⁶⁸Ga-PSMA PET/CT targeted biopsy for the diagnosis of clinically significant prostate cancer compared with transrectal ultrasound guided biopsy: a prospective randomized single-centre study

This study was approved by the Nanjing First Hospital, and all patients signed an informed consent. A total of 120 patients with elevated serum prostate-specific antigen (PSA) levels (> 4.0 ng/ml) were consecutively enrolled and randomized into PSMA-PET and TRUS groups. Each patient was assigned a random computer-generated number, and patients with odd numbers were assigned to the PSMA-PET group and patients with even numbers to the TRUS group. All patients in the PSMA-PET group (n = 60) underwent ⁶⁸Ga-PSMA-11 PET/CT. PSMA-TB was subsequently performed if PSMA PET/CT was highly suggestive of PCa, and systematic TRUS-GB was performed if PSMA PET/CT was negative. For patients with multiple intra-prostatic PSMA-avid lesions, the lesion with the highest uptake was selected as the puncture target. If PSMA-TB were negative, systematic TRUS-GB plus two cores of suspicious-lesion biopsies were performed within 2 days. All patients in the TRUS group (n = 60) underwent direct systematic TRUS-GB. Clinically significant PCa was defined as any Gleason score ≥ 7 (3 + 4) and a lesion diameter > 0.5 cm³, or T3/T4 clinical stage. The CONSORT diagram for this study is shown in Fig. 1.

The precursor PSMA-HBED (DKFZ-PSMA-11; GMP-compliant grade) was obtained from ABX advanced biochemical compounds GmbH (Germany). ⁶⁸Ga was obtained from a ⁶⁸Ge/⁶⁸Ga generator (ITM Company, Germany). ⁶⁸Ga-PSMA-11 was radiolabelled using an automated module (ITM). All products were prepared using good manufacturing practice and were non-pyrogenic and sterile. Radiochemical purity and stability were determined by analytical reverse phase high-performance liquid chromatography. The radiochemical purity of all products administered to patients for imaging was > 99%. PET/CT was performed (uMI780, United Imaging, China) 45–60 min after injection of 111–185 × 10⁶ MBq ⁶⁸Ga-PSMA-11 (3–5 mCi). CT images were used for attenuation correction and accurate localization. PET imaging was acquired immediately after CT scanning (matrix 256) with a 15.5 cm field of view, with 3 min acquisition for each bed position. Pelvic imaging in the prone position was specifically acquired for prostate TB. Reconstruction was conducted using an ordered subset expectation maximization algorithm with four iterations/eight subsets, and Gauss-filtered to an in-plane spatial resolution of 3 mm at full-width at half-maximum. PET/CT fusion was performed using a uMI 780 workstation. All lesions were displayed in three planes (transaxial, coronal and sagittal), and the region of interest (ROI) was delineated from the PET/CT fusion image.

Images were examined by one radiologist and one nuclear medicine physician who were blinded to the clinical characteristics and pathology. Visual and semi-quantitative analyses were used to detect the primary lesion and lymph node metastasis. Region of interest (ROI) was drawn around the primary prostatic lesion with 40% maximal standardized uptake values (SUVmax) cut-off in the 1 h postinjection fusion image, SUVmax of the prostatic primary lesions and metastasis were acquired from the ROI. In this study, lesions with abnormal focal uptake in the prostate gland, tracer activity higher than the surrounding background and a SUVmax higher than the cut-off value of 8.0 were defined as intra-prostatic primary lesions, and suspected with csPCa based on our previous data and other clinical studies [13, 20]. Considering the lymph node drainage pattern of prostate cancer, any pelvic lymph node with focal increased uptake was regarded as a metastatic lymph node. For csPCa detection, a PSMA-avid prostatic lesion and SUVmax ≥ 8.0 were considered necessary prerequisites for transgluteal PSMA-TB. Lesion size, position and number in the prostate bed were identified by a urologist, radiologist and nuclear medicine physician on PET/CT. The volume of each lesion was calculated using the equation: lesion volume = 1/2 × A × B2 (mm³) where A and B represent the long and short diameters of the lesion, respectively. For multiple positive lesions, the lesion with the highest SUVmax, which may present with the highest aggressiveness, was selected as the puncture target. Patients with lower uptake in the primary prostatic lesion (SUVmax < 8.0) underwent systematic TRUS-GB.

All patients signed consent for this novel targeted biopsy procedure. Enemas and prophylactic antibiotics were not required. Aspirin and other anticoagulants were withdrawn for 7 days prior to the procedure. Pelvic CT was performed in the prone position, and the optimal cognitive fusion of PET and CT images was conducted. Once the index lesion in the prostate was determined, the puncture path, angle and depth were pre-simulated on the CT image. Metal palisade labelling was used to locate the puncture point on the patient’s buttock, and a single-puncture percutaneous transgluteal approach was implemented with 1% lidocaine for local anaesthesia. Under CT guidance, an Argon Angiotech 17G biopsy trocar (Athens, TX, USA) was introduced through the gluteus maximus, the ischial anal fossa, and the anal elevator muscle and positioned cognitively at the target point (Fig. 2). Two to four biopsy specimens were taken in the target area using the co-axial needle technique. Tissue samples were analysed by a pathologist. A post-biopsy pelvic CT (5–10 min after puncture) was performed to exclude immediate complications such as active pelvic bleeding or rectal bleeding. The procedure was well-tolerated, with an overall procedure time of approximately 20 min.

