Three-dimensional Printed Model of Prostate Anatomy and Targeted Biopsy-proven Index Tumor to Facilitate Nerve-sparing Prostatectomy

Precise knowledge of the three-dimensional (3D) location of the index cancer (lesion with highest Gleason score and/or largest volume) and its proximity to the prostate capsule and neurovascular bundle (NVB) has implications for both nerve-sparing robotic radical prostatectomy (NS-RRP) and focal therapy [1–3]. We report proof of the concept for use of customized, patient-specific printed 3D models of the prostate gland and index cancer lesion to aid in prostate cancer surgery.

Five patients with clinically localized prostate cancer were included. The median patient age was 69 yr (range 62–73), median prostate-specific antigen was 7.9 ng/ml (range 4.5–14.7), and clinical stage was cT1c (n = 4) or cT2 (n = 1). Prebiopsy magnetic resonance imaging (MRI) was performed using a 3-T body coil (3-mm steps), followed by 3D MRI-transrectal ultrasound image-fusion targeted biopsy using a Urostation system (Koelis, Paris, France) [4,5]. The index lesion in all five prebiopsy MRI scans was the MRIvisible dominant lesion, rated as Prostate Imaging Reporting and Data System (PI-RADS) 4 or 5 with a high probability of microscopic extracapsular extension (ECE) according to the MRI-determined contact length [5], a new quantitative MRI parameter to predict microscopic ECE, defined as the circumferential length of a cancer lesion in contact with the prostate capsule on MRI. We determined that all five lesions had the highest Gleason score or greatest cancer core length in each case. Using our prototype computer software, the 3D volume data from MRI results were transferred for manual segmentation of (1) the entire prostate gland, (2) the biopsy-proven index lesion, and (3) bilateral NVBs [1]. Using these digitized, surface-rendered images, we collaborated with a commercial 3D printer manufacturer (Fasotec, Chiba, Japan) to create life-size 3D printed models to demonstrate three key anatomic aspects (Fig. 1): a translucent prostate gland (light orange), a surface-rendered index lesion (red), and bilateral NVBs (yellow).

Fig. 1 –

Index tumor (PI-RADS 4) in contact with the posterior lateral capsule, outlined in red, on (A) T2-weighted magnetic resonance imaging (MRI) and (B) apparent diffusion coefficient map for a 73-yr-old man with prostate-specific antigen of 8.3 ng/ml and estimated prostate volume of 26ml. MRI transrectal ultrasound fusion targeted biopsy revealed Gleason 3 + 4 disease with 90% cancer core involvement. (C) Radical prostatectomy pathology revealed negative surgical margins with pT3a and Gleason 4 + 3 = 7 prostate cancer. (D,E) Printed 3D model showing a translucent prostate gland (light orange resin), surface-rendered index cancer (red resin), and neurovascular bundles (yellow). Since the location, size, and extent of the index cancer and its proximity to the prostate capsule were well recognized from the 3D model, the surgeon could focus on avoiding any capsulotomy and took a wider dissection of periprostatic tissue around the location suggested by the model. The wider periprostatic dissection around the cancer showed very accurate concordance with the real cancer location in direct contact with the capsule on pathology. (F) Real-size printed 3D models for all five cases.

The translucence of the 3D whole-gland prostate models allows easy visualization of the 3D location, size, and extent of the index lesion inside the prostate. The life-size reality of the 3D printed models facilitates understanding of the distance or proximity of the index cancer to the prostate capsule and NVB. The surgeon also found it useful to inspect the 3D models before and during surgery as a reference tool. Such detailed preoperative knowledge of three key aspects of prostate and cancer anatomy provided a novel opportunity to enhance the intraoperative precision and confidence of the robotic surgeon.

Because the MRI-visible index lesions had a high probability of ECE, the surgical recommendation was to dissect a slightly wider (1 mm) area of periprostatic tissue at that precise site to achieve negative margins during NS-RRP. Pathology examination of step-sectioned prostatectomy specimens revealed negative surgical margins in all these challenging high-risk cases (pT2c [n = 1], pT3a [n = 2], and pT3b [n = 2]). Accurate concordance between the 3D printed model and the histologic location of the index lesion/ECE was noted, resulting in negative margins.

Limitations of this report include the additional cost of creating 3D printed prostate models (approx. US$500 per case). Further studies with greater number of patients are necessary to evaluate the cost-effectiveness of this approach.