Current guidelines recommend CT, MRI, and bone scintigraphy for primary staging of high-risk prostate cancer (PC).1 Distinguishing between disease confined to the prostate gland from that already spreading to the lymph nodes or distant organs is essential for planning therapy. However, lymph node assessment by CT and MRI is primarily based on morphological information such as lymph node size and irregular shape,2 resulting in a low sensitivity of these imaging modalities.3 Bone scintigraphy has a high sensitivity and, when performed as hybrid imaging together with CT, also a high specificity4,5 but allows detection of osseous spread only. For PET/CT, 18F-fluorocholine6 and 11C-choline7 have widely been used for PC staging,8,9 although they lack sensitivity and specificity at low prostate-specific antigen (PSA) levels.10,11 Recently, small molecule inhibitors targeting the prostate-specific membrane antigen (PSMA) have entered the arena of radionuclide imaging.12 Prostate-specific membrane antigen is a cell surface protein expressed 100- to 1000-fold higher in PC cells than in benign prostate tissue or normal tissue of most other organs.12–14 The PET tracer 68Ga-PSMA-HBED-CC is so far the most common clinically used small-molecule PSMA inhibitor offering higher diagnostic accuracy than conventional imaging methods and choline PET.15 Because of the broader availability of SPECT systems and their lower instrumentation and radionuclide costs compared with PET, the development of a 99mTc-labeled small molecule inhibitor is of some clinical interest. MIP-1404 (Progenics Pharmaceuticals, Inc, New York, NY) is a novel glutamate urea amino acid heterodimeric inhibitor that binds to PSMA on the cell surface of PC cells.16 Biodistribution and pharmacokinetics of MIP-1404 are favorable, with rapid blood clearance, low urinary activity, and high lesion contrast, allowing detection of even subcentimeter lymph node metastases.17 Evidence on its clinical value is so far scarce.18 Recently, we reported a high diagnostic accuracy of 99mTc-MIP-1404 SPECT/CT to detect the correlate of a biochemical recurrence of PC in a group of 225 patients.19 So far, there are no clinical data concerning the detection rates and interobserver variability of MIP-1404 for primary whole-body staging in patients with PC. Prior to widespread clinical adoption of PSMA-targeted SPECT, knowledge of its capability for detection of primary PC and distant metastases, as well as on interobserver variability and reproducibility, is needed.
In this article, we report data from 93 patients with PC examined by 99mTc-MIP-1404 SPECT/CT prior to first therapy. In this group of subjects, we determined detection rates for primary PC, local lymph nodes, and distant metastases. Furthermore, we investigated interobserver agreement of reading 99mTc-MIP-1404 SPECT/CT scans and assessed the correlation between clinical, biochemical, and histological parameters on the one hand with detection rates of metastases on the other.
PATIENTS AND METHODS
Patients and Standard of Reference
From April 2013 to May 2017, a total of 400 patients were examined with 99mTc-MIP-1404 SPECT/CT at the Clinic of Nuclear Medicine at the University Hospital Erlangen. From this collective, we retrospectively selected all subjects conforming to the following criteria: histopathologically confirmed PC by needle biopsy and no primary therapy (eg, prostatectomy with or without lymph node dissection, radiation therapy, androgen deprivation therapy).
After applying these criteria, 93 patients were ultimately included in the study. For each subject, a complete medical history including histology of confirmed PC, PSA values, clinical T stage, and demographic information was collected. Furthermore, follow-up information including therapeutic strategy after MIP imaging and the rate of biochemical recurrence of PC after primary therapy was obtained and available for 52 (56%) of 93 patients. Patient characteristics are summarized in Table 1.
Being an investigational radiopharmaceutical 99mTc-MIP-1404 is not yet US Food and Drug Administration or European Medicines Evaluation Agency approved. It was therefore used under the conditions outlined in §13(2b) of the Arzneimittelgesetz (German Medicinal Products Act). Our protocol was approved by the institutional review board of the University Hospital Erlangen, and written informed consent was obtained from all subjects.
