Karim Fizazi 1, Gero Kramer 2, Jean-Christophe Eymard 3, Cora N Sternberg 4, Johann de Bono 5, Daniel Castellano 6, Bertrand Tombal 7, Christian Wülfing 8, Michael Liontos 9, Joan Carles 10, Roberto Iacovelli 11, Bohuslav Melichar 12, Ásgerður Sverrisdóttir 13, Christine Theodore 14, Susan Feyerabend 15, Carole Helissey 16, Stéphane Oudard 17, Gaetano Facchini 18, Elizabeth M Poole 19, Ayse Ozatilgan 19, Christine Geffriaud-Ricouard 20, Samira Bensfia 20, Ronald de Wit
Main idea: Since cabazitaxel improved pain response, time to pain progression, time to symptomatic skeletal events, clinicians and patients with metastatic castration-resistant prostate cancer can be reassured that cabazitaxel will not reduce quality of life when compared with treatment with a second androgen signalling-targeted inhibitor.
Background: In the CARD study, cabazitaxel significantly improved radiographic progression-free survival and overall survival versus abiraterone or enzalutamide in patients with metastatic castration-resistant prostate cancer previously treated with docetaxel and the alternative androgen signalling-targeted inhibitor. Here, we report the quality-of-life outcomes from the CARD study.
Methods: CARD was a randomised, multicentre, open-label, phase 4 study involving 62 clinical sites across 13 European countries. Patients (aged ≥18 years, Eastern Cooperative Oncology Group (ECOG) performance status ≤2) with confirmed metastatic castration-resistant prostate cancer were randomly assigned (1:1) by means of an interactive voice-web response system to receive cabazitaxel (25 mg/m2 intravenously every 3 weeks, 10 mg daily prednisone, and granulocyte colony-stimulating factor) versus abiraterone (1000 mg orally once daily plus 5 mg prednisone twice daily) or enzalutamide (160 mg orally daily). Stratification factors were ECOG performance status, time to disease progression on the previous androgen signalling-targeted inhibitor, and timing of the previous androgen signalling-targeted inhibitor. The primary endpoint was radiographic progression-free survival; here, we present more detailed analyses of pain (assessed using item 3 on the Brief Pain Inventory-Short Form [BPI-SF]) and symptomatic skeletal events, alongside preplanned patient-reported outcomes, assessed using the Functional Assessment of Cancer Therapy-Prostate (FACT-P) questionnaire and the EuroQoL-5 dimensions, 5 level scale (EQ-5D-5L). Efficacy analyses were done in the intention-to-treat population. Pain response was analysed in the intention-to-treat population with baseline and at least one post-baseline assessment of BPI-SF item 3, and patient-reported outcomes (PROs) were analysed in the intention-to-treat population with baseline and at least one post-baseline assessment of either FACT-P or EQ-5D-5L (PRO population). Analyses of skeletal-related events were also done in the intention-to-treat population. The CARD study is registered with ClinicalTrials.gov, NCT02485691, and is no longer enrolling.
Findings: Between Nov 17, 2015, and Nov 28, 2018, of 303 patients screened, 255 were randomly assigned to cabazitaxel (n=129) or abiraterone or enzalutamide (n=126). Median follow-up was 9·2 months (IQR 5·6-13·1). Pain response was observed in 51 (46%) of 111 patients with cabazitaxel and 21 (19%) of 109 patients with abiraterone or enzalutamide (p<0·0001). Median time to pain progression was not estimable (NE; 95% CI NE-NE) with cabazitaxel and 8·5 months (4·9-NE) with abiraterone or enzalutamide (hazard ratio [HR] 0·55, 95% CI 0·32-0·97; log-rank p=0·035). Median time to symptomatic skeletal events was NE (95% CI 20·0-NE) with cabazitaxel and 16·7 months (10·8-NE) with abiraterone or enzalutamide (HR 0·59, 95% CI 0·35-1·01; log-rank p=0·050). Median time to FACT-P total score deterioration was 14·8 months (95% CI 6·3-NE) with cabazitaxel and 8·9 months (6·3-NE) with abiraterone or enzalutamide (HR 0·72, 95% CI 0·44-1·20; log-rank p=0·21). There was a significant treatment effect seen in changes from baseline in EQ-5D-5L utility index score in favour of cabazitaxel over abiraterone or enzalutamide (p=0·030) but no difference between treatment groups for change from baseline in EQ-5D-5L visual analogue scale (p=0·060).