Lymph node surgery and CDK4/6 inhibitors in early breast cancer: a pooled analysis from five randomised trials.

Abstract

Omitting sentinel lymph node biopsy (SLNB) and completion axillary lymph node dissection (cALND) in early-stage breast cancer reduces morbidity without compromising overall survival or recurrence rates. However, reduced axillary surgery restricts staging information, impacting eligibility for CDK4/6 inhibitors. We evaluated the impact of omitting SLNB or cALND on CDK4/6 inhibitor eligibility and the associated benefits and harms. We used data from the SOUND, INSEMA, SENOMAC, monarchE, and NATALEE randomised trials. These five trials evaluated the safety of de-escalated lymph node surgery and adjuvant CDK4/6 inhibitors in patients with early breast cancer. The primary outcome was the number needed to diagnose and treat with CDK4/6 inhibitor-ie, number of axillary surgeries leading to adjuvant CDK4/6 inhibitor use, preventing one invasive disease-free survival event, distant disease-free survival event, or overall survival event at 5 years. Between Jan 31, 2015, and Dec 31, 2021, 19 541 patients were randomly assigned and analysed within five trials, of whom 19 475 (99·7%) were women and 66 (0·3%) were men; median follow-up was 57·2 months (IQR 44·1-81·6). For SLNB and ribociclib, the number needed to diagnose and treat for one prevented event was 123 for invasive disease-free survival, 129 for distant disease-free survival, and 345 for overall survival. For cALND and abemaciclib, the number needed to diagnose and treat for one prevented event was 106 invasive disease-free survival, 119 for distant disease-free survival, and 807 for overall survival. Performing SLNB or cALND solely to determine eligibility for CDK4/6 inhibition requires very high number needed to diagnose and treat, adds substantial morbidity, and is costly, with only marginal overall survival benefit. None.