Radiosynthesis of 99mTc-MIP-1404
99mTc-MIP-1404 was synthesized as described previously.16,20 The radiopharmaceutical 99mTc-MIP-1404 was formulated in a radioactivity concentration of 234 ± 19 MBq/mL (6.3 ± 0.5 mCi) (referred to the time of expiry, n = 141), with high radiochemical purity of 96.5% ± 1.3% and specific radioactivity of 18.5 ± 1.6 GBq/mg.
Image Acquisition and Reconstruction
The injected radioactivity of 99mTc-MIP-1404, administered at 3 to 4 hours prior to image acquisition, ranged from 516 to 794 MBq (13.9–21.4 mCi), with a mean radioactivity of 706 ± 72 MBq (19.1 ± 1.9 mCi) for this study cohort. Image acquisition was performed using a Symbia T2 SPECT/CT system (Siemens Healthcare, Hoffman Estates, Ill). After performing whole-body planar scans (anterior and posterior views) using low-energy high-resolution collimation and a scan speed of 12 cm/min using a body-contouring protocol, SPECT/CT of the abdomen and thorax was done. If any suspect lesion outside the thorax or abdomen occurred on planar scintigraphy, SPECT/CT of the respective region was performed. SPECT scans were obtained likewise with low-energy high-resolution collimation, a 128 × 128 matrix with 4.8-mm pixel size, and a total of 120 projections over 360 degrees with a projection dwell time of 15 seconds. SPECT scans were followed by low-dose CT (130 kV, 30 mAs) using adaptive dose modulation (CARE Dose 4D; Siemens Healthcare). CT data were reconstructed with 3- and 5-mm slice thicknesses using B70s and B41s kernels for image analysis. SPECT data set reconstruction was performed using Siemens' implementation of an ordered subset maximization algorithm with 4 subsets and 8 iterations, including point-spread-function modeling (Siemens Flash 3D) with CT-based attenuation correction and dual-energy window scatter correction.
All MIP-1404 scans were analyzed with commercially available software (Syngo.via; Siemens Molecular Imaging, Hoffman Estates, IL), allowing review of SPECT, CT, and fused imaging data. Visual evaluation was performed by 2 nuclear medicine physicians independently from each other. Before evaluation of interobserver differences, both observers performed consensus reading of 5 MIP-1404 data sets (not included in study analysis) to provide for basic agreement in definition of pathological tracer uptake in PSMA-positive lesions. The following information was disclosed to both observers before image interpretation: indication (primary staging in all patients), sex (male or female), age (years), height (cm), weight (kg), injected dose (MBq), and PSA level (ng/mL).
The observers were masked to all other data. Images were visually interpreted for the following information: overall scan result for presence or absence of disease, presence or absence of pathological uptake in the prostate region indicating primary tumor, lymph nodes affected (yes or no), number of lymph node regions affected, location of lymph node regions affected, number of lymph node metastases detected, presence of bone metastases (yes or no), number of bone metastases, disseminated bone metastases (yes or no), number of organs affected, and number of organ metastases detected.
In addition, in all 93 patients, quantitative analysis of tracer uptake in PSMA-positive lesions in the prostate and normal tissues of the gluteal muscle was performed using a protocol developed by our group.21 For this, volumes of interest (VOIs) were drawn, and the extracted counts were converted to absolute activity concentrations in units of kBq/mL and subsequently multiplied by patient weight and divided by injected dose in order to yield the respective SUV values. For normal tissues, SUVmean was determined on spherical VOIs in the gluteal muscle covering a representative amount of tissue and a minimum VOI size of at least 10 mL in order to minimize statistical fluctuations and partial volume effects. For suspected pathological lesions, SUVmean, SUVmax, and VOI volume were determined by VOIs with isocontours set at 40% of the maximum uptake within the respective focus. Tumor-to-normal ratios (TNRs) were calculated for each detected primary lesion individually by dividing the SUVmax of the lesion by the SUVmean of the reference regions (gluteal muscle for soft tissue lesions) of the respective patient. In addition, we determined a surrogate parameter for each subject using the quantitative SPECT images by determining the amount of activity (mean activity concentration multiplied by volume of the SPECT VOI) in the prostate gland and multiplying this by subject weight and normalizing by injected activity, as shown in the following formula.
