Use of natural language understanding to facilitate surgical de-escalation of axillary staging in patients with breast cancer

ABSTRACT Natural language understanding (NLU) may be particularly well-equipped for enhanced data capture from the electronic health record (EHR) given its examination of both content- and context-driven extraction. We developed and applied a NLU model to examine rates of pathological node positivity (pN+) and rates of lymphedema to determine if omission of routine axillary staging could be extended to younger patients with ER+/cN0 disease. We found that rates of pN+ and arm lymphedema were similar between patients 55-69yo and ≥70yo, with rates of lymphedema exceeding rates of pN+ for clinical stage T1c and smaller disease. Data from our NLU model suggest that omission of SLNB might be extended beyond Choosing Wisely recommendations, limited to those over 70 years old, to all postmenopausal women with early-stage ER+/cN0 disease. These data support the recently-reported SOUND trial results and provide additional granularity to facilitate surgical de-escalation.