Background: Comprehensive geriatric assessment (CGA) is recommended to inform treatment decisions in older patients with cancer. Patients ≥ 70 years are referred to the Geriatric Oncology and Liaison (GOAL) clinic in University Hospital Waterford, for a pre-treatment CGA, which includes assessment of mobility, cognitive, nutritional, functional, and psychological status, along with comprehensive geriatric pharmacological assessment. Informed treatment recommendations are made using CGA outcomes and risk assessed using the Cancer and Aging Research Group toxicity calculator. Limited information exists on treatment decisions based on age subgroups within the geriatric oncology population. Methods: Data from patients with gastrointestinal (GI) malignancies attending the GOAL clinic over a 3-year period to September 2022 were included. Patients ≥ 70 years with a G-8 (screening tool) score ≤14, and all patients aged ≥ 80 underwent CGA. An age-based subgroup analysis (< 80 vs ≥ 80) of systemic anti-cancer therapy (SACT) decisions was performed using the Chi-square test. Results: In total, 206 patients with GI malignancies were assessed, comprising 67 (33%) patients with gastroesophageal cancer, 40 (19%) with pancreaticobiliary and 92 (45%) with colorectal cancer. 40% of patients were ≥ 80 years (n = 83). The majority of patients had stage III/IV disease (82%, n = 168). Across all groups, 58% (n = 120) of patients were treated with SACT, comprising 70% of those < 80 (n = 86) and 41% (n= 34) of those ≥ 80. Across all stages (I-IV), patients ≥ 80 years were less likely to be treated with SACT compared to patients < 80 years (p < 0.01). There was no significant difference between age groups when comparing treatment intent (curative vs palliative) (p = 0.15). There was no significant difference between age groups for baseline dose reductions vs full dose (p = 0.18). Patients ≥ 80 years were less likely to receive polychemotherapy regimens in comparison with the < 80 subgroup (p < 0.01). Conclusions: Older patients with GI malignancies are a heterogeneous population. In the setting of a CGA, patients ≥ 80 years are less likely to receive SACT across any stage group, and are more likely to receive monochemotherapy regimens. The use of CGA ensures a considered and personalised approach to identify older adults most likely to benefit from active treatment and those where best supportive care alone is more appropriate.