Session Item

The spatial confinement and longer treatment time to deliver online MR-guided adaptive radiotherapy on the MR-Linac (MRL), necessitates reliable and comfortable patient positioning. This work describes the steps taken to determine the optimal position and image sequence for accelerated partial breast irradiation (APBI), prior to clinical implementation, on a 1.5T Unity MRL (Elekta AB, Sweden).

A 2-phased approach was used. Phase 1 recruited 5 non-patient volunteers to an approved imaging study. Participants underwent 2 imaging sessions comparing 2 positioning approaches: supine, arms over head on an in-house manufactured thorax board, with and without an individualised (vac)uum-moulded-bag (Klarity Medical Products, USA). Participants completed a questionnaire using a 10-point numeric rating scale after each scan to assess comfort. Responses were analysed using descriptive statistics in Microsoft Excel. Qualitative responses adopted a simple thematic analysis. Image quality for all sequences, including vendor provided T1 and T2 weighted (W) images, were assessed using visual grading analysis (VGA) by 3 independent observers (2 Radiographers, 1 Oncologist). Results of phase 1 informed the positioning choice for phase 2. Sequences were also optimised.

Phase 2 recruited 5 patients to undergo 2 scans in the same position, with the questionnaire and VGA repeated. Reproducibility of positioning was evaluated by an automatic bone (sternum, ribs, anterior-lateral chest wall) match of the 2 scans in Monaco (v5.40.01, Elekta AB, Sweden).

Phase 1 participants, independent of vac-bag status, reported identical comfort scores at the start of the scan (27.5%). At the end of the scan, levels of discomfort were greater in the non vac-bag cohort than with the vac-bag (82.5% vs 52.5%). Paraesthesia in the arms, with pain in the shoulders was reported by 60% of participants without the vac-bag. VGA signified overall structure visibility was greater (very clear/clear) without the vac-bag compared to with (64.38% vs 56.48%). All participants preferred the vac-bag (Fig 1), which was selected for phase 2. Sequence parameters were modified to retain signal-to-noise ratio and spatial resolution (Table 1).

Phase 2 patients reported a higher comfort level than the volunteers with the vac-bag indicating good tolerability. Paraesthesia in the hands with no other pain was described by 40% of patients. VGA determined overall conspicuity was better on the T1W sequence than T2W (58.3% vs 45%). Reproducibility of positioning was satisfactory, with overall mean displacements of -0.29, 0.57 and 0.50cm in the right-left, superior-inferior and antero-posterior directions respectively.

The addition of the vac-bag provided participants with greater comfort and acceptable reproducibility, with little compromise to image quality. The tissue weighting selected for APBI daily online image registration is T1W. Work continues to investigate the dosimetric implication of this positioning.