Loading ...

While colorectal cancer (CRC) has long been considered a disease of aging — with about 90% of cases occurring in people older than 50 years — rates of early-onset CRC (EOCRC), defined as diagnosis before age 50, have been rising worldwide. This trend has puzzled researchers as traditional risk factors do not fully explain it.

A new study offers a potential clue, pointing to environment exposures, including a commonly used agricultural pesticide, as potential contributors. Rather than focusing solely on known risk factors, researchers in Spain took a novel “exposome” approach, examining the cumulative environmental and lifestyle exposures individuals experience over their lifetime.

The analysis included more than 450 cases across datasets comparing patients with EOCRC with those diagnosed at age 70 or older. Investigators evaluated 29 exposome-related factors, spanning lifestyle variables (eg, diet and education level), environmental exposures (eg, air pollution), and multiple agricultural pesticides.

Surprisingly, tumor biology showed few meaningful differences between early- and late-onset cases.

“That led us to think that something external must be contributing something that triggers or influences tumor development,” said Jose Seoane, PhD, the study’s senior author and group leader of the Cancer Computational Biology Group at the Vall d’Hebron Institute of Oncology in Barcelona, Spain.

To better capture lifetime exposures, which are difficult to measure directly, the team used DNA methylation as an innovative proxy. These chemical modifications attach to DNA and influence gene expression, reflecting past exposures. These patterns were used to generate “methylation risk scores” from tumor samples.

Using data from The Cancer Genome Atlas and nine additional validation datasets, they identified one herbicide, picloram, as having the strongest and most consistent association with EOCRC.

In an interview with Medscape Medical News, Seoane discussed the key findings of the study, which was published in Nature Medicine.

What led you to explore the association between EOCRC and environmental pesticides and other lifestyle exposures?

The idea came from conversations with our CRC clinical team. They were seeing more cases of EOCRC and asked us a fundamental question: How are these tumors different from those in older patients?

We turned to known risk factors — obesity, diabetes, smoking, and alcohol — but most of that data are based on patient surveys, which sometimes aren’t accurate. CRC develops over decades — often 20-30 years — so a tumor diagnosed in a young person may reflect exposures from much earlier in life. Surveys typically don’t capture that accurately.

We had to figure out a way to focus on the tumors themselves instead of relying on patient recall.

One approach is analyzing “mutational signatures.” For example, a recent study identified a link between EOCRC and colibactin, a toxin produced by a particular strain of Escherichia coli, which leaves a distinct pattern of DNA mutations that can be detected and tied back to earlier infection.

Instead of focusing on mutation patterns, we analyzed DNA methylation — an epigenetic mark that changes upon environmental exposures. We interrogate the methylation profiles in the tumor to identify the exposure.

Among the 14 pesticides analyzed, one herbicide — picloram — showed a strong and consistent association with EOCRC. Why might this be? Why would younger people have greater exposure to this particular herbicide?

The picloram finding was unexpected. We initially focused more on environmental pollution and lifestyle exposures, but we included pesticides based on prior research. When we systematically tested exposures, picloram had the strongest and most consistent association with EOCRC.

As to why younger people might show a stronger association, this ties into our broader hypothesis. Traditionally, CRC is considered an age-related disease because tumors typically arise through a gradual accumulation of mutations over decades.

What we think may be happening here is different: certain environmental exposures — like this herbicide — might trigger or speed up tumor development earlier in life. That’s still a hypothesis, but it could help explain rising rates in younger individuals.

Could timing of pesticide introduction play a role?

Picloram has been used since the 1960s. It’s possible that people born after that period had greater lifetime exposure compared to earlier generations, which could contribute to the pattern we’re seeing. However, we can’t say that definitively.

You also used US county-level data on pesticide use from 1992 to 2012 to analyze the relationship between pesticide application and EOCRC incidence rates. What did the results show after adjusting for socioeconomic factors?

In the initial analysis adjusting for socioeconomic variables, but not other pesticides, glyphosate showed the strongest association, with picloram ranking third.

However, because pesticides are often used together, we ran additional models adjusting for the use of other pesticides. After doing so, picloram emerged as the strongest and most significant association, while glyphosate dropped to second.

Importantly, picloram and glyphosate use were not strongly correlated with each other, suggesting that the picloram signal is independent.

How are people being exposed to this pesticide?

Because this pattern is geographic, it suggests environmental exposure rather than dietary intake. Picloram persists in soil for long periods and can travel beyond where it’s originally applied, meaning exposure could occur even without direct use or consumption of locally treated crops.

Our working hypothesis is this may not be about what people eat, but where they live.

That said, some of the lifestyle findings are also informative. Lower adherence to a Mediterranean diet and lower education levels — both proxies for poorer diet quality — were also associated with EOCRC risk in our analysis. These findings are consistent with large epidemiologic studies linking diet, obesity, and smoking to early-onset disease.

Another innovative aspect of your study was its attempt to distinguish between a patient’s chronological age and the biological ‘age’ of their tumor. Can you talk about how you used the mutational signature known as SBS1, which reflects the number of cell divisions a tumor has undergone, to estimate tumor age independently of the patient’s age?

We wanted to go beyond using a patient’s chronological age because that doesn’t always reflect when a tumor began developing.

When we used this approach to classify tumors as “younger” or “older” based on their SBS1 levels — rather than the patient’s age — we found consistent results. In fact, the association became even stronger: picloram exposure was more closely linked to tumors with lower SBS1 scores, meaning biologically younger tumors.

This supports the idea that certain environmental exposures may accelerate tumor development, leading to cancers that arise earlier and progress more quickly.

What are the next steps in this research process?

One of the main limitations of our study is we did not directly measure picloram exposure. Instead, we inferred it using DNA methylation, which although powerful, limits our ability to draw firm conclusions about the exact role of the herbicide.

This is also an observational study, so we can identify associations but not establish causality. Different types of studies, such as experimental models, are still needed to directly test whether picloram contributes to tumor development.

Seoane reported having no relevant disclosures.

Jennifer Lubell is a freelance medical writer in the Greater Washington Area.