EMIM 2019
To search for a specific ID please enter the hash sign followed by the ID number (e.g. #123).

Nuclear Imaging in Diagnosis & Treatment of Cancer

Session chair: Marleen Keyaerts (Brussels, Belgium); Sven Hermann (Münster, Germany)
Shortcut: PS 01
Date: Wednesday, 20 March, 2019, 12:30 p.m.
Room: ALSH | level 0
Session type: Parallel Session


Click on an contribution to preview the abstract content.

12:30 p.m. PS 01-01

Introductory Lecture

Tim Witney

London, UK

This talk provides an overview of state-of-the-art research and refers to the following presentations selected from abstract submissions.

12:48 p.m. PS 01-02

SGLT2 is a diagnostic and therapeutic target for early-stage lung adenocarcinoma (#500)

Brendon Villegas1, Gihad Abdelhady1, Jie Liu1, Dean Wallace1, David Elashoff1, Jane Yanagawa1, Denise Aberle1, Jorge Barrio1, Steven Dubinett1, David Shackelford1, Claudio R. Scafoglio1

1 University of California, Los Angeles, Los Angeles, California, United States of America


Lung cancer is the leading cause of cancer-related death, and adenocarcinoma (LUAD) is the most frequent histology. Early diagnosis and treatment are essential; however, there are no targeted approaches for early LUAD. An important hallmark of cancer is the increased glucose uptake via GLUTs, imaged in cancer by positron emission tomography (PET) with 2-[18F] fluorodeoxyglucose (FDG). FDG PET is standard for staging, but has low sensitivity for early-stage LUAD. We found that early LUADs use a novel system of metabolic supply not detected by FDG PET: the sodium glucose transporter 2 (SGLT2)1.


We characterize SGLT2 expression and activity in human samples and in mouse models. We analyzed SGLT2 and GLUT1 expression by immunohistochemistry in 60 archival samples of human LUAD, in a Kras-driven, p53-null genetically engineered murine model (GEMM), and in 4 different patient-derived xenograft (PDX) models. We used microPET/CT imaging with the SGLT-specific tracer methyl-4-[18F] fluorodeoxyglucose (Me4FDG) to quantify SGLT activity in vivo in GEMMs and PDXs, and to compare Me4FDG and FDG uptake. Finally, we performed therapeutic trials with a specific SGLT2 inhibitor in GEMMs and PDXs. In the PDXs, we performed Me4FDG and FDG PET/CT before and after the treatment. We are now starting to recruit patients diagnosed with lung nodules for a pilot study of Me4FDG PET in humans.


SGLT2 is expressed in human lung premalignancy and early-stage LUADs, which are GLUT1-negative2. Me4FDG detects early-stage, FDG-negative LUAD in mouse models2. Targeting SGLT2 with FDA-approved inhibitors significantly reduces tumor growth and prolongs survival in GEMMs and PDXs, confirming an important role of SGLT2 in early-stage LUAD2. Me4FDG uptake predicts the rate of tumor growth and reliably monitors the response to SGLT2 inhibitors2.

 The reliance of early stage LUAD on SGLT2 opens exciting opportunities. The National Lung Screening Trial showed a 20% reduction in lung cancer mortality in high risk individuals using computed tomography (CT). CT is highly sensitive for detecting lung nodules, but has low specificity, especially for lesions presenting as ground-glass opacities. These can be benign lesions or early LUAD, and may persist for years before transforming into invasive disease. New biomarkers to predict the malignant potential of these nodules are needed.


We conclude that in vivo detection of SGLT activity is a promising marker to diagnose early-stage LUAD and to monitor the response to treatment with SGLT2 inhibitors. Our next step is the performance of pilot trials of Me4FDG PET in patients with lung ground glass opacities.


1Scafoglio et al. Proc Natl Acad Sci U S A 2015;112(30):E4111-4119

2Scafoglio et al. Sci. Transl. Med. 10, eaat5933 (2018)


This study was supported by the following grants: NIH/National Center for Advancing Translational Science (NCATS) UCLA Clinical and Translational Science Institute KL2 Translational Science Award, UL1TR001881; Integrated Molecular, Cellular, and Imaging Characterization of Screen-Detected Lung Cancer, NCI 1U01CA196408; American Cancer Society Research Scholar Grant 130696-RSG-17-003-01-CCE; Tobacco-Related Disease Research Program High Impact Research Project Award 2016TRDRP0IR00000143977; STOP Cancer Foundation Seed Grant; Saul Brandman Foundation grant.

