Trends in Early-Onset Colorectal Cancer in Singapore: Epidemiological Study of a Multiethnic Population.

Increasing Incidence Rates of Colorectal Cancer at Ages 50–54 Years

Background. Colorectal cancer (CRC) incidence rates have increased in younger individuals worldwide. A recent US study reported increases among both non- Hispanic whites (NHW) and non-Hispanic blacks (NHB), but did not consider other racial/ethnic groups. Thus, we examined the most recent CRC rates for the US by age, race/ethnicity, and anatomic location. Methods. Age-standardized incidence rates (ASR, per 100,000) of CRC were calculated for 2001-2002 through 2015-2016 using the US Cancer Statistics Database, which includes data from the National Program of Cancer Registries and the Surveillance, Epidemiology, and End Results Program. US Cancer Statistics covers the entire US population. Results were cross- classified by age (20-49 [early-onset] and 50-74 years [late-onset]), race/ethnicity (NHW, NHB, Hispanic, American Indian/Alaskan Native [AIAN], Asian/Pacific Islander [API]), and location (proximal, distal, rectal). To examine the change in ASR over time, annual percent change (APC) was calculated using joinpoint regression models. Results. Historically, NHB have had the highest rates of early-onset CRC. However, in 2015-2016, early-onset CRC rates were highest in AIAN (ASR=14.72), followed by NHB (ASR=13.20) and NHW (ASR=12.31). The rate of early-onset proximal colon cancer was highest in NHB (ASR=4.83), which was 62% higher than the rate in NHW (ASR=2.98). Early-onset distal colon and rectal cancer rates were highest in AIAN (ASR=3.98 and 6.43, respectively). Between 2001-2002 and 2015-2016, early-onset CRC rates significantly increased among NHWs (APC=1.62%) and AIAN (APC=3.33%). Racial disparities in incidence of early-onset CRC have decreased between NHB and NHW, due to increasing rates in NHW. Among NHW, early-onset CRC has significantly increased for all CRC locations, with the largest increase for rectal cancer (APC=2.22%). Early-onset rates of rectal cancer also significantly increased for AIAN (APC=4.38%) and Hispanics (APC=0.76%), while rates of proximal colon cancer significantly increased for AIAN (APC=2.21%). Rates of late-onset CRC have significantly decreased in all racial/ethnic groups. However, compared to NHW (ASR=82.48), late-onset CRC rates remain 32% higher in NHB (ASR=108.91) and 18% higher in AIAN (ASR=97.06). Conclusion. Early-onset CRC rates have been increasing, while late-onset CRC rates have decreased. The racial disparity between NHB and NHW in incidence of early-onset CRC has decreased, but this is due to increases in NHW—not decreases in NHB. NHB and AIAN have the highest rates of both early-and late-onset CRC. To counter these trends, research should focus on identifying predictors of early-onset CRC to determine who should be screened prior to age 50 and identify underlying causes of early-onset CRC. Further, ongoing prevention efforts must ensure access to screening colonoscopies for AIAN and NHB. Citation Format: Jessica L. Petrick, Lauren E. Barber, Shaneda Warren Andersen, Andrea A. Florio, Julie R. Palmer, Lynn Rosenberg. Racial disparities in early- and late-onset colorectal cancer incidence, 2001- 2016 [abstract]. In: Proceedings of the AACR Virtual Conference: Thirteenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2020 Oct 2-4. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(12 Suppl):Abstract nr PO-223.

Novel, Emerging Risk Factors for Colorectal Cancer Remain Understudied

Evaluation of Early-Life Factors and Early-Onset Colorectal Cancer Among Men and Women in the UK Biobank

