Clinical Hemorheology and Microcirculation 61 (2015) 143–150 DOI 10.3233/CH-151986 IOS Press
143
Evaluation of multimodality imaging using image fusion with MRI and CEUS in an experimental animal model P.M. Paprottkaa,∗ , P. Zengelb , C.C. Cyrana , K.J. Paprottkaa , M. Ingrischa , K. Nikolaoua , M.F. Reisera and D.A. Cleverta a b
Institute for Clinical Radiology, Ludwig Maximilian University Hospital, Munich, Germany Institute for Ear, Nose and Throat Medicine, Munich, Germany
Abstract. PURPOSE: To evaluate the diagnostic benefits of multimodality imaging using image fusion with magnetic-resonance-imaging (MRI) and contrast-enhanced-ultrasound (CEUS) in an experimental small-animal-squamous-cell-carcinoma-model for the assessment of tissue hemodynamics and morphology. MATERIAL AND METHODS: Human hypopharynx-carcinoma-cells were injected subcutaneously into the left flank of 15 female athymic nude rats. After 10 daysof subcutaneous tumor growth, CEUS and MRI measurements were performed using a high-end-ultrasound-system and 3-T-MRI. After successful point-to-point or plan registration, the registered MR-images were simultaneously shown with the respective ultrasound sectional plane. Data evaluation was performed using the digitally stored video sequence data sets by two experienced radiologists using a subjective 5-point scale. RESULTS: CEUS and MRI are well-known techniques for the assessment of tissue hemodynamics (score: mean 3.8 ± 0.4 SD and score 3.8 ± 0.4 SD). Real-time image fusion of MRI and CEUS yielded a significant (p < 0.001) improvement in score (score 4.8 ± 0.4 SD). Reliable detection of small necrotic areas was possible in all animals with necrotic tumors. No significant intraobserver and interobserver variability was detected (kappa coefficient = +1). CONCLUSION: Image fusion of MRI and CEUS gives a significant improvement for reliable differentiation between different tumor tissue areas and simplifies investigations by showing the morphology as well as surrounding macro-/microvascularization. Keywords: Image fusion, multimodality imaging, magnetic resonance imaging (MRI), contrast-enhanced ultrasound (CEUS), experimental animal model
1. Introduction The main reasons for the poor prognosis of patients suffering from squamous cell carcinoma of the head and neck are occult tumor cells that can lead to locoregional recurrence and distant metastases [9, 16]. Angiogenesis is an important, well-known mechanism influencing tumor growth and metastasis, and a number of molecular drugs represent a promising approach to inhibit tumor angiogenesis [6]. Established methods of monitoring therapy, such as assessing the size and growth behavior of a tumor during therapy ∗
Corresponding author: Philipp Marius Paprottka, Institute for Clinical Radiology, Ludwig Maximilian University Hospital, Munich, Marchioninistrasse 15, 81377 Munich, Germany. Tel.: +49 89 7095 3620; Fax: +49 89 7095 8832; E-mail: [email protected]. 1386-0291/15/$35.00 © 2015 – IOS Press and the authors. All rights reserved
144
P.M. Paprottka et al. / Image fusion with MRI and CEUS
(using RECIST criteria) or progression-free survival of patients,are less sensitive and not sufficiently specific to detect the subtle effects of these new molecular therapeutics in the early stages of therapy. Although tissue hemodynamics can be assessed in MRI using dynamic contrast-enhanced measurements [2–5, 10, 12, 17], its use and clinical applicability is still limited due to the high costs, the long required measurement times and, in particular, the lack of standardized post-processing. CEUS, on the other hand, is cheap and easily applicable in a clinical setting. In addition ultra sound enables the noninvasive depiction of different organs with high spatial and temporal resolution and without the use of ionizing radiation. Ultrasound contrast agents (e.g.Sonovue® ) contain a gas that is exhaled via the lungs. Elimination is usually within a few minutes. Since these ultrasound contrast agent are not eliminated via renal excretion (unlike iodine and gadolinium), they are not contraindicated for patients with impaired renal function. However, its application is strongly user-dependent, thus hindering a reliable assessment of longitudinal studies. By using standard, contrast-enhanced MRI images for reliable and reproducible slice-positioning, this limitation of CEUS may be overcome. It is the purpose of this study to evaluate the quality and diagnostic benefits of multimodality imaging using image fusion with magnetic resonance imaging (MRI) and contrast-enhanced ultrasound (CEUS) in an experimental small-animal squamous cell carcinoma model for the assessment of tissue hemodynamics and morphology in a preclinical setting. 