This parameter (in the following referred to as TUprostate) consequently reflects the total amount of pathological uptake of MIP-1404 in the prostate gland for the respective patient.
For binary data, agreement between both observers was determined by calculation of percentage of concordant findings. In addition, 95% confidence intervals (95% CIs) are reported for these values. For nonbinary data, agreement among observers was evaluated by calculation of percent of concordant findings (eg, concordant numbers of positive lymph nodes) and mean differences (Δ) ± SD in case of discrepant numbers.
In patients who underwent radical prostatectomy with lymph node dissection, sensitivity and specificity were calculated using histopathology results as criterion standard.
Receiver operating characteristic analyses and area under the curve (AUCs) with 95% CIs were calculated for TUprostate and TNR values in patients with distant metastases to determine optimal cutoff values. All statistical analyses were performed using MATLAB version R2012b (The MathWorks Inc, Natick, MA) and SPSS version 21.0 (IBM Corp, Armonk, NY).
On a patient basis, PC was detected in 90 of 93 patients (detection rate, 97%; 95% CI, 0.91–0.99). Primary tumor was also diagnosed in 90 of 93 patients (detection rate, 97%; 95% CI, 0.91–0.99). A total of 48 MIP-1404–positive lymph nodes occurred in 16 patients (see Fig. 1 for MIP-1404 imaging in a patient with lymph node metastasis). Twenty-nine bone metastases were present in 9 patients. Three patients showed disseminated bone metastases (see Fig. 2 for MIP-1404 imaging of a patient with metastatic bone disease). No distant organ metastases were detected.
An overview of the interobserver agreement for visual image interpretation is shown in Table 2. Interobserver agreement was high for the overall scan result, the detection of primaries, and for lymph node or bone involvement (all >96%). The mean absolute difference between observers was low for number of detected lymph node and bone metastases (each Δ ~1). Both observers identified the same lymph node regions as location of disease.
Follow-up data were available in 52 patients (56%) (Table 3). The median follow-up period was 18 months (range, 2–36 months). Twenty-five (48%) of 52 patients underwent radical prostatectomy with lymph node dissection at least 1 week but not more than up to 4 weeks after MIP-1404 imaging. Nineteen patients underwent radiation therapy (37%), and 8 patients underwent androgen deprivation therapy (15%). Recurrence of PC occurred in 5 patients (10%). All 5 patients suffered from local recurrence in the prostate region.
Sensitivity and Specificity
In the 25 patients undergoing prostatectomy with lymph node dissection, 25 prostate glands and a total of 315 lymph nodes were removed.
All removed glands were positive for PC by pathology. Of these, 24 glands were true-positive, and 1 gland was false-negative in MIP-1404 imaging, which results in a sensitivity of 97% for detection of primary tumor. Because all patients had histologically confirmed adenocarcinoma of the prostate, specificity could not be calculated.
Only 2 lymph nodes of 1 patient were positive for PC by pathology. Both nodes were true-positive in MIP-1404 imaging, with no false-negatives. Sensitivity was not calculated because of the low number of lymph nodes that were positive by pathology. Of the 312 nodes that were negative by pathology, all nodes were true-negative in MIP-1404 imaging, with no false-positives, which results in a specificity of 100% for detection of lymph node metastases.
