Reducing mastectomy rates in invasive lobular carcinoma by high-resolution 3D breast CT

点击翻译按钮获取中文摘要

 DESCRIPTION (provided by applicant): Cancers of the breast are the second most common type of breast cancers (after invasive ductal carcinoma, IDC) and account for 10-15% of all breast cancers. More than any other breast cancer, lobular cancers present as multifocal, multicentric and bilateral disease. Among histological types, re- excision rates are highest (28.3%) for invasive lobular carcinoma and substantially more than invasive ductal carcinoma (19.1%). For subjects diagnosed with lobular carcinoma, breast MRI is currently the preferred modality for evaluating disease extent as ultrasound and mammography have been shown to be inferior in accurately estimating tumor size. Meta-analysis of pooled data have shown that breast MRI has a sensitivity of 93.3% for detecting lobular carcinoma with additional lesions detected in 32% and 7% of patients in the ipsilateral and contralateral breasts, respectively. However, for approximately 25% to 35% of tumors, the size estimated from breast MRI differs from pathology by more than 1 cm. The use of pre-operative breast MRI for evaluating disease extent has been associated with increased odds for mastectomy. A recent study analyzing 243 patients with invasive lobular carcinoma (ILC) concluded that ILC can be safely treated with conservative surgery but a more accurate preoperative evaluation of tumor size and multifocality is needed to reduce re-excision rate. We hypothesize the high spatial resolution 3D imaging provided by dedicated breast CT that is capable of resolving features in the 200 to 250 microns range would improve the proportion of tumors that are concordant in size with histopathology and hence would increase the likelihood of subjects being treated with breast conserving surgery. Dedicated breast CT does not require physical compression of the breast and takes 10 seconds for a scan. This prospective clinical study is designed to address if all foci observed with contrast-enhanced breast MRI are also visible with contrast-enhanced breast CT (sensitivity) and if the tumor size determined from breast CT is more concordant with pathology than breast MRI. The study will also investigate two automated segmentation and tumor size quantification methods to determine, which quantitative algorithm is more accurate, with tumor size from surgical pathology serving as reference standard. Thus, the proposed study challenges existing paradigms on the accuracy of tumor size measurements and paves for the way for reducing mastectomy rates.