Dose perturbations from implanted helical gold markers in proton therapy of prostate cancer

Abstract

Implanted gold fiducial markers are widely used in radiation therapy to improve targeting accuracy. Recent investigations have revealed that metallic fiducial markers can cause severe perturbations in dose distributions for proton therapy, suggesting smaller markers should be considered. The objective of this study was to estimate the dosimetric impact of small gold markers in patients receiving proton therapy for prostate cancer. Small, medium, and large helical wire markers with lengths of 10 mm and helix diameters of 0.35 mm, 0.75 mm, and 1.15 mm, respectively, were implanted in an anthropomorphic phantom. Radiographic visibility was confirmed using a kilovoltage x-ray imaging system, and dose perturbations were predicted from Monte Carlo simulations and confirmed by measurements. Monte Carlo simulations indicated that size of dose perturbation depended on marker size, orientation, and distance from the beam's end of range. Specifically, the perturbation of proton dose for the lateral-opposed-pair treatment technique was 31% for large markers and 23% for medium markers in a typical oblique orientation. Results for perpendicular and parallel orientations were respectively lower and higher. Consequently, these markers are not well suited for use in patients receiving proton therapy for prostate cancer. Dose perturbation was not observed for the small markers, but these markers were deemed too fragile for transrectal implantation in the prostate.

Figure 1

Three representations of a helical gold marker. Kilovoltage radiograph of the medium marker in the presence of bone (anterior) and soft tissue (posterior) in an anthropomorphic phantom. The marker was imaged with the large aperture $(18\text{cm}\times 18\text{cm})$ and the following technique: 110 kVp, 320 mA, 200 ms or 64 mAs. Optical density map with accompanying color scale for the large marker in the distal position. The unperturbed dose was approximately 490 cGy with a standard deviation of film response estimated at 10.474 cGy (i.e. about 2%). The dose behind each coil was approximately 460 cGy, with dose enhancement up to 590 cGy. Kilovoltage radiograph with the medium marker inserted between femoral heads in a RANDO phantom. The marker was imaged with the small aperture $(11\text{cm}\times 11\text{cm})$ and the following technique: 115 kVp, 500 mA, 400 ms or 200 mAs.

Figure 2

Plot of relative absorbed dose, D, and a function of distance along the beam axes from parallel opposed fields with the large and medium markers. The center of the prostate and the center of modulation (COM) were located at isocenter $(\mathrm{z}=0\text{cm})$. The markers were located at $\mathrm{z}=4\text{cm}$. For clarity, the curves were offset by adding multiples of 0.2. The dose shadow of each profile is compared to the area at the COM of the control (i.e. the area around $\mathrm{z}=0\text{cm}$).