To explore the role of preoperative MRI prostate shape in urinary incontinence after robot-assisted radical prostatectomy (RARP). Patients were stratified into four groups based on the mpMRI prostatic apex shape: Group A (prostatic apex overlapping the membranous urethra anteriorly and posteriorly), Group B and C (overlap of the prostatic apex of the anterior or posterior membranous urethra, respectively) and Group D (no overlap). Preoperative variables and intraoperative data were compared. Continence recovery was defined as no pad/day or 1 safety pad/day by an outpatient evaluation performed at 1, 3, 6, and 12 months after RARP. One hundred patients underwent RARP were classified as belonging to Group A (n = 30), Group B (n = 16), Group C (n = 14), and Group D (n = 40). Group D showed a significantly more favorable urinary continence recovery after RARP respect to all the other shapes presenting any forms of overlapping (HR = 1.9, 95% CI 1.2-3.1, p = 0.007). The estimated HR remained substantially unchanged after adjusting by age, body mass index, CCI, prostate volume, and bladder neck sparing (HR = 1.9, 95% CI 1.1-3.2, p = 0.016). The continence recovery median time was 9 months for Group A + B + C (95% CI 5-11) and 4 months for Group D (95% CI 2-6) (p = 0.023). Shape D showed a better continence recovery when compared to other shapes presenting any kind of overlapping of the prostatic apex over the membranous urethra.