Visual Abstract
Abstract
18F-PSMA-1007 PET is used in the management of patients with prostate cancer. However, recent reports indicate a high rate of unspecific bone uptake (UBU) with 18F-PSMA-1007, which may lead to a false-positive diagnosis. UBU has not been evaluated thoroughly. Here, we evaluate the frequency of UBU and bone metastases separately for 18F-PSMA-1007 and 68Ga-PSMA-11 in biochemical recurrence (interindividual comparison). Additionally, we investigate UBU seen in 18F-PSMA-1007 through follow-up examinations (intraindividual comparison) using 68Ga-PSMA-11 PET, bone scintigraphy, and MRI. Methods: First, all patients (n = 383) who underwent 68Ga-PSMA-11 PET between January 2020 and December 2020 and all patients (n = 409) who underwent 18F-PSMA-1007 PET between January 2020 and November 2021 due to biochemical recurrence were included for an interindividual comparison of bone metastases and UBU rate. In a second approach, we regarded all patients with UBU in 18F-PSMA-1007, characterized by focal bone uptake with an SUVmax > 4 and prostate-specific antigen (PSA) ≤ 5 ng/mL, who underwent additional 68Ga-PSMA-11 PET (n = 17) (interindividual comparison). Of these, 12 patients also had bone scintigraphy and whole-body MRI within a 1- to 5-wk interval. Bone uptake seen on 18F-PSMA-1007 but not on any of the other 4 modalities (CT, MRI [n = 1], bone scanning, and 68Ga-PSMA-11 PET) was recorded as false-positive. Results: Patients scanned with 18F-PSMA-1007 PET had a significantly higher rate of UBU than those scanned with 68Ga-PSMA-11 (140 vs. 64; P < 0.001); however, the rate of bone metastases was not significantly different (72 vs. 64; P = 0.7). In the intraindividual comparison group, workup by CT, MRI, bone scanning, and 68Ga-PSMA-11 PET resulted in a positive predictive value for 18F-PSMA-1007 focal bone uptake (mean SUVmax, 6.1 ± 2.9) per patient and per lesion of 8.3% and 3.6%, respectively. Conclusion: In patients with PSA ≤ 5 ng/mL and SUV > 4 at biochemical recurrence, most 18F-PSMA-1007 focal bone uptake is likely to be false-positive and therefore due to UBU. In the case of low clinical likelihood of metastatic disease, 18F-PSMA-1007 bone uptake without morphologic surrogate should be assessed carefully with regard to localization and clinical context. However, the rate of bone metastases was not higher with 18F-PSMA-1007 in the clinical routine, indicating that experienced reporting physicians adjust for UBU findings.
Up to 60% of prostate cancer patients develop biochemical recurrence (BCR) after initial radiotherapy or radical prostatectomy in 10 y of follow-up (1). Local salvage therapy and complete metastatic ablation of oligometastatic prostate cancer may provide a curative pathway and an alternative to initiation of palliative androgen deprivation therapy (2). Therefore, to determine location and extent of recurrent PC is of the utmost importance for directing salvage therapy.
The recent European Association of Urology Prostate Cancer guideline recommended that prostate-specific membrane antigen (PSMA) PET should be offered to BCR patients with a persistent prostate-specific antigen (PSA) level greater than 0.2 ng/mL if the results will influence subsequent treatment decisions (3). PSMA PET readers need proper training, as each PSMA ligand features distinct characteristics (4,5).
More recently, 68Ga-labeled PSMA ligands are increasingly replaced by 18F-labeled compounds offering mostly technical and logistic advantages including lower positron energy; improved spatial resolution; longer half-life; high-yield production in cyclotrons; and large batch production, thereby enabling long-distance distribution and potential cost savings (4). Moreover, 18F-PSMA-1007 exhibits blood clearance through the liver that leads to only minimal urinary excretion, yielding potential advantages for pelvic tumor assessment (6,7). However, unspecific bone uptake (UBU) on 18F-PSMA-1007 PET, reported in a considerable fraction of patients, may lead to false-positive findings as metastasis; this in turn may result in overstaging, leading to inadequate therapy (4,8,9). However, despite large observational data, UBU have not been correlated systematically by other imaging, including 68Ga-PSMA-11 PET/CT, MRI, and bone scanning.
