University Medical Centre Utrecht, Netherlands

Introduction Previous learning curve studies are focused on short term surgical outcomes of robot-assisted surgery. We are the first to study the influence of learning curve on recurrence rate and quantify its impact on survival in patients with early stage cervical cancer using a validated analysis.

Methods We included all patients with early stage cervical cancer treated consecutively by one tertiary surgical team with robot-assisted laparoscopy (2007 – 2017). To establish the number of procedures needed to ascertain oncological proficiency, we performed a risk-adjusted cumulative sum (RA-CUSUM) analysis. The impact of the learning curve on oncological outcomes was quantified with Kaplan-Meier survival analysis.

Results In total, 165 patients received radical robot-assisted laparoscopy and had a median follow-up of 57 months (range 3–132 months). Multivariate RA-CUSUM analysis demonstrated a learning phase of 61 procedures. The recurrence rate decreased significantly with increasing experience in robot-assisted laparoscopy. No significant differences in baseline characteristics were observed between patients treated during the learning phase (n=61) and experienced phase (n=104). The 5-year disease-free, disease-specific and overall survival significantly increased in patients treated after the learning phase; 80.2% to 91.1% (P=0.040), 84.7% to 97.7% (P=0.002) and 84.7% to 94.8% (P=0.018), respectively.

Conclusions/Implications In this single institution study we found a learning phase for robot-assisted laparoscopy in cervical cancer of 61 procedures with significant impact on the survival rates. With our analysis we aim to encourage others performing robot-assisted radical surgery for cervical cancer to assess their own learning curves.