Can virtual non-contrast images replace true non-contrast images to calculate dose in children’s craniospinal irradiation proton therapy plan?
Abstract
Background: There is consensus that non-contrast computed tomography (CT) image should be used for radiotherapy dose calculation. Although the enhanced CT can bring better target delineation for radiotherapy, it may also bring deviation to the Hounsfield unit (HU) values and furthermore dosimetry errors, especially in proton radiotherapy. Since the literature proposes that virtual non-contrast (VNC) CT can be used for photon radiotherapy dose calculation, this study aimed to investigate whether VNC can also be used for proton radiotherapy dose calculation, especially in children’s craniospinal irradiation (CSI) proton therapy plans. Methods: Five patients who underwent dual-energy computed tomography (DECT) for CSI CT simulation were retrospectively analyzed. During the arterial phase of the CT scan, the spinal volume was the clinical target volume (CTV) and organs at risk (OARs) were delineated. Meanwhile, a protective ring was created by expanding 3 mm around the CTV. Subsequently, these were transferred to both true non-contrast (TNC) images and VNC images generated from DECT. We generated the planning target volume (PTV) by a 2–4 mm isotropic expansion from the CTV. The Eclipse treatment planning system generated radiotherapy treatment plans for each patient, with dose-volume histograms (DVH) calculated based on the dose distribution. VNC-based plans were transplanted onto the TNC images to simulate the real in vivo dose distribution. The results for both VNC and TNC images were evaluated. Results: During CT value measurement, the attenuation of the VNC images was lower than that of the TNC images. To further investigate this difference, a protective ring added 3 mm outside the target area can not only prevent dose damage caused by patient movement, but also enable a more detailed observation of changes in the isodose lines. Additionally, the volume difference between the CTV and prescribed dose was calculated for both plans. Analysis of the DVH curves of the patients revealed that the entire spinal cord radiotherapy plan had the largest difference. In the comparison of the TNC and VNC radiotherapy plans, the maximum doses differed in the doses covering 100% and 95% of the volume in the whole spinal cord plans by 8%. Conclusions: Our results suggest that VNC imaging cannot replace TNC imaging for proton radiotherapy planning in patients with CSI.
Related
- Unilateral Congenital Ptosis Is Associated With Impaired Binocular Vision and Reduced Contrast Sensitivity in Children
- Multicentre phase III randomised controlled trial to evaluate effectiveness of the Both EARS (BEARS) virtual reality training package to maximise hearing abilities in children and young people with bilateral cochlear implants: the BEARS protocol
- TRPM2-mediated Ca2+ entry drives cellular senescence to irradiation-induced salivary gland dysfunction
- Confirmation of gamma irradiation–mediated inactivation of a vero-adapted African swine fever virus Lisbon 60 strain for molecular assays
- Evaluation of the nutritional and digestive impact of irradiation in nasopharyngeal cancers: the French multicenter retrospective DRAKAR study
BibTeX
@article{liu2026virtual,
title = {{Can virtual non-contrast images replace true non-contrast images to calculate dose in children’s craniospinal irradiation proton therapy plan?}},
author = {Tianye Liu and Min Fu and Guobin Qu and Cheng Tao and Chengqiang Li and Antoine Simon and Jian Zhu},
journal = {Translational Pediatrics},
year = {2026},
month = oct,
doi = {10.21037/tp-2025-607},
url = {https://hal.science/hal-05830487v1},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Can virtual non-contrast images replace true non-contrast images to calculate dose in children’s craniospinal irradiation proton therapy plan?
AU - Tianye Liu
AU - Min Fu
AU - Guobin Qu
AU - Cheng Tao
AU - Chengqiang Li
AU - Antoine Simon
AU - Jian Zhu
PY - 2026
DA - 2026/10/21
JO - Translational Pediatrics
DO - 10.21037/tp-2025-607
UR - https://hal.science/hal-05830487v1
AB - Background: There is consensus that non-contrast computed tomography (CT) image should be used for radiotherapy dose calculation. Although the enhanced CT can bring better target delineation for radiotherapy, it may also bring deviation to the Hounsfield unit (HU) values and furthermore dosimetry errors, especially in proton radiotherapy. Since the literature proposes that virtual non-contrast (VNC) CT can be used for photon radiotherapy dose calculation, this study aimed to investigate whether VNC can also be used for proton radiotherapy dose calculation, especially in children’s craniospinal irradiation (CSI) proton therapy plans. Methods: Five patients who underwent dual-energy computed tomography (DECT) for CSI CT simulation were retrospectively analyzed. During the arterial phase of the CT scan, the spinal volume was the clinical target volume (CTV) and organs at risk (OARs) were delineated. Meanwhile, a protective ring was created by expanding 3 mm around the CTV. Subsequently, these were transferred to both true non-contrast (TNC) images and VNC images generated from DECT. We generated the planning target volume (PTV) by a 2–4 mm isotropic expansion from the CTV. The Eclipse treatment planning system generated radiotherapy treatment plans for each patient, with dose-volume histograms (DVH) calculated based on the dose distribution. VNC-based plans were transplanted onto the TNC images to simulate the real in vivo dose distribution. The results for both VNC and TNC images were evaluated. Results: During CT value measurement, the attenuation of the VNC images was lower than that of the TNC images. To further investigate this difference, a protective ring added 3 mm outside the target area can not only prevent dose damage caused by patient movement, but also enable a more detailed observation of changes in the isodose lines. Additionally, the volume difference between the CTV and prescribed dose was calculated for both plans. Analysis of the DVH curves of the patients revealed that the entire spinal cord radiotherapy plan had the largest difference. In the comparison of the TNC and VNC radiotherapy plans, the maximum doses differed in the doses covering 100% and 95% of the volume in the whole spinal cord plans by 8%. Conclusions: Our results suggest that VNC imaging cannot replace TNC imaging for proton radiotherapy planning in patients with CSI.
SN - 2224-4336
SN - 2224-4344
ER -