Rare De Novo Germline Copy-Number Variation in Testicular Cancer

e15675 Background: The incidence of early onset (EO) colorectal cancer is on the rise in the western world, with approximately 10% of all new cases diagnosed at or before the age of 50. We sought to characterize EO vs average onset (AO) colorectal cancer patients at a dedicated cancer center over a ten-year period. Methods: A retrospective review was performed of an institutional database that prospectively collects clinical and demographic data of all patients with colon or rectal (including rectosigmoid) cancers. Patients were then categorized by age range and their clinical characteristics recorded. Results: A total of 5,890 patients with colorectal cancer were identified between the years 2014 and 2023, with 1,316 (22.3%) diagnosed at or before age 50. Overall, median follow up duration was 1082.5 days (IQR 588-1916). Within the EO group, 800 (60.8%) had colon cancer and 516 (39.2%) had rectal cancer. Yearly rates of colorectal cancer in both EO and overall cancer groups, as well as the ratio of EO cases, remained stable within the study period. EO colorectal cancer was associated with lower overall mortality rate (37.5% vs 41.6%, p = 0.009), as well as lower histologic grade at diagnosis with fewer poorly differentiated (9.2% vs 11.1%) and undifferentiated tumors (2.1% vs 3.4%) when compared to AO group (p = 0.002). There were no differences in the rate of local (0.3% vs 0.5%), regional (0.5% vs 0.8%) or distant recurrences (5.6% vs 4.9%) in the EO and AO groups (p = 0.56). In the subset analysis between colon and rectal cancers, colon cancer was more common within the AO group as compared to EO group (64.3% vs.60.8%, p = 0.02). While there were no differences seen in overall mortality between the groups for colon cancer, rectal cancer had higher overall mortality in the AO group (39.7% vs 30.6%, p = 0.0002). Conclusions: EO colorectal cancer at or before the age 50 may confer lower mortality rate as well as lower histologic grade when compared to AO colorectal cancer patients. In the AO population as compared to the EO group, colon cancer appears to be more prevalent albeit with higher overall mortality rate for rectal cancers. We did not find any yearly difference in proportion of EO patients. Characteristics and outcomes of early onset versus average onset colorectal cancer. EO CRC (n=1,316) AO CRC (n=4,574) P-value Gender (n, %) 0.37 Male 691 (52.5) 2495 (54.5) Female 625 (47.5) 2079 (45.5) Primary sites (n, %) 0.02 Colon 800 (60.8) 2670 (64.3) Rectum 516 (39.2) 1904 (35.7) Grade (n, %) 0.002 Well differentiated 38 (8.0) 85 (3.8) Moderately differentiated 291 (61.1) 995 (56.7) Poorly differentiated 44 (9.2) 293 (11.6) Undifferentiated 10 (2.1) 65 (3.7) Recurrences (n, %) 0.56 Local 4 (0.3) 23 (0.5) Regional 6 (0.5) 37 (0.8) Distant 73 (5.6) 226 (4.9) Overall mortality (n, %) 494 (37.5) 1904 (41.6) 0.009 EO: early onset, AO: average onset, CRC: colorectal cancer.

BackgroundThe incidence rate of colorectal cancer (CRC) in Asian countries is increasing. Furthermore, recent studies have shown a concerning rise in the incidence of CRC among younger patients aged less than 50 years. This study aimed to analyze the incidence trends and clinicopathological features in patients with early-onset CRC (EOCRC) and later-onset CRC (at age ≥ 50 years).MethodsA retrospective analysis was performed on 946 patients with CRC diagnosed from 1997 to 2017 at Universiti Kebangsaan Malaysia Medical Centre. The time trend was assessed by dividing the two decades into four 5-year periods. The mean age-standardized and age-specific incidence rates were calculated by using the 5-year cumulative population of Kuala Lumpur and World Health Organization standard population. The mean incidence was expressed per 100,000 person-years.ResultsAfter a stable (all age groups) CRC incidence rate during the first decade (3.00 per 100,000 and 3.85 per 100,000), it sharply increased to 6.12 per 100,000 in the 2008–2012 period before decreasing to 4.54 per 100,000 in the 2013–2017 period. The CRC incidence trend in later-onset CRC showed a decrease in the 2013–2017 period. Contrariwise, for age groups of 40–44 and 45–49 years, the trends showed an increase in the latter 15 years of the study period (40–44 years: 1.44 to 1.92 to 2.3 per 100,000; 45–49 years: 2.87 to 2.94 to 4.01 per 100,000). Malays’ EOCRC incidence rate increased from 2008–2012 to 2013–2017 for both the age groups 40–44 years (1.46 to 2.89 per 100,000) and 45–49 years (2.73 to 6.51 per 100,000). Nearly one-fifth of EOCRC cases were diagnosed at an advanced stage (Dukes D: 19.9%), and the majority of them had rectal cancer (72.8%).ConclusionThe incidence of EOCRC increased over the period 1997–2017; the patients were predominantly Malays, diagnosed at a later stage, and with cancer commonly localized in the rectal region. All the relevant stakeholders need to work on the management and prevention of CRC in Malaysia.

