Dosimetric Comparison of Plans Obtained by Applying 3DCRT-VMAT and Tomotherapy Radiotherapy Methods in Patients with Thyroid" Ophthalmopathy Diagnosis

While TO is defined as an effective treatment, many different approaches have been proposed in the literature for RT technique. In the study by Li et al.,[] the Lateral Counter-Field (LOF) technique is traditionally recommended with the advantage of easy and fast application; however, the most prominent disadvantage of the technique is interpreted as blocking the anterior part of the globes to minimize the dose to the lenses. As a result, it caused an insufficient dose in the anterior part of the retro-orbital adipose tissue defined as the target volume. In addition, it is difficult to achieve homogeneous dose distribution within the target with the LOF technique.

IMRT, an evolutionary form of 3DCRT, has the ability to provide a dose distribution around a more irregular and complex target volume. Furthermore, steeper dose gradients are achieved between the target and normal structures, so that planning can reduce the dose delivered to surrounding tissues without compromising target coverage. In conclusion, IMRT may be advantageous in retro-orbital delivery due to the highly irregular target volume of retro-orbital structures.

Li et al.[] reported the dosimetric superiority of IMRT in retro-orbital radiation in a dosimetric study involving 10 TO patients treated with IMRT. They concluded that IMRT has a significantly superior conformity index and homogeneity index than 3DCRT and LOF and can provide better dose savings to the eyes, lenses, and optic nerves. However, clinical efficacy and side effect results have not been reported since this was a dosimetrically retrospective study.

In our study, similar results to Li et al.[] were obtained in target volume coverage. VMAT and HT techniques representing IMRT applications are superior in target volume coverage in CI and HI terminology compared to the 3DCRT technique. When evaluated in terms of eyes, lenses, and optic nerves, VMAT and HT techniques were observed to be lower than 3DCRT in terms of maximum doses, supporting the results obtained by Li et al.[]

In a retrospective evaluation of 14 consecutive patients diagnosed with bilateral TO and treated with retro-orbital irradiation between August 2012 and August 2014, San-Miguel et al.[] determined that a dose of 10 Gy in 10 fractions was prescribed for LOF, 3DCRT, and VMAT techniques. It confirmed that VMAT provided a significantly better CI compared to 3DCRT (p=0.001) and LOF (p<0.001). Furthermore, 3DCRT was superior to LOF (p=0.007). The median HI produced by VMAT, 3DCRT, and LOF was 1.05 (1.03-1.08), 1.08 (1.05-1.14), and 1.60 (1.06-4.60), respectively, with no statistically significant difference between the three groups.

When we compared the results obtained in the study by San-Miguel et al.[10] with our current results, VMAT and HT gave more conformal results than 3DCRT when analyzed in terms of the CI of PTV (p=0.007-p=0.004). In addition to this study, although there was no significant difference between VMAT and HT in our study, it was clearly seen that HT wrapped the target volume more conformally (p=1.000).

In a study by San-Miguel et al.,[] the dose PTV values of VMAT were compared with LOF or 3DCRT. The study showed that lower minimum values (p=0.004; p=0.040) and higher maximum values (p<0.001, p=0.004) were obtained with VMAT. However, there was no difference reported in mean dose, median, D05, or D01. Additionally, no significant difference was observed between 3DCRT and LOF. In our study, unlike this situation, when evaluated in terms of minimums for PTV, 3DCRT was the lowest, and the VMAT technique was the highest. Comparison of 3DCRT with both VMAT and HT was statistically significant (p=0.000; p=0.010) and it was found to have a lower dose value. When evaluated in terms of PTV maxima, no statistical difference was found for the three techniques.

MU assessment is important for treatment in terms of low-dose exposure. The mean MU values obtained by San-Miguel et al.[] were higher for VMAT (268 MU) compared to 3DCRT (174 MU) and LOF (120 MU) (p=0.00; p=0.00). In our study, 3DCRT (258.7 MU), VMAT (497.2 MU), and HT (4569.1 MU) were found. Considering that San-Miguel et al.[] applied a dose of 10 Gy in 10 fractions, consistent MU values are observed in the current study. In addition, in our study, it was reported that the MU value of the 3DCRT technique was lower than the other two techniques.

In the study by San-Miguel et al.,[] comparing the protection of organs at risk for three different techniques, VMAT was shown to provide better protection in the eyes than 3DCRT or LOF for all dosimetric parameters (p<0.001), if the minimum dose distribution width was ignored. Significant differences between 3DCRT and LOF were reported only in mean and minimum doses. Better protection is reported to be achieved in all dosimetric parameters, with no statistical difference between VMAT and the other two techniques. When comparing 3DCRT with LOF, it should be noted that the lacrimal gland shows lower minimum and mean doses with the LOF technique; in contrast, all dosimetric parameters in the lens appear to fail significantly with 3DCRT. These results can be explained by the fact that in LOF, the anterior part of the eye where the lacrimal glands or the lens are located is blocked from radiation. While there is no significant difference in dosimetric parameters for the optic nerves between the three techniques, a different situation emerges in terms of the protection of the optic chiasm. Compared to LOF, VMAT and 3DCRT techniques resulted in a clear increase in the radiation dose for all dosimetric parameters, resulting in a pulse with a spread of the dose distribution. This results in an increase of the low-dose fields of the beams directed from the posterior area in the VMAT and 3DCRT techniques, as expected in the optic chiasm.

