Imaging Overestimates the Local Extent of Breast Tumor After Neoadjuvant Chemotherapy

The current study was performed to analyze the accuracy of imaging in determining the amount of residual tumor after neoadjuvant chemotherapy. The hypothesis was that imaging overestimated the local extent of tumor in breast in a significant proportion of patients, contributing to the selection of mastectomy as the definitive surgical procedure.

A retrospective analysis was performed for patients with newly diagnosed invasive breast cancer treated with NAC from 2015 through 2023. Imaging characteristics after NAC were compared with tumor size on final pathology. Imaging overestimation and underestimation were defined as size on imaging > 1 cm either greater than or less than the final size on pathology, respectively.

Of 274 patients receiving NAC, there were 21(17%) patients who had false-positive and 24(28%) with false-negative imaging post-NAC. There were 181 patients who did not have radiographic complete response post-NAC, for whom mean radiographic tumor size post-NAC was 3.2 cm. Comparison of T size on post-NAC imaging versus final pathology yielded a correlation coefficient of 0.579. Radiographic assessment overestimated final pathology tumor size by > 1 cm in 81 (45%) of 181 patients overall. Rates of overestimation were not associated with tumor biomarker status or multi-focality/centricity but were significantly higher on post-NAC imaging for ycT2 (55%, p < 0.001) and ycT3 (67%, p < 0.001) versus ycT1 (25%, Ref.) tumors.

These results cast uncertainty on the reliability of post-NAC imaging to guide selection of the breast surgical procedure, especially when there is ycT3 tumor remaining after neoadjuvant systemic therapy.