15th European Molecular Imaging Meeting
supported by:

To search for a specific ID please enter the hash sign followed by the ID number (e.g. #123).

Intra-Operative Imaging

Session chair: Victor Goncalves (Bourgogne, France); Elnaz Yaghini (London, UK)
Shortcut: PS 07
Date: Wednesday, 26 August, 2020, 12:00 p.m. - 1:30 p.m.
Session type: Parallel Session


Abstract/Video opens by clicking at the talk title.

12:00 p.m. PS 07-01

Introductory Lecture

Sophie Hernot1

1 Vrije Universiteit Brussel, Laboratory for In-vivo Cellular and Molecular Imaging, Brussels, Belgium

Keywords: Intra-Operative Imaging, Introductory talk, EMIM 2020
12:18 p.m. PS 07-02

A proof-of-concept real-time fluorescence-lifetime-guided surgery imaging system

Hans Ingelberts1, Thomas Lapauw1, Thomas Van den Dries1, Pieterjan Debie2, Sophie Hernot2, Maarten Kuijk1

1 Vrije Universiteit Brussel, ETRO-LAMI, Elsene, Belgium
2 Vrije Universiteit Brussel, BEFY-ICMI, Jette, Belgium


Fluorescence has emerged as a safe and real-time surgical guidance tool being regularly used for imaging blood perfusion, with many more applications such as tumor imaging in various stages of research and clinical translation. The spectral fluorescence imaging employed in current (pre-) clinical guidance systems faces limitations such as limited specificity and multiplexing, and background fluorescence. Fluorescence-lifetime imaging offers a potential solution to many current issues but is challenging because of the near-infrared and (sub-)nanosecond lifetime nature of the typical dyes used.


We propose a system, optimized for fluorescent dyes in the “800nm-channel” such as ICG and IRDye 800, consisting of our fast-gated CAPS camera1,2 and pulsed illumination from a supercontinuum LASER (NKT Photonics SuperK Fianium) providing 25 mW of illumination power in a 20 nm bandwidth centered around 780 nm to a flexible "pointer-style" illuminator, diffusing the light under a 20° scatter angle. The pointer can be moved far (larger illumination field) or close (larger illumination power density) to the object of interest to enable adequate results under variable conditions. The camera has an excitation block filter (NF785-33) and emission pass filter (FF01-835/70-25) in front of the lens. A time-domain multi-gate scheme is employed to resolve lifetimes under variable pointer distances.


We demonstrate realtime fluorescence lifetime imaging, in vitro and in vivo, of fluorescent dyes ICG and IRDye 800CW in clinically relevant concentrations in an open setup, exposed to lab room lights.
In normal imaging mode, the fluorescence lifetime is mapped to a typical rainbow color gradient and we are able to clearly distinguish both dyes even though they have a lifetime difference of only 200 ps. In a “high-contrast” mode we map the expected lifetimes of the two dyes to two contrasting colors to demonstrate the dye multiplexing potential.


We show that by using our CAPS camera, realtime NIR fluorescence lifetime imaging becomes a possibility. A pointer style illumination in combination with real-time multi-window lifetime analysis can make the high-speed pulsed illumination power requirements feasible.

[1] H. Ingelberts et al., “A proof-of-concept fluorescence lifetime camera based on a novel gated image sensor for fluorescence-guided surgery,” in Proc. SPIE 10862 (2019)
[2] T. Lapauw et al., "Sub-nanosecond time-gated camera based on a novel current-assisted CMOS image sensor," in Proc. SPIE 1092506 (2019)
Keywords: Fluorescence-guided surgery, fluorescence lifetime, CAPS camera
12:30 p.m. PS 07-03

A first-in-human study of multispectral endoscopy for delineation of adenoma in pituitary surgery

Dale J. Waterhouse1, 2, Daniele Borsetto3, Thomas Santarius3, James Tysome3, Sarah E. Bohndiek1, 2

1 University of Cambridge, Department of Physics, Cambridge, United Kingdom
2 University of Cambridge, Cancer Research UK Cambridge Institute, Cambridge, United Kingdom
3 Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom


Pituitary adenomas are benign tumours of the pituitary gland, a pea-sized organ situated behind the nose, attached to the base of the brain. During transsphenoidal endoscopic surgery to remove pituitary adenomas, it is important to distinguish normal pituitary tissue from adenoma, both to maximize the completeness of resection, and to minimize damage to healthy tissue, thus preserving endocrine function (Fig 1A). Unfortunately, the current standard of care (SOC) for intraoperative imaging, white light endoscopy, often displays low contrast between healthy pituitary tissue and adenoma.


