B rain metastases are quite common in cancer patients and their incidence has increased over the last two decades. That may be due to both improvements in diagnostic methods and longer survival of metastatic patients. The most common primary site for patients with brain metastases is the lung and approximately 20% of patients with non-small-cell lung cancer (NSCLC) will develop brain metastases at some point in the course of their disease.Most patients with brain metastases are symptomatic at diagnosis and their presence should be suspected in any patient with a history of lung cancer who presents with a new neurological sign or symptom. Magnetic resonance imaging is the test of choice for diagnosing brain metastases because of its superior resolution and accuracy compared with computed tomography.Overall, the outcome for NSCLC patients who develop brain metastases remains poor, with an average survival of 3-6 months. A number of prognostic indexes have been proposed and validated with a view to predicting more accurately the prognosis of patients with brain metastases [1]. Essentially, all these index scores help us to better select patients for the different options available to treat brain metastases: surgery, stereotactic radiosurgery, whole brain radiotherapy, chemotherapy and corticosteroids. However, these index scores are less than perfect and more precise prognostic tools are clearly needed.For patients with single or 1-3 brain metastases as the only site of disease, or with systemic disease which can be radically treated, the main objective is to completely eradicate or control the brain lesion. Surgical resection is usually employed for potentially resectable lesions, which are symptomatic and/or affecting silent areas of the brain as surgery can provide immediate relief in patients with a good baseline performance status. Another advantage of surgical resection is the pathologic confi rmation of the malignant nature of the lesion. Postoperative radiotherapy has demonstrated a decrease in the recurrence of brain metastases, a decrease in the mortality due to neurological deterioration and a delay in neurologic clinical deterioration [2,3].Radiosurgery has emerged as a non-invasive option to treat patients with 1-3 brain metastases, providing similar local control rates to surgery [4][5][6]. However, no randomised trial has compared the two approaches. Stereotactic radiosurgery is preferable to open surgery in patients with brain metastases with lesions no greater than 30-35 mm who have no symptomatic oedema or mass effect and those unlikely to achieve complete surgical resection. Whole brain radiotherapy is the standard of care in patients with multiple brain metastases. Several randomised trials have explored different doses and fractions; a total of 30 Gy in 10 fractions is considered appropriate in the majority of patients [7]. In patients with brain metastases and symptomatic extracranial metastatic disease, starting treatment with systemic chemotherapy and delaying whole brain radiotherapy could ...