Effects of magnetic resonance-guided high-intensity focused ultrasound ablation on bone mechanical properties and modeling
- 2.1k Downloads
Magnetic resonance-guided high-intensity focused ultrasound (MR-HIFU) is a promising technique for palliative treatment of bone pain. In this study, the effects of MR-HIFU ablation on bone mechanics and modeling were investigated.
A total of 12 healthy rat femurs were ablated using 10 W for 46 ± 4 s per sonication with 4 sonications for each femur. At 7 days after treatments, all animals underwent MR and single photon emission computed tomography/computed tomography (SPECT/CT) imaging. Then, six animals were euthanized. At 1 month following ablations, the remaining six animals were scanned again with MR and SPECT/CT prior to euthanization. Thereafter, both the HIFU-treated and contralateral control bones of three animals from each time interval were processed for histology, whereas the remaining bones were subjected to micro-CT (μCT), three-point bending tests, and micro-finite element (micro-FE) analyses.
At 7 days after HIFU ablations, edema formation around the treated bones coupled with bone marrow and cortical bone necrosis was observed on MRI and histological images. SPECT/CT and μCT images revealed presence of bone modeling through an increased uptake of 99mTc-MDP and formation of woven bone, respectively. At 31 days after ablations, as illustrated by imaging and histology, healing of the treated bone and the surrounding soft tissue was noted, marked by decreased in amount of tissue damage, formation of scar tissue, and sub-periosteal reaction. The results of three-point bending tests showed no significant differences in elastic stiffness, ultimate load, and yield load between the HIFU-treated and contralateral control bones at 7 days and 1 month after treatments. Similarly, the elastic stiffness and Young’s moduli determined by micro-FE analyses at both time intervals were not statistically different.
Multimodality imaging and histological data illustrated the presence of HIFU-induced bone damage at the cellular level, which activated the bone repair mechanisms. Despite that, these changes did not have a mechanical impact on the bone.
KeywordsHIFU Focused ultrasound Bone Ablation Bone mechanical properties
High-intensity focused ultrasound (HIFU) is a non-invasive thermal therapy, which uses acoustic energy to locally heat the tissue to ablative temperature, thereby leading to cell death [1, 2]. This technique is often performed under the guidance of magnetic resonance imaging (MR-HIFU) to aid treatment planning, enable temperature monitoring to ensure precise heating during treatments, and immediate follow-up assessments thereafter. To date, MR-HIFU has been utilized in the clinic for treatment of uterine fibroids [3, 4, 5] and is currently under investigation for treatment of prostate [6, 7, 8], breast [9, 10], liver [11, 12], and pancreatic [13, 14] cancers.
In the recent years, MR-HIFU has also been approved as an alternative palliative treatment method for cancer-induced bone pain in radiation refractory patients. This approach was assumed to alleviate pain through periosteal denervation. In patients with bone metastases, MR-HIFU ablations provided pain relief in 60–100 % of the patients [15, 16, 17, 18, 19, 20]. Similarly, in patients with osteoid osteoma, complete pain relief was observed in 90 % of the patients from 1 month after treatments until the 12-month follow-up period . Interestingly, the palliative efficacy of MR-HIFU ablation has been extended to treatment of osteoarthritic pain of the facet joints [22, 23] and knee . In the knee osteoarthritic pain study, the visual analog scale (VAS) scores of 75 % of the patients were reduced from the third day following treatments and the observed pain relief effects remained at 6-month follow up . In addition to pain palliation, preclinical  as well as clinical [16, 18] results have shown the prospect of HIFU-induced osteogenesis. Bucknor et al. demonstrated that MR-HIFU ablations of healthy pig femurs caused new bone formation at the treated sites . In clinical studies, ablation of osteolytic lesions led to de novo mineralization of cortical bone [16, 18].
