Covariate clustering: women with breast cancer in southwestern Parana, Brazil

Due to the high incidence and aggressiveness of breast cancer, the understanding of specific factors associated with the profile of the disease is necessary. In this context, the aim of the study was to analyze data from 155 patients with breast cancer, attended at a reference hospital for Oncology of the Unified Health System (SUS), in the period 2015-2020, in the southwest region of Parana, Brazil. Using multivariate statistical analysis, sample data were divided into three clusters. The heterogeneity between clusters was obtained by Ward9s method. The clinical and pathological variables obtained from the patients9 medical records were: the presence of intratumoral emboli, presence of lymph nodes, menopausal status, molecular subtype of breast cancer, histological grade, TNM staging of the disease, tumor size (cm), age at diagnosis (years), weight (kg), height (m2) and body mass index (BMI) (kg/m2). From the data of the total sample, it is observed that 70% of the patients were in menopause at diagnosis, 31.5% had tumors containing emboli, and 41% had positive lymph nodes. The prevalence of Luminal subtype B tumors, intermediate histological grade, and TNM staging II was verified. Furthermore, the prevalence of the disease was higher in women aged over 50 years, representing 66% of cases. The BMI of the patients ranged from 17.63 kg/m2 to 51.26 kg/m2, with 26.45% of the patients with a BMI below 25 kg/m2, 40.65% with a BMI between 25 kg/m2 and 30 kg/m2 and 32.9% with BMI above 30 kg/m2. Cluster analysis, using the spatial distribution of patients, showed that the region of Vale do Iguassu was the region with the worst averages for clinical-pathological variables, while the region of Vale do Marrecas had the highest number of breast cancer cases.