Main Findings
Our results show that a novice robotic surgeon, who was trained in accordance with the SERGS structured learning curriculum and started under proctor supervision, can pass through a learning phase without compromising the surgical and oncological outcomes of cervical cancer patients treated with robot-assisted surgery. The expanded RA-CUSUM analysis (Figure 1 ), used to assess the learning curve, revealed no new peak in the last 61 cases (group 3), which indicates the observed recurrence rate remained as expected after introduction of the novice surgeon. Also, the mean operation time remained equal compared to the preceding experienced phase (group 2). Furthermore, short-term surgical outcomes, defined as severe (grade 3-5) adverse events occurring intraoperatively or within the first 42 days postoperatively, were not significantly increased. Using Kaplan-Meier curves (Figure 2 ), the 5-year DFS, DSS and OS during introduction of the novice (group 3) showed to be lower than in the experienced phase (group 2) and higher than in the early learning phase (group 1), but no significant difference in survival outcomes existed. Although based on a single surgical team, our results suggest that structured training (including proctoring) performs better than unstructured training (consisting of case observation, box training, and limited proctoring), which was available for novices during the initial phase (group 1).8