Author: Qianyi Xu, Inhwan Yeo 👨🔬
Affiliation: INOVA Schar Cancer Institute, Thomas Jefferson University 🌍
Purpose:
Four-dimensional computed tomography(4DCT) is prone to a geometrical error when respiration changes upon patient shift. 4DCT with a longer-length detector(16cm), uninvolving the shift, has been shown to improve geometrical accuracy, compared with a conventional CT(4cm), with some HU error. The two were dosimetrically validated.
Methods:
A lung-mimicking phantom with three tumor-shaped targets and their respiratory traces(from treatments) was respectively imaged when the lung was moved to each of the ten-positions/phases (1) step-wisely by helical scan at each movement(ground truth; GT), (2) continuously by 4D scan with 16cm detectors, and (3) the same with 4cm detectors. Irregularities(baseline and/or amplitude change of the traces) affected #3 only in its shifted position. Treatments plans included volumetric modulated arc therapy(VMAT) and intensity-modulated proton therapy(IMPT) on averaged CT images from #2 and 3. IMPT was targeted to internal target volumes with density overrides(1.0 g/cm3), while VMAT to ITVs+3mm, with the objective Vprescribed dose≥95%. The plans were recalculated on each phase image of #1(GT) and those of #2 were additionally calculated on its own phase images. The calculated doses were transferred to the maximum-exhale image, and summed.
Results:
(1) 16cm CT: two IMPT plans met the objective, despite under-doses in some phase images, but one(patient #2) fell short by 94.5% at Vprescribed dose. Two VMAT plans met the objective, but not for patient #1(93%). (2) 4cm CT: all IMPT plans met the objective. VMAT plans provided 83%(#1), 91%(#2), and 93.5%(#3). Although the 16cm CT provided better results, affected by inherent limitations(motion averaging, phase sorting) they were not fully satisfying. (3) 4D dose of the 16cm CT agreed with that of GT by <2% in IMPT and <1-5% in VMAT.
Conclusion:
The 16cm 4DCT showed superior dosimetry, but more data are needed. Extra margins for planning on averaged CT may reduce the 4D dosimetry failure.