A Risk Model based on Ultrasound, Ultrasound Elastography, and Histologic Parameters for Predicting Axillary Lymph Node Metastasis in Breast Invasive Ductal Carcinoma

Abstract

To develop a risk model for predicting axillary lymph node metastasis (LNM) in patients with breast invasive ductal carcinoma (IDCs) using ultrasound (US), US elastography of virtual touch tissue imaging (VTI) and virtual touch tissue imaging & quantification (VTIQ), and histologic parameters. This study included 162 breast IDCs in 162 patients. Univariate and multivariate analyses were used to identify the risk factors and a risk model was created. The results found that 64 (39.5%) of 162 patients had axillary LNMs. The risk score (RS) for axillary LNM was defined as following: RS = 1.3 × (if lesion size ≥20 mm) + 2.6 × (if taller than wide shape) + 2.2 × (if VTI score ≥5) + 3.9 × (if histological grade III) + 1.9 × (if positive C-erbB-2). The rating system was divided into 6 stages (i.e. Stage I, Stage II, Stage III, Stage IV, Stage V, and Stage VI) and the associated risk rates in terms of axillary LNM were 0% (0/19), 6.1% (2/33), 7.7% (3/39), 65.5% (19/29), 92.3% (24/26), and 100% (16/16), respectively. The risk model for axillary LNM established in the study may facilitate subsequent treatment planning and management in patients with breast IDCs.

Figure 1

Images in a 60-year-old patient with breast invasive ductal carcinoma, no axillary lymph node metastasis (LNM), histologic grade II, positive estrogen receptor (ER), negative progesterone receptor (PR), and negative C-erbB-2. (a) A solid, heterogeneous hypoechogenicity, and poorly defined margin lesion (arrows) is shown on ultrasound (US). (b) Poor internal flow (i.e. dotted signals on color Doppler US) is found on color Doppler flow image (arrows). (c) Virtual touch tissue imaging (VTI) score of the lesion (arrows) is 3. (d) On virtual touch tissue imaging & quantification image, the lesion (arrows) is heterogeneous with a mean SWS value of 8.24 m/s. (e) Pathological examination confirms the diagnosis of invasive ductal carcinoma (Hematoxylin-eosin stain, x200). (f) Pathological examination confirms the diagnosis of normal lymph node (Hematoxylin-eosin stain, x200).

Figure 2

Images in a 58-year-old patient with breast invasive ductal carcinoma, axillary lymph node metastasis (LNM), histologic grade III, negative estrogen receptor (ER), negative progesterone receptor (PR), and positive C-erbB-2. (a) A solid, marked hypoechogenicity, well defined margin, irregular, and taller than wide shape lesion (arrows) is shown on US. (b) Rich internal flow (i.e. 3 linear or tree-like signals) is found on color Doppler flow image (arrows) of the breast invasive ductal carcinoma. (c) Virtual touch tissue imaging (VTI) score of the lesion (arrows) is 4. (d) On virtual touch tissue imaging & quantification image, the lesion (arrows) is heterogeneous with a mean SWS value of 4.75 m/s. (e) Pathological examination confirms the diagnosis of invasive ductal carcinoma (Hematoxylin-eosin stain, ×200). (f) Pathological examination confirms the diagnosis of axillary LNM (Hematoxylin-eosin stain, ×200).

Figure 3

Virtual Touch tissue imaging elasticity scores of the lesions (arrows): (a) score 2; (b) score 3; (c) score 4; (d) score 5; and (e) score 6.

Figure 4

Representative immunohistochemical staining of estrogen receptor (ER), progesterone receptor (PR), and C-erbB-2. (a) ER status (negative); (b) ER status (positive); (c) PR status (negative); (d) PR status (positive); (e) C-erbB-2 expression (negative); and (f) C-erbB-2 expression (positive).

Figure 5

Receiver operating characteristic (ROC) curves of (a) the lesion size ≥20 mm (AUROC = 0.681), (b) taller than wide shape (AUROC = 0.722), (c) VTI score ≥5 (AUROC = 0.735), (d) histological grade III (AUROC = 0.824), (e) C-erbB-2 positive (AUROC = 0.696), and (f) the predictive equation (AUROC = 0.958) for prediction of axillary lymph node metastasis (LNM).