2. Material and methods 2.1. Animal model and experimental protocol The study was performed with the approval of the Institutional Committee for Animal Research in accordance with the guidelines of the National Institute of Health for the care and use of laboratory animals. 6 × 106 human hypopharynx carcinoma cells were injected subcutaneously into the left abdominal flank of 15 female athymic nude rats (Charles River® , Sulzfeld, Germany/7-8 weeks old/180–220 g body weight). The animals were inspected daily to assess general appearance and tumor growth. When tumors reached a volume of ∼300 mm3 based on caliper measurements in 3 dimensions (tumor volume mm3 = a×b×c×0.5/median 10 days of subcutaneous tumor growth), MRI and CEUS measurements were performed using a 3-T MRI (MagnetomVerio® , Siemens Healthcare® , Erlangen, Germany) and a highend ultrasound system (GE® , E9® , Medical Systems, Milwaukee, Wisc.). For examinations, animals were anesthetized with intraperitoneal injections of Ketamine® (100 mg/kg bodyweight, Ketavet® , Pfizer Inc. ©, New York, NY) and Xylazine® (10 mg/kg bodyweight, Rompun® 2%, Bayer, Leverkusen, Germany). A 22-gauge butterfly catheter (B. Braun AG® , Melsungen, Germany) was inserted into a tail vein for the fast manual bolus injection of contrast media. 2.2. MRI 6 minutes after administration of 0.1 mmol/kg BWGadobutrol® (Gadovist® , Bayer Schering, Berlin, Germany), a small-molecular, macrocyclic, clinically available contrast medium, contrast-enhanced MRI was performed with a clinical 3-tesla system with the animals supine, using a customized H-phasedarray coil (Rapid Biomedical® , Rimpar, Germany) for rat hearts (actively decoupled receiving coil, 4 channels, resonance frequency of 123.3 MHz). For data acquisition, we used a 3Dflash (fast low-angle shot) sequence (48 slices, slice thickness 0.75 mm, in-plane resolution 0.5 mm, 128 × 128 matrix size, flip angle 19◦ , TR = 10.7 ms, TE = 4.9 ms).
P.M. Paprottka et al. / Image fusion with MRI and CEUS
145
2.3. Image fusion For image fusion, a magnetic field generator and a linear transducer (6–15 MHz) connected to two sensors on the transducer are the hardware components required. Dedicated software that can detect the transducers by means of a positioning system is also needed. The positioning system was used to calculate the exact position of the sensors connected to the transducer. A magnetic field generator was placed on the animal’s left side. The DICOM data set was uploaded to the ultrasonic device for data registration. Standard DICOM MRI data sets were used for image fusion. Unlike the MRI examinations, which were performed in the supine position, the ultrasound examinations were performed in the right lateral position to obtain optimum ultrasound investigation conditions; the image was adjusted and calibrated accordingly prior to the image fusion investigations. Registration was done manually using plane registration; therefore the axial contrast enhanced MRI slices were used as baseline and the ultrasound images were linked to the same plane. In cases where this technique was not successful, point-to-point registration with a minimum of three common registration points within the tumor was selected. After successful registration and image fusion the registered MR images were simultaneously shown with the respective ultrasound sectional plane [13, 14]. All special ultrasound techniques such as gray-scale US, duplex US or contrast-enhanced ultrasound were be included in the image fusion examination. The co-registration procedure took approximately 8–10 minutes, including DICOM data upload. 2.4. CEUS Technical developments over the past decade have focused on different microbubble consistencies and effective methods of detecting their nonlinear signals. The low mechanical index allows the production of real-time gray-scale images [1]. Contrast-specific techniques use a low applied acoustic pressure to produce images based on nonlinear acoustic interaction between the ultrasound system and stabilized microbubbles. These microbubbles oscillate and resonate, giving continuous contrast enhancement to gray-scale images [1, 11]. SonoVue® (Bracco® , Milan, Italy) is a second-generation contrast agent consisting of stabilized microbubbles of sulfur hexafluoride gas, which is eliminated via the respiratory system. It has low solubility, and is innocuous, isotonic with human plasma, and devoid of antigenic potential, since it contains no proteinaceous material [7, 8]. The required dose for a single injection was 0.3 ml followed by 0.3 ml of saline to improve the detection of contrast enhancement in the tumor tissue.