Rate of Metastasis
In patients with PSA levels of 10 ng/mL or greater (n = 58), the rate of MIP-1404–positive metastases was significantly higher (Wilcoxon rank sum test, P < 0.05) compared with patients with PSA levels of less than 10 ng/mL (n = 35) (Fig. 3). Patients with a Gleason score of 8 or greater (n = 51) showed significantly more MIP-1404–positive metastatic lesions compared with patients with a Gleason score of 7 or less (n = 42) (Wilcoxon rank sum test, P < 0.05) (Fig. 4).
Quantitative analysis of 99mTc-MIP-1404–positive primary lesions in the prostate yielded SUVmax values of, on average, 9.6 ± 8.1, ranging from 1.0 to 47.8, and TNR of 110.9 ± 100.5, ranging from 8.7 to 533.5.
In addition, the TUprostate of 99mTc-MIP-1404 was significantly higher in patients with Gleason score of 8 or greater compared with patients with Gleason score of 7 or less (Wilcoxon rank sum test, P < 0.05) (Fig. 5). TUprostate of MIP-1404 was significantly (Wilcoxon rank sum test, P < 0.001) higher for patients with metastases compared with patients without metastatic lesions (Fig. 6).
Receiver operating characteristic (Fig. 7) analysis for TUprostate revealed an optimal cutoff value of 26 for discriminating between patients with and without metastatic disease, with a sensitivity and specificity of 82% and 76% (AUC, 0.81; 95% CI, 0.68–0.94). For TNR, the optimal cutoff value was 96.2 for discriminating between patients with and without metastatic disease, with a sensitivity and specificity of 71% and 64% (AUC, 0.75; 95% CI, 0.61–0.89).
99mTc-MIP-1404, a novel SPECT-compatible PSMA ligand, demonstrated high sensitivity for detection of primary PC with high interobserver agreement. The overall detection rate of PSMA-positive lesions in 93 patients was 97% in both observers. High sensitivity for detection of primary PC (97%) confirmed by postoperative histology was shown. In addition, it was seen that patients at higher-risk groups (Gleason score ≥8 and PSA ≥10 ng/mL) had a higher prevalence of metastasis, as expected.
So far the only comparable study by Goffin et al22 in 104 patients who underwent 99mTc-MIP-1404 SPECT/CT and MRI, a sensitivity of 94% for detection of primary PC was found, which is in the range of our results. From the same study, a sensitivity and specificity of 50% and 87% for detecting lymph node metastasis were reported. Unfortunately, we cannot reasonably calculate this for our data because of a very low rate of lymph nodes that were diagnosed as PC positive in pathology.
As of now, there are no data reported in literature for the detection of bone metastases in primary staging using MIP-1404. In this regard, our data extend those published by Goffin et al22 by demonstrating that osseous spread may be detected prior to therapy in a significant number of patients. This is an important finding because the presence of distant bone metastases has a considerable effect on mortality and morbidity and alters therapy regimen.23
For the PSMA-targeting PET tracers, for example, Eiber et al24 found detection rates of 92% for 68Ga-PSMA PET/CT in 53 patients with primary PC and a higher rate of 98% for 68Ga-PSMA PET/MRI. Sachpekidis et al25 reported overall detection rates of 96%. Regarding lymph node involvement, Maurer et al26 detected 117 lymph node metastases in 41 of 130 patients, using 68Ga-PSMA PET/CT, resulting in sensitivity and specificity values of 66% and 99%. Leeuwen et al27 even reported a sensitivity and specificity of 46% and 95% for preoperative lymph node assessment using 68Ga-PSMA PET/CT.
Interobserver applicability is an important aspect of clinical applicability; whereas studies regarding PSMA PET/CT are available,28 so far there are no data regarding whole-body staging in PSMA SPECT/CT. Our findings indicate that MIP-1404 imaging has a high interobserver reliability for staging in patients with primary PC.
