- Email: [email protected]
Robotic Transperineal Prostate Biopsy: Pilot Clinical Study
Technology and Engineering Robotic Transperineal Prostate Biopsy: Pilot Clinical Study H. Ho, J. S. P. Yuen, P. Mohan, E. W. Lim, and C. W. S. Cheng OBJECTIVE
METHODS
RESULTS
CONCLUSION
To develope a robot (BioXbot) that performs mapping transperineal prostate biopsy (PB) with two perineal skin punctures under ultrasound guidance. Our pilot study’s clinical endpoints were complications and its technical endpoints were the duration for each phase. This institution review board–approved prospective clinical trial included patients with indications for PB. Two urologists performed these PBs. In the lithotomy position and under general anesthesia, the transrectal biplane ultrasound probe acquired transverse images of the prostate gland. The urologist defined its boundaries and planned the biopsy. It guided the PB in 3 axes, passing through a single perineal skin puncture for each prostate side. After each biopsy, it automatically moved to the next position. The steps were repeated on the contralateral side. Our 20 patients had a mean prostate-specific antigen of 8.4 ⫾ 4.9 ng/mL. Two patients had 2 previous biopsies, whereas the rest had one. The mean number of biopsies taken was 28.5 ⫾ 6.2 in a mean total procedure time of 32.5 ⫾ 3.2 minutes. We detected 3 patients with prostate cancer with Gleason score 3 ⫹ 3. Two patients required brief bladder catheterization after their biopsy. Their prostate volumes were ⬎50 mL and the number of biopsies taken was ⬎30 cores. There was no mechanical failure, sepsis, bleeding per-rectal, or perineal hematoma. This pilot study demonstrated BioXbot’s safety and feasibility as a biopsy platform. It can potentially be used for image-guided PB and focal therapy. UROLOGY 78: 1203–1208, 2011. © 2011 Elsevier Inc.
S
everal recent developments mold the scene of prostate cancer (PCa) management. Together with downward stage migration, there is also tumor volume reduction at diagnosis.1 PCa screening trials reveal the numbers of screened cases needed to bring benefit to 1 patient who underwent radical prostatectomy.2,3 Alternative forms of PCa management, such as active surveillance and focal therapy, have also emerged.4 They are largely based on limited clinical and biopsy information. Meanwhile, PCa detection with imaging is improving, and the promise of image-guided biopsy or treatment is far from being fulfilled.5 Prostate biopsy (PB) surpasses its traditional role of cancer diagnosis to include characterization and localization. Current decision-making on PCa management is dependent on its characteristics as predicted by PB. The flaws of transrectal PB are rectal bleeding and potentially life-threatening postbiopsy sepsis.6 Its inherent inability to sample the anterior and apical regions leads to undersampling and a false-negative rate of 30%.7 Accurate localization knowledge of the positive biopsy site can also Patent is held by Singhealth, Pte. Ltd. Funding Support: This research was supported in part by funding from the NMRC grant 0659/2004 and the Enterprise Challenge by the Prime Minister’s Office. From the Department of Urology, Singapore General Hospital, Singapore Reprint requests: Dr. Henry Ho, Department of Urology, Singapore General Hospital, Outram Road, Singapore 169608, Singapore. E-mail: [email protected]. Submitted: March 29, 2011, accepted (with revisions): July 9, 2011
© 2011 Elsevier Inc. All Rights Reserved
be lacking in the current variable clinical setting.8 Moreover, there is poor Gleason score and locality correlation between the PB and pathologic prostatectomy specimen.9 These inadequacies call for a new approach that approximates the truth of PCa by accurately sampling all parts of the prostate with minimal or no complication. To achieve this, we envisage a paradigm shift whereby transperineal PB replaces the transrectal approach. With these criteria in mind, we developed a novel robotic device (BioXbot) that performs transperineal PB. In this report, we present the results of its pilot clinical study.
MATERIAL AND METHODS We had previously described the BioXbot’s technical details, both the hardware and software design.10 In the following, we emphasized on its clinically relevant features. Bio-Xbot is a robotic system that guides transperineal prostate biopsy (Fig. 1A). The mobile cart has a powered lift that supports a movable platform with 6 degrees of rotational freedom. The robot positioning system (RPS), which sits on the movable platform, houses the controls for x- and y-axes, whereas the biopsy gun holder (Fig. 1C) controls the z axis. It calculates every biopsy point in the 3 axes and guides the biopsy needle to the predetermined position. Image acquisition is done via a video cable that connects the ultrasound machine output to digital image archiving in BioXbot. The urologist inputs via the touch-screen monitor and keyboard. 0090-4295/11/$36.00 doi:10.1016/j.urology.2011.07.1389
1203
Figure 1. (A) BioXbot G1 Cart. (B) Gantry and ultrasound probe holder. (C) Biopsy gun holder.11
We use the double dual-conical concept to ensure complete access to the prostate gland with the least skin puncture (Fig. 2B). Pivot points are angulation points on the perineal skin puncture through which the needle passes through to the various locations within the prostate. Regardless of the location and number of biopsies, all needle trajectories within 1 prostate lobe pass through 1 pivot point. As such, the entire prostate can be sampled with 2 skin punctures. Our software development incorporates clinical and safety requirements. Biopsy planning is not permitted outside the predefined prostate boundary. This prevents needle path outside the prostate, injuring the bladder or rectum. For improved apical sampling, each biopsy core always starts at the apex of the prostate. In a larger prostate, 2 cores are taken in the same trajectory, first at the apex and the other nearer the base, allowing better coverage of the whole gland.