Table 1 Evaluation of tissue hemodynamics and morphology using gray-scale US, duplex US or CEUS and image fusion (score 0, 1, 2, 3, 4, 5) Score 0 Score 1 Score 2 Score 3 Score 4 Score 5
No detection possible, technical difficulties Morphologydetectable Macrovascularizationdetectable Microvascularizationdetectable Reliable differentiation between different areas of tumor microvascularization possible Reliable differentiation between different areas of tumor microvascularization and surrounding morphology as well as surrounding macro-/microvascularization
146
P.M. Paprottka et al. / Image fusion with MRI and CEUS
2.5. Data analysis Data evaluation was performed blind by two experienced radiologists (10 and 5 years’ experience) using the digitally stored video sequence data sets and the following criteria and scores (please see Table 1) for the analysis of all special ultrasound techniques such as gray-scale US, duplex US or contrast-enhanced ultrasound and the image fusion examinations. The authors complied with the ethical guidelines for publication in Investigative Radiology. 2.6. Statistical analysis Continuous variables are presented as the mean and standard deviation. The imaging scores were compared using the Wilcoxon matched-pairs test. Inter observer agreement was tested by means of a weighted Kappa test. Analyses were carried out in SPSS® for Windows (SPSS® , version 19, IBM® , USA). P values of 0.05) improvement in diagnosis was observed with the above-mentioned ultrasound techniques using image fusion (Figs. 1d, 1e, 2d and 2e). CEUS (Figs. 1c and 2c), and MRI are well-known techniques for the assessment of tissue hemodynamics (score: mean 3.8 ± 0.4 SD and
a
d
b
e
c
f
Fig. 2. a) Gray-scale US: Notumor inhomogeneity detectable. Small necrosis possible, cannot be excluded with certainty. b) Duplex US: No tumor inhomogeneity detectable. Small necrosis possible, cannot be excluded with certainty. No tumor vascularization detectable. c) CEUS: Homogeneously opacified tumor tissue, no necrosis detectable. d) Real-time image fusion of gray-scale US and MRI: Good correlation of the tumor margin and surrounding tissue. Reliable detection of tumor size is possible. e) Real-time image fusion of duplex US and MRI: No central detection of tumor vascularization shown by MRI is possible using duplex US. f) Image fusion of CEUS and MRI: Homogeneously opacified tumor tissue, no necrosis detectable. Investigation simplified by presentation of the surrounding morphology as well as surrounding macro-/microvascularization.
148
P.M. Paprottka et al. / Image fusion with MRI and CEUS Table 2 The lesions were evaluated by two experienced radiologists using a subjective 5-point scale
Animal No.
Gray-scale US
Duplex US
CEUS
MRI
Image fusion MRI and CEUS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Stand. dev. Mean
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 2
4 4 4 4 4 3 3 3 4 4 4 4 4 4 4 0.41 3.8
4 4 4 4 4 4 3 3 4 4 4 4 4 3 4 0.41 3.8
5 5 5 5 5 4 5 5 5 5 4 5 5 4 5 0.41 4.8
score 3.8 ± 0.4 SD/Table 2). Real-time image fusion of MRI and CEUS (Figs. 1f and 2f) yielded a significant (p < 0.001) improvement in score (score 4.8 ± 0.4 SD). Reliable detection of small necrotic areas was possible in all animals with necrotic tumors. Three animals showed no histologically-proven tumor necrosis Fig. 3b. No significant intraobserver and interobserver variability was detected (kappa coefficient = +1). 4. Discussion CEUS and MRI are well-known techniques for the assessment of tissue hemodynamics and achieved magnificent scores (score: mean 3.8 ± 0.4 SD and score 3.8 ± 0.4 SD). Not surprisingly, only the morphology or macrovascularization was detectable with conventional US techniques. Image fusion of MRI and CEUS produced a significant (p < 0.001) improvement in score (score 4.8 ± 0.4 SD) and allowed reliable differentiation between different microvascularization tumor areas and surrounding morphology as well as surrounding macro-/microvascularization. Intraobserver and interobserver variability was low and confirmed the reproducibility of the results. Multiple follow-up examinations at short intervals are necessary to monitor the subtle antiangiogenic treatment effect. The problems of (DCE-) MRI are the long exposure time, high costs, and poor availability. However, second imaging with ultrasound can be difficult and fairly time-consuming, particularly with small or multifocal lesions. Additionally ultrasound is strongly user-dependent, thus hindering a reliable assessment of longitudinal studies. By using standard, contrast-enhanced MRI images for reliable and reproducible slice-positioning, this limitation of CEUS may be overcome. In our study we therefore analyzed the potential advantages of image fusion as a new technical development in terms of the identifiability and assessment of tumor tissue. Using CEUS with the image fusion mode yielded better results as
P.M. Paprottka et al. / Image fusion with MRI and CEUS
a
149
500 µm
b Fig. 3. Hematoxylin-eosin staining of two different tumors. a) Homogeneous tumorcell-tissue of squamous cell carcinoma of head and neck (FaDu). b) Necrotic areal with destruktion of the normal tumorcell-formation (black arrow).
