Session Item

Fiducial markers can assist in target localization during cone beam CT (CBCT) guided radiotherapy. The purpose of this study is to assess technical feasibility of fiducial marker implantation for use in pre-operative gastric cancer radiotherapy.

Fiducial markers were placed endoscopically in the stomach of gastric cancer patients (MaagART-01 trial, No. NL7036) prior to radiation treatment (45Gy in 25 fractions; daily CBCT imaging). Each procedure was performed under mild or deep sedation by one of four gastroenterologists. All of the currently 12 included patients received gold markers (Visicoil, Core Oncology, CA, USA; Ø 0.35mm x 10mm length), which were individually backloaded into a 22-gauge endoscopic needle prior to each placement. The last 5 patients also received liquid markers (BioXmark, Nanovi, Kongens Lyngby, Denmark; injection volume 0.08−0.20 mL). The liquid marker was loaded once per patient in a 25-gauge injection needle, allowing placement of multiple subsequent markers without retraction of the needle.

We evaluated duration of implantation (from first loaded needle entering the endoscope to final fiducial placed) and occurrence of complications (e.g. bleeding, perforation, fever). For each marker, we determined whether (a) its implantation was assessed as successful at time of implantation (i.e., marker secured in tissue), (b) it was present on the first scan post-implantation (pre-treatment CT or first CBCT (on reconstructed CBCT or projection images)) and (c) it was present on the last fraction’s CBCT. Finally, we assessed adequacy of location, i.e. in or near the stomach wall.

In total, we attempted to implant 79 markers (5−8 per patient; 61 Visicoil and 18 BioXmark; Fig 1). Duration of implantation was 16−38 min (mean=24 min), implantation time per marker was 2.5−5.4 min (mean=3.7 min). There were no procedure-related complications. At time of implantation, 68 markers were assessed to be implanted successfully (Fig 2). When markers were assessed as not successful at time of implantation (all Visicoil), this was mostly due to either technical difficulties (e.g. unable to push marker out of needle), or incomplete implantation caused by peristaltic motion.

On the first scan post-implantation, 65 markers were present. Of those, 59 were still present on the last fraction’s CBCT, 4 disappeared prior to radiation treatment (within 1−10 days post-implantation), and 2 disappeared during treatment (15 and 22 days post-implantation; fractions 5 and 8). Four markers (all Visicoil) were not placed correctly: one in diaphragm, one in spleen, and two in fatty tissue >1 cm from the stomach. For the markers that remained full treatment, no migration was observed.

Fiducial marker placement in the stomach was technically feasible and successful, for both Visicoil (38 of 61 attempts, 62%) and BioXmark (17 of 18, 94%). Visibility and benefit of fiducial markers in image-guided radiotherapy (using 4DCT and 4DCBCT) will be evaluated in an ongoing study.