Abstract GS5-04: Accuracy of post-neoadjuvant chemotherapy image-guided breast biopsy to predict the presence of residual cancer: A multi-institutional pooled analysis

Abstract

Abstract Background: Image-guided biopsy of the residual imaging abnormality / tumor bed at the end of neoadjuvant chemotherapy (NAC) is increasingly used to assess residual disease in the breast, facilitate risk-adaptive surgery and potentially identify exceptional responders who may not require surgical intervention. The aim of this pooled analysis was to further assess the accuracy of post-NAC, image-guided biopsy to predict residual disease in the breast. The findings could help define an optimal post-NAC biopsy protocol to support trials omitting surgery in selected groups of patients. Methods: Multi-institutional, individual patient data on post-NAC image-guided biopsy was collected and pooled following Institutional Review Board (IRB) approval from each centre (Royal Marsden Hospital, Seoul National University Hospital, MD Anderson Cancer Center). Biopsy sampling accuracy was defined as representative if pathology features suggestive of tumor bed or residual cancer were identified. Pathologic complete response (pCR) was defined as no residual disease in the breast (ypT0). Biopsy predictive accuracy was calculated using final surgical pathology as the reference standard. Simple descriptive statistics and non-parametric analyses were performed. Results: Data were analyzed from 166 women who underwent post-NAC image-guided biopsy. Median age was 49 years (range: 25-76). The majority (n=160) had invasive ductal carcinoma (IDC) with phenotype distribution of 31 (18.7%), 47 (28.3%), 29 (17.5%) and 59 (35.5%) respectively for hormone receptor (HR) positive / Human Epidermal Growth Factor Receptor (HER)-2 negative, HR and HER2 positive, HR negative / HER2 positive and triple negative (TN). Median tumor size on pre-treatment imaging was 33.5 mm (range: 12-100). The overall pCR rate was 51.2% [16.1% for HR positive / HER2 negative, 44.7% for HR positive / HER2 positive, 69% for HR negative / HER2 positive and 66.1% for TN]. The majority (n=143) underwent vacuum assisted biopsy (VAB) and 23 had core cut (CC) biopsy. Median size of the biopsy needle was 10 gauge (range: 7-14) and median number of samples was 6 (range: 2-18). The biopsy was performed under ultrasound (n=129) or stereo guidance (n=37) and in 159 cases was representative of the tumor bed. When the image-guided biopsy (VAB and CC) was representative, the false negative rate (FNR) across the whole cohort was 18.7% (95% CI 9.8-26.8). Exploratory analysis of accuracy of VAB in cases with a residual imaging abnormality < 2 cm to allow adequate sampling and at least 6 representative biopsies taken (n=76) showed a FNR of 3.2% (95% CI 0-8.8), a negative predictive value (NPV) of 97.4% (95% CI 84.6-99.6) and an overall accuracy of 89.5% (95% CI: 80.3-95.3). Subgroup analysis in patients with HER2 positive or TN cancer (n=66) who are the most likely to be exceptional responders and achieve pCR showed similar accuracy of the technique with a FNR of 4.2% (95% CI 0-10.7), a negative predictive value (NPV) of 97.2% (95% CI 83.6-99.6) and an overall accuracy of 87.9% (95% CI: 77.5-94.6). Conclusions: This large pooled multicenter data suggests that a standardized protocol using image-guided VAB of a tumor bed measuring up to 2 cm with at least 6 samples allows reliable prediction of residual disease and pCR. These results should inform the design of de-escalation trials in NAC exceptional responders testing the safety of eliminating surgery. Citation Format: Marios Konstantinos Tasoulis, Han-Byoel Lee, Wei Yang, Romney Pope, Savitri Krishnamurthy, Soo-Yeon Kim, Nariya Cho, Victoria Teoh, Gaiane M Rauch, Benjamin D Smith, Vicente Valero, Wonshik Han, Royal Marsden Hospital MDT, Fiona MacNeill, Henry M Kuerer. Accuracy of post-neoadjuvant chemotherapy image-guided breast biopsy to predict the presence of residual cancer: A multi-institutional pooled analysis [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr GS5-04.