We investigated which timing of cumulative dose measurement is most strongly associated with acute urinary toxicity in magnetic resonance (MR)-guided linear accelerator (Linac) treatment and also explored additional contributing factors. A total of 114 patients with prostate cancer treated with five-fraction MR-Linac therapy were included. Cumulative bladder doses were calculated using deformable image registration at three time points: before treatment (pre-MRI), during position verification (PV-MRI), and after irradiation (post-MRI). Acute urinary toxicity was defined as an increase of ≥ 10 points in the International Prostate Symptom Score within 3 months from baseline. The relationships between cumulative bladder dose and acute urinary toxicity were analyzed, along with non-dosimetric contributing factors. Cumulative bladder doses calculated from PV- and post-MRI showed stronger associations with acute urinary toxicity than those from pre-MRI. The area under the curve values for bladder V37Gy (cc) and V37Gy (%) on post-MRI were 0.69 and 0.70. A higher frequency of motion correction applied for target displacement during PV-MRI was associated with a lower incidence of acute urinary toxicity (p = 0.04). Secondary adapt-to-shape-based recontouring and replanning shortened the MRI-to-irradiation interval, stabilized prostate positioning during irradiation, and enabled adaptation to an enlarged bladder, which may have reduced bladder dose. Therefore, cumulative bladder doses assessed on PV- and post-MRI are moderately associated with acute urinary toxicity. PV-MRI-based target motion correction and replanning may mitigate this toxicity; however, the correlation between motion correction and post-MRI bladder V37Gy (%) was weak despite statistical significance and should be interpreted with caution.
Tanaka et al. (Mon,) studied this question.