Tumor budding in cervical cancer as a prognostic factor and its possible role as an additional intermediate-risk factor.

Abstract Objective To evaluate the prognostic value and its possible role as an additional intermediate-risk factor of tumor budding (TB) in cervical cancer following radical hysterectomy. Methods In total, 136 patients with cervical cancer who underwent radical hysterectomy with pelvic and/or paraaortic lymphadenectomy were included. We assessed the status of TB in available hematoxylin and eosin-stained specimens. Univariate and multivariate analyses for predicting tumor recurrence and death were performed using TB and other clinicopathologic parameters. To evaluate additional intermediate-risk factors of TB, patients who had at least one high-risk factor were excluded, and a total of 81 patients were included. We added TB to three conventional intermediate-risk models and compared their performance with new and conventional models using the log-rank test and receiver operating characteristic analysis. Results High TB was defined as ≥5 per high-power field for disease-free survival and ≥ 8 per high-power field for overall survival. Multivariate analysis revealed that high TB was an independent prognostic factor for predicting overall survival (hazard ratio, 4.96; 95% confidence intervals, 1.06–23.29; p = .0423). The addition of TB to the conventional intermediate-risk models improved the accuracy of recurrence prediction. Among the risk models, the new model using at least two risk factors, including tumor size (≥ 4 cm), deep stromal invasion (outer one-third of entire cervical thickness), lymphovascular invasion, and high TB, was the most accurate for predicting tumor recurrence (area under the curve, 0.708, hazard ratio, 4.25; p = .0231). Conclusion High TB may be a prognostic biomarker of cervical cancer. Moreover, the addition of TB to the conventional intermediate-risk models improves the stratification of tumor recurrence.