We developed objective definitions of extreme phenotypes of obstructive sleep apnea (OSA) using a multivariate approach, and demonstrate their utility for identifying characteristics that confer predisposition towards or protection against OSA in a new prospective sample.In a large international sample, we calculated race-specific liability scores from a weighted logistic regression including age, gender and body mass index (BMI). Extreme Cases were defined as individuals with an apnea-hypopnea index (AHI) ≥30 events/hour, but low likelihood of OSA based on age, gender and BMI (liability scores >90th percentile). Similarly, Extreme Controls were individuals with AHI<5, but high likelihood of OSA (liability scores <10th percentile). Definitions were applied to a prospective sample from the Sleep Apnea Global Interdisciplinary Consortium (SAGIC) and differences in photography-based craniofacial and intraoral phenotypes evaluated.Retrospective data included 81,338 individuals. We identified 4,168 Extreme Cases and 1,432 Extreme Controls using liability scores. Extreme Cases were younger (43.1±14.7 years), overweight (28.6±6.8 kg/m2), and predominantly female (71.1%). Extreme Controls were older (53.8±14.1 years), obese (34.0±8.1 kg/m2), and predominantly male (65.8%). These objective definitions identified 29 Extreme Cases and 87 Extreme Controls among 1,424 SAGIC participants with photography-based phenotyping; comparisons suggest greater cervicomental angle increases risk for OSA in the absence of clinical risk factors and smaller facial widths are protective in the presence of clinical risk factors.This objective definition can be applied in sleep centers throughout the world to consistently define OSA extreme phenotypes for future studies on genetic, anatomic and physiological pathways to OSA.