Multispectral imaging (MSI), which allows simultaneous collection of morphological (spatial) and biochemical (spectral) information, can help to more effectively delineate tissue types. This motivated the design and construction of a clinically translatable endoscope capable of capturing multispectral images during surgery. This device employs a spectrally resolved detector array (SRDA) with 9 spectral filters coupled to a standard 4mm rigid endoscope for in vivoimaging (Fig 1B).

Subjects (enrolment still open, n=8 enrolled to date) due to undergo transsphenoidal surgery to resect pituitary adenoma were enrolled in a pilot clinical study (REC reference 18/EE/0165). MSI was performed before, during and after resection by switching from the SOC to the MSI endoscope (Figs 1C, 1D).


SOC images were captured prior to each MSI session. Post-surgery, the SOC images were manually annotated with regions of interest (ROIs) by the surgeon (Fig 2A), principally to delineate pituitary tissue and adenoma where possible. Multispectral images were manually coregistered with the annotated SOC images and the ROIs transferred to the multispectral images (Fig 2B), allowing spectra for pituitary and adenoma to be determined from the annotated regions (Fig 2C).

Using the spectra for pituitary tissue from an interim analysis of the patient data acquired so far, pixel-by-pixel spectral angle mapping was applied to all multispectral images (Fig 2D). For each pixel, the spectral similarity to pituitary tissue, as quantified by (1-cosθ) where θ is the spectral angle between them, was overlaid onto the multispectral images to create abundancy maps highlighting pituitary tissue (Fig 2E). These maps show promise for delineating the pituitary gland and compare favourably to SOC images.


Following this promising interim analysis, we plan to recruit 12 additional patients over the next 6 months. Should the results of the completed study recapitulate those seen in this interim analysis, this technique has the potential to increase contrast for pituitary tissue and adenoma, thus facilitating more complete of resection of adenoma, whilst conserving healthy pituitary tissue, thus improving clinical outcomes in these patients.

AcknowledgmentThis work was funded by CRUK (C14303/A17197, C9545/A29580, C47594/A16267, C47594/A21102, C55962/A24669), EPSRC (EP/N014588/1, EP/R003599/1) and the EU FP7 agreement FP7-PEOPLE-2013-CIG-630729.
Figure 1. Multispectral imaging of adenoma in pituitary surgery (MAPS) schematic
An overview of the MAPS study. A. Schematic of transsphenoidal endoscopic resection of pituitary adenoma. B. A novel multispectral endoscope was designed and built to increase contrast during surgery. This device employs a spectrally resolved detector array (SRDA) with 9 spectral filters coupled to a standard 4mm rigid endoscope for in vivo imaging. C. The multispectral endoscope camera head can easily be switched with the standard of care (SOC) camera head. D. The camera heads are switched at three points during the surgery to allow multispectral imaging of pituitary and tumour tissue.
Figure 2. Data analysis pipeline for multispectral images captured in the MAPS study

A. The SOC images are annotated by a surgeon post-surgery. B. These images are manually coregistered with the multispectral images and the ROIs transferred to multispectral images selected for their similarity to the SOC images. C. This allows endmember spectra to be determined from the ROIs. These are shown for 5 patients where coregistration was possible. D. The endmember spectra are used to perform pixel-by-pixel spectral angle mapping on the remaining multispectral images and produce maps highlighting pituitary tissue. E. These maps show promise for delineating pituitary tissue.

Keywords: endoscopy, intraoperative, pituitary adenoma, multispectral, clinical trial
12:42 p.m. PS 07-04

Optimization of a fluorescent somatostatin analog for high-contrast intraoperative imaging

Servando Hernandez Vargas1, Solmaz AghaAmiri1, Michael P. Luciano2, Sukhen C. Ghosh1, Sarah Kim1, Martin J. Schnermann2, Ali Azhdarinia1

1 The University of Texas Health Science Center at Houston, Institute of Molecular Medicine, McGovern Medical School, Houston, United States of America
2 National Cancer Institute, Chemical Biology Laboratory, Center for Cancer Research, Frederick, United States of America


Fluorescence-guided surgery (FGS) has shown increasing clinical utility for intraoperative tumor imaging (1). Patients with neuroendocrine tumors (NETs) are excellent candidates for FGS since (i) the current surgical standard-of-care leads to incomplete resections in 15-45% of cases, and (ii) surgery can still improve overall survival in metastatic disease. Building on our work with the near-infrared fluorescent (NIRF) somatostatin analog, MMC(IR800)-TOC (2), we examined the relationship between dye charge, non-specific binding, and tumor contrast using the novel FGS agent, MMC(FNIR-Tag)-TOC.