Keywords: lung cancer, early diagnosis, cancer metabolism
1:00 p.m. PS 01-03

[89Zr]Zr-DFO-Panitumumab for the evaluation of EGFR expression in mouse models of pancreatic ductal adenocarcinoma (#316)

Sophie Poty1, Komal Mandleywala1, Jason S Lewis1

1 Memorial Sloan Kettering, Radiology, New York, New York, United States of America


Panitumumab is an anti-Epidermal Growth Factor Receptor (EGFR) monoclonal antibody that was FDA approved in 2006 for the treatment of EGFR-positive metastatic colorectal cancer.1 Panitumumab was previously radiolabeled with 89Zr/111In and its potential as a PET/SPECT tracer was demonstrated in colorectal, skin and ovarian carcinomas mouse models. Pancreatic ductal adenocarcinoma (PDAC) currently lacks non-invasive diagnostic tools and EGFR is expressed in 55% of patient samples,2 [89Zr]Zr-DFO-panitumumab is therefore investigated for the evaluation of EGFR expression in PDAC mouse models.


PDAC xenografted mouse models were screened by immunohistochemistry for EGFR expression. Panitumumab was functionalized with the DFO chelator and radiolabeled with 89Zr. Stability study in PBS and human serum were performed. Cell membrane binding and internalization of [89Zr]Zr-DFO-panitumumab was evaluated in EGFR-positive cells. [89Zr]Zr-DFO-panitumumab (11.0 MBq, 50 μg) was intravenously injected in mice bearing subcutaneous and orthotopic EGFR-positive xenografts (e.g. BxPC3, AsPC-1, Suit-2). PET/CT images were acquired at 4h, 24h, 48h, 72h and 144h post-injection. After the last imaging time point, mice were sacrificed, organs of interest were collected, weighted and counted on a gamma counter. Blocking studies were performed with the co-injection of 300 μg of Panitumumab.


Immunohistochemistry of PDAC mouse xenografts revealed strong EGFR expression in a broad range of PDAC tumors including AsPC1, BxPC3 and Suit-2. Radiolabeling with 89Zr yielded the desired radioimmunoconjugate with excellent radiochemical yield (>99%) and good specific activity (220 MBq/mg). [89Zr]Zr-DFO-panitumumab showed prolonged stability (>93%) up to 6 days. Cell binding assay on AsPC-1, BxPC3 and Suit-2 cell revealed internalization of the radioconjugate (23.3-39.8% of added radioactivity) and membranous retention (2.5-8.6% of added radioactivity). PET images showed a strong and persisting accumulation of the radioimmunoconjugate at the tumor (Fig1B). Biodistribution confirmed the high tumor uptake (38.6 ± 9.1%ID/g in BxPC3) at 144 h post-injection (Fig1C). Blocking study with co-injection of excess panitumumab (300 μg), resulted in a consistent decreased tumor uptake (16.6 ± 3.4%ID/g in BxPC3) and validated the specificity of the radioconjugate for EGFR (Fig1D).


Our study highlights the potential of [89Zr]Zr-DFO-panitumumab as an immuno-PET tracer for the non-invasive evaluation of EGFR expression in PDAC. Such molecular imaging approaches are critical at the clinical stage to guide therapy selection and optimal dose finding. Panitumumab's high affinity for the EGFR receptor should also be evaluated in a theranostic approach with the conjugation of therapeutic isotopes for targeted radionuclide therapy.


1. Ray GL, Baidoo KE, Wong KJ, et al. Preclinical evaluation of a monoclonal antibody targeting the epidermal growth factor receptor as a radioimmunodiagnostic and radioimmunotherapeutic agent. Br J Pharmacol. 2009;157(8):1541-1548.

2. Guo M, Luo G, Liu C, et al. The prognostic and predictive role of Epidermal Growth Factor Receptor in surgical resected pancreatic cancer. Int J Mol Sci. 2016;17(7).