Study Population This was a prospective clinical trial (http://ClinicalTrial.gov; registration identifier: NCT01102361) with ethical committee 1204
approval from our institution review board. Two urologists performed these PBs. We included patients with no contraindication for general anesthesia (GA) who had persistent indications for repeat PB, such as rising prostate-specific antigen (PSA) or abnormal initial biopsy histology (high-grade prostatic intra-epithelial neoplasia or atypical small acinar proliferation). They were mandated to read and understand the clinical trial’s patient information sheets and gave written informed consent before participation. They were also required to document the presence of fever, gross hematuria, or perineal hematoma for one week after PB. They were reviewed at 1 week and informed of their histology results.
Biopsy Workflow Patient Preparation. With the patient under GA and in the lithotomy position, the perineum is exposed and cleansed with iodine solution. Meanwhile, the device is initialized by running through its internal software checklist and full range of motion of the r.p.s. in 3 axes. UROLOGY 78 (5), 2011
Figure 2. (A) Single dual-cone coverage. (B) Double dual-cone coverage.11 Image Acquisition. Aloka ultrasound machine (SSD-1700, Aloka, Tokyo, Japan) with multifrequency biplane convex and linear transrectal transducer ultrasound probe (Aloka UST672-5/7.5) was used for this study. It was inserted into an ultrasound-compatible rectal sheath. The latter maintains the prostate’s position by supporting it when the ultrasound probe moves to acquire transverse images of the prostate. The transrectal ultrasound probe and sheath is held by the ultrasound probe holder, which moves it to any required position in the craniocaudal direction. BioXbot is adjusted until good-quality transverse images of prostate are acquired. The urologist defines the craniocaudal limits of scan along the length of the prostate gland, which are set beyond its base and apex. Next, the device automatically moves the probe holder between the defined limits to acquire the transverse ultrasound images of the prostate gland at 0.5 mm apart. These images are digitalized into the patient’s database. Prostate Modeling. From these transverse ultrasound images, the urologist selects 5-10 slices of transverse prostate images to define the prostate’s boundaries at various levels of the prostate, from the base to the apex, after which a 3-D model of the prostate is created (Fig. 3A, B).
Biopsy Planning Biopsy planning is based on the transverse view of the prostate gland. BioXbot contains software that automatically individualizes a biopsy plan for each patient (Fig. 3C, D). It is based on the prostate length and volume, with emphasis in the peripheral zone. The urologist can add more biopsy sites or modify the sampling locations depending on the clinician’s preferences. When the urologist is satisfied with biopsy plan and reviews the 3-D biopsy trajectories with their prostate coverage, it is finalized. Pivot Point Preparation. The sterile patient contact pads are inserted at the front of the gantry (Fig. 1B). They are in direct UROLOGY 78 (5), 2011
contact with the patient’s sterile perineum when the gantry is in the biopsy position. Next, the urologist inserts a sterile needle sheath (16G) (Bard needle guide C1610B, Bard, West Sussex, UK) through the ring guide of the gantry and one of the openings in the patient’s contact pad. When the gantry swings into biopsy position, the urologist makes a 5-mm stab incision where the needle sheath indents the perineal skin. They traverse the skin and dermis to avoid inadvertent creation of multiple skin punctures by the biopsy needle. Biopsy. The biopsy sequence begins with the first biopsy position being automatically calculated by BioXbot and positioned in the x-, y-, and z-axes. The urologist inserts the loaded needle biopsy gun (Bard Magnum biopsy gun) with a sterile biopsy needle (16G needle, Magnum MN1620, Bard) into the needle sheath. The depth of needle insertion for each biopsy site is determined by biopsy gun holder, which is also adjusted by BioXbot. The urologist can verify its position on the transverse view of real-time ultrasound images. The biopsy needle is seen as a dot with the echogenic shadow, and the entire trajectory is verified by a series of ultrasound images at various level of the prostate. He fires the biopsy gun and the specimen is retrieved. After each core, BioXbot automatically repositions itself to the next planned biopsy position and the process is repeated for subsequent biopsy. After the planned biopsies on one lobe are completed, the needle sheath shifts to the other opening in the patient’s contact pad. A second stab incision is made and the needle sheath goes into the new pivot point to biopsy the other prostate lobe. At completion, the rectal sheath with ultrasound probe and BioXbot are removed together. Pressure is applied to the 2 perineal puncture wounds for hemostasis before skin dressing is applied. Our clinical endpoints are complications arising from the procedures, and the technical endpoints are the duration for each phase of our procedure, as recorded by a single research 1205
Figure 3. Software screen shots. (A) 2-D images with boundaries drawn, (B) 3-D model creation, (C) target planning, and (D) 3-D model.11
assistant. Setup time starts from the insertion of the transrectal ultrasound probe, and BioXbot is adjusted until satisfactory ultrasound images are obtained. Ultrasound image acquisition time is the time taken for the urologist to define the limits of the length of prostate and complete the scan between the predefined limits to acquire the transverse prostate images. Three-dimensional prostate modeling time is the duration when the urologist defines the prostate boundary in the transverse view. Biopsy planning time includes the time taken for “autoplanning,” addition of new biopsy cores, and adjustment of the planned biopsy sites. It ends when the urologist finalizes and approves his plan. Biopsy time is time taken for the prostate biopsy to be completed, which is dependent on the total number of biopsies taken. In this feasibility trial, we use descriptive statistics (mean, median, range).
