Polymer gel dosimetry for experimental verification of conformal small animal irradiation at a preclinical research platform

In translational research, there is a need for preclinical studies of high quality that corresponds to the conformal dose distributions conventionally delivered to humans in the modern radiotherapy clinic. To facilitate this need, novel preclinical systems consist of preclinical radiation platforms and small animal treatment planning systems. However, small-field dosimetry is challenging and requires dosimeters with high spatial resolution. In this study we demonstrate the feasibility of experimentally validating the dose distribution in small fields at a preclinical X-ray research platform using a polymer gel dosimeter and MRI-readout.


Introduction
Preclinical experiments with small animals are an important step between in vitro research and clinical implementation.To provide translatable results, preclinical experiments are required to follow the development of modern radiotherapy techniques with conformal dose delivery [1].Radiotherapy platforms dedicated for preclinical research, such as XStrahl's XenX, as well as novel treatment planning systems (TPS) adapted for small animals, allow for precise irradiation of tumor and normal tissue [2,3].For high quality studies, it is desirable to experimentally verify the dose distribution calculated by the treatment planning system, and complex treatment plans require precise dosimetry tools.A polymer gel dosimeter is a tissue equivalent dosimeter with a high 3D spatial resolution, high reproducibility, and the potential to provide accurate small field dosimetry [4].
The aim of this study was to 1) investigate the dose response and reproducibility of a polymer gel with MRI readout in small fields at a preclinical X-ray research platform, and 2) investigate if the polymer gel dosimeter can verify the conformal small field dose distribution calculated by the preclinical TPS µRayStation.

Manufacturing of a polymer gel dosimeter
The polymer gel N-isopropylamide (NIPAM), with low toxic monomer composition, was used [5,6].Manufacturing of the polymer gel was performed according to the recipe of Waldenberg et al., under normoxic conditions in a fume hood.Water was heated and stirred using an IKA ® C-MAG HS 7 control.After slowly adding 5% w/w gelatine from porcine skin (gel strength 300 type A, Sigma-Aldrich) and letting it dissolve, 3 % w/w NIPAM (97 % Sigma-Aldrich) and 3% w/w N,N'-methylene-bis-acrylamide (99 % Sigma-Aldrich) was added.After the mixture was cooled to 38°C, 21-26 mM of the antioxidant THPC (80 % solution in water, Sigma-Aldrich) was added.The gel mixture was dispensed in 15 ml vials.Before irradiation, the gels were stored in a dark room-temperature environment for 24h.

Polymer gel irradiation
Vials with gel were positioned one-by-one on top of solid water slabs at a source-to-surface distance (SSD) of 33 cm inside the preclinical research platform XenX (XStrahl Ltd, UK, Figure 1a), and irradiated with a 220 kV X-ray beam.To investigate the reproducibility in the gel response, seven vials were irradiated at each dose of 0.5, 2, 5 and 10 Gy (batch I).To investigate the dose response of the gel, vials were irradiated in the range 0.5-40 Gy (n=2/dose level, batch II).To investigate the feasibility to use the gel dosimeter for absorbed dose measurements in small field irradiations, vials were positioned in a narrow gutter in the XenX cabinet (Figure 1b) and irradiated at an SSD of 34 cm using three beams with gantry angles 135°, 180° and 225°, and field sizes of 5x5 mm 2 (batch III).A calibration curve was obtained by irradiating gels in the same batch with known absorbed doses.

Polymer gel read-out
At 24h after irradiation, vials were scanned at a General Electric Architect 3T MRI system using a fast spin echo (FSE) pulse sequence with a repetition time of 4000 ms and various effective echo times (TEeff), to obtain T2 weighted MR images in the coronal plane (Figure 2).The signal (S) of the MR images were extracted using Mice Toolkit 2021.2.1, and the relaxation time (R2) was obtained by fitting the data to a monoexponential model  = (0) −  * 2 .

Quantifying dose distribution
The R2 map obtained from MR images was calibrated to absorbed dose and compared to calculated dose distribution from the Monte Carlo based preclinical TPS µRayStation (RaySearch, Sweden) commissioned for the 220 kV beam of the XenX irradiator.In the TPS, a vial (ρ=2.33g/cm 3 ) with gel  (ρ=1g/cm 3 ) was built and the treatment set-up was reconstructed.A calculated dose profile was extracted along the beam axis of the 180° beam, and compared to the corresponding measured dose profile from the MR image.

Polymer gel dose response reproducibility and linearity
The T2 weighted signal from the MR images decreased with increasing dose.For the reproducibility assessment (batch I), the mean R2 ± SD for vials (n=7/group) irradiated with 0.5, 2, 5 and 10 Gy respectively, were 1.405 ± 0.016 s -1 , 1.477 ± 0.010 s -1 , 1.743 ± 0.015 s -1 and 2.231 ± 0.021 s -1 , showing a high reproducible gel response.For the dose response assessment (batch II), the R2 values increased linearly (at a sample correlation coefficient r 2 of >0.99) up to a dose of 12 Gy, after which the response saturated (Figure 3).For higher doses, a second-degree polynomial function could provide a more accurate description of the dose response.Based on the linear fit, the dose-R2 sensitivity of the gel dosimeter (i.e., the slope of the dose-response curve) in the dose range 0.5-40 Gy was 0.1 s -1 Gy -1 .In future work, attempts will be made to improve the sensitivity by refining the gel recipe, e.g., by increasing the percentage of NIPAM or adding acetone as a co-solvent to allow for increased concentration of N,N'-methylene-bis-acrylamide [7].

Polymer gel dosimeter to quantify dose disposition in preclinical experiments
As illustrated in Figure 4, the agreement between the absolute dose distribution measured with the polymer gel dosimeter and the dose distribution calculated with the TPS was good.In the central part of the profile (± 3 mm from isocenter) the absolute mean deviation was 2.9%.The mean absolute deviation over the entire profile was 6.3%.These results indicate that the polymer gel investigated can be useful for small field dosimetry.

Conclusion
In this study we demonstrate the feasibility of experimentally validating the dose distribution of conformal small animal irradiations at a preclinical research platform using polymer gel dosimetry and MRI-readout.In future work we will refine the gel mixing procedure for increased sensitivity and optimize the MRI-readout to reduce noise.

Figure 1 .
Figure 1.The preclinical research platform XenX (a) and the setup used for irradiation of gel with 220 kV using 5x5 mm 2 beams (b).

12th
International Conference on 3D and Advanced Dosimetry Journal of Physics: Conference Series 2630 (2023) 012034

Figure 2 .
Figure 2. Coronal MR images of a gel batch.Images were acquired using a fast spin echo pulse sequence with TR = 4000 ms and sixteen different TEeff ranging from 14 ms to 1140 ms.The MR images correspond to the shortest (left panel), middle (middle panel), and longest (right panel) TEeff used.

Figure 3 .
Figure 3. Dose response for the polymer gel irradiated with 220 kV X-rays.The R2 values are fitted with a second-degree polynomial function (solid line).For doses ≤12 Gy, the data are also fitted with a linear function (dotted line).

Figure 4 .
Figure 4.The MR image (a), MR dose image (b), and treatment plan (c) of the vial irradiated with three 5x5 mm 2 beams, as well as a comparison between an absolute dose profile extracted from the TPS (blue) and the MR dose image (orange) (d).The yellow line in the MR image indicates the position of the profile.