Eur Radiol DOI 10.1007/s00330-013-2916-9 INTERVENTIONAL Simultaneous dual-isotope SPECT/CT with 99mTcand 111In-labelled albumin microspheres in treatment planning for SIRT Moritz Palmowski & Andreas Goedicke & Andreas Vogg & Götz Christ & Georg Mühlenbruch & Hans J. Kaiser & Rolf W. Günther & Christiane K. Kuhl & Felix M. Mottaghy & Florian F. Behrendt Received: 24 January 2013 / Revised: 24 April 2013 / Accepted: 6 May 2013 # European Society of Radiology 2013 Abstract Objectives To investigate simultaneous dual-isotope SPECT/ CT with two differently radioisotope-labelled albuminmicrosphere fractions for treatment planning of hepatic radioembolisation. M. Palmowski : A. Goedicke : A. Vogg : G. Christ : G. Mühlenbruch : H. J. Kaiser : F. M. Mottaghy : F. F. Behrendt Department of Nuclear Medicine, RWTH-Aachen University, Aachen, Germany M. Palmowski : C. K. Kuhl Department of Diagnostic and Interventional Radiology, RWTH-Aachen University, Aachen, Germany M. Palmowski Academic Radiology Baden-Baden, Diagnostic and Interventional Radiology University Medical Center, Heidelberg, Germany G. Mühlenbruch Department of Neuroradiology, RWTH-Aachen University, Aachen, Germany R. W. Günther Department of Radiology, Charité, Humboldt-University, Berlin, Germany F. M. Mottaghy Department of Nuclear Medicine, Maastricht University Medical Center, Maastricht, The Netherlands M. Palmowski (*) Department of Nuclear Medicine, Department of Diagnostic and Interventional Radiology, RWTH-Aachen University, Pauwelsstrasse 30, 52074, Aachen, Germany e-mail: mpalmowski@ukaachen.de A. Goedicke Research Laboratories, Philips Technologie GmbH, Innovative Technologies, Aachen, Germany Methods In addition to 99mTechnetium-labelled albumin microspheres (commercially available), we performed labelling with 111Indium. Binding stability of 111Indium-labelled microspheres was tested in vitro and in vivo in mice. Simultaneous dual-isotope SPECT/CT imaging was validated in an anthropomorphic torso phantom; subsequently, dualisotope SPECT/CT was performed under in-vivo conditions in pigs (n=3) that underwent transarterial injection of 99mTechnetium- and 111Indium-labelled microspheres in the liver (right and left hepatic artery, respectively), in both kidneys and in the gluteal musculature. In total, n=18 transarterial injections were performed. Results In-vitro testing and in-vivo studies in mice documented high binding stability for both 99mTechnetium-labelled and 111Indium-labelled microsphere fractions. In phantom studies, simultaneous dual-isotope SPECT/CT enabled reliable separation of both isotopes. In pigs, the identified deposition of both isotopes could be accurately matched with intended injection targets (100 %, 18/18 intended injection sites). Furthermore, an incidental deposition of 99mTechnetium-labelled microspheres in the stomach could be correlated to the test injection into a right hepatic artery. Conclusion Simultaneous dual-isotope SPECT/CT after transarterial injection with 99mTechnetium- and 111Indiumlabelled microspheres is feasible. Thus, it may offer additional, valuable information compared to single 99mTechnetiumlabelled albumin examinations. Key Points • Simultaneous dual-isotope SPECT/CT with 111In- and 99m Tc-labelled albumin microspheres is feasible. • Differentiation of two microsphere fractions after transarterial injection is possible. • The origin of an extra-hepatic microsphere deposition can be correlated to the corresponding artery. Eur Radiol • This technique could reduce the setup time for selective internal radiation treatment. Keywords Liver . SIRT . Albumin microspheres . Simultaneous . Dual-isotope SPECT/CT Abbreviations SPECT/CT Single photon emission computed tomography/computed tomography MAA Macro aggregated albumin HSAM Human serum albumin microspheres SIRT Selective internal radiation treatment 99m 99m Tc Technetium 111 111 In Indium In this proof-of-concept study, we investigated the question whether we can allocate an extra-hepatic intestinal isotope-deposition to the right of left hepatic artery. Therefore, we applied human serum albumin microspheres labelled with 