Ultrasound is at least as good as magnetic resonance imaging in predicting tumour size post-neoadjuvant chemotherapy in breast cancer

Background: The aim of this study was to evaluate the accuracy of clinical imaging of the primary breast tumour post-neoadjuvant chemotherapy (NAC) related to the post-neoadjuvant histological tumour size (gold standard) and whether this varies with breast cancer subtype. In this study, results of both magnetic resonance imaging (MRI) and ultrasound (US) were reported.
Methods: Patients with invasive breast cancer were enrolled in the INTENS study between 2006 and 2009. We included 182 patients, of whom data were available for post-NAC MRI (n = 155), US (n = 123), and histopathological tumour size.
Results: MRI estimated residual tumour size with <10-mm discordance in 54% of patients, overestimated size in 28% and underestimated size in 18% of patients. With US, this was 63%, 20% and 17%, respectively. The negative predictive value in hormone receptor-positive tumours for both MRI and US was low, 26% and 33%, respectively. The median deviation in clinical tumour size as percentage of pathological tumour was 63% (P25 = 26, P75 = 100) and 49% (P25 = 22, P75 = 100) for MRI and US, respectively (P = 0.06).
Conclusions: In this study, US was at least as good as breast MRI in providing information on residual tumour size post-neoadjuvant chemotherapy. However, both modalities suffered from a substantial percentage of over- and underestimation of tumour size and in addition both showed a low negative predictive value of pathologic complete remission (Gov nr: NCT00314977).