Large study finds bacteria and viruses living inside most cancers are likely contamination
Researchers using the world's largest cancer dataset found that most tumours lack a genuine microbiome, suggesting earlier studies may have been fooled by lab contamination—but cancers of the mouth, oesophagus, stomach and bowel do host real microbial communities.
For decades, cancer researchers thought of tumours as purely human disease—rogue cells multiplying out of control. But over the past few years, a growing body of research has suggested otherwise: some cancers appear to come with their own microbiomes, with bacteria, viruses, and fungi living inside the tumour tissue itself.
The problem is that the field has been left in confusion. Competing studies have produced contradictory results, and researchers have used wildly different methods. In one high-profile case, findings could not be replicated and had to be retracted. Without an agreed-upon standard for detecting microbial signals in tumours, the field has been at an impasse—and the stakes matter. If microbes really do help some cancers grow or resist treatment, they could become targets for new therapies.
Settling the question with rigorous data
A team of researchers set out to answer the question properly, drawing on the world's largest collection of cancer genetic data: Genomics England's 100,000 Genomes Project, which includes DNA from over 16,000 tumours. They built what they describe as the most rigorous analysis pipeline yet developed for this kind of work, specifically designed to strip out sources of error—including laboratory contamination from equipment and handling by scientists. They then applied the method to the entire dataset.
The findings were sobering for much of the existing research. Most cancers—including those of the brain, breast, and kidneys—showed no distinguishable microbiome signal beyond background noise. This suggests that earlier studies claiming to find microbial life in these tumours were likely picking up contamination rather than genuine biological organisms.
The particular mix of species varied depending on where in the digestive tract the cancer was and was linked to features such as the cancer's subtype and how many genetic mutations it carried.
But some cancers are genuinely different
However, the picture was strikingly different for cancers of the digestive system. Tumours of the mouth, oesophagus, stomach, and bowel showed clear, consistent evidence of microbial life. The researchers found not just bacteria, but also viruses, fungi, and archaea—organisms similar to bacteria but genetically distinct. In some cases, they even detected trichomonas, a single-celled protozoan parasite.
Importantly, the mix of microbes varied depending on the location of the cancer within the digestive tract and was linked to specific features of the tumour, such as its subtype and the number of genetic mutations it carried. This biological coherence suggests these findings are genuine, not artefacts of the laboratory process.
The challenge: real signals versus contamination
Distinguishing between genuine microbial signals and laboratory contamination proved to be the hardest part of the project. When researchers sequence a tumour, they read every strand of DNA in the sample—both human and non-human. Most cancer researchers simply ignore the non-human portion. This team took the opposite approach: they discarded the human DNA and matched everything left over against known microbial genomes to identify what was actually present.
The work addresses a long-standing frustration in the field. Every research group had been using its own methods and level of rigour, with no agreed-upon benchmark to check new findings against. That inconsistency meant that genuine signals could be lost among false positives, and time, money, and precious patient samples could be wasted chasing contamination artefacts.
According to the reporting, the findings suggest that earlier contradictions in the cancer microbiome literature may stem partly from contamination rather than genuine biological disagreement. For most cancers, the signal appears to be noise. But for digestive-tract cancers, the signal appears real—opening a genuine avenue for future research into how these microbial communities might affect tumour behaviour and treatment response.