99mTc (based on commercial kits) and microspheres labelled with 111In after derivatisation of the spheres with the chelator DTPA. Qualitative discrimination of both isotopes by SPECT/CT has been verified in an anthropomorphic torso phantom. Consecutively, we tested the feasibility of simultaneous dual-isotope SPECT/CT imaging under in vivo conditions in a large animal model (pigs) that underwent angiography, protective embolisation and final test injections of 111In- and 99mTc-labelled microspheres. Materials and methods Introduction Labelling of albumin microspheres with 99m Tc and 111 In 90 Y-microsphere selective internal radiation treatment (SIRT) has emerged as a valuable tool for local treatment of primary and secondary liver tumours [1–3]. Despite its low toxicity profile in general, this therapeutic approach is associated with the risk of severe gastrointestinal ulcer disease [4]. This complication is due to extra-hepatic deposition of 90Y-loaded microspheres caused by hepatointestinal collaterals. Therefore, prior to SIRT a protective embolisation of all extra-hepatic non-target vessels and the proof thereof by applying test injections with 99mTc-labelled macro-aggregated albumin (MAA) are mandatory. If intestinal deposition of 99mTc-MAA is detected by SPECT/CT, the supporting extra-hepatic artery has to be identified and (re-)embolised in a second angiographic intervention [5]. As a consequence, the SIRT has to be suspended and postponed. According to our clinical experience, such an unfavourable situation occurs in roughly 1 out of 15 patients. To shorten the preliminary investigations preceding SIRT, it would be beneficial to delimit the origin of a possible extrahepatic non-target vessel in one session. An elegant approach would be the simultaneous use of two microsphere fractions labelled with different nuclides, which may be detected independently by a simultaneously performed SPECT/CT. Injection of fraction A and B in the right and left hepatic artery, respectively, would enable correlating an extra-hepatic MAA deposition either originating from the right or left hepatic artery. As a consequence for the clinical workflow, SIRT via the non-affected liver artery could be performed much earlier, e.g. during the second protective angiographic intervention for the contralateral side. Technically, simultaneous SPECT imaging and consecutive differentiation of two isotopes is possible [6–9]. Dualisotope imaging is enabled by different gamma photon energies of the two nuclides, e.g. 140 keV for 99mTc or 171 keV and 247 keV for 111In, thus allowing a discrimination of both isotopes. 99m Tc microspheres Human serum albumin microsphere (HSAM) labelling with 99m Tc was performed with commercially available kits (“HSA Microspheres B20”, containing 2.5 mg HSAM), which were purchased (ROTOP Pharmaka AG, Dresden, Germany). 99mTc labelling was done as described detailed in the package insert. No purification step was applied. Radiochemical labelling yield identical with radiochemical purity was >99 %. 111 In-DTPA microspheres Crude microspheres were provided by ROTOP. Derivatisation was conducted using excess agent SCN-DTPA (N-[(R)-2amino-3-(p-isothiocyanato-phenyl)propyl]-trans-(S,S)-cyclohexane-1,2-diamine-N,N,N’,N”,N”-pentaacetic acid) from Macrocyclics (Dallas, TX, USA, product B-355). The synthesis was allowed to proceed for 12 h in sodium hydrogen carbonate solution. Purification of the DTPA microspheres was performed similarly as described [10]. Labelling with 111 In was achieved using a mixture of 2.3 mg DTPA microspheres, 200 μl 3 M NH4OAc and 40 μl 111InCl3 (200 MBq in 0.04 M HCl, PerkinElmer International). The mixture was heated for 30 min at 60 °C. The spheres were purified by means of centrifugation and re-suspended in 4 ml of saline containing 1 mg/ml Poloxamer-188 (ROTOP Pharmaka AG). Radiochemical labelling yield was >95 %. After purification final radiochemical purity was >99 %. 