Comparison of postoperative CT- and preoperative MRI-based breast tumor bed contours in prone position for radiotherapy after breast-conserving surgery

Yinping Dong,Wanhu Li,Yiru Chen,Yongqing Li,Wei Huang,Jinhu Chen,Qian Zhao,Yang Liu

doi:10.1007/s00330-020-07085-0

Abstract

ObjectivesTo compare the target volume of tumor bed defined by postoperative computed tomography (post-CT) in prone position registered with or without preoperative magnetic resonance imaging (pre-MRI).MethodsA total of 22 patients were included with early-stage breast invasive ductal cancer, who have undergone breast-conservative surgery and received the pre-MRI and post-CT in prone position. The MRI sequences (T1W, T2W, T2W-SPAIR, DWI, dyn-eTHRIVE, sdyn-eTHRIVE) were delineated and manually registered to CT, respectively. The clinical target volumes (CTVs) and planning target volumes (PTVs) were contoured on CT and different MRI sequences, respectively. Differences were measured in terms of consistence index (CI), dice coefficient (DC), geographical miss index (GMI), and normal tissue index (NTI).ResultsThe differences of delineation volumes among CT and MRIs were significant, both in the CTVs (p = 0.035) and PTVs (p < 0.001). The values of CI and DC for sdyn-eTHRIVE registration to CT were the largest among all MRI sequences, but GMI and NTI were the smallest. No obvious linear correlation (p > 0.05) between the CI derived from the registration of CT and sdyn-eTHRIVE of CTV with the breast volume, the cavity visualization score (CVS) of CT, time interval from surgery to CT simulation, the maximum diameter of the intraoperative mass, and the number of titanium clips, respectively.ConclusionsThe CTVs and PTVs in MRI sequences were all smaller than those in CT. The pre-MRI, especially the sdyn-eTHRIVE, could be used to optimize the post-CT-based target delineation of breast cancer.Key Points• Registered pre-MRI to post-CT in order to improve the accuracy of target volume delineation of breast cancer.• The CTVs and PTVs in MRI sequences were all smaller than those in CT.• The sdyn-eTHRIVE of pre-MRIs may be a better choice to improve the delineation of CT-based CTV and PTV.

Highlights

Boost to the tumor bed (TB) following whole-breast radiotherapy could improve the local recurrence rates in patients with conservative breast cancer [1,2,3]
The sdyn-eTHRIVE of preoperative magnetic resonance imaging (pre-magnetic resonance imaging (MRI)) may be a better choice to improve the delineation of computed tomography (CT)-based clinical target volumes (CTVs) and planning target volumes (PTVs)
Radiation oncologists usually outline the TB in a variety of ways, but the position and the shape of the TB could not be determined accurately. All these elements, including the patient’s surgical records, surgical scar [4], postoperative palpation changes, clips placed around the bed [5, 6], and presurgical or postoperative breast imaging examinations, such as mammography, ultrasonography, computed tomography (CT) imaging, and magnetic resonance imaging (MRI), were all taken into account

Summary

Introduction

Boost to the tumor bed (TB) following whole-breast radiotherapy could improve the local recurrence rates in patients with conservative breast cancer [1,2,3]. Radiation oncologists usually outline the TB in a variety of ways, but the position and the shape of the TB could not be determined accurately All these elements, including the patient’s surgical records, surgical scar [4], postoperative palpation changes, clips placed around the bed [5, 6], and presurgical or postoperative breast imaging examinations, such as mammography, ultrasonography, computed tomography (CT) imaging, and magnetic resonance imaging (MRI), were all taken into account. Because of its superb soft tissue contrast [11, 12] and better definition of TB [13], MRI, especially the postoperative MRI (post-MRI), has been incorporated into target volume definition under the premise of CT-positioned and CTdetermined TB, as well, the co-registration of the MRI and CT imaging modalities could result in improved interobserver concordance in delineation [7, 14]. In order to improve the accuracy of the TB delineation, our previous study delineated the lumpectomy cavity (LC) volume on the basis of the registration imaging of post-MRI and postoperative CT (postCT) in prone position, which pointed out that MRI could improve the visibility of LCs compared with CT, and therein the MRI-STIR sequence showed the highest visibility [16]

Objectives

Methods

Results

Discussion

Conclusion