Patients underwent a standard systematic TRUS-GB protocol. Briefly, oral ciprofloxacin was routinely administered for 3 days, and an enema was performed 1 h before the procedure. Patients were placed in the left lateral position and local anaesthesia was applied to the anus. Systematic prostate biopsy was performed using an 18G needle (Bard, Covington, GA, USA) under B ultrasound guidance (Hitachi Noblus, Tokyo, Japan). A total of 12 core aspirations were performed in four sagittal planes of the medial line of the prostate lobe on both sides and the lateral side of the peripheral belt.

Biopsy specimens and the tumour cell ratio of the whole tissue sample were analysed by a pathologist with 5 years’ experience in urinary pathology. Histopathology positive biopsy specimens were compared with gross tissue histopathology in patients who underwent radical prostatectomy. The final pathology for each patient was confirmed as follows: (1) gross tissue histopathology after radical prostatectomy; (2) biopsy histopathology if patients did not undergo radical prostatectomy; and (3) biopsy specimen histopathology of TRUS-GB plus two core suspicious lesions if patients were negative for PSMA-TB.

Statistical analysis was performed using SPSS, version 24.0 (Chicago, IL, USA). The normality of the data was tested by Kolmogorov–Smirnov analysis. Continuous variables with a normal distribution were presented as mean (± standard deviation, SD) and analysed by Student’s t tests. Continuous variables with a non-normal distribution were presented as median (interquartile range) and analysed by Mann–Whitney U tests. χ² or Fisher’s exact tests were used to compare the differences between categorical variables. A P value < 0.05 was considered statistically significant.

The clinical data are shown in Table 1. Age, prostate volume and PSA level were comparable between the PSMA-PET and TRUS groups.

The overall detection rates of PCa and csPCa were 37.5% (45/120) and 32.5% (39/120), respectively. PSMA-PET detected 43.3% (26/60) PCa and 40.0% (24/60) of csPCa, whilst TRUS detected 31.6% (19/60) PCa and 25.0% (15/60) of csPCa. There was no significant difference in detection rates for PCa (χ² = 1.74, P > 0.05) and csPCa (χ² = 3.08, P > 0.05).

Amongst 25 patients with PSMA-avid lesions (SUVmax 22.5 ± 14.72), 23 were true positive, of whom 21 were diagnosed by a single needle puncture (including two patients with negative TRUS-GB prior to this study). Two patients were initially negative on PSMA-TB but confirmed by supplementary TRUS-GB, and two patients who received both PSMA-TB and TRUS-GB followed by transurethral resection of the prostate gland were true negative, with a final diagnosis confirmed as benign disease. In the PSMA-PET group, the detection rates of PCa and csPCa were significantly higher in PSMA-PET-positive compared with PSMA-PET-negative patients (23/25, 92.0% vs 3/35, 8.6%, χ² = 41.34, P < 0.01 and 22/25, 88.0% vs 2/35, 5.7%, χ² = 42.68, P < 0.01, respectively) (Fig. 3, Table 2).

PSMA-TB detected PCa in 21/25 (84.0%) and csPCa in 20/25 (80.0%) patients with PSMA-avid lesions, whereas TRUS-GB detected PCa in 19/60 (31.6%) (χ² = 19.4, P < 0.01) and csPCa in 15/60 (25.0%) (χ² = 22.0, P < 0.01) TRUS control patients. In one patient with previous abdomino-perineal resection due to rectal cancer (without rectal access), PCa (Gleason 4 + 3) was finally confirmed by this novel puncture technique (Fig. 4).

Amongst patients with PSA 4.0–20.0 ng/ml, the detection rate of csPCa was significantly higher in the PSMA-PET (10/37, 27.02%) than that of TRUS group (3/34, 8.82%) (χ² = 3.93, P < 0.05). However, no significant difference existed in csPCa detection between PSMA-PET and TRUS in patients with PSA > 20.0 ng/ml (60.9%, 14/23 vs 46.2%, 12/26, χ² = 1.06, P > 0.05) (Table 3).

Amongst 25 patients who underwent PSMA-TB, 13 with PCa were confirmed by gross tissue histopathology after radical prostatectomy, and 10 patients with PCa confirmed by biopsy histopathology. Gleason score was underestimated by biopsy histopathology in three patients, and adjusted from 3 + 4 to 4 + 3 in one and from 4 + 3 to 4 + 4 in two patients after radical prostatectomy.

In the 21 of 25 cases of PCa diagnosed by single needle puncture, the proportions of cancer tissues were 100% in two, 90% in two, 80% in three, 70% in two, 50% in seven, and < 50% in five cases.

We assessed the relationship between histopathology and lesion size in the 25 patients with PSMA-avid lesions (volume, long diameter, short diameter and long/short diameter ratio). There was no significant differences in lesion size between patients with higher Gleason scores (≥ 7) and lower Gleason scores (< 7) (all P > 0.05) (Table 4).

Compared with systematic TRUS-GB, the needle did not pass via the rectum for PSMA-TB and no bowel preparation, including preoperative enema and pre-antibiotics, was therefore needed. The total time for PSMA-TB was thus only 20–30 min, and no postoperative antibiotics were administered during and after puncture. PSMA-TB was well-tolerated, with only one patient experiencing a few episodes of haematuria after the TB. However, haematuria occurred in 10 patients, urine retention in three patients and rectal infection in one patient after TRUS-GB.