2174

Towards MR-guided Hyperthermia at Low Field: a Proof-of-Concept Investigation

Marco Fiorito¹, Mauro Spreiter¹, Maksym Yushchenko¹, Mathieu Sarracanie¹, and Najat Salameh¹
¹Department of Biomedical Engineering, University of Basel, Allschwil, Switzerland

Synopsis

MR-guided hyperthermia at low magnetic field is challenging due to the inherently low SNR, which typically calls for multiple signal averages. Dedicated coils are crucial at this field regime, to maximise signal acquisition. In this study we built a custom RF coil which integrates a water-based heating system to reproduce hyperthermia conditions in hands. Using a fast T₁-mapping sequence, we could successfully monitor the temperature variations generated in a phantom over 120min. A first in vivo employment in a volunteer’s hand at room temperature was also shown, achieving a spatial and temporal resolution compatible with hyperthermia requirements.

Introduction

Hyperthermia treatments rely on exposing the tissue to mild temperatures (39-43°C range) for 30 to 90min^1,2 in parallel to anti-tumour treatments, such as chemotherapy and radiotherapy, in order to enhance their efficacy^3–7. Accurate mapping of the local temperature distribution in the treated region is crucial to assess the deposited thermal dose and to avoid hot spots. Although MR thermometry established itself as a fast and reliable modality for temperature mapping, to this day invasive temperature probes are still favoured^8,9. This is mainly due to the high costs associated to the use of clinical MR scanners, the limited availability of fully integrated MRI/hyperthermia systems, and the technique limitations in adipose tissues or moving organs¹⁰. Low field scanners could represent a valid alternative for MR thermometry, thanks to the reduced economic footprint and increased flexibility. In this work we present a proof-of-concept set-up for low field hyperthermia in hands. A custom-made RF coil was developed, with an integrated water-circulating for heat delivery. Using a T₁-based temperature mapping sequence, we could successfully follow the temperature variations in a phantom heated up to 50°C over 120min. Using the same sequence/set-up, a T₁ map of a volunteer’s hand could be acquired at room temperature in 4min58s.

Methods

To reproduce hyperthermia conditions, a heating-box based on circulating hot water was built (Fig.1). A custom transceiver cylindrical RF coil with inductive coupling was wound around a 3D-printed case shaped to accommodate a hand while maximising the filling factor. The electrical characteristics measured with a VNA (E5061B-3L4, Keysight Technologies, USA) were as follows: working frequency=4.35MHz, S11=-7dB, BW=18kHz, Q=237. A flip angle map was measured using the LAM method¹¹, with a nominal flip angle (FA) of 5°. A PVC-tube (outer/inner diameters=8/6mm), connected to a temperature-controlled water circulating system (PD7LR-20, PolyScience AG, Switzerland), was integrated in the case to generate homogeneous heating. A proof-of-concept hyperthermia experiment was carried out using an agar sample (3.5%w/w agar, 0.3mM MnCl₂, 0.5%w/w ADBAC) fully fitting in the case. The temperature-control system was set to 65°C for 120min. Ground truth measures were provided by three temperature probes (OTP-A temperature sensors, Althen, Germany) inserted in the phantom (Fig.2b). T₁ maps were acquired every 15 min using a Look-Locker-based mapping technique with a bSSFP readout and an initial saturation recovery excitation. Each acquisition lasted 2min29s with the following imaging parameters: FA=5°, matrix=64$$$\times$$$27$$$\times$$$11, resolution=3$$$\times$$$3$$$\times$$$15mm³, undersampling factor=50%, TR/TE=1000/4.1ms, TI=8ms, pixel bandwidth=312Hz. The same sequence parameters were also employed to acquire one in vivo T₁ map of a volunteer’s in 4min58s (TR=2000ms). All images were acquired using a resistive biplanar MRI system (Bouhnik S.A.S., France) operating at 0.1T.

Results

The three temperature traces obtained from the agar phantom show a monotonic temperature rise (Fig.2a). T₁ maps at the different time points are displayed in Fig.2c. Temperature values computed from T₁ maps are in good agreement with the ground truth measures (not shown). Fig.3 presents the T₁ map acquired in vivo at room temperature, together with an anatomical image acquired using a 3D spoiled GRE sequence with the same resolution for comparison.

Discussion

The water-based hyperthermia set-up could successfully achieve temperatures as high as 50°C in the agar phantom over the 120-min exposure. While the probes reported comparable temperatures at the first time point, the divergence of the three traces over time suggests a temperature gradient within the phantom. This was confirmed at the end of the experiment by a thermal camera image (Fig.2b). Although this could have been predicted due to the non-uniform geometry of the box and phantom, further iterations in the design are required to achieve a more even heat distribution. The acquired T₁ maps proved able to capture the temperature variations in the front, middle and back of the phantom. An increasing gradient was also observed between subsequent time points. A first in vivo map of a volunteer’s hand was obtained in 4min58s. The acquisition time was doubled in vivo because of a doubled TR to better sample longer T₁s expected in the human body. Despite the coarse resolution, the main anatomical structures of the hand could be identified, with T₁s of different tissue types ranging roughly between 200ms to 400ms. Generally shorter T₁s were found towards the wrist, however we believe that this could be ascribed to inhomogeneities in the B₁⁺ profile (Fig1.b). Indeed, a similar gradient could also be identified in the phantom in the absence of external heating (Fig2.c). Although T₁ modelling appears fully independent on the employed flip angle α (provided that α$$$\rightarrow$$$0)¹² further investigation into the impact of the flip angle on T₁ recovery using the proposed sequence is mandatory.

