Comparision of the results of the "predict" online survival prediction calculator with real life data in patients diagnosed with non-metastatic breast cancer

Background: Predict is a calculator that can make survival predictions using the clinical and pathological data of patients diagnosed with breast cancer. We planned our study to evaluate the concordance of Predict's survival estimates with real-life data in our patients diagnosed with non-metastatic breast cancer. Our study aims to compare the 5-year and 10-year survival data of breast cancer patients followed at Bolu Abant Izzet Baysal University Izzet Baysal Training and Research Hospital between 2010 and 2024 with the values predicted by the Predict calculator. With this study, we aimed to assess the effectiveness of survival prediction tools in planning breast cancer treatment. Material and Methods: We retrospectively analyzed the data of 276 breast cancer patients followed in the Medical Oncology outpatient clinic of our hospital. 5-year and 10-year survival predictions were calculated for each patient using the Predict survival calculator. For survival analyses, patients' demographic, clinical, and treatment-related data were recorded into the SPSS database. The survival estimates obtained from the Predict calculator were compared with the patients' actual survival durations. Results: The median follow-up period for the patients was calculated as 117.5 months. During this time, 225 patients (81.5%) were alive, and 51 patients (18.5%) had died. The median age of the patients was 51 years. Factors affecting disease-free survival included the method of disease detection, whether chemotherapy was administered, tumor focality, Ki67 level and the type of chemotherapy given (p=0.008; p=0.027; p=0.001; p=0.001). Factors influencing overall survival were determined to be the patients' age at diagnosis, menopausal status, number of metastatic lymph nodes, tumor stage, focality, tumor subtype, whether chemotherapy was administered in lymph node-negative patients and the type of chemotherapy received (p=0.014; p=0.044; p=0.013; p=0.003; p=0.014; p=0.003; p=0.044; p=0.024). When comparing the predicted and observed rates for 5-year survival using the Predict calculator, the predicted survival rate was 94.2%, and the observed survival rate was 94.6%. This difference was not statistically significant (p=0.838). Similarly, for 10-year survival, the survival rate calculated with Predict was 87.6%, while the observed survival rate was 85.3%; this difference was also not statistically significant (p=0.528). Conclusion: This study evaluated the reliability of the Predict survival calculator in our own patient population with non-metastatic breast cancer and determined that Predict is a reliable tool for survival predictions. The findings suggest that Predict can be used as a clinical decision support tool in patients diagnosed with non-metastatic breast cancer in our country. KEYWORDS: Breast Cancer, Survival, Predict, Prognostic Factors