Prospective Study of the Radiolabeled GRPR Antagonist BAY86-7548 for Positron Emission Tomography/Computed Tomography Imaging of Newly Diagnosed Prostate Cancer

Abstract

Background:

Current imaging techniques may not detect all prostate cancer (PCa) lesions.

Objective:

To evaluate positron emission tomography (PET)/computed tomography (CT) using the radiolabeled GRPR antagonist probe BAY86–7548 (⁶⁸Ga-RM2) for localization of newly diagnosed PCa in comparison with multiparametric magnetic resonance imaging (mpMRI), histopathology, and immunohistochemistry (IHC).

Design, setting, and participants:

This was a prospective study of 16 men with biopsyproven PCa (2 low, 8 intermediate, and 6 high risk). ⁶⁸Ga-RM2 PET/CT was performed within 4 wk after mpMRI and within 2 wk before radical prostatectomy and extended bilateral pelvic lymph node dissection.

Outcome measurements and statistical analysis:

The presence of cancer was evaluated by blinded specialists using a 5-point Likert scale, with lesions scoring 4 or 5 considered positive, on ⁶⁸Ga-RM2 PET/CT, mpMRI, and ⁶⁸Ga-RM2 PET/CT-mpMRI fused images for each of 12 anatomic areas of the prostate. Whole-mount, step-section pathology served as the reference standard. Expression of GRPR and prostate-specific membrane antigen (PSMA) was analyzed via IHC of tumor paraffin sections.

Results and limitations:

Of 192 areas analyzed, 128 contained cancer. The sensitivity, specificity, and accuracy of ⁶⁸Ga-RM2 PET/CT imaging and mpMRI did not differ significantly; fusing the images maximized the sensitivity and accuracy (85.2% and 83.9%, respectively) and averaged the specificity (81.3%). The area under the receiver operating characteristic curve was 0.76 for PET visual analysis, 0.72 for PET quantitative analysis, 0.76 for mpMRI, and 0.85 for combined PET/CT and mpMRI analysis. ⁶⁸Ga-RM2 uptake did not correlate with Gleason score. IHC analysis revealed weaker staining for GRPR than for PSMA, and the expression of these markers was not correlated (r = 0.3882). The major limitation is the small sample size.

Conclusions:

⁶⁸Ga-RM2 PET/CT is promising for detection and localization of primary PCa, and complements mpMRI. GRPR expression appears to be independent from PSMA expression, suggesting that GRPR- and PSMA-targeted PET imaging may be complementary.

Patient summary:

This pilot prospective study shows that a positron emission tomography probe that binds to a marker of prostate cancer, GRPR, improves the ability of magnetic resonance imaging to detect prostate cancer

1. Introduction

Appropriate treatment selection for prostate cancer (PCa) hinges on accurate determination of its grade and anatomic distribution. This task is aided by multiparametric magnetic resonance imaging (mpMRI) [1], which is now standard in PCa care, but mpMRI may fail to detect some lesions and has limited diagnostic accuracy for pelvic lymph node metastases [1,2]. Recent advances in the understanding of PCa biology have led to the development of positron emission tomography (PET) probes that may more reliably detect cancer foci [3,4]. Among these probes, those targeting prostate-specific membrane antigen (PSMA), which is widely expressed on PCa cells, so far appear to be more sensitive and specific than mpMRI for detection of cancer in the prostate and lymph nodes [5–9]. However, relatively few studies have compared PSMA-targeted PET with a thorough histologic evaluation, and case reports have identified multiple sources of false-positive findings [10–12]. Thus, alternative imaging targets complementary to PSMA are needed.

GRPR, one of the four known bombesin receptors, is consistently overexpressed in PCa [13]. It has been shown that GRPR antagonist probes, such as ⁶⁸Ga-RM2 (BAY86–7548), can image cancer foci in the prostate and lymph nodes with high contrast on PET/computerized tomography (PET/CT) [14,15]. The first clinical studies evaluating GRPR antagonists in primary PCa are promising, with good sensitivity in detecting the primary tumor [15–17]. Nevertheless, none of these studies compared GRPR antagonists in PET and mpMRI and none analyzed the images by 12 regions of interest. In this pilot study, we compared the performance of ⁶⁸Ga-RM2 PET/CT for localizing newly diagnosed PCa with mpMRI. We also compared GRPR and PSMA expression in tumors via immunohistochemistry (IHC) to understand the relationship between the two targets.