The authors gratefully acknowledge the Radiochemistry and Molecular Imaging Probes core facility, which was supported in part by NIH, grant P30 CA08748. We gratefully acknowledge Mr. William H. and Mrs. Alice Goodwin and the Commonwealth Foundation for Cancer Research, the Center for Experimental Therapeutics of Memorial Sloan Kettering Cancer Center (JSL), the National Institutes of Health (U01 CA221046, JSL), and the fellowship from the François Wallace Monahan Fellowship from the JLM Benevolent Fund (SP).

Figure 1
Keywords: EGFR, PET imaging, Preclinical, Pancreatic cancer
1:12 p.m. PS 01-04

Translational tracking of a cell/gene cancer therapy using 89Zr-oxine and PET imaging (#91)

P. Stephen Patrick1, Krishna Kolluri2, Adam Edwards2, Beth Sage2, May Zaw Thin1, Thomas Sanderson3, John Dickson3, Sam Janes2, Mark Lythgoe1, Tammy Kalber1

1 University College London, Centre for Advanced Biomedical Imaging, Division of Medicine, London, United Kingdom
2 University College London, Lungs for Living Research Centre, Division of Medicine, London, United Kingdom
3 University College London Hospital, Institute of Nuclear Medicine, London, United Kingdom


The unknown bio-distribution dynamics of cell therapies in patients leads to questions of safety and scientific understanding of clinical trials. We are evaluating cord-derived MSCs expressing the anti-cancer protein TRAIL (TNF-Related Apoptosis Inducing Ligand) as a new therapy for lung adenocarcinoma1 in a first-in-man phase I/II clinical trial (TACTICAL; Targeted stem Cells expressing TRAIL as therapy for lung Cancer). In a phase II arm of TACTICAL, we plan to use 89Zr-oxine to label MSCs2, and PET to monitor bio-distribution. We present here our pre-clinical demonstration of feasibility.


Human cord-derived MSCs were transduced with TRAIL-encoding lentivirus at >95%. MSCs were labelled with 89Zr-oxine doses (79 to 332 kBq/106 cells) approximating 28 to 116 MBq per 70kg patient given 5x106 cells/kg. MSC phenotype (CD73, CD90, CD105 +Ve; HLAII, CD14, CD34, CD45 -Ve) was probed with flow cytometry, and ATP and NADH metabolism via Cell Titre Glo and XTT assays. TRAIL-induced apoptosis was tested via 24hr incubation of MSCs with lung cancer cell lines H28 and PC9, and measured using flow cytometry. Interpleural injection of human CRL2081 mesothelioma cells into NSG mice gave an orthotopic lung tumour model, which was detected using MRI (Bruker 1T Icon) and CT. Bioluminescence was imaged using an IVIS Lumina, and PET-CT with a Mediso Nanoscan.


Within the tested 89Zr-oxine dose range MSCs retained TRAIL expression and therapeutic capacity (A), radio-label (B), and MSC phenotype. The effects of radiolabelling on metabolism and cell cycle showed a dose and time dependent effect on proliferation suggesting labelling at <100 kBq/106 cells, or ~37 MBq per patient will balance signal and toxicity. In an orthotopic model of lung cancer i.v.-injected 89Zr-oxine labelled TRAIL-MSCs co-localised with lung tumours, as visualised with MRI and PET/CT (1C-E). MSCs were tracked up to 1 week (F), revealing their migration dynamics to the lungs, liver, and spleen (F,G). Using separate GFP-Luciferase transduced cord MSCs we confirmed viable cell location in vivo correlated up to a week with PET signal using bioluminescence imaging, and in ex vivo cryosections using fluorescence microscopy and matched autoradiography. Organ-specific and whole-body effective doses were estimated for patients using OLINDA software and biodistribution data.


Cord-derived MSCs were labelled with 89Zr-oxine in a clinically relevant time of <40 min between thawing and injection. Longitudinal imaging with PET/CT shows delivery and retention dynamics across the body.  This promises to provide valuable feedback on target and off-target cell delivery and migration within the TACTICAL trial. This will give insight on mechanism of action and patient-specific responses for this and other cell-based therapies.