regards lesion identifiability than MRI or CEUS used separately, as could be shown in other studies [15]. The combination of standard US and image fusion was inferior to MRI or CEUS alone. The detection of small necrotic areas in all animals, especially within these small tumors (∼318 mm3 ), by means of image fusion indicates that detection of the tumor response during antiangiogenic treatment may be possible in the near future. The well-known limitations of US such as obesity, meteorism and noncompliance are of course valid for CEUS too, with or without image fusion mode, but due to the superficial position of the tumor and the intraperitoneal anesthetized injection, these limitations had no influence in our study. The main advantage of image fusion does however seem to be the ability to evaluate the vascularization of the tumor tissue in real time and to compare the results directly with the respective contrast-enhanced MRI images. It is not possible to discuss the consistency or inconsistency of our results in the context of previous studies because there have not previously been any published results. The limitations of the study could be that our measurements were made in an experimental animal model with optimized experimental conditions and are therefore not 100% transferable to clinical practice. 5. Conclusion To our knowledge, we are the first research group to use ultrasound-MRI image fusion in a small animal model to detect tumor necrosis and prove the great accuracy of this method in extremely small tumors (∼318 mm3 ). The image fusion of MRI and CEUS yields a significant improvement for reliable
150
P.M. Paprottka et al. / Image fusion with MRI and CEUS
differentiation between different tumor tissue areas and simplifies the investigation by presenting the surrounding morphology as well as surrounding macro-/microvascularization. Using contrast-enhanced ultrasound it is possible to depict and differentiate vital from non-vital tissue, which is particularly important for the detection of early responders and nonresponders during antiangiogenic therapy in the future. Disclosure Nothing to disclose. References [1] A. Bauer, L. Solbiati and N. Weissman, Ultrasound imaging with SonoVue: low mechanical index real-time imaging, Academic radiology 9(Suppl 2) (2002), S282–4. [2] W. Cai and X. Chen, Multimodality molecular imaging of tumor angiogenesis, J Nucl Med 49(Suppl 2) (2008), 113S–128S. [3] H.L. Cheng, C. Wallis and Z. Shou, et al., Quantifying angiogenesis in VEGF-enhanced tissue-engineered bladder constructs by dynamic contrast-enhanced MRI using contrast agents of different molecular weights, J MagnReson Imaging 25 (2007), 137–145. [4] C.C. Cyran, P.M. Paprottka and B. Schwarz, et al., Perfusion MRI for monitoring the effect of sorafenib on experimental prostate carcinoma: A validation study, American journal of Roentgenology 198 (2012), 384–391. [5] R. De Visschere, W. Oosterlinck and G. De Meerleer, et al., Clinical and imaging tools in the early diagnosis of prostate cancer, a review, JBR-BTR2010 93 62–70. [6] K. Dietze, I. Slosarek, T. Fuhrmann-Selter, C. Hopperdietzel, J. Plendl and S. Kaessmeyer, Isolation of equine endothelial cells and life cell angiogenesis assay Clin Hemorheol Microcirc 58 (2014), 127–146. [7] C. Greis, Technical aspects of contrast-enhanced ultrasound (CEUS) examinations: Tips and tricks. Clin Hemorheol Microcirc 58 (2014), 89–95. [8] C. Greis, Technology overview: SonoVue (Bracco, Milan), European Radiology 14(Suppl 8) (2004), P11–P15. [9] S. Lang, B. Wollenberg and M. Dellian, et al., Clinical and epidemiological data of patients with malignomas of the head and neck, Laryngorhinootologie 81 (2002), 499–508. [10] K.S. Lazanyi, A. Abramyuk and G. Wolf, et al., Usefulness of dynamic contrast enhanced computed tomography in patients with non-small-cell lung cancer scheduled for radiation therapy, Lung Cancer (2010) doi:S0169-5002(10)00117-0 [pii]10.1016/j.lungcan.2010.03.004. [11] R. Lencioni, D. Cioni and C. Bartolozzi, Tissue harmonic and contrast-specific imaging: Back to gray scale in ultrasound, European Radiology 12 (2002), 151–65. [12] K.A. Miles, Molecular imaging with dynamic contrast-enhanced computed tomography Clin Radiol 65 (2010), 549–556. [13] W. Wein, S. Brunke, A. Khamene, et al., Automatic CT-ultrasound registration for diagnostic imaging and image-guided intervention, Medical Image Analysis 12 (2008), 577–585. [14] W. Wein, A. Khamene and D.A. Clevert, et al., Simulation and fully automatic multimodal registration of medical ultrasound, Medical image computing and computer-assisted intervention : MICCAI International Conference on Medical Image Computing and Computer-Assisted Intervention 10 (2007), 136–143. [15] H. Wobser, R. Wiest, B. Salzberger, W.A. Wohlgemuth, C. Stroszczynski and E.M. Jung, Evaluation of treatment response after chemoembolisation (TACE) in hepatocellular carcinoma using real time imagefusion of contrast-enhanced ultrasound (CEUS) and computed tomography (CT)–preliminary results, Clin Hemorheol Microcirc 57 (2014), 191–201. [16] B. Wollenberg, A. Ollesch, K. Maag, et al., Micrometastases in bone marrow of patients with cancers in the head and neck area, Laryngorhinootologie 73 (1994), 88–93. [17] X. Wu, E.K. Jeong and L. Emerson, et al., Noninvasive evaluation of antiangiogenic effect in a mouse tumor model by DCE-MRI with Gd-DTPA cystamine copolymers, Mol Pharm 7 (2010), 41–48.
