Prostate biopsy remains the gold standard for histologic confirmation of prostate cancer diagnosis, with over 1 million biopsies performed yearly in the US and Europe and many requiring repeated biopsies over their lifetime.^1,2 The transrectal approach continues to be ubiquitous despite the virtual elimination of infection and sepsis without the need for antibiotic prophylaxis for the transperineal approach. There has been a slow adoption of the technique due to the perceived need for anesthesia due to lower pain tolerance and the lack of a training platform.³ The PREVENT trial demonstrated that transperineal biopsy (TPBx) can be completed in office-based settings with comparable cancer detection rates and low infection risk compared to the transrectal approach¹; therefore, there is a need for a training platform that reinforces correct transperineal technique and spacing of biopsies to optimize zonal sampling. 3D printing offers a unique tool for surgical education, allowing for customizable and patient-specific models. Combined with hydrogel molding, realistic simulators can be developed for surgical simulation, training, and education. Using our previously validated and published approach of 3D printing and hydrogel molding, we developed a nonbiohazardous training model with built-in metrics for real-time feedback during TPBx training.

Previously, we developed a high-fidelity simulator for transrectal ultrasound biopsy⁴ and modified this for the transperineal approach. Archival 3T MRI images were selected from a prospective database and segmented to incorporate essential anatomy: perineum, prostate, seminal vesicles, vas deferens, bladder, urethra, rectum, pubic bone, pelvic diaphragm, and ischiocavernosus muscles. Using the 10-sector template, the 3D computer-aided design was modified to include 4 different color-coded zones that provide immediate feedback after biopsy: transition zone, central zone, peripheral zone, and anterior zone (Figure 1, A and B). Tissue tensile strength was modeled after a series of cadaver mechanical tests (Instron Universal Testing System).⁵ Additional mechanical testing was completed comparing the model and cadaver using a novel needle force mechanical system.⁶ A hydrogel model was then created for simulations with realistic ultrasound echogenicity and the various anatomical components (Figure 1, C-E).

Six experts and 4 novices completed TPBx on the model targeting a 12-core template (Figure 2, A and B). Colored biopsy cores were collected and measured for accuracy, core length, and number of attempts. Novices were trained by an expert prior to collecting biopsy cores for assessment (Figure 2, C and D). Experts successfully biopsied the cores at a higher accuracy than novices (80% vs 67%, P < .001) and superior length (12.1 vs 7.9 mm, P < .001), but with no difference in the number of attempts (P = .14). The base posterior, lateral, and anterior cores demonstrated the greatest difference in core length among experts and novices (P < .05, .01, and .05, respectively; Figure 3). In addition, the experts assessed the model via survey using a 5-point Likert scale. The results concluded the model adequately replicated ultrasound appearance (82%), procedural realism (71%), and educational effectiveness (86%).

This high-fidelity hydrogel simulator provided a portable nonbiohazardous training platform for transperineal prostate biopsy. Participants rated the model highly for realism and educational effectiveness. Real-time feedback was provided to participants in the form of accuracy (core color), precision (correct color core percentage), and quality (length of core). This model can be modified to accommodate any preferred prostate template and adapted to variable prostate anatomy while blunting the learning curve to facilitate adoption. Our next step includes a multicenter validation study and including lesions for targeted biopsy to train in this technique and encourage further adoption of the transperineal approach for prostate biopsy.