This study is a branch-off of another project that we have been looking at. It is related to this thing, it’s a SMS88 colibactin mutational signature. Just to take a step back with that, colibactin is a genotoxin that is produced by E. coli that are pks island positive. This signature is an aggregate of mutations that are in the tumour that are caused by colibactin exposure and this has been validated in some experimental studies. So we know that tumours that have this signature had been exposed to colibactin at some point in time, we’re not sure exactly when the time-point is for that.
Colorectal cancer is a disease that has many different subtypes, so we were curious if these traditional risk factors like BMI, alcohol, smoking, dietary factors, if these associations with risk and survival were different based on if the tumour had this colibactin signature or if it didn’t. So that was the main objective of our study here.
This is a study within the GECCO Consortium, so that’s the Genetics and Epidemiology of Colorectal Cancer Consortium. It’s a case-controlled study so we had colorectal cancer cases and then controls from different studies that are part of GECCO. So we measured this SBS88 or colibactin signature in our cases and then we looked at risk associations where we looked at colibactin positive cases compared to controls and then also colibactin negative cases compared to controls to look at differential risk. Then we did a survival analysis stratified by colibactin signature status to look at the association for these different epidemiologic factors’ association with colorectal cancer specific survival among SBS88 positive or negative or colibactin positive or negative cases.
The main result that we thought was most interesting was that we found that among cases that had the colibactin signature there was a stronger association between higher BMI category and worse CRC specific survival among those who had the colibactin positive signature. We did not see this association among those colibactin negative. So that was the strongest association that we found. We also found some suggestion of some associations with differential CRC risk for some different factors and we found some of these associations among cohort studies only because in GECCO we have some case-controlled studies and then we also have some case-controlled studies that are nested within larger cohorts. So the larger cohorts have their environmental factors at an earlier time-point generally. So, some of the risk assessments we need future research on but the BMI category association with colorectal cancer specific survival that was differential by this signature is the most interesting finding that we found so far.
What are the next steps for this study?
One thing that I alluded to at the beginning about colibactin is that this signature is in the tumour but we don’t really know when the tumour was exposed or when the person was exposed to this pks positive E. coli. So we just know at some point in the past you were exposed to this E. coli that has this genotoxin and then the signature is the evidence that remains but we don’t know when that happened. There is some evidence in the literature that this could potentially happen in an early childhood exposure which would make it more difficult to intervene on. But knowing the timing of colibactin exposure and when we could potentially intervene with patients would be really interesting to study in the future.
Then also just generally as we know, and as I mentioned earlier, colorectal cancer has many different subtypes and these can be treated differently and have different risk factors. The more we know about these different subtypes, the more we will be able to help patients and clinicians in the future make decisions about either their treatment or what their risk factors are. So the more information we can gather about these tumours that are related to colibactin or not related to colibactin, is just going to improve everyone’s knowledge about this different subtype.