2. Patients and methods

2.1. Patients

This prospective clinical study (NCT02559115) was performed after institutional review board approval (IND 125834). Written informed consent was obtained from all patients. Patients with newly diagnosed, biopsy-proven PCa scheduled to undergo surgical resection were enrolled from the urology clinic. Two patients with low-risk, eight with intermediaterisk, and six with high-risk PCa were included (16 patients in total). All patients underwent pelvic mpMRI no more than 4 wk before ⁶⁸Ga-RM2 PET/CT, which was performed in all patients no more than 2 wk before surgery. Surgery consisted of laparoscopic radical prostatectomy and extended pelvic lymph node dissection, including the external iliac, hypogastric, and obturator fossa nodal packets bilaterally. Additional areas with ⁶⁸Ga-RM2 uptake were included in the dissection.

2.2. MRI protocol

MRI was performed using an MR750 3-T whole-body unit (General Electric, Boston, MA, USA). Transverse T1-weighted and transverse, coronal, and sagittal T2-weighted fast spinecho sequences were obtained. Diffusion-weighted MRI was acquired in the transverse plane using a single-shot echo-planar imaging sequence with multiple b values (0–1000) using a pelvic coil.

2.3. ⁶⁸Ga-RM2 PET/CT imaging

⁶⁸Ga-RM2 (150–200 MBq) was injected intravenously as a slow bolus. PET/CT scans extending from the mid-thigh to the top of the skull were performed 60 min (±10 min) after injection on a Discovery 710 PET/CT system (GE Healthcare) in time-of-flight mode using an acquisition time of 3 min per bed position. Low-dose CT was used for attenuation correction and anatomic localization (80 mAs, 120 kVp). Images were reconstructed iteratively using the ordered subset expectation maximization (OSEM) algorithm provided by the manufacturer (2 iterations of 16 subsets with point-spread-function modeling followed by 6.4-mm full width at half-maximum Gaussian in-plane and 0.25, 0.5, and 0.25 z-axis post filters). ⁶⁸Ga-RM2 PET/CT Imaging was performed at an average of 2.6 mo after biopsy (range 27 d–8.5 mo).

2.4. Image analysis

⁶⁸Ga-RM2 PET/CT and mpMRI scans were evaluated by an experienced nuclear medicine physician and an experienced radiologist, respectively. Each reader was blinded to the other imaging findings, biopsy results, and all other clinical data, except for the patient’s PCa diagnosis. After independent analysis of each modality, fused PET/CT-mpMRI images were analyzed. The fusion was achieved using Hermes software. For fusion, manual overlay between the CT of the PET images and the axial T1 scan of the mpMRI was performed using anatomic landmarks (seminal vesicles, hip bone, anterior rectal wall, and posterior bladder wall).

The first analysis focus was to determine the dominant prostatic lesion for each patient. A dominant lesion was defined as the higher focal uptake on PET images and the larger lesion on mpMRI. Then each reader interpreted the prostate imaging by area. The prostate was divided into 12 areas: right/left, anterior/posterior, and base/mid gland/apex. For PET/CT, the reader recorded the level of suspicion for cancer in each area using a 5-point scale: 1 = very low; 2 = low; 3 = intermediate-equivocal; 4 = high; and 5 = very high. For mpMRI, the reader used the 5 point Prostate Imaging-Reporting and Data System classification. If a lesion was localized between two areas, both areas were considered as positive.

A retrospective analysis of benign prostatic hyperplasia (BPH) nodules was also performed. Maximum standardized uptake (SUVmax) values for normal prostatic tissue, BPH, and dominant tumor were compared.

On PET images, the minimum intensity of uptake for a positive lesion was defined as the mean plus two times the standard deviation in normal prostate tissue. SUVmax values (normalized to patient’s body weight) were recorded for each area and for each dominant lesion on PET. The presence of extraprostatic extension (EPE), seminal vesicle involvement (SVI), and lymph node or bone metastasis was also recorded.

2.5. Pathology

Radical prostatectomy specimens were coated with India ink to delineate surgical margins and were then fixed in 10% formalin. Prostate and seminal vesicles were step-sectioned transversely (or sagittally for apical tissue) at 3–4-mm intervals. Specimens were examined for location of cancer, grade groups (Gleason sum), pathologic stage, SVI, bladder neck invasion, and EPE. All lymph node specimens were counted and then fixed in neutral buffered 4% formaldehyde for 24 h, separated from the adjoining adipose tissue, and counted manually. Each retrieved node was cut in 3-mm slices, which were separately embedded in paraffin, stained with hematoxylin and eosin, and examined microscopically.

