Colorectal cancer (CRC) is the third-most common type of malignancy in the U.S., and the second-leading cause of cancer deaths. CRC mostly affects older adults and has been declining in this population due to better prevention and early detection. However, CRC is on the rise in adults under age 50, with incidence rates more than doubling in the U.S. since the 1990s. Several other countries have seen similar increases. Typically, this early-onset CRC (EOCRC) is diagnosed at more advanced stages and is more aggressive than CRC diagnosed in older people.

“It’s only in the most recent 10 years that more urgent attention has turned toward early-onset disease,” said Yi-Qian Nancy You, MD, associate professor of colorectal surgical oncology and associate medical director of the clinical cancer genetics program at University of Texas MD Anderson Cancer Center in Houston.

Robust research efforts aim to understand what’s driving the rise in EOCRC, its specific clinical and molecular features, and the best ways to diagnose and prevent this disease. In response to their findings, the U.S. Preventive Services Task Force in October 2020 issued draft guidance that calls for CRC screening in the general population to begin at age 45 rather than 50. But because a research focus on EOCRC is still relatively new, “there’s a great open field for novel applications, tests, and techniques” to detect the disease, You noted.

Diet and More

A slew of recent papers has identified possible culprits behind the rise in EOCRC, particularly poor diet. For instance, researchers at Washington University in St. Louis and the Harvard T.H. Chan School of Public Health showed an association between poor diet quality and risk for early-onset, high-risk distal and rectal adenomas (J Natl Cancer Inst 2020; 10.1093/jnci/djaa164). This analysis of the prospective Nurses’ Health Study II cohort compared associations of a Western diet and recommended healthier eating approaches with early-onset colorectal adenomas—which usually turn into cancer—and those of high malignant potential as surrogate endpoints. The authors found that a Western diet was associated with increased early-onset adenoma risk, whereas healthier diets had an inverse association with adenomas.

This research is among the latest in a series on risk factors for EOCRC by senior author Yin Cao, ScD, associate professor of surgery at Washington University in St. Louis. Her team and collaborators have also found evidence implicating obesity, sedentary behavior, and metabolic syndrome as risk factors, she noted. “Our work suggests that similar to later-onset CRC, EOCRC is a multifactorial disease. Identification of the set of risk factors that contribute to the rising incidence is a priority,” said Cao.

An editorial accompanying Cao’s recent paper calls for stepping up efforts to identify preventable EOCRC risk factors. The authors note that her team looked at adenomas because existing prospective studies have too few cases in younger people to properly investigate risk factors. To address this dearth of EOCRC research subjects, the editorial authors—Neil Murphy, PhD, Peter Campbell, PhD, and Marc Gunter, PhD—recently established the Colorectal Cancer Pooling Project (C2P2), an international effort of more than 25 prospective cohort studies.

C2P2 will comprehensively examine potential risk factors and biomarkers for CRC diagnosed in people in different age groups and provide an infrastructure for unraveling the etiology of EOCRC. C2P2 also plans to study potential biomarkers that might be intermediates of established lifestyle risk factors related to metabolic health and gut dysbiosis, or microbial imbalance. Investigations will begin with a population of EOCRC cases and controls and later expand to older patients for comparison, said editorial co-author and C2P2 co-founder Campbell, scientific director of population science at the American Cancer Society.

The Gut Microbiome

A recent review identified several possible EOCRC risk factors, including a Western diet, with heavy intake of red and processed meats, high-fructose corn syrup, synthetic dyes, and monosodium glutamate, as well as stress, antibiotic use, and sedentary behavior (Nat Rev Gastroenterol Hepatol 2020;17:352-64). The paper calls inflammation and the gut microbiome “guardians of the colon.” First author Lorne J. Hofseth, PhD, a professor of pharmacy and director of the Center for Clinical Colon Cancer Research at University of South Carolina, emphasized the outsized influence of diet: “We have much to learn about the role of diet. But its role in prevention is clear. The low-hanging fruit is to change your freaking diet.”

Among many others, Georgetown University researchers also have noted evidence that the human gut microbiome might play a role in cancer pathogenesis (Curr Oncol Rep 2019;21:3). They describe epidemiologic shifts in CRC incidence and mortality across age groups and differences between younger and older patients in clinicopathologic, molecular, treatment, and survival characteristics. More studies of the microbiome might elucidate bacterial causes of CRC in younger individuals, they note.

