Intraoperative evaluation of sentinel lymph nodes in patients with breast cancer treated with systemic neoadjuvant therapy: a systematic review and meta-analysis of diagnostic studies

Background Frozen section (FS) and touch imprint (TI) are common intraoperative evaluation (IOE) techniques for sentinel lymph nodes (SLNs) in breast cancer surgery. Their accuracy in patients receiving neoadjuvant systemic therapy (NST) remains variable. Objective To summarise evidence on the diagnostic accuracy of FS and TI in the NST context. Methods A systematic search of PubMed, Embase, Scopus and Web of Science was conducted through April 2024 for studies evaluating FS and/or TI in SLNs of breast cancer patients treated with NST. Meta-analysis was performed using the logit function, and Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies-2 and Grading of Recommendation, Assessment, Development and Evaluation were employed to assess risk of bias and evidence certainty. Results 20 studies were included. At the lymph node level, TI demonstrated pooled sensitivity and specificity of 0.54 and 1.00, respectively, while FS achieved 0.85 and 0.99. At the patient level, TI showed sensitivity and specificity of 0.72 and 1.00, while FS reached 0.82 and 0.98. Combined, FS and TI presented pooled sensitivity and specificity of 0.59 and 1.00 at the patient level. Risk of bias was frequently unclear, and the certainty of evidence for both techniques ranged from low to very low. Conclusion FS exhibits higher sensitivity and specificity than TI at both lymph node and patient levels, but evidence certainty remains limited. Further prospective, blinded studies are needed to validate these findings and optimise IOE methods for NST-treated patients.