Background and purpose: Adaptive MR-guided radiotherapy (MRgRT) is an innovative approach for delivering stereotactic body radiotherapy (SBRT) in prostate cancer (PC). Despite the increased clinical use of SBRT for PC, there is limited data on the relation between the actual delivered dose and toxicity. We aimed to identify dose parameters based on the total accumulated delivered bladder dose (DOSEACCTX). Furthermore, for future personalization, we studied whether prospective accumulation of the first 3 of 5 fractions (DOSEACC3FR) could be used as a representative of DOSEACCTX. Materials and methods: We deployed a recently validated deformable image registration-based dose accumulation strategy to reconstruct DOSEACCTX and DOSEACC3FR in 101 PC patients treated with stereotactic MRgRT. IPSS scores at baseline, end of MRgRT, at 6 and 12 weeks after treatment were analyzed to identify a clinically relevant increase of acute urinary symptoms. A receiver operator characteristic curve analysis was used to investigate the correlation of an increase in IPSS and bladder DOSEACCTX (range V5–V36.25 Gy, D1cc, D5cc) and DOSEACC3FR (range V6–V21.8 Gy, D1cc, D5cc) parameters. Results: A clinically relevant increase in IPSS in the three months following MRgRT was observed in 25 patients. The V20Gy-32Gy from DOSEACCTX and V15Gy-18Gy from DOSEACC3FR showed good correlation with IPSS increase with area under the curve (AUC) values ranging from 0.71 to 0.75. In contrast, baseline dosimetry showed a poor correlation with AUC values between 0.53 and 0.62. Conclusion: DOSEACCTX was superior to baseline dosimetry in predicting acute urinary symptoms. Because DOSEACC3FR also showed good correlation, this can potentially be used to optimize MRgRT for the remaining fractions.