PET Imaging in Urology: A Rapidly Growing Successful Collaboration

Andrea Farolfi; Samantha Koschel; Declan G. Murphy; Stefano Fanti


Curr Opin Urol. 2020;30(5):623-627. 

In This Article

Abstract and Introduction


Purpose of review: To discuss and highlight the recent findings in urological oncology focusing on nuclear medicine advances on imaging and therapy.

Recent findings: Testicular tumors: 18F-FDG as the standard positron emission tomography (PET) tracer with proven good accuracy in detecting metastatic testicular cancer; urothelial cancer: good accuracy of 18F-FDG PET in detecting distant metastases but poor results in detecting local disease; prostate cancer: prostate-specific membrane antigen (PSMA) is a promising target for imaging prostate cancer with unprecedented accuracy in both staging and restaging and prospective studies were recently published. In castration-resistant prostate cancer, PSMA-targeting radionuclide therapy is showing potential as a curative possibility (e.g. using 177Lu-PSMA); renal cell cancer (RCC): besides FDG other PET radiotracers are under clinical evaluation (e.g. antibody-based molecular imaging, metabolic radiotracers and PSMA-based tracers). PSMA-based imaging may have applications in staging clear-cell RCC and in the selection and in the antiangiogenic treatment's response assessment. Possible role of PSMA-targeting radionuclide therapy?

Summary: In urological oncology, the use of 18F-FDG has been limited by a generally low tumor uptake and physiological 18F-FDG excretion by the urinary system. Other radiotracers are increasing the urologist's portfolio allowing imaging of several biochemical pathways. Theragnostic possibilities are also under investigation thanks to PSMA-based tracers.


Imaging of the urinary tract has long been a fascination of the urologist. Urologists were among the first physicians to introduce imaging into daily practice; however, until a few years ago, nuclear medicine did not play an essential role in the evaluation of urological malignancies. The only available and approved procedure was 111In-capromab pendetide scintigraphy for prostate cancer imaging, whereas 99Tc-bone scan is a long established, but poorly performing standard for the evaluation of bone metastases in prostate cancer. Urologists have joked in the past about the role of 'unclear' rather than 'nuclear' medicine in genitourinary oncology. The employment of positron emission tomography (PET) combined with computed tomography (CT) or MRI imaging has opened up an exciting new era for genitourinary oncology with a range of opportunities because of different radiotracers (Table 1). It is very clear that the nuclear medicine physician is now an integral part of the genitourinary oncology multidisciplinary team.[21] In this special edition of Current Opinion