CDK4/6-inhibitor or Chemotherapy, in Combination with ENDOcrine Therapy, for Advanced Breast Cancer / KENDO

Group sequential response adaptive randomized clinical trial of concomitant chemotherapy plus endocrine therapy versus cyclin-dependent Kinase 4/6 (CDK4/6) inhibitor plus endocrine therapy for advanced hormone receptor-positive, HER2-negative breast cancer Primary Objective: To compare the efficacy of concomitant CDK4/6 inhibitor plus endocrine therapy versus chemotherapy plus endocrine therapy (administered either concomitantly from the beginning or sequentially) in terms of progression-free survival (PFS). Secondary objectives: To compare between treatment arms: * quality of life (EORTC quality of life questionnaire(QLQ) QLQ -C30 and QLQ-BR23) * toxicity (CTCAE version 5.0) * time to treatment failure * best response rate * duration of response * clinical benefit rate * overall survival (OS) * PFS and clinical benefit with the subsequent line of treatment after cross-over: CDK4/6 inhibitors plus endocrine therapy in patients treated with chemotherapy plus endocrine therapy, chemotherapy (with or without endocrine therapy) in patients treated with CDK4/6 inhibitors plus endocrine therapy * correlative biomarkers of response to CDK4/6 inhibitors and chemotherapy: * tissue markers (on the primary tumor and / or metastatic tissue) * circulating markers (e.g. CTCs, ctDNA) The patients will be allocated according to block randomization until two events are observed in each arm, and then according to the time-to-event adaptation of the group sequential Doubly-adaptive Biased Coin Design (DBCD) whose allocation probabilities are computed at the end of the block randomization and after around 70% and 85% of the 150 maximum patients are enrolled during a 23 month period. At these last two (i.e. after 105 and 128 patients, respectively), interim analysis on efficacy will be carried out allowing for early stopping. At the end of the 16-month follow up, administrative censoring is introduced. Therefore, the total study duration is 39 months. Previous results on palbociclib and fulvestrant combination in second line and the characteristics of our target population lead us to assume a median PFS of 8 and 12 months for arm A and B, respectively. Under this scenario, for a sample size of at the most 150 patients, the proposed design strategy has led to a simulated power of 0.911 compared with a 0.717 one for the Complete Randomisation design.