MRI evaluation of neoadjuvant low-dose fractionated radiotherapy with concurrent chemotherapy in patients with locally advanced breast cancer

Objectives

We address the diagnostic performance of breast MRI and the efficacy of neoadjuvant radiochemotherapy (NRC) treatment (NRC protocol) vs conventional neoadjuvant chemotherapy (NAC) in patients with locally advanced breast cancer.

Methods

The NRC protocol consists of six anthracycline/taxane cycles and concomitant low-dose radiotherapy on breast tumour volume. Breast MRI was performed at baseline and after the last therapy cycle in 18 and 36 patients undergoing the NRC protocol or conventional NAC (propensity matching).

Results

In both groups, we observed reduced tumour dimensions after the last cycle (p<0.001), and the response evaluation criteria in solid tumours (RECIST) class directly correlated with the tumour regression grade class after the last cycle (p<0.001). Patients in the NRC group displayed a higher frequency of complete/partial response than those in the NAC group (p=0.034). 17 out of 18 patients in the NRC group met the criteria for avoiding mastectomy based on final MRI evaluation. The RECIST classification displayed a superior diagnostic performance in the prediction of the response to treatment [area under the receiver operating characteristic curve (AUC)=0.72] than time-to-intensity curves and apparent diffusion coefficient (AUC 0.63 and 0.61). The association of the three above criteria yielded a better diagnostic performance, both in the general population (AUC=0.79) and in the NRC and the NAC group separately (AUC=0.82 and AUC=0.76).

Conclusions

The pathological response is predicted by MRI performed after the last cycle, if both conventional MRI and diffusion imaging are integrated. The NRC treatment yields oncological results superior to NAC.

Advances in knowledge

MRI could be used to establish the neoadjuvant protocol in breast cancer patients.Neoadjuvant chemotherapy is currently widely employed in patients with locally advanced breast cancer (LABC) in order to improve the rate of breast-conserving surgery (up to 98% of patients) and systemic control of the disease [1,2]. The coupling of pre-operative radiotherapy (RT) cycles with neoadjuvant chemotherapy has been proposed for other cancer types. In particular, taxanes could have a synergistic effect with RT when administered concurrently [3-5]. Nonetheless, few data are currently available on the efficacy of concurrent neoadjuvant RT in patients with LABC, although evidence exists that such a strategy is safe and feasible [6], and is supported by preliminary investigations [7,8]. Radiation doses below 0.5 Gy have been demonstrated to enhance the effectiveness of continuous-infusion taxanes. This phenomenon has been termed low-dose hyper-radiosensitivity [9,10].MRI is a reliable tool to evaluate the breast cancer response to chemotherapy by measuring tumour diameter changes and by assessing the viability of residual tumour areas [11-13]. Nevertheless, MRI may under- or overestimate the burden of residual tumour by confounding a fibrotic scar with viable tumour tissue, or vice versa. Diffusion-weighted imaging (DWI) has been shown in such contexts to improve the diagnostic performance of MRI [14]. It has not been clarified whether MRI retains its diagnostic performance even in the context of breast RT. The latter is known to trigger tissue oedema, which may potentially impair the diagnostic accuracy [15,16]. The purpose of the present work is to ascertain (1) the diagnostic performance of MRI and DWI-MRI in the context of concurrent low-dose fractionated RT (LD-FRT) and chemotherapy in the prediction of response to neoadjuvant treatment; (2) whether the adoption of concurrent neoadjuvant LD-FRT and chemotherapy yields better oncological results in LABC than neoadjuvant chemotherapy alone.