Background: American Society for Gastrointestinal Endsocopy (ASGE) guidelines for assessing minimal competence in EUS are based on expert opinion and retrospective studies. Objective: To prospectively define learning curves in EUS among advanced endoscopy trainees (AETs). Design: Prospective trial. Setting: Three tertiary-care referral centers. Patients: AETs with no prior EUS experience. Intervention: AETs were evaluated by attending endosonographers at intervals of 10 EUS examinations (beginning at the 25th examination) during a 12-month training period. A standardized data collection form was used to grade examination of EUS anatomic stations and, when applicable, lesion of interest, accurate uTNM staging, wall layer origin of subepithelial lesions, and technical success with FNA. Main Outcome Measurements: Cumulative sum analysis was applied to assess competency and produce a learning curve for each trainee for overall performance and for each anatomic station. Acceptable and unacceptable failure rates of 10% and 20%, respectively, were used. Results: Five AETs were included, with a total of 1412 EUS examinations (AET1-225, T2-175, T3-402, T4-315, T5-295). Two AETs crossed the threshold for acceptable performance at cases number 255 and 295, two AETs showed a trend toward acceptable performance after 225 and 196 cases but needed ongoing training, and 1 AET demonstrated the need for ongoing training after 402 cases. Similar variable results were noted for individual stations. Limitations: Results from this study may not be generalizable to other centers' AETs. Conclusion: We observed substantial variability in achieving competency and a consistent need for more supervision in all AETs than current recommendations (150 cases). Future studies should focus on standardization of trainee performance, definition of competency, and widespread applicability of AET evaluation.