Comparing the drawn PDD data in Fig. 5 reveals that a steep dose gradient along the central axis would be observed in the case of 125I and 106Ru brachytherapy plaques, while a rather uniform dose distribution would be found along the Z-axis (along the tumor depth) in the case of the proton beam. For example, by increasing the distance from the surface to 6 mm away (the spatial range at which the tumor is deeply distributed), the PDD value will respectively reduce by about 78%, 76%, and 7% for COMS plaque, CGD plaque, and proton beam. The main reason for the observed steep dose gradient for COMS and CGD plaques is the relatively low energy nature of the emitted radiations from 125I and 106Ru radioactive sources. In this regard, the maximum electron energy which is emitted during the different beta-minus decay channels of 106Ru is about 3.5 MeV [49]. The mean emitted photon energy from 125I radioactive material is also about 28.37 keV [50]. This finding explicitly indicates that a more uniform dose distribution inside the target volume and along the Z-axis would be obtained when the proton beam is employed for irradiation of the intraocular malignancies in comparison with both photon and electron beam (125I and 106Ru brachytherapy plaques, respectively). On the other hand, for regions that are located behind the tumor volume (depths beyond 6 mm in the current study), the dose fall-off along the central axis for the proton beam is more appreciable in comparison with both CGD plaque and COMS plaque. For example, with moving from 6 mm to 10 mm depth, the PDD data decrements by about 95%, 90%, and 59% for proton beam, CGD plaque, and COMS plaque, respectively). Although the mean energy of the emitted photons from 125I is quite low, some numbers of photons will always pass the tumor and can consequently increase the received dose along the central axis at the regions which are located behind the tumor volume.
For regions before the tumor volume (such as sclera, as shown in Fig. 2), the dose during the plaque-based brachytherapy would be several times higher than the administered dose by the proton beam. In this regard, the COMS plaque would deliver the maximum dose to the overlaying regions (i.e., the presented PDD data in Fig. 5 indicates that the PDD value at the surface would be about 462%, 419%, and 108% for COMS plaque, CGD plaque, and proton beam, correspondingly).
As illustrated by Fig. 6, the lateral spreading of the isodose curves is more obvious in the case of the CGD brachytherapy plaque with respect to other studied modalities. The main reason for this issue is the multiple electrons scattering into the lateral regions, especially at lower energies, which can finally lead to the spreading of the isodose lines to the lateral regions. The special design of the CGD plaque which contains a large radioactive aperture with a 20.3 mm diameter can also contribute to this finding. This lateral spreading in isodose lines may lead to the overexposure of the laterally distributed OARs which are located in close vicinity of the tumor volume such as the ciliary body. On the other hand, the less lateral spreading of the isodose curve is found in the case of 125I COMS plaque, according to Fig. 6, which can reduce the administered dose to the lateral region adjacent to the tumor volume. The minimum lateral spreading was found for the proton beam, as demonstrated in Fig. 6. Because of high mass, the proton beam will experience small deviations from its original direction and accordingly, it can be expected that a negligible lateral spreading would be observed for the relevant isodose curve to the proton beam. Therefore, the proton beam can minimize the received dose by laterally-distributed healthy tissues among the treatment modalities understudy and consequently more efficiently spare such OARs.
As demonstrated in Fig. 6, the most desirable dose uniformity within the tumor region is relevant to the proton beam. This fact is mainly linked to the much more uniform depth dose distribution of the proton beam in comparison with the considered brachytherapy plaques of CGD and COMS. In this regard, the tumor volume falls within the SOBP region for the proton beam, while in the case of COMS and CGD plaques, the tumor region lies within a descending part of the PDD curve with a very steep dose gradient (as shown in Fig. 5).
Comparison among the obtained transverse dose profiles for the proton beam, COMS, and CGD plaque in Fig. 7 clearly shows that the uniformity of the lateral dose distribution at all studied depths is more remarkable for the proton beam. This fact is mainly linked to the uniform dose painting by applying the active beam scanning technique for the proton dose delivery system.
