Avaliação da capacidade prognóstica da inflamação sistêmica, funcionalidade e estado nutricional em pacientes com câncer avançado

Abstract

Introduction: In the context of advanced cancer, treatment effectiveness and adequate care planning depend on an accurate prediction of patient survival. Classic prognostic tools, widely used in clinical practice, have presented limitations in terms of accuracy due to their complexity and subjectivity, resulting in inaccurate prognostic estimates. Therefore, it is necessary to develop new structures for prognostic evaluation that incorporate validated and objective factors that are easy to apply in practice. Objectives: To compare established prognostic factors (performance status, systemic inflammatory response and body mass index [BMI]) in two international cohorts of patients with advanced cancer; to establish a prognostic framework for treatment in this cohort; to examine the prognostic value of The Global Leadership Initiative on Malnutrition (GLIM) criteria including BMI, weight loss and systemic inflammatory response in patients with advanced cancer. Material and methods: In the first article of this thesis, the following prognostic factors were analyzed in two international cohorts of patients with advanced cancer (n=1,518): performance status (Eastern Cooperative Oncology Group Performance Status [ECOG-PS]), BMI and systemic inflammatory response (modified Glasgow Prognostic Score [mGPS]). The relationship between these and survival was examined using Kaplan−Meier and Cox regression methods. In the second article, the prognostic value of the GLIM criteria was assessed in a combined Brazilian and European cohort (n=1303). The relationship between the factors and patient survival was assessed using Cox regression method. Results: In the first article, according to multivariate analysis, in the European cohort the most highly predictive factors were BMI <20 kg/m2 (hazard ratio [HR] 1.644), BMI 20-21.9 kg/m2 (HR 1.347), ECOG-PS (HR 1.597−11.992) and mGPS (HR 1.843−2.365). In the Brazilian cohort, the most highly predictive factors were ECOG-PS (HR 1.678−8.938) and mGPS (HR 2.103−2.837). A prognostic framework ECOG-PS/mGPS was proposed. In the second article, the survival rate at 3 months varied according to BMI, mGPS, and weight loss (WL). According to multivariate analysis, the most highly predictive factors were BMI/WL grade 3 (HR 1.454, P=0.004), BMI/WL grade 4 (HR 2.285, P<0.001) and mGPS (HR 1.887–2.545, all P < 0.001). A prognostic framework BMI/WL grade/mGPS was proposed. Conclusion: Prognostic factors compared in the Brazilian and European cohorts showed similar prognostic capacity. A high ECOG-PS and mGPS scores, and a high BMI/weight loss score are consistently associated with poorer survival of patients with advanced cancer. Furthermore, the two prognostic frameworks suggested in the articles (ECOG-PS/mGPS and BMI/WL grade/mGPS) can be incorporated into the routine assessment of patients due to their simplicity and clinical utility.