KEY POINTS
- This single-centre retrospective dosimetric study generated 30 VMAT plans from 26 imaging datasets drawn from the randomized DART trial and a subsequent clinical simulation-free RT workflow. Nineteen plans treated spinal targets and 11 pelvic or proximal femoral bone targets.
- All cases were retrospectively replanned to 20 Gy in 5 fractions using an automated VMAT script. Planned dose was calculated on an existing diagnostic CT, while an identical plan was recalculated on the first-fraction CBCT as a pragmatic estimate of delivered dose.
- Target coverage was remarkably stable: mean relative PTV D95 was 99.6% ± 2.4%, with a median of 99.7%. Every plan maintained a delivered PTV D95 of at least 97% of its planned value, with no clinically meaningful undercoverage.
- PTV V90 changed only minimally, from a mean 100.0% on diagnostic-CT planning to 99.7% on CBCT recalculation, an absolute difference of −0.3%. Relative D95 did not differ between spine and pelvic/proximal femur targets (p=0.52).
- OAR differences were generally small. Spinal canal D0.03 cm³ changed by only 0.45% ± 1.75% (p=0.33); selected kidney metrics differed statistically, although six cases had only partial kidney visualization within the CBCT field of view.
- Differences in Hounsfield units between diagnostic CT and CBCT showed only weak relationships with dose variation. A homogeneous-density sensitivity analysis still produced a relative PTV D95 of 99.9% ± 1.7%.
- Important limitations are that these VMAT plans were not actually delivered, only first-fraction anatomy was assessed, and CBCT truncation sometimes required missing-tissue compensation. Careful diagnostic-image selection and standard IGRT remain essential.
CLINICAL TAKEAWAY
For selected palliative bone RT patients, a separate simulation CT may not be necessary to generate a robust conformal VMAT plan. This study strengthens the dosimetric case for nonadaptive diagnostic-CT-based workflows using standard linacs and CBCT, but clinical implementation still depends on careful imaging eligibility and reproducible setup.