Survival in patients undergoing surgical resection for brain metastasis from lung cancer and utility of different prognostic scales

Abstract

Brain metastases (BM) from lung cancer are among the most common intracranial tumors. Several studies have published scales to estimate the survival of patients with BM. Routine access to molecular diagnostics and modern oncologic treatments, including targeted therapy and immunotherapy, is limited in low- and middle-income countries (LMICs); therefore, incorporating them into recent prognostic scales may diminish the reliability of the scales in LMICs. This retrospective study aimed to determine the survival of 55 patients who were surgically treated for BM from lung cancer at a Brazilian public tertiary teaching hospital between 2012 and 2022. We determined clinical factors associated with survival, and compared observed survival rates with the estimated survival on prognostic scales. The mean overall survival (OS) was 9.3 months (range:0.2–76.5). At univariate analysis, female sex and improved postoperative Karnofsky performance status (KPS) score were associated with longer survival. The median survival did not differ between groups when classified using the Graded Prognostic Assessment (GPA)-2008, Lung-molecular GPA-2017, and Lung-GPA-2021 scales. According to the Diagnosis-Specific (DS)-GPA-2012 scale, there was a significant difference between the groups. In the multivariate Cox regression survival analysis, a higher DS-GPA-2012 and improved postoperative KPS score remained significantly associated with longer survival. In conclusion, this cohort showed a mean OS of < 1 year. Improved KPS score after surgery was associated with increased survival. This cohort DS-GPA scale demonstrated the highest concordance with observed survival, indicating its potential as a valuable tool for patient stratification in surgical treatment decision-making in LMICs.