For each patient the pathologist described the dominant lesion corresponding to the larger one. For each area the presence or absence of tumor was recorded.

True positive (TP), false positive (FP), true negative (TN), and false negative (FN) mpMRI or PET/CT results were determined on the basis of histopathologic findings.

2.6. Immunohistochemistry

Formalin-fixed paraffin-embedded tumor was sliced into 5-μm sections and stained at the Molecular Cytology Core Facility of Memorial Sloan Kettering Cancer Center using a Discovery XT processor (Ventana Medical Systems, Oro Valley, AZ, USA). Sections were deparaffinized with EZPrep buffer (Ventana Medical Systems) and then antigens were retrieved using CC1 buffer (Ventana Medical Systems) and blocked for 30 min with Background Buster solution (Innovex, Lincoln, RI, USA), followed by avidin-biotin for 8 min (Ventana Medical Systems).

2.6.1. Anti-GRPR

Slides were incubated with anti-GRPR (Abcam, Cambridge, UK; 5 μg/ml) for 5 h and then with biotinylated goat anti-rabbit (Vector Labs, Burlingame, CA, USA) for 60 min at 1:200 dilution. Staining was detected using a DAB kit (Ventana Medical Systems). Slides were counterstained with hematoxylin and coverslipped with Permount (Fisher Scientific, Hampton, NH, USA).

2.6.2. Anti-PSMA

Sections were incubated with anti-PSMA (Proteintech, Chicago, IL, USA; 1 μg/ml) for 5 h and then with biotinylated horse anti- rabbit (Vector Labs) for 60 min at 1:200 dilution. Staining was detected and slides were counterstained as for anti-GRPR.

Staining was classified using a 5-point scale (0 = no staining to 4 = very strong staining) adapted for each antibody (staining was weaker for GRPR than for PSMA). Six fields of view across the tumor area for each patient were scored at high magnification by two blinded readers, and the average of the two scores was used for further analysis.

2.7. Statistical analysis

Receiver operating characteristic (ROC) curves were generated for PET visual analysis, PET quantitative analysis (SUVmax), mpMRI, and fused PET/CT-mpMRI, and the area under the curve (AUC) was calculated. For calculation of sensitivity and specificity, scores of 4 or 5 were considered positive for malignancy. Because multiple areas were evaluated, statistical analysis was adjusted for clustering [18]. Sensitivity, specificity, and accuracy were compared using McNemar’s test [19] and ROC curves were compared using a nonparametric method [20]. All analyses were performed using R (www.r-project.org). A p value of <0.05 was considered statistically significant for all analyses.

Correlation of SUVmax for the dominant lesion and Gleason score was evaluated using the Wilcoxon test, and correlation between PSMA and GRPR staining scores was assessed using the Spearman rank test.

3. Results

A dominant PCa lesion was identified in all 16 patients. Gleason scores ranged between 6 (3 + 3) and 9 (4 + 5), and the median PSA at the time of imaging was 12.5 ng/ml (range 1.4–64; Table 1). ⁶⁸Ga-RM2 PET and mpMRI each correctly identified 15 dominant lesions and missed one; for the two lesions that were missed, each one was missed by one imaging modality and detected by the other. The average SUVmax for dominant lesions on ⁶⁸Ga-RM2 PET was 9.1 (1.5–27.8; Fig. 1). SUVmax for the dominant lesion did not correlate with Gleason score.

Table 1–

Patient characteristics

Characteristic	Result
AUA risk (n)
Low	2
Intermediate	8
High	6
Median age, yr (range)	60 (46–68)
Median PSA, ng/ml (range)	12.5 (1.4–64)
Pathologic grade group (n)
I	1
II	7
III	5
IV	2
V	1
Median lymph nodes removed, n (IQR)	16 (5–42)

AUA = American Urological Association; PSA = prostate-specific antigen; IQR = interquartile range.

Fig. 1–

Maximum standardized uptake (SUVmax) of the dominant lesion according to Gleason score.

