Clinical feasibility of single-source dual-spiral 4D dual-energy CT for proton treatment planning within the thoracic region

Purpose: Single-source dual-spiral dual-energy computed tomography (DECT) provides additional patient information but is prone to motion between both consecutively acquired CT scans. Here, the clinical applicability of dual-spiral time-resolved DECT (4D-DECT) for proton treatment planning within the thoracic region was evaluated.

Methods and Materials: Dual-spiral 4D-DECT scans of three lung-cancer patients were acquired. For temporally averaged datasets and 4 breathing phases, the geometrical conformity of 80/140kVp 4D-DECT scans before image post-processing was assessed by normalized cross correlation (NCC). Additionally, the conformity of the corresponding DECT-derived 58/79keV pseudo-monoenergetic CT datasets (MonoCTs) after image post-processing including deformable image registration (DIR) was determined. To analyze the reliability of proton dose calculation, clinical (PlanClin) and artificial worst-case (PlanWorstCase, targeting diaphragm) treatment plans were calculated on 140kVp and 79keV datasets and compared with gamma analyses (0.1% dose-difference, 1mm distance-to-agreement criterion). The applicability of patient-specific DECT-based stopping-power-ratio (SPR) prediction was investigated and proton range shifts compared to the clinical heuristic CT-number-to-SPR conversion (HLUT) were assessed. Finally, the delineation variability of an experienced radiation oncologist was quantified on DECT-derived datasets.

Results: Dual-spiral 4D-DECT scans without DIR showed a high geometrical conformity with average NCC (±1SD) of 98.7(±1.0)% including all patient voxel or 88.2(±7.8)% considering only lung. DIR clearly improved the conformity leading to average NCC of 99.9(±0.1)% and 99.6(±0.5)%, respectively. PlanClin dose distributions on 140kVp and 79keV datasets were similar with average gamma passing rate of 99.9% (99.2%-100%). The worst-case evaluation still revealed high passing rates (average: 99.3%, minimum: 92.4%). Clinically relevant mean range shifts of 2.2(±1.2)% were determined between patient-specific DECT-based SPR prediction and HLUT. The intra-observer delineation variability could be slightly reduced by additional DECT-derived datasets.

Conclusions: 79keV MonoCT datasets can be consistently obtained from dual-spiral 4D-DECT and are applicable for proton dose calculation. Patient-specific DECT-based SPR prediction performed appropriately and potentially reduces range uncertainty in proton therapy of lung-cancer patients.