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).

New Tools for Cancer Imaging

Session chair: Emmet McCormack (Bergen, Norway); Frank Denat (Dijon, France)
Shortcut: PS 13
Date: Thursday, 27 August, 2020, 10:00 a.m. - 11:30 a.m.
Session type: Parallel Session


Abstract/Video opens by clicking at the talk title.

10:00 a.m. PS 13-01

Introductory Lecture

Susanne Kossatz1

1 TU Munich, Munich, Germany

Keywords: Cancer Imaging, Introductory Talk, EMIM 2020
10:18 a.m. PS 13-02

Multiplexed whole animal imaging with reversibly switchable optoacoustic proteins

Kanuj Mishra1, Mariia Stankevych1, Juan P. F. Werner1, Simon Grassmann2, Vipul Gujrati1, 3, Uwe Klemm1, Veit R Buchholz2, Vasilis Ntziachristos3, 1, Andre C. Stiel1

1 Helmholtz Zentrum München, Institute of Biological and Medical Imaging, München, Germany
2 Technische Universität München, Immunology and Hygiene/Institute for Medical Microbiology, München, Germany
3 Technische Universität München, Chair of Biological Imaging and Center for Translational Cancer Research (TranslaTUM), München, Germany


Optoacoustic (OA) imaging enables high-resolution, real-time in-vivo imaging well beyond the 1mm penetration depth1,2 In Photo-switchable proteins the absorption bands can be modulated by light and this nodulation separate them from non-modulating background3,4. Here we report a superior near-infrared reversible switchable optoacoustic protein (rsOAP) from the wild type Bacteriophytochrome of Rhizobium etli and demonstrate its potential for simultaneous tracking of different cellular processes through temporal multiplexing in in-vivo using widely accessible commercial imaging system.


Screening and structure-guided truncation strategy was applied for the development of superior rsOAP.OA characterization of rsOAPs was done as described elsewhere5. All OA imaging was done using multispectral OA tomography device-MSOT (iThera Medical). For unmixing,machine learning-based algorithm was used. To assess the sensitivity, dorsal implants of Matrigel with different numbers of Jurkat T lymphocytes & E.Coli expressing rsOAP were imaged. Using rsOAP we also studied the development of tumor models - colorectal and breast cancer. Later,we imaged mice brain implanted with 4T1 cells expressing rsOAP. Finally, multiplexing was shown in in-vitro phantoms  and in-vivo by using 4T1 expressing fast switching rsOAP injected with either E.Coli or Jurkat T cells expressing slow switching rsOAP


The screening and truncation strategy resulted in superior rsOAP, Re-PCM (PCM = photosensory core module). RePCM is a small monomer with high absorption which shows a 2-fold larger change in OA signal,>5-fold faster switching and greater resistance to photo-fatigue(Fig.1) compare to other rsOAPs. Using RePCM as label and ML-based unmixing algorithm, we achieve a sensitivity of 500 immune-cells and 14000 E.Coli in-vivo (Fig.2). We also show the infiltration and development of tumor models over time, such as colorectal and breast cancer(Fig.2). We also identified 1.4 x 105 4T1 cells expressing fast rsOAP at the depth of 3.6mm in brain tissue (Fig.2). Finally, the fast switching speed of Re-PCM allows it to be used together with other rsOAPs with slower switching speeds effectively imaging mixed cell-populations labeled with the different rsOAPs(Fig.2). This demonstrates the potential for simultaneous tracking of different cellular processes through temporal multiplexing.


Transgenic rsOAP labels in OA allow the tracking of specific cell populations in-vivo, which opens up new possibilities for longitudinal studies in diverse fields such as immunology, cancer research, etc. here, we described next-generation rsOAPs that have faster switching and greater resistance to photo-fatigue, allowing highly sensitive detection, and importantly true multiplexing, without interference from endogenous absorbers in-vivo.