Comparison of trends in early-onset colorectal cancer in North America and Europe

Racing to stem a surge in early-onset colorectal cancer: After identifying a growing wave of early-onset colorectal cancer cases, researchers are hoping to investigate potential culprits and ramp up screening efforts.

The incidence of early-onset colorectal cancer (CRC) (age, <50 years) continues to increase globally within high-income countries. To examine and compare rates of synchronous neoplasia found in patients at colonoscopic diagnosis of early-onset CRC with rates found at diagnosis of average-onset CRC. In this multisite retrospective and cross-sectional study conducted at Mayo Clinic sites and in the Mayo Clinic Health System from January 1, 2012, to December 31, 2022, 150 randomly selected patients with early-onset CRC were identified from the electronic health record and matched with 150 patients with average-onset CRC based on sex and colonoscopic indication. Patients with known hereditary syndromes, past history of CRC, or inflammatory bowel disease were excluded. Colonoscopic findings (polyp size, number, site) and related histopathologic findings (adenoma, advanced adenoma, sessile serrated polyp) were analyzed in association with cancer clinicopathologic features and molecular data (mismatch repair status, KRAS, and BRAFV600E). Among 300 patients (156 men [52%]), the median age at diagnosis was 43 years (IQR, 39-47 years) for those with early-onset CRC and 67 years (IQR, 57-76) for those with average-onset CRC. Overall, 85% of patients were symptomatic at CRC diagnosis. Cancer stage, grade, molecular features, body mass index, and family history did not differ significantly between these groups. Among patients with colon cancer, the overall prevalence of synchronous neoplasia was similar, yet advanced adenomas were 3 times more frequent in those with early-onset vs average-onset cancers (31 of 75 [41%] vs 10 of 75 [13%]; P < .001). This difference was not associated with cancer stage or primary location. Among patients with rectal cancer, nonadvanced adenomas were less frequent among the early-onset group than the average-onset group (21 of 75 [28%] vs 36 of 75 [48%]), and although the prevalence of advanced adenomas was similar (11 of 75 [15%] vs 14 of 75 [19%]), they were more commonly located in the rectum (early onset, 5 of 11 [45%] vs average onset, 1 of 14 [7%]). Patients with early-onset cancer of the colon were significantly more likely than those with early-onset cancer of the rectum to have a synchronous advanced adenoma (31 of 75 [41%] vs 11 of 75 [15%]; P < .001). In this cross-sectional study, synchronous advanced adenomas were more commonly found in patients with early-onset colon cancer compared with average-onset colon cancer, and they were distributed throughout the colon. In contrast, advanced adenomas were not increased in patients with rectal cancer and, when detected, were predominantly located in the rectum.