Therefore, the aim of this study was 2-fold. First, we evaluated the rate of UBU and bone metastases reported in clinical reads separately for 18F-PSMA-1007 and 68Ga-PSMA-11 PET to estimate the relevance of UBUs (interindividual group). Second, we present a single-center experience with 18F-PSMA-1007 UBU in 17 patients, who underwent follow-up examinations to clarify the nature of the bone uptake. In those patients, we evaluated 18F-PSMA-1007 UBUs intraindividually with bone scanning, 68Ga-PSMA-11 PET, and MRI.
MATERIALS AND METHODS
Patient Characteristics
Patient characteristics are shown in Table 1 and Supplemental Table 1 (supplemental materials are available at http://jnm.snmjournals.org). All patients were recruited at the Department of Nuclear Medicine of the University Hospital Essen. The analysis was performed retrospectively, and the need for study-specific written consent was waived (Ethics approval no. 22-10694-BO and 21-9865-BO). Briefly, 2 patient cohorts were investigated: First, the rate of UBU and bone metastases in all patients scanned with 68Ga-PSMA-11 in the last year before the introduction of 18F-PSMA-1007 was compared with the respective rates in all patients scanned with 18F-PSMA-1007 in the first year of its use in our Department (interindividual comparison group). Additionally, patients who received 18F-PSMA-1007 and underwent 68Ga-PSMA-11 due to 18F-PSMA-1007 UBU clinical workup were included (intraindividual group).
Patient Characteristics (n = 17)
Inclusion Criteria of the Interindividual Comparison Group
All patients who received 68Ga-PSMA-11 PET between January 2020 and December 2020 and all patients who received 18F-PSMA-1007 PET between January 2020 and November 2021 were regarded for the interindividual comparison group. Of these, 383 and 409 patients were referred to PET due to BCR or persistence and further analyzed with regard to the rate of UBU and bone metastases. For this group of patients, bone-related imaging findings were retrospectively extracted from our archives regardless of the finding’s SUVmax and regardless of preimaging prostate-specific antigen (PSA) values in the case of histologically proven prostate cancer and biochemical recurrence (BCR) or PSA persistence without any known metastases.
The incidence of UBU and bone metastases on 18F-PSMA-1007 and 68Ga-PSMA were compared in different preimage PSA–level groups (PSA < 1 ng/mL vs. 1–5 ng/mL vs. > 5 ng/mL).
Inclusion Criteria of the Intraindividual Comparison Group
The SUVmax of UBU was reported among different studies with similar image acquisition, and the measurements ranged between 3.6 and 21.1 (4,10). Therefore, in this study, UBU was defined as focally increased 18F-PSMA-1007 uptake in the bone with an SUVmax higher than 4 and clear visualization in the maximum-intensity-projection images without CT correlate (no lytic or osteoplastic reaction). Patients with 18F-PSMA-1007 PET UBU were offered additional workup in the case of histologically proven prostate cancer, BCR of prostate cancer, PSA levels at the time of imaging ≤ 5 ng/mL, and no known distant metastases.
Patients underwent additional clinical whole-body 68Ga-PSMA-11 PET/MRI and bone scanning (together with SPECT/CT). Patient datasets were analyzed retrospectively.
Imaging and Image Interpretation of the Intraindividual Comparison Group
Tracer precursors (PSMA-11 and PSMA-1007) were obtained from ABX advanced biochemical compounds (ABX GmbH). 18F-PSMA-1007 and 68Ga-PSMA-11 were synthesized on site using a kit-based approach on automated platforms with comprehensive pH, radiochemical, chemical, and radionuclide purity control tests.
