Comparison of ⁶⁸Ga-PSMA-I/T PET-CT and Multiparametric MRI for Locoregional Staging of Prostate Cancer Patients: A Pilot Study

Prostate cancer (PCa) is the most frequent type of malignancy in men and has the second highest mortality rate among male malignant carcinomas. Radical prostatectomy (RP) is one of the commonly used first-line treatment modalities for patients with localized PCa [1]. Current European guidelines recommend MRI for local staging and CT/bone scan to achieve accurate staging for those with distant metastases before treatment of PCa [2]. However, these conventional imaging modalities have low sensitivity and specificity to identify small lesions. Moreover, misdiagnosis of benign lesions may lead to incorrect treatment or overtreatment in non-metastatic PCa patients. Novel imaging modalities, such as ⁶⁸Ga-DOTAGA-(3-iodo-y)fk(Sub-KuE) (⁶⁸Ga-PSMA-I/T PET-CT) and multiparametric magnetic resonance imaging (mp-MRI; including diffusion-weighted imaging [DWI] and dynamic contrast-enhanced T1-weighted imaging), have been developed for the treatment planning of PCa patients [2].

MRI plays an essential role in detecting suspicious lesions for PCa, guiding the biopsy procedure, and local staging in biopsy-proven PCa patients. Recent advances in mp-MRI have led to the detection of clinically significant disease and reduced unnecessary biopsies and treatment [3‒7]. mp-MRI, including T1- and T2-weighted DWI and T1-weighted dynamic contrast-enhanced sequences, is used to increase the diagnostic accuracy of the MRI for PCa [8‒13]. This modality is also widely used for active surveillance and guiding the repeat biopsies in these patients [14].

Prostate-specific membrane antigen (PSMA) is a type II transmembrane glycoprotein that is expressed in normal prostate gland epithelium and overexpressed on the surface of >80% of the primary and metastatic PCa [15, 16]. ⁶⁸Ga-PSMA 11 binds to the receptor of transmembrane folate hydrolase and can be imaged using positron emission tomography-computed tomography (PET-CT) scan [17, 18]. Recently, ⁶⁸Ga-PSMA-I/T PET-CT has been used in PCa detection, diagnosis, and staging [19, 20]. However, no study has so far compared the diagnostic performance of mp-MRI and ⁶⁸Ga-PSMA-I/T PET-CT. Therefore, we aimed to compare the diagnostic performance of ⁶⁸Ga-PSMA-I/T PET-CT and mp-MRI in PCa, using histopathological findings as the gold standard.

The study protocol was approved by Pamukkale University School of Medicine Human Research Ethics Committee (approval No. 60116787-020/56122) and the study was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. Patients who agreed to participate in the study and planned to undergo laparoscopic RP were included and patients underwent both mp-MRI and ⁶⁸Ga-PSMA-I/T PET-CT preoperatively. After imaging, patients underwent laparoscopic RP (LRP) within 3 days by a single urologist with 5 years of experience. Individuals with a history of treated PCa and a concurrent diagnosis of any other primary malignancy were excluded from the study. Additionally, patients who had more than the 6-week time interval between imaging to LRP and PSMA uptake related to PCa within the distant metastases were excluded. In all patients the TNM (tumor, node, and metastasis) system was used for PCa staging [21]. Patients with confirmed PCa were stratified by the D’Amico risk stratification [22].

mp-MRI was performed using a 1.5-T superconducting magnet (Ingenia, Philips Medical Systems, The Netherlands) and a 32-channel torso coil. Transverse, coronal, and sagittal T2-weighted turbo spin-echo images were acquired, respectively (3-mm slice thickness without gap, 2,290 ms time-to-repetition, 100 ms time-to-echo, number of signals acquired 2). Axial DWI were obtained using spin-echo echo-planar imaging sequences and two b-values (3-mm slice thickness without gap, 4,000 ms time-to-repetition, 90 ms time-to-echo, number of signals acquired 12, and b-values were 0 and 1,800 s/mm²). Fat-suppressed T1-weighted isovolumetric images after intravenous gadolinium-based contrast medium (0.1 mmol/kg, gadoterate meglumine, Dotarem, Guerbet, France) injection were obtained for dynamic contrast-enhanced images (6-mm slice thickness, intersection gap 3 mm, 3.9 ms time-to-repetition, 1.83 ms time-to-echo, number of signals acquired 1, 7-s dynamic scan time). Apparent diffusion coefficient maps were generated from DWI images using a dedicated software (Extended Workspace, Philips Healthcare, The Netherlands). MR images were evaluated by a urogenital radiologist with 22 years of experience (Observer 1) who was unaware of the clinical features, pathology, and PET-CT results of the patients. The Prostate Imaging – Reporting and Data System (PI-RADS) version 2 was used to calculate mp-MRI scores [23].