Incidence rates (IRs) of early-onset colorectal cancer (EOCRC) are increasing, whereas average-onset colorectal cancer (AOCRC) rates are decreasing. However, rural-urban and racial/ethnic differences in trends by age have not been explored. The objective of this study was to examine joint rural-urban and racial/ethnic trends and disparities in EOCRC and AOCRC IRs. Surveillance, Epidemiology, and End Results data on the incidence of EOCRC (age, 20-49 years) and AOCRC (age, ≥50 years) were analyzed. Annual percent changes (APCs) in trends between 2000 and 2016 were calculated jointly by rurality and race/ethnicity. IRs and rate ratios were calculated for 2012-2016 by rurality, race/ethnicity, sex, and subsite. EOCRC IRs increased 35% from 10.44 to 14.09 per 100,000 in rural populations (APC, 2.09; P < .05) and nearly 20% from 9.37 to 11.20 per 100,000 in urban populations (APC, 1.26; P < .05). AOCRC rates decreased among both rural and urban populations, but the magnitude of improvement was greater in urban populations. EOCRC increased among non-Hispanic White (NHW) populations, although rural non-Hispanic Black (NHB) trends were stable. Between 2012 and 2016, EOCRC IRs were higher among all rural populations in comparison with urban populations, including NHW, NHB, and American Indian/Alaska Native populations. By sex, rural NHB women had the highest EOCRC IRs across subgroup comparisons, and this was driven primarily by colon cancer IRs 62% higher than those of their urban peers. EOCRC IRs increased in rural and urban populations, but the increase was greater in rural populations. NHB and American Indian/Alaska Native populations had particularly notable rural-urban disparities. Future research should examine the etiology of these trends.

Early-onset colorectal cancer (EO-CRC) occurring in individuals under age 50 is rapidly increasing globally, while the incidence of late-onset colorectal cancer (LO-CRC) has decreased over recent years. Previous studies have identified metabolites linked to CRC biology, however tumor-specific differences between EO-CRC and LO-CRC have not been explored. This study aimed to compare the tumor metabolome of EO-CRC and LO-CRC patients to reveal the unique biochemical state of EO-CRC. Mass spectrometry-based untargeted metabolomics was performed on tumor and patient-matched normal tissues from EO-CRC (n=53) and LO-CRC (n=314) patients to identify metabolites significantly altered in tumors (q≤0.05). Metabolite set enrichment analysis, metabolic pathway, and network analyses were performed, to identify the relationship between the altered metabolites and biological function. Analysis revealed 155 metabolites significantly altered between normal and tumor tissues. Homovanillic acid (HVA), a metabolite of dopamine, was uniquely downregulated in EO-CRC. Despite shared changes to HVA-metabolizing genes between EO- and LO-CRC the disruption in catecholamine metabolism may be specific to EO-CRC biology. Pathway and network analysis, supported by gene expression validation, showed that PD-L1 was uniquely decreased in EO-CRC suggesting immunosuppression. Additionally, phospholipid signaling was favored in EO-CRC, whereas LO-CRC tumors showed alterations to EGFR signaling and oxidative stress-related genes. In summary, this study reveals the metabolic nuances in tumor tissues from patients with EO-CRC and LO-CRC, indicating catecholamine metabolism, phospholipid signaling and immunosuppression in the biology of EO-CRC. These findings provide new insight into the metabolism of EO-CRCs that may inform new therapeutic strategies for this group of CRC patients. Citation Format: Caroline Johnson, Abhishek Jain, Allison Janak, Oladimeji Aladelokun, Rolando Garcia-Milian, Xiaomai Ma, Philip Paty, Sajid Khan. Distinct metabolic and genetic alterations in tumors from early-onset versus late-onset colorectal cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: The Rise in Early-Onset Cancers—Knowledge Gaps and Research Opportunities; 2025 Dec 10-13; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(23_Suppl):Abstract nr C009.