AcknowledgmentThe authors wish to thank Prof. Andreas Möglich for providing the wild-type BphPs, Ruth Hillermann for technical assistance and Armando C. Rodríguez for discussions on the manuscript.
[1] Ntziachristos, V. & Razansky, D. Molecular Imaging by Means of Multispectral Optoacoustic Tomography (MSOT). Chem. Rev. 110, 2783–2794 (2010).
[2] Wang, L. V. & Hu, S. Photoacoustic tomography: In vivo imaging from organelles to organs. Science 335, 1458–1462 (2012).
[3] Yao, J. et al. Multi-scale photoacoustic tomography using reversibly switchable bacterial phytochrome as a near-infrared photochromic probe. Nat. Methods 13, 67–73 (2016).
[4] Mishra, K., Fuenzalida Werner, J. P., Ntziachristos, V. & Stiel, A. C. Photo-Controllable Proteins for Optoacoustic Imaging. Anal. Chem. acs.analchem.9b01048 (2019). doi:10.1021/acs.analchem.9b01048
[5] Vetschera, P. et al. Characterization of Reversibly Switchable Fluorescent Proteins in Optoacoustic Imaging. Anal. Chem. 90, 10527–10535 (2018).
In vitro characterization of ReBphP-PCM and RpBphP1-PCM in comparison to another rsOAP DrBphP-PCM

a Principle of photo-switching in BphP-derived rsOAPs .

b Absorbance spectra of Pr and Pfr states of the three rsOAPs used in this study in comparison to HbO2 and Hb.

c Switching cycles of the rsOAPs at 770nm.

d Single switching-cycle from panel c.

e Photo fatigue of the proteins per cycle.

f Absorbance ratio between the Pfr and Pr state for different wavelengths.

g Absorbance (filled bars) and OA signal intensity (hollow) ratio between the Pfr and Pr a.t 770nm

h Matthew ́s coefficient shown as a function of number of cycles and pulses

MSOT imaging of Re-PCM and other rsOAPs

a Schematic of OA tomography

b 4T1 cells (0.8 x 106 injected subcutaneously) stably expressing Re-PCM

c 4T1 cells (0.7 x 106 injected intracranially) stably expressing Re-PCM imaged at a depth of 3.6 mm in the brain.

d,e HCT116 cells (1.5 x 106 injected intraperitoneally) stably expressing Re-PCM imaged on day 14, Volume representation of the same mice at consecutive time points.

f Certainty of prediction, indicating quality of discerning label signal or background of ROIs shown in(d)

g,h Different numbers of Jurkat T lymphocytes & E.Coli expressing rsOAP

i,j,k,l in-vitro and in-vivo multiplexing.


Keywords: Optoacoustics, Photocontrollable Proteins, Cancer Imaging, Machine learning, Multiplexing
10:30 a.m. PS 13-03

Water-soluble aza-BODIPYs: biocompatible organic dyes for high contrast in vivo NIR-II molecular imaging

Ghadir Kalot1, Amélie Godard2, Jacques Pliquett2, Benoit Busser1, 3, Xavier Le Guével1, David Wegner4, Franck Denat2, Jean-Luc Coll1, Ewen Bodio2, Christine Goze2, Lucie Sancey1

1 Institute for Advanced Biosciences, IAB U1209 UMR5309 UGA, Grenoble, France
2 ICMUB, UMR 6302, Dijon, France
3 Grenoble University Hospital, Grenoble, France
4 BAM institute, Berlin, Germany


Optical imaging has gained in popularity because of its use in clinical image-guided surgery, and possible integration with other imaging modalities. NIR-I imaging allows the detection of dyes with high sensitivity, but still suffers from light scattering, photon attenuation, and autofluorescence. NIR-II or SWIR imaging (>1,000 nm) provides better contrast and tissue penetration, and reduced scattering. The development of biocompatible organic dyes for NIR-II imaging is thus an active field. Here, we present a new class of aza-BODIPYs as contrast agent for molecular imaging in the NIR-II.


Aza-BODIPYs were modified by boron functionalization1 to become water-soluble and were named SWIR-WAZABY. The photophysical characteristics were evaluated in terms of absorbance and emission. Serial dilutions were performed in various media, including saline buffered solutions, BSA 40 g/L, and plasma. The safety profile of SWIR-WAZABY was studied on healthy human cells (fibroblasts, renal and endothelial cells), while its distribution was studied in human cancer cells from glioblastoma origin (including U87MG), cultured in 2D and 3D. In vivo, SWIR-WAZABY was administrated in mice for pharmacokinetics and NIR-II imaging investigation.