After intravenous injection (111 ± 20 min) of 18F-PSMA-1007 (350.6 ± 61.8 MBq), PET/CT was obtained between the base of the skull and midthighs with the patient supine. A Biograph Vision and Biograph mCT were used for image acquisition (all: Siemens Healthineers). Full-dose CT was acquired for attenuation correction (210 mAs, 120 keV, 512 × 512 matrix, 128 × 0.6 mm slice thickness). PET emission data were attenuation corrected by help of the CT data and iteratively reconstructed (Vision—4 iterations; 5 subsets; voxel size, 3.3 × 3.3 × 3 mm3; Gauss filtering, 4 mm, and mCT—3 iterations; 21 subsets; voxel size; 4.07 × 4.07 × 3 mm3; Gauss filtering; 4 mm) with time-of-flight information and point-spread function correction (HD PET).
68Ga-PSMA-11 PET/MR (n = 14) or PET/CT (n = 3) was used to acquire coregistrated images. The mean injected dose and mean imaging delay were 133.3 ± 81.2 MBq and 67 ± 14 min, respectively. PET/MRI examination was performed with an integrated 3.0-T Biograph mMR scanner (Siemens Healthineers), and simultaneous PET and 3D Dixon-volumetric interpolated breath-hold examination (VIBE) sequences for MRI-based scatter correction were performed, followed by a standardized whole-body MRI protocol. The following MR sequences of choice were acquired: high-resolution T2-weighted fast spin-echo sequences (axial, coronal, and sagittal planes), diffusion-weighted sequences (b values, b = 0, 500, 1,000 s/mm2), and dynamic contrast-enhanced imaging sequences (namely, T1-weighted VIBE sequence obtained every 7 s during 5–10 min). PET emission data were iteratively reconstructed (3 iterations; 21 subsets; voxel size, 2.09 × 2.09 × 2.03 mm3, Gauss filtering, 4 mm).
Whole-body planar bone scintigraphy imaging was performed after 2.5–4 h of the administration of the median dose of 628.5 MBq (range, 584–652 MBq) 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid radiopharmaceutical in a continuous mode at a rate of 15 cm/min on a 256 × 1,024 acquisition matrix of anterior and posterior planes with a dual-head γ-camera equipped with a low-energy, high-resolution collimator (Symbia T2 or Intevo; Siemens Healthineers). In all cases of uncertain radionuclide accumulations on bone scan, SPECT/CT images were acquired (15 s/view step and shoot with 128 × 128 matrix).
The time interval between the PET image acquisitions was between 1 and 5 wk. Images were interpreted using a dedicated workstation and software (SyngoVia; Siemens). All available imaging modalities were present for retrospective image reading. All PET and bone scintigraphy images were interpreted by 2 nuclear medicine physicians, and MR images were interpreted by 2 radiologists. Two nuclear medicine physicians performed semiquantitative analyses of the PET data retrospectively in consensus. For example, a focal bone uptake of 18F-PSMA-1007 (Fig. 1A) showing contrast enhancement (Fig. 1D), diffusion restriction (Fig. 1E), and radiotracer uptake in 68Ga-PSMA PET (Fig. 1B) and bone scintigraphy (Fig. 1C) was rated as bone metastasis. Conversely, a focal 18F-PSMA-1007 uptake of the bone without any suspicious finding on bone scan, 68Ga-PSMA-11, and MRI was rated as false-positive (Figs. 1F–1H).
Exemplary cases of UBU regarded as true-positive and false-positive. Axial slices of a patient with suspected UBU on 18F-PSMA-1007 PET (A, arrow). Suggestive uptake was seen on 68Ga-PSMA-11 PET/MRI (B, arrow) and on bone scan SPECT/CT (C, arrow). Corroborating these findings, the MRI showed contrast enhancement (D, arrowhead) and diffusion restriction (E, arrowhead). Therefore, the bone uptake was rated as true-positive. A second patient is shown in F–H. Axial slices of a patient with unspecific 18F-PSMA-1007 uptake rated as false-positive in left inferior pubic ramus (SUVmax, 5.6) without any CT correlate (F, arrow) are shown. There was no suspicious finding either in 68Ga-PSMA-11 PET/MRI (G, dashed circle) or bone SPECT/CT (H, dashed circle). Therefore, this bone uptake was considered as false-positive.
