Accuracy of morphologic change measurements by ultrasound in predicting pathological response to neoadjuvant chemotherapy in triple-negative and HER2-positive breast cancer

Background: Neoadjuvant chemotherapy (NAC) is standard therapy in triple-negative breast cancer (TNBC) and HER2-positive breast cancer (HER2 + ve BC). There are concerns about the accurate imaging modalities to measure residual tumor during or after NAC. Up to now no standard imaging method for monitoring the efficacy of NAC has been established, and few reports showed ultrasonographic change. We aimed to assess the echogenicity in ultrasonography (US) as the predictive marker of pathological complete response (pCR) for not only TNBC, but also HER2 + ve BC. Furthermore, we also investigated the change in depth (D) and width (W) of the tumor as the predictive value of pCR. Methods: We retrospectively reviewed a consecutive 59 patients with TNBC and 41 patients with HER2 + ve BC who received NAC. In all of 100 patients, echogenicity, D and W of the tumor were measured before (pre-NAC) and after NAC (post-NAC). The tumor echogenicity was measured at representative region of interest (ROI), and calculated as the relative comparative assessment with fat echogenicity (ROI ratio). Results: pCR was significantly associated with higher post-NAC ROI ratio in TNBC (p = 0.010), while there was no association in HER2 + ve BC (p = 0.885). pCR was significantly associated with smaller sizes of post-NAC D and W in TNBC (p = 0.001, 0.003), while no trend was observed in HER2 + ve BC (p = 0.259, 0.435). The area under the curve (AUC) for post-NAC ROI ratio and D were 0.701, 0.755, respectively. Combined with them, AUC became higher up to 0.762. Conclusion: TNBC and HER2 + ve BC showed different morphologic features of residual disease. Echogenicity and tumor size after NAC were both useful to predict pCR for TNBC, but not HER2 + ve BC. In future, radiological imaging needs to be analyzed in terms of breast cancer subtypes.