Before its boron functionalization, the compound possesses remarkable NIR-II fluorescence in DMSO. In saline or isotonic solutions, SWIR-WAZABY was non-fluorescent, but, when diluted in BSA and plasma, SWIR-WAZABY revealed 2 emission peaks centered around 915 and 1,280 nm. Its fluorescence properties were boosted in presence of plasma, especially when enriched by HDL or cholesterol (F2).
In vitro, SWIR-WAZABY was non-toxic, rapidly internalized in 2D and 3D cultured cells, in small cytoplasmic vesicles.
After IV injection, SWIR-WAZABY circulated several hours in the blood, was mainly eliminated by hepatic route, and did not accumulated in other organs. The tumor uptake was fast and persistent (F1). The tumor/skin ratio increased until 48h post-injection, reaching a maximum value of 4, before being stable from days 3-7, with values ~3. In parallel, the tumor/muscle ratio was around 10, reaching 30 at day7. The ex-vivo tumor analysis indicated the tumor cell internalization of SWIR-WAZABY.


The water-soluble organic fluorophore SWIR-WAZABY has attractive contrast properties for optical NIR-II imaging. The tumor uptake of SWIR-WAZABY occurred within a few hours, and was persistent during more than a week, with favorable tumor to skin/healthy tissue ratios. This compound can be coupled to specific molecules for molecular imaging. Next in vivo experiments will demonstrate its potential as a contrast agent for human applications.


The authors acknowledge the support of the French funding agencies: FLI (France Life Imaging) for the project Thera-BODIPY, the CNRS Mission for Transversal and Interdisciplinary Initiatives for the project BREVET-ISOTOP, and the GEFLUC Grenoble Dauphiné Savoie, the French Research National Agency (ANR) via project JCJC ‘‘SPID” ANR-16-CE07-0020 and project JCJC “WazaBY” ANR-18-CE18-0012 are gratefully acknowledged. This work is part of the projects ‘‘Pharmacoimagerie et agents théranostiques” and ‘‘Chimie durable, environnement et agroalimentaire” supported by the Université de Bourgogne and the Conseil Régional de Bourgogne through the Plan d’Actions Régional pour l’Innovation (PARI) and the European Union through the PO FEDER-FSE Bourgogne 2014/2020 programs.

[1] Pliquett, J.; Dubois, A.; Racoeur, C.; Mabrouk, N.; Amor, S.; Lescure, R.; Bettaieb, A.; Collin, B.; Bernhard, C.; Denat, F.; Bellaye, P. S.; Paul, C.; Bodio, E.; Goze, C., A Promising Family of Fluorescent Water-Soluble aza-BODIPY Dyes for in Vivo Molecular Imaging. Bioconjug Chem 2019, 30 (4), 1061-1066.

Fluorescence of SWIR-WAZABY in various solutions.
The NIR-II contrat agent was diluted in PBS, NaCl 0.9%, Glocuse 5%, BSA 40 g/L and plasma. Its fluorescence was collected from 1300 to 1700 nm.
In vivo distribution and behavior of SWIR-BODIPY in mice-bearing sub-cutaneous U87MG. 
SWIR-WAZABY circulated in the blood stream, was mainly eliminated by the liver, and and accumulated several days in the tumor with favorable Tumor-to-Muscle and Tumor-to-Skin ratios.
Keywords: NIR-II contrast agent, water-soluble aza-BODIPY, SWIR imaging
10:42 a.m. PS 13-04

­­Imaging system xc- activity using the novel PET radiotracer [18F]FRPG

Hannah Greenwood1, Richard Edwards1, Norman Koglin2, Mathias Berndt2, Friedrich Baark1, Jana Kim1, George Firth1, Andre Mueller2, Timothy H. Witney1

1 King's College London, Department of Imaging Chemistry and Biology, London, United Kingdom
2 Life Molecular Imaging GmbH, Berlin, Germany


The amino acid transporter system xc- is over-expressed in multiple cancer types, providing intracellular cysteine for glutathione biosynthesis. System xc- activity in tumours is altered in response to chemotherapy and in drug-resistant cells, which can be imaged by PET using the L-glutamate analogue [18F]FSPG (1-2). Little is known regarding transporter-recognition of non-natural amino acids or their utility for cancer imaging. Motivated by this, we compared the in vitro and preclinical characteristics of [18F]FSPG to its stereoisomer [18F]FRPG (Fig. 1A) in models of cancer and inflammation.