111 In microspheres For a comparative stability assay, native non-derivatised microspheres were labelled by incubation with 111In. The Eur Radiol reaction mixture contained 400 μl 3 M NH4OAc (with 10 mg/ml Poloxamer-188), 60 μl microsphere suspension (75.8 mg/ml 15 % EtOH) and 77 μl 111InCl3 in 0.1 M HCl. Reaction time was 30 min at 70 °C. Purification was done as for 111In-DTPA microspheres. Radiochemical labelling yield was 60 %. After purification final radiochemical purity was >99 %. Analytics and quality control procedure The amount of radiochemical labelling yield (RCY) in all three labelled HSAM species was determined by separating free radio-metallic ions from microsphere-bound activity using sterile filters. Unbound activity was found in the filtrate, while labelled spheres quantitatively remained on the filter. Both the filtrate and filter were measured for radioactivity and the fraction of HSAM associated activity was that fraction on the filter. In vitro stability of microspheres 111 In microspheres and 111In-DTPA In vitro binding stability was tested by incubation of both In-DTPA microspheres and directly labelled 111In microspheres in blood plasma. During an examination period of up to 4 days, fractions of bound and free activity were determined. 111 In vivo stability of 99m Tc-HSAM and 111 In-DTPA-HSAM 111 In microspheres (without DTPA) were not applied for in vivo studies because of the insufficient binding stability (see Results section). All animal experiments were approved by the governmental review committee of animal care. To verify binding stability under in vivo conditions, 2-3 MBq 99m Tc microspheres or 111In-DTPA microspheres were injected intravenously in n=3 mice each. Each group of animals received only one isotope. One hour after injection, the animals were killed and then underwent SPECT imaging using a multi-pinhole approach for small animals as described [11]. After imaging, the lungs were resected and activity within the lungs and the remaining cadaver was measured by γ-counting. Activity within the lungs was assigned to the activity attached to microspheres. Simultaneous Dual-Isotope SPECT/CT Simultaneous dual-isotope SPECT/CT phantom studies and in vivo animal (pig) studies were performed on a Symbia TruePoint SPECT/CT T16 (Siemens Medical Solutions, Malvern, PA, USA, Inc.). A Medium Energy General Purpose (MEGP) collimator was used for scintigraphic and SPECT imaging. In SPECT mode, 64 projections (grid size: 128×128) were acquired in one (phantom) or two (pigs) bed positions on a 360° trajectory. In each position, detector events within five energy windows were collected during 20 s as separate data sets (window centre/width in keV/%): (a) 140/15, (b) 172/15, (c) 247/15, (a1) 117/20 and (c1) 212/15. Individual CT images (25 mAs/110 kV; grid size: 512×512; pixel spacing: 0.977×0.977 mm2; 0.6 mm slices) were acquired in each experiment as anatomical/object reference and for attenuation/scatter correction. Reconstruction of spatial activity distribution (voxel cube edge length: 4.8 mm) was carried out based on the Ordered Subset Expectation Maximisation (OSEM) algorithm using Flash3D™ on a Siemens MI workplace. CT-based attenuation correction was applied during the iterative reconstruction for both involved isotopes (99mTc: 8 iterations/16 subsets, 111In: 4 iterations/8 subsets). Additionally, scatter correction was enabled for reconstruction of the Technetium projections based on energy window data a1/c1, as described in [6, 7]. Furthermore, Butterworth filtering (n: 5; fc: 0.4) was applied to reduce pixel noise artefacts. In scintigraphy mode, planar anterior and posterior views were acquired (axial speed: 8 cm/min; matrix size: 256× 1024) using the same energy window settings as for SPECT. Subsequently, a geometric mean of these views was separately calculated for all matching data sets to account for attenuation effects. The Technetium images were corrected for scatter contamination by manually subtracting the weighted sum of the geometric mean images for scatter windows a1 and c1. Thereby, the applied weighting factor (k) was heuristically found. Starting from zero, k was increased in steps of 0.1 until visual over-correction artefacts occurred (e.g. SPECT untypical sharp structural edges and physically not reasonable areas containing