The ABRAX trial is multicentre retrospective cohort study. Included are patients with negative LN in clinical staging, in whom LN involvement is detected intraoperatively. Completion or abandonment of planned cervical procedure stratifies the cohort in two subgroups in which oncological outcome and morbidity will be compared.

The investigators hypothesise that in patients with intraoperative LN involvement, a completion of radical hysterectomy or other cervical procedure does not improve oncological outcome of definitive chemoradiation.

The management of patients with intraoperative detection of lymph node (LN) involvement currently varies widely. Options include completing or abandoning radical hysterectomy, performing or abandoning pelvic lymph node dissection and even continuing with inframesenteric or infrarenal paraaortic lymph node dissection. The most significant aspect is the decision regarding the performance of radical hysterectomy, due to the high morbidity caused by combined treatment composed of radical parametrectomy and adjuvant radiotherapy in these patients - patients are referred to pelvic radiotherapy after distal ureters, bladder and rectum have been surgically dissected [1].

Data on the oncologic outcome of patients with LN involvement after radical hysterectomy and adjuvant radiotherapy are broadly available. Recent figures show five-year survival in stage IB at around 70-85 % [2,3]. In contrast, the data on patients in whom radical hysterectomy was abandoned due to intraoperative detection of LN involvement are scarce. Available literature mostly refers to small groups of cases with grossly involved LN detected during surgery [4-7].

The goal of this study is to obtain the best data available from an adequate number of patients treated by both types of management in the same period of time and to analyse the risks and benefits of the performance of radical hysterectomy if LN involvement is detected intraoperatively in spite of non-suspicious preoperative radiological assessment.

ABRAX is an international multicenter retrospective trial. Data of cervical cancer patients with preoperatively negative LN in whom nodal involvement is detected during operation will be evaluated. The protocol has been developed to be inclusive and reflect current clinical practice. All surgical approaches are eligible for the study. Lymph node infiltration can be detected either by intraoperative pathology assessment or by gross assessment (macroscopic suspicion has to be confirmed by the final histology). Sentinel or any other pelvic lymph node can be subject of intraoperative assessment. Any type of nodal metastasis (macrometastasis, micrometastasis ot isolated tumor cells) is considered a positive LN. All types of cervical procedures (such as conisation, simple hysterectomy, trachelectomy, any type radical hysterectomy or radical trachelectomy) are eligible for the trial and administration of neoadjuvant chemotherapy is not an exclusion criterion. The abandonment of planned cervical procedure depends solely on the decision of the surgeon and on local institutional guidelines. The completion or abandonment of cervical procedure stratifies the cohort in two subgroups.

The primary end-point is the progression free survival assessed separately in both subgroups. Prevalence of ≥ G2 treatment related morbidity (CTCAE), overall and pelvic progression free survival are the secondary end-points.

Oncological outcome will be stratified according to the prognostic parameters such as tumour size, number of involved LN, type of LN metastases, presence of LVSI (lymphovascular space invasion), tumor type, administration of neoadjuvant chemotherapy, performance of systematic pelvic lymphadenectomy.

Inclusion Criteria:

• Histologically confirmed squamous cell carcinoma, adenocarcinoma, adenosquamous carcinoma
• Stage pT1a - pT2b
• Patient referred for primary surgical treatment such as radical / simple hysterectomy or fertility-sparing procedure (FST) and LN staging
• Intraoperative detection of LN involvement (any type of metastasis):
• Macroscopic involvement = grossly involved lymph nodes (if confirmed by final pathology) OR Microscopic involvement = SLN / LN intraoperative pathologic evaluation (frozen section)
• Follow-up data available for ≥ 2 years
• Surgery performed between January 2005 and December 2015

Exclusion Criteria:

• Preoperative evidence of grossly involved LN
• Histologic subtypes other than those noted in the Inclusion criteria
• Negative pelvic LN
• LN involvement reported by the final histology but not detected during the surgery
• Unavailability of follow-up data

Consecutive participant sampling of patients with pT1a - pT2b cervical cancer and no evidence of LN involvement in preoperative imaging who were referred for primary surgery with curative intent and in whom nodal involvement was discovered intraoperatively.