Simple SummaryThe incidence of non-hereditary cancer in the left colon and rectum is increasing in young patients worldwide for unknown reasons. To understand this phenomenon, the biology of early-onset colorectal cancer needs to be established. Here, we investigated the immune response to tumors by selecting a highly representative group of patients younger than 45 years matched to those aged 70–75 years, excluding hereditary cases. Both T-cell distribution in tumors and expression of 770 immune-related genes were overall similar between the groups. The findings suggest that the immune response in cancer of the left colon and rectum is not dependent on age and that early-onset colorectal cancer is likely not related to immune response deficiencies.The incidence of left-sided colon and rectal cancer in young people are increasing worldwide, but its causes are poorly understood. It is not clear if the tumor microenvironment is dependent on age of onset, and little is known about the composition of tumor-infiltrating T cells in early-onset colorectal cancer (EOCRC). To address this, we investigated T-cell subsets and performed gene expression immune profiling in sporadic EOCRC tumors and matched average-onset colorectal cancer (AOCRC) tumors. Left-sided colon and rectal tumors from 40 cases were analyzed; 20 EOCRC (<45 years) patients were matched 1:1 to AOCRC (70–75 years) patients by gender, tumor location, and stage. Cases with germline pathogenic variants, inflammatory bowel disease or neoadjuvant-treated tumors were excluded. For T cells in tumors and stroma, a multiplex immunofluorescence assay combined with digital image analysis and machine learning algorithms was used. Immunological mediators in the tumor microenvironment were assessed by NanoString gene expression profiling of mRNA. Immunofluorescence revealed no significant difference between EOCRC and AOCRC with regard to infiltration of total T cells, conventional CD4+ and CD8+ T cells, regulatory T cells, or γδ T cells. Most T cells were located in the stroma in both EOCRC and AOCRC. Immune profiling by gene expression revealed higher expression in AOCRC of the immunoregulatory cytokine IL-10, the inhibitory NK cell receptors KIR3DL3 and KLRB1 (CD161), and IFN-a7 (IFNA7). In contrast, the interferon-induced gene IFIT2 was more highly expressed in EOCRC. However, in a global analysis of 770 tumor immunity genes, no significant differences could be detected. T-cell infiltration and expression of inflammatory mediators are similar in EOCRC and AOCRC. This may indicate that the immune response to cancer in left colon and rectum is not related to age of onset and that EOCRC is likely not driven by immune response deficiency.

Background: The causes of the rising incidence of early-onset colorectal cancer (EOCRC), defined as CRC in patients aged &amp;lt;50, remain unknown. In this study, we evaluated tumor genomic differences in patients with EOCRC versus late-onset CRC (LOCRC, age &amp;gt;60). Methods: The international cohort included 13,262 patients diagnosed with stages I-III colon or rectal cancer who had ctDNA testing using a personalized and tumor-informed multiplex PCR assay (Signatera™ 16-plex bespoke mPCR NGS assay), from which whole-exome sequencing (WES) on the surgically resected tumor was performed. Tumor mutational burden (TMB) and microsatellite instability (MSI) status were derived from WES analysis. The prevalence of gene-wide mutations, pathogenic gene variants, and mutations in known oncogenic pathways was compared between EOCRC and LOCRC groups, stratified by TMB and MSI status. Fisher’s exact test was used to test significance between the groups and p-values were adjusted using the FDR method for multiple test correction. Results: A total of 3,093 patients with EOCRC (70.8% colon, 27.4% rectal, 1.9% unknown) and 10,169 patients with LOCRC (79.9% colon, 18.3% rectal, 1.7% unknown) were included, where 9.0%/37.3%/53.7% were AJCC stages I, II, and III, respectively. Early-onset patients compared to late-onset patients had fewer cases of stage II CRC (30.7% vs. 39.3%, p&amp;lt;0.01) and more cases of stage III CRC (60.9% vs 51.6%, p&amp;lt;0.01). Adjusted by stage, patients with EOCRC were less likely to be MSI-H compared to patients with LOCRC (10% vs. 17%, p&amp;lt;0.01), or have high tumor mutational burden (TMB-H) (15% vs. 19%, p&amp;lt;0.01). Genes of the Hippo, NOTCH, WNT, and RTK-RAS oncogenic pathways were less commonly mutated in the EOCRC cohort (p&amp;lt;0.01). The BRAF V600E mutation was less prevalent in the EOCRC group (3% vs. 15%, p&amp;lt;0.01), regardless of TMB and MSI status. In the TMB-low/MSS group, TP53 mutations were more common in EOCRC (8% vs. 5%, p&amp;lt;0.01), but APC gene mutations were less common in EOCRC (56% vs. 66%, p&amp;lt;0.01). When comparing EOCRC and LOCRC in the TMB-H/MSI-H group, BRAF V600E (4% vs. 60%), RNF43 G659V (16% vs. 45%), and WNT1 G619A (6% vs. 20%) mutations were less prevalent in EOCRC (p&amp;lt;0.01 for all mutations); however, patients with EOCRC had more mutations in PIK3CA H1047R (22% vs. 9%), APC R1468* (11% vs. 3%), and KRAS A146T (7% vs. 2%) gene variants (p&amp;lt;0.01 for all mutations). In the TMB-H/MSS group, EOCRC patients were more likely to have driver mutations in the PI3K pathway (74% vs. 56%, p&amp;lt;0.01). Further, POLE P286R mutations were more common in TMB-H/MSS patients with EOCRC compared to LOCRC (38% vs. 13%, p&amp;lt;0.01), whereas ACVR2A K437R was less common (11% vs. 30%, p&amp;lt;0.01). Conclusion: Patients with LOCRC were more likely to have pathogenic gene variants and mutations in established pathways of CRC carcinogenesis. Tumors in EOCRC cases harbored unique genomic alterations that varied between the TMB-low/MSS, TMB-H/MSI-H, and TMB-H/MSS subpopulations. Citation Format: Eric M. Lander, Samuel Rivero-Hinojosa, Vasily N. Aushev, Jesús Izaguirre-Carbonell, Adham Jurdi, Minetta C. Liu, Cathy Eng. Genomic alterations associated with early-onset and late-onset colorectal cancer. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6696.

