Cediranib with weekly paclitaxel or olaparib versus weekly paclitaxel for advanced or recurrent endometrial cancer (COPELIA): a multicentre, open-label, randomised, phase 2 trial in the UK.

Background: Patients with advanced or recurrent endometrial cancer have poor survival outcomes because effective treatment options remain limited. We investigated the efficacy and safety of cediranib plus weekly paclitaxel or olaparib versus weekly paclitaxel alone after prior platinum-based chemotherapy. Methods: COPELIA was an open-label, randomised, phase 2 trial undertaken at 15 centres in the United Kingdom. Eligible participants were aged 16 years or older with histologically confirmed endometrial cancer, Eastern Cooperative Oncology Group performance status 0-1, a life expectancy greater than 16 weeks, and at least one previous line of platinum-based chemotherapy. Patients were randomly assigned to paclitaxel 80 mg/m Findings: Between May 1, 2018, and January 11, 2022, 124 patients were enrolled and randomised to arm 1 (n = 41), arm 2 (n = 41), or arm 3 (n = 42). Median follow-up in arms 1, 2, and 3 was 34.4 months (interquartile range [IQR] 11.6-37.6), 26.3 months (11.8-not evaluable), and 23.7 months (15.0-not evaluable), respectively. PFS at 3 months was significantly higher in arm 2 than in arm 1 (73.2% versus 48.8%; adjusted odds ratio [aOR] 3.2, lower limit of one-sided 80% confidence interval [CI] 2.1; p = 0.01). RECIST response was also higher in arm 2 versus arm 1 (56.4% versus 28.2%; aOR 5.7, 95% CI 1.8-17.6; p < 0.001). No differences were observed between arm 3 and arm 1 for PFS at 3 months (aOR 1.0, lower limit of one-sided 80% CI 0.71; p = 0.46) or RECIST response (aOR 0.8, 95% CI 0.3-2.6; p = 0.73). Median OS was 12.8 months (95% CI 7.3-16.8) in arm 1, 18.1 months (9.5-26.4; log-rank p = 0.42 versus arm 1) in arm 2, and 13.9 months (11.2-18.3; log-rank p = 0.86 versus arm 1) in arm 3. There was no difference in PFS at 6 months and median PFS between arm 1 and arm 2 or arm 3. Grade 3 adverse events occurring in ≥10% of patients in arm 2 included hypertension (15%), neutropenia (12%), and diarrhoea (10%). Quality-of-life differences favoured arm 1 over arm 2 for diarrhoea and gastrointestinal symptoms (p < 0.001). In multivariable analyses, cediranib-treated patients who achieved a Tie2-defined vascular response had significantly improved PFS (hazard ratio 0.54, 95% CI 0.33-0.88; p = 0.014). Interpretation: Paclitaxel plus cediranib improved 3-month PFS and RECIST response compared with paclitaxel alone in advanced or recurrent endometrial cancer. However, this effect diminished over time; with no difference in PFS observed at 6 months, nor median PFS or median OS. Plasma Tie2 identified cediranib-treated patients with improved PFS. Further evaluation of paclitaxel plus cediranib in patients previously treated with platinum-based chemotherapy and immunotherapy is warranted, ideally incorporating plasma Tie2 as a response biomarker to guide treatment continuation. Funding: AstraZeneca.