Prognostic Scores in Patients with Brain Metastases from Non-small Cell Lung Cancer

Introduction: Prognostic scores might be useful tools in both clinical practice and clinical trials, where they can be used as stratification parameter. The five available scores for the general population of patients with brain metastases have never been tested specifically in patients with non-small cell lung cancer (NSCLC). No comparison to the two NSCLC-specific scores has been made either. Hence, it remains unclear which score is most appropriate for these patients. Methods: We evaluated seven previously published prognostic scores in a group of 183 patients with brain metastases from NSCLC. All patients had been treated outside of clinical studies with whole brain radiotherapy with or without radiosurgery or surgical resection. Results: The three scores with significant prognostic impact were the recursive partitioning analysis classes, the basic score for brain metastases, and the graded prognostic assessment (GPA) score. All three score systems were developed in a general population of patients with brain metastases. The GPA score performed better than the others. In this four-tiered system, the 1-year survival was 43, 20, 8, and 0%, respectively. Conclusion: Three prognostic scoring models describe the survival of patients with brain metastases from NSCLC to a satisfactory degree. In the Current patient population, GPA performs better than the others. The two NSCLC-specific scores were developed in radiosurgery-treated populations, and their prognostic impact in a general NSCLC population seems limited.