Background and purpose Currently, there is no consensus on dose prescription in image guided adaptive brachytherapy (IGABT) in locally advanced cervical cancer. The purpose of this study was to provide evidence based recommendations for tumor dose prescription based on results from a multi-center patient series (retroEMBRACE). Materials and methods This study analyzed 488 locally advanced cervical cancer patients treated with external beam radiotherapy ± chemotherapy combined with IGABT. Brachytherapy contouring and reporting was according to ICRU/GEC-ESTRO recommendations. The Cox Proportional Hazards model was applied to analyze the effect on local control of dose-volume metrics as well as overall treatment time (OTT), dose rate, chemotherapy, and tumor histology. Results With a median follow up of 46 months, 43 local failures were observed. Dose (D90) to the High Risk Clinical Target Volume (CTVHR) (p = 0.022, HR = 0.967 per Gy) was significant for local control, whereas increasing CTVHR volume (p = 0.004, HR = 1.017 per cm3), and longer OTT (p = 0.004, HR = 1.023 per day) were associated with worse local control. Histology (p = 0.084), chemotherapy (p = 0.49) and dose rate (p = 1.00) did not have significant impact on local control. Separate analyses according to stage of disease showed that dose to CTVHR, residual gross tumor volume (GTVres), and Intermediate Risk CTV (CTVIR) has significant impact on local control. Conclusion CTVHR dose of ⩾85 Gy (D90) delivered in 7 weeks provides 3-year local control rates of >94% in limited size CTVHR (20 cm3), >93% in intermediate size (30 cm3) and >86% in large size (70 cm3) CTVHR. CTVIR and GTVres dose of ⩾60 Gy and ⩾95 Gy (D98) leads to similar local control. A dose of 5 Gy (CTVHR) is required to compensate an increase of OTT by one week. Increased CTVHR volume by 10 cm3 requires additional 5 Gy for equivalent local control.