Multiparametric magnetic resonance imaging/transrectal ultrasound fusion prostate biopsy with semi-robotic navigation in the Chinese population: initial results

Abstract

Dear Editor, The current standard 12-core systematic transrectal ultrasound (TRUS)-guided biopsy in prostate cancer diagnosis has the disadvantages of overdetecting indolent tumors, while failing to identify clinically significant cancers in up to 35%.1 Advances in imaging techniques, notably the multiparametric magnetic resonance imaging (mpMRI), allow clinically significant prostate cancers to be detected with more precise localization.2 Fusion of MRI and real-time TRUS enables targeted biopsy of suspicious lesions, overcoming human error, and inability of biopsy site tracking in cognitive fusion.3 Multiple Western series have suggested a higher detection rate, higher specificity, and better risk stratification with fusion biopsy.45678 Study on the Chinese population, however, is sparse. We report the initial results of the first prospective study on mpMRI/US fusion-guided prostate biopsy in a Chinese population using the Artemis™ (Eigen, Grass Valley, CA, USA) platform, an elastic fusion device dedicated for prostate biopsy. Its distinctive features include elastic registration of MRI and TRUS images, probe stabilization with a semi-robotic navigation arm, and patient movement compensation, allowing accurate 3D modeling and lesion tracking.45 The ethics approval was obtained and a matched comparison analysis was performed with the systematic 12-core biopsy method. From July 2015 to April 2016, all men with elevated serum prostate-specific antigen (PSA) level of 4–20 ng ml−1 were counseled for prebiopsy mpMRI of the prostate or conventional systematic TRUS biopsy. MRI prostate was performed with 1.5T MRI scanner (MAGNETOM Aera: Siemens Healthcare, Erlangen, Germany), using a combination of 18-channel phase array body coil and 32-channel spine array coil. All mpMRI were independently reviewed by a panel of three experienced genitourinary radiologists, with suspicious lesions defined according to the Prostate Imaging – Reporting and Data System (PI-RADS) version 2 published in 2015.9 There were five PI-RADS assessment categories (1–5) correlating with probability of having significant cancer. The software Profuse (Eigen, Grass Valley, CA, USA) was used to outline the contour of the prostate and any PI-RADS 2–5 lesions. The Digital Imaging and Communication in Medicine data sets were then imported into the Artemis workstation for fusion with real-time TRUS images (Figure 1). TRUS with power Doppler was performed using an F37 with semi-Compound Pulse Wave Generator (Hitachi Aloka Medical, Tokyo, Japan) equipped with a UST-676P 3.0–7.5 MHz end-fire transrectal probe, under regional periprostatic block with 1% lignocaine (10 ml). Patients with any PI-RADS 2–5 lesions underwent a targeted biopsy (1–3 cores from each lesion depending on lesion size) and template 12-core biopsy (an extended sextant template from the lateral and medial aspects of the base, mid, and apical prostate bilaterally). Clinically significant cancers were defined as pathologic Gleason grade ≥3 + 4, or 3 + 3 with a maximum cancer core length more than 4 mm. All biopsy slides were reviewed by two genitourinary histopathologists of over 10 years’ experience.Figure 1: (a) The Artemis elastic fusion platform. (b) An example shows a PIRADS 3 lesion at the left peripheral base and a PIRADS 4 lesion at the right peripheral base of the prostate in a fusion biopsy patient. (c) During the biopsy procedure, the ultrasound probe is docked to the robot-like mechanical arm with encoders (angle-sensing devices) within each joint for tracking of probe position. After fusion of MR images with real-time TRUS images, target lesions will show up as areas of interests in the three-dimensional prostate model. MR: magnetic resonance imaging; TRUS: transrectal ultrasound.A total of seventy patients underwent prebiopsy mpMRI in the period. Two non-Chinese ethnic patients were excluded. Among the 68 eligible patients, 57 (83.8%) had positive MRI findings with a total of 98 PI-RADS 2–5 l