The ⁶⁸Ga-PSMA-I/T was carried out using a Scintomics synthesis unit (Lindach, Fürstenfeldbruck, Germany) by a qualified radiochemist under the recommended laboratory conditions. Quality control of the ⁶⁸Ga-PSMA-I/T radionuclide was performed by high-performance liquid chromatography and thin layer chromatography, validated according to previously described methods [19, 20], after a single ⁶⁸Ga-PSMA-I/T injection (mean 185 MBq; range 125–317 MBq) followed by a 60-min uptake period. Then, image acquisitions were carried out while the patient was lying in the supine position. PET-CT images were obtained using a PET-CT unit (Gemini TF TOF PET-CT; Philips, Cleveland, OH, USA). No intravenous contrast was used during emission scans, and image acquisition was from the skull vertex to knee. The transmission images were obtained using a low-dose CT protocol (50–120 mA, 90–140 kV, 5-mm slice thickness). For the PET images, attenuation correction was performed using CT features and the ordered subsets-expectation maximization algorithm (33 subsets, three iterations). The iterative method was used to reconstruct the PET images.

Two observers (Observer 2 and Observer 3) with 12 years of experience in PET-CT imaging, and who were unaware to the clinical features, pathology, and PET-CT results of the patients, evaluated the images, reaching a consensus. Image analysis was performed qualitatively and semiquantitatively. For prostate gland lesions to be considered positive when focal uptake was performed on the background prostatic uptake, values of maximum standardized uptake (SUV_max) normalized to body weight were used to assess the ⁶⁸Ga-PSMA-I/T uptake in the primary tumor focus. The SUV_max value in the prostate was noted at the site of the most intense uptake in the prostate gland. Seminal vesicle invasion (SVI), extracapsular extension (ECE), lymph node metastasis (LNM), and bladder neck invasion (BNI) were evaluated visually in each patient.

All LRP operations were performed by a urologist who had 5 years of experience in LRP. An extraperitoneal antegrade laparoscopic approach was performed as previously described with some modifications [24].

After abdominal access and entering of the Retzius space, the endopelvic fascia was opened and the dorsal vein complex reached. After ligation of the dorsal vein complex, incision of the bladder neck was performed and on reaching the vas deferens dissection was completed. Controlling the prostatic pedicles, neurovascular bundle dissection was performed towards the urethra. Afterwards, the urethra was cut and by removing the prostate from the area urethrovesical anastomosis was performed. Pelvic lymphadenectomy was also performed in those who had an intermediate or high risk of PCa.

The pathological evaluation of each case was reviewed by a uropathologist with 10 years of experience in uropathology. The evaluation of prostate histopathology was made according to the 2014 International Society of Urological Pathology (ISUP) [25]. ECE, SVI, BNI, LNM, and the number of positive lymph nodes were recorded.

The diagnostic performance of mp-MRI and ⁶⁸Ga-PSMA-I/T PET-CT in detecting ECE, SVI, LNM, and BNI were verified using the histopathological examination. All demographic data analyses were performed using an independent χ² with significance at the 0.05 level. The one-sample t test and the χ² test were used to analyze the correlation between mp-MRI, ⁶⁸Ga-PSMA-I/T PET-CT, and histopathological findings. The McNemar χ² test was used to compare the sensitivity and specificity of diagnostic examinations.

Clinical data of the patients are presented in Table 1. Histopathologically, all cases were acinar adenocarcinomas. The mean age of the patients was 65.07 ± 8.01 years (range 46–82). The mean serum prostate-specific antigen (PSA) level was 9.49 ± 6.97 ng/mL (range 1.3–27). Among the 30 patients, SVI, BNI, and ECE were present in 6 (20%), 4 (13.3%), and 17 (56.7%) cases, respectively. Lymph node dissection was performed in 22 patients, and 1 patient had lymph node metastasis.