On area-based analysis, 128 of 192 areas (66.7%) contained PCa on histology. ⁶⁸Ga-RM2 PET/CT was 84.4% sensitive, 67.2% specific, and 78.7% accurate. mpMRI was 73.4% sensitive, 82.8% specific, and 76.6% accurate (Table 2). None of these parameters differed significantly. ⁶⁸Ga-RM2 PET imaging and mpMRI concordantly correctly detected tumor in 64.8% of the areas containing cancer (83 of 128) and concordantly correctly predicted its absence in 60.9% of the areas without cancer (39 of 64). ⁶⁸Ga-RM2 PET failed to detect tumor in 8.6% of the areas containing cancer (11 of 128), while mpMRI failed in 19.5% (25 of 128). The sensitivity, specificity, and accuracy of fused ⁶⁸Ga-RM2 PET/CT-mpMRI were 85.2%, 81.3%, and 83.9%, respectively. If a score of 3 was considered as positive, ⁶⁸Ga-RM2 PET/CT was 89% sensitive and 55% specific and mpMRI was 76.5% sensitive and 56% specific. However, considering score of 3 as positive is less accurate for imaging interpretation.

Table 2–

Tumor detection by each imaging modality on area-based analysis^a

Score	mpMRI Histology		68Ga-RM2 PET Histology
	+	−	+	−
1	25	28	1	9
2	5	8	13	25
3	4	17	6	9
4	22	3	44	8
5	72	8	64	13
Total	128	64	128	64

mpMRI = multiparametric magnetic resonance imaging; PET = positron emission tomography.

^aBold values are true positives and true negatives (where scores of 4 and 5 are considered indicative of cancer).

From the ROC curves, the AUC calculated was 0.76 for PET visual analysis, 0.72 for PET, 0.76 for mpMRI, and 0.85 for fused PET/CT-mpMRI 0.85 (Fig. 2).

Fig. 2–

Receiver operating characteristic curves for each imaging modality and maximum standardized uptake (SUV) of ⁶⁸Ga-RM2 on PET. PET = positron emission tomography; MRI = magnetic resonance imaging.

This analysis showed that eight patients had BPH nodules. In total, there were 12 BPH areas, five of which were within tumor areas. Therefore, SUVmax for BPH was calculated for seven lesions only. SUVmax for normal prostatic tissue was also recorded. The average SUVmax was 2.1 (0.6–3.3) for normal prostate tissue, 3.7 (0.45–7.1) for BPH, and 9.1 (1.5–27.8) for dominant tumor (Fig. 3). Two BPH nodules were interpreted as cancer by ⁶⁸Ga-RM2 PET (FP) but correctly identified by mpMRI (TN).

Fig. 3–

Maximum standardized uptake (SUVmax) of normal prostate tissue (PT), benign prostatic hyperplasia (BPH) and dominant tumor.

Of the three patients with lymph node metastasis on pathology, one was correctly diagnosed by ⁶⁸Ga-RM2 PET and missed by mpMRI (Fig. 4) and two were correctly diagnosed by mpMRI but missed on ⁶⁸Ga-RM2 PET/CT. In these two cases, ⁶⁸Ga-RM2 PET was either negative (SUVmax 1.5) or showed low uptake (SUVmax 3.2) in the dominant lesion in the prostate (Fig. 5). In addition, there was one FP result for lymph node metastasis on mpMRI, which was correctly classified as negative by ⁶⁸Ga-RM2 PET.

Fig. 4–

Lymph node metastasis on pathology correctly diagnosed by ⁶⁸Ga-RM2 PET and missed by mpMRI. The patient was 61 yr of age and had prostate-specific antigen of 9.1 ng/ml. (A) ⁶⁸Ga-RM2 PET/CT maximum intensity projection image showing the physiologic distribution of ⁶⁸Ga RM2. (B) Transverse ⁶⁸Ga-RM2 PET/CT-MRI fusion image reveals a lesion (maximum standardized uptake 21.9) involving all of the prostate, predominant on the left side, with potential extracapsular extension. (C) Transverse T2-weighted MRI and (D) apparent diffusion coefficient map revealing the same lesion with a low signal. (E) Histopathology section shows the lesion has a Gleason score of 9 (4 + 5). (F) Transverse ⁶⁸Ga-RM2 PET/CT-MRI fusion image shows high focal uptake in a left internal iliac lymph node highly suspicious of metastasis (confirmed on histology). (G) On transverse T1-weighted MRI, the left internal iliac lymph node does not appear enlarged and is considered nonsuspicious (false-negative lymph node). PET = positron emission tomography; mpMRI = multiparametric magnetic resonance imaging; CT = computed tomography.

