Assessing response in breast cancer with dynamic contrast-enhanced magnetic resonance imaging: are signal intensity-time curves adequate?

Abstract

Quantitative DCE-MRI parameters including K(trans) (transfer constant min(-1)) can predict both response and outcome in breast cancer patients treated with neoadjuvant chemotherapy (NAC). Quantitative methods are time-consuming to calculate, requiring expensive software and interpretive expertise. For diagnostic purposes, signal intensity-time curves (SITCs) are used for tissue characterisation. In this study, we compare the ability of NAC-related changes in SITCs with K(trans) to predict response and outcomes. 73 women with primary breast cancer underwent DCE-MRI studies before and after two cycles of NAC. Patients received anthracycline and/or docetaxel-based chemotherapy. At completion of NAC, patients had local treatment with surgery & radiotherapy and further systemic treatments. SITCs for paired DCE-MRI studies were visually scored using a five-curve type classification schema encompassing wash-in and wash-out phases and correlated with K(trans) values and to the endpoints of pathological response, OS and DFS. 58 paired patients studies were evaluable. The median size by MRI measurement for 52 tumours was 38mm (range 17-86mm) at baseline and 26mm (range 10-85mm) after two cycles of NAC. Medianbaseline K(trans) (min(-1)) was 0.214 (range 0.085-0.469), and post-two cycles of NAC was 0.128 (range 0.013-0.603). SITC shapes were significantly related to K(trans) values both before (χ (2)=43.3, P=0.000) and after two cycles of NAC (χ (2) =60.5, P=0.000). Changes in curve shapes were significantly related to changes in K(trans) (χ (2)=53.5, P=0.000). Changes in curve shape were significantly correlated with clinical (P=0.005) and pathological response (P=0.005). Reductions in curve shape of ≥1 point were significant for overall improved survival using Kaplan-Meier analysis with a 5-year OS of 80.9 versus 68.6% (P=0.048). SITCs require no special software to generate andprovide a useful method of assessing the effectiveness of NAC for primary breast cancer.