143
Evaluation of multimodality imaging using image fusion with MRI and CEUS in an experimental animal model P.M. Paprottkaa,∗ , P. Zengelb , C.C. Cyrana , K.J. Paprottkaa , M. Ingrischa , K. Nikolaoua , M.F. Reisera and D.A. Cleverta a b
Institute for Clinical Radiology, Ludwig Maximilian University Hospital, Munich, Germany Institute for Ear, Nose and Throat Medicine, Munich, Germany
Abstract. PURPOSE: To evaluate the diagnostic benefits of multimodality imaging using image fusion with magnetic-resonance-imaging (MRI) and contrast-enhanced-ultrasound (CEUS) in an experimental small-animal-squamous-cell-carcinoma-model for the assessment of tissue hemodynamics and morphology. MATERIAL AND METHODS: Human hypopharynx-carcinoma-cells were injected subcutaneously into the left flank of 15 female athymic nude rats. After 10 daysof subcutaneous tumor growth, CEUS and MRI measurements were performed using a high-end-ultrasound-system and 3-T-MRI. After successful point-to-point or plan registration, the registered MR-images were simultaneously shown with the respective ultrasound sectional plane. Data evaluation was performed using the digitally stored video sequence data sets by two experienced radiologists using a subjective 5-point scale. RESULTS: CEUS and MRI are well-known techniques for the assessment of tissue hemodynamics (score: mean 3.8 ± 0.4 SD and score 3.8 ± 0.4 SD). Real-time image fusion of MRI and CEUS yielded a significant (p < 0.001) improvement in score (score 4.8 ± 0.4 SD). Reliable detection of small necrotic areas was possible in all animals with necrotic tumors. No significant intraobserver and interobserver variability was detected (kappa coefficient = +1). CONCLUSION: Image fusion of MRI and CEUS gives a significant improvement for reliable differentiation between different tumor tissue areas and simplifies investigations by showing the morphology as well as surrounding macro-/microvascularization. Keywords: Image fusion, multimodality imaging, magnetic resonance imaging (MRI), contrast-enhanced ultrasound (CEUS), experimental animal model
1. Introduction The main reasons for the poor prognosis of patients suffering from squamous cell carcinoma of the head and neck are occult tumor cells that can lead to locoregional recurrence and distant metastases [9, 16]. Angiogenesis is an important, well-known mechanism influencing tumor growth and metastasis, and a number of molecular drugs represent a promising approach to inhibit tumor angiogenesis [6]. Established methods of monitoring therapy, such as assessing the size and growth behavior of a tumor during therapy ∗
Corresponding author: Philipp Marius Paprottka, Institute for Clinical Radiology, Ludwig Maximilian University Hospital, Munich, Marchioninistrasse 15, 81377 Munich, Germany. Tel.: +49 89 7095 3620; Fax: +49 89 7095 8832; E-mail: [email protected]. 1386-0291/15/$35.00 © 2015 – IOS Press and the authors. All rights reserved
144
P.M. Paprottka et al. / Image fusion with MRI and CEUS
(using RECIST criteria) or progression-free survival of patients,are less sensitive and not sufficiently specific to detect the subtle effects of these new molecular therapeutics in the early stages of therapy. Although tissue hemodynamics can be assessed in MRI using dynamic contrast-enhanced measurements [2–5, 10, 12, 17], its use and clinical applicability is still limited due to the high costs, the long required measurement times and, in particular, the lack of standardized post-processing. CEUS, on the other hand, is cheap and easily applicable in a clinical setting. In addition ultra sound enables the noninvasive depiction of different organs with high spatial and temporal resolution and without the use of ionizing radiation. Ultrasound contrast agents (e.g.Sonovue® ) contain a gas that is exhaled via the lungs. Elimination is usually within a few minutes. Since these ultrasound contrast agent are not eliminated via renal excretion (unlike iodine and gadolinium), they are not contraindicated for patients with impaired renal function. However, its application is strongly user-dependent, thus hindering a reliable assessment of longitudinal studies. By using standard, contrast-enhanced MRI images for reliable and reproducible slice-positioning, this limitation of CEUS may be overcome. It is the purpose of this study to evaluate the quality and diagnostic benefits of multimodality imaging using image fusion with magnetic resonance imaging (MRI) and contrast-enhanced ultrasound (CEUS) in an experimental small-animal squamous cell carcinoma model for the assessment of tissue hemodynamics and morphology in a preclinical setting. 