Quantitative assessment of tumor lesions by determination of tracer uptake is getting more and more established in PSMA PET/CT. Our quantification protocol is based on previous work from our group.21,29 Using this protocol, we evaluated PSMA-positive lesions for tracer uptake in absolute units. Whereas Goffin et al22 demonstrated that visual assessment and TNRs could discriminate normal prostate tissue and lower-grade PC from higher-grade PC and predict lymph node involvement, we implemented the concept of total uptake in the prostate gland (TUprostate) to achieve a standardized observer independent uptake parameter. Using this approach, we could demonstrate that total uptake values in primary prostate tumors using PSMA SPECT/CT can discriminate between risk groups based on Gleason score and PSA level. This goes in line with previous findings based on ex vivo results of tumor immunohistochemistry.30
One of the most important findings in this study is that total uptake in the primary tumor is a predictor of lymph node and bone metastasis with a sensitivity of 82% and a specificity of 76%. Whereas Goffin et al22 reported that TNRs might predict lymph node invasion with a sensitivity of 90% and specificity of 67%, we could demonstrate for the first time that total uptake values might also predict distant bone metastases. In our study, TUprostate performed slightly better in predicting metastasis, compared with TNRs, although not statistically significant. Total uptake values in the primary tumor could be used to identify patients at higher risk of lymph node or bone metastasis and might be a molecular marker to improve patient allocation to different risk groups and therapeutic strategies.
MIP-1404 SPECT/CT has a high accuracy and low interobserver variability in the diagnosis of PC and allows detection of lymph node and bone metastases in a significant proportion of as yet untreated PC patients. Total tracer uptake of the primary tumor is correlated with Gleason score and PSA serum concentration and allows prediction of the occurrence of lymph node and bone metastases with moderate accuracy at primary staging.
1. Heidenreich A, Bastian PJ, Bellmunt J, et al. EAU guidelines on prostate cancer
. Part 1: screening, diagnosis, and local treatment with curative intent-update 2013. Eur Urol
2. Sathianathen NJ, Lamb A, Nair R, et al. Updates of prostate cancer
staging: prostate-specific membrane antigen. Investig Clin Urol
3. Hövels A, Heesakkers R, Adang E, et al. The diagnostic accuracy of CT and MRI in the staging of pelvic lymph nodes in patients with prostate cancer
: a meta-analysis. Clin Radiol
4. Römer W, Nömayr A, Uder M, et al. SPECT-guided CT for evaluating foci of increased bone metabolism classified as indeterminate on SPECT in cancer patients. J Nucl Med
5. Helyar V, Mohan HK, Barwick T, et al. The added value of multislice SPECT/CT
in patients with equivocal bony metastasis from carcinoma of the prostate. Eur J Nucl Med Mol Imaging
6. Bauman G, Belhocine T, Kovacs M, et al. 18
F-fluorocholine for prostate cancer
imaging: a systematic review of the literature. Prostate Cancer Prostatic Dis
7. Fuccio C, Rubello D, Castellucci P, et al. Choline PET/CT for prostate cancer
: main clinical applications. Eur J Radiol
8. Hartenbach M, Hartenbach S, Bechtloff W, et al. Combined PET/MRI improves diagnostic accuracy in patients with prostate cancer
: a prospective diagnostic trial. Clin Cancer Res
9. Giovacchini G, Picchio M, Garcia-Parra R, et al. 11
C-choline PET/CT predicts prostate cancer
–specific survival in patients with biochemical failure during androgen-deprivation therapy. J Nucl Med
10. Castellucci P, Ceci F, Graziani T, et al. Early biochemical relapse after radical prostatectomy: which prostate cancer