[18F]FRPG and [18F]FSPG uptake was assessed in H460 lung cancer cells (0.25MBq; 1h), with efflux measured 30 min after removal of exogenous activity. Specificity of [18F]FRPG for system xc- was further examined following transporter inhibition and blocking studies with system xc- substrates (1mM).

Tissue uptake of [18F]FRPG and [18F]FSPG was quantified in mice bearing subcutaneous A549, H460, VCAP and PC3 tumours by gamma counting 60 min post-injection. Additionally, mice bearing A549 tumours were dynamically imaged by PET/CT over 60 min (3 MBq). Lung inflammation was induced in BALB/C mice through intratracheal administration of lipopolysaccharide (LPS; 1.5mg/kg; 24h). [18F]FRPG and [18F]FSPG uptake in LPS-treated lungs were compared to PBS controls by PET (3 MBq; 40-60min post-injection).


[18F]FRPG uptake was specific for the glutamate/cystine antiporter system xc- (Fig 1B), matching that of [¹⁸F]FSPG (3). In H460 cells, [18F]FRPG uptake was lower than [18F]FSPG over 60 min (7.9±0.2 %activity/100,000 cells vs. 15.1±0.7 %activity/100,000 cells; n = 3; P = 0.004; Fig 1C). However, the percentage of radiotracer retained in cells 30 min after removal of exogenous activity was higher for [18F]FRPG than [18F]FSPG (91% vs. 69%; n = 3; P = 0.002; Fig 1D). In vivo biodistribution studies showed low [18F]FRPG retention in the blood, liver and pancreas (Fig 2A-C). Tumour uptake of [18F]FRPG was lower compared to [18F]FSPG in prostate VCAP tumours and lung A549 tumours (Fig 2D). However, [18F]FRPG tumour-to-blood was either elevated or equal to that of [18F]FSPG (Fig 2F). [18F]FSPG uptake in inflammatory cells may complicate the interpretation of tumour uptake (1). Following LPS treatment, uptake of both tracers was significantly increased compared to control mice (Fig 2H&I).


These results confirm the specificity of [18F]FRPG for system xc- and desirable properties for cancer imaging.  High [18F]FRPG uptake was measured in a range of human tumours, with background tissue uptake reduced in comparison to [18F]FSPG. LPS-induced inflammation significantly increased the uptake of both radiotracers. Future work will determine the redox sensitivity of [18F]FRPG and its ability to image response and resistance to therapy.

[1] McCormick PN, Greenwood HE, Glaser M, Maddocks ODK, Gendron T, Sander K, Gowrishankar G, Hoehne A, Zhang T, Shuhendler AJ, Lewis DY, Berndt M, Koglin N, Lythgoe MF, Gambhir SS, Arstad E, Witney TH. Assessment of tumor redox status through (S)-4-(3-[ 18 F]fluoropropyl)-L-glutamic acid PET imaging of system x c activity. Cancer Res. 2019;79:853-863.
[2] Greenwood HE, McCormick PN, Gendron T, Glaser M, Pereira R, Maddocks ODK, Sander K, Zhang T, Koglin N, Lythgoe MF, Arstad E, Hochhauser D, and Witney TH. Measurement of tumor antioxidant capacity and prediction of chemotherapy resistance in preclinical models of ovarian cancer by positron emission tomography. Clin Cancer Res. 2019;25:2471-2482.
Figure 1. In vitro characterisation of [18F]FRPG.

A. The chemical structures of [18F]FRPG and [18F]FSPG. B. [18F]FRPG uptake in H460 cells was blocked in the presence of FSPG, L-glutamate and the xCT inhibitor p-carboxyphenylglycine (CPG) (1 mM; 30 min uptake; n = 3). C. Time course of [18F]FRPG and [18F]FSPG uptake over 90 min in H460 cells (n = 3).  D. [18F]FRPG and [18F]FSPG efflux in media alone or with the addition of L-glutamate. Cell uptake was allowed to proceed for 60 min prior to removal of exogenous activity and assessment of cell-associated activity after a further 30 min. **, P<0.01; ***, P<0.001. L-Glu, L-glutamate. 

