A first in human feasibility study of 18F-FDG Cerenkov luminescence imaging (CLI)for intraoperative assessment of tumour resection margins in breast-conserving surgery (BCS)


Session type:


Massimiliano Cariati1,Maarten Grootendorst1,Ashutosh Kothari2,Michael Douek1,Tibor Kovacs2,Hisham Hamed2,Amit Pawa2,Fiona Nimmo2,Vernie Ramalingam2,Sweta Sethi1,Sanjay Mistry1,Kunal Vyas3,David Tuch4,Alan Britten5,Mieke Van Hemelrijck1,Gary Cook1,Sarah Allen2,Christopher Sibley-Allen2,Sarah Pinder1,Arnie Purushotham1
1King's College London,2Guy's and St Thomas' NHS Foundation Trust,3Sagentia,4Lighpoint Medical Ltd,5St George's University Hospital NHS Foundation Trust



CLI combines optical and molecular imaging properties by detecting visible light emitted by 18F-FDG accumulating in cancer cells. Its high-resolution and portable imaging equipment make it attractive for intraoperative margin assessment.


Following informed consent, 23 BCS patients received 5 MBq/kg of 18F-FDG pre-operatively. An increased technetium-99m dose was used to ensure SLN detection. Following excision, specimens were inked for orientation, incised, and imaged using an investigational CLI imager (Lightpoint Medical Ltd, UK). The first 10 patients constituted a run-in dataset that was used to optimize the imaging protocol; the following thirteen patients were included in the analysis dataset. Images were analysed by calculating mean radiance ± std (photons/s/cm2/sr) and tumour-to-background ratios (TBRs). Tumour size and margin status (as informed by CLI images) were independently estimated by two surgeons. CLI results were correlated with histopathology. A one-sample Wilcoxon test and Bland-Altman analysis were used to determine the degree of agreement on tumour size between CLI and histopathology. Agreement between surgeons was evaluated using an intraclass correlation coefficient (ICC) analysis. Radiation doses to staff were measured using body dosimeters.


Elevated radiance was identified in 10/13 patients. Mean radiance and TBR were 560 ± 160 photons/s/cm2/sr and 2.41 ± 0.54. In total, fifteen margins were assessed. Complete concordance between CLI and histopathology was observed with regards to identification of clear margins of excision. A non-significant underestimation of invasive tumour size on CLI was found (p = NS). Whole tumour size was significantly smaller on CLI (p ≤ 0.026). Agreement between surgeons was excellent (ICC = 0.97). SLNs were successfully detected in all patients. Radiation dose to staff was low; operating surgeons received the highest dose at 33 ± 15 μSv.


Intraoperative 18F-FDG CLI is a feasible and low risk procedure that shows promise in the intra-operative assessment of resection margins in breast-conserving surgery.