Abstract Background: Breast cancer (BC) is the most common type of cancer among women in the world and in Brazil as well. Age is the most important risk factor for the development of cancer. The treatment of BC in elderly women is particularly challenging, as physiological changes, functional deficits, comorbidities and the use of polypharmacy can alter the risk of toxicity. We conducted this trial to determine the risk factors for deaths in less than 180 days after admission in elderly cancer patients with BC treated at the Instituto de Medicina Integral Prof. Fernando Figueira (IMIP), Recife, Brazil, and correlate with specific aspects of this type of cancer, such as type of treatment and immunohistochemical profile. Methods: Prospective cohort, in women aged ≥ 60 years, diagnosed with BC, admitted to the IMIP Oncogeriatric Service. At admission, sociodemographic and clinical variables and 10 tests that make up the CGA (Comprehensive Geriatric Assessment) were evaluated, including Mini Nutritional Assessment Short Form (MNA-SF), Polypharmacy and KPS (Karnofsky Performance Status). The follow-up period was at least six months. Descriptive, bivariate analysis using the Fisher and Chi-square tests, a survival study using the Kaplan-Meier method and univariate analysis were performed, including those with p < 0.20 in the multivariate Cox proportional hazards model controlled by Age and Tumor staging. All subjects involved in the research were informed about the objectives of the study and were only included after voluntarily agreeing to participate, signing the informed consent form. Results: The study sample consisted of 326 patients diagnosed with BC in the period from January 2015 to December 2020, all female, with a mean age of 73.46 years (SD ± 7.81). Sociodemographic, clinical and treatment characteristics are described in Table 1. Of the 326 patients, 46 (14,1%) died, and of these, 21 (45,7%) had an infection and 18 (39,1%) required hospitalization. Regarding the domains of the CGA, the characteristics of patients both in the general population and those who died are described in Table 2. After multivariate analysis by Cox regression of clinical variables related to the Comprehensive Geriatric Assessment (CGA) according to death, were identified as a risk factor for death: MNA-SF < 12 (HR = 2,76, IC95%, 1,49-9,48, p =0,001), KPS ≤ 50% (HR = 4,39, IC95%, 2,03-9,48,p < 0,001). Conclusions: In order to improve the efficacy and safety of the treatment of elderly women with BC and to improve their quality of life, the instruments of CGA should be performed routinely. In this population, KPS and Mini Nutritional Assessment Short Version (MNA – SF) were identified as important predictors of early death (180 days after admission). Therefore, in elderly patients with BC who are at risk for malnutrition and have low KPS, the treatment should be analyzed with caution, as well as the evaluation for concomitant early palliative care. Table 1. Sociodemographic, clinical and treatment characteristics of elderly patients treated for BC at IMIP, Recife, 2015-2020. Table 2. Characteristics of CGA (Compreensive Geriatric Assessment) of elderly patients treated for BC at IMIP, Recife, 2015-2020. Citation Format: Ana Cecilia De Vasconcelos, Candice Santos, Jurema Telles, Maria Julia Mello, Bárbara Luz. RISK FACTORS FOR DEATH IN ELDERLY PATIENTS WITH BREAST CANCER TREATED IN AN ONCOLOGY REFERENCE HOSPITAL IN THE CITY OF RECIFE, BRAZIL [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-18-02.