Title: Evaluation of Gold Fiducial Marker Migration After Implantation in Prostate Image Guided Radiation Therapy (IGRT)
Abstract: Gold fiducial markers are commonly placed prior to delivery of EBRT to the prostate for prostate cancer. Studies indicate that there is little migration over the course of therapy, but it is unknown whether there is clinically significant migration of the markers immediately after placement. Because of this uncertainty, it is common for RT planning studies to be done several days after marker placement, resulting in delay in therapy and added costs. We sought to evaluate the degree of seed migration between the day of marker placement (Day 0) and the CT planning study (Day 7). Fifty patients with transperineal (TP, n = 29) or transrectal (TR, n = 21) placement of 3mm-long 0.9mm-diameter gold fiducials underwent 3mm slice thickness CT scans on Day 0 within 15 minutes after placement and on Day 7. In the Treatment Planning System, points-of-interest were placed on the geometrical center of the implanted markers in both scans to form fiducial pairs, which were used by an automated, point-based algorithm to register the two images. The resulting mean distance between fiducial pairs (d) was recorded to assess the degree of seed migration. Next, by using the fiducial pairs as a frame of reference, prostate volumes from Day 0 and Day 7 were compared. The same attending radiation oncologist confirmed prostate contours on both scans, and the Day 0 prostates were uniformly expanded by 1,3 and 5mm. The percentage volumes of the Day 7 prostate that were not covered by these expansions were computed to assess whether prostate contours, if done at Day 0, adequately cover the prostate at Day 7. The average d for all patients was 0.806±0.5mm, with TP patients (0.921±0.5mm) having a higher (p<0.05) d than TR patients (0.647±0.4mm). 88% (44/50) of patients had d < 1.5mm; of the 6 men who did not, 5 had TP placement (p = 0.08). 50% (25/50) of patients had a history of ADT use, but this as well as other factors, such as prostate volume and closest distance of each seed to the prostate capsule, were not shown to be associated with a difference in d. 1mm expansions of Day 0 prostate contours covered >97% of the volume of Day 7 prostate contours in all patients (median 99.69%, IQR 99.0-99.8%). Furthermore, all 3mm expansions covered >99% of Day 7 prostate volumes, and all 5mm expansions covered 100%. The minimal changes in the relative positions of the seeds (average d < 1.0mm) indicated that there is unlikely any clinically significant migration of the intraprostatic gold markers placed with either TP or TR techniques. Additionally, the Day 0 prostate contours adequately covered the Day 7 prostates with clinically reasonable margins. These results suggest that the CT images obtained on the day of fiducial implantation could be reliably used for treatment planning.