Original article
Annals of Oncology 14: 1274–1277, 2003
DOI: 10.1093/annonc/mdg344
Imaging gastrointestinal tumours using vascular endothelial
growth factor-165 (VEGF165) receptor scintigraphy
S. Li1*, M. Peck-Radosavljevic2, O. Kienast1, J. Preitfellner1, G. Hamilton3, A. Kurtaran1, C. Pirich1,
P. Angelberger4 & R. Dudczak1
1
Department of Nuclear Medicine, 2Department of Internal Medicine IV, Division of Gastroenterology and Hepatology, and 3Department of Surgery,
University of Vienna, Vienna; 4Austrian Research Center, Seibersdorf, Austria
Received 1 April 2003; accepted 23 April 2003
Background: Recent studies have shown that vascular endothelial growth factor (VEGF) receptor is overexpressed in vascular endothelial cells of various human tumours as well as in human tumour cells. The aim of
this study was to evaluate the usefulness of scanning with VEGF165 labeled with 123I for tumor localisation in
patients with gastrointestinal tumours.
Patients and methods: Human recombinant VEGF165 was radiolabelled with 123I by electrophilic radioiodination using the chloramine T method. [123I]VEGF165 was administered intravenously [mean dose 184 ± 18 MBq
(≤130 pmol; ≤5 µg) per patient] to 18 patients with gastrointestinal tumours. Dynamic acquisition was initiated
immediately after administration and carried out until 30 min post-injection. Whole body images were done in
anterior and posterior views at various time points. All patients underwent single-photon emission tomography
imaging 1.5 h post-injection. Scanning with [123I]VEGF165 was compared with computed tomography and
magnetic resonance imaging.
Results: Intravenous injection of [123I]VEGF165 did not cause any side-effects. Binding of [123I]VEGF165
to primary tumours and metastases was visible shortly after injection. In patients with pancreatic adenocarcinomas, primary tumours were visualised in seven of nine, lymph node metastases in three of four, liver
metastases in three of six and lung metastases in one of three. Cholangiocarcinomas were visualised by imaging
in one of two patients. Hepatocellular carcinomas were visible by imaging in two of four patients. [123I]VEGF165
scans were weakly positive in one patient with abdominal schwannoma and in one patient with peritoneal
carcinosis.
Conclusions: These results indicate that scanning with [123I]VEGF165 can visualise gastrointestinal tumours
and metastases expressing receptors for VEGF165. [123I]VEGF165 receptor scintigraphy may be useful for visualisation of tumour angiogenesis.
Key words: angiogenesis, gastrointestinal, receptors, scintigraphy, tumours, VEGF
Introduction
Vascular endothelial growth factor-165 (VEGF165) is a 38 kDa
dimeric glycoprotein, which has been suggested to play an important role in the process of tumour angiogenesis [1]. VEGF165
may be secreted by tumour cells [2]. VEGF acts directly on cells
by binding to specific cell surface receptors [1]. Two receptors
for VEGF165, namely, VEGF receptor (VEGFR)-1 (Flt-1) and
VEGFR-2 (KDR) have been characterised [3, 4]. VEGFR-1 and
VEGFR-2 are members of a receptor tyrosine kinase family.
VEGF165 binds with high affinity to VEGFR-1 and VEGFR-2
[3, 4].
*Correspondence to: Dr S. Li, Department of Nuclear Medicine,
University of Vienna, Waehringer Guertel 18–20, A-1090 Vienna, Austria.
Tel: +43-1-40400-4538; Fax: +43-1-2605866;
E-mail: [email protected]
© 2003 European Society for Medical Oncology
Recent studies have demonstrated that various tumour cells,
including gastrointestinal tumour cells and endothelial cells of
proliferating tumours, overexpress receptors for VEGF165 [5–7].
In our previous studies we have shown that significantly higher
amounts of VEGF receptors were found in various human tumour
cells and tumour tissues compared with adjacent normal tissues or
peripheral blood cells [8]. These observations led us to develop
an [123I]VEGF165 receptor scintigraphy to explore a possible role
of VEGF receptor scintigraphy in the staging and follow-up of
patients with solid tumours. Furthermore, knowledge about the
VEGF receptor status in vivo may bring about fascinating aspects
for tumour-specific antiangiogenic therapies and strategies [9].