Fig. 5–

Lymph node metastasis on pathology correctly diagnosed by mpMRI and missed by ⁶⁸Ga-RM2 PET. The patient was 56 yr of age and had prostate-specific antigen of 1.4 ng/ml. (A) ⁶⁸Ga-RM2 PET/CT maximum intensity projection image showing the physiologic distribution of ⁶⁸Ga RM2. (B) Transverse ⁶⁸Ga-RM2 PET/CT fusion image reveals very low uptake (maximum standardized uptake 1.5) in the left mid-gland posterior, considered equivocal evidence of cancer. Of note, the avid area near the pubic bone is due to urine at the bladder neck. (C) Transverse T2-weighted and (D) apparent diffusion coefficient MRI show a low-signal lesion in the left posterior peripheral zone, consistent with prostate cancer. (E) Histopathology section shows the lesion has a Gleason score of 7 (4 + 3). (F) Transverse ⁶⁸Ga-RM2 PET/CT-MRI fusion image reveals no uptake in an enlarged right internal iliac lymph node, considered negative for metastasis by ⁶⁸Ga-RM2 PET. (G) Transverse T1-weighted MRI shows that the lymph node is enlarged and thus suspicious for metastasis (confirmed on histology). PET = positron emission tomography; mpMRI = multiparametric magnetic resonance imaging; CT = computed tomography.

GRPR staining intensity ranged from 1.4 to 4 (median 2.2) and PSMA staining from 0.3 to 4 (median 2.8). PSMA and GRPR staining scores did not correlate (r = 0.3882; Fig. 6). In two patients, PSMA staining was negative and GRPR staining was intermediate (scores of 2.6 and 2, respectively). These patients had PSA levels of 4.3 and 6.4 ng/ml and Gleason scores of 7 (grade group 2). SUVmax for the tumors on ⁶⁸Ga-RM2 PET/CT was 11.3 and 8.3 (Fig. 7). Conversely, another two patients had low GRPR staining (scores of 1.4), but intermediate PSMA staining (scores of 2.5 and 2.4). These patients had PSA levels of 9.7 and 3.2 ng/ml and Gleason scores of 7 (grade II). SUVmax for the tumors on ⁶⁸Ga-RM2 PET/CT was 4.5 and 4.4 (Fig. 8).

Fig. 6–

Relationship between immunohistochemistry (IHC) staining scores for prostatespecific membrane antigen (PSMA) and GRPR.

Fig. 7–

Tumor negative for prostate-specific membrane antigen (PSMA). The patient had prostate-specific antigen of 4.3 ng/ml and a Gleason score of 7 (3 + 4). (A) Hematoxylin and eosin section revealing tumor on the left. (B) IHC for PSMA at (B1) low magnification and (B2) 20× magnification; score = 0.3. (C) Transverse ⁶⁸Ga-RM2 PET/CT-MRI fusion image shows high tumoral uptake (maximum standardized uptake 11.3) in posterior lesions involving both sides of the prostate, predominant in the left, evocative of prostate cancer. (D) IHC for GRPr; (D1) low magnification, (D2) 20x magnification; score = 2.6. IHC = immunohistochemistry; PET = positron emission tomography; mpMRI = multiparametric magnetic resonance imaging; CT = computed tomography.

Fig. 8–

Tumor negative for GRPR. The patient had prostate-specific antigen of 9.7 ng/ml and a Gleason score of 7 (3+4). (A) Hematoxylin and eosin section shows tumor on the right. (B) IHC for PSMA at (B1) low magnification and (B2) 20× magnification; score = 2.5. (C) Transverse ⁶⁸Ga-RM2 PET/CT-MRI fusion image shows moderate tumoral uptake (maximum standardized uptake 4.5) in the center of the prostate evocative of prostate cancer. (D) IHC for GRPR at (D1) low magnification and (D2) 20× magnification; score = 1.4. IHC = immunohistochemistry; PET = positron emission tomography; mpMRI = multiparametric magnetic resonance imaging; CT = computed tomography.

4. Discussion

This study shows that combining GRPR-targeted PET with mpMRI improves the accuracy of PCa localization compared with mpMRI alone. We also found that expression levels of GRPR and PSMA are not correlated, suggesting that GRPr- and PSMA-targeted PET may provide complementary information.

