Background: At least 75% of bladder cancer (BC) are diagnosed in adults aged ≥65 years, with a median of 72 years. Older adults have poor outcomes but are also an heterogeneous population in which the functional status often does not reflect the chronological age. ASCO and SIOG recommend the use of geriatric assessment (GA) to guide management of these patients (pts) but very few data are present to date. Methods: We investigated the role of GA and an oncological version of MPI in pts ≥70 years old affected by advanced/metastatic urothelial carcinoma. Data were retrieved from a prospectively maintained database from 2010 to 2022. Comprehensive GA (CGA) is a multidimensional tool used to classify pts as fit, vulnerable and frail according to Balducci’s criteria. Onco-MPI was calculated by a validated algorithm derived from different GA domains and tumour characteristics. We also collected characteristics of pts, tumors and treatment regimens. Correlation between GA and treatments used and oncological outcomes were analysed. Results: Complete data and follow up were available for 181 patients. Pts characteristics are reported in Table 1. According to Onco-MPI all the pts but 7 were at high risk. The mFU was 46.2 months, while mOS was 8.2 months. Balducci’s CGA and Bellmunt groups had a concordance of 58%, p<0.01. Strong correlations between these scores and type of treatment received were found (pts fit or 0 score received more often a platinum-based treatment without reduction dose while frail or 2-3 score received more often treatment with a reduced dose or no treatment at all, p<0.01). The two scores have a strong prognostic value (18.5 vs 10.8 vs 3.8 months, p<0.01; 21.9 vs 8.5 vs 4.1 months, p<0.01, respectively for CGA and Bellmunt subgroups). At univariate analysis also timing of mets, number of sites, type of treatment and dose were good prognostic variables. All these variables retained their prognostic value at a multivariate analysis incorporating each of the two scores separately. Conclusions: CGA has a great prognostic value in older adults with metastatic urothelial cancer. This tool could help in decision making and patients stratification but validation in prospective cohort is needed. Onco-MPI could also be useful but new cut offs must be set in this specific population.