Outcomes of Retreatment for Local Recurrence After Surgery for Metastatic Spine Cancer: A Multicenter Study.

Abstract

Local recurrence (LR) after surgery for metastatic spine cancer (MSC) is a challenging complication, and evidence guiding optimal retreatment strategies remains limited. To evaluate prognosis following retreatment for LR after surgery for MSC, compare treatment outcomes based on retreatment modality, and identify risk factors associated with retreatment failure. A multicenter retrospective cohort study. Ninety-nine patients from four tertiary hospitals who developed symptomatic LR after surgery for MSC and underwent reoperation or radiation therapy (RT). Motor grade, pain visual analog scale (VAS) score, ambulatory function, retreatment failure rate, and overall survival after LR diagnosis were assessed. Patients were grouped based on retreatment modality into reoperation (n=36), RT (n=38), and conservative management (n=25) groups. Conservative management included symptomatic care for LR without surgery or RT, with or without systemic treatments such as chemotherapy or hormonal therapy. Retreatment failure was defined as symptom progression despite treatment or recurrence after initial improvement. Clinical outcomes were compared among the groups, and logistic regression was performed to identify factors associated with retreatment failure. Reoperation significantly improved motor grade (p=0.041) and pain VAS score (p=0.002), whereas RT and conservative treatment showed no significant improvement. Ambulatory status decreased significantly in the RT group (p=0.031) but was preserved in the reoperation group. Retreatment failure occurred in 41.9% of patients (15/36 after reoperation, 16/38 after RT; p=0.414) and was associated with increased pain (p=0.042) and reduced ambulation (p=0.012). The only significant predictor of retreatment failure was a shorter interval between initial surgery and LR (odds ratio, 0.953; p=0.045). A cutoff of 12.5 months was predictive of retreatment failure (area under the curve, 0.713; p=0.056). Only reoperation resulted in significant functional improvement after retreatment for LR. RT was associated with decreased ambulatory function and did not provide significant symptom relief. Retreatment failure was common and associated with increased pain and diminished function. A significant predictor of retreatment failure was a shorter interval-less than 12.5 months-between the initial surgery and LR. These findings highlight the importance of patient selection and the limited utility of retreatment in aggressive or early-recurrent cases.