2. Material and methods 2.1. Animal model and experimental protocol The study was performed with the approval of the Institutional Committee for Animal Research in accordance with the guidelines of the National Institute of Health for the care and use of laboratory animals. 6 × 106 human hypopharynx carcinoma cells were injected subcutaneously into the left abdominal flank of 15 female athymic nude rats (Charles River® , Sulzfeld, Germany/7-8 weeks old/180–220 g body weight). The animals were inspected daily to assess general appearance and tumor growth. When tumors reached a volume of ∼300 mm3 based on caliper measurements in 3 dimensions (tumor volume mm3 = a×b×c×0.5/median 10 days of subcutaneous tumor growth), MRI and CEUS measurements were performed using a 3-T MRI (MagnetomVerio® , Siemens Healthcare® , Erlangen, Germany) and a highend ultrasound system (GE® , E9® , Medical Systems, Milwaukee, Wisc.). For examinations, animals were anesthetized with intraperitoneal injections of Ketamine® (100 mg/kg bodyweight, Ketavet® , Pfizer Inc. ©, New York, NY) and Xylazine® (10 mg/kg bodyweight, Rompun® 2%, Bayer, Leverkusen, Germany). A 22-gauge butterfly catheter (B. Braun AG® , Melsungen, Germany) was inserted into a tail vein for the fast manual bolus injection of contrast media. 2.2. MRI 6 minutes after administration of 0.1 mmol/kg BWGadobutrol® (Gadovist® , Bayer Schering, Berlin, Germany), a small-molecular, macrocyclic, clinically available contrast medium, contrast-enhanced MRI was performed with a clinical 3-tesla system with the animals supine, using a customized H-phasedarray coil (Rapid Biomedical® , Rimpar, Germany) for rat hearts (actively decoupled receiving coil, 4 channels, resonance frequency of 123.3 MHz). For data acquisition, we used a 3Dflash (fast low-angle shot) sequence (48 slices, slice thickness 0.75 mm, in-plane resolution 0.5 mm, 128 × 128 matrix size, flip angle 19◦ , TR = 10.7 ms, TE = 4.9 ms).
P.M. Paprottka et al. / Image fusion with MRI and CEUS
145
2.3. Image fusion For image fusion, a magnetic field generator and a linear transducer (6–15 MHz) connected to two sensors on the transducer are the hardware components required. Dedicated software that can detect the transducers by means of a positioning system is also needed. The positioning system was used to calculate the exact position of the sensors connected to the transducer. A magnetic field generator was placed on the animal’s left side. The DICOM data set was uploaded to the ultrasonic device for data registration. Standard DICOM MRI data sets were used for image fusion. Unlike the MRI examinations, which were performed in the supine position, the ultrasound examinations were performed in the right lateral position to obtain optimum ultrasound investigation conditions; the image was adjusted and calibrated accordingly prior to the image fusion investigations. Registration was done manually using plane registration; therefore the axial contrast enhanced MRI slices were used as baseline and the ultrasound images were linked to the same plane. In cases where this technique was not successful, point-to-point registration with a minimum of three common registration points within the tumor was selected. After successful registration and image fusion the registered MR images were simultaneously shown with the respective ultrasound sectional plane [13, 14]. All special ultrasound techniques such as gray-scale US, duplex US or contrast-enhanced ultrasound were be included in the image fusion examination. The co-registration procedure took approximately 8–10 minutes, including DICOM data upload. 2.4. CEUS Technical developments over the past decade have focused on different microbubble consistencies and effective methods of detecting their nonlinear signals. The low mechanical index allows the production of real-time gray-scale images [1]. Contrast-specific techniques use a low applied acoustic pressure to produce images based on nonlinear acoustic interaction between the ultrasound system and stabilized microbubbles. These microbubbles oscillate and resonate, giving continuous contrast enhancement to gray-scale images [1, 11]. SonoVue® (Bracco® , Milan, Italy) is a second-generation contrast agent consisting of stabilized microbubbles of sulfur hexafluoride gas, which is eliminated via the respiratory system. It has low solubility, and is innocuous, isotonic with human plasma, and devoid of antigenic potential, since it contains no proteinaceous material [7, 8]. The required dose for a single injection was 0.3 ml followed by 0.3 ml of saline to improve the detection of contrast enhancement in the tumor tissue.
