This study used cone-beam computed tomography (CBCT) via voxel-based superimposition to evaluate lip and incisor changes after orthodontic treatment with four premolar extractions (Ext) versus nonextraction (Non-Ext) among African American (AA) and White (W) patients. A total of 240 CBCTs of 120 adolescent orthodontic patients with Class I skeletal/dental relationships were included. Patients were initially divided according to treatment, and then each group was subdivided according to patients' ethnicity (Ext/W = 30, Ext/AA = 30, Non-Ext/W = 30, and Non-Ext/AA = 30). CBCTs were imported into Invivo6 for voxel-based superimposition. Lip and incisor measurements were recorded. Independent t tests and two-way ANOVA were used for statistical assessment. The Non-Ext/W group had a greater increase in all outcome variables compared with the Non-Ext/AA group, with a significant change in volume and position of upper (UL) and lower lips (LL), inclination of upper (U1) and lower incisors (L1), and position of U1. The Ext/AA group had a greater decrease in all measured outcomes compared with the Ext/W group, with significant change in inclination of U1 and L1, and position and volume of LL. W patients had more crowding than AA patients in both treatment approaches. Retraction ratios of 6.5:1 and 2.1:1 were recorded between the U1 and UL positions, and L1 and LL positions, respectively. There were no direct interaction effects between ethnicity and treatment, nor were there any significant effects of ethnicity after controlling for the covariates. Ethnicity alone has no impact on incisor and lip position after treatment. However, ethnicity in the form of initial presentation of malocclusion can have a significant influence.