RESULTS The mean age of the 20 patients enrolled was 61.8 years (range 56-71). The mean PSA at the time of biopsy was 8.4 ⫾ 4.9 ng/mL (range 7.4-21.9). All except 2 patients had 1 previous negative biopsy. These 2 patients had 2 previous negative PBs. The mean 3-D prostate volume is 33.3 ⫾ 12.7 mL (range 30.0-66.6). The mean time from previous biopsy was 15.0 ⫾ 8.6 months (6-22 months). 1206
Table 1. Results of clinical study Technical Endpoints (time)
Mean ⫾ SD (min)
Range (min)
Device setup Ultrasound scanning Prostate modeling Biopsy planning Prostate biopsy Total procedure
4.85 ⫾ 1.84 2.7 ⫾ 0.6 4.1 ⫾ 2.1 2.8 ⫾ 1.3 24.2 ⫾ 6.9 32.5 ⫾ 3.2
2-8 2-4 2-8 2-6 17-43 25-48
None of the patients had histologic (high-grade prostatic intraepithelial neoplasia, atypical small acinar proliferation) indications for repeat biopsy. The mean number of biopsies taken was 28.5 ⫾ 6.2 (range 24-40). Our technical endpoints results are shown in Table 1. We detected 3 patients with PCa in this cohort and their histology was Gleason 3 ⫹ 3 adenocarcinoma. The number of positive cores ranged from 2-6 cores. The remaining patients had benign prostatic hyperplasia. There were no major complications such as sepsis or intractable hematuria. No bleeding per-rectal or perineal hematoma were recorded. There was a case of mild hematuria that resolved spontaneously after 1 day. Two UROLOGY 78 (5), 2011
patients needed an in-out bladder catheterization after their biopsy. Their prostate volumes were ⬎50 mL and number of biopsies taken was ⬎30 cores.
COMMENT Recent evolution of PCa management has brought about new demands on PB. It is an increasingly important decision-making tool for patients and urologists. Beyond the purpose of cancer detection, PB should provide information on accurate cancer characterization and localization. PCa surveillance necessitates frequent PB because such patients’ safety cannot be overlooked.11 Thus, a device that can deliver these requirements is needed. This is the first clinical trial report of an ultrasoundguided robotic transperineal PB device. Its unique feature is the “conical” approach fanning out caudal-cranially from the apex to the base of the prostate gland. As such, there will be 2 perineal skin punctures regardless of the number of biopsies (see Fig. 2). This differs to the brachytherapy template-based transperineal PB where multiple skin punctures are made.12,13 This preliminary clinical trial demonstrated clinical feasibility and safety of our robotic device in performing transperineal PB. Our mean procedure time was 32.5 minutes for a mean number of 28.5 biopsy cores. Longer operative time was noted in 4 patients with prostate volume of ⬎50 mL. These patients had a mean time of more than 30 minutes. Two patients required more than 35 cores biopsy. The other reason was our inability to obtain planned biopsy in the retropubic part of the prostate because of pubic arch interference. There were no mechanical failure, bleeding, or infection complications. We detected 3 cancers in this cohort of repeat PB patients. This is lower than other studies because the incidence of PCa is lower in Asia. Moreover, this feasibility study is too small to conclude on its cancer detection capability. Transperineal intervention to the prostate gland has distinct advantages over the transrectal approach. Sterility via perineal skin punctures is achievable, which can reduce the rate of postbiopsy sepsis.14 It also has improved coverage of the prostate gland and access to the anterior and apical parts of the prostate via the conical approach.15 These are regions in the prostate where cancer is usually found, especially in repeat PB patients, and they are not easily accessible from the transrectal route.16 This is the basis of our clinical trial: to include patients who need repeat PB. Its value may be extrapolated to patient selection for nonradical treatment options, such as active surveillance or focal therapy localization. Moreover, sterility will be of utmost importance in the potential setting of treatment delivery for PCa. Our safety advantage is enhanced with our ability to verify each needle trajectory with real-time ultrasound images during the biopsy. Freehand transperineal PB has a substantial learning curve, which impedes its widespread adoption. Its acUROLOGY 78 (5), 2011
curacy and the precision of each biopsy location can be variable.17,18 To ensure a systematic and thorough biopsy of the entire prostate gland, most transperineal PBs are based on the brachytherapy template with multiple perineal skin punctures.19 This can be overcome with BioXbot. Our transperineal approach distinguishes us from other ultrasound-based PB devices (TargetScan; Envisioneering Medical Technologies, St. Louis, MO).20,21 The latter uses the transrectal approach, replicating the same problems associated with the current practice. Other investigators have developed their devices based on magnetic resonance imaging (MRI) and are awaiting clinical trial.22 Their challenges are different, namely, MRI-compatible material and limited space in the MRI tunnel. As such, their cost will be high and limited by the availability of MRI. These factors will restrict widespread urological application outside of a clinical trial in tertiary centers. At this preliminary stage of development, we face several problems. Our procedure time is long compared with standard transrectal PB. This is caused by our evolving workflow, learning curve, and adherence to strict research protocols. GA is a clear disadvantage for this transperineal PB protocol. Under pilot study environment, we need optimal patient conditions to evaluate the device and refine the procedural workflow. However, many centers perform freehand transperineal PB under local anesthesia.23 Thus, we envisage that with future technical modifications and improved operative steps, this can be an office-based local anesthetic procedure. Another limitation was our inability to “reach” the retropubic part of a large prostate gland because of pubic arch interference. In 3 cases where prostate volumes were ⬎50 mL, we were not able to perform the planned biopsy in these parts, a problem also faced by practitioners of prostate brachytherapy seed implant.24 Current horizontal transperineal approach as mandated by the cart limited our access to them. One possible solution was the extreme lithotomy position to rotate the pelvis cranially. This would rotate the pubic bone anteriorly, exposing the anterior part of the prostate gland for better access. The other solution was platform angulation, such that the skin entry points were at an angle instead of being parallel to the ground. This preliminary clinical experience reinforces our pursuit for transperineal dual “conical” prostate intervention. Current disadvantages are related to learning curve and hardware design. The next version contains solutions to overcome the inaccessible parts of the prostate and a clinician-centered workflow. Study to compare our biopsy plan with a whole-mount radical prostatectomy specimen is in progress. Prospective randomized clinical trials that compare transrectal or brachytherapy gird template PB are necessary to validate our clinical findings, including cancer detection rate, complications, and postbiopsy pain score. In addition to accurate systematic PB, 1207
BioXbot can also be guided by ultrasound fusion software with any emerging imaging modality, such as MRI for targeted PB. It can also be a positioning device for treatment delivery of ablative energy in focal therapy for PCa.