Conclusions

In this proof-of-concept investigation we could successfully monitor a mimicked hyperthermia experiment in phantom using a 0.1T scanner. With the presented set-up, specifically designed for hands, we could acquire maps both in phantom and in vivo with a spatial and temporal resolution compatible with the general requirements for hyperthermia². T₁ dependence on temperature needs to be investigated in vivo to provide temperature maps and this step will follow once the required ethical approval is obtained.

Acknowledgements

This research was funded by the Swiss National Science Foundation Grants No. 170575, No. 186861 and No. 198905. The authors would like to thank the COST Action CA15209 for the insightful discussions.

References

¹Overgaard J, Suit HD. Time-Temperature Relationship in Hyperthermic Treatment of Malignant and Normal Tissue in Vivo. Cancer Res. 1979;39(8):3248-3253.

²Feddersen T V, Hernandez-Tamames JA, Franckena M, van Rhoon GC, Paulides MM. Clinical Performance and Future Potential of Magnetic Resonance Thermometry in Hyperthermia. Cancers (Basel). 2020;13(1):31. doi:10.3390/cancers13010031

³Vujaskovic Z, Song CW. Physiological mechanisms underlying heat-induced radiosensitization. Int J Hyperth. 2004;20(2):163-174. doi:10.1080/02656730310001619514

⁴Datta NR, Gómez Ordóñez S, Gaipl US, et al. Local hyperthermia combined with radiotherapy and-/or chemotherapy: Recent advances and promises for the future. Cancer Treat Rev. 2015;41(9):742-753. doi:10.1016/j.ctrv.2015.05.009

⁵Colombo R, Salonia A, Leib Z, Pavone-Macaluso M, Engelstein D. Long-term outcomes of a randomized controlled trial comparing thermochemotherapy with mitomycin-C alone as adjuvant treatment for non-muscle-invasive bladder cancer (NMIBC). BJU Int. 2011;107(6):912-918. doi:10.1111/j.1464-410X.2010.09654.x

⁶Hurwitz M, Stauffer P. Hyperthermia, radiation and chemotherapy: The role of heat in multidisciplinary cancer care. Semin Oncol. 2014;41(6):714-729. doi:10.1053/j.seminoncol.2014.09.014

⁷Horsman MR, Overgaard J. Hyperthermia: a Potent Enhancer of Radiotherapy. Clin Oncol. 2007;19(6):418-426. doi:10.1016/j.clon.2007.03.015

⁸Fani F, Schena E, Saccomandi P, Silvestri S. CT-based thermometry: An overview. Int J Hyperth. 2014;30(4):219-227. doi:10.3109/02656736.2014.922221

⁹Kok HP, Wust P, Stauffer PR, Bardati F, van Rhoon GC, Crezee J. Current state of the art of regional hyperthermia treatment planning: A review. Radiat Oncol. 2015;10(1):1-14. doi:10.1186/s13014-015-0503-8

¹⁰Odéen H, Parker DL. Magnetic resonance thermometry and its biological applications – Physical principles and practical considerations. Prog Nucl Magn Reson Spectrosc. 2019;110:34-61. doi:10.1016/j.pnmrs.2019.01.003

¹¹Balezeau F, Eliat PA, Cayamo AB, Saint-Jalmes H. Mapping of low flip angles in magnetic resonance. Phys Med Biol. 2011;56(20):6635-6647. doi:10.1088/0031-9155/56/20/008

¹²Schmitt P, Griswold MA, Jakob PM, et al. Inversion Recovery TrueFISP: Quantification of T₁, T₂, and Spin Density. Magn Reson Med. 2004;51(4):661-667. doi:10.1002/mrm.20058

Figures

Fig.1: a) Design and implementation of the RF coil with integrated water-based heating system. The case, 3D-printed using acrylonitrile butadiene styrene, is shaped to optimise contact with the hand, to maximise the filling factor of the coil. The water tube follows the carving in the top and bottom inner sides of the case, to provide homogeneous heat distribution. b) Flip angle map of the custom coil obtained using a nominal flip angle of 5°.

Fig.2: a) Temperature traces measured with the three thermal probes. The grey areas represent the time needed for the various T₁ maps. b) Thermal camera image of the agar phantom acquired at the end of the heating experiment. While the probes were inserted inside the phantom while heating, here they are placed on top of it to show their position during the experiment. c) T₁ maps acquired while heating the phantom.

Fig.3: Anatomical image and T₁ map of a volunteer’s hand at room temperature. The latter was acquired in 4min58s using a TR=2000ms.

Proc. Intl. Soc. Mag. Reson. Med. 30 (2022)

2174

DOI: https://doi.org/10.58530/2022/2174