Session Item

Adaptive proton therapy aims to deliver the radiation dose accurately in the presence of anatomical changes by practicing a CT imaging feedback loop on a regular basis to systematically monitor anatomical changes and adapt the treatment plan if required. However, the series of these repeated CT scans results in an accumulated additional patient dose, in particular, if 4DCT is performed to account for breathing effects. We hypothesized that the signal-to-noise ratio provided by conventional 4DCT protocols is higher than needed for proton therapy dose calculation. In this study, we aim to assess the effect of reducing CT dose on proton dose calculation in esophageal cancer.

A standard-dose 4DCT scan [Siemens SOMATOM Drive, 120 kVp, Q.ref mAs = 50 mAs, 10 breathing phases] for an esophageal cancer patient was used to investigate the effect of CT dose reduction on proton dose calculation. Low-dose CTs (LDCTs) with gradual reduction in dose were simulated using our in-house LDCT simulator. Several phantom studies previously confirmed the accuracy of the simulated images in providing realistic LDCT images.

An intensity modulated proton therapy plan with two posterior beams (150˚ and 180˚) was generated in RayStation (Version 9B) with the average image of the standard-dose 4DCT as planning CT. The prescribed dose of 50.40 Gy RBE (28 fractions of 1.80 Gy RBE) was optimized on the iCTV using robustness settings of 7mm setup error and 2.6% range error. To avoid additional variation due to the statistical uncertainty of the Monte Carlo simulation, a pencil beam engine was used for plan optimization and final dose calculation. The dose distributions were then recalculated for the LDCTs using the optimized plan and the results were compared with that of the standard-dose scan. 

The dose distributions on standard-dose CT and LDCTs were compared using the Dose-Volume Histogram (DVH) and gamma index tests. Figure 1 compares the dose distributions using the DVH curves of organs at risk and the target volume. The DVH curves are on top of each other, indicating the similarity of the dose distributions. Table 1 shows the gamma passing rate of the dose distributions acquired from LDCTs. It can be seen that the gamma passing rate decreases by reducing the CT dose, however, it is always higher than 99.50% for a 1%/1mm criterion, confirming the similarity of the dose distributions calculated on the standard-dose and LDCTs.

Our results on patient data suggested the possibility of reducing the CT dose for the purpose of dose calculation during the course of treatment. An aggressive dose reduction by a factor of up to 10 did not have a significant effect on the dose distributions. A quantitative analysis utilizing more patient data is ongoing.