Patterns of recurrence in a diverse, population-based sample of adults with early-onset colorectal cancer.

43 Background: The incidence of early-onset colorectal cancer (CRC) among adults under 50 has increased by 22% over the past two decades. Despite advances in survivorship care, current guidelines do not adequately address the unique challenges faced by survivors of early-onset CRC, particularly in tailoring surveillance intensity to recurrence risk. This study aimed to characterize recurrence patterns in a diverse, population-based sample to better inform surveillance and survivorship care. Methods: We constructed a population-based cohort of adults <=50 newly diagnosed with stages I-III CRC (2015-2018) and treated with curative-intent surgery from three SEER registries (Georgia, Los Angeles County, and Kentucky, n = 2995). Recurrences were identified through novel methodology utilizing human review of electronic pathology reports, which were then linked to SEER demographic and clinical characteristics. Pathologic-confirmed recurrence was identified in the follow-up period 6 months to five years post-curative intent surgery. Recurrence risk was estimated using proportional hazard regression, adjusting for key demographic and clinical characteristics. Results: Overall, in this population-based sample, 12.5% (n = 373) experienced a pathologic recurrence within five years. Recurrence proportions varied by SEER site: 15% in Georgia, 16% in Kentucky and 7% in California. Recurrence also varied by stage: 6% in stage I, 11% in stage II, and 17% in Stage III. Non-white race/ethnicity, advanced clinical stage, and poor pathologic features were significantly associated with increased recurrence risk (all p < 0.05). Age, biologic sex, poverty, rurality and tumor location were not significantly associated with recurrence. Both Hispanic patients (adjusted HR 1.7, 95% CI: 1.2–2.5) and Black patients (adjusted HR 1.6, 95% CI: 1.2–2.1) had a higher risk of recurrence when compared to white patients. Conclusions: In this diverse, population-based cohort of adults with early-onset CRC, we provide recurrence risk estimates, which also reveal critical disparities by sociodemographic and clinical features. Identifying recurrence patterns and risk phenotypes may enhance current survivorship guidelines and improve outcomes in this growing patient population.