Use of the 3D specimen tomosynthesis slices intra-operatively could reduce re-excision rates compared to 2D composite view by 20% when images are reviewed by the surgeon alone intraoperatively. If a breast radiologist were to report the 3D images intra-operatively, 50% of patients could have avoided re-excision. This would lead to a significant improvement in patient outcomes and a reduction in healthcare provision costs.

‍

Introduction

70% of women undergoing surgical intervention for breast cancer will undergo breast-conserving surgery (BCS). 20-25% of women who undergo BCS require at least a second operation due to positive/close margins on histopathology. Our institution follows UK Association of Breast Surgeons guidelines defining a positive margin as <1 mm of healthy tissue for invasive cancer with associated DCIS and <2mm for pure DCIS.

Intraoperative specimen radiography for non-palpable lesions has become standard practice in the UK, however, it is reviewed solely by the surgeon. Surgeons review 2D or rarely 3D images of the specimen intraoperatively to assess if further excision is necessary. Advanced 3D imaging using a Kubtec® machine with tomosynthesis allows the evaluation of the entire height of the specimen via 1mm slices, although the 3D function is not always utilised.

This retrospective review of specimen tomosynthesis images obtained using the Kubtec® 3D system looks at whether a review of all tomography slices by the surgeon, a radiologist or a combined assessment could reduce the need for subsequent operations.

‍

Methods and Materials

This study retrospectively reviewed specimen images that a surgeon had intra-operatively reviewed and interpreted as radiologically clear margins using the Kubtec Mozart ® System. However, the final histopathology identified a positive margin. During the retrospective review, an additional senior surgeon and two senior consultant radiologists assessed the four radial margins using the 3D tomosynthesis function. All investigators were blinded to the histopathology results.

All patients undergoing BCS who required a return to theatre for further surgery from July 2021 to June 2023 were included. Patients with either primary in- situ or invasive disease and those who underwent primary surgery or surgery following neoadjuvant systemic therapy were included. Patients who did not have intra-operative specimen imaging, or who had a positive margin diagnosed intra- operatively that was revised at the time, or who had suboptimal specimen orientation were excluded from the study.

The Kappa coefficient was calculated to assess agreement between the surgeon and the radiologist's interpretation of positive radial margins based on the 3D images. Sensitivity and specificity were calculated for both surgeon and radiologist assessment of margins as compared with the gold standard of histopathological assessment. All statistical analyses were conducted in Stata (version 18.0, StataCorp LLC, College Station, TX).

‍

Results

The review included 54 specimens from 54 patients, diagnosed with invasive carcinoma and pure ductal carcinoma in situ (DCIS) that underwent primary or post-neoadjuvant chemotherapy (NACT) surgery.

81 margins out of 216 (37.5%) were positive on histopathology, 44 (54.3%) of these were assessed as having involved margins by radiologists and 25 (30.9%) by surgeons using the 3D tomosynthesis function. The radiologist individually reported margins with better sensitivity (54.3% vs 30.9%) but poorer specificity (95.6% vs 100%) than the surgeon; however, there was a fair agreement between their interpretation (Kappa agreement 78.3%, Kappa statistic 0.26, p<0.0001). The dual assessment showed a better accuracy of 85.2% (Radiologist 80.1% and Surgeons 74.1%).

Overall, 27 (50%) patients could have avoided returning to the theatre if the images had been reported intra-operatively by a breast radiologist alongside the surgeon. At our intuition, a day case procedure for re-excising margins costs circa £3,638 per patient, saving approximately £200,000 a year.

‍

Conclusions

Use of the 3D specimen tomosynthesis slices intra-operatively could reduce re- excision rates compared to 2D composite view by 20% when images are reviewed by the surgeon alone intraoperatively. If a breast radiologist were to report the 3D images intra-operatively, 50% of patients could have avoided re-excision. This would lead to a significant improvement in patient outcomes and a reduction in healthcare provision costs.