Figure 2. In vivo comparison of [18F]FRPG and [18F]FSPG.
A-D. Biodistribution of radiotracers. E. Radiotracer uptake in s/c tumour models, 60 min p.i. F. Tumour:blood ratios for [18F]FRPG and [18F]FSPG. G. PET/CT MIP images representing the same mouse bearing an A549 s/c tumour imaged with [18F]FRPG and [18F]FSPG (40-60 min p.i.). Images taken 24 h apart. H. [18F]FRPG and [18F]FSPG uptake in an LPS model of lung inflammation. Representative single slice axial PET/CT image, 40-60 min p.i. I.  Quantification of radiotracer uptake (n = 5-6/group). White arrows, tumour; SG, salivary glands; P, pancreas; B, bladder. *, P<0.05; **, P<0.01. 
Keywords: [18F]FRPG, [18F]FSPG, system xc-, positron emission tomography
10:54 a.m. PS 13-05

Repurposing 18F-FMISO for clinical PET/CT imaging of nitroreductase gene directed enzyme prodrug therapy

Elvira García de Jalón1, 2, Gorka Ruiz de Garibay1, Endre Stigen1, Mihaela Popa1, Mireia M. Safont1, Cecilie B. Rygh3, Heidi Espedal3, Torill M. Barrett4, Bengt Erik Haug2, Emmet McCormack1

1 University of Bergen, Centre for Cancer Biomarkers (CCBIO), Department of Clinical Science, Bergen, Norway
2 University of Bergen, Department of Chemistry and Centre for Pharmacy, Bergen, Norway
3 University of Bergen, Molecular imaging Center, Department of Biomedicine, Bergen, Norway
4 University of Bergen, Department of Pathology, Haukeland University Hospital, Bergen, Norway


Nitroreductase (NTR) is an enzyme used in gene directed enzyme prodrug therapy (GDEPT) that following tumor targeting and transgene expression selectively activates nitroaromatic prodrugs (e.g. CB1954) inducing cytotoxicity in the targeted cells (1). The clinical development of NTR-based GDEPT has, in part, been hampered by the lack of translational imaging modalities assessing gene transfer and drug efficacy non-invasively (2-4). This study presents translational preclinical PET/CT imaging using a clinically approved radiotracer, 18F-FMISO, as a reporter of NTR expression.


The applicability of 18F-FMISO PET/CT imaging to report NTR expression in all the steps of GDEPT was validated in NTR- and NTR+ breast carcinoma xenografts (MDA-MB-231 and L-149 NTR expression vector (5)). Tumor growth in subcutaneous (SC) models and metastatic disease progression (weeks 8-12) in orthotopic tumors were monitored. Response to CB1954 (40 mg/kg q5d x 2 i.p.) and in vivo intratumoral NTR transduction (8 x 107 TU/tumor) were evaluated in SC tumors. Fluorescence imaging (IVIS Spectrum) was performed 1 h post i.v. dosing of CytoCy5S (100 μmol), a validated NTR reporter (5). PET/CT (nanoScan, Mediso) was performed 1.5 h post i.v. dosing of 18F-FMISO (8-12 MBq). Bioluminescence imaging (In-Vivo FX Pro, Carestream) was performed 10 min post i.p. dosing of D-luciferin (150 mg/kg).


Significantly higher 18F-FMISO PET contrast was detected in NTR+ versus NTR- subcutaneous tumors (Fig. 1A), with SUVmax values at week 4 of 3.64 ± 0.13 g/ml and 1.23 ± 0.15 g/ml respectively (p<0.0001).
In the orthotopic tumors, PET signal was observed in the axillary area at week 9, reaching a maximum at week 12, with SUVmax values of 4.25 ± 1.18 g/ml.
A reduction on PET signal, SUVmax values of 3.19 ± 0.15 g/ml (p<0.0001), was detected in NTR+ tumors 3 days after CB1954 treatment, concomitant with a 40% reduction in tumor volume (Fig. 1B).
4 weeks after in vivo NTR transduction SUVmax values were significantly higher than in NTR- tumors, 5.67 ± 1.81 and 2.03 ± 0.10 g/ml respectively (p<0.05, Fig. 1A).


18F-FMISO PET/CT imaging proved useful for monitoring tumor growth, metastatic progression, response to CB1954, and in vivo NTR transduction. These results support 18F-FMISO as a readily implementable PET imaging probe to aid pre-clinical and clinical optimization of the NTR GDEPT system.