negative background counts). Finally, pixel noise in the corrected results was diminished using 3×3 average (weighted ‘boxcar’) filtering. In vitro experiments in an anthropomorphic torso phantom For in vitro studies, a static Anthropomorphic Torso Phantom™ (Data Spectrum Corp., Hillsborough, USA) was used comprising a body-shaped cylinder with lung, liver and spine inserts. Two balloons were placed inside the liver insert and filled from outside with water. The balloon in the right and left liver contained 780 ml and 390 ml, respectively. The lung compartments were filled with Styrofoam® beads and water to simulate lung tissue density. Furthermore, n=3 extra-hepatic hot spots (simulating “extra-hepatic microsphere accumulation”) were introduced using 5-ml syringes radially fixed to the phantom’s spine insert. The complete setup is illustrated in Fig. 1. For the simultaneous dual isotope scintigraphy, the phantom was filled with clinically realistic activity concentrations as in humans: Eur Radiol Fig. 1 a The construction of the anthropomorphic phantom consisting of a lung, liver and spine insert with additionally introduced syringes radially fixed to the phantom’s spine. b The distribution of the two isotopes technetium and indium in the phantom setup. c-f Images of the simultaneous dual-isotope SPECT/CT in the 99mTc-window (c), 111 In-window (d) or as fused images (e, f) The liver contained a total activity of 90 MBq: right lobe: 58.1 MBq 99mTc; left lobe: 31.8 MBq 111In. The lungs were filled with a total activity of 3.6 MBq: right lung: 1.2 MBq 99mTc and 0.7 MBq 111In; left lung: 0.5 MBq 99mTc and 1.2 MBq 111In. The lung shunt was calculated to be 3.8 %. The simulated “extra-hepatic” depositions were filled as follows: beside the right liver: 0.5 MBq 99mTc; beside the left liver: 0.5 MBq 111In; cranially of the liver: 0.25 MBq 99mTc and 0.25 MBq 111In. Transarterial liver embolisation with 99mTc microspheres and 111In-DTPA microspheres and in pigs All animal experiments were approved by the governmental review committee of animal care. For anaesthesia, n=3 domestic female pigs (weight approximately 60 kg) received premedication with azaperone and 0.1 ml ketamine per kilogram body weight followed by intubation and mechanical ventilation with an oxygen–air mixture containing 1.0 vol% isoflurane. Maintenance of anaesthesia was achieved by infusion of propofol and fentanyl. Constant saline infusion was administered in all animals to prevent dehydration. For the angiographic procedure, animals were placed on the patient table of the SPECT/CT. Before intervention, diagnostic contrast-enhanced CT was performed to identify the celiac trunk and the renal arteries. Consecutively, arterial access was obtained by a 5-F sheath placed in the right femoral artery. A 5-F cobra catheter (Cook Medical, Bloomington, IN, USA) was inserted for catheterisation of the celiac trunk. Image guidance was performed using a C-arm X-ray imaging unit (Ziehm Vision, Ziehmimaging, Nurenberg, Germany). Super-selective catheterisation of the gastroduodenal artery and of the hepatic arteries was performed using a micro-catheter (2.7 F Progreat, Terumo, Eschborn Germany). The gastroduodenal artery was embolised via a fibred coil occlusion system (Interlock, Boston Scientific, Natick, MA, USA). Consecutively, suitable hepatic branches were identified that reached either the left or the right liver. We have to mention that the arterial situation in pigs is very different from humans. In contrast to the mostly clear differentiation into the left and right hepatic Eur Radiol artery in humans, in pigs there are many small segmental arteries together with branches leading to the stomach. Into a right liver segment, 100 MBq 99mTc microspheres was administered, and into a left liver segment 100 MBq 111InDTPA microspheres. Additionally to the liver, we injected control microsphere fractions into the kidneys to get another clearly separated anatomic deposition of both microsphere fractions. The right kidney received 30 MBq 99mTc microspheres and the left kidney 30 MBq 111In-DTPA microspheres. Finally, a left gluteal artery was