Part I | Chapter 7 144 Acute patient-reported urinary toxicity was observed in 30% of our cohort in the first three months following treatment. While these numbers seem a bit high compared to other reports, we used a different definition of ‘patient-reported acute urinary toxicity’ compared to IPSS+10 only.16,29,30 In our study, only 15.4% (20/130) reported an increase in IPSS of ³ 10 points. However, many patients started with alpha-blocking medication during or shortly after treatment, in case of (severe) irritative urinary complaints. Our pre-defined hypothesis was that the use of alpha-blocking medication might therefore mask an increase in IPSS during follow-up and thus dilute the correlation if not considered. As a sensitivity analysis (data not shown), we performed univariable logistic regression analysis for IPSS+10 as the only outcome. Although the trends of the OR were consistent with the presented results, this yielded no significant correlations. This may partly be attributed to a lack of power with only 20 cases. Nevertheless, significant correlations between dosimetry and IPSS+10 have been reported previously.16,24 This discrepancy might be caused by using different assessment time-points during follow-up; Bohoudi et al.16 measured IPSS also at the end of treatment and Henderson et al.24 measured IPSS two weeks after the end of treatment compared to our first measurement at 1 month post-treatment. This leads to a difference in the sensitivity, as it is likely that a larger proportion will report significant urinary toxicity shortly after treatment. This is also reflected in the differences in incidence of IPSS+10. Furthermore, differences in prescription of alpha-blocking medication could also attribute to this effect. Additionally, a sensitivity analysis was performed in those without alpha-blocking medication at baseline (n = 116, data not shown). This did not alter the results (OR) in a meaningful way, besides resulting in slightly higher p-values due to a smaller sample size. To our knowledge, only Bohoudi et al.16 investigated the relationship between the accumulated dose to the bladder and urinary toxicity in prostate cancer patients treated with adaptive MRIguided SBRT on a 0.35 T MR-Linac. Considering only the dose to the entire bladder, similar mediumhigh accumulated dose levels (V20-32Gy in cm3) were correlated with IPSS+10. No correlations were found for the pre-treatment plan. No relative dose parameters were assessed. Our results do not confirm the significant correlations found for absolute bladder dose parameters. This might have several explanations, besides the difference in primary outcome definition. First, as previously mentioned, the outcome was measured at different follow-up moments, leading to differences in toxicity rates. This could lead to differences in the sensitivity of the analyses that were performed. Finally, no multivariable analysis was performed, and therefore it is unclear if the observed correlations would persist when corrected for other (potentially) important clinical variables. Still, this study clearly indicates that the pre-treatment dose cannot necessarily be used as a substitute for the actual delivered dose. Therefore, the accumulated dose should be considered. Our study has some limitations. First, the exploratory aspects of the study should be interpreted with caution since no independent validation has yet been performed. For the bladder wall, however, the results were significant for almost the entire range of dose parameters, even after correction for baseline characteristics that are potentially related to the outcome, thus strongly suggesting the important role of bladder wall dose in the development of acute urinary toxicity. Although we did correct for several baseline characteristics, we cannot conclude that the found correlations are
RkJQdWJsaXNoZXIy MTk4NDMw