Tumor Budding Significance as a Biomarker for Clinicopathology and Prognostic Evaluation in Gynecological Malignancy: A Bayesian Meta-Analysis

Abstract

OBJECTIVE: Tumor budding (TB) has been recommended as a marker for prognosis and therapeutic decision-making in various types of cancer, yet it has not been comprehensively studied in gynecological malignancies. This study aimed to evaluate the relationship between TB and clinicopathological features, as well as prognosis, in patients with gynecological malignancies, using a Bayesian meta-analysis design. METHODS: The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 protocol was followed. A systematic literature search was conducted using PubMed, ScienceDirect, and Cochrane databases. A meta-analysis was performed to assess the relationship between TB and clinicopathological parameters using odds ratios (ORs). Prognostic outcomes were analyzed using hazard ratios (HRs). Specific Bayesian priors were applied to each variable. Data analysis was conducted using R (version 4.4.0). RESULTS: Eighteen cohort studies (n = 3,320) involving patients with cervical and endometrial cancer were included. Bayesian meta-analysis showed that TB was significantly associated with clinicopathological parameters, specifically cancer stage (OR=2.91; 95%CrI: 1.86-4.41; prediction interval (PI) OR=2.92; 95%CrI: 0.82-9.92; τ2=0.53), grading (OR=5.00; 95%CrI: 2.83-8.76; PI OR=5.00; 95% CrI: 0.89-27.87; τ2=0.77), nodal involvement (OR=3.63; 95%CrI: 2.41-5.47; PI OR=3.63; 95%CrI: 0.85-15.52; τ2=0.66), and lymphovascular invasion (LVI) (OR=4.22; 95%CrI: 2.52-6.92; PI OR=4.22; 95%CrI: 0.63-27.80; τ2=0.89). Overall survival (OS) showed an HR of 2.14 (95%CrI: 1.27-3.63; PI HR=2.14; 95%CrI: 0.83-5.58; τ2=0.25) and DFS showed an HR of 1.20 (95%CrI: 0.77-1.59; PI HR=1.21; 95%CrI: 0.14-2.20; τ2=0.42). CONCLUSION: Tumor budding is significantly associated with clinicopathological features and prognosis in patients with gynecological malignancies.