Tumor budding associates with adverse clinicopathological features and reduced overall survival in cervical and endometrial cancer.

This study is a Bayesian meta-analysis of observational cohort studies investigating the prognostic significance of tumor budding in patients with gynecological malignancies. The analysis pooled data from 18 cohort studies, encompassing a total sample size of 3320 patients diagnosed with either cervical cancer or endometrial cancer. The specific setting of the individual studies was not reported in the available data. The primary exposure of interest was the presence of tumor budding, defined as the presence of single cells or small clusters of cells at the invasive front of the tumor, compared to its absence. The study design relies on retrospective or prospective observational data rather than randomized controlled trials, which inherently limits the ability to establish causality.

The analysis focused on the association between tumor budding and various clinicopathological parameters as well as survival outcomes. For the primary outcome of cancer stage, the presence of tumor budding demonstrated a significant association with higher stage disease. The effect size was reported as an odds ratio (OR) of 2.91, with a 95% credible interval (CrI) ranging from 1.86 to 4.41. This indicates that tumors exhibiting budding were nearly three times more likely to be associated with advanced staging compared to those without budding, though absolute numbers were not reported.

Regarding histological grading, tumor budding showed a robust association with higher grade tumors. The calculated odds ratio was 5.00, with a 95% CrI of 2.83 to 8.76. This suggests a strong link between the presence of budding and poorly differentiated tumor characteristics. Furthermore, the analysis found a significant association with nodal involvement, with an OR of 3.63 (95% CrI: 2.41-5.47). This finding implies that tumor budding is a marker for increased risk of regional lymph node metastasis in these patient populations.

Lymphovascular invasion was another significant parameter evaluated. The presence of tumor budding was associated with this feature with an OR of 4.22 (95% CrI: 2.52-6.92). This association highlights the potential utility of tumor budding as a marker for aggressive biological behavior involving vascular invasion. In terms of survival outcomes, the analysis examined overall survival. The presence of tumor budding was significantly associated with reduced overall survival, yielding a hazard ratio (HR) of 2.14 (95% CrI: 1.27-3.63). This indicates that patients with tumor budding faced more than double the risk of death compared to those without budding.

In contrast to overall survival, the association with disease-free survival did not reach statistical significance. The hazard ratio was 1.20, with a 95% CrI of 0.77 to 1.59. While the point estimate suggests a potential trend toward worse disease-free survival, the wide credible interval crossing unity indicates uncertainty in this specific outcome. Safety and tolerability data, including adverse events, serious adverse events, and discontinuations, were not reported, as the study assessed prognostic markers rather than therapeutic interventions. Consequently, no safety profile for tumor budding itself is applicable, as it is a pathological feature rather than a drug or procedure.

Methodological limitations inherent to the study design must be acknowledged. The analysis is based entirely on observational data, meaning that tumor budding is a biomarker and not an intervention. Therefore, the results describe associations rather than causal relationships. The Bayesian approach utilized credible intervals and prediction intervals to reflect heterogeneity among the included studies, but specific limitations regarding study quality, potential biases, or funding conflicts were not reported. Additionally, the disease-free survival result was not statistically significant, which tempers the overall prognostic strength of the marker for recurrence specifically. The scope is also limited to cervical and endometrial cancers, restricting generalizability to other gynecological or non-gynecological malignancies.

Clinically, these results suggest that tumor budding may serve as a useful adjunctive prognostic marker in cervical and endometrial cancer pathology reports. The significant associations with stage, grade, nodal status, and overall survival support its potential role in risk stratification. However, clinicians should interpret these findings with caution, recognizing that the evidence is derived from observational cohorts and does not establish that tumor budding causes worse outcomes. The lack of significance in disease-free survival further limits its utility for predicting recurrence in isolation. Further research is needed to validate these findings in prospective settings and to determine if tumor budding status should influence clinical management decisions, such as adjuvant therapy recommendations, in these specific cancer types.

Key questions remain unanswered regarding the optimal threshold for defining tumor budding and its integration into standard prognostic models. The heterogeneity of the included studies and the lack of reported limitations suggest that the true magnitude of the association may vary across different populations or pathological assessment methods. Until prospective validation is available, tumor budding should be viewed as an associative finding that correlates with known adverse features but requires confirmation in larger, standardized cohorts before altering practice patterns.