April 23, 2025

expert reaction to study suggesting childhood exposure to bacterial toxin from certain E. coli strains may be associated with colorectal cancer rate increase among the young

A study published in Nature looks at E. coli strains and an association with increased colorectal cancer rates.

Professor Trevor Graham, Professor of Genomics and Evolution and Director of the Centre for Evolution and Cancer at The Institute of Cancer Research, London, said:

Is this good quality research?  Are the conclusions backed up by solid data?

“This is very good quality research. The authors have collected bowel cancers from countries around the world and performed whole genome sequencing on them, which detects the pattern of mutations across the genome. Different mutational processes (such as the mutations that happen normally through ageing, or the mutations caused by smoking) make different patterns. Therefore by looking at the mutational patterns, the authors could learn what likely caused the mutations in the bowel cancers. They found different patterns of mutations in cancers from different countries, although it isn’t clear what caused these differences. They also found that, in countries where bowel cancer was more common, there were increased numbers of mutations caused by a special type of E. coli that can live in the bowel (called pks+ E. coli that make a mutagen called colibactin). These colibactin caused mutations which also plausibly caused the cancers to grow in the first place, although we can’t say that from this study. Most importantly, the colibactin mutations were also more common in people who got bowel cancer before the age of 50. This suggests the mutations caused by these bugs in the bowel could be a cause of early-onset bowel cancer, although further studies are needed to confirm this.

“We’ve known for a while that colibactin made by a particular strain of E. coli causes mutations linked to bowel cancers, and also that these mutations likely had a role in causing the cancers to grow in the first place. It had been proposed before that these “bad bugs” could have a role in causing early-onset disease: this work provides strong data yet that the hypothesis is correct.”

Is this an association or causation? Do we know yet that this toxin actually causes the colorectal cancer? 

“The study is correlation only. The data are very suggestive that the colibactin producing E. coli may have a causative role in bowel cancer development, but they do not prove it directly.

“It is certain, from previous work, that colibactin causes mutations in the bowel, and that pks+ E. coli make colibactin. These new data are definitive that the burden of colibactin mutations is greater in early onset bowel cancers.

“It’s not clear when and how the particular strain of E. coli gets into the bowel in the first place, and why they are usually gone again by the time the cancer starts to grow. Therefore, if we were to think about eradicating these “bad bugs”, how we might be able to do that isn’t clear either.

“It’s also possible that these apparently “bad bugs” are actually playing a role in maintaining the overall health of the bowel microbiome: indeed these bugs have been used as a probiotic in some countries in the past years. Eradicating them might have unexpected consequences for gut health.”

How common are these strains of E.coli? Are they the same ones that cause food poisoning?

“E. coli is a usual (normal) part of the human gut microbiome. Pks+ strains are very common in Europe and in many places around the world (where their frequency correlates with cancer incidence). Usually these strains wouldn’t cause food poisoning: indeed they have been used as probiotics to treat intestinal problems.”

Does having this toxin mean you will definitely get young onset colorectal cancer?  Or is this association only seen in some cases where the patient has the genetic signature?

“We don’t have definitive data on whether having the toxin means you will definitely get young-onset colorectal cancer: this study only looked at cancers themselves, not at the bowels of healthy people without cancer. So, it’s quite possible that pks+ E. coli are very common and only a few people with the “bad bugs” will actually go on to get bowel cancer. I think it is very likely cancer only occurs in some cases, because even though someone might have the “bad bugs” that cause mutations, those bugs have to cause the right mutations to make a cancer grow.”

Prof Julian Peto, Professor of Epidemiology, London School of Hygiene & Tropical Medicine, said:

“The paper shows that mutational signatures in colorectal cancer vary between countries, which is good evidence of regional differences in environmental exposures. However, the observation that colibactin signatures are commoner in young colorectal cancers is not good evidence that increasing colibactin exposure is an important cause of the recent increase in colorectal cancer in successive generations born since the 1950s. An equally plausible explanation of these data is that the incidence of colorectal cancers (CRCs) caused by early exposure to colibactin (like CRCs caused by inherited APC mutations) rises less steeply with age than for the majority of CRCs. Their hypothesis should be tested by similar studies on stored histology samples from CRCs diagnosed in successive periods. That would show whether there has been a secular increase in the proportion of CRCs with these signatures.”

Comment provided by our friends at the Spanish SMC:

Dr Isabel Portillo, coordinator of Colorectal and Prenatal Cancer Screenings for the Basque Health Service-Osakidetza, researcher in the Cancer Biomarkers group at the Biobizkaia Health Research Institute, and secretary of the Board of Directors of the Spanish Society of Epidemiology, said:

Does the press release accurately reflect the study?

“Yes, it reflects the issue and the possible association with a biomarker that was under study.”

Is the study of good quality?

“Yes, it presents a well-founded discussion and acknowledges the uncertainties and the need for further studies to confirm and demonstrate the role of the microbiome and its potential beneficial effect, although the association of colibactin with colorectal cancer in young people and in other age groups requires more retrospective and prospective research. It’s surprising that there is no mention of possible environmental and dietary risk factors or healthy habits, which are also related.”

How does this work fit in with the existing evidence?

“It is new evidence of a possible marker associated with colorectal cancer.”

Have the authors considered confounding factors?

“They have been considered; however, the focus is more on genetic analyses than on the interaction with other factors.”

What are the real-world implications?

“Basic research. It is still too early to say that there is only one biomarker, or whether it can be neutralized and how.

“I believe this research is highly relevant for advancing our understanding of the origin and development of cancer (both colorectal cancer and others related to the same markers).”

‘Geographic and age variations in mutational processes in colorectal cancer’ by Marcos Díaz-Gay et al. was published in Nature at 16:00 UK time on Wednesday 23 April 2025.

DOI: 10.1038/s41586-025-09025-8 (2025)

Declared interests

Dr Isabel Portillo: no conflicts of interest.

For all other experts, no response to our request for DOIs was received.