Challenging the age barrier: comparative outcomes in octogenarian and non-octogenarian colorectal cancer patients

The treatment of colorectal cancer (CRC) in elderly patients, particularly in octogenarians, presents unique challenges due to comorbidities and presumed treatment intolerance. However, real-world data on outcomes in this age group remain limited.

This study aims to evaluate and compare the clinical characteristics, treatment modalities, toxicities, and survival outcomes of octogenarian CRC patients with those of younger cohorts.

In this retrospective cohort study, 111 patients diagnosed with stage I-III CRC between March 2021 and December 2023 were analyzed. Patients were stratified into three age groups: octogenarians (≥ 80 years), non-octogenarian elderly (65-79 years), and non-elderly (< 65 years). Data on demographics, comorbidities, ECOG performance status, tumor features, treatment modalities, toxicities, and survival were collected. Survival outcomes were analyzed using Kaplan-Meier and log-rank tests.

Among the patients, 35.1% were octogenarians. Octogenarians had a significantly higher comorbidity burden but similar ECOG scores compared to younger groups. Combined local and systemic treatment was administered in 78.2% of octogenarians and was well tolerated, with comparable hematologic (15.4%) and non-hematologic (28.2%) toxicities across age groups. Disease-free survival (DFS) and overall survival (OS) were favorable in octogenarians, with 36-month OS reaching 91.8%, comparable to or slightly better than non-elderly patients (77.6%). These findings should be interpreted cautiously due to the limited sample size. The non-octogenarian elderly group demonstrated the highest survival rates overall.

Octogenarians with non-metastatic CRC can achieve favorable treatment outcomes comparable to or even better than younger patients, supporting the feasibility of curative-intent treatment in this age group. Age alone should not preclude standard oncologic management in elderly CRC patients. However, the single-center design and modest sample size limit generalizability, and further multicenter validation is required.