Although MR-HIFU ablation provides pain relief and promotes skeletal remodeling, this treatment method also induces cortical bone damage. As illustrated by different preclinical studies using healthy bones, osteocyte necrosis was detected in the ablated cortical bone [26, 27, 28]. Therefore, aggressive ablation of bone may compromise bone strength. Recently, the effects of MR-HIFU treatments on the mechanical properties of bone were first examined by Herman et al. in pig ribs . In their study, a reduction in bone mechanical properties was observed at 6 weeks after HIFU treatments, but these changes reversed at 12 weeks after ablations. A drawback of their study is that non-weight-bearing bones were used and that ribs have a different morphology and function than long bones. As weight-bearing bones are more at risk of pathological fractures, assessing the effects of MR-HIFU on the biomechanical properties of weight-bearing bones is essential and warranted . Therefore, in this study, the effects of MR-HIFU ablation on the mechanical properties of weight-bearing bones in a rat animal model were investigated by using three-point bending tests and micro-finite element (micro-FE) analyses. The animal model was chosen as it allows a full comparative study with small animal imaging, histology, and eventually tumor-bearing bones. We chose the time points of 7 days and 1 month for our study to reduce transient effects directly after HIFU treatments, while keeping it comparable to other ablation studies in small animals [31, 32, 33, 34]. Subsequently, the findings were correlated with bone damage and modeling as assessed by multimodality imaging techniques and histological analysis.
Multimodality imaging assessments of MR-HIFU ablation
The soft tissue changes due to MR-HIFU ablation were assessed using MR images taken with a whole body SENSE rat coil (Rapid Biomedical, Germany) and a 3-T MRI scanner (Philips Healthcare, The Netherlands). T2-weighted MR images were acquired using a TSE sequence (TR = 8724 ms, TE = 28 ms, FOV = 70 × 70 × 25 mm3, voxel size = 0.35 × 0.35 × 0.70 mm3, NSA = 4) for evaluation of edema. T1-weighted FFE (TR = 20 ms, TE = 5.3 ms, FOV = 70 × 70 × 25 mm3, voxel size = 0.35 × 0.35 × 0.70 mm3, NSA = 4) images were obtained prior to and after injection of DOTAREM® at 0.2 mmol/kg body weight to assess tissue damage.
The area with high bone turnover following HIFU treatments was investigated using a high resolution small animal SPECT/CT scanner (NanoSPECT/CT®, Bioscan, USA) equipped with four detector heads and converging 9-pinhole collimators (pinhole diameter = 2.5 mm) in combination with 99mTc-MDP. 99mTc-MDP was injected at 76 ± 3 MBq. The CT of both femurs was acquired with 360 projections, 2-s exposure time per projection and a peak tube voltage of 65 kV, to provide anatomical reference. At 2 h after injection, SPECT images of both femurs were acquired with 32 projections and 200 s per projection. SPECT and CT images were reconstructed, and qualitative assessments of the SPECT images were done using InVivoScope software (Bioscan).
The bone micro-architecture of the treated and contralateral femur was acquired using a vivaCT 40 scanner (Scanco Medical AG, Switzerland). μCT images of the whole femur were acquired with a voxel size of 25 μm (45 kVp, 175 μA, 500 projections per 180°, 300-ms integration time).
Three-point bending tests
Micro-FE analyses of bone biomechanical properties
The μCT scans were rotated in such a way that their orientation corresponded to that in the three-point bending experiments. In order to reduce the total number of elements for the micro-FE analyses, the image resolution was reduced to 50 μm. Thereafter, images were thresholded using a single threshold of 500 per mille of the maximum possible value to segment solely the normally mineralized bone tissue. Then, the images were cropped to include only the region between the bottom rollers plus 0.5 mm on each side. The images were converted to micro-FE models using a voxel conversion technique . Linear elastic material properties were assigned to all materials with a Young’s modulus of 10 GPa for the bone elements while the Poisson’s ratio was set to 0.3. Boundary conditions were chosen to represent the rollers’ support conditions at the bottom while a vertical displacement was prescribed at the location where the top roller was in contact with the bone. The bone stiffness was defined similarly to the three-point bending experiment. Finally, the ratio of the experimentally determined stiffness and the micro-FE-calculated stiffness was determined. After multiplying this ratio with the bone tissue value of 10 GPa prescribed in the micro-FE models, the actual tissue modulus can be back calculated . All image processing and micro-FE simulations were done using IPL v5.16 (Scanco Medical AG, Brüttisellen, Switzerland).