CONCLUSIONS This first clinical report of robotic transperineal PB demonstrates its feasibility and safety. It serves as an important tool that answers the new demands on PB, beyond which it represents the beginning of a new era of accurate and targeted biopsy tool and a glimpse of the future focal PCa treatment delivery system. Acknowledgment. We acknowledge the late Associate Professor Ng Wan Sing and CIMIL, Nanyang Technological University. References 1. Caso JR, Mouraviev V, Tsivian M, et al. Prostate cancer: an evolving paradigm. J Endourol. 2010;24:805-809. 2. Schroder FH, Hugosson J, Roobol MJ, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360:320-328. 3. Andriole GL, Crawford ED, Grubb RL, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009; 360:1310-1319. 4. Klotz L. Active Surveillance with selective delayed intervention: using natural history to guide treatment in good risk prostate cancer. J Urol. 2004;172(5)(2):S48-S50. 5. Hambrock T, Somford DM, Hoeks C, et al. Magnetic resonance imaging guided prostate biopsy in men with repeat negative biopsies and increased prostate specific antigen. J Urol. 2010;183:520527. 6. Hadway P, Barrett LK, Waghorn DJ, et al. Urosepsis and bacteraemia caused by antibiotic-resistant organisms after transrectal ultrasonography-guided prostate biopsy. BJU Int. 2009;104:15561558. 7. Levine MA, Ittman M, Melamed J, et al. Two consecutive sets of transrectal ultrasound guided sextant biopsies of the prostate for the detection of prostate cancer. J Urol. 1998;159:471-476. 8. Schulte RT, Wood DP, Daignault S, et al. Utility of extended pattern prostate biopsies for tumor localization: Pathologic correlations after radical prostatectomy. Cancer. 2008;113:1559-1565. 9. Gofrit ON, Zorn KC, Taxy JB, et al. Predicting the risk of patients with biopsy Gleason score 6 to harbor a higher grade cancer. J Urol. 2007;178:1925-1928.
1208
10. Ho H, Mohan P, Lim EW, et al. Robotic ultrasound-guided prostate intervention device: system description and results from phantom studies. Int J Med Robot Comput Assist Surg. 2009;5(1):51-58. 11. Loblaw A, Zhang L, Lam A, et al. Comparing prostate specific antigen triggers for intervention in men with stable prostate cancer on active surveillance. J Urol. 2010;184:1942-1946. 12. Merrick GS, Taubenslag W, Andreini H, et al. The morbidity of transperineal template-guided prostate mapping biopsy. BJU Int. 2008;101(12):1524-1529. 13. Merrick GS, Gutman S, Andreini H, et al. Prostate cancer distribution in patients diagnosed by transperineal template-guided saturation biopsy. Eur Urol. 2007;52:715-723. 14. Hara R, Jo Y, Fujii T, et al. Optimal approach for prostate cancer detection as initial biopsy: Prospective randomized study comparing transperineal versus transrectal systematic 12-core biopsy. J Urol. 2008;71:191-195. 15. Mohan P, Ho H, Yuen J, et al. A 3D computer simulation to study the efficacy versus transrectal biopsy of the prostate. Int J Comput Assist Radiol Surg. 2007;1:351-360. 16. Taira AV, Merrick GS, Galbreath RW, et al. Performance of transperineal template-guided mapping biopsy in detecting prostate cancer in the initial and repeat biopsy setting. Prostate Cancer Prostatic Dis. 2010;13:71-77. 17. Emiliozzi P, Longhi S, Scarpone LP, et al. The value of a single biopsy with 12 transperineal cores for detecting prostate cancer in patients with elevated prostate specific antigen. J Urol. 2001;166: 845-850. 18. Emiliozzi P, Corsetti A, Tassi B, et al. Best approach for prostate cancer detection: a prospective study on transperineal versus transrectal six-core prostate biopsy. Urology. 2003;61:961-966. 19. Onik G, Miessau M, Bostwick DG. Three-dimensional prostate mapping biopsy has a potentially significant impact on prostate cancer management. J Clin Oncol. 2009;27:4321-4326. 20. Andriole GL, Bullock TL, Belani JS, et al. Is there a better way to biopsy the prostate? Prospects for a novel transrectal systematic biopsy approach. Urology. 2007;70:22-26. 21. Megwalu II, Ferguson GG, Wei JT, et al. Evaluation of a novel precision template-guided biopsy system for detecting prostate cancer. BJU Int. 2008;102:546-550. 22. Van den Bosch MR, Moman MR, van Vulpen M, et al. MRIguided robotic system for transperineal prostate interventions: Proof of principle. Phys Med Biol. 2010:55:N133-N140. 23. Kubo Y, Kawakami S, Numao N, et al. Simple and effective local anaesthesia for transperineal extended prostate biopsy: application to three-dimensional 26-core biopsy. Int J Urol. 2009;16: 240-243. 24. Gibbons EP, Smith RP, Beriwal S, et al. Overcoming pubic arch interference with free-hand needle placement in men undergoing prostate brachytherapy. Brachytherapy. 2009;8(1):74-78.
