Surgery-Based Salvage for Locoregional Recurrent Salivary Gland Carcinoma: Comparative Survival Evidence from a Systematic Review and Meta-analysis

Objectives.: To compare survival outcomes between surgery-based salvage and non-surgical management in patients with locoregional recurrent salivary gland carcinoma (RSGC), and to explore whether surgery-based salvage is associated with more favorable outcomes in appropriately selected patients. Methods.: A search was performed in MEDLINE (Ovid), EMBASE, and the Cochrane Library. Observational studies were included that reported comparative outcomes for surgery-based salvage (resection ± adjuvant therapy) versus non-surgical management (re-irradiation, brachytherapy, systemic therapy, or palliative treatment) in locoregional recurrent disease. Patients with distant metastasis at recurrence were excluded from the pooled analysis. Primary outcomes were overall survival (OS) and disease-free survival (DFS). Hazard ratios (HRs) were calculated as non-surgical management versus surgery-based salvage and pooled using a random-effects model. Risk of bias was assessed using RoBANS 2.0. Results.: Five retrospective comparative studies (N = 163) met the inclusion criteria. Four studies contributed to a meta-analysis for OS (N = 153) and DFS (N = 145). In pooled observational analyses, non-surgical management showed higher hazard estimates for OS (pooled HR = 3.40; 95% confidence interval [CI], 1.17-9.84; p = 0.024) and DFS (pooled HR = 3.09; 95% CI, 1.21-7.91; p = 0.019) than surgery-based salvage. Heterogeneity was considerable for OS (I² = 80.1%) and moderate for DFS (I² = 56.4%). Most studies had moderate-to-high risk of bias, particularly for participant comparability and residual confounding. Conclusion.: Available comparative observational evidence suggests that surgery-based salvage may be associated with more favorable survival outcomes in appropriately selected patients with locoregional RSGC. However, the evidence base is small, retrospective, and clinically heterogeneous; therefore, these findings should be interpreted as exploratory rather than confirmatory, and treatment decisions should be individualized within a multidisciplinary approach.