TOPLINE:

Contrast-enhanced MRI demonstrated a sensitivity of 91.3% and a specificity of 68.8% in assessing high-risk breast lesions, with the potential to help patients avoid unnecessary surgery, according to a systematic review and meta-analysis.

METHODOLOGY:

• Researchers conducted a systematic review and meta-analysis of seven studies and examined 479 patients with 493 high-risk lesions undergoing breast MRI until August 2024.
• Studies included five retrospective and two prospective analyses, with most using dedicated breast coils and examining patients in a prone position.
• The analysis included histopathological examination or imaging follow-up of at least 12 months as the reference standard and two readers who were blinded to histopathological results. The average breast cancer prevalence was 17% among the analysed cases.
• Researchers evaluated the diagnostic performance of contrast-enhanced MRI in ruling out malignancy among high-risk lesions.

TAKEAWAY:

• The pooled sensitivity was 91.3% (95% CI, 82.8%-95.8%) and the pooled specificity was 68.8% (95% CI, 50.3%-82.8%) for contrast-enhanced MRI in assessing high-risk lesions.
• The pooled negative likelihood ratio was 0.135 (95% CI, 0.082-0.222), whereas the pooled positive likelihood ratio was 2.875 (95% CI, 1.810-4.566).
• Only 6 out of 493 malignancies were missed by contrast-enhanced MRI, all being small low-grade ductal carcinoma in situ. The negative predictive value was 100% for invasive cancer.
• Contrast-enhanced MRI could reliably rule out malignancy in lesions with pretest probabilities up to 13.1%.

IN PRACTICE:

"This approach shows very promising results and high clinical relevance, potentially identifying patients who might avoid surgery, saving costs and time, as well as reducing patient anxiety and morbidity," the authors wrote.

SOURCE:

This study was led by Giulia Vatteroni, Department of Biomedical Sciences, Humanitas University, Milan, Italy. It was published online on January 17, 2026, in European Radiology.

LIMITATIONS:

Significant heterogeneity was noted between the included studies, which could not be fully explained due to limited data.

DISCLOSURES:

This study received open access funding from the Medical University of Vienna. The authors declared having no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.