The somatostatin receptor subtype-2 (SSTR2)-targeting peptide, TOC, was conjugated to a multimodality chelator (MMC) on solid-phase, conjugated to IR800 (negatively charged) or FNIR-Tag (charge balanced (3)) via click chemistry, and radiolabeled with 68Ga as previously described (2). Spectral analysis and phantom imaging were conducted to measure the optical properties of the conjugates. To examine SSTR2 binding properties, flow cytometry and radioactive uptake studies were performed using HCT116-SSTR2 and HCT116-WT cells. For in vivo studies, mice implanted with HCT116-SSTR2 xenografts were injected with 2 nmol of non-radioactive MMC(IR800)-TOC or MMC(FNIR-Tag)-TOC (n=4/agent) and imaged at 1 and 3 h p.i.. Key tissues were resected for ex vivo imaging and histological characterization.


Spectral analysis of the fluorescent conjugates revealed nearly identical absorption (772-778 nm) and emission (795-799 nm) maxima, and comparable absolute emissions of ~300,000 AUs at 0.5 μM. Similar radiant efficiency was measured by phantom imaging. Cell-based experiments showed SSTR2-mediated binding for both agents (3606±68 vs. 3003±42 FAUs and 13.4±1.6 vs. 17.6±3.0 %ID for MMC(IR800)-TOC and MMC(FNIR-Tag)-TOC, respectively), which compared favorably to 68Ga-DOTA-TOC (16.9±1.6 %ID) (Fig. 1). The effects of dye charge on pharmacokinetics were evident in vivo, as shown by the notably lower background signal for charge-balanced MMC(FNIR-Tag)-TOC (Fig. 2). While the more negatively charged MMC(IR800)-TOC had ~30% higher tumor accumulation, it also exhibited 2 to 4-fold higher off-target binding that ultimately led to higher TBRs for MMC(FNIR-Tag)-TOC. Histological and mesoscopic analyses of FFPE sections further confirmed the in vivo and ex vivo imaging results.


Incorporation of the net-neutral zwitterionic dye, FNIR-Tag, produced a fluorescent somatostatin analog with remarkably lower background signal than its IR800 counterpart. The lower off-target binding of MMC(FNIR-Tag)-TOC may enable more sensitive detection of microlesions and residual cancer in the wound bed. Furthermore, its rapid elimination from the blood and tissues makes it possible to perform FGS at earlier time points after injection.

[1] Hernot, S., et al., 2019, 'Latest developments in molecular tracers for fluorescence
image-guided cancer surgery', Lancet Oncol., 20.7: e354–e367.
[2] Hernandez Vargas, S., et al., 2019, 'Specific Targeting of Somatostatin Receptor Subtype-2 for Fluorescence-Guided Surgery', Clin. Cancer Res., Vol.25(14), pp.4332-4342.
[3] Luciano, M. P., et al., 2019, 'A Nonaggregating Heptamethine Cyanine for Building Brighter Labeled Biomolecules', ACS Chem. Biol., 14, 934−940
Figure 1

Uptake and blocking of radiopeptides in HCT116-SSTR2 and HCT116-WT cells. Blocking was done with 100-fold excess of octreotide. Data are presented as mean ± SD.

Figure 2
Representative in vivo NIRF images of fluorescently-labeled MMC-TOC conjugates at 1 and 3 h. Dashed arrows indicate tumor, solid arrows indicate kidney.
Keywords: intra-operative imaging, NIRF, somatostatin receptor, dual labeling
12:54 p.m. PS 07-05

Ex vivo assessment of multimodal PSMA-targeting ligands in human prostate cancer samples.

