PURPOSE: To compare dosimetric characteristics, clinical outcomes, and toxicity in patients with locally advanced cervical cancer(LACC) treated with transperineal interstitial brachytherapy(PISBT), using CT-based planning or MRI-based planning with a monoinstitutional MRI-compatible applicator, Template Benidorm(TB). MATERIALS AND METHODS: We retrospectively analyzed 106 women treated with P-ISBT for LACC from 2006 to 2022. P-ISBT was CT-based, with Martinez Universal Perineal Interstitial Template(MUPIT) between 2006 and 2012, or MRI-based planning (TB) since 2013. Volumetric, clinical, and dosimetric parameters were compared between applicators. RESULTS: Forty-six (43.4%) patients were treated with MUPIT, and 60 (56.6%) with TB. Median follow-up was 42 months (4-188). No significant difference was observed in 5-year local control (LC), regional control (RC), distant failure (DF), overall survival (OS), and disease-specific survival (DSS) between applicators. The TB cohort had fewer needles ( p = 0.001 ), halved median CTV volume ( p < 0.001 ), increased median D90%CTV dose ( p < 0.005 ), and lower D2cc values in rectum/bladder ( p < 0.001 ). G3 chronic toxicity was 15% in the MUPIT group and 6% in TB ( p = 0.15 ), with no G4 toxicity reported. Rectal hemorrhage was lower in the TB group ( p = 0.002 ). Larger CTV volumes correlated with higher rectal D2cc and rectal hemorrhage ( p = 0.001 ; p = 0.029 , respectively), and enterovaginal fistula ( p = 0.03; p < 0.001 , respectively). Median CTV volume was 152.82 cc (35.3-256.78) in patients with chronic rectal toxicity and 102.9 cc (21.4-329.41) in those without ( p = 0.001 ). CONCLUSION: MRI's superior CTV-volume definition results in smaller treatment volumes, lower D2ccfor rectum and bladder, and a trend towards higher CTV D-90% . Rectal hemorrhage was significantly lower in the MRI-based group. A significant correlation was observed between larger CTV-volumes, higher rectal D2cc and rectal hemorrhage/enterovaginal fistula. (c) 2024 American Brachytherapy Society. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.