UROLOGY 78 (5), 2011
METHODS
RESULTS
CONCLUSION
To develope a robot (BioXbot) that performs mapping transperineal prostate biopsy (PB) with two perineal skin punctures under ultrasound guidance. Our pilot study’s clinical endpoints were complications and its technical endpoints were the duration for each phase. This institution review board–approved prospective clinical trial included patients with indications for PB. Two urologists performed these PBs. In the lithotomy position and under general anesthesia, the transrectal biplane ultrasound probe acquired transverse images of the prostate gland. The urologist defined its boundaries and planned the biopsy. It guided the PB in 3 axes, passing through a single perineal skin puncture for each prostate side. After each biopsy, it automatically moved to the next position. The steps were repeated on the contralateral side. Our 20 patients had a mean prostate-specific antigen of 8.4 ⫾ 4.9 ng/mL. Two patients had 2 previous biopsies, whereas the rest had one. The mean number of biopsies taken was 28.5 ⫾ 6.2 in a mean total procedure time of 32.5 ⫾ 3.2 minutes. We detected 3 patients with prostate cancer with Gleason score 3 ⫹ 3. Two patients required brief bladder catheterization after their biopsy. Their prostate volumes were ⬎50 mL and the number of biopsies taken was ⬎30 cores. There was no mechanical failure, sepsis, bleeding per-rectal, or perineal hematoma. This pilot study demonstrated BioXbot’s safety and feasibility as a biopsy platform. It can potentially be used for image-guided PB and focal therapy. UROLOGY 78: 1203–1208, 2011. © 2011 Elsevier Inc.
S
everal recent developments mold the scene of prostate cancer (PCa) management. Together with downward stage migration, there is also tumor volume reduction at diagnosis.1 PCa screening trials reveal the numbers of screened cases needed to bring benefit to 1 patient who underwent radical prostatectomy.2,3 Alternative forms of PCa management, such as active surveillance and focal therapy, have also emerged.4 They are largely based on limited clinical and biopsy information. Meanwhile, PCa detection with imaging is improving, and the promise of image-guided biopsy or treatment is far from being fulfilled.5 Prostate biopsy (PB) surpasses its traditional role of cancer diagnosis to include characterization and localization. Current decision-making on PCa management is dependent on its characteristics as predicted by PB. The flaws of transrectal PB are rectal bleeding and potentially life-threatening postbiopsy sepsis.6 Its inherent inability to sample the anterior and apical regions leads to undersampling and a false-negative rate of 30%.7 Accurate localization knowledge of the positive biopsy site can also Patent is held by Singhealth, Pte. Ltd. Funding Support: This research was supported in part by funding from the NMRC grant 0659/2004 and the Enterprise Challenge by the Prime Minister’s Office. From the Department of Urology, Singapore General Hospital, Singapore Reprint requests: Dr. Henry Ho, Department of Urology, Singapore General Hospital, Outram Road, Singapore 169608, Singapore. E-mail: [email protected]. Submitted: March 29, 2011, accepted (with revisions): July 9, 2011
© 2011 Elsevier Inc. All Rights Reserved
be lacking in the current variable clinical setting.8 Moreover, there is poor Gleason score and locality correlation between the PB and pathologic prostatectomy specimen.9 These inadequacies call for a new approach that approximates the truth of PCa by accurately sampling all parts of the prostate with minimal or no complication. To achieve this, we envisage a paradigm shift whereby transperineal PB replaces the transrectal approach. With these criteria in mind, we developed a novel robotic device (BioXbot) that performs transperineal PB. In this report, we present the results of its pilot clinical study.
MATERIAL AND METHODS We had previously described the BioXbot’s technical details, both the hardware and software design.10 In the following, we emphasized on its clinically relevant features. Bio-Xbot is a robotic system that guides transperineal prostate biopsy (Fig. 1A). The mobile cart has a powered lift that supports a movable platform with 6 degrees of rotational freedom. The robot positioning system (RPS), which sits on the movable platform, houses the controls for x- and y-axes, whereas the biopsy gun holder (Fig. 1C) controls the z axis. It calculates every biopsy point in the 3 axes and guides the biopsy needle to the predetermined position. Image acquisition is done via a video cable that connects the ultrasound machine output to digital image archiving in BioXbot. The urologist inputs via the touch-screen monitor and keyboard. 0090-4295/11/$36.00 doi:10.1016/j.urology.2011.07.1389
1203
Figure 1. (A) BioXbot G1 Cart. (B) Gantry and ultrasound probe holder. (C) Biopsy gun holder.11
We use the double dual-conical concept to ensure complete access to the prostate gland with the least skin puncture (Fig. 2B). Pivot points are angulation points on the perineal skin puncture through which the needle passes through to the various locations within the prostate. Regardless of the location and number of biopsies, all needle trajectories within 1 prostate lobe pass through 1 pivot point. As such, the entire prostate can be sampled with 2 skin punctures. Our software development incorporates clinical and safety requirements. Biopsy planning is not permitted outside the predefined prostate boundary. This prevents needle path outside the prostate, injuring the bladder or rectum. For improved apical sampling, each biopsy core always starts at the apex of the prostate. In a larger prostate, 2 cores are taken in the same trajectory, first at the apex and the other nearer the base, allowing better coverage of the whole gland.
