Prognostic assessment of breast cancer using clinical and pathological variables of more than 1 million patients.

e13170 Background: Understanding the prognosis of cancer patients based on clinical characteristics is crucial for personalized treatment strategies and clinical decision-making. We aimed to analyze survival probabilities of breast cancer patients across various clinical subgroups over time and establish a novel web-based tool. Methods: We utilized data from the SEER Research Plus for the years 2000–2021. Inclusion criteria were adult patients (≥18 years) with available overall survival data. After excluding incomplete records, 1,083,849 cases were analyzed. Clinical variables included histology, stage, grade, TNM, age, race, treatment, receptor & nodal status. Survival probabilities at 24, 60, and 120 months were computed using Cox proportional hazards regression. Results: Survival probabilities in infiltrating ductal carcinoma showed significant variation across clinical characteristics. STAGE: Patients with Stage I disease had the highest survival (24 months: 0.99, 60 months: 0.96, 120 months: 0.93). Survival declined markedly with advanced stages, with Stage III showing survival of 0.88, 0.74, and 0.63 at 24, 60, and 120 months, respectively, and Stage IV having the lowest survival (0.60, 0.33, 0.18). GRADE: Tumor grade strongly influenced survival outcomes. Patients with Grade 1 tumors demonstrated survival probabilities of 0.99, 0.98, and 0.95 at 24, 60, and 120 months, respectively. Grade 2 tumors had slightly lower survival (0.97, 0.93, 0.86), while survival decreased progressively with Grade 3 (0.92, 0.82, 0.76) and Grade 4 tumors (0.90, 0.80, 0.73). TUMOR SIZE: Tumors measuring 10–19.9 mm had the highest survival rates at 24, 60, and 120 months (0.99, 0.98, 0.95). Larger tumors were associated with decreased survival, with survival probabilities of 0.98, 0.95, and 0.91 for tumors 20–39.9 mm, 0.95, 0.87, and 0.79 for tumors 40–99.9 mm, and 0.86, 0.70, and 0.60 for tumors ≥100 mm. RACE: Survival differences by race were notable. Asian/other populations showed the highest survival probabilities (0.97, 0.92, 0.86), followed by White patients (0.96, 0.90, 0.84). Black patients exhibited the lowest survival rates (0.91, 0.82, 0.74). Our extended web-based platform at www.kmplot.com enables real-time uni- and multivariate analysis of clinical sub cohorts. Conclusions: We provide a comprehensive survival analysis for breast cancer, stratified by clinical variables. The established web-based platform offers a user-friendly tool for prognosis assessment, patient cohort size estimation, benefiting clinicians, researchers, and pharmaceutical developers.