Further supporting the use of both GRPr- and PSMA-targeted PET probes, their biodistribution is quite different. While ⁶⁸Ga-PSMA-11 accumulates to some extent in the small intestine and bladder, ⁶⁸Ga-RM2 does not, which might favor ⁶⁸Ga-RM2 for detection of retroperitoneal and pelvic lesions owing to lower interference. This advantage has been confirmed by a small comparative study of ⁶⁸Ga-RM2 and ⁶⁸Ga-PSMA-11 in patients with recurrent PCa [21].

The overall accuracy of GRPR PET for PCa localization was similar to that reported for PSMA-targeted PET [6,22–24}. Eiber et al [6] reported an AUC of 0.83 for ⁶⁸Ga-PSMA-11 PET and 0.88 for combined ⁶⁸Ga-PSMA-11 PET-mpMRI, similar to the AUC values of 0.76 for ⁶⁸Ga-RM2 PET and 0.85 for combined ⁶⁸Ga-RM2 PET-mpMRI in the present study. Two smaller studies of PSMA PET/CT reported sensitivity of 49% [22] and 92% [23] and specificity of 92% [24] and 95% [22], similar to the values for GRPR-targeted PET/CT in the present study.

The diagnostic accuracy of ⁶⁸Ga-RM2 PET/CT for PCa localization was also similar to that found in an initial study on the same probe, despite differences in study design. Kahkonen et al [15] reported sensitivity, specificity, and accuracy of 88%, 81%, and 83%, respectively, following an analysis of 132 areas (57 containing cancer and 75 not) in 11 patients. Enrollment in that study required the presence of cancer in at least 20% of the biopsy material.

The feasibility of targeting PCa using other GRPR ligands, including ⁶⁴Cu-CB-TE2A-AR06, ¹⁸F-BAY 864367, and ⁶⁸Ga-NeoBOMB1, has been demonstrated in very small studies [14,16,25]. In addition, Zhang et al [17] found that PET/CT using the GRPR antagonist ⁶⁸GaRM26 is more sensitive than mpMRI. In 17 patients with primary PCa and 11 with biochemical recurrence, the probe detected 15 tumors, 19 lymph node metastases, and 21 bone metastases versus 12 tumors and six lymph node metastases according to mpMRI. In our study, ⁶⁸Ga-RM2 PET/CT yielded two FPs, which turned out to be BPH. Although autoradiographic studies indicate that GRPR is overexpressed in PCa cells and prostatic intraepithelial neoplasia relative to normal prostate and BPH [26,27], BPH-related FPs were also described by Kahkonen et al [15]. Further studies are needed to determine the degree to which BPH expression of GRPR may affect interpretation of GRPR-targeted PET. We found no correlation between ⁶⁸Ga-RM2 SUVmax and Gleason grade, in contrast to the findings of Beer et al [28], who observed lower GRPR expression in more aggressive PCa. These discrepant results may be explained by the fact that GRPR antibodies and ⁶⁸Ga-RM2 target different parts of the GRPR protein, and binding of ⁶⁸Ga-RM2 in vivo requires the presence of GRPR on the cell surface, while IHC detection does not.

While our subdivision of the prostate into 12 regions of interest rather than the six used in previous studies yields greater precision, takes into consideration anterior/posterior distribution, and reduces the likelihood of FPs, this approach still has limitations. Given the multifocal nature of PCa and the fact that lesions do not exactly fit within any predetermined regions, a positive signal within a region is not necessarily a thorough map. In addition, matching of regions defined on imaging to whole-mount pathology specimens is imprecise. We recognize that the number of patients included in this pilot study is small, limiting comparison with published data for other molecular imaging modalities. Nevertheless, these preliminary data are encouraging and confirm by strict correlation with whole-mount sections and mpMRI that GRPR is a valid target for imaging primary PCa. Moreover, fusion of mpMRI and ⁶⁸Ga-RM2 PET/CT improves the detection of primary disease and has a potential additive value.

5. Conclusions

Our results demonstrate that ⁶⁸Ga-RM2 PET/CT is promising for the detection and localization of primary PCa, and complements mpMRI. GRPR expression appears to be independent from PSMA expression, suggesting that GRPR- and PSMA-targeted PET imaging may be complementary. Further investigation of ⁶⁸Ga-RM2 PET performance is warranted and will provide insight into its utility either in lieu of or in combination with PSMA imaging across the clinical PCa spectrum.