Statistical Analysis
SPSS Statistics (version 22; IBM Inc.) was used for statistical analyses. The compliance of variables to normal distribution was determined using the Kolmogorov–Smirnov test. Patient characteristics were presented as median (interquartile range [IQR] or range) or mean ± SD in accordance with the data distribution. The χ2 or Pearson goodness-of-fit tests were used to compare the differences of bone metastases and UBU in between 2 PSMA PET agents. A P value of less than 0.05 was considered statistically significant. A Sankey diagram was designed with the online Diagram Generator (Acquire Procurement Services, http://sankey-diagram-generator.acquireprocure.com).
RESULTS
Rate of Reported Bone Metastases and UBU in the Interindividual Group (Comparing 68Ga-PSMA-11 and 18F-PSMA-1007 Cohorts, n = 792)
A total of 792 PSMA PET scans of patients with BCR were included (n = 409 for 18F-PSMA-1007 and n = 383 for 68Ga-PSMA-11) to evaluate the frequency of UBU and bone metastases. Among the patients who were imaged with 18F-PSMA-1007, 332 (81.2%), 33 (8%), 13 (3.2%), 3 (0.1%), and 115 (28.1%) patients underwent radical prostatectomy, definitive radiotherapy, transurethral prostate resection, local ablative treatments, and adjuvant/salvage radiotherapy as previous local therapy, respectively. Among the patients who were imaged with 68Ga-PSMA-11, 324 (84.6%), 28 (7.3%), 7 (1.8%), 1 (0.2%), and 99 (25.8%) patients underwent radical prostatectomy, definitive radiotherapy, transurethral prostate resection, local ablative treatments, and adjuvant/salvage radiotherapy as previous local therapy, respectively. Overall, there was no statistically significant difference for the bone metastases rate when the final reports of 18F-PSMA-1007 and 68Ga-PSMA-11 were compared (72 vs. 64; P = 0.7). Stratifying by PSA value, 229 of 397 (57.7%) patients undergoing 18F-PSMA-1007 and 201 of 360 (55.8%) patients undergoing 68Ga-PSMA-11 PET had PSA levels lower than 1 ng/mL. A fraction of the 138 of 397 (34.8%) patients undergoing 18F-PSMA-1007 and the 147 of 360 (40.8%) patients undergoing 68Ga-PSMA-11 PET had PSA levels between 1 and 5 ng/mL. Thirty of 397 (7.6%) patients undergoing 18F-PSMA-1007 and 12 of 360 (3.3%) patients undergoing 68Ga-PSMA-11 PET had >5 ng/mL of PSA. There was no statistically significant difference of bone metastasis detection between 18F-PSMA-1007 and 68Ga-PSMA-11 among different PSA groups (P = 0.2, 0.2, and 0.6 for PSA levels groups < 1 ng/mL, 1–5 ng/mL, and > 5 ng/mL, respectively) (Fig. 2).
Frequency of bone metastases is presented separately for PSA groups and PET tracers (18F-PSMA-1007 or 68Ga-PSMA-11). There was no statistically significant difference of bone metastasis detection between 18F-PSMA-1007 and 68Ga-PSMA-11 among 3 different PSA level groups. mets = metastasis.
UBU was reported at a significantly higher rate with 18F-PSMA-1007 than it was in the 68Ga-PSMA-11 group (140 [34.2%] vs. 64 [16.7%]; P < 0.001). Moreover, there was at least 1 identifiable benign bone lesion with focal PSMA uptake in 22 (5.4%) and 11 (2.9%) of the 18F-PSMA-1007 and 68Ga-PSMA-11 PET reports, respectively. There was no significant difference between the PSA-level groups and UBU rate for both agents (P = 0.4 and 0.6, respectively, for 18F-PSMA-1007 and 68Ga-PSMA-11).