AcknowledgmentThis work was supported by The Research Council of Norway, the Norwegian Cancer Society, the Western Norway Regional Health Authority, CCBIO and the Bergen Research Foundation.
[1] Knox RJ, et al. 1988, ‘The nitroreductase enzyme in Walker cells that activates 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB 1954) to 5-(aziridin-1-yl)-4-hydroxylamino-2-nitrobenzamide is a form of NAD(P)H dehydrogenase (quinone) (EC’, Biochemical pharmacology, 37, 4671-4677
[2] Zhang J, et al. 2015, ‘Gene-directed enzyme prodrug therapy’, AAPS J, 17, 102-110
[3] Shah K, et al. 2004, ‘Molecular imaging of gene therapy for cancer’, Gene therapy, 11, 1175-1187
[4] Sekar TV, et al. 2014, ‘Noninvasive theranostic imaging of HSV1-sr39TK-NTR/GCV-CB1954 dual-prodrug therapy in metastatic lung lesions of MDA-MB-231 triple negative breast cancer in mice’, Theranostics, 4, 460-474
[5] McCormack E, et al. 2013, ‘Nitroreductase, a near-infrared reporter platform for in vivo time-domain optical imaging of metastatic cancer’, Cancer Res, 73, 1276-1286
Representative 18F-FMISO PET/CT and CytoCy5S FLI images
11:06 a.m. PS 13-06

CAIX-targeting VHHs as agents for SPECT imaging of tumor hypoxia

Sanne A. M. van Lith1, Fokko J. Huizing1, 2, Bianca Hoeben2, Sofia Doulkeridou3, Martin Gotthardt1, Paul M. P. van Bergen en Henegouwen3, Johan Bussink2, Sandra Heskamp1

1 Radboudumc, Radiology and Nuclear Medicine, Nijmegen, Netherlands
2 Radboudumc, Radiation Oncology, Nijmegen, Netherlands
3 Utrecht University, Cell Biology, Utrecht, Netherlands


Hypoxia is present in the majority of solid tumors and is associated with poor outcome and radiotherapy resistance. Carbonic anhydrase IX (CAIX) is upregulated by cells under hypoxic conditions. Non-invasive imaging of CAIX could therefore be of prognostic value and used for radiotherapy planning. The aim of this study was to optimize preclinical SPECT imaging of CAIX expression using an 111In-labeled anti-CAIX single domain antibody (VHH) B9. Furthermore, the effect of blood half-life extension on biodistribution and tumor uptake was determined by fusing B9 to an albumin binding domain (ABD).


CAIX targeting VHHs B9 and B9-ABD, provided with a C-terminal cysteine, were site-specifically functionalized with maleimide-DTPA. Upon labeling with 111In, effective inhibitory concentrations (IC50) and binding affinity (KD) of the conjugates were determined using CAIX expressing SKRC-52 cells. Subsequently, a dose-escalation study was performed for [111In]In-DTPA-B9 in athymic nude mice carrying subcutaneous SCCNij153 head and neck cancer xenografts, which are known to express CAIX in hypoxic areas. Finally, biodistribution of [111In]In-DTPA-B9 was compared to that of [111In]In-DTPA-B9-ABD. Tracer uptake in the tumor was visualized with SPECT imaging, and autoradiography images of tumor sections were analyzed for correlation with CAIX immunohistochemistry images.


Both [111In]In-DTPA-B9 and [111In]In-DTPA-B9-ABD showed CAIX specific binding to SKRC-52 cells (KD = 13.85 ± 1.03 nM and 94.12 ± 4.92 nM, respectively). Tumor uptake of [111In]In-DTPA-B9 in SCCnij153 xenografts was 0.75 ± 0.13%ID/g at 4h post injection, with an excellent tumor-to-blood ratio (8.39 ± 1.26). Fusion to ABD led to a significantly increased tumor uptake (4.11 ± 1.72%ID/g; p = 0.01), however with a significantly lower tumor-to-blood ratio (0.16 ± 0.06; p = 0.0001) at 4h post injection. Maximal tumor uptake (5.80 ± 1.02%ID/g) and tumor-to-blood ratio (2.21 ± 0.60) were observed for [111In]In-DTPA-B9-ABD at 72h post injection. Tumor uptake of both tracers could be visualized with SPECT, and co-localization analysis showed better correlation of [111In]In-DTPA-B9 localization and CAIX expression (mean R = 0.32 ± 0.22; p = 0.07) at 4h post injection compared to that of [111In]In-DTPA-B9-ABD localization and CAIX expression (mean R = 0.06 ± 0.24; p = 0.73) at 72h post injection.