1. Sage et al. 2014 Thorax. 2. Ferris et al. 2014 Dalton Trans.

Figure 1
(A) TRAIL-MSCs induce apoptosis in TRAIL-sensitive (H28 mesothelioma) and partially resistant (PC9 adenocarcinoma) human lung cancer cells as effectively as recombinant TRAIL (50 ng/mL) and were unaffected by 89Zr-oxine labelling at low (79 kBq/106 cells) or high (332 kBq/106 cells) 89Zr-oxine doses. (B) TRAIL-MSCs retain 89Zr over 1 week. Orthotopic lung xenografts visualised with (C) MRI (T2 RARE, TE = 55 ms), (D) X-ray CT, (E) CT overlaid with 89Zr PET showing co-localisation of TRAIL-MSCs. (F) Full body BLI and PET/CT imaging of 89Zr-oxine and GFP-luciferase labelled MSCs over 7 days. Ex vivo autoradiography (G) and (H) fluorescence microscopy showing  location of cells in lung tissue (I) % Injected dose/organ over 7 days shows the distribution dynamics of labelled TRAIL MSCs. Points show means, error bars SD (n=3).
Keywords: Cell Tracking, Stem Cells, MSCs, Cell Therapy, PET
1:24 p.m. PS 01-05

[18F]FPyGal: In Vivo Imaging and Monitoring of Tumor Senescence with an emerging ß-Galactosidase specific PET Tracer (#45)

Benyuan Zhou1, Jonathan Cotton1, Katharina Wolter2, 3, Johannes Schwenck1, 4, Anna Kuehn1, Kerstin Fuchs1, Andreas Maurer1, Christian la Fougère4, Lars Zender2, 3, Marcel A. Krueger1, Bernd J. Pichler1

1 Eberhard Karls University of Tuebingen, Department of Preclinical Imaging and Radiopharmacy, Werner Siemens Imaging Center, Tuebingen, Germany
2 University Hospital Tuebingen, Department of Internal Medicine VIII, Tuebingen, Germany
3 Eberhard Karls University of Tuebingen, Institute of Physiology, Department of Physiology I, Tuebingen, Germany
4 Eberhard Karls University of Tuebingen, Department of Nuclear Medicine and Clinical Molecular Imaging, Tuebingen, Germany


Senescence, a process by which cells cease proliferation and enter a state of stable cell-cycle arrest without cell death, can be induced by various stress factors during cancer treatment and conversely influencing the therapeutic outcomes. Therefore it is of great value to have an in vivo monitoring tool to detect senescent cells in cancer patients. Here we developed the PET tracer [18F]FPyGal for non-invasive imaging of tumor senescence through one of the major biomarkers - elevated Senescence-associated ß-galactosidase (SA-ß-gal) activity.


In vitro, senescence was induced in HCT116 cells by doxorubicin, in a liver progenitor cell line by p53-reactivation and in two liver carcinoma cell lines by a ribosomal checkpoint inhibitor (RCI). Cells were incubated with [18F]FPyGal and the radioactivity was measured in a gamma counter.

In vivo, the cell lines described previously were injected s.c. in mice. To induce senescence, tumor bearing mice were appropriately treated with doxorubicin, doxycycline and RCI, respectively. PET/MR measurements were performed after i.v. injection of [18F]FPyGal.

After toxicology studies in rats, a pilot first-in-man study was performed in a colorectal cancer patient with liver metastases treated with senescence inducing alisertib.


In vitro, tracer uptake increased significantly in all four cell lines compare to the respective non-senescent cells. For one of the liver carcinoma cell lines we observed more than 4-fold increase.

In vivo, the tracer uptake in the tumors (ID%/cc) and the tumor-to-muscle ratios (TMR) were quantified. In line with the in vitro results, we had significantly increased tracer uptake in all four tumor models compare to the respective non-treated tumors. Additionally, ex vivo autoradiography and X-gal staining confirmed the correlation between tracer uptake and SA-ß-gal activity.

The compound passed the toxicology tests. A first patient study showed elevated uptake in a liver metastasis under treatment with the senescence inducer alisertib.


[18F]FPyGal showed increased uptake in vitro and in vivo in senescent cells and tumors, as well as low toxicity and promising results in a first-in-man study. Additional clinical trials are currently under preparation to determine the clinical value of this approach.