catheterised super-selectively. Here, we administered a combination of 15 MBq 99mTc microspheres and 15 MBq 111In-DTPA microspheres within the same artery, thus simulating a simultaneous deposition of both isotopes. After angiography, dual-isotope SPECT/CT imaging with acquisition in different energy windows was performed as described above. Following the image acquisition, the animals were killed. Finally, the data were analysed regarding the achieved distribution and especially the isotopespecific differentiation of both microsphere fractions. Fig. 2 Stability assay of both directly labelled 111In microspheres and 111 In-DTPA microspheres in blood plasma at 37 °C: Fraction of human serum albumin microspheres bound activity as a function of incubation time microspheres, 95.54±0.04 % of the injected activity was found to be in the lungs. In vitro experiments in an anthropomorphic torso phantom Results Labelling of human serum albumin microspheres with 99mTc and 111In The excellent stability of the commercial compound 99mTc microspheres has been verified. In contrast to 111In microspheres the derivatised 111InDTPA microspheres provided high in vitro binding stability. After 2 h, about 1 % of 111In-DTPA microspheres but 25 % of 111In microspheres had lost 111In to plasma proteins. Finally, after 3 days in human blood plasma, 95 % of the activity was still bound to the DTPA microspheres, while directly labelled native 111In microspheres had lost almost 60 % of activity, thus revealing a significantly lower binding stability. Corresponding time-stability curves for the plasma assay are shown in Fig. 2. These stability assays strongly recommended the preferred use of chelator-derivatised microspheres rather than native microspheres for labelling with 111 In. In vivo binding stability was investigated by i.v. injection of either 99mTc microspheres or 111In-DTPA microspheres in n=3 mice each. Because of the discouraging in vitro result, directly labelled 111In microspheres were not administered. One hour after i.v. injection, μSPECT demonstrated accumulation of the activity within the lungs and excluded other focal accumulation of activity (e.g. within the urinary bladder) (Fig. 3). Ex vivo γ-counting of the explanted lungs showed that for 99mTc microspheres, 94.93±0.01 % of the injected dose was located within the lungs. For 111In-DTPA For planar scintigraphy, manual subtractive scatter correction using a weighting factor of k=0.6 was visually confirmed to deliver superior results. Simultaneous dual-isotope SPECT/CT imaging of the anthropomorphic torso phantom demonstrated that a clear differentiation of 99mTc and 111In is feasible. The two isotopes were detectable precisely in the liver inserts and the introduced syringes. In the 99mTc-window, tracer activity could be confirmed only in the right liver lobe, in the 99mTc-filled syringe, the syringe filled with the mixture of the two isotopes and slightly in both lungs (Fig. 1c). The 99mTc related scatter contamination in the 111 In -window was found to be negligible. In the 111Inwindow, activity was observable only in the left liver lobe, the 111In-filled syringe, the syringe filled with the two isotopes and slightly in both lungs (Fig. 1d). Results showed that the right and left liver lobe could be precisely differentiated from each other. Fusion images represented activity findings in all sites, accordingly (Fig. 1e, f). Transarterial injection of 99mTc microspheres and 111In-DTPA microspheres in pigs In n=3 pigs, selective transarterial injection of both HSAM microsphere fractions was performed into five areas in the each session. 99mTc microspheres were injected into a right hepatic branch and the right renal artery. 