Yvonne Derks1, Mark Rijpkema1, Helene Amatdjais-Groenen2, Annemarie Kip1, Michiel Sedelaar3, 5, Diederik Somford4, 5, Michiel Simons6, Peter Laverman1, Martin Gotthardt1, Dennis Löwik2, Sandra Heskamp1, Susanne Lütje7

1 Radboud university medical center, Department of Radiology and Nuclear Medicine, Nijmegen, Netherlands
2 Raboud University Nijmegen, Institute for Molecules and Materials, Systems Chemistry, Nijmegen, Netherlands
3 Radboud university medical center, Department of Urology, Nijmegen, Netherlands
4 Canisius Wilhelmina Hospital, Department of Urology, Nijmegen, Netherlands
5 Prosper Clinics, Nijmegen, Netherlands
6 Radboud university medical center, Department of pathology, Nijmegen, Netherlands
7 University hospital Bonn, Department of Nuclear Medicine, Bonn, Germany


Incomplete resection of prostate cancer (PCa) may lead to disease recurrence and consequently poor patient outcome. To achieve complete resection, prostate specific membrane antigen (PSMA) targeting multimodal ligands, containing both a radiolabel and a fluorophore, might be useful for intraoperative detection and delineation of tumor tissue. Previously, we tested 12 PSMA-targeting ligands preclinically as prostate cancer imaging agents. Now, we selected two high-potential ligands and evaluated their PSMA-specificity and imaging potential in an ex vivo incubation study on human PCa biopsies.


We included 10 patients scheduled for radical prostatectomy, and obtained from the surgically removed prostate two biopsies from the tumor (one per ligand) and one from contralateral healthy prostate tissue. Biopsies were incubated with 0.08 nmol (26.3 MBq/nmol, 4 hrs) of 111In-labeled and IRDye700DX-conjugated PSMA-N064 (DOTAGA chelator) or PSMA-N140 (DOTA chelator). Cellular distribution of the ligands was evaluated macroscopically using fluorescence flatbed scanning and 111In autoradiography. Biopsies were snapfrozen, sectioned (4 µm) and stained (H&E, PSMA). Based on the H&E cryosections, tumor regions within the tumor biopsy were drawn by a pathologist. Localization of the marked tumor regions was compared with microscopic autoradiography, fluorescence, and PSMA immunohistochemistry.


In biopsies from two patients, no malignant tissue was detected and they were excluded from the study. Macroscopic fluorescence imaging and autoradiography of 111In-PSMA-N064 and 111In-PSMA-N140 showed preferential accumulation in tumor tissue, whereas signal in healthy tissue was negligible, leading to a clear distinction between tumor and normal tissue (Fig.1A). Tumor uptake, quantified using fluorescence intensity, was similar for the two ligands (P = 0.14). Combined analysis of both ligands showed a significant difference in tracer accumulation between tumor regions (mean fluorescence intensity; 67877 ± 24385) and either a normal tissue region within the same tumor biopsy (24498 ± 8233) or the contralateral control biopsy (20586 ± 10714; P < 0.001, Fig. 1B), with a mean tumor-to-normal ratio ranging from 1.8 to 10.4. Moreover, accurate microscopic co-localization of the radiosignal, fluorescent signal and PSMA expression with the marked tumor regions was observed (Fig. 1C).


We demonstrated highly efficient PSMA-specific accumulation of  111In-PSMA-N064 and 111In-PSMA-N140 ligands in human PCa samples. Tumor regions were clearly visualized using both radionuclide and fluorescence imaging, bridging the gap to the clinical translation. In the future, these ligands may be used for intraoperative tumor detection to improve the surgical outcome of PCa patients.


This work was supported by EKFS (2016-A64) and the Dutch Cancer Society  (NKB-KWF 10443/2016-1).

High PSMA-specific accumulation of multimodal ligands in human prostate cancer biopsies
(A) Representative macroscopic fluorescence image and autoradiography of 111In-PSMA-N064 or 111In-PSMA-N140 incubated tumor biopsy and 111In-PSMA-N064 incubated contralateral control biopsy, taken directly after surgical removal of the prostate. (B) Mean fluorescence intensity of tumor regions within the tumor biopsy were compared with intensity in normal regions of the tumor biopsy and the control biopsy, as defined by a pathologist. (C) Co-localization in 4μm cryosections PSMA staining, autoradiography and fluorescence imaging with marked tumor regions (based on H&E, outlined in black).
Keywords: intraoperative, fluorescence, radionuclide, prostate cancer, ex vivo
1:06 p.m. PS 07-06