On the other hand, a completely non-uniform lateral dose distribution would be observed for the COMS and CGD plaques. In this regard, substantial horns are found at the filed edges in both X and Y dimensions for the CGD plaque. Increasing the off-axis distance from the plaque central axis decreases the distance between the intended dosimetry point and the inner plaque surface. Consequently, it can be expected that a more radiation dose would be received at these laterally located points and therefore, such horns will have appeared in the measured transverse dose profiles on both X-axis and Y-axis. The other potential reasons for such horns can be the non-uniform coating of the 106Ru radioactive layer upon the inner surface of the employed CGD plaque as well as the multiple electrons scattering into the lateral regions. The presence of these horns can remarkably contribute to the overexposure of the laterally distributed healthy tissues when the CGD plaque is applied for treatment. Besides, these lateral dose non-uniformities can disrupt the homogeneity of the dose distribution within the target volume.
The measured transverse dose profiles for the COMS plaque show a peak at the mid-region in both X and Y directions for all considered depths. Then, the dose value sharply reduces with moving away from the central region. Such behavior can be justified by the arrangement of 125I brachytherapy seeds loaded within the COMS plaque. As shown in Fig. 4, the radioactive 125I seeds are located at the central part of the COMS plaque and therefore it can be expected that the corresponding regions inside the phantom will absorb the maximum dose value. On the other hand, the peripheral regions of COMS plaque are empty. Therefore, much lower dose values would be received to the corresponding lateral distances inside the phantom.
More uniform dose distribution for the proton beam concerning that of the COMS and CGD plaque is also obvious when the reader compares the relevant 2D isodose distributions to the considered intraocular tumor treatment techniques in Fig. 8. Furthermore, the COMS plaque also shows better dose uniformity with respect to the CGD plaque the in X-Y plane. The presence of the horns in measured dose profiles for CGD plaque can affect the uniformity of the 2D isodose distributions in the X-Y plane. As indicated in Fig. 8, some hot spots have been formed at the lateral regions which are associated with the presence of the horns in measured transverse dose profiles for CGD plaque.
The more uniform coverage of the tumor volume by the proton beam with respect to the COMS plaque and CGD plaque is evident from Fig. 9, a fact that was previously discussed in detail. Besides, the remarkable lateral spreading of the relevant isodose curves to the CGD plaque with respect to the proton beam and COMS plaque is also obvious from the illustrated results in Fig. 9. This issue was also fully explained in previous sections.
According to a more uniform dose distribution inside the tumor region during the proton beam irradiation, it can be expected that the relevant DVH to the tumor volume shows a wide plateau region, as declared in Fig. 10. On the other hand, non-uniform dose delivery to the tumor region through applying the COMS and CGD plaque results in a descending DVH for tumor volume with a very short plateau region. The reported DVH data in Fig. 10 also demonstrates that dose uniformity inside the tumor volume is superior in the case of COMS plaque with respect to the CGD plaque. This fact was also confirmed by the obtained 3D isodose distribution data in Fig. 9. Therefore, using the proton beam and electron beam respectively results in the best and the worst performance viewpoint on the dose uniformity inside the target volume.
The demonstrated data in Fig. 10 indicates that the proton beam can more efficiently spare the distributed OARs around the tumor volume in comparison with the COMS and CGD brachytherapy plaques. This issue is mainly attributed to the minimum lateral spreading of the proton beam during the passage from the eyeball which was discussed earlier.
As illustrated by Fig. 10, healthy organs such as the eye lens, optic disc, optic nerve, and macula can be better spared using the CGD plaque rather than the COMS one. This finding can be well justified by the presented results in Fig. 6. As shown in Fig. 6, the extension of isodose lines within the above-mentioned OARs is lower in the case of CGD plaque with respect to the COMS one. Viewpoint to the fact that the eye lens, optic disc, and optic nerve are sensitive organs, using the CGD plaque would be preferred to the COMS plaque, provided that sparing other healthy organs during the radiotherapy is not to be considered.
On the other hand, healthy organs such as the sclera, choroid, retina, ciliary body, cornea, iris, and vitreous humor are better spared using the COMS plaque in comparison with the CGD one. This finding can be also justified by the presented results in Fig. 6. The spreading of the isodose lines within the above-mentioned OARs is lower in the case of COMS plaque (refer to Fig. 6) and therefore, it can be deduced that lower dose values would be received to these healthy organs when the CGD plaque is substituted by the COMS plaque.