Clinical Report: Effects of Virtual Dynamic Contrast-Enhanced Breast MRI
Overview
This retrospective study evaluated virtual dynamic contrast-enhanced (vDCE) breast MRI generated from noncontrast MRI data, including transfer of a previously described approach to data from a scanner not used for model training. The supplied article excerpt describes the cohort and study design but does not provide diagnostic results, lesion-size comparisons, BI-RADS findings, or reader-confidence outcomes.
Background
Breast MRI protocols commonly use dynamic contrast-enhanced (DCE) imaging after intravenous gadolinium-based contrast agent (GBCA) administration to detect and characterise enhancement patterns. GBCA use entails costs and has raised environmental concerns, while noncontrast approaches such as diffusion-weighted imaging may require interpretation strategies different from those used for DCE images. Deep-learning methods have demonstrated technical feasibility in generating virtual DCE images, including series at multiple timepoints. The study was designed to assess whether a vDCE approach could transfer to a scanner absent from model training and support BI-RADS-based assessment compared with multiparametric DCE-MRI.
Data Highlights
Study characteristic
Reported information
Design and setting
Retrospective study at University Hospital Erlangen
Diagnostic results and comparisons of lesion size, BI-RADS assessment, and reader confidence are not provided
Key Findings
The study included 540 women who underwent clinically indicated breast MRI between January 2017 and June 2020; mean age was 52 years (SD ±12).
Indications included preoperative exclusion of multifocal disease, high-risk screening, exclusion of recurrent breast cancer, and assessment of indeterminate findings or clinical complaints.
Patients with breast implants were excluded because distortion artefacts frequently occur on diffusion-weighted imaging.
Data from one 3-T scanner were used to limit potential domain shifts; the cohort was divided into training (377), validation (81), and holdout test (82) sets.
A secondary test set comprised 52 consecutive examinations, and the study aimed to evaluate transfer of a previously described vDCE approach to a scanner absent from model training.
The supplied material does not report study results for lesion size, BI-RADS assessment, or reader confidence; related context describes DCE-MRI as a central component of breast MRI and notes research on radiomics and lesion visibility, without providing directly comparable outcomes for this vDCE study.
Clinical Implications
The supplied excerpt establishes the study population, imaging approach, and intended comparison with multiparametric DCE-MRI, but it does not provide outcome data to support conclusions about diagnostic performance or reader confidence. The background describes vDCE as an investigated approach to reducing GBCA use, not as an established replacement for contrast-enhanced breast MRI.
Conclusion
This retrospective investigation was designed to assess vDCE transferability and BI-RADS-based assessment against multiparametric DCE-MRI. The provided source material does not include the results needed to determine its effects on lesion size, BI-RADS categorisation, or reader confidence.
by Andrzej Liebert, Hannes Schreiter, Lorenz A. Kapsner, Jessica Eberle, Chris M. Ehring, Dominique Hadler, Sabine Ohlmeyer, Frederik B. Laun, Ramona Erber, Julius Emons, Michael Uder, Evelyn Wenkel, Sebastian Bickelhaupt
Patients with estrogen-receptor-positive (ER+) HER-2-negative advanced breast cancer experienced significantly improved progression-free survival (PFS) when treated with an oral combination regimen that included giredestrant, a novel, next-generation selective ER degrader and full antagonist (SERD), compared to a standard combination approach.
