TOPLINE:

Contrast-enhanced mammography (CEM) demonstrated diagnostic performance comparable to MRI in assessing nipple-areola complex (NAC) involvement in preoperative patients with breast cancer. CEM achieved a sensitivity of 60% and a specificity of 96.3%, whereas MRI showed a sensitivity of 50% and a specificity of 96.4%, with no significant differences observed between two modalities.

METHODOLOGY:

• Researchers conducted a retrospective cohort study of 195 women with biopsy-proven breast cancer who underwent either CEM (n = 91; median age, 66 years) or MRI (n = 104; median age, 53 years) for preoperative assessment at a single institution in Italy between January and December 2022.
• Institutional policy allocated women aged 60 years or older and/or those with relative or absolute contraindications to MRI to CEM, and MRI was reserved for all other patients.
• Two radiologists independently evaluated images for NAC involvement on the basis of the predefined criteria including nipple retraction and invasion, abnormal enhancement of the NAC, tumour nipple enhancement, periareolar skin thickening, abnormal morphology or asymmetry of the NAC, and tumour-to-nipple distance ≤ 10 mm.
• Performance was evaluated against the reference standard of histopathologic evaluation of surgical specimens.
• Multivariable logistic regression was used to identify independent imaging predictors of NAC involvement for both modalities.

TAKEAWAY:

• NAC involvement was histologically confirmed in 11.0% of CEM cases and 19.2% of MRI cases, with a difference that was not statistically significant (P = .110).
• The sensitivity for detecting NAC involvement was 60% (95% CI, 26%-87%) for CEM and 50% (95% CI, 27%-73%) for MRI, with no significant difference (P = .897); however, the specificity was nearly identical at 96.3% (95% CI, 89%-99%) for CEM and 96.4% (95% CI, 83%-99%) for MRI (P = .709).
• Independent predictors of NAC involvement included periareolar thickening (odds ratio [OR], 26.3; P = .0007) for CEM and abnormal NAC enhancement (OR, 14; P < .001) and shorter tumour-to-nipple distance (OR, 0.28; P < .001) for MRI.
• In situ components were frequently observed in NAC-positive cases (90% in the CEM group and 75% in the MRI group), but this finding was not statistically predictive of NAC involvement in either cohort.

IN PRACTICE:

"Notwithstanding a difference in patient age between our CEM and MRI cohorts, which mirrors the different clinical use of the two imaging modalities, CEM and MRI show comparable diagnostic performance in the assessment of NAC involvement in breast cancer patients, suggesting their usefulness to guide surgical planning for NAC-sparing procedures," the authors wrote.

"Given the high specificity, great accessibility, fast acquisition time, favorable tolerability, and simple interpretation, CEM may represent a clinically viable alternative to MRI in appropriately selected patients," they added.

SOURCE:

This study was led by Michele Lorenzon, Azienda Sanitaria Universitaria Friuli Centrale, Udine, Italy. It was published online on February 24, 2026, in European Radiology.

LIMITATIONS:

This study was limited by its retrospective design and a significant difference in median age between the two cohorts, mirroring different clinical applications of the modalities. The relatively small number of NAC-positive cases may have limited the robustness of some statistical analyses.

DISCLOSURES:

Open access funding was provided by Università degli Studi di Udine within the CRUI-CARE Agreement. One author reported serving as a deputy editor for European Radiology but did not participate in the selection or review processes of this article.

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