Keywords: Senescence, Cancer, SA-ß-gal, PET tracer
1:36 p.m. PS 01-06

Analysis of intracellular uptake, processing and distribution of radiolabeled polymersomes and in vivo bio distribution (#410)

Stefan J. Roobol1, 2, Thomas A. Hartjes3, Robin M. de Kruijff7, Janneke Molkenboer-Kuenen8, Sandra Heskamp8, Guzman Torrelo7, Marion de Jong2, Roland Kanaar1, 4, Dik C. van Gent1, Adriaan B. Houtsmuller3, Antonia G. Denkova7, Martin E. van Royen3, 6, Jeroen Essers1, 4, 5

1 Erasmus MC, Molecular Genetics, Rotterdam, Netherlands
2 Erasmus MC, Radiology & Nuclear Medicine, Rotterdam, Netherlands
3 Erasmus MC, Pathology/Erasmus Optical Imaging Centre (OIC), Rotterdam, Netherlands
4 Erasmus MC, Radiation Onocoloy, Rotterdam, Netherlands
5 Erasmus MC, Vascular Surgery, Rotterdam, Netherlands
6 Erasmus MC, Cancer Treatment Screening Facility (CTSF), Rotterdam, Netherlands
7 Delft University of Technology, Radiation Science and Technology, Delft, Netherlands
8 Radboud University Medical Centre, Radiology & Nuclear Medicine, Nijmegen, Netherlands


Polymersomes, composed of amphiphilic block copolymers PB-b-PEO (polybutadiene - d - polyethylene oxide), have emerged as promising robust customizable nano-carriers for high-LET radionuclides in radionuclide therapy. In this study, we analyzed on the uptake mechanism of polymersomes in cells in vitro using high-content screening , confocal (live) cell and super resolution microscopy. In addition, we evaluate the DNA damage induction in vitro and bio distribution in vivo using radioisotope labeling of these novel nano-carriers.


Polymersomes were formed using an inverse nanoprecipitation method and characterization was done by Cryo-Transmission Electron Microscopy (TEM) and Dynamic Light Scattering (DLS). We compared epithelial, cancer and macrophage cell lines using high-content screening and confocal microscopy. In addition, co-localization experiments were performed using cells transfected and incubated with Rab4A or Lysotracker for endocytic analysis, respectively. Structured Illumination Microscopy (SIM) was used for intracellular distribution analysis. 53BP1-GFP was used as DNA damage marker to evaluate DNA damage capacity of 231Bi labeled polymersomes. Polymersomes were  loaded with 111In for subsequent in vivo bio-distribution in MDA-MB-231 tumor bearing and control Balb/c nude mice.


Formed polymersomes showed a size distribution around 100 nm in diameter, measure by DLS. Using fluorescent dyes for polymersome labeling we reached up to 7,64E13 polymersomes/ml for in vitro use. High-content screening showed polymersome uptake is affected by size (smaller than 100 nm), concentration (higher than 2E10) and specific cell type (large vs. small cells). Live cell imaging demonstrated polymersome uptake events and quick microtubule mediated processing, reaching intracellular speeds up to 1.0 μm/s. Co-localization of polymersomes and early endosomes (Rab4a-YFP) or lysosomes (Lysotracker) showed maturation of endosomes to lysosomes containing polymersomes 3 hours post uptake (20% vs. 50%). In cells with intracellular 231Bi labeled polymersomes we see 2-fold DNA damage induction compared to no intracellular polymersomes. Bio distribution analysis showed no difference in polymersome size and high uptake in liver, spleen and bone-marrow compared to limited uptake in the tumor.


With a combination of high-content screening, confocal and super resolution we determined the endocytic uptake, intracellular processing and distribution of fluorescent polymersomes in vitro using various cell types. Radiolabeled polymersomes show  DNA damage capabilities limited to uptake, enabeling very specific treatment options. Tumor specific uptake needs optimization for further treatment options.

