Lung cancer represents one of the leading causes of death with non–small cell lung cancer (NSCLC), accounting for approximately 85% of patient cases.
1,2 Brain metastases are frequent, with up to 50% of patients developing brain metastasis within the course of their disease.
2,3 Additionally, at least 10% of patients present with brain metastases at initial clinical presentation.
4 Historically, brain metastases have been regarded as a terminal disease stage harboring a poor prognosis with a median overall survival (OS) of 3 months with best supportive care or up to 6 months with whole brain radiation therapy and up to 8 months in selected patients undergoing surgical metastasis removal combined with adjuvant treatment.
1,4,5 In a subset of patients with good functional status and surgically accessible or symptomatic brain metastases, aggressive treatment—including craniotomy with neurosurgical resection followed by local irradiation and systemic treatment—is common practice.
4-6 However, there is a lack of specific and prospective randomized studies in the context of neurosurgically treated patients comparing different treatment options, including local therapy and systemic treatment using chemotherapy or immune checkpoint inhibitors (ICIs). More evidence is available for upfront local radiation and systemic treatment with either ICIs or small molecule inhibitors.
7-11 Retrospective data may help in further evaluating the role of local and systemic treatment modalities specifically in cohorts of surgically treated patients.
11,12 This retrospective single-center comparative effectiveness study describes clinical, radiological, and histological characteristics; prognostic factors; and OS in a large cohort of patients with NSCLC brain metastases who underwent craniotomy and brain metastasis resection. We compare outcomes in patients treated with radiation therapy and chemotherapy vs those receiving radiation therapy and ICI after surgical brain metastasis removal using a 1:1 propensity score matching approach to account for bias and heterogeneity among the patient groups.
13