patients may benefit from a restaging 11
C-choline PET/CT scan before salvage radiation therapy? J Nucl Med
11. Suardi N, Gandaglia G, Gallina A, et al. Long-term outcomes of salvage lymph node dissection for clinically recurrent prostate cancer
: results of a single-institution series with a minimum follow-up of 5 years. Eur Urol
12. Afshar-Oromieh A, Babich JW, Kratochwil C, et al. The rise of PSMA ligands for diagnosis and therapy of prostate cancer
. J Nucl Med
13. Sokoloff R, Norton KC, Gasior CL, et al. A dual-monoclonal sandwich assay for prostate-specific membrane antigen: levels in tissues, seminal fluid and urine. Prostate
14. Silver DA, Pellicer I, Fair WR, et al. Prostate-specific membrane antigen expression in normal and malignant human tissues. Clin Cancer Res
15. Maurer T, Eiber M, Schwaiger M, et al. Current use of PSMA-PET in prostate cancer
management. Nat Rev Urol
16. Hillier SM, Maresca KP, Lu G, et al. 99m
Tc-labeled small-molecule inhibitors of prostate-specific membrane antigen for molecular imaging of prostate cancer
. J Nucl Med
17. Vallabhajosula S, Nikolopoulou A, Babich JW, et al. 99m
Tc-labeled small-molecule inhibitors of prostate-specific membrane antigen: pharmacokinetics and biodistribution studies in healthy subjects and patients with metastatic prostate cancer
. J Nucl Med
18. Reinfelder J, Kuwert T, Beck M, et al. First experience with SPECT/CT
Using a 99m
Tc-labeled inhibitor for prostate-specific membrane antigen in patients with biochemical recurrence of prostate cancer
. Clin Nucl Med
19. Schmidkonz C, Hollweg C, Beck M, et al. 99m
for the detection of PSMA-positive lesions in 225 patients with biochemical recurrence of prostate cancer
20. Lu G, Maresca KP, Hillier SM, et al. Synthesis and SAR of 99m
Tc/Re-labeled small molecule prostate specific membrane antigen inhibitors with novel polar chelates. Bioorg Med Chem Lett
21. Zeintl J, Vija AH, Yahil A, et al. Quantitative accuracy of clinical 99m
using ordered-subset expectation maximization with 3-dimensional resolution recovery, attenuation, and scatter correction. J Nucl Med
22. Goffin KE, Joniau S, Tenke P, et al. Phase 2 study of 99m
to identify and localize prostate cancer
in intermediate- and high-risk patients undergoing radical prostatectomy and extended pelvic LN dissection. J Nucl Med
23. Bäuerle T, Semmler W. Imaging response to systemic therapy for bone metastases. Eur Radiol
24. Eiber M, Weirich G, Holzapfel K, et al. Simultaneous 68
Ga-PSMA HBED-CC PET/MRI improves the localization of primary prostate cancer
. Eur Urol
25. Sachpekidis C, Kopka K, Eder M, et al. 68
Ga-PSMA-11 dynamic PET/CT imaging in primary prostate cancer
. Clin Nucl Med
26. Maurer T, Gschwend JE, Rauscher I, et al. Diagnostic efficacy of 68 gallium-PSMA positron emission tomography compared to conventional imaging for lymph node staging of 130 consecutive patients with intermediate to high risk prostate cancer
. J Urol
27. Leeuwen PJ, Emmett L, Ho B, et al. Prospective evaluation of 68
Gallium-prostate-specific membrane antigen positron emission tomography/computed tomography for preoperative lymph node staging in prostate cancer
. BJU Int
28. Fendler WP, Calais J, Allen-Auerbach M, et al. 68
Ga-PSMA-11 PET/CT interobserver agreement for prostate cancer
assessments: an international multicenter prospective study. J Nucl Med
29. Ritt P, Vija H, Hornegger J, et al. Absolute quantification
in SPECT. Eur J Nucl Med Mol Imaging
30. Perner S, Hofer MD, Kim R, et al. Prostate-specific membrane antigen expression as a predictor of prostate cancer
progression. Hum Pathol
Keywords:Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
99mTc-PSMA; absolute quantification; primary staging; prostate cancer; SPECT/CT