BackgroundThe tyrosine kinase inhibitor pazopanib is used for treatment of sarcoma. Recent studies have suggested that the use of pazopanib may lead to the development of pneumothorax, an unexpected adverse effect in patients with sarcoma metastatic to the chest.MethodsWe conducted a retrospective case control study of patients with sarcoma with metastases to the chest with pneumothorax (cases) and without pneumothorax (controls). The control population was selected from tumor registry in a 1:4 (cases to controls) ratio. The primary outcome of interest was the association between pazopanib and pneumothorax risk in patients with sarcoma metastatic to the chest. Secondary objective was to evaluate risk factors for pneumothorax.ResultsWe identified 41 cases and 164 controls. Using purposeful selection method the odds of developing pneumothorax while being on pazopanib was not significant in univariate (p = .06) and multivariable analysis (p = .342). On univariate analysis risk factors of pneumothorax in patients with sarcoma were age, male sex, African American race, the presence of cavitary lung nodules/masses, and the presence of pleural-based nodules/masses. On multivariate analysis, only the presence of cavitary lung nodules/masses (P < .001) and the presence of pleural-based nodules/masses (P < .001) remained as risk factors for developing pneumothorax.ConclusionPazopanib does not increase the risk of pneumothorax in patients with sarcoma and evidence of metastatic disease to the chest. Presence of cavitary lung nodules/masses and the presence of pleural-based nodules/masses were found to be risk factors for pneumothorax.