Study Population This was a prospective clinical trial (http://ClinicalTrial.gov; registration identifier: NCT01102361) with ethical committee 1204
approval from our institution review board. Two urologists performed these PBs. We included patients with no contraindication for general anesthesia (GA) who had persistent indications for repeat PB, such as rising prostate-specific antigen (PSA) or abnormal initial biopsy histology (high-grade prostatic intra-epithelial neoplasia or atypical small acinar proliferation). They were mandated to read and understand the clinical trial’s patient information sheets and gave written informed consent before participation. They were also required to document the presence of fever, gross hematuria, or perineal hematoma for one week after PB. They were reviewed at 1 week and informed of their histology results.
Biopsy Workflow Patient Preparation. With the patient under GA and in the lithotomy position, the perineum is exposed and cleansed with iodine solution. Meanwhile, the device is initialized by running through its internal software checklist and full range of motion of the r.p.s. in 3 axes. UROLOGY 78 (5), 2011
Figure 2. (A) Single dual-cone coverage. (B) Double dual-cone coverage.11 Image Acquisition. Aloka ultrasound machine (SSD-1700, Aloka, Tokyo, Japan) with multifrequency biplane convex and linear transrectal transducer ultrasound probe (Aloka UST672-5/7.5) was used for this study. It was inserted into an ultrasound-compatible rectal sheath. The latter maintains the prostate’s position by supporting it when the ultrasound probe moves to acquire transverse images of the prostate. The transrectal ultrasound probe and sheath is held by the ultrasound probe holder, which moves it to any required position in the craniocaudal direction. BioXbot is adjusted until good-quality transverse images of prostate are acquired. The urologist defines the craniocaudal limits of scan along the length of the prostate gland, which are set beyond its base and apex. Next, the device automatically moves the probe holder between the defined limits to acquire the transverse ultrasound images of the prostate gland at 0.5 mm apart. These images are digitalized into the patient’s database. Prostate Modeling. From these transverse ultrasound images, the urologist selects 5-10 slices of transverse prostate images to define the prostate’s boundaries at various levels of the prostate, from the base to the apex, after which a 3-D model of the prostate is created (Fig. 3A, B).
Biopsy Planning Biopsy planning is based on the transverse view of the prostate gland. BioXbot contains software that automatically individualizes a biopsy plan for each patient (Fig. 3C, D). It is based on the prostate length and volume, with emphasis in the peripheral zone. The urologist can add more biopsy sites or modify the sampling locations depending on the clinician’s preferences. When the urologist is satisfied with biopsy plan and reviews the 3-D biopsy trajectories with their prostate coverage, it is finalized. Pivot Point Preparation. The sterile patient contact pads are inserted at the front of the gantry (Fig. 1B). They are in direct UROLOGY 78 (5), 2011
contact with the patient’s sterile perineum when the gantry is in the biopsy position. Next, the urologist inserts a sterile needle sheath (16G) (Bard needle guide C1610B, Bard, West Sussex, UK) through the ring guide of the gantry and one of the openings in the patient’s contact pad. When the gantry swings into biopsy position, the urologist makes a 5-mm stab incision where the needle sheath indents the perineal skin. They traverse the skin and dermis to avoid inadvertent creation of multiple skin punctures by the biopsy needle. Biopsy. The biopsy sequence begins with the first biopsy position being automatically calculated by BioXbot and positioned in the x-, y-, and z-axes. The urologist inserts the loaded needle biopsy gun (Bard Magnum biopsy gun) with a sterile biopsy needle (16G needle, Magnum MN1620, Bard) into the needle sheath. The depth of needle insertion for each biopsy site is determined by biopsy gun holder, which is also adjusted by BioXbot. The urologist can verify its position on the transverse view of real-time ultrasound images. The biopsy needle is seen as a dot with the echogenic shadow, and the entire trajectory is verified by a series of ultrasound images at various level of the prostate. He fires the biopsy gun and the specimen is retrieved. After each core, BioXbot automatically repositions itself to the next planned biopsy position and the process is repeated for subsequent biopsy. After the planned biopsies on one lobe are completed, the needle sheath shifts to the other opening in the patient’s contact pad. A second stab incision is made and the needle sheath goes into the new pivot point to biopsy the other prostate lobe. At completion, the rectal sheath with ultrasound probe and BioXbot are removed together. Pressure is applied to the 2 perineal puncture wounds for hemostasis before skin dressing is applied. Our clinical endpoints are complications arising from the procedures, and the technical endpoints are the duration for each phase of our procedure, as recorded by a single research 1205
Figure 3. Software screen shots. (A) 2-D images with boundaries drawn, (B) 3-D model creation, (C) target planning, and (D) 3-D model.11
assistant. Setup time starts from the insertion of the transrectal ultrasound probe, and BioXbot is adjusted until satisfactory ultrasound images are obtained. Ultrasound image acquisition time is the time taken for the urologist to define the limits of the length of prostate and complete the scan between the predefined limits to acquire the transverse prostate images. Three-dimensional prostate modeling time is the duration when the urologist defines the prostate boundary in the transverse view. Biopsy planning time includes the time taken for “autoplanning,” addition of new biopsy cores, and adjustment of the planned biopsy sites. It ends when the urologist finalizes and approves his plan. Biopsy time is time taken for the prostate biopsy to be completed, which is dependent on the total number of biopsies taken. In this feasibility trial, we use descriptive statistics (mean, median, range).