111In-DTPA microspheres were injected into a left hepatic branch and the left renal artery. Both isotopes were injected simultaneously into the left gluteal musculature. Eur Radiol Fig. 3 CT and multi-pinhole SPECT images of mice 1 h after i.v. injection of either 99mTc microspheres (a) or 111In-DTPA microspheres (b). Both isotopes are located within the lungs; no relevant extrapulmonary accumulation (e.g. within the urinary bladder) can be observed. Thus, a stable binding of both isotopes to human serum albumin microspheres under in vivo conditions can be concluded In vivo planar imaging confirmed the results from the anthropomorphic torso phantom regarding a differentiation of both MAA microsphere fractions. In the planar images of the 99mTc window, activity was found only in the right liver lobe, the right kidney and the left gluteal muscle. Planar images of the 111In window showed only activity in the left liver, the left kidney and the left gluteal muscle. In the fused planar images, activity was found in both liver lobes, both kidneys and the left gluteal muscle (Fig. 4). In order to facilitate dual-isotope (HSAM) scintigraphy and data reviewing in the clinical routine, we performed a colour coding of both isotopes on the cross-sectional SPECT/CT images: red was chosen for 99mTc and blue for 111In. Using this colour coding, fused dual-isotope SPECT/CT images Fig. 4 Planar scintigraphy images of a pig after transarterial embolisation with 111 In-DTPA microspheres and 99m Tc microspheres. In the 99m Tc-window (a), activity is found only in the right liver (*), the right kidney (**) and the left gluteal muscle (***). Images of the 111In window (b) show activity predominantly within the left liver (*), the left kidney (**) and the left gluteal muscle (***). In the fused images (c), activity is found in both liver lobes, both kidneys and the left gluteal muscle. No extrahepatic shunting to the gastrointestinal tract is observed showed a clear differentiation of both MAA microsphere fractions within the liver and kidneys. The deposition of both MAA microsphere fractions (red and blue) within the gluteal muscle was highlighted using the combination colour (magenta). The use of the combination colour facilitates the depiction of depositions of both MAA microsphere fractions on the cross sectional images (Fig. 5). In one animal, we observed an extra-hepatic MAA deposition of the 99mTc microsphere fraction within the stomach. Due to the dual-isotope approach, we were able to correlate the gastric MAA deposition to the hepatic branch where the test injection was performed (Fig. 6). In the other two animals, no extra-hepatic deposition in the gastrointestinal tract was observed. Eur Radiol Fig. 5 Simultaneous dual-isotope SPECT/CT images of a pig after transarterial embolisation with 111In-DTPA microspheres (colour coded in blue) and 99mTc microspheres (colour coded in red). a Coronal and axial view of the liver after embolisation of a right and left liver segment. Both isotopes can be clearly distinguished. b In addition to the liver, the kidneys were embolised with 111In-DTPA microspheres (left kidney) and 99mTc microspheres (right kidney) as a control. c To simulate a tumour with a supply from the left and right hepatic artery, both isotopes were injected within a left gluteal artery leading to a clearly visual combination colour enhancement in the left gluteal muscles Discussion with precaution and patients were selected appropriately [4]. On the other hand, also severe side effects and complications have been reported to be caused by extra-hepatic accumulation of 90Y-loaded microspheres. These complications include severe gastrointestinal ulceration or bleeding, pancreatitis, cholecystitis and radiation pneumonitis [13–16]. The cause for these complications is that radionuclideloaded microspheres are inadvertently injected into extrahepatic non-target vessels and hepatovenous collaterals during the therapy. Thus, it is an essential precondition to identify non-target vessels on an angiogram and, consecutively, to perform protective coil embolisation. However, in about 15 % of patients these hepatointestinal arteries may not be detectable during the initial interventional procedure because they are too small [17]. Therefore, another crucial integral part after the protective coil embolisation is a test injection of 99mTc-labelled macroaggregated albumin (MAA) followed by a planar scintigraphy or SPECT to exclude extra-hepatic deposition [17, 18]. Results of this study demonstrate that the MAA test injection