Most of the patients (96.7%) had PI-RADS 4 and 5 lesions in mp-MRI. All of the patients with SVI, ECE, LNM, and who were BNI positive were in the high-risk group. PI-RADS 5 lesions were reported in all patients who had a positive SVI and BNI status and also PI-RADS 4 and 5 lesions were reported in all patients who had a positive ECE status. Moreover, preoperative PSA levels and SUV_max-T values were significantly higher in patients with SVI, and SUV_max-T was higher in patients with BNI (p < 0.05; Table 2).

Table 3 shows the association between imaging variables and Gleason scores. Both mp-MRI and ⁶⁸Ga-PSMA-I/T PET-CT had no statistically significant results for the assessment of SVI, BNI, and ECE-positive or negative tumors according to prostate pathology. However, both ⁶⁸Ga-PSMA-I/T PET-CT LNM and mp-MRI predicted LNM significantly. Although statistically not significant, mp-MRI had higher overall sensitivity for ECE, overall specificity for SVI, ECE, and BNI, and higher positive predictive value (PPV) for ECE, SVI, and BNI. Furthermore, ⁶⁸Ga-PSMA-I/T PET-CT had higher overall sensitivity for BNI and negative predictive value (NPV) for BNI and LNM, as shown in Table 4 and Figures 1 and 2.

An accurate test must provide valuable information about the initial evaluation, diagnosis, and staging of patients with suspected PCa. Specificity and sensitivity are much more critical with cancer imaging modalities because false-positive results create anxiety and lead to unnecessary and invasive follow-up tests such as biopsies. The most valuable imaging technique for the diagnosis and staging is still controversial for patients with PCa.

Despite a small number of patients, to the best of our knowledge this is the first study to compare both ⁶⁸Ga-PSMA-I/T PET-CT and mp-MRI to histological findings from the RP specimen. Preoperative accurate assessment of lymph node involvement, SVI, ECE, and BNI status is critical in intermediate- to high-risk PCa patients. Unfortunately, the most commonly used conventional imaging has limited specificity and sensitivity for preoperative assessment, so the clinical status of the patients is largely underestimated with these traditional techniques [26, 27]. We evaluated the potential of ⁶⁸Ga-PSMA-I/T PET-CT for imaging primary PCa by intra-individual comparison to mp-MRI. Our results showed that both the ⁶⁸Ga-PSMA-I/T PET-CT and mp-MRI had a significantly high sensitivity to detect LNM. However, ⁶⁸Ga-PSMA-I/T PET-CT and mp-MRI showed no statistically significant results for detecting SVI, BNI, and ECE. mp-MRI had higher overall sensitivity for ECE, overall specificity for SVI, ECE, and BNI, and PPV for ECE, SVI, and BNI. ⁶⁸Ga-PSMA-I/T PET-CT had higher overall sensitivity for BNI and NPV for BNI and LNM. A comparison of imaging modalities to histopathologic results was the fundamental strength of this study.

Evaluation of SVI and ECE in the case of PCa is critical to determine the correct therapeutic approach because SVI or ECE invasion are crucial prognostic factors for recurrence after RP. Patients with an advanced local disease with ECE and SVI generally have a worse prognosis because the risk of a positive surgical margin and the incidence of LNM is increased [28]. Berger et al. [29] reported that specificity of both PSMA PET-CT and mp-MRI were similar for diagnosing SVI (92.7 vs. 95.0%, p = 0.39). Still, mp-MRI showed superior sensitivity in detecting SVI (75.0%) compared to PSMA PET-CT (11.1%). Yilmaz et al. [30] showed that the SVI assessment of mp-MRI was superior to ⁶⁸Ga-PSMA-11 PET-CT, although PET-CT and MRI were associated with clinically and statistically significant results. In our study, ⁶⁸Ga-PSMA-I/T PET-CT and mp-MRI were similar in terms of sensitivity, but specificity was superior with mp-MRI over ⁶⁸Ga-PSMA-I/T PET-CT. However, the results were not statistically significant, probably due to the number of included patients. For further confirmation, studies with a higher number of patients are needed.

