| Sumario: | Background This study aimed to explore the ability of a minimal invasive biopsy to diagnose a pathological complete response (pCR = ypT0) in the breast. Methods Ultrasound-guided, vacuum-assisted, minimal invasive biopsy (VAB) was performed in 50 patients after neoadjuvant chemotherapy and before breast surgery. Negative predictive values (NPV) and false negative rates (FNR) to predict a pCR in surgical specimen were the main outcome measures. To assess the possible sampling error, the representativeness of the sample was evaluated by the biopsy performing physician (subjectively based on the visibility in ultrasound), by radiography of the biopsy specimen, and by the pathologist (based on histopathology). Results The cohort (n = 50) consisted of 15 (30%) triple negative breast cancers, 13 (26%) human epidermal growth factor receptor 2 (HER2) positive and 22 (44%) hormone receptor positive/HER2 negative cancers. ypT0 was diagnosed in 23 (46%) cases. In the overall cohort (n = 50), VAB yielded an NPV of 76.7% and an FNR of 25.9%. Given a representative VAB sample, according to the histopathological evaluation (n = 38), the NPV was 94.4% (95% CI 87.1–100.0) and the FNR 4.8% (95% CI 0.0–11.6). Non-representative VABs were mainly due to bad visibility of the target lesion in ultrasound. Conclusion A VAB can accurately diagnose a pCR, given a histopathologically representative sample.
|
|---|