Table 1 Evaluation of tissue hemodynamics and morphology using gray-scale US, duplex US or CEUS and image fusion (score 0, 1, 2, 3, 4, 5) Score 0 Score 1 Score 2 Score 3 Score 4 Score 5
No detection possible, technical difficulties Morphologydetectable Macrovascularizationdetectable Microvascularizationdetectable Reliable differentiation between different areas of tumor microvascularization possible Reliable differentiation between different areas of tumor microvascularization and surrounding morphology as well as surrounding macro-/microvascularization
146
P.M. Paprottka et al. / Image fusion with MRI and CEUS
2.5. Data analysis Data evaluation was performed blind by two experienced radiologists (10 and 5 years’ experience) using the digitally stored video sequence data sets and the following criteria and scores (please see Table 1) for the analysis of all special ultrasound techniques such as gray-scale US, duplex US or contrast-enhanced ultrasound and the image fusion examinations. The authors complied with the ethical guidelines for publication in Investigative Radiology. 2.6. Statistical analysis Continuous variables are presented as the mean and standard deviation. The imaging scores were compared using the Wilcoxon matched-pairs test. Inter observer agreement was tested by means of a weighted Kappa test. Analyses were carried out in SPSS® for Windows (SPSS® , version 19, IBM® , USA). P values of 0.05) improvement in diagnosis was observed with the above-mentioned ultrasound techniques using image fusion (Figs. 1d, 1e, 2d and 2e). CEUS (Figs. 1c and 2c), and MRI are well-known techniques for the assessment of tissue hemodynamics (score: mean 3.8 ± 0.4 SD and
a
d
b
e
c
f
Fig. 2. a) Gray-scale US: Notumor inhomogeneity detectable. Small necrosis possible, cannot be excluded with certainty. b) Duplex US: No tumor inhomogeneity detectable. Small necrosis possible, cannot be excluded with certainty. No tumor vascularization detectable. c) CEUS: Homogeneously opacified tumor tissue, no necrosis detectable. d) Real-time image fusion of gray-scale US and MRI: Good correlation of the tumor margin and surrounding tissue. Reliable detection of tumor size is possible. e) Real-time image fusion of duplex US and MRI: No central detection of tumor vascularization shown by MRI is possible using duplex US. f) Image fusion of CEUS and MRI: Homogeneously opacified tumor tissue, no necrosis detectable. Investigation simplified by presentation of the surrounding morphology as well as surrounding macro-/microvascularization.
148
P.M. Paprottka et al. / Image fusion with MRI and CEUS Table 2 The lesions were evaluated by two experienced radiologists using a subjective 5-point scale
Animal No.
Gray-scale US
Duplex US
CEUS
MRI
Image fusion MRI and CEUS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Stand. dev. Mean
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 2
4 4 4 4 4 3 3 3 4 4 4 4 4 4 4 0.41 3.8
4 4 4 4 4 4 3 3 4 4 4 4 4 3 4 0.41 3.8
5 5 5 5 5 4 5 5 5 5 4 5 5 4 5 0.41 4.8
score 3.8 ± 0.4 SD/Table 2). Real-time image fusion of MRI and CEUS (Figs. 1f and 2f) yielded a significant (p < 0.001) improvement in score (score 4.8 ± 0.4 SD). Reliable detection of small necrotic areas was possible in all animals with necrotic tumors. Three animals showed no histologically-proven tumor necrosis Fig. 3b. No significant intraobserver and interobserver variability was detected (kappa coefficient = +1). 4. Discussion CEUS and MRI are well-known techniques for the assessment of tissue hemodynamics and achieved magnificent scores (score: mean 3.8 ± 0.4 SD and score 3.8 ± 0.4 SD). Not surprisingly, only the morphology or macrovascularization was detectable with conventional US techniques. Image fusion of MRI and CEUS produced a significant (p < 0.001) improvement in score (score 4.8 ± 0.4 SD) and allowed reliable differentiation between different microvascularization tumor areas and surrounding morphology as well as surrounding macro-/microvascularization. Intraobserver and interobserver variability was low and confirmed the reproducibility of the results. Multiple follow-up examinations at short intervals are necessary to monitor the subtle antiangiogenic treatment effect. The problems of (DCE-) MRI are the long exposure time, high costs, and poor availability. However, second imaging with ultrasound can be difficult and fairly time-consuming, particularly with small or multifocal lesions. Additionally ultrasound is strongly user-dependent, thus hindering a reliable assessment of longitudinal studies. By using standard, contrast-enhanced MRI images for reliable and reproducible slice-positioning, this limitation of CEUS may be overcome. In our study we therefore analyzed the potential advantages of image fusion as a new technical development in terms of the identifiability and assessment of tumor tissue. Using CEUS with the image fusion mode yielded better results as
P.M. Paprottka et al. / Image fusion with MRI and CEUS
a
149
500 µm
b Fig. 3. Hematoxylin-eosin staining of two different tumors. a) Homogeneous tumorcell-tissue of squamous cell carcinoma of head and neck (FaDu). b) Necrotic areal with destruktion of the normal tumorcell-formation (black arrow).
