KEY POINTS
- The study retrospectively evaluated 10 adult patients treated with craniospinal irradiation in the lateral decubitus position at the New York Proton Center during 2024–2025. Treatments used pencil-beam scanning in a fixed-beam room with brain, upper-spine and lower-spine isocenters; 30 Gy in 10 fractions was the typical prescription.
- Two workflows were compared. Continuous full-field image guidance aligned all three isocenters before delivery at every fraction, whereas selective field-based image guidance used full alignment at the first fraction followed by field-specific verification during subsequent treatments.
- Translational accuracy was similar across workflows: mean deviations for continuous versus selective guidance were 0.09 vs 0.08 cm for brain, 0.17 vs 0.16 cm for upper spine, and 0.12 vs 0.12 cm for lower spine, with no significant differences. Mean rotational deviations also remained below 1.5° across all isocenters.
- Dosimetric robustness was maintained. Mean change in D95 was −0.87% with continuous guidance versus −0.19% with selective guidance (p=0.119), and differences in dose gradients across field junctions were small and not statistically significant.
- Selective imaging reduced mean setup time from 23.9 ± 7.1 to 18.5 ± 9.3 minutes, although the difference did not reach statistical significance (p=0.069). Approximately 90% of selective-guidance sessions versus 80% of continuous-guidance sessions were completed within 30 minutes.
- The cohort was small, setup assessment relied predominantly on two-dimensional oblique kV imaging with limited sensitivity to roll, rotational errors were not included in dose recalculation, and respiratory or intrafraction motion was not explicitly measured. Continuous full-field alignment may therefore remain useful when immobilization reproducibility deteriorates or global realignment is required.
CLINICAL TAKEAWAY
For lateral-decubitus proton CSI in a fixed-beam room, selective field-based imaging appears capable of reducing setup burden without measurable loss of geometric or dosimetric performance. The workflow is attractive as a default strategy, but the evidence remains technical and institution-specific, with comprehensive imaging still appropriate for selected difficult setups.