RESULTS The mean age of the 20 patients enrolled was 61.8 years (range 56-71). The mean PSA at the time of biopsy was 8.4 ⫾ 4.9 ng/mL (range 7.4-21.9). All except 2 patients had 1 previous negative biopsy. These 2 patients had 2 previous negative PBs. The mean 3-D prostate volume is 33.3 ⫾ 12.7 mL (range 30.0-66.6). The mean time from previous biopsy was 15.0 ⫾ 8.6 months (6-22 months). 1206
Table 1. Results of clinical study Technical Endpoints (time)
Mean ⫾ SD (min)
Range (min)
Device setup Ultrasound scanning Prostate modeling Biopsy planning Prostate biopsy Total procedure
4.85 ⫾ 1.84 2.7 ⫾ 0.6 4.1 ⫾ 2.1 2.8 ⫾ 1.3 24.2 ⫾ 6.9 32.5 ⫾ 3.2
2-8 2-4 2-8 2-6 17-43 25-48
None of the patients had histologic (high-grade prostatic intraepithelial neoplasia, atypical small acinar proliferation) indications for repeat biopsy. The mean number of biopsies taken was 28.5 ⫾ 6.2 (range 24-40). Our technical endpoints results are shown in Table 1. We detected 3 patients with PCa in this cohort and their histology was Gleason 3 ⫹ 3 adenocarcinoma. The number of positive cores ranged from 2-6 cores. The remaining patients had benign prostatic hyperplasia. There were no major complications such as sepsis or intractable hematuria. No bleeding per-rectal or perineal hematoma were recorded. There was a case of mild hematuria that resolved spontaneously after 1 day. Two UROLOGY 78 (5), 2011
patients needed an in-out bladder catheterization after their biopsy. Their prostate volumes were ⬎50 mL and number of biopsies taken was ⬎30 cores.
COMMENT Recent evolution of PCa management has brought about new demands on PB. It is an increasingly important decision-making tool for patients and urologists. Beyond the purpose of cancer detection, PB should provide information on accurate cancer characterization and localization. PCa surveillance necessitates frequent PB because such patients’ safety cannot be overlooked.11 Thus, a device that can deliver these requirements is needed. This is the first clinical trial report of an ultrasoundguided robotic transperineal PB device. Its unique feature is the “conical” approach fanning out caudal-cranially from the apex to the base of the prostate gland. As such, there will be 2 perineal skin punctures regardless of the number of biopsies (see Fig. 2). This differs to the brachytherapy template-based transperineal PB where multiple skin punctures are made.12,13 This preliminary clinical trial demonstrated clinical feasibility and safety of our robotic device in performing transperineal PB. Our mean procedure time was 32.5 minutes for a mean number of 28.5 biopsy cores. Longer operative time was noted in 4 patients with prostate volume of ⬎50 mL. These patients had a mean time of more than 30 minutes. Two patients required more than 35 cores biopsy. The other reason was our inability to obtain planned biopsy in the retropubic part of the prostate because of pubic arch interference. There were no mechanical failure, bleeding, or infection complications. We detected 3 cancers in this cohort of repeat PB patients. This is lower than other studies because the incidence of PCa is lower in Asia. Moreover, this feasibility study is too small to conclude on its cancer detection capability. Transperineal intervention to the prostate gland has distinct advantages over the transrectal approach. Sterility via perineal skin punctures is achievable, which can reduce the rate of postbiopsy sepsis.14 It also has improved coverage of the prostate gland and access to the anterior and apical parts of the prostate via the conical approach.15 These are regions in the prostate where cancer is usually found, especially in repeat PB patients, and they are not easily accessible from the transrectal route.16 This is the basis of our clinical trial: to include patients who need repeat PB. Its value may be extrapolated to patient selection for nonradical treatment options, such as active surveillance or focal therapy localization. Moreover, sterility will be of utmost importance in the potential setting of treatment delivery for PCa. Our safety advantage is enhanced with our ability to verify each needle trajectory with real-time ultrasound images during the biopsy. Freehand transperineal PB has a substantial learning curve, which impedes its widespread adoption. Its acUROLOGY 78 (5), 2011
curacy and the precision of each biopsy location can be variable.17,18 To ensure a systematic and thorough biopsy of the entire prostate gland, most transperineal PBs are based on the brachytherapy template with multiple perineal skin punctures.19 This can be overcome with BioXbot. Our transperineal approach distinguishes us from other ultrasound-based PB devices (TargetScan; Envisioneering Medical Technologies, St. Louis, MO).20,21 The latter uses the transrectal approach, replicating the same problems associated with the current practice. Other investigators have developed their devices based on magnetic resonance imaging (MRI) and are awaiting clinical trial.22 Their challenges are different, namely, MRI-compatible material and limited space in the MRI tunnel. As such, their cost will be high and limited by the availability of MRI. These factors will restrict widespread urological application outside of a clinical trial in tertiary centers. At this preliminary stage of development, we face several problems. Our procedure time is long compared with standard transrectal PB. This is caused by our evolving workflow, learning curve, and adherence to strict research protocols. GA is a clear disadvantage for this transperineal PB protocol. Under pilot study environment, we need optimal patient conditions to evaluate the device and refine the procedural workflow. However, many centers perform freehand transperineal PB under local anesthesia.23 Thus, we envisage that with future technical modifications and improved operative steps, this can be an office-based local anesthetic procedure. Another limitation was our inability to “reach” the retropubic part of a large prostate gland because of pubic arch interference. In 3 cases where prostate volumes were ⬎50 mL, we were not able to perform the planned biopsy in these parts, a problem also faced by practitioners of prostate brachytherapy seed implant.24 Current horizontal transperineal approach as mandated by the cart limited our access to them. One possible solution was the extreme lithotomy position to rotate the pelvis cranially. This would rotate the pubic bone anteriorly, exposing the anterior part of the prostate gland for better access. The other solution was platform angulation, such that the skin entry points were at an angle instead of being parallel to the ground. This preliminary clinical experience reinforces our pursuit for transperineal dual “conical” prostate intervention. Current disadvantages are related to learning curve and hardware design. The next version contains solutions to overcome the inaccessible parts of the prostate and a clinician-centered workflow. Study to compare our biopsy plan with a whole-mount radical prostatectomy specimen is in progress. Prospective randomized clinical trials that compare transrectal or brachytherapy gird template PB are necessary to validate our clinical findings, including cancer detection rate, complications, and postbiopsy pain score. In addition to accurate systematic PB, 1207
BioXbot can also be guided by ultrasound fusion software with any emerging imaging modality, such as MRI for targeted PB. It can also be a positioning device for treatment delivery of ablative energy in focal therapy for PCa.
