Quantification of MRI visibility and artefacts at 3T of BioXmark® fiducial marker in a pancreas tissue mimicking phantom

Purpose: X-ray based position verification of the target volume in image-guided radiation therapy (IGRT) of patients with pancreatic ductal adenocarcinoma (PDAC) is currently performed on rigid fiducial markers that are implanted under endoscopic ultrasonography. A new biodegradable liquid fiducial marker has recently been introduced. To assess its potential use for magnetic resonance imaging (MRI) guided radiotherapy of PDAC, the MRI visibility and artefacts of this marker were quantified and compared against rigid gold markers.

Material and Methods: Different spherical volumes (10 µL, 25 µL, 50 µL and 100 µL) of BioXmark® (Nanovi Radiotherapy A/S) as well as four differently oriented Gold Anchor™ (Naslund Medical AB; Ø 0.28 mm, 1 cm and 2 cm length) and three VisiCoil™ (IBA Dosimetry; Ø 0.35 mm, 5 mm and 10 mm length) were implanted in a spherical gel phantom mimicking the proton spin relaxation properties of healthy pancreatic tissue at 3 Tesla. MR relaxometry was performed to quantify the size and magnitude of the decrease in T¬2* relaxation time and relative proton density ρ(H) as a measure of visibility, and to quantify the size and magnitude of the increase in magnetic field inhomogeneity ΔB_0 as a measure of signal artefacts. The phantom was scanned in a 3.0 T Philips Ingenuity TF PET/MR scanner with an 8-channel head coil. For T2*- and ΔB_0-mapping a spoiled 3D multi-echo gradient echo sequence (GRE) was performed. ρ(H)-mapping was based on the signal intensity at TE = 0 ms relative to ultrapure water extrapolated from the T2-decay curve deduced from a spin echo sequence with different echo times. The signal was corrected for its T1 decay and B1- field inhomogeneity, for which an inversion-recovery spin echo sequence with multiple inversion times and a gradient echo sequence with different repetition times was utilized respectively.

Results: The rigid markers showed a direct linear relationship between size of visibility and artefact. BioXmark® showed a tendency towards larger size of visibility at smaller artefacts. BioXmark® markers up to 100 µL created volumes of visibility comparable to the size of visibility of VisiCoil™ and Gold Anchor™ markers. The magnitude of visibility was the highest for BioXmark® from 25µL – 100µL showing no correlation with the magnitude of artefact. The rigid markers show a non-linear correlation between magnitude of visibility and artefact where Gold Anchor™ induce the strongest artefacts.

Conclusion: BioXmark® causes signal voids on MRI due to its low proton density without strongly affecting the magnetic field in the surrounding tissue. The latter was found to be the main effect leading to the visibility of the rigid markers especially in GRE sequences. Hence, especially when a low level of image distortion is required, MRI characteristics of the BioXmark® surpass those of rigid gold markers currently being used for IGRT of PDAC.