Long-term survival in patients with brain metastases-clinical characterization of a rare scenario.

PURPOSE: This study aimed to assess clinical, treatment, and prognostic features in patients with brain metastases (BM) from solid tumors achieving long-term survival (LTS). Further, the accuracy of diagnosis-specific Graded Prognostic Assessment scores (ds-GPA) to predict LTS was evaluated. METHODS: Patients admitted for radiotherapy of BM between 2010 and 2020 at a large tertiary cancer center with survival of at least 3 years from diagnosis of BM were included. Patient, tumor, treatment characteristics and ds-GPA were compiled retrospectively. RESULTS: From a total of 1248 patients with BM, 61 (4.9%) survived ≥ 3 years. In 40 patients, detailed patient charts were available. Among LTS patients, median survival time from diagnosis of BM was 51.5 months. Most frequent primary tumors were lung cancer (45%), melanoma (20%), and breast cancer (17.5%). At the time of diagnosis of BM, 11/40 patients (27.5%) had oligometastatic disease. Estimated mean survival time based on ds-GPA was 19.7 months (in 8 cases estimated survival < 12 months). Resection followed by focal or whole-brain radiotherapy (WBRT) was often applied (60%), followed by primary stereotactic radiotherapy (SRT) (20%) or WBRT (20%). 80% of patients received systemic treatment, appearing particularly active in specifically altered non-small lung cancer (NSCLC), melanoma, and HER2-positive breast cancer. Karnofsky performance score (KPS) and the presence of oligometastatic disease at BM diagnosis were persisting prognostic factors in LTS patients. CONCLUSION: In this monocentric setting reflecting daily pattern of care, LTS with BM is heterogeneous and difficult to predict. Effective local treatment and modern systemic therapies often appear crucial for LTS. The impact of concomitant diseases and frailty is not clear.

Analysis of frequency of deep white matter metastasis on cerebral MRI.

Supratentorial white matter is an important part of the brain and a major site of detrimental effects after whole brain radiotherapy (WBRT). It is not known if prevalence of metastases in white matter justifies standard inclusion of white matter in whole brain treatment. In this retrospective analysis we examined the frequency of metastasis in supratentorial deep cerebral white matter with cerebral magnetic resonance imaging (MRI). Deep white matter (DWM) was defined as white matter in corpus callosum with forceps anterior and posterior and centrum semiovale. Lesions extending from grey matter, gyrus or ventricles into white matter were not classified as DWM metastases. Brain MRI of 198 patients from two centres were analyzed. In total 1330 metastases were counted and only 4.6% were located in DWM. Metastases in DWM were small (median diameter 6 mm). Only 1/41 patients (2%) with a singular metastasis had a DWM metastasis, 2/35 patients (6%) with 2 metastases had a DWM metastasis, 14/79 patients (18%) with 3-9 metastases and 12/43 patients (28%) with >9 metastases had a single or more DWM metastases (p = 0.003). There appeared to be tumor related differences with renal cell carcinoma showing significantly more DWM metastasis (6/17, 35%), than NSCLC (11/85, 13%, p = 0.024), breast cancer (1/20, 5%, p = 0.019) or colorectal cancer (0/10, 0%, p = 0.033). Overall, relevant preservation of DWM from metastases, especially in oligometastatic disease, was shown. This implies that DWM in patients with only few brain metastases is unnecessarily damaged by conventional WBRT.