The femurs were decalcified in 12.5 % ethylenediaminetetraacetic acid pH 7.4 solutions. Next, samples were processed for paraffin embedding. The femurs were sectioned at a 4-μm thickness using a microtome. Subsequently, the samples were machine-stained (ArtisanTM Link Pro, Dako, Belgium) with hematoxylin and eosin (H&E). Histological images were acquired at 20× magnification (ScanScope XT, Aperio, USA).
The Wilcoxon signed rank test was used for all paired comparisons of the HIFU-treated versus the contralateral control bones in the three-point bending tests and micro-FE analyses. The two-sided t test was used to compare if the bone mechanical properties (elastic stiffness, ultimate load, yield load, and Young’s modulus) changed from 7 days to 1 month due to HIFU treatments, to investigate the changes in mechanical properties of bones due to growth from 7 days to 1 month, and to evaluate the differences in stiffness obtained from the three-point bending tests and micro-FE analyses. Analyses were performed using SPSS Statistics (v22, IBM, USA). Data are presented as the mean ± standard deviation (sd), with P < 0.05 considered significant.
Multimodality imaging of MR-HIFU ablation
At 1 month after MR-HIFU ablation, edema around the treated femurs had resolved (Fig. 4c). Interestingly, the two lines with hypo-intense signal still remained, but the distance between these lines had shrunk from 18.6 to 12.6 mm (Fig. 4b, c). The T1-weighted images showed comparable hypo-intense signal lines within the bone marrow (Fig. 4f, i). Following injection of contrast agents, the non-perfused region in the surrounding tissue had diminished (Fig. 4i). The intramedullary space showed regions of hypo-intense signal, indicating bone marrow damage (Fig. 4i). In addition, similar to MR images, the distance between the HIFU-induced hotspots in SPECT/CT images had decreased from 19.3 to 13.9 mm, confirming progression of bone modeling (Fig. 4k, l). Moreover, increased uptake of 99mTc-MDP was observed on the cortical bone surface. As expected, visual inspection of the μCT images confirmed the presence of woven bone formation and cortical thickening (Fig. 4o).
Three-point bending test
At 1 month after MR-HIFU treatments, the surrounding soft tissue had started healing leading to a smaller area of soft tissue damage, which was in line with the MR images (Figs. 4 and 7g), while the cortical bone remained necrotic (Fig. 7h). A part of the necrotic marrow had been replaced by an immature collagenous matrix made up of fibrovascular structures, an indication of bone marrow repair (Fig. 7i). Similar to the 7-day time point, woven bone was present in the intramedullary space (Fig. 7j). Consistent with the imaging results (Fig. 4), the distance between the woven bones in the intramedullary space had decreased (Fig. 7g). Although the woven bone in the intramedullary space had moved inward to allow for the return of the viable marrow cells, the adjacent cortical bone remained necrotic (Fig. 7k), suggesting a faster bone marrow healing process compared to the cortical bone. In agreement with the μCT results (Fig. 4o), the woven bone on the periosteal surface was more calcified, thereby contributing to cortical thickening (Fig. 7l).
Bone pain as a result of tumor growth affects 75 % of patients with advanced cancer . These patients experience not only severe pain but also reduced quality of life. As radiation therapy (RT) is a suboptimal pain rescue treatment approach for localized painful lesions, development of alternative treatments for patients who do not respond to RT is crucial [37, 38]. MR-HIFU ablation is a promising technique for radiation refractory patients. Through understanding the bone functional changes and bone modeling induced by MR-HIFU, this technique can be applied in a safer and more effective manner. Herein, using multimodality imaging techniques, three-point bending tests, micro-FE, and histological analyses, the effects of MR-HIFU ablation on weight-bearing bone were investigated.
In this study, by means of multimodality imaging and histological analyses, we show that MR-HIFU ablations led to necrosis of bone marrow, cortical bone, and the surrounding soft tissue at 7 days post treatments, which is in agreement with previously published results [26, 27, 28]. At 1 month after treatment, soft tissue healing was noted, marked by a reduction in amount of soft tissue damage surrounding the bones. Moreover, as illustrated by H&E images, bone healing was evident through formation of woven bone and fibrovascular structures along with an immature collagenous matrix in the intramedullary space at the investigated time points. These bone healing processes are similar to those that occurred in response to mechanically induced bone marrow ablation in long bones [33, 34, 39]. With time, resorption of the woven bone will ensue to allow re-establishment of the hematopoietic and fat cells in the bone marrow cavity , a phenomenon also present in this study. In addition, sub-periosteal reaction adjacent to ablation sites, an indicator of intramembranous bone regeneration, was detected and resembles the results observed by others [26, 28, 34].