Purpose: Despite declines in the overall incidence and mortality of colorectal cancer (CRC), there have been increases in the incidence and mortality of CRC among those &amp;lt; 50 years old (early-onset CRC). Although the reasons for the increases in early-onset CRC are unknown, one hypothesis is that temporal changes in dietary patterns has led to changes in the consumption of foods that are associated with increased CRC risk, in particular red and processed meats. Thus, the purpose of our analyses was to evaluate differences in dietary factors between early-onset and late-onset CRC patients. Methods: We used the Puget Sound SEER cancer registry to identify a population-based sample of patients diagnosed with CRC from 4/1/2016 through 12/31/2018. CRC patients were recruited to the study via mail and telephone, and consented patients completed a questionnaire assessing patient demographics, medical history, and CRC risk factors, including diet. For dietary factors, we ascertained information on the average number of servings per week of fruits, vegetables, red meat, processed meat, and “spicy” foods two years prior to CRC diagnosis. We used logistic regression to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) comparing dietary intake in early-onset vs. late-onset CRC. ORs were adjusted for sex, race/ethnicity, cigarette smoking history, diabetes, hyperlipidemia, and alcohol consumption. Results: There were 304 early-onset and 1,150 late-onset CRC patients in our analyses. Compared to late-onset patients, those with early-onset CRC were less likely to be current smokers or to have a history of diabetes or hyperlipidemia. After adjustment for these differences, compared to late-onset CRC, early-onset CRC patients had higher intake of red meat (OR for quartile 4 vs. quartile 1 = 1.49 CI: 1.02 - 2.17), processed meat (OR for quartile 4 vs. quartile 1 = 1.63 CI: 1.16 -2 .29), and spicy food (OR for quartile 4 vs. quartile 1 = 1.80 CI: 1.25 - 2.58). However, there was not a statistically significant difference between early- and late-onset CRC patients with regard to fruit (OR for quartile 4 vs. quartile 1 = 1.07 CI: 0.73 - 1.57) or vegetable (OR for quartile 4 vs. quartile 1 = 0.87 CI: 0.62 - 1.22) consumption. Conclusion: Our results suggest that dietary patterns differ between early- and late-onset CRC patients; in particular, early-onset CRC patients had higher intake of red meat, processed meat, and spicy food. This may reflect differences is dietary patterns by age; additional research with population-based controls is needed to determine the association between diet and early-onset colorectal cancer. Citation Format: Andrea N. Burnett-Hartman, Mimi (Trucmai) Ton, Chad (Qianchuan) He, Rachel C. Malen, Amanda I. Phipps, Julia D. Labadie, Heather Spencer Feigelson, Polly A. Newcomb. A comparison of dietary factors between early-onset and late-onset colorectal cancer patients [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr LB-161.