• Bryan D Choi, Ganesh M Shankar, Ahilan Sivaganesan, Laura A Van Beaver, Kevin Oh, and John H Shin.
• Department of Neurosurgery, Massachusetts General Hospital, Boston, MA.
• Spine. 2018 Nov 1; 43 (21): E1274-E1280.

Study DesignA retrospective cohort study.ObjectiveWe performed a retrospective study of patients treated at our institution over the last 7 years to ascertain whether gene expression signatures in patients with advanced metastatic disease are associated with survival, when the disease has progressed to the spine.Summary Of Background DataSpinal metastases are a major cause of morbidity in patients with cancer. Molecular profiling strategies to characterize lung cancer have identified several genetic biomarkers that may lead to more effective prognostication.MethodsWe queried our institutional database for patients with metastatic lung cancer who underwent treatment for spinal metastases between 2011 and 2017. Genetic mutations in ALK, MET, ROS1, EGFR, and KRAS were chosen a priori for study based on availability by standard SNaPshot Lung Tumor Genotyping Analysis. Survival time was the duration between treatment for spinal metastases and death. Kaplan-Meier methods and the log-rank test were applied to characterize survival data.ResultsTwenty-six patients met criteria for inclusion. Median survival after surgery was 0.67 years. Median overall survival (OS) after diagnosis was 2.7 years. The presence of molecular abnormalities in patients with spinal metastases was significantly associated with increased OS (HR 0.38, 95% CI 0.12-1.22, P = 0.03).ConclusionMolecular phenotyping may provide prognostic insight in patients undergoing surgery for spinal metastases. This is the first study to demonstrate an association between genetic mutational data and OS in this patient population. It also represents the largest published series of such patients (n = 26) for which genetic mutational data are reported. Future models estimating survival for patients with spinal metastases may be enhanced by incorporation of molecular criteria.Level Of Evidence4.

31 keywords...

hide…