Besides that, data from the three-point bending tests show that the elastic stiffness, ultimate load, and yield load of the HIFU-treated bones did not significantly differ from the contralateral control bones at 7 days and 1 month after treatments, which were in agreement with previously published results on non-weight-bearing bones . As expected, between the 7 days and 1 month time intervals, significant differences were observed in mechanical properties, which was due to the presence of normal growth. However, changes in mechanical properties due to MR-HIFU ablations alone were not discerned between the investigated time intervals. By combining the experimental and micro-FE results, the Young’s moduli of bone tissue were obtained to examine potential changes at bone tissue level, which could be caused, for instance, by micro-cracks or collagen damage. Since no significant differences in the tissue Young’s moduli were found at both time intervals following treatments, we conclude that this is not the case. In line with previous observations, no HIFU-induced fractures were observed following treatments [17, 29]. One possible explanation for the lack of any significant difference in bone stiffness and strength between the treated and control bones is the fact that necrotic bone at the investigated time intervals has the same mechanical properties as healthy bone. In our study, we deliberately assessed the effect of HIFU ablation using healthy bones to first understand HIFU effects as long as the bone structure and stability are not compromised by the presence of lesions. In treatment of bone lesions, particularly those with osteolytic phenotype, the effects of MR-HIFU would differ as these tumor-bearing bones are more prone to formation of fractures, but HIFU may also cause re-mineralization of the lesion. Therefore, future studies should be performed on lesion-bearing bones using the tools presented here.
In conclusion, using MRI, SPECT/CT, μCT, and histological analyses, we demonstrated that MR-HIFU ablations induced bone damage at the cellular level, thereby triggering bone repair and modeling. However, based on three-point bending tests and micro-FE simulations, the resulting cellular damage did not compromise the mechanical function of the bone or cause micro-cracks at the bone tissue level.
We would like to thank Suzanne Kivits, Monique Berben, (Philips Research, The Netherlands), Caren van Kammen, and Marije Janssen (Maastricht University, The Netherlands) for their excellent support with the in vivo experiments. This work was supported by the Center for Translational Molecular Medicine (HIFU-CHEM, Grant 03O-301).
- 7.Dickinson L, Ahmed HU, Kirkham AP, Allen C, Freeman A, Barber J, et al. A multi-centre prospective development study evaluating focal therapy using high intensity focused ultrasound for localised prostate cancer: the INDEX study. Contemp Clin Trials. 2013;36:68–80.PubMedCentralPubMedCrossRefGoogle Scholar
- 13.Sung HY, Jung SE, Cho SH, Zhou K, Han J-Y, Han ST, et al. Long-term outcome of high-intensity focused ultrasound in advanced pancreatic cancer. Pancreas. 2011;1080–1086.Google Scholar
- 19.Huisman M, Van Den Bosch MAAJ, Wijlemans JW, Van Vulpen M, Van Der Linden YM, Verkooijen HM. Effectiveness of reirradiation for painful bone metastases: a systematic review and meta-analysis. Int J Radiat Oncol Biol Phys. 2012;8–14.Google Scholar
- 25.Bucknor MD, Rieke V, Seo Y, Horvai AE, Hawkins RA, Majumdar S, et al. Bone remodeling after MR imaging-guided high-intensity focused ultrasound ablation: evaluation with MR imaging, CT, Na(18)F-PET, and histopathologic examination in a swine model. Radiology. 2014;132605.Google Scholar
- 32.Wijlemans JW, Deckers R, van den Bosch MAAJ, Seinstra BA, van Stralen M, van Diest PJ, et al. Evolution of the ablation region after magnetic resonance–guided high-intensity focused ultrasound ablation in a Vx2 tumor model. Invest Radiol. 2013;48.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.