The value of preoperative molecular breast imaging for the size assessment of invasive carcinoma and ductal carcinoma in situ

PO044 The value of preoperative molecular breast imaging for the size assessment of invasive carcinoma and ductal carcinoma in situ S. Galassi1, M.L. Stazza1, M. Rondini1, A. Lazzarato1, S. Nuvoli1, G. Madeddu1, A. Spanu1 1Unit of Nuclear Medicine, Department of Clinical, Surgical and Experimental Sciences, University of Sassari, Sassari, Italy Background-aim: Molecular Breast Imaging (MBI) with the technetium- labelled cationic lipophilic radiotracers has proved to be able to play an important complementary role to conventional imaging procedures in the diagnosis of primary breast cancer as well as in the preoperative local staging of disease. MBI also demonstrated a high accuracy in the identification of residual disease following neoadjuvant therapy. The aim of the present study was to assess the value of preoperative MBI for the size assessment of invasive carcinoma and ductal carcinoma in situ (DCIS) in patients with newly diagnosed primary breast cancer. Methods: A consecutive series of 190 patients (age: 31–78 years) with newly diagnosed primary breast cancer scheduled to surgery was retrospectively reviewed. Prior to surgery, all patients underwent MBI, in both craniocaudal and mediolateral oblique projection (600 s/ view), using a high-resolution semiconductor-based device for imaging acquisition. MBI data were correlated with surgical histopathological findings. Pearson’s correlation coefficient was calculated to assess differences in tumor size between MBI and histopathological examination. Concordance was defined as a difference B 0.5 cm between MBI and pathology. Results: At surgery, 108 out of 190 patients had an invasive primary tumor (invasive ductal: 95 cases, invasive lobular: 7 cases, mucinous: 3 cases, other types: 3 cases) with a median tumor size of 1.80 ± 1.83 cm at histopathological analysis and 1.90 ± 1.94 cm at MBI (r = 0.986, p\0.00001) which over-estimated lesion size in 8/108 cases. Thirty-five/190 patients had a DCIS with median tumor size of 1.60 ± 1.23 cm at histopathological analysis and 2.15 ± 1.50 cm at MBI (r = 0.629, p\0.00005). In this group of patients, lesion size was overestimated at MBI in 11 cases and underestimated in one. The remaining 47/190 patients had an invasive tumor (invasive ductal: 41 cases, invasive lobular: 6 cases) and an associated DCIS component. In 38 of these 47 cases, the DCIS component was marginal; median tumor size was 1.5 ± 1.18 cm at histology and 1.85 ± 1.16 cm at MBI (r = 0.921, p\0.00001) which overestimated lesion size in 10 cases and underestimated in one case. In the remaining 9/47 patients, the DCIS associated to the invasive tumor was extensive and MBI gave an accurate size definition of both components in 7/9 cases, overestimating in one case and underestimating in the remaining case. Conclusions: MBI proved a reliable diagnostic tool in the preoperative assessment of tumor size in patients with newly diagnosed primary breast cancer. MIBI lesion size strongly correlated with histological findings mainly in pure invasive carcinomas. Lesion size overestimation was more frequent in DCIS. However, the size of extensive DCIS component associated to invasive carcinomas was correctly predicted in a high percentage of cases. These data seem to suggest a wider application of MBI in the surgical planning of breast cancer patients. However, they need to be confirmed in larger prospective studies.