regards lesion identifiability than MRI or CEUS used separately, as could be shown in other studies [15]. The combination of standard US and image fusion was inferior to MRI or CEUS alone. The detection of small necrotic areas in all animals, especially within these small tumors (∼318 mm3 ), by means of image fusion indicates that detection of the tumor response during antiangiogenic treatment may be possible in the near future. The well-known limitations of US such as obesity, meteorism and noncompliance are of course valid for CEUS too, with or without image fusion mode, but due to the superficial position of the tumor and the intraperitoneal anesthetized injection, these limitations had no influence in our study. The main advantage of image fusion does however seem to be the ability to evaluate the vascularization of the tumor tissue in real time and to compare the results directly with the respective contrast-enhanced MRI images. It is not possible to discuss the consistency or inconsistency of our results in the context of previous studies because there have not previously been any published results. The limitations of the study could be that our measurements were made in an experimental animal model with optimized experimental conditions and are therefore not 100% transferable to clinical practice. 5. Conclusion To our knowledge, we are the first research group to use ultrasound-MRI image fusion in a small animal model to detect tumor necrosis and prove the great accuracy of this method in extremely small tumors (∼318 mm3 ). The image fusion of MRI and CEUS yields a significant improvement for reliable
150
P.M. Paprottka et al. / Image fusion with MRI and CEUS
differentiation between different tumor tissue areas and simplifies the investigation by presenting the surrounding morphology as well as surrounding macro-/microvascularization. Using contrast-enhanced ultrasound it is possible to depict and differentiate vital from non-vital tissue, which is particularly important for the detection of early responders and nonresponders during antiangiogenic therapy in the future. Disclosure Nothing to disclose. References [1] A. Bauer, L. Solbiati and N. Weissman, Ultrasound imaging with SonoVue: low mechanical index real-time imaging, Academic radiology 9(Suppl 2) (2002), S282–4. [2] W. Cai and X. Chen, Multimodality molecular imaging of tumor angiogenesis, J Nucl Med 49(Suppl 2) (2008), 113S–128S. [3] H.L. Cheng, C. Wallis and Z. Shou, et al., Quantifying angiogenesis in VEGF-enhanced tissue-engineered bladder constructs by dynamic contrast-enhanced MRI using contrast agents of different molecular weights, J MagnReson Imaging 25 (2007), 137–145. [4] C.C. Cyran, P.M. Paprottka and B. Schwarz, et al., Perfusion MRI for monitoring the effect of sorafenib on experimental prostate carcinoma: A validation study, American journal of Roentgenology 198 (2012), 384–391. [5] R. De Visschere, W. Oosterlinck and G. De Meerleer, et al., Clinical and imaging tools in the early diagnosis of prostate cancer, a review, JBR-BTR2010 93 62–70. [6] K. Dietze, I. Slosarek, T. Fuhrmann-Selter, C. Hopperdietzel, J. Plendl and S. Kaessmeyer, Isolation of equine endothelial cells and life cell angiogenesis assay Clin Hemorheol Microcirc 58 (2014), 127–146. [7] C. Greis, Technical aspects of contrast-enhanced ultrasound (CEUS) examinations: Tips and tricks. Clin Hemorheol Microcirc 58 (2014), 89–95. [8] C. Greis, Technology overview: SonoVue (Bracco, Milan), European Radiology 14(Suppl 8) (2004), P11–P15. [9] S. Lang, B. Wollenberg and M. Dellian, et al., Clinical and epidemiological data of patients with malignomas of the head and neck, Laryngorhinootologie 81 (2002), 499–508. [10] K.S. Lazanyi, A. Abramyuk and G. Wolf, et al., Usefulness of dynamic contrast enhanced computed tomography in patients with non-small-cell lung cancer scheduled for radiation therapy, Lung Cancer (2010) doi:S0169-5002(10)00117-0 [pii]10.1016/j.lungcan.2010.03.004. [11] R. Lencioni, D. Cioni and C. Bartolozzi, Tissue harmonic and contrast-specific imaging: Back to gray scale in ultrasound, European Radiology 12 (2002), 151–65. [12] K.A. Miles, Molecular imaging with dynamic contrast-enhanced computed tomography Clin Radiol 65 (2010), 549–556. [13] W. Wein, S. Brunke, A. Khamene, et al., Automatic CT-ultrasound registration for diagnostic imaging and image-guided intervention, Medical Image Analysis 12 (2008), 577–585. [14] W. Wein, A. Khamene and D.A. Clevert, et al., Simulation and fully automatic multimodal registration of medical ultrasound, Medical image computing and computer-assisted intervention : MICCAI International Conference on Medical Image Computing and Computer-Assisted Intervention 10 (2007), 136–143. [15] H. Wobser, R. Wiest, B. Salzberger, W.A. Wohlgemuth, C. Stroszczynski and E.M. Jung, Evaluation of treatment response after chemoembolisation (TACE) in hepatocellular carcinoma using real time imagefusion of contrast-enhanced ultrasound (CEUS) and computed tomography (CT)–preliminary results, Clin Hemorheol Microcirc 57 (2014), 191–201. [16] B. Wollenberg, A. Ollesch, K. Maag, et al., Micrometastases in bone marrow of patients with cancers in the head and neck area, Laryngorhinootologie 73 (1994), 88–93. [17] X. Wu, E.K. Jeong and L. Emerson, et al., Noninvasive evaluation of antiangiogenic effect in a mouse tumor model by DCE-MRI with Gd-DTPA cystamine copolymers, Mol Pharm 7 (2010), 41–48.