We demonstrated that both [111In]In-DTPA-B9 and [111In]In-DTPA-B9-ABD bind specifically to CAIX expressing cells. [111In]In-DTPA-B9-ABD showed higher tumor uptake in vivo, but correlation with CAIX expression was better for [111In]In-DTPA-B9. In order to identify the optimal VHH-based imaging compound with high tumor uptake and CAIX specificity, experiments with alternative B9-ABD variants are ongoing.

Figure 1. Biodistribution of B9 and B9-ABD
Biodistribution study showing tracer uptake in organs and tumors as determined by ex vivo radioactivity counting at 4, 24 or 72 hours after injection of equimolar quantities of [111In]In-DTPA-B9 or [111In]In-DTPA-B9-ABD.
Figure 2. Tracer uptake and correlation with CAIX expression in vivo
Tracer (co)-localization of [111In]In-DTPA-B9 at 4 hours post injection (A) and [111In]In-DTPA-B9-ABD at 72 hours post injection (B). In both rows from left to right: in vivo SPECT imaging of the tumor, autoradiography and CAIX immunohistochemistry. The latter two are visualizations of consecutive sections of the same tumor. Note the colocalization of the tracer uptake in SPECT and autoradiography with CAIX immunohistochemistry, which is more apparent for [111In]In-DTPA-B9 compared to [111In]In-DTPA-B9-ABD.
Keywords: Hypoxia, Carbonic Anhydrase IX, VHH, SPECT
11:18 a.m. PS 13-07

Development of pH sensitive probes for in vivo Cerenkov imaging

Andrea E. Guzman2, Alejandro D. Arroyo1, Natalia Rubtcova1, Anatoliy Popov1, Kavindra Nath1, Edward J. Delikatny1

1 University of Pennsylvania, Radiology, Philadelphia, United States of America
2 University of Pennsylvania, Pharmacology, Philadelphia, United States of America


Increased acidity in the tumor microenvironment plays a role in the invasion and metastasis of cancer cells and contributes to chemotherapeutic resistance1. Cerenkov radiation emitted by β-particles can be employed as a tool for pH imaging in vivo2,3. pH-sensitive fluorophores can attenuate Cerenkov radiation’s continuous and multispectral emission through selective bandwidth quenching (SBQ) and produce Cerenkov Radiation Energy Transfer (CRET)4,5. SBQ and CRET provide a quantitative readout that can be correlated with a pH value, resulting in a non-invasive technique for pH imaging in vivo.


The pH sensor 5,6-carboxynaphthofluorescein succinimidyl ester was conjugated via nucleophilic substitution to 4-aminobutyl DOTA, in a 1:1 molar ratio, to synthesize a pH-sensitive probe suitable for chelation with Ga. The product, NFbD, was chelated with 68Ga to obtain the Cerenkov-active complex, NFbD-Ga. In vitro Cerenkov imaging was performed as a function of pH to construct standard curves. Intramolecular (100 µCi NFbD-68Ga) and intermolecular (100 µCi 68Ga mixed with NFbD-67,69Ga) data were acquired. In vivo Cerenkov imaging was performed following intratumoral injection of 40 µCi NFbD-68Ga into nude mice bearing 4175-Luc+ triple negative breast cancer xenografts. CRET values were used to estimate in vivo tumor pH and validated with 31P Magnetic Resonance Spectroscopy (MRS).


