Evaluation of target volume margins for radiotherapy of prostate implanted with fiducial markers

The planning target volume (PTV) depends on the method of radiotherapy guidance. This study aimed to measure the systemic and random errors using an online marker matching and offline bone structure matching to estimate PTV_marker, PTV_bone, or PTV_laser for treatment verification and radiotherapy guidance, especially in centers lacking radiotherapy fiducial markers (FMs). Thirty patients with localized prostate cancer who were treated with FM-based dose escalation protocol were included. The initial set-up was done with laser marks and daily megavoltage images were acquired. The systematic and random errors were calculated. PTV_marker, defined as the sum of maximum marker migration, and PTV calculated to compensate for the difference between online marker matching and offline analysis of marker matching. PTV_marker was added to estimated PTV from online marker matching to obtain PTV_laser. PTV_{skin marks migration}, was calculated and deducted from PTV_laser to acquire PTV_bone. The mean maximum marker migration was 2 ± 1.2 mm. The resultant values of PTV_marker were 2.7 ± 0.6 mm, 3.3 ± 1.1 mm, and 4.4 ± 2.2 mm, in the lateral (lat.), longitudinal (long) & vertical (vert.) directions, respectively, whereas values of PTV_laser were 13 ± 0.6 mm, 17.7 ± 1.1 mm, and 15.8 ± 2.2 mm, and PTV_bone were 5.9 ± 0.6 mm, 8.6 ± 1.1 mm, 7.2 ± 2.2 mm, respectively, in the lat., long., and vert. directions. Our results show that PTV needed with FM-based image guidance ranged between 3 and 4 mm in the three cardinal directions, was 10 mm smaller than that required with laser skin marks guidance, and narrower by 5 mm compared to that obtained by offline bone structure image matching.