Análise da sobrevivência em mulheres com câncer de colo do útero

Abstract

Cervical cancer is one of the leading causes of death from neoplasms among women in Brazil, especially in regions with lower socioeconomic development. Although it is preventable and treatable when detected early, it still presents high incidence and mortality rates, particularly in the Northeast region and in the state of Paraíba. Prevention involves HPV vaccination, Pap smear screening, and educational actions; however, challenges such as low adherence, lack of information, and limited access to health services compromise the effectiveness of these strategies. Treatment varies according to the stage of the disease and tends to be more complex and less effective in advanced stages. Given this context, this study aims to evaluate survival and prognostic factors associated with mortality among women with cervical cancer in the state of Paraíba. This is an observational, retrospective study with a quantitative approach, focusing on the survival analysis of women diagnosed with cervical cancer using data from patients treated at a reference hospital for oncological care in Paraíba, which is part of the network of institutions with data available on the TriNetX platform. The final sample size depended on the availability and completeness of records in the platform, respecting the ethical and technical principles of the study, resulting in data from 570 patients. The results showed that the median age of participants was 55 years (SD ± 13.58), ranging from 21 to 87 years, including diagnoses that occurred between 2020 and 2025. Of the total number of patients registered in the database, 73 (12.81%) progressed to death due to cervical cancer, while 497 (87.19%) had no record of the analyzed outcome in the database and were therefore considered censored. The overall survival analysis using the Kaplan–Meier method showed a higher concentration of deaths in the first months after diagnosis, followed by stabilization of the survival curve. The Weibull Mixture Model (WMM) with a cure fraction showed good agreement with the empirical curve, indicating adequate modeling of the temporal survival pattern. Stratified analyses demonstrated that clinical staging is the main prognostic determinant, exhibiting a clear prognostic gradient: patients in stages I–II presented the highest survival probabilities, those in stage III showed intermediate behavior, and patients in stage IV experienced a marked decline in survival with stabilization at a substantially lower level. Additionally, radiotherapy was associated with better survival probabilities throughout the follow-up period.