Decoding the mysterious surge in early-onset colorectal cancers

The change began slowly. In the 1990s, colorectal cancer rates started creeping up by a few percentage points every year in adults who often were told they were too young to have cancer. At the same time, the incidence decreased markedly in those over the age of 55 years. Researchers noticed the diverging trends but were unable to determine why they continued year after year.

By 2019, epidemiological data revealed that the incidence of early-onset colorectal cancer had risen by an alarming 63% in less than 3 decades.¹ Although the mystery behind its increase remains, several large research projects are digging into a wide range of modifiable and nonmodifiable risk factors to determine what may have changed in the 1950s or 1960s to cause the persistent and worrisome uptick.

“The short answer is we don’t know what’s causing it,” says Robin Mendelsohn, MD, clinical director of the Gastroenterology, Hepatology, and Nutrition Service at Memorial Sloan Kettering Cancer Center in New York. “Obviously we want to figure this out, but we also want people to be taken care of promptly because the earlier cancer’s detected, the better the prognosis and treatment.”

In March 2018, Memorial Sloan Kettering opened its Center for Young Onset Colorectal Cancer to focus on patients who are diagnosed before they turn 50 years old. In January 2021, because of growing demand, the center was expanded to include all gastrointestinal cancers. “We’ve unfortunately been busy,” says Dr Mendelsohn, codirector of what is now called the Center for Young Onset Colorectal and Gastrointestinal Cancer. Although the number of early-onset cancers is highest in the 40- to 49-year-old age group, she notes, the biggest increase in incidence has been in the 20- to 29-year-old group.

To date, the center has seen more than 5500 patients, including nearly 3800 with colorectal cancer. Beyond providing coordinated care to patients who often have far different needs than older patients, Dr Mendelsohn says that the center has become a significant source of studies into why their risk has increased. All patients receive a lengthy questionnaire that asks about their history of medications, exposures, and habits to help to answer what might have shifted over the past 70 years.

Similar detective work has been launched by the new Colorectal Cancer Pooling Project (C2P2), which is also aiming to understand the potential role of a long list of modifiable and non-modifiable risk factors. Peter Campbell, PhD, a project leader and a professor of epidemiology and population health at Albert Einstein College of Medicine in New York, says that the logical starting point was a list of 12 known risk factors linked to regular-onset colorectal cancer, including physical inactivity; cigarette smoking; alcohol use; antibiotic use; eating red or processed meat; and eating low levels of fiber, fruits and vegetables, or calcium.

In puzzling over what exposures or risk factors have changed over time, many researchers have zeroed in on rising rates of obesity. Although some data have suggested that obesity may play a role, the connection has proven tenuous. Dr Mendelsohn says that the average patient at her early-onset cancer center is more likely than not to be overweight or obese. Even so, she points out that patients in the group still have a lower likelihood of being overweight or obese than their counterparts in a national cohort without cancer.

“Though some of them may be obese and that may be contributing, it’s definitely not the whole answer,” she says. Nor is smoking, as smoking rates have declined overall and the center’s patients are less likely than the general population to have ever smoked. The center’s patients are a bit more likely to have diabetes than the general population, however. Given studies that have pointed to diabetes as an independent risk factor, Dr Mendelsohn and her colleagues are exploring whether metabolic disease might be another contributor.

As people age, their gut microbiome diversity generally declines over time, and some studies have associated lower diversity with lower health status. Dr Mendelsohn’s studies also have suggested that patients with early-onset colorectal cancer have a lower level of microbiome diversity than those with average-onset cancer. Even so, she says separating cause and effect could prove difficult. A microbiome can be altered through environmental exposures such as antibiotics and diet, but even then, she says that most people might be hard-pressed to recall their history of antibiotic use and what they ate during adolescence and young adulthood. Despite the limitations, Dr Mendelsohn is hopeful that correlations with diet, medicine use, or other exposures could yet emerge from patient-reported recollections captured by the center’s data set.

Dr Campbell says that antibiotic use was also initially “very high” on his research group’s list of early-onset risk factors to investigate but agrees that it has proven difficult to study. So far, some research has linked overall colorectal cancer risk with past use of antibiotics, but no specific connection has yet been found with early-onset cancer. “It’s a great idea, the biology fits, the timing fits. There are a lot of things that really add up for antibiotic use to be part of the story, but so far, the data just aren’t there to support it,” Dr Campbell says.

will include 19,000 people diagnosed with a verified colon or rectal cancer, including approximately 1500 who were diagnosed before the age of 50 years—a study population approximately an order of magnitude larger than those of most prospective studies to date. “It’s a proof-of-principle study on steroids: It’s a big study,” Dr Campbell says.

Given the lack of definitive answers so far, Dr Campbell says that the project’s large size means that its results could put researchers in a “win–win position” no matter what they suggest. “If the results are very underwhelming, that would actually be very encouraging,” he says. A lack of big signals between early-onset cancer risk and obesity, diet, or other variables being investigated could point in new directions.

“That might relate to environmental exposures. It might relate to infectious agents and processes. It might relate more to social determinants of health and air pollution and things along those lines,” he says. “That’s the beauty with discovery like this: No matter what you get, you kind of win because we didn’t know that before.”

If all goes well and the funding comes through, the project could more than double in size for its second stage. “Our goal is to really build a research infrastructure, put the data together, harmonize it, make it available to the scientific community, and support collaborative research,” Dr Campbell says. Critically, the work could provide an important new resource for other researchers. “We can’t do all of this on our own,” he says. “So we hope to be collaborative with others and explore their hypotheses too.”

One emerging hypothesis, Dr Campbell says, is that early-onset cancers may be linked to accelerated biological aging. “It’s not so much that things are different, just things are happening faster,” he says. Although the exposures may not be necessarily new, in other words, they might be present in much greater amounts. “So we’re quickening that process,” he says.

So far, Dr Campbell says that the general idea fits with epidemiological and molecular data suggesting that colorectal tumors in younger patients lacking a known genetic predisposition are not dramatically different from the cancers in older patients. Instead of a new exposure affecting colon epithelial tissue since 1960, he says, “I think it’s just an accumulation of a lot of bad things happening in that younger age group.”

Even so, both he and Dr Mendelsohn agree that patients may have different exposure–susceptibility combinations that converge to drive early-onset disease. In the end, Dr Mendelsohn says, a two-hit hypothesis may provide the likeliest explanation: For a more susceptible individual, a second hit from exposure to a risk factor may trigger tumor formation.

Whatever that combination may be, she says that one big take-home message is that no adult is too young for colorectal cancer. Recently changed recommendations now suggest that adults should start colonoscopies when they turn 45 years old. Although further lowering that age cutoff may not be logistically feasible, new insights into the most relevant risk factors could help physicians to take a more targeted approach to prevention. “We do know that screening definitely decreases the incidence of mortality. So we just need to figure out how we can screen younger people effectively,” Dr Mendelsohn says.