MRI-Derived Tumour-to-Breast Volume Is Associated with the Extent of Breast Surgery

Abstract

The tumour-to-breast volume ratio (TBVR) is a metric that may help surgical decision making. In this retrospective Ethics-Committee–approved study, we assessed the correlation between magnetic resonance imaging (MRI)-derived TBVR and the performed surgery. The TBVR was obtained using a fully manual method for the segmentation of the tumour volume (TV) and a growing region semiautomatic method for the segmentation of the whole breast volume (WBV). Two specifically-trained residents (R1 and R2) independently segmented T1-weighted datasets of 51 cancer cases in 51 patients (median age 57 years). The intraobserver and interobserver TBVR reproducibility were calculated. Mann-Whitney U, Spearman correlations, and Bland-Altman statistics were used. Breast-conserving surgery (BCS) was performed in 31/51 cases (61%); mastectomy was performed in 20/51 cases (39%). The median TBVR was 2.08‰ (interquartile range 0.70–9.13‰) for Reader 1, and 2.28‰ (interquartile range 0.71–9.61‰) for Reader 2, with an 84% inter-reader reproducibility. The median segmentation times were 54 s for the WBV and 141 s for the TV. Significantly-lower TBVR values were observed in the breast-conserving surgery group (median 1.14‰, interquartile range 0.49–2.55‰) than in the mastectomy group (median 10.52‰, interquartile range 2.42–14.73‰) for both readers (p < 0.001). Large scale prospective studies are needed in order to validate MRI-derived TBVR as a predictor of the type of breast surgery.