Patient Characteristics of the 18F-PSMA-1007 and 68Ga-PSMA-11 PET Intraindividual Comparison Cohort (n = 17)
Seventeen prostate cancer patients (mean age, 70.9 y; median duration of disease, 43.7 mo [IQR, 18.6–122.9]) underwent both 68Ga-PSMA-11 and 18F-PSMA-1007 PET due to clinical indication. The median time interval between PET scans was 22 d (IQR 8.0–29.5) days. Most patients were also evaluated with bone scanning and SPECT/CT (n = 14) and whole-body MRI (n = 15), and 12 patients were evaluated with all 4 modalities. All the patients had PSA recurrence after radical prostatectomy, and 8 of 17 patients also had undergone adjuvant or salvage radiation therapy. Twelve of 17 patients had a PSA level lower than 1, and 5 of 17 had PSA levels between 1 and 5 ng/mL. Further characteristics of the patients are outlined in Table 1.
Local recurrence was detected on 18F-PSMA-1007 in 7 (41.1%) of the patients with a median SUVmax of 8.1 (range, 3.48–24.6); 41.1% (7/17) of the patients were rated as pelvic lymph node–positive on 18F-PSMA-1007 PET. The median SUVmax and size of the most prominent pelvic lymph node was 10.9 (range, 3.2–37.6) and 0.5 cm (range, 0.4–1.2), respectively. Moreover, 4 patients were staged as extrapelvic lymph node–positive (n = 2 inguinal and 2 retroperitoneal; median SUVmax = 5.1 [range: 3.4–10.2]) by 18F-PSMA-1007 PET.
Intraindividual Analysis of 18F-PSMA-1007 Bone Uptake by Bone Scanning and 68Ga-PSMA-11 PET/MRI
In 18F-PSMA-1007 PET, 34 suggestive bone uptake findings (in 17 patients) were seen (Supplemental Figs. 1–17 for details on patients). Evaluation of the UBUs and final decisions are summarized in Figure 3. Eleven patients (64.7%) showed unifocal, 4 patients (23.5%) showed oligofocal, and 2 patients (11.8%) showed multifocal 18F-PSMA-1007 bone uptake without any correlative lesion on CT (n = 13 ribs, n = 10 pelvis, n = 4 vertebrae, n = 3 scapula, n = 2 sternum, n = 1 clavicula, n = 1 humerus head). Distribution of the false-positive bone uptake on 18F-PSMA-1007 is presented in Figure 4.
Sankey Diagram summarizing the evaluation of UBUs seen in 18F-PSMA-1007 PET. BS = bone scan; FP = false-positive; TP = true-positive; PSMA = prostate-specific membrane antigen; ⊖ = no suspicious finding; ⨁ = suspicious finding. A total number of 34 UBUs were detected on 18F-PSMA-1007 PET. One lesion was regarded as true-positive (bone metastasis) and 1 lesion was rated as benign because of characteristic MRI findings. Thirty-three UBU were rated as false-positive on 18F-PSMA-1007 PET and 4 instances of false-positive bone uptake were seen on 68Ga-PSMA-11 PET (triple validation was only available in a subcohort).
Anatomic distribution of UBU seen on 18F-PSMA-1007 PET. Thirty-two instances of bone uptake were seen on 18F-PSMA-1007 PET (in multiple regions) and 4 instances of bone uptake were seen on 68Ga-PSMA-11 (all located in ribs). Most common UBU localizations for 18F-PSMA-1007 were ribs and pelvis.
The per-patient true-positive rate was 8.3%, the per-lesion (n = 28) true-positive rate was 3.6%; the positive predictive value of bone uptake seen in 18F-PSMA-1007 PET was 8.3% (95% CI, −7%–23.8%) per patient (n = 12) and 3.6% (95% CI, −3.3%–10.5%) per lesion (n = 28) (only n = 12 patients with all modalities, that is, MRI, bone scanning, and 68Ga-PSMA-11 PET, were included).
