EORTC 2019.

Study characteristics
Methods	Randomised phase III study (EORTC55994). Multicentre, multinational (Austria, Belgium, France, Italy, Netherlands, Portugal, Spain, the UK). Between May 2002 and June 2014 620 women with FIGO Stage IB2–IIB were randomised to NACT + surgery (311 women) or standard CCRT (309 women). Estimated primary completion date July 2019.
Participants	Women with FIGO Stage IB2, IIA > 4 cm or IIB cervical cancer. Histologically confirmed cervical cancer, including the following subtypes: squamous cell carcinoma; adenocarcinoma (excluding small cell, clear cell, and other rare variants of the classical adenocarcinoma). Age Range: 18–75 years Performance status WHO 0–2
Interventions	NACT + surgery Women received neoadjuvant cisplatin‐based chemotherapy on day 1. Treatment repeats every 21 days. Within 6 weeks after the last chemotherapy course, and with a cumulative minimum of 225 mg/m2, women underwent a type III‐V Piver‐Rutledge RH. Women with positive lymph nodes or tumour invasion into the parametria or < 5 mm from the resection borders after surgery receive standard adjuvant external‐beam radiotherapy once daily, 5 days/week, for 5.0–5.6 weeks (25–28 treatment days) followed by external boost radiotherapy or brachytherapy for 1 or 2 days. CCRT Radiotherapy 45–50 Gy + boost concurrent with weekly cisplatin chemotherapy (40 mg/m2/week). Adjuvant hysterectomy allowed, but not recommended, in cases of histologically confirmed residual tumour.
Outcomes	Primary endpoint OS at 5 years Secondary endpoints OS PFS Toxicity Quality of life
Notes	Results Median follow‐up was 8.2 years (95% CI 7.8 to 8.6) and similar between arms. Total of 191 deaths (31%) occurred. Age, stage and histological cell type were balanced between arms. Protocol treatment was completed in 459 (74%) women (71% for NACT + surgery and 82% for concurrent chemotherapy). In NACT + surgery arm, 238 (76%) women underwent surgery. Main reasons for not having surgery as per‐protocol were toxicity (25/74, 34%), progressive disease (18/74, 24%) and insufficient response to NACT (12/74, 16%). 198 SAEs occurred: 145 in the NACT + surgery arm vs 53 in the concurrent chemotherapy arm. Within the group of SAEs, there were 109 serious adverse reactions in the NACT + surgery arm and 35 in the concurrent chemotherapy arm. Nearly all were chemotherapy related. Protocol treatment was discontinued in 41 women due to toxicity (6.5%). In the NACT + surgery arm, surgery in 23 women was abandoned due to NACT‐related toxicity. Most frequent reported were nausea and vomiting, infections, metabolism disturbance, and renal and urinary disorders. There were grade 3/4 complications related to surgery in the following number of women: 8 (3.3%) bleeding, 10 (4.2%) operative lesions to ureter or bladder, 3 (1.2%) fistula, 7 (2.9%) others (sepsis, urinary tract infection and wound dehiscence). Additional radiotherapy given to 113 (36.3%) women in the NACT + surgery arm. Additional surgery was performed in 9 (2.9%) women in the concurrent chemotherapy arm. Short‐term severe adverse events (≥ grade 3) occurred more frequently in the NACT + surgery arm (35%) than in the concurrent chemotherapy arm (21%) (P < 0.001). 5‐year OS was 72% in the NACT + surgery arm and 76% in the concurrent chemotherapy arm (difference 4.0%, 95% CI –4% to 12%; HR 0.87, 95% CI 0.65 to 0.15; P = 0.332). 5‐year DFS in the NACT + surgery group was 56.9% compared with 65.6% in the CCRT arm, mostly for Stage IIB disease. However. the authors noted that the first imaging measurement occurred systematically earlier in the NACT + surgery arm than in the CCRT arm. Conclusions: these preliminary results revealed no difference in 5‐year OS between NACT + surgery and CCRT, indicating that quality of life and long‐term toxicity are important to decide optimal treatment. Overall toxicity was acceptable, occurred more frequently in the NACT + surgery arm and was mainly related to NACT. The final results were expected by April 2019, including long‐term toxicity and treatment effect across prognostic factors. Clinical trial information: NCT00039338.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Blinding not possible but should not have introduced bias for objectives outcomes such as OS and PFS.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Unclear.
Selective reporting (reporting bias)	Unclear risk	In abstract form only, so unclear.
Other bias	Unclear risk	Insufficient information to permit judgement.