The presence of ECE is a poor prognostic factor and known to increase the risk of PCa-related mortality [31]. Furthermore, nerve-sparing surgery should be performed to avoid positive surgical margins at the high-risk area of ECE [2]. Reported sensitivity and specificity of mp-MRI in the detection of ECE was not found to be sufficient (57% in a meta-analysis) [27]. However, the diagnostic performance of ⁶⁸Ga-PSMA PET-CT in the detection of EPE remains unclear. Prompt diagnosis of ECE is essential for the decision to perform nerve-sparing surgery and is a further benefit for preoperative use of ⁶⁸Ga-PSMA PET-CT. Yilmaz et al. [30] reported that the determination of EPE status was significantly possible using mp-MRI, which had a considerable range in the accuracy (87.5%), specificity (85.7%), and sensitivity (90%) that was verified with the final pathology findings for RP. In the current study, we also found that the sensitivity, specificity, PPV, and NPV of mp-MRI were higher for the detection of ECE. However, these results were not statistically significant.

Our study showed that both mp-MRI and ⁶⁸Ga-PSMA-I/T PET-CT had a similar and relatively inadequate ability to determine BNI accurately. Yilmaz et al. [30] found that mp-MRI and ⁶⁸Ga-PSMA PET-CT showed similar sensitivity, specificity, PPV, NPV, and accuracy in the diagnosis of BNI (33.3, 100, 100, 81.8, and 83.3%, respectively). ⁶⁸Ga-PSMA PET-CT has the disadvantage of bladder activity, which can be a problem for the assessment of BNI [26]. Since there are limited data in the published literature on this topic, it is currently not possible to compare our data with others until further studies are published, preferably with a higher number of patients.

LNM on routine histology is a robust negative predictor of survival [32]. Therefore, evaluation of lymph nodes should be performed optimally before treatment decisions are made. Petersen et al. [33] reported that both MRI and CT as well as DW-MRI had a lower sensitivity compared with ⁶⁸Ga-PSMA PET-CT. MRI/CT (8 and 100%) and DW-MRI (36 and 83%) had lower sensitivity and specificity for LNM detection compared to ⁶⁸Ga-PSMA PET-CT (39 and 100%, retrospectively). In general, the sensitivity of ⁶⁸Ga-PSMA PET-CT was about 66% in most trials at the patient level. Moreover, ⁶⁸Ga-PSMA PET-CT has a high specificity (98.9%) and high accuracy (88.5%) [34]. In the current study, ⁶⁸Ga-PSMA-I/T PET-CT and mp-MRI were similar and able to predict LNM in terms of sensitivity, specificity, PPV, NPV, and accuracy. However, our positive lymph node dissection group was limited.

Our study is not without limitations. Firstly, due to its retrospective nature, patients represented a heterogeneous group, and the study included a limited number of patients. The small number of patients with positive lymph nodes was another limiting factor. The results need to be validated in a prospective study comprising a larger dataset.

In this study, we demonstrated that both mp-MRI ⁶⁸Ga-PSMA PET-CT had significant results for detecting LNM in patients with PCa. Although the results for detecting ECE, SVI, and BNI were not statistically significant due to the limited number of patients, mp-MRI had superior specificity, sensitivity, NPV, PPV, and accuracy for assessing ECE and SVI in patients with PCa when compared to ⁶⁸Ga-PSMA PET-CT overall. For the detection of BNI, however, ⁶⁸Ga-PSMA PET-CT had higher sensitivity and NPV, and mp-MRI had higher specificity and PPV. We suggest that the use of mp-MRI and ⁶⁸Ga-PSMA PET-CT as new and advanced imaging modalities is a promising field in the management of patients with PCa, and further studies with higher numbers are urgently required to provide a more accurate conclusion and enable these modalities to be used more widely.

All patients included in this study gave written informed consent. This study complies with the guidelines for human studies and animal welfare regulations. This study protocol was approved by Pamukkale University School of Medicine Human Research Ethics Committee (approval No. 60116787-020/56122) and the study was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

The authors have no conflicts of interest to declare.

No funding was acquired for this work.

Sinan Çelen: protocol development, data collection, data analysis, and manuscript writing. Aziz Gültekin, Yusuf Özlülerden, Aslı Mete, Ergin Sağtaş, Baki Yağcı: data collection. Furkan Ufuk, Ali Ersin Zümrütbaş, Doğangün Yüksel: protocol development and manuscript editing.