One lesion with PSMA expression (SUVmax 6.7 and 3 on 18F-PSMA-1007 and 68Ga- PSMA-11 PET, respectively) in the left ischiopubical junction without any correlative CT lesion was regarded as true-positive because it was also positive on the bone scan and showed contrast enhancement on T1-weighted images with diffusion restriction on MRI (Fig. 1). The patient with true-positive pelvic bone metastasis had a PSA level of 0.91 ng/mL, PSA doubling time of 1 mo, 83 IU/L of alkaline phosphatase, and 21.5 ug/L of bone-specific alkaline phosphatase.
One lesion with PSMA expression (SUVmax 6.1 and 2.3 on 18F-PSMA-1007 and 68Ga-PSMA-11 PET, respectively) without any significant CT correlation was evaluated as enchondroma on MRI (Supplemental Fig. 17). Follow-up examinations of the bone findings are summarized in Figure 3.
All other sites of 18F-PSMA-1007 focal bone uptake were rated as false-positive and likely UBU.
DISCUSSION
In this article, we investigated 18F-PSMA-1007 PET UBU in patients with BCR by 68Ga-PSMA-11 PET, MRI, and bone scanning correlation. In patients with correlative imaging, the positive predictive value of 18F-PSMA-1007 PET for bone metastases was very low. We present a systematic confirmation of 18F-PSMA-1007 PET UBU. However, the higher rate for 18F-PSMA-1007 than for 68Ga-PSMA-11 PET did not translate into more frequent diagnosis of bone metastases if images were read by experienced readers.
PSMA PET has become the reference standard examination of the staging and restaging of patients with prostate cancer (11,12). It was shown previously that PSMA PET is superior to CT and bone scanning in primary staging of patients with high-risk prostate cancer (12). PSMA-11 was assessed in most prospective trials on PSMA-directed imaging, which led to recent Food and Drug Administration approval. Several other PSMA ligands have been studied. For example, 18F-DCF-Pyl showed high diagnostic accuracy and was also approved by the Food and Drug Administration (13). Head-to-head comparison of 18F-DCF-Pyl and 18F-PSMA-1007 revealed near equal tumor detection in a small group of patients with newly diagnosed prostate cancer (14). In France, the ligand 18F-PSMA-1007 is available through expanded access.
PSMA ligands show comparable tumor uptake and distribution, but also have distinctive biodistribution features (5). 18F-PSMA-1007 has a liver-dominant excretion, which offers advantages for the assessment of local prostate cancer infiltration (6). Because of lesser ligand accumulation in the bladder, the differentiation between true tracer uptake and urinary background activity is often easier, which facilitates the detection of local recurrence.
The rise of 18F-PSMA-1007 is mainly caused by the ease of cyclotron-based 18F-fluorine production, which enables the syntheses of larger quantities of PSMA ligands compared with 68Ga generators (4). Additionally, 18F-fluorine offers a longer half-life than 68Ga, enabling an optimized patient management (4). Moreover, the lower positron energy of 18F enables a higher spatial resolution and higher signal-to-background ratio than 68Ga (4).
Despite these benefits of 18F-PSMA-1007, it has been reported that the rate of UBU is notably higher than that of 68Ga-PSMA-11 (8,15). In our study, 33 UBUs have been reported for 18F-PSMA-1007 PET and 4 for 68Ga-PSMA-11 (triple validation was available only in a subcohort). This makes the clear delineation of bone metastases challenging in patient cohorts in which bone metastases have a low prevalence, such as in men with biochemically recurrent prostate cancer at low PSA level. The false-positive assessment of bone uptake may potentially lead to inadequate treatment when anticipating that 18F-PSMA-1007 has the same high specificity of other PSMA ligands.
UBU has also been reported in other PSMA-targeting tracers. For example, preliminary reports indicate that rhPSMA-7 also shows UBU (16). The cause of UBU is not yet known. Unconjugated fluorine, activated bone marrow granulocytes (15), and PSMA expression in nonprostate cancer tissue have been discussed previously (17,18).
