Evaluation of ThinkQA (v2.0.1.11) as an online secondary dose check for MR guided radiation therapy with the Elekta Unity MR-Linac.

To evaluate ThinkQA (TQA), a collapsed cone convolution-based secondary dose check, as an alternative to MU2net (Clarkson-based, point-dose) for online adaptive planning on the Elekta Unity 1.5 T MR-Linac at Townsville University Hospital. Commissioning followed MPPG 5.b tests. The reference-dose agreement, magnetic-field modelling, directional dependence, output factors, off-axis points, heterogeneous slab geometries and calculation properties were assessed. Nine step-and-shoot IMRT plans (courtesy of Elekta) and 226 retrospectively analysed adapted fractions (prostate and pelvic nodes; planning target volumes 1.9-170.0 cm3) were compared between TQA and Monaco by gamma analysis (global 10.0% threshold; 2.0%/2.0 mm, 3.0%/2.0 mm). At 10.0 cm depth under TQA reference conditions, the mean absolute point-dose difference versus Monaco was 0.4%. TQA reproduced models the magnetic-field-induced cross-plane asymmetry with close agreement to Monaco. Directional dependence differences were ≤  ± 1.2% except when traversing the couch (± 1.8%). Output factors agreed within ≤ 1.0% (SSD 133.5 cm) and ≤ 2.0% (SSD 138.5 cm). In 226 clinical fractions, 3.0%/2.0 mm (global) yielded 93.0% passes in the high-dose region and 100.0% in other regions; 2.0%/2.0 mm yielded 25% high-dose passes. TQA results were available within about 1 min post Monaco export. TQA provides accurate, rapid, volumetric secondary dose verification for Unity, improves agreement with Monaco, and reduces console time by eliminating dose point re-selection. A 3.0%/2.0 mm global gamma criterion is a clinical acceptance level, with tighter criteria reserved for targeted investigations.