as an integrated part of the pre-procedural evaluation of SIRT can be performed with 99mTc- and 111In-labelled microspheres. In contrast to previous investigations directly 111 In-labelled native human serum albumin microspheres revealed only poor blood plasma stability [12]. DTPAderivatised microspheres used in this study, however, were found to bind 111In with sufficient plasma stability. Differentiation of both isotopes by SPECT/CT was proven in a phantom setup and findings were confirmed during in vivo conditions in a pig model using clinical realistic angiographic conditions including protective coil embolisation of the gastroduodenal artery followed by dual-isotope microsphere injection. SIRT has become an established treatment method for patients with primary and metastatic liver cancer. In general, an advantage of SIRT is that the incidence of complications is low provided that preceding examinations are carried out Eur Radiol Fig. 6 Simultaneous dual-isotope axial SPECT/CT images of a pig. a Contrast-enhanced CT images of the upper abdomen at the level of the liver, stomach and duodenum. b A traditional MAA SPECT/CT image on which both isotopes are highlighted in the same colour. Here, extrahepatic MAA deposition within the intestine is detected. Dual-isotope SPECT/CT with isotope specific colour mapping (red: 99mTc microspheres, blue: 111In microspheres) enables clear differentiation and provides evidence that the extrahepatic HSAM deposition belongs to the fraction of 99mTc microspheres that were injected in a left-sided hepatic branch In some cases, an extra-hepatic focus is found in MAA SPECT/CT with the consequence of a mandatory second interventional angiography to localise and embolise the responsible artery [17, 18]. In these cases, the dual-isotope MAA injection would be very helpful as the origin of the shunt artery could be determined safely because of the clear differentiation of the two isotopes in the same session, as demonstrated in our study. This procedure will have some relevant benefits for the clinical work flow and cost. For instance, it would be feasible to begin with a SIRT of the other liver lobe from which it is known that there is no hepatointestinal vascular connection. Thus, the dualisotope MAA would prevent a further delay of the therapy. This seems to be important especially given the fact that the protective embolisation of the extra-hepatic non-target arteries should be performed close to the intended date of SIRT, since revascularisation may occur quickly [17]. It also might be feasible to directly combine SIRT of one liver lobe with a further MAA test injection of the contralateral liver lobe (from which the intestinal MAA deposition originated). The impact of the resulting Bremsstrahlung scatter in the 99mTc energy window originating from the decay of the therapeutic agent can be corrected using basically the same approach as for the isotope cross-contamination correction applied in this study [6, 7]. Due to the reportedly higher sensitivity of hybrid crosssectional imaging with SPECT/CT compared to planar scintigraphy alone [18], we assigned each isotope an individual colour (red for 99mTc and blue for 111In) during image post-processing and performed image fusion with the diagnostic contrastenhanced CT images. Thereby, we could identify both differently labelled MAA microsphere fractions within the livers and kidneys. The extra-hepatic (incidental) gastric MAA deposition, which was observed in one animal, could be clearly depicted clearly on the dual isotope SPECT/CT images. In the current study, we performed a qualitative differentiation of the two isotopes to allocate an extrahepatic MAA deposition to the right or left hepatic artery. The second task of the MAA test injection is to quantitatively investigate the hepatopulmonary shunt volume. In principle, quantitative determination of the hepatopulmonary shunt volume could be affected to a certain degree by a scatter contamination (“spill-over radiation”) of 111In on the 99mTc window. A correction of the scatter contamination seems feasible by utilisation of multiple different energy