Novel imaging techniques for intraoperative assessment of tumour resection margins in breast-conserving surgery

There is a clear need for more accurate techniques to assess tumour resection margins intraoperatively in breast-conserving surgery (BCS), as to date an average 10 – 30% of patients undergoing BCS require a re-operation to achieve clear margins. This thesis evaluates two novel imaging techniques for...

Full description

Bibliographic Details
Main Author: Grootendorst, Maarten Ruben
Other Authors: Purushotham, Anand David
Published: King's College London (University of London) 2017
Subjects:
Online Access:https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.718598
Description
Summary:There is a clear need for more accurate techniques to assess tumour resection margins intraoperatively in breast-conserving surgery (BCS), as to date an average 10 – 30% of patients undergoing BCS require a re-operation to achieve clear margins. This thesis evaluates two novel imaging techniques for intraoperative tumour margin assessment, Terahertz Pulsed Imaging (TPI) and Cerenkov Luminescence Imaging (CLI), that have been developed with a view to reducing reoperation rates in BCS. Both techniques were evaluated in a first-in-human, single centre study to demonstrate proof-of-principle and feasibility. A TPI handheld probe system (Teraview Ltd., UK) was used to scan breast samples ex vivo, and the TPI data was correlated with histopathology to assess diagnostic accuracy. CLI was evaluated intraoperatively by scanning excised BCS specimens from patients that received 2- deoxy-2-(18F)fluoro-D-glucose (18F-FDG) preoperatively using an investigational CLI imaging system (Lightpoint Medical Ltd., UK). An increased Technetium-99m (99mTc) nanocolloid activity of 150 MBq was used to facilitate sentinel node detection. Radiation doses to theatre staff were measured. CLI images were analysed postoperatively, and margin status correlated with histopathology results. The TPI handheld probe discriminated invasive breast cancer from benign breast tissue with a high sensitivity (86%) and an encouraging degree of accuracy (75%). Accurate discrimination of cancer from tissue containing a high percentage of fibrous cells proved challenging due to the similarities in the THz pulse between these two types of tissue. Intraoperative 18F-FDG CLI showed to be a feasible and low-risk procedure. Good agreement was found between CLI and histopathology for clear margins of excision. Sentinel nodes could be successfully detected using the gamma probe and blue dye, and radiation dose to staff was low. Elimination of image artefacts from tissue dissection with the monopolar diathermy device is needed to further refine CLI. In conclusion, both TPI and 18F-FDG CLI are promising techniques for intraoperative assessment of tumour resection margins in BCS, warranting larger studies to assess the diagnostic accuracy of each technique on different cancer types including DCIS, and the impact on re-operation rates.