Detection of Papillary Thyroid Cancer nodal metastasis after intravenous administration of a fluorescent tracer targeting MET

Pascal Jonker1, Madelon Metman1, Luc Sondorp1, 2, Mark Sywak3, Anthony Gill4, 5, 6, Bettien van Hemel7, Liesbeth Jansen1, Paul van Diest8, Gooitzen M. van Dam10, 11, 13, Rob Coppes2, Rudolf Fehrmann12, Schelto Kruijff1

1 University Medical Center Groningen, Department of Surgical Oncology, Groningen, Netherlands
2 University Medical Center Groningen, University of Groningen, Department of Biomedical Sciences of Cell & Systems – section molecular cell biology, Groningen, Netherlands
3 Royal North Shore Hospital, Department of Endocrine Surgery and Surgical Oncology, St Leonards, Australia
4 Royal North Shore Hospital, Department of Anatomical Pathology, St Leonards, Australia
5 University of Sydney, Sydney Medical School, Sydney, Australia
6 Royal North Shore Hospital, Cancer Diagnosis and Pathology Group Kolling Institute of Medical Research, St Leonards, Australia
7 University Medical Center Groningen, Department of Pathology, Groningen, Netherlands
8 University Medical Center Utrecht, Department of Pathology, Utrecht, Netherlands
9 University Medical Center Groningen, Department of Surgery, Groningen, Netherlands
10 University Medical Center Groningen, University of Groningen, Medical Imaging Center, Groningen, Netherlands
11 University Medical Center Groningen, Department of Medical Oncology, Groningen, Netherlands
12 AxelaRx/Tracer BV, Groningen, Netherlands


Central compartment lymph node dissection (CLND) or watchful waiting (WW) are currently two strategies in papillary thyroid cancer (PTC) treatment. Choosing between these two strategies for an individual patient is difficult due to low specificity of preoperative imaging for nodal metastases (NM). CLND reduces recurrence rates (RR) but exposes pN0 patients to undesirable morbidity. Reoperation for recurrence after WW is challenging and causes even more morbidity. Molecular fluorescence guided imaging of PTC NM with EMI-137 might improve patient selection, ultimately reducing RR and morbidity.


Patients with fine needle aspiration proven PTC or recurrent disease undergoing a lymph node dissection were included in a phase 1 dose escalation study (NCT0347025) to evaluate tracer safety and optimal dosage. Adverse events during hospital admission were registered. Fresh whole specimen and formalin fixed sliced tissue (BLS) were imaged ex vivo (IVIS Spectrum, PerkinElmer). Based on hematoxylin and eosin (H&E) stained slides, median fluorescence intensities (MFI in p/sec/cm2/sr, Interquartile Range (IQR)) were determined and compared (Mann–Whitney U-test) between BLS of nodal metastases (NM) and normal lymph nodes (NLN).


A total of 15 patients were included and received 0.09 mg/kg (n=3), 0.13 mg/kg (n=6) or 0.18 mg/kg (n=6) EMI-137. Patients were treated with a total thyroidectomy (TTX) with (prophylactic) CLND (n=6), completion TTX with left PCLND (n=1), TTX, (prophylactic) CLND and uni- or bilateral lymph node dissection (n=4) or selective lymph node dissection for recurrent disease (n=4). A total of 54 NM and 237 NLN were yielded. No serious adverse events occurred. Only dosage cohorts 0.13 mg/kg (p = 0.01) and 0.18 mg/kg (p=0.02) were extended based on MFI difference between NM and NLN during interim analysis. Following dose extension, only patients receiving 0.13 mg/kg had a significant difference (p<0.0001) in MFI between 24 PTC NM (3.74x1010 p/sec/cm2/sr, IQR 8.54x1010 p/sec/cm2/sr) and 128 NLN (9.67x109 p/sec/cm2/sr, IQR 1.71x1010 p/sec/cm2/sr).


Detection of PTC nodal metastasis with 0.13 mg/kg EMI-137 is safe and feasible. Further extension of this dosage cohort will provide more insight in sensitivity and specificity of molecular fluorescence guided imaging with EMI-137 for metastasis detection.

AcknowledgmentDutch Cancer Society
Keywords: Thyroid Cancer, EMI-137, Molecular fluorescence guided imaging, nodal metastasis