Interestingly, UBUs of 18F-PSMA-1007 show a distinct distribution pattern. Especially, uptake in the ribs and pelvis can be observed, yet the explanation for this is unknown. Despite a higher UBU rate for 18F-PSMA-1007, the rate of bone metastases was not different in the cohorts of patients imaged with 18F-PSMA-1007 versus 68Ga-PSMA-11 in patients with BCR. For this, all patients scanned in the year before transition to 18F-PSMA-1007 were compared with all patients scanned in the year after the tracer switch. This observation indicates that experienced nuclear medicine physicians can detect the UBU pattern and identify the lesions as unspecific. The distinctive pattern of UBU at the above-described locations may contribute to this observation. Current knowledge on UBU for 18F-PSMA-1007 and radioligands with similar bone pattern should be summarized in a comprehensive reader training before local implementation of these tracers.
This study comes with limitations. First, the comparisons of patient cohorts before and after the change of PSMA tracers (from 68Ga-PSMA-11 to 18F-PSMA-1007) were analyzed retrospectively. Therefore, the analysis might be prone to selection bias and missing information. In the subgroup of patients undergoing MRI, bone scanning, and 68Ga-PSMA-11 as well as18F-PSMA-1007 PET, the additional PSMA PET and bone scintigraphy were performed only when clinically indicated and after patient approval and the data collection was done retrospectively. Therefore, our cohort with 4 imaging assessments was relatively small, and the results may not be transferable to larger cohorts. Finally, histopathologic confirmation and follow-up imaging were not acquired for this study.
CONSLUSION
In patients with BCR of prostate cancer and PSA ≤ 5 ng/mL, focal bone uptake on 18F-PSMA-1007 PET (SUV > 4) was most often false-positive/UBU when compared with 68Ga-PSMA-11 PET, MRI, and bone scanning. 18F-PSMA-1007 false-positive/UBU findings were most commonly located in the ribs and pelvis. Bone uptake in 18F-PSMA-1007 and 18F radioligands with similar bone pattern should therefore be evaluated carefully with regards to the location and clinical context. Most likely due to reader experience, the rate of bone metastases was not higher when clinical cohorts of patients with BCR imaged with 68Ga-PSMA-11 and 18F-PSMA-1007 were compared. To prevent false bone upstaging and consequently incorrect therapy management of the patients, 18F-PSMA-1007 PET should be performed by experienced physicians with knowledge of UBU distribution pattern and characteristics
DISCLOSURE
Wolfgang P. Fendler reports fees from SOFIE Bioscience (research funding), Janssen (consultant, speakers bureau), Calyx (consultant), Bayer (consultant, speakers bureau, research funding), Parexel (image review), and AAA (speakers bureau) outside of the submitted work. Boris Hadaschik has had advisory roles for ABX, AAA/Novartis, Astellas, AstraZeneca, Bayer, Bristol Myers Squibb, Janssen R&D, Lightpoint Medical, Inc., and Pfizer; has received research funding from Astellas, Bristol Myers Squibb, AAA/Novartis, German Research Foundation, Janssen R&D, and Pfizer; and has received compensation for travel from Astellas, AstraZeneca, Bayer and Janssen R&D. Robert Seifert has received research funding from the Else Kröner-Fresenius-Stiftung and Boehringer Ingelheim Fonds. No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: How clinically relevant is the previously reported occurrence of UBU on 18F-PSMA-1007 PET in prostate cancer?
PERTINENT FINDINGS: Bone uptake seen on 18F-PSMA-1007 PET in patients with BCR, PSA ≤ 5 ng/mL, and SUV > 4 is likely false-positive. Common locations for false positive findings were ribs and pelvis. However, in the clinical routine, the rate of reported bone metastases of patients imaged with 18F-PSMA-1007 or 68Ga-PSMA-11 is comparable, indicating that reporting physicians adapt to the tracer characteristics.
IMPLICATIONS FOR PATIENT CARE: When metastatic disease is suspected in biochemical recurrent prostate cancer, osseous 18F-PSMA-1007 uptake without morphologic correlate has to be carefully assessed.
ACKNOWLEDGMENT
We thank Andrei Iagaru for his support in revising the article.
Footnotes
Published online Dec. 2, 2022.
- © 2023 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication August 13, 2022.
- Revision received November 17, 2022.