Archived

This content is available here for research, reference, and/or recordkeeping.

Date Available

5-1-2028

Year of Publication

2026

Document Type

Graduate Capstone Project

Degree Name

Master of Public Health (M.P.H.)

College

Public Health

Department/School/Program

Public Health

Faculty

Steven Browning, PhD

Committee Member

Krystal Kuhs, PhD

Faculty

Jaclyn McDowell, DrPH

Abstract

Background: Early age onset colorectal cancer (EAO CRC) incidence has increased in recent decades, and Kentucky has historically experienced a higher CRC burden than the United States overall. Prior studies suggest that clinicodemographic characteristics, tumor presentation, and histologic subtype distributions may differ between EAO CRC and non-early age onset CRC (non-EAO CRC), but these patterns have not been well described in Kentucky.

Methods: This retrospective, population-based cohort study used Kentucky Cancer Registry (KCR) data for first primary CRC cases diagnosed from 2000 to 2022. Differences between EAO CRC (diagnosed < 50 years of age) and non-EAO CRC (diagnosed > = 50 years of age) were assessed using descriptive statistics, chi-square tests, Kaplan-Meier methods, and log-rank tests. Cox proportional hazards regression was utilized to evaluate predictors of overall survival among EAO CRC cases. Histology was categorized as adenocarcinoma, not otherwise specified (NOS); mucinous adenocarcinoma; or signet ring cell carcinoma. Covariates included sex, race, tobacco use, family history of CRC, rurality, Appalachian status, subsite, and stage at diagnosis.

Results: Among 37,124 CRC cases, 3,811 (10.3%) were EAO and 33,313 (89.7%) were non-EAO. Histologic subtype distributions differed significantly by age-of-onset group (p < 0.0001), with signet ring cell carcinoma being more common among EAO CRC cases (2.39% vs 1.13%). EAO CRC cases were also more likely to be male, Black, have a family history of CRC, have distal or rectal tumors, and be diagnosed at late stage (all p < 0.05). Overall survival differed significantly by age-of-onset group (log-rank p < 0.0001), with higher 5-year survival among EAO CRC cases than non-EAO CRC cases (62.8% vs 50.0%). Among EAO CRC cases, adjusted Cox models showed worse survival for signet ring cell carcinoma versus adenocarcinoma, NOS (adjusted hazard ratio [aHR]=1.79, 95% CI: 1.40-2.27), Black race (aHR=1.41, 95% CI: 1.19-1.67), Appalachian residence (aHR=1.23, 95% CI: 1.11-1.35), tobacco use (aHR=1.28, 95% CI: 1.15-1.42), and late-stage disease (aHR=3.32, 95% CI: 2.85-3.88). The adjusted histology by subsite interaction was significant (p = 0.061), with signet ring cell carcinoma associated with worse survival among distal tumors (aHR=2.70, 95% CI: 1.83-3.98) and rectal tumors (aHR=2.08, 95% CI: 1.32-3.27), but not proximal tumors (aHR=1.16, 95% CI: 0.77-1.75).

Conclusion: EAO CRC cases in Kentucky differed from non-EAO CRC cases in terms of histologic subtype distribution, tumor location, stage at diagnosis, and overall survival. Among EAO CRC cases, late-stage disease was the strongest predictor of worse survival. Furthermore, the prognostic impact of signet ring cell carcinoma differed by anatomic subsite. These findings help clarify age-specific CRC patterns in Kentucky and support future research on histology, tumor location, and survival disparities among younger CRC patients in this high-burden population.

Download

Available for download on Monday, May 01, 2028

Contact Author

Included in

Public Health Commons

Share

COinS