966 SENTIX – accuracy of preoperative local staging in the sentix trial (CEEGOG-CX01; ENGOT-CX2; NCT02494063)

Introduction/Background* The SENTIX is a prospective cohort international study on sentinel lymph node (SLN) biopsy without pelvic lymph node dissection (PLND) in patients with early-stage cervical cancer. The primary end point is a recurrence rate at 24 months´ follow-up after the surgery. Either magnetic resonance imaging (MRI) or expert ultrasound (EUS) was mandatory as a preoperative staging method. The aim of this study is to report the accuracy of preoperative local staging. Methodology Forty-seven sites from 18 countries participated in the study. Patients with stages T1a1/LVSI+ – T1b1 (FIGO 2009), common histological types and no suspicious lymph nodes on imaging were eligible. Patients were excluded from further study if SLN were not detected on both sides and if SLN was positive on frozen section histological evaluation. Compared were results from preoperative imaging with final pathology reports. Result(s)* From May 2016 to October 2020, 733 registered patients underwent surgery, 132 were excluded intraoperatively, data from 708 were analysed in this study. Patients’ characteristics are in table 1. Out of 90 patients clinically staged as 1A tumours, 42 (46.7%) were upstaged to IB1 (86% ≤ 2 cm, 14% 2-4cm, 0% > 4cm); 76.3% had conisation as diagnostic procedure. Fourteen out of 547 preoperatively IB1 tumours (2.6%) were upstaged to IB2 > 4cm. Analogously 33 patients (6%) with IB tumours were downstaged to IA. Preoperatively unrecognized parametrial involvement was found by pathology only in 22 out of 637 patients (3.5%). EUS and MRI were used equally in the study (53.5% vs 56.1%), both were comparable in the accuracy of tumour size measurement (2 cm size categories shift in stage IB) (p=1.000) and in the failure to detect parametrial involvement (2.9% vs 4.0%) (p=0.535). Chart 1. Conclusion* Clinical staging with EUS and MRI failed to detect positive parametria only in 3.5% of patients in the Sentix trial. Upstaging from IA tumours was frequent, mostly after previous conization. Only 2.6% of patients were upstaged to IB2 tumours >4 cm (IB3 FIGO 2018). Both EUS and MRI were equally reliable in tumour size and parametrial involvement assessment.