Purpose <p>Natural language processing (NLP, artificial intelligence) can enable automated identification of records in large datasets. The purpose of this study was to evaluate the feasibility of NLP in identifying breast cancer-associated lung metastases and to understand the clinical characteristics and challenges of this common site of breast cancer recurrence.</p> Methods <p>NLP was applied to a large dataset of institutional pathology reports at an academic center to identify patients with pathologically confirmed breast cancer-associated lung metastases seen between 3/2012 and 5/2019. Chart review was conducted to confirm breast cancer-associated lung metastases and ascertain clinical and pathological features.</p> Results <p>Altogether, NLP identified 32 patients with pathology reports describing breast cancer-associated lung metastases from a pool of approximately 91,000 records. There was pathologic confirmation from lung biopsy tissue in the majority of cases (75%; <i>n</i> = 24) and from pleural fluid specimens (25% <i>n</i> = 8) on the remainder. After this dataset was defined using NLP, we were able to analyze clinical and pathologic features of the breast cancer-associated lung metastases.</p> Conclusions <p>NLP can be applied to identify organ-specific metastases from pathology reports, such as breast cancer-associated lung metastases as done here, which can then facilitate observational, translational, and clinical research to characterize and address challenges posed by this common site of breast cancer recurrence. This cohort of patients highlights the potential application of NLP for disease characterization and clinical research in oncology.</p>