window data [6, 7]. A systematic investigation of this question in an anthropomorphic torso phantom and in animals is currently the subject of an ongoing study. In conclusion, dual-isotope SPECT/CT with 111In- and 99m Tc-labelled human serum albumin microspheres is feasible in phantom studies and in pigs under realistic in vivo angiographic conditions. Thus, it offers a potential alternative for single 99mTc-labelled MAA prior to radioembolisation of the liver. Consequently, the next step after the successful animal experiments is the translation to the patient. Due to the fact that all compounds (99mTc-, 111In- and DTPA-coupled human Eur Radiol serum albumin microspheres) are established, the process of a clinical translation should be feasible. Acknowledgements M. Palmowski and A. Goedicke contributed equally to this work. 8. 9. 10. References 11. 1. Salem R, Lewandowski RJ, Mulcahy MF et al (2010) Radioembolization for hepatocellular carcinoma using yttrium-90 microspheres: a comprehensive report of long-term outcomes. Gastroenterology 138:52–64 2. Hilgard P, Hamami M, Fouly AE et al (2010) Radioembolization with yttrium-90 glass microspheres in hepatocellular carcinoma: European experience on safety and long-term survival. Hepatology 52:1741–1749 3. Prompers L, Bucerius J, Brans B, Temur Y, Berger L, Mottaghy FM (2011) Selective internal radiation therapy (SIRT) in primary or secondary liver cancer. Methods 55:253–257 4. Murthy R, Nunez R, Szklaruk J et al (2005) Yttrium-90 microsphere therapy for hepatic malignancy: devices, indications, technical considerations, and potential complications. Radiographics 25(Suppl 1):S41–S55 5. Lenoir L, Edeline J, Rolland Y et al (2012) Usefulness and pitfalls of MAA SPECT/CT in identifying digestive extra-hepatic uptake when planning liver radioembolization. Eur J Nucl Med Mol Imaging 39:872–880 6. Jaszczak RJ, Greer KL, Floyd CE Jr, Harris CC, Coleman RE (1984) Improved SPECT quantification using compensation for scattered photons. J Nucl Med 25:893–900 7. Ogawa K, Harata Y, Ichihara T, Kubo A, Hashimoto S (1991) A practical method for position-dependent Compton-scatter 12. 13. 14. 15. 16. 17. 18. correction in single photon emission CT. IEEE Trans Med Imaging 10:408–412 Zhou R, Thomas DH, Qiao H et al (2005) In vivo detection of stem cells grafted in infarcted rat myocardium. J Nucl Med 46:816–822 Perault C, Schvartz C, Wampach H, Liehn JC, Delisle MJ (1997) Thoracic and abdominal SPECT-CT image fusion without external markers in endocrine carcinomas. The group of thyroid tumoral pathology of Champagne-Ardenne. J Nucl Med 38:1234–1242 Schiller E, Bergmann R, Pietzsch J et al (2008) Yttrium-86labelled human serum albumin microspheres: relation of surface structure with in vivo stability. Nucl Med Biol 35:227–232 Nuyts J, Vunckx K, Defrise M, Vanhove C (2009) Small animal imaging with multi-pinhole SPECT. Methods 48:83–91 Watanabe N, Shirakami Y, Tomiyoshi K et al (1997) Direct labeling of macroaggregated albumin with indium-111-chloride using acetate buffer. J Nucl Med 38:1590–1592 Murthy R, Brown DB, Salem R et al (2007) Gastrointestinal complications associated with hepatic arterial yttrium-90 microsphere therapy. J Vasc Interv Radiol 18:553–561, quiz 562 Carretero C, Munoz-Navas M, Betes M et al (2007) Gastroduodenal injury after radioembolization of hepatic tumors. Am J Gastroenterol 102:1216–1220 Salem R, Parikh P, Atassi B et al (2008) Incidence of radiation pneumonitis after hepatic intra-arterial radiotherapy with yttrium90 microspheres assuming uniform lung distribution. Am J Clin Oncol 31:431–438 Atassi B, Bangash AK, Lewandowski RJ et al (2008) Biliary sequelae following radioembolization with yttrium-90 microspheres. J Vasc Interv Radiol 19:691–697 Ahmadzadehfar H, Sabet A, Biermann K et al (2010) The significance of 99mTc-MAA SPECT/CT liver perfusion imaging in treatment planning for 90Y-microsphere selective internal radiation treatment. J Nucl Med 51:1206–1212 Hamami ME, Poeppel TD, Muller S et al (2009) SPECT/CT with 99mTc-MAA in radioembolization with 90Y microspheres in patients with hepatocellular cancer. J Nucl Med 50:688–692