In our study, it was observed that VMAT provided better protection in the eyes, similar to the results obtained by San-Miguel et al.[] (p=0.003). When we evaluated the retina, there was a significant statistical difference in the VMAT treatment technique compared to 3DCRT, with better protection in favor of VMAT (p=0.003). In the lacrimal area, although there was no significant statistical difference in VMAT compared to 3DCRT (p=0.353), results in favor of VMAT were obtained. In this case, as stated by San-Miguel et al.,[] better protection can be provided in VMAT for lacrimals. However, when evaluated in terms of lenses, although there was no statistically significant difference between 3DCRT and VMAT in terms of maximum doses (p=0.072), it is observed that 3DCRT has a lower value. When analyzed in terms of optic nerves, there was no finding supporting San-Miguel et al.[] in terms of maximum doses. When the results of 3DCRT and VMAT were evaluated in terms of the optic chiasm, similar results were observed. However, it was found that both techniques were more inadequate than the HT technique.

In the study published by Wang et al.[] in 2020, the target dose for patients was prescribed as 20 Gy delivered in 10 fractions to cover 95% of the PTV, and the beam angle for double partial arc VMAT plans was 240° to 120° clockwise and 120° to 240° counterclockwise. The gantry angle for the 7-fixed-beam IMRT plans was 0°, 30°, 70°, 120°, 240°, 290°, and 330°. There was no significant statistical difference in CI between VMAT and IMRT (p=0.0673, p>0.05). When the VMAT values of the CI in this study were compared with the values in our study, the mean value of 0.606 was reported similar to the mean CI of 0.697 in our study. When compared with VMAT in terms of HI, a superior HI was observed in IMRT (p=0.0014). When the VMAT values of HI were compared with our study, Wang et al.[] reported a mean HI of 0.1175, while the mean HI in our study was 0.0941.

The results obtained by Wang et al.[] showed that VMAT produced a lower D_min (p=0.0009), higher D_max (p=0.0105), and D_mean (p=0.0276) than IMRT. Although the dose received by the maximum 5% of the PTV was higher in VMAT, there was no significant statistical difference in the dose received by the maximum 95%. Considering this result and ICRU criteria, PTV %2 and PTV %98 values were also evaluated in our study to make a statistical inference in terms of PTV. The lowest value for PTV%2 was obtained in VMAT compared to other techniques. For D98%, it was observed that the maximum dose of the VMAT technique was higher than the other techniques.

Wang et al.,[] similar to PTV, reported that VMAT produced a higher dose distribution in the left lens (D_max: p=0.0246, D_mean: p=0.0114) and a higher D_max in the right lens (p=0.0463). VMAT had the highest dose value, achieving a similar result for the lenses. The rings of both lenses received a higher D_max (left: p=0.0161 and right: 0.0034). Wang et al.[12] achieved similar results in the optic nerve (left: p=0.0050 and right: p=0.0225) and eyes (left: p=0.0045 and right: p=0.0031).

Nguyen et al.[] compared HT dosimetry and conventional half-beam technique (HBT) or nonsplit- beam technique (NSBT) in terms of target coverage and radiation dose to the lacrimal glands and lens. In the evaluation of 7 patients with CTR who received radiotherapy for disease progression at high steroid doses, 3 patients were treated with HT and 4 patients with HBT. Patients were selected from those with TO unresponsive to high steroid doses (30 mg prednisone orally every day) and severe exophthalmos and/or extraocular muscle paralysis resulting in blurred vision and/or diplopia. No patient had optic nerve compression, keratitis, or retinopathy. The anterior borders of the fields were determined as the lateral bone canthus, and the posterior borders were determined by the peripheral target volume (PTV), which included the retro-orbital soft tissues and extraocular muscles to avoid marginal misses. A total dose of 2,000 cGy in 200 cGy/fraction was delivered to the PTV. After 2008, when the HT unit (6MV photons) was installed, 3 patients were treated with an intensity-modulated radiotherapy technique.

Compared to HBT, the most common technique for TO, HT provides better coverage of the target volume and more optimal sparing of the lacrimal glands. However, since the cataract threshold for fractionated radiotherapy is estimated at 500 cGy, the maximum lens dose is higher with HT, possibly leading to an increased risk of cataracts. The undivided beam technique has been observed to provide adequate coverage of the target volume at the expense of excessive irradiation to the lens and lacrimal glands. In a retrospective study of seven patients by Nguyen et al.,[] HT, conventional HBT, and NSCT were compared dosimetrically. Target dose delivery was better with HT, and values supporting this result are reported in our study. However, lens doses were higher in this study, and unlike this study, the lowest lens dose was obtained with HT in our study.

Kargıoğlu et al.[] compared 3DCRT, VMAT, HT, and HYPERARC on single patient data. Extraocular eye muscles and retrobulbar adipose tissue were contoured as clinical target volumes. The lens, lacrimal gland, macula, brain, pituitary gland, and hippocampus were identified as organs at risk. In their treatment plans, they aimed for 95% of the target volume to receive 95% of the prescribed dose. When the maximum dose results obtained were analyzed, close values were measured between the planning techniques in the right and left lacrimal glands and the right and left retina, which are critical organs. In the right and left lenses, the best values are seen in the HT technique. The MU value was by far the highest in the HT technique (4715.5 MU).

When the HT results of our study were compared with this study, similar results were observed for the retina, lacrimal glands, and lens doses, which are critical organs. When the same study is evaluated in terms of MU values, we see that the 3DCRT, VMAT, and HT in our study support the average MU values obtained as a result of the planning. In addition, in the study by Pete et al.,[] devices with helical tomotherapy units, such as TomoTherapy Hi-Art, offer many more monitor units than conventional linear accelerators. To combat the higher head leakage resulting from this increased machine output, the jaws are made thicker, and the accelerator head is heavily shielded. This protection is effective, even reported to reduce the total peripheral dose to a lower value than conventional accelerators under certain conditions.