Brain metastases (BM) represent the most common intracranial malignancy in adults. Limitations of existing prognostic models reduce their predictivity and clinical applicability. The aim of this study is to validate the cumulative intracranial tumour volume prognostic assessment (CITVPA) as a new prognostic score system for patients with BM treated by Stereotactic Radiosurgery (SRS). Between January 2001 and December 2015, 1894 patients underwent Gamma Knife SRS treatment. The CITVPA model was implemented and validated as follows: the CITV cut-offs were identified thanks to a receiver-operating characteristic (ROC) curve analysis; the survival predictive factors were selected through a Cox proportional hazard model; its prognostic power was compared to RPA, SIR and GPA through the Harrel concordance index (HCI). According to the ROC curve analysis, the CITV cut-off values were set at 1.5 and 4.0 cc. Based on the multivariate analysis, the CITVPA model included: age (OR 1.010, 95% CI 1.005-1.015, p < 0.001), KPS (OR 0.960, 95% CI 0.956-0.965, p < 0.001), extracranial metastases (OR 1.287, 95% CI 1.154-1.437, p < 0.001), BM number (OR 1.193, 95% CI 1.047-1.360, p = 0.008), and CITV (OR 1.028, 95% CI 1.020-1.036, p < 0.001). A score between 0 and 1 was attributed to each prognosticator; a global CITVPA score ranging from 0 to 5 was assigned with higher results corresponding to worse outcomes. The CITVPA (HCI = 0.64) exhibited a significantly (p < 0.001) higher prognostic power compared to RPA (HCI = 0.55), SIR (HCI = 0.55) and GPA (HCI = 0.61). The CITVPA represents a reliable prognostic system for patients with BM treated by SRS. However, further prospective and multicentric studies are necessary before its applicability in clinical practice.
Keywords: Brain metastases; Cumulative intracranial tumour volume; Prognostic score system; Stereotactic radiosurgery.