Clinical implementation of dual-energy CT for proton treatment planning to reduce CT-based range uncertainties

Purpose/Objective:

Particle treatment planning is particularly afflicted by CT-based range uncertainties. A clinical application of dual-energy CT (DECT) provides additional tissue information to potentially achieve more precise range predictions compared to single-energy CT (SECT). Therefore, the clinical implementation of DECT was aimed to be reached in this study.

Materials/Methods:

To define an optimal DECT protocol (Siemens Somatom Definition AS: 80/140kVp, kernel D34), CT scan settings were experimentally analyzed concerning beam hardening, image quality and influence on the heuristic conversion of CT numbers into stopping-power ratios (SPRs) per look-up table (HLUT) using phantoms consisting of tissues and tissue surrogates. Differences in range prediction and dose distribution between SECT and pseudo-monoenergetic CT datasets (MonoCT), derived by a weighted sum of both DECT scans, were quantified for phantoms and patients.

Results:

For treatment planning a DECT-based MonoCT of 79 keV is optimal, since CT-based HLUT uncertainties can be reduced (Figure 1). Dose distributions planned on SECT and MonoCT datasets reveal mean range deviations of 0.3mm, gamma passing rates (1%,1mm) greater than 99.9% and no clinically relevant changes in dose-volume histograms. Therefore, DECT was clinically implemented for patients treated with protons. 70 planning and 400 control DECT scans of overall 90 patients were acquired until January 2016.

Conclusions:

More precise range predictions and a wider diagnostic variety are feasible with DECT-based MonoCTs. Further improvements are expected from a direct, patient-specific, non-heuristic SPR determination. To quantify their possible benefits, first investigations of intra- and interpatient variations were performed on the still growing patient database.