Malignancy Rates of Pure Radial Scars Detected on Digital Breast Tomosynthesis: A Meta-analysis

To estimate the pooled malignancy rate associated with pure radial scars (RS) detected solely on digital breast tomosynthesis (DBT) and evaluate the consistency of these findings across published literature.

A systematic review and meta-analysis were conducted following the preferred reporting items for systematic reviews and meta-analyses guidelines. Nine single-arm studies, including 593 histologically confirmed pure RS lesions detected on DBT were identified through Medline and Embase database searches up to January 2025. The primary endpoint was the pooled malignancy rate. Fixed-effects meta-analysis with Wilson confidence intervals was used to calculate proportions. Heterogeneity was assessed using Cochran's Q, I2, and Tau2. Publication bias was evaluated using a funnel plot.

Across the nine included studies, nine of 593 pure RS lesions were upstaged to malignancy, yielding a pooled malignancy rate of 1.9% (95% CI: 1.04-3.29%). There was no significant heterogeneity across studies (Q=1.19, I2=0%, Tau2=0.00). Funnel plot analysis revealed no asymmetry, suggesting an absence of publication bias.

The risk of malignancy associated with pure RSs detected only on DBT is low and consistent across studies. These findings support a conservative management approach, especially when radiologic-pathologic concordance is established, rather than surgical excision.