Findings from a large analysis of real-life patient data presented at the ESMO Virtual Congress 2020 detailed patient characteristics and outlined the changes in patient outcome as well as diagnostic and initial treatment strategies in patients with newly diagnosed brain metastasis occurring over the course of time from 1986 to 2020. Dr Ariane Steindl of the Division of Oncology, Department of Medicine, Medical University of Vienna in Vienna, Austria presented data regarding the clinical characterisation of patients with brain metastases, including prognostic assessments in patients with brain metastasis.
Dr Steindl and colleagues reviewed the Vienna Brain Metastasis Registry to identify patients newly diagnosed with brain metastasis from 1986 to 2020. This study included 6001 of these patients.
The investigators found that the frequency of the primary tumour remained unchanged during the observation period. Lung cancer was identified as the primary tumour that most frequently was the source of brain metastasis in 50.5% of patients. Other sources of brain metastasis occurring at high frequency included breast cancer in 16.0% of patients, melanoma in 11.1%, colorectal cancer in 5.8% and renal cell carcinoma in 5.5% of cases. The primary tumour was not identified in 2.5% of cases, which were defined as unknown primary.
Brain metastasis in patients without neurological symptoms was increasingly diagnosed by screening
This analysis uncovered a rising incidence of brain metastasis that was diagnosed by screening in the absence of neurological symptoms. This incidence was 15.3% in the period from 1986 to 1999, which rose to 25.5% from 2000 to 2009, and was recorded as 31.4% during the period from 2010 to 2020 (p < 0.001).
Patient outcome was seen to improve in latter years and prognostic assessments significantly associated with overall survival
The outcome of patients newly diagnosed with brain metastasis improved during the later decades studied; between the years 1986 to 1999 patients had a poor median overall survival of just 5 months, which increased to 7 months in patients diagnosed between 2000-2009 or 2010-2020 (p < 0.005).
In patients with brain metastasis derived from lung cancer, both the diagnostic-specific graded prognostic assessment (DS-GPA: hazard ratio [HR] 1.42; p < 0.001), and the GPA for lung cancer using molecular markers (Lung-molGPA: HR 1.67; p < 0.001) were significantly associated with overall survival.
The cohort of patients with brain metastasis derived from melanoma showed no statistically significant association with overall survival using molecular markers (Melanoma-molGPA: HR 1.46; p = 0.011), class 2 patients (median OS: 14 months) patients presented with a longer median overall survival than class 1 patients (median OS: 19 months).
This analysis reflected a shift in treatment strategies over time period
The initial approaches to treatment of brain metastasis were found to change significantly over the duration of time studied; from 1986 to 1999, fewer patients (36.8%) received stereotactic radiosurgery (SRS) as compared with 56.3% of patients during the interval from 2000 to 2009 (p < 0.001).
Systemic treatment was also shown to increase from 0.4% of patients treated between 1986 to 1999 to 2.4% of patients from 2010 to 2020. On the contrary, the use of neurosurgical resection as initial treatment of brain metastasis decreased from 39.9% (1986-1999) to 20.4% of patients from 2010 to 2020.
The authors noted that these data are may aid in guiding individualised treatment decisions and clinical trial development.
According to this analysis, initial treatment patterns have changed from 1986 to 2020 in patients with brain metastasis towards increased use of SRS and systemic therapies, whereas the use of neurosurgical resection has declined. The duration of overall survival has improved in more recent years.
In addition, the investigators commented that their data confirmed refinement of brain metastasis-specific prognostic scores by the incorporation of molecular data in lung cancer patients.
No external funding was disclosed for this study.
363O – Steindl A, Kreminger J, Moor E, et al. Clinical characterization of a real-life cohort of 6001 patients with brain metastases from solid cancers treated between 1986-2020. ESMO Virtual Congress 2020.