CONCLUSIONS This first clinical report of robotic transperineal PB demonstrates its feasibility and safety. It serves as an important tool that answers the new demands on PB, beyond which it represents the beginning of a new era of accurate and targeted biopsy tool and a glimpse of the future focal PCa treatment delivery system. Acknowledgment. We acknowledge the late Associate Professor Ng Wan Sing and CIMIL, Nanyang Technological University. References 1. Caso JR, Mouraviev V, Tsivian M, et al. Prostate cancer: an evolving paradigm. J Endourol. 2010;24:805-809. 2. Schroder FH, Hugosson J, Roobol MJ, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360:320-328. 3. Andriole GL, Crawford ED, Grubb RL, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009; 360:1310-1319. 4. Klotz L. Active Surveillance with selective delayed intervention: using natural history to guide treatment in good risk prostate cancer. J Urol. 2004;172(5)(2):S48-S50. 5. Hambrock T, Somford DM, Hoeks C, et al. Magnetic resonance imaging guided prostate biopsy in men with repeat negative biopsies and increased prostate specific antigen. J Urol. 2010;183:520527. 6. Hadway P, Barrett LK, Waghorn DJ, et al. Urosepsis and bacteraemia caused by antibiotic-resistant organisms after transrectal ultrasonography-guided prostate biopsy. BJU Int. 2009;104:15561558. 7. Levine MA, Ittman M, Melamed J, et al. Two consecutive sets of transrectal ultrasound guided sextant biopsies of the prostate for the detection of prostate cancer. J Urol. 1998;159:471-476. 8. Schulte RT, Wood DP, Daignault S, et al. Utility of extended pattern prostate biopsies for tumor localization: Pathologic correlations after radical prostatectomy. Cancer. 2008;113:1559-1565. 9. Gofrit ON, Zorn KC, Taxy JB, et al. Predicting the risk of patients with biopsy Gleason score 6 to harbor a higher grade cancer. J Urol. 2007;178:1925-1928.
1208
10. Ho H, Mohan P, Lim EW, et al. Robotic ultrasound-guided prostate intervention device: system description and results from phantom studies. Int J Med Robot Comput Assist Surg. 2009;5(1):51-58. 11. Loblaw A, Zhang L, Lam A, et al. Comparing prostate specific antigen triggers for intervention in men with stable prostate cancer on active surveillance. J Urol. 2010;184:1942-1946. 12. Merrick GS, Taubenslag W, Andreini H, et al. The morbidity of transperineal template-guided prostate mapping biopsy. BJU Int. 2008;101(12):1524-1529. 13. Merrick GS, Gutman S, Andreini H, et al. Prostate cancer distribution in patients diagnosed by transperineal template-guided saturation biopsy. Eur Urol. 2007;52:715-723. 14. Hara R, Jo Y, Fujii T, et al. Optimal approach for prostate cancer detection as initial biopsy: Prospective randomized study comparing transperineal versus transrectal systematic 12-core biopsy. J Urol. 2008;71:191-195. 15. Mohan P, Ho H, Yuen J, et al. A 3D computer simulation to study the efficacy versus transrectal biopsy of the prostate. Int J Comput Assist Radiol Surg. 2007;1:351-360. 16. Taira AV, Merrick GS, Galbreath RW, et al. Performance of transperineal template-guided mapping biopsy in detecting prostate cancer in the initial and repeat biopsy setting. Prostate Cancer Prostatic Dis. 2010;13:71-77. 17. Emiliozzi P, Longhi S, Scarpone LP, et al. The value of a single biopsy with 12 transperineal cores for detecting prostate cancer in patients with elevated prostate specific antigen. J Urol. 2001;166: 845-850. 18. Emiliozzi P, Corsetti A, Tassi B, et al. Best approach for prostate cancer detection: a prospective study on transperineal versus transrectal six-core prostate biopsy. Urology. 2003;61:961-966. 19. Onik G, Miessau M, Bostwick DG. Three-dimensional prostate mapping biopsy has a potentially significant impact on prostate cancer management. J Clin Oncol. 2009;27:4321-4326. 20. Andriole GL, Bullock TL, Belani JS, et al. Is there a better way to biopsy the prostate? Prospects for a novel transrectal systematic biopsy approach. Urology. 2007;70:22-26. 21. Megwalu II, Ferguson GG, Wei JT, et al. Evaluation of a novel precision template-guided biopsy system for detecting prostate cancer. BJU Int. 2008;102:546-550. 22. Van den Bosch MR, Moman MR, van Vulpen M, et al. MRIguided robotic system for transperineal prostate interventions: Proof of principle. Phys Med Biol. 2010:55:N133-N140. 23. Kubo Y, Kawakami S, Numao N, et al. Simple and effective local anaesthesia for transperineal extended prostate biopsy: application to three-dimensional 26-core biopsy. Int J Urol. 2009;16: 240-243. 24. Gibbons EP, Smith RP, Beriwal S, et al. Overcoming pubic arch interference with free-hand needle placement in men undergoing prostate brachytherapy. Brachytherapy. 2009;8(1):74-78.
UROLOGY 78 (5), 2011