Development of a predictive model for the prognosis of breast cancer patients

• Grant number: 23/10706-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: October 01, 2023
• End date: February 23, 2025
• Field of knowledge: Biological Sciences - Morphology - Cytology and Cell Biology
• Principal Investigator: Luciana Rodrigues Carvalho Barros
• Grantee: Patricia Honorato Moreira
• Host Institution: Instituto do Câncer do Estado de São Paulo Octavio Frias de Oliveira (ICESP). Coordenadoria de Serviços de Saúde (CSS). Secretaria da Saúde (São Paulo - Estado). São Paulo , SP, Brazil

Abstract

Breast cancer is one of the most prevalent neoplasms among women, representing a quarter of diagnosed cases and being responsible for a significant portion of annual deaths. Treatment and prognosis changed according to tumor characteristics, such as histological subtype and advanced diagnosis. Given the heterogeneity of responses to treatment, biomarkers that help predict outcomes are highly convertible. Recently, the neutrophil to lymphocyte ratio (NLR) has emerged as a promising biomarker. In this context, we intend to carry out a study to analyze retrospective data from patients diagnosed with breast cancer between 2008 and 2022, focusing on the evaluation of NLR and its clinical faith. The analysis will be conducted with absolute values of these markers through complete blood count tests performed before and during treatment. As a preliminary result, we used the RNL value of 3.5 as a treshold, below being indicative of a normal index. To assess whether the NLR is a planned factor, we constructed the Kaplan-Meier curve and performed the log-rank test. A univariate Cox regression analysis was also performed with other blood count cells, identifying significant prognostic factors. This preliminary retrospective analysis suggests that NLR may be a prognostic growth factor in breast cancer. In view of the results, the next stage of the research is to create and validate predictive models. We seek to implement machine learning techniques to develop algorithms capable of predicting clinical outcomes based on NLR and other hemogram markers. Thus, the objective is to evaluate NLR as a biomarker in breast cancer and to offer tools that help health professionals in making therapeutic decisions. The ultimate goal is to maximize therapeutic success and improve the quality of life of patients affected by this neoplasm.