Molecular Features

Other research suggests EOCRC has distinct clinical and molecular features in particular age groups, a finding that might warrant more consideration of patient age in testing and in managing the disease (Cancer 2019;125:2002-10). This review of more than 36,000 CRC patients showed that compared with older patients, those with early onset disease were more likely to have microsatellite instability (MSI), although an earlier paper reported that EOCRC is less likely to involve MSI (JAMA Oncol 2017;3:464-71). The Cancer paper also found that in comparison to older patients those with EOCRC were more likely to have primary tumors in the distal colon and rectum, and fewer BRAF V600 mutations. In patients younger than 40, consensus molecular subtype (CMS)1 was the most common, whereas CMS3 and CMS4 were rare.

The review published in Cancer also suggests greater use of genetic testing and whole genome sequencing (WGS), especially for high-risk patients and family members of patients with EOCRC. 

Hofseth noted that WGS and other genetic tests might identify molecular changes in several genes already well-associated with early disease, such as KRAS and P53. These genes, as well as LINE-1 hypomethylation, are very common in nonhereditary cancers and might also serve as EOCRC biomarkers.

Other studies have noted more genetic variations associated with EOCRC. One that observed a higher rate of hereditary cancer syndromes in EOCRC patients found pathogenetic variations in cancer genes not previously associated with CRC, such as BRCA1 and BRCA2 (JAMA Oncol 2017;3:464-71). The Ohio Colorectal Cancer Prevention Initiative examined variations in 450 patients who were prospectively recruited at 51 of the state’s hospitals. Among EOCRC patients, 16% had a pathogenic variant in at least one cancer susceptibility gene. Of these, about half had Lynch syndrome, the most common cause of hereditary CRC, while the other half did not. Of those without Lynch syndrome, about a third had pathogenic variants in genes not previously associated with CRC. Senior author Heather Hampel, MS, CGC, a clinical professor of human genetics at The Ohio State University, said that a forthcoming paper on 3,310 individuals involved in the initiative shows similar results.

National Comprehensive Cancer Network guidelines recommend that all CRC patients under 50 get genetic evaluations, Hampel pointed out. In light of this, her hospital has a liaison between the pathology and genetics departments who identifies EOCRC patients and navigates them to genetics consults.

DNA and Stool-Based Tests

Many groups are developing less invasive CRC methods that rely on circulating tumor cell (ct)DNA for screening, diagnosing, and monitoring treatment for the disease. Most patients—especially younger people—find them easier to accept than colonoscopies. Hofseth noted that fecal samples can reveal changes in the genome, metabolome, microbiome, and inflammation that point to disease, and stool-based DNA tests can find mutations associated with CRC.

But ctDNA and stool tests have downsides, including being less sensitive for detecting adenomas. ctDNA has been used successfully to monitor metastatic disease, but detecting minimal residual disease (MRD), early-stage cancer, or small premalignant, molecular changes has been more challenging. “All of the tests have detection limits. Ongoing efforts aim to improve sensitivity while maintaining specificity,” said You, who is a co-leader in MD Anderson’s Colorectal Cancer Moon Shot Program.

The plethora of ctDNA testing platforms and methods used in symptomatic CRC patients was an impetus for a recent National Cancer Institute Colon and Rectal-Anal Task Force meeting on integrating ctDNA technology in clinical care. You co-authored a white paper based on the proceedings, which notes that ctDNA has potential to detect MRD, monitor responses to therapy, and track clonal dynamics in response to targeted therapies and other systemic treatments. But first, the field must establish optimal assay characteristics for various clinical scenarios, standardize assay quality control and reference materials, and ensure reliable pre-analytical variables. Related clinical trials should be collaborative and benefit from new platforms that help research groups share data and collaborate, the paper adds (Nat Rev Clin Oncol 2020;17:757-70).

In the meantime, “the message remains that early-onset disease remains insufficiently understood and tends to have distinct clinical behavior,” said You. The field is trying to understand underlying molecular differences at various levels, such as DNA and RNA, tumor epigenetics, and the microenvironment including the tissue cellular matrix and microbiome. You emphasized, “Although there has not been a big breakthrough yet, there is a lot of very active ongoing research.”