Rapid uptake, microtubule processing and co-localization of the endocytic pathway of polymersomes
Additive figure 1
Polymersome bio distribution
A) Balb/c nude mice (n=5) bearing MDA-MB-231 tumors were injected intra-venous with polymersomes (100 nm, 111In-labaled, 15-20 MBq) and imaged at 4  and 24 hours post injection (h.p.i.) using SPECT/CT. (B) Animals were sacrificed at 24 hours post injection. Organs and tumor were harvested, weighted and radioactivity was measured in a Gamma Counter. 
Keywords: Nano carriers, biodistribution, microscopy, alpha particle irradiation, DNA damage
1:48 p.m. PS 01-07

Imaging DNA damage response to 177Lu-DOTATATE radionuclide therapy (#427)

Edward O'Neill1, James Knight1, Veerle Kersemans1, Samantha Terry2, Nadia Falzone1, Simon Smart1, Marion de Jong3, Julie Nonnekens3, Katherine Vallis1, Bart Cornelissen1

1 University of Oxford, Cancer Research UK and Medical Research Council Oxford Institute for Radiation Oncology, Oxford, United Kingdom
2 King’s College London, Department of Imaging Chemistry and Biology, School of Biomedical Engineering and Imaging Sciences, London, United Kingdom
3 Erasmus Medical Center, Department of Radiology and Nuclear Medicine, Rotterdam, Netherlands


Despite the success of 177Lu-DOTATATE treatment in patients with neuroendocrine tumours, there is still much potential to improve standard treatment of 4 x 7.4 GBq. Although physical dosimetry is needed, it cannot answer the fundamental question: ‘what is the biological effect and long-term outcome of the treatment’? The DNA damage repair (DDR) marker γH2AX has been proposed as a biodosimeter of radiation-induced toxicity [1]. We previously demonstrated that a radiolabelled antibody-conjugate, anti-γH2AX-TAT, can non-invasively image DDR after external beam radiation by in vivo SPECT [2].


CA20948 SSTR-positive xenograft-bearing Balb/c nu/nu mice (n=5 per treatment group), were intravenously administered 177Lu-DOTATATE (20 MBq), with subsequent intravenous injection of 111In-anti-γH2AX-TAT or non-selective 111In-IgG-TAT control (5 MBq, 5 μg). Dual isotope SPECT imaging was undertaken at various times up to 72 h, followed by ex vivo analysis, including autoradiography and γH2AX foci counting. Clonogenic survival of CA20948 and other neuroendocrine cancer cells following in vitro exposure to 177Lu-DOTATATE was compared to their DNA damage repair response.


Autoradiography and immunohistochemistry showed that 177Lu-DOTATATE administration resulted in heterogeneous uptake of 177Lu in tumour tissue. The number of γH2AX foci per cell was higher in areas of increased 177Lu uptake (Pearson correlation coefficient: 0.84 [P<0.0001]; Figure 1).

Ex vivo and SPECT VOI analysis at 72 h showed a significantly higher uptake of 111In-anti-γH2AX-TAT in 177Lu-DOTATATE-treated versus untreated animals (7.7±0.9 vs. 5.9±1.1%ID/mL; P = 0.04) (Figure 2). No increase by 177Lu-DOTATATE treatment was observed for tumour uptake of 111In-IgG-TAT (5.6±1.5 vs. 4.9±1.2%ID/mL, respectively; P > 0.05).


111In-anti-γH2AX-TAT allows in vivo imaging of DNA damage response following 177Lu-DOTATATE peptide receptor radionuclide therapy.


[1]        Denoyer, D. et al. J. Nucl. Med. 2015, 56, 505–511.
[2]        Cornelissen, B. et al. Cancer Res. 2011, 71, 4539–4549.

Figure 1 - Ex vivo analysis

Figure 1 – ex vivo analysis of CA20948 tumour cyrosections (10 µm thick) from mice treated with just 177Lu-DOTATATE (20 MBq) at 72 h, immunofluorescence with anti-γH2AX antibody (a) and (b) autoradiography on an adjacent slide.

Figure 2 – Dual isotope SPECT

Figure 2 – Dual isotope SPECT 111In-anti-γH2AX-TAT (5 MBq/5 µg) and 177Lu-DOTATATE (20 MBq) in CA20948 tumour bearing mice at 72 h.

Keywords: PRRT, Lu-177, antibody, DOTATATE, DNA damage repair