BACKGROUNDConcurrent chemoradiotherapy has been the standard of care for locally advanced cervical cancer for over 20 years; however, 30-40% of treated patients have recurrence or progression within 5 years. Immune checkpoint inhibition has improved outcomes for patients with PD-L1 positive metastatic or recurrent cervical cancer. We assessed the benefit of adding durvalumab, a PD-L1 antibody, with and following chemoradiotherapy for locally advanced cervical cancer.METHODSThe CALLA randomised, double-blind, phase 3 trial included 105 hospitals across 15 countries. Patients aged at least 18 years with previously untreated locally advanced cervical cancer (adenocarcinoma, squamous, or adenosquamous; International Federation of Gynaecology and Obstetrics [FIGO] 2009 stage IB2-IIB lymph node positive, stage ≥III any lymph node status) and WHO or Eastern Cooperative Oncology Group performance status of 0 or 1 were randomly assigned (1:1) through an interactive web response system using a permuted block size of 4 to receive durvalumab (1500 mg intravenously once every 4 weeks) or placebo with and following chemoradiotherapy, for up to 24 cycles. Chemoradiotherapy included 45 Gy external beam radiotherapy at 5 fractions per week concurrent with intravenous cisplatin (40 mg/m2) or carboplatin (area under the concentration-time curve 2) once weekly for 5 weeks, followed by image-guided brachytherapy (high-dose rate, 27·5-30 Gy or low-dose/pulse-dose rate, 35-40 Gy). Randomisation was stratified by disease stage status (FIGO stage and node status) and geographical region. Chemoradiotherapy quality was continuously reviewed. The primary endpoint was progression-free survival, assessed by the investigator using Response Evaluation Criteria in Solid Tumors, version 1.1, in the intention-to-treat population. Safety was assessed in patients who received at least one dose of study treatment. This study is registered with ClinicalTrials.gov, NCT03830866.FINDINGSBetween Feb 15, 2019, and Dec 10, 2020, 770 women were randomly assigned (385 to durvalumab and 385 to placebo; median age 49 years [IQR 41-57]). Median follow-up was 18·5 months (IQR 13·2-21·5) in the durvalumab group and 18·4 months (13·2-23·7) in the placebo group. At data cutoff, median progression-free survival had not been reached (95% CI not reached-not reached) for either group (HR 0·84; 95% CI 0·65-1·08; p=0·17); 12-month progression-free survival was 76·0% (71·3-80·0) with durvalumab and 73·3% (68·4-77·5) with placebo. The most frequently reported grade 3-4 adverse events in both groups were anaemia (76 [20%] of 385 in the durvalumab group vs 56 [15%] of 384 in the placebo group) and decreased white blood cells (39 [10%] vs 49 [13%]). Serious adverse events occurred for 106 (28%) patients who received durvalumab and 89 (23%) patients who received placebo. There were five treatment-related deaths in the durvalumab group (one case each of urinary tract infection, blood loss anaemia, and pulmonary embolism related to chemoradiotherapy only; one case of endocrine disorder related to durvalumab only; and one case of sepsis related to both durvalumab and chemoradiotherapy). There was one treatment-related death in the placebo group (pneumonia related to chemoradiotherapy).INTERPRETATIONDurvalumab concurrent with chemoradiotherapy was well tolerated in participants with locally advanced cervical cancer, however it did not significantly improve progression-free survival in a biomarker unselected, all-comers population. Concurrent durvalumab plus chemoradiotherapy warrants further exploration in patients with high tumoral PD-L1 expression. Rigorous monitoring ensured high chemoradiotherapy compliance with advanced technology and allowed patients to receive optimal care.FUNDINGAstraZeneca.