NFbD-Ga maintained its pH indicator capability after conjugation and chelation with Ga (pKa= 7.7, λex= 600 nm, λem= 669 nm at pH 9). NFbD-Ga showed increased photostability, compared to naphthofluorescein. In vitro Cerenkov imaging of NFbD-Ga showed marked SBQ and CRET at basic pH (>6.8) values measured at 600 nm and 700 nm, respectively, and standardized to 840 nm, Fig.1. Cerenkov pH titration curves showed an SBQ pKa of 6.70 and a CRET pKa of 6.96. Intermolecular Cerenkov imaging using 68Ga added to a non-radiolabeled probe exhibited identical pH-dependent SBQ and CRET properties as intramolecular Cerenkov imaging using the radiolabeled NFbD-68Ga. Cerenkov imaging with NFbD-Ga can detect pH with a sensitivity of 0.2 pH units.  SBQ and CRET are observed in vivo in mice bearing 4175-Luc+ tumors, after injection with NFbD-Ga. Preliminary in vivo average radiance data was validated by 31P MRS, resulting in tumor pH values of 6.92 ± 0.04 (n=5), Fig. 2.


Average radiance readout from Cerenkov imaging showed detectable CRET signals for each tumor. For NFbD-Ga, CRET occurs at wavelengths which are more favorable for in vivo imaging due to decreased tissue scattering and absorption. This novel method allowed for precise pH measurements of the tumor microenvironment, validating the potential of Cerenkov radiation for critical biomedical applications and expanding the realm of optical imaging.

AcknowledgmentWe would like to acknowledge the Small Animal Imaging Facility, the Cyclotron Facility at the University of Pennsylvania and our funding sources: NIH grants R01 EB018645, F31 EB029309 and T32 GM008076.
[1] Estrella, V., Chen, T., Lloyd, M., Wojtkowiak, J., Cornnell, H.H., Ibrahim-Hashim, A., Bailey, K., Balagurunathan, Y., Rothberg, J.M., Sloane, B.F., et al. (2013). Acidity generated by the tumor microenvironment drives local invasion. Cancer Res. 73, 1524–1535.
[2] Thorek, D.L., Robertson, R., Bacchus, W.A., Hahn, J., Rothberg, J., Beattie, B.J., and Grimm, J. (2012). Cerenkov imaging - a new modality for molecular imaging. Am J Nucl Med Mol Imaging 2, 163–173.
[3] Czupryna, J., Kachur, A.V., Blankemeyer, E., Popov, A.V., Karp, J.S., and Delikatny, E.J. (2015). Cerenkov specific contrast agents for detection of pH in vivo. J Nucl Med 56, 483–488.
[4] Dothager, R.S., Goiffon, R.J., Jackson, E., Harpstrite, S., and Piwnica-Worms, D. (2010). Cerenkov Radiation Energy Transfer (CRET) Imaging: A Novel Method for Optical Imaging of PET Isotopes in Biological Systems. PLoS One 5, e13300.
[5] Arroyo, A.D., Guzmán, A.E., Kachur, A.V., Saylor, S.J., Popov, A.V., and Delikatny, E.J. (2019). Development of fluorinated naphthofluoresceins for Cerenkov imaging. Journal of Fluorine Chemistry 225, 27–34.
Figure 1. pH dependent intermolecular Cerenkov imaging of NFbD.
Left: Cerenkov image of NFbD with 68Ga at various pH values (100 µCi). Maximum SBQ is observed at 600 nm due to absorption of NFbD in basic form. Maximum CRET is observed between 700 nm and 720 nm. Top Right: Cerenkov spectrum at various pH values of NFbD with 68Ga, normalized to 68Ga control. Bottom Right: Cerenkov pH titration curve for NFbD with 68Ga at Max SBQ (600 nm) and Max CRET (700 nm).  SBQ pKa = 6.698 & CRET pKa = 6.963.
Figure 2. Validation of Cerenkov pH measurements by 31P Magnetic Resonance Spectroscopy.
31P MRS was performed on athymic nude mice bearing 4175-Luc+ breast cancer tumors. 48 hours later, Cerenkov imaging was performed on the same mice, injecting NFbD-Ga (40 µCi) intratumorally. Top Left: Cerenkov imaging of athymic nude mice bearing 4175-Luc+ tumors show SBQ and CRET, at 600 nm and 700 nm, respectively. Bottom Left: Average radiance signal from in vivo Cerenkov imaging for each mouse, as a function of wavelength, normalized to 840 nm. Right: Intramolecular Cerenkov imaging pH titration curve fits pH values obtained from each mouse, after validation with 31P MRS.
Keywords: Cerenkov Imaging, Tumor Microenvironment pH, Molecular Imaging, Radiochemistry, Gallium-68