2366 - Dosimetric Analysis for Clinical Treatment Volumes Using a Novel CBCT Imaging System

Purpose/Objective(s): This work evaluates dosimetric accuracy of a novel high quality cone beam CT (CBCT) platform, which has demonstrated improved image quality over standard image guidance CBCT¹. Dosimetric accuracy in clinical treatment planning was compared between novel and standard CBCT imaging systems through analysis of dose volume histograms (DVHs) of planning target volumes (PTVs) and organs at risk (OARs).

Materials/

Methods: Thirty subjects were recruited to a clinical imaging study, where deep inspiration or deep expiration breath hold was required during imaging and radiation treatment of tumors within liver, abdomen, pancreas, breast, chest wall, and lung sites. Subjects were imaged with fan beam CT (FBCT), standard image guidance CBCT, and the novel CBCT imaging system. Both CBCT image sets were rigidly registered to FBCT, contours were copied to the registered images, and clinical treatment plans were forward calculated. Four participants were excluded as GTV motion between CBCT and FBCT exceeded 0.5 cm. Differences in DVH data between FBCT and both CBCT modalities were calculated across all subjects, for the PTVs and adjacent OARs varying by treatment site.

Results: Table 1 shows the average differences in the DVH data of PTVs between FBCT and novel CBCT, and between FBCT and standard CBCT (N=28 PTVs across 26 subjects).  No statistically significant differences were observed between these two groups; however, the interquartile ranges show consistently less variance in the novel group compared to the standard group.  For the V105 and D99 measures, the interquartile ranges for the standard group were over double that of the novel group. When considering dose to the ipsilateral lung for breast and chest wall plans (N=10), D2 and D5 (indicative of dose hot spot within the lung), and V50 (indicative of low dose distribution within the lung) measures are statistically significantly more accurately calculated on novel CBCT over standard CBCT (p<0.05, Wilcoxon signed rank), as evaluated against FBCT. Furthermore, these measures are consistently lower for plans calculated on standard CBCT relative to FBCT.

Conclusion: The results indicate an underrepresentation of ipsilateral lung dose on standard CBCT that is improved with novel CBCT, suggesting that the novel CBCT platform may outperform standard CBCT in dosimetric accuracy at boundaries with sharp changes in Hounsfield units. This novel imaging system may have the potential for expanding the role of CBCT beyond image guidance to treatment planning in adaptive radiotherapy.

[1] Robar JL et al, PRO 2023