First‐line CDK4/6 inhibitor‐based combinations for HR+/HER2– advanced breast cancer: A Bayesian network meta‐analysis

• Published in: Journal of evidence-based medicine 2024-03, Vol.17 (1), p.106-118
• Main Authors: Guo, Xianan, Zhou, Yunxiang, Zhang, Kun, Lu, Wei, Zhong, Xi, Wu, Shijie, Shen, Lu, Chen, Huihui, Chen, Yiding
• Format: Article
• Language: English
• Subjects: Aminopyridines; Benzimidazoles; Breast cancer; Breast Neoplasms - drug therapy; CDK4/6 inhibitors; Chemotherapy; Clinical outcomes; Cyclin-Dependent Kinase 4 - therapeutic use; Endocrine therapy; Female; Fulvestrant - therapeutic use; HR+/HER2− advanced breast cancer; Humans; Inhibitor drugs; Network Meta-Analysis; Protein Kinase Inhibitors - therapeutic use; Purines; Side effects
• ISSN: 1756-5391; 1756-5383; 1756-5391
• DOI: 10.1111/jebm.12571

Background International guidelines recommend cyclin‐dependent kinase 4/6 inhibitor (CDK4/6i)‐based first‐line therapy for hormone receptor‐positive, human epidermal growth factor receptor 2‐negative (HR+/HER2−) advanced breast cancer (ABC). However, direct drug comparisons are lacking. We aimed to identify the most effective and safe therapy through network meta‐analysis (NMA). Methods We searched PubMed, Embase, Web of Science, Cochrane Central Register of Controlled Trials, and OpenGrey up to September 30, 2023. Eligible studies included randomized controlled trials (RCTs) assessing endocrine therapy alone or in combination with CDK4/6i as first‐line endocrine treatment for HR+/HER2− ABC patients. The hazard ratios for progression‐free survival (PFS) and overall survival (OS) and relative risks for objective response rate and adverse events (AEs) were available in selected trials. We performed a Bayesian NMA following PRISMA guidelines. Results Thirteen RCTs, involving 10 treatments, were included. Most studies were at low risk of bias. Regarding PFS, ribociclib+fulvestrant ranked first with a surface under the cumulative ranking curve (SUCRA) of 85.0%, followed by dalpiciclib+nonsteroidal aromatase inhibitor (NSAI) (SUCRA = 78.9%). Considering OS, the top three ranked treatments were ribociclib+fulvestrant (SUCRA = 94.1%), abemaciclib+NSAI (SUCRA = 69.9%), and ribociclib+NSAI (SUCRA = 68.5%). Out of four CDK4/6is, ribociclib minimized the grade 3/4 AEs, while dalpiciclib demonstrated the worst safety. Publication bias could not be ignored in our analyses, and the certainty of evidence was downgraded primarily due to imprecision. Conclusions Ribociclib+fulvestrant probably represents the best option in a first‐line setting. When combined with NSAI, dalpiciclib likely showed the best efficacy but the worst safety. Abemaciclib+NSAI and ribociclib+NSAI could also be promising treatments, while palbociclib presented inferiority. (PROSPERO Registration No. CRD42022370271)