Intracranial solitary fibrous tumor (SFT) is a rare central nervous system (CNS) tumor that lacks a reliable prognostic clinical model. Uncertainty persists regarding the treatment outcomes of surgery and adjuvant radiotherapy (ART). To address this, we investigated the efficacy of ART and applied machine learning (ML) to develop accurate prognostic models. The SEER database was used for this study's analysis. To identify the prognostic variables, we conducted Cox regression analysis and constructed prognostic models using five ML algorithms to predict 5-year survival. A validation method incorporating the area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to validate the accuracy and reliability of the models. We investigated the role of ART and surgery using Kaplan Meier survival analysis, competing risk analysis and BRACE method. The study population comprised 747 patients. Among them are 316 patients with "surgery" and 431 patients with "surgery + ART." The therapeutic groups showed significant differences in overall survival (OS). Multivariate Cox regression analysis revealed that older age and surgery alone were poor prognostic factors. The most significant prognostic factors were the local tumor excision, followed by lobectomy and age. Although ART did not lead to a substantial decrease in cancer-specific deaths, it did improve OS. This underscores the broader health benefits of ART, including effective management of comorbid conditions. Caution is advised when interpreting these survival benefits because of potential confounding factors in patient health and treatment management. Our web tool and ML models aid in clinical decision-making.
Read full abstract