Recurrence and survival after robotic-assisted radical hysterectomy (RRH) for early-stage cervical cancer (CC): Experience may matter.

Abstract

5535 Background: The phase III LACC Trial found that minimally invasive surgery (MIS) / radical hysterectomy was inferior to open radical hysterectomy (ORH) with reduced disease-free survival (86% v 96.5%) and a higher disease-specific death rate (DSDR) (4.4% v 0.6%). We evaluated our experience with attention to the learning curve. Methods: Patients (pts) with early-stage CC (4/2007-12/2017) who underwent RRH with a uterine manipulator were evaluated in a contemporaneously maintained database. First 10 learning curve cases per surgeon (Group A) were compared to all subsequent cases (Group B). Inclusion criteria mirrored the LACC trial: > one-year follow-up, adenocarcinoma, adenosquamous, or squamous carcinoma, stage IA2 or IB1 using FIGO 2014 guidelines, and pathologic tumor size (TS) of 4 cm or less. Atypical histologies and lesions > 4 cm were excluded. Study parameters assessed included recurrence free survival (RFS), DSDR, and pattern of recurrence. Results: 144 RRH pts were identified and 90 met inclusion criteria with mean age of 45.6±14.3 years. Exclusions included stage 1A1 without LVSI (n = 13), atypical histology (n = 10), lost to follow-up (n = 13), and occult stage 1B2 (n = 18). 40 pts met Group A and 50 met Group B criteria. Median follow-up was 61±34.3 months (A = 71.5, B = 52.5). The 5-year RFS was 92% (95 CI ±4%) and the DSDR 5.5% (n = 5). There were 7 (7.8%) recurrences with median time to recurrence of 12±8.3 mos. Recurrence in Group A (n = 6, 15%) exceeded Group B (n = 1, 2%), p= 0.025. DSDR was 10% Group A v 2% B ( p= 0.184). The 4.5 yr RFS was 84.8% (95 CI ±7%) in Group A v 98% (95 CI ±3%) in Group B. There were no differences in risk factors for recurrence between A & B (TS > 2 cm, LN (+), adjuvant therapy (AT), and LVSI p> 0.05), except (+) vaginal margin status (A = 10% v B = 0%, p= 0.034). Three recurrences involved carcinomatosis, which may be insufflation related. All recurrent cases had TS > 2 cm and 5 received AT. Conclusions: In this study, recurrence of disease in early-stage CC clustered in the first 10 cases per surgeon and occurred in TS > 2 cm. This data suggests a possible learning curve effect and argues against a uterine manipulator cause. Carcinomatosis may be insufflation related, unique to MIS, and deserves further study.