For this purpose we used VEGF receptor scintigraphy in patients
with histologically confirmed gastrointestinal tumours to determine the extent of [123I]VEGF165 accumulation in the tumours
and/or its metastases. The results of [123I]VEGF165 receptor scinti-
1275
Table 1. Comparison of various imaging procedures for localisation of primary and metastatic lesions
No. of cases
Pancreatic adenocarcinomas
9
No. of lesions
22
Detection rate of lesions (%)
VEGFS
CT/MRI
14 (64)
21 (95)
Hepatocellular carcinomas
4
8
4 (50)
8 (100)
Cholangiocarcinomas
2
4
3 (75)
4 (100)
Abdominal schwannoma
1
3
1 (33)
3 (100)
Peritoneal carcinosis
1
2
1 (50)
1 (50)
Hepatic focal nodular hyperplasia
Total
1
1
18
40
Cold spot
23 (58)
1 (100)
38 (95)
VEGFS, [123I]VEGF165 receptor scintigraphy; CT, computed tomography; MRI, magnetic resonance imaging.
graphy were also compared with those of computed tomography
(CT) scanning and magnetic resonance imaging (MRI).
Patients and methods
The study protocol was approved by the Ethical Committee of the Medical
Faculty of the University of Vienna. The 18 patients enrolled in this study gave
written informed consent to participate in the study.
Nine patients had a pancreatic adenocarcinoma, four had a hepatocellular
carcinoma, two had a cholangiocarcinoma, one had a peritoneal carcinosis of a
gastric adenocarcinoma, one had an abdominal schwannoma and one had a
benign hepatic focal nodular hyperplasia. In all patients, diagnosis and stage of
disease were established according to WHO criteria. The location and size of
primary tumours and/or spread of metastases were investigated by conventional CT scan, MRI, endoscopy or surgery.
At time of scintigraphic evaluation, only one patient with pancreatic adenocarcinoma and the patient with abdominal schwannoma had undergone
chemotherapy 3 weeks before scintigraphic evaluation. The patient with
peritoneal carcinosis of gastric adenocarcinoma had undergone gastroectomy
1 year previously.
Recombinant human VEGF165 (rhVEGF165) (PromoCell GmbH, Heidelberg,
Germany) was labelled with 123I by electrophilic radioiodination using chloramine T, essentially as described previously [8]. Briefly, 10 µg (0.26 nmol) of
rhVEGF165 were labelled with ∼30 mCi [123I]Na (Research Center, Karlsruhe,
Germany) and chloramine T. After 3 min the reactions were stopped by addition
of 4 µl sodium metabisulphite (Na2S2O5, 50 nmol). The reaction mixture was
diluted with phosphate-buffered saline containing 0.1% human serum albumin
and applied onto a size-exclusion chromatography column (Sephadex G-25 M).
The first 123I peak eluting from the column was collected, analysed and filtered
using a sterile membrane (Millex GV 0.2 µm).
The study protocol included planar, single-photon emission tomography
(SPET) and whole-body gamma camera imaging as well as blood and urine
collections over 24 h.
[123I]VEGF165 was administered slowly as a single intravenous bolus injection over 3 min in a dose of 184 ± 18 MBq [≤130 pmol (≤5 µg) VEGF165 per
patient]. In order to determine the haemodynamic effects of VEGF, blood
pressure and heart rate were monitored during tracer application and scintigraphy. The patients received 400 mg sodium perchlorate three times daily
over 3 days for thyroid blockage.
Standard techniques were applied for recording and visualisation. Wholebody acquisitions were performed simultanously in anterior and posterior
views with a double-headed and a large feld of view camera (Millenium VG with
Hawkeye; GE Medical Systems, Milwaukee, WI, USA) employing mediumenergy high-resolution collimators (10 cm/min; matrix 256 × 1024 pixels).
Serial whole body images were obtained 1, 2, 18 and 24 h post-injection. A 30
min dynamic image was recorded starting at the time of injection (matrix 128
× 128 pixels). To verify that the activity localisation corresponded to the
tumour lesions documented by CT scans, SPET studies in combination with
CT scans were obtained 1.5 h post-injection (60 projections over a 360 rotation, 40 s per step and a 64 × 64 pixel matrix). Planar images were acquired 40
min post-injection including anterior and posterior views of abdomen and thorax (matrix 256 × 256 pixels; 800 kcts preset). Additional lateral or oblique
views of above regions or other regions were obtained when necessary.
Scintigrams were viewed separately and independently by two experienced
observers unaware of tumour findings. [123I]VEGF165 receptor scintigraphy
results were compared with the conventional CT and/or MRI. Quantitative
analysis of [123I]VEGF165 receptor scintigraphic results with image quality was
done in clinically or radiologically known involved sites. Lesion to background uptake ratios were determined on 1.5 h SPET images using regions of
interest over the lesion and over the ipsilateral or contralateral side of normal
uptake. The mean counts over the lesions and background regions were calculated and ratios were obtained.
