects it, with treating periodontal disease the clearest example, and then test at least annually alongside whatever colorectal screening you are already doing. If you have been treated for colorectal cancer, the post-surgical persistence data argue for closer tracking, since a signal that does not clear is the finding with the most predictive weight behind it.

One caution about what retesting can and cannot confirm. The periodontal treatment studies measured stool Fusobacterium directly, so a drop after dental work is a real, measurable thing. Most other interventions have thinner evidence, and a small change between two samples is at least as likely to be sampling noise as a genuine shift.

What to Do With an Unexpected Result

An elevated result is a reason to look, not a reason to panic. The first move is to establish whether the signal is coming from a tumor, from inflammation, or from your mouth.

• Run FIT at the same time, always. The two tests catch partly different cancers, and the combined result is what the evidence actually supports. An elevated Fusobacterium level with a positive FIT is a much stronger prompt for colonoscopy than either alone.
• Check inflammation. Fecal calprotectin distinguishes an inflamed bowel from a quiet one. If calprotectin is high, inflammation is a plausible explanation for the bacterial signal, and the workup shifts toward gastroenterology rather than oncology.
• Check your mouth. If you have untreated gum disease, the oral source is a live possibility, and periodontal treatment has been shown to lower stool levels.
• Colonoscopy settles it. No stool test replaces direct visualization. If you are at or past screening age, or you have a family history of colorectal cancer, an elevated result is a reason to schedule the colonoscopy rather than to keep testing stool.

If you have symptoms that point at the bowel, rectal bleeding, a persistent change in stool caliber or frequency, or unexplained weight loss, act on those directly. They outrank any stool biomarker, and they warrant a gastroenterologist regardless of what this test shows.