P5-11-01: The Accuracy of Preoperative Ultrasonography Guided Vacuum-Assisted Breast Biopsy in Determining Histological Type, ER Status, PgR Status, HER2 Status and Ki67 Level in Invasive Breast Cancer.

Abstract

Abstract Background Recently, ultrasonography guided vacuum-assisted breast biopsy (US-guided VABB) has been widely used as alternative to surgical open biopsy. Enough breast tissue samples obtained by US-guided VABB are important because of increasing neoadjuvant treatment. Although the information obtained from US-guided VABB may be the only information available for determining the candidates for neoadjuvant treatment, only few studies evaluated the concordance of histological type, estrogen receptor (ER) status, progesterone receptor (PgR) status, human epidermal growth factor-2 (HER2) status, and Ki67 level between US-guided VABB and surgical specimen. The aim of this study was to evaluate the accuracy of preoperative US-guided VABB. Materials and Methods: In 439 breast cancer patients without neoadjuvant treatment who underwent US-guided VABB and surgical resection from April 2004 and March 2011 at Aichi Cancer Center hospital, we examined the concordance of Histological type, ER status, PgR status, HER2 status, and Ki67 level between US-guided VABB and surgical specimen. All the US-guided VABB were performed using 11-gauge Mammotome® or 10-gauge VACORA®. The ER and PgR status were assessed using Allreds scoring system by IHC. These statuseswere categorized as positive when the total score was more than two. HER2 expression status was tested by IHC and FISH. HER2 3+ by IHC, or 2+ and FISH positive were judged as HER2 positive. In this study, the Ki67 cut-off level for positivity was defined at 20% (Penault-Llorca et al, JCO 2009). The agreement on histological type, ER status, PgR status, HER2 status, and Ki67 level were tested using the absolute concordance rate and the kappa statistic values. Results: The concordance rate of histological types between US-guided VABB and surgical specimens was 93.4% (410 of 439 cases) with a Kappa statistic value of 0.82. In 115 cases diagnosed as DCIS by US-guided VABB, 28 cases (24.3%) were subsequently diagnosed as invasive cancer by surgical specimens. However, among these cases, 78.6% (22/28) were T1mic and T1a. And one case (0.3%, 1/324) diagnosed as invasive cancer by US-guided VABB changed DCIS by surgical specimens. The concordance rate of ER, PgR, and HER2 status were 96.6% (112/116), 89.5% (102/116), and 97.4%(113/116), respectively (kappa statistic value of 0.99, 0.76, and 0.90). In HER2 status, the concordance rate between US-guided VABB and surgical specimens was better than between core needle biopsy and surgical specimens (the concordance rate: 88%, kappa statistic value: 0.65, Usami et al, Jpn J Clin Oncol 2007). The agreement of Ki67 level was 85.7% (24/28) with a Kappa statistic value of 0.71. Conclusions: The judgment of histological type, ER status, and HER2 status by preoperative US-guided VABB can be used with confidence due to determine the treatment strategies according to molecular subtype. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-11-01.