For statistical analyses, the χ2 test was used to compare sites detection by
two techniques. Statistical analysis was using Student’s t-test and analysis of
variance at a confidence level of 95%.
Results
[123I]VEGF165 was routinely analysed by paper-electrophoresis on
Whatman Nr.3 MM in 0.1 M barbital buffer (pH 8.6) using a field
of 300 V for 10 min, and by trichloroacetic acid precipitation.
[123I]VEGF165 was obtained in a radiochemical yield of ∼27%, in
a specific activity of ∼1 mCi/µg and in a radiochemical purity of
>96%. The biological activity of unlabelled and labelled VEGF165
were identical as assessed by [3H]thymidine uptake assay on
human umbilical vein endothelial cells.
After intravenous injection of [123I]VEGF165, patients showed
no clinical adverse reaction, and no side-effects were noted.
The results of a lesion-by-lesion comparison of [123I]VEGF165
receptor scintigraphy with CT and MRI are given in Table 1.
Primary pancreatic adenocarcinomas (Figure 1) and liver metastases were visualised by [123I]VEGF165 receptor scanning shortly
(30 min) after injection of [123I]VEGF165, and were still visible at
2–3 h after application. For liver and lung metastases an enhanced
tracer uptake was observed in some of the lesions, whereas
others showed a heterogenous accumulation of the tracer compared with normal liver or lung tissue. The overall sensitivity of
1276
strated the heterogenous accumulation of tracer in the abdomen.
The majority of false-negative [123I]VEGF165 receptor scanning
results were obtained by tumours and metastases with maximum
median diameter of <2 cm. In comparison, almost all primary
tumours and metastases were visualised by CT and MRI with the
exception of peritoneal carcinosis and lymph node metastases
(Table 1).
Discussion
Figure 1. The anterior planar image obtained from a patient with a primary
pancreatic adenocarcinoma 2 h post-injection indicating an increased tracer
accumulation in the tumour region.
[123I]VEGF165 receptor scintigraphy for detecting pancreatic
tumors and their metastases was 64%. [123I]VEGF165 receptor
scintigraphy visualised primary pancreatic adenocarcinomas in
seven of nine patients (sensitivity 78%). Lymph node metastases
were seen in three of four patients (75%), and liver metastases in
three of six patients (50%). Lung metastases were detected only in
one of three patients (33%). The uptake ratios between pancreatic
primary tumour and background ranged from 1.5 to 2.6. A falsenegative [123I]VEGF165 receptor scanning for a primary pancreatic
cancer lesion was obtained in one patient, who had received the
last cycle of chemotherapy (docetaxel and gemcitabine) 3 weeks
previously. In one patient with a large primary pancreatic cancer
in the pancreatic head, CT scan showed the median diameter to be
6 cm, while [123I]VEGF165 receptor scintigraphy indicated the
presence of a ring-shaped tracer accumulation around the lesion;
histological examination after surgery revealed the presence of
a central necrosis in the primary tumor. For hepatocellular
carcinoma (Figure 2) and cholangiocarcinoma, an enhanced tracer
uptake was observed in some of the lesions, whereas others showed
a heterogenous accumulation of the tracer compared with normal
liver tissue. Two of four patients with hepatocellular carcinoma
and one of two patients with cholangiocarcinoma could be visualised by [123I]VEGF165 receptor scintigraphy. Interestingly, one
patient with a large liver lesion was examined by [123I]VEGF165
receptor scintigraphy because of suspicion of having hepatocellular
carcinoma. The [123I]VEGF165 receptor scintigraphy showed a large
cold spot in the liver lesion. The histological examination confirmed a diagnosis of benign hepatic focal nodular hyperplasia.
In one patient with abdominal schwannoma, the [123I]VEGF165
receptor scintigraphy showed only the liver metastases. In the
patient with peritoneal carcinosis who had undergone gastroectomy 1 year previously, the [123I]VEGF165 receptor scan demon-
In this study, our results demonstrate that there are no adverse
side-effects after intravenous application of [123I]VEGF165. An
overall [123I]VEGF165 receptor scan sensitivity of 58% was obtained
for gastrointestinal tumours, varying from 78% for primary pancreatic cancers to 33% for lung metastases. In contrast to CT/MRI,
the detection rate of [123I]VEGF165 receptor scintigraphy for
gastrointestinal tumour was low in our series. However, the most
negative [123I]VEGF165 receptor scannings were found in the primary tumours and metastases with maximum median diameter
of <2 cm, as well as in the patient who had recently received
chemotherapy. More interestingly, [123I]VEGF165 receptor scanning showed a ring-shaped tracer accumutation around the necrotic
lesions. Furthermore, malignant liver lesion can be visualised by
[123I]VEGF165 receptor scan; however, benign liver hyperplasia
was shown as a cold spot in the [123I]VEGF165 receptor scan.
Therefore, [123I]VEGF165 receptor scintigraphy might indicate
some biological characteristics (activity or angiogenetic state) of
the disease and might be useful for the differential diagnosis of
liver lesions.
The precondition for the accumulation of VEGF165 in the scintigram is the presence of receptors for VEGF165 on the tumour cells
and tumour vascular endothelial cells. Using a direct receptor
binding assay, a class of saturable VEGF165 receptors has been
identified on various gastrointestinal tumour cell lines and tumour
tissues [8]. The cases in which the [123I]VEGF165 receptor scintigraphy was negative may be explained by the low number or
absence of VEGF165 receptors or by the blockade of VEGF165
receptors by locally produced ligands or small size of lesions.
Tumour cell necrosis or changes in the blood supply of the cancers
might also have an effect on cell binding properties. With regard
to lung metastases, only about one-third of metastases were found
on [123I]VEGF165 receptor scanning, although the largest lung
lesions imaged in our studies were only in the range of 2.5 cm in
diameter. This might be due to relatively high background activities, indicating a high degree of physiological VEGF165 uptake in
the lungs. We observed that there was no substantial VEGF165
uptake by normal gastrointestinal tissue. This characteristic of
VEGF165 biodistribution may be an apparent advantage for the
detection of gastrointestinal tumours.
In conclusion, our preliminary results demonstrate the safety of
imaging with [123I]VEGF165, and that a variety of gastrointestinal
tumours and metastases may be visualised by VEGF165 receptor
scintigraphy. Although CT/MRI is superior to [123I]VEGF165
receptor scintigraphy for the visualisation of the gastrointestinal
tumours and metastases, [123I]VEGF165 scintigraphy might be a
useful tool for visualisation of viable tumour angiogenesis.
1277
Figure 2. Imaging of a hepatocellular carcinoma by means of [123I]VEGF165 receptor scintigraphy. The left panel shows a transverse slice of computed
tomography scanning and the middle panel the corresponding transverse slice of the single-photon emission tomography study. The right panel demonstrates
the image fusion performed 1.5 h post-injection, showing tracer accumulation in the liver lesion.
Acknowledgements
We would like to thank Karoline Wiesner and Christian Poetzi for
their excellent technical assistance. This study was supported in
part by the ‘Jubiläumsfonds’ of the Austrian National Bank
(project no. 8320) and Pfeiffer Scholarship of Austrian Society of
Nuclear Medicine.
References
1. Senger DR, Van-de-Water L, Brown LF et al. Vascular permeability factor
(VPF, VEGF) in tumor biology. Cancer Metastasis Rev 1993; 12: 303–
324.
2. Ferrara N, Houck KA, Jakeman LB, Leung DW. Molecular and biological
properties of the vascular endothelial growth factor family of proteins.
Endocr Rev 1992; 13: 18–32.
3. De Vries C, Escobedo JA, Ueno H et al. The fms-like tyrosine kinase, a
receptor for vascular endothelial growth factor. Science 1992; 255: 989–
991.
4. Terman BI, Dougher-Vermazen M, Carrison ME et al. Identification of
the KDR tyrosine kinase as a receptor for vascular endothelial cell growth
factor. Biochem Biophys Res Commun 1992; 187: 1579–1586.
5. Harada Y, Ogata Y, Shirouzu K. Expression of vascular endothelial growth
factor and its receptor KDR (kinase domain-containing receptor)/Flk-1
(fetal liver kinase-1) as prognostic factors in human colorectal cancer. Int
J Clin Oncol 2001; 6: 221–228.
6. Boocock CA, Charnock-Jones DS, Sharkey AM et al. Expression of
vascular growth factor and its receptors flt and KDR in ovarian carcinoma.
J Natl Cancer Inst 1995; 87: 506–516.
7. Plate KH, Breier G, Weich HA, Risau W. Vascular endothelial growth
factor is a potential tumor angiogenesis factor in human gliomas in vivo.
Nature 1992; 359: 845–848.
8. Li S, Peck-Radosavljevic M, Koller E et al. Characterization of (123)Ivascular endothelial growth factor-binding sites expressed on human
tumour cells: possible implication for tumour scintigraphy. Int J Cancer
2001; 91: 789–796.
9. Fidler IJ, Ellis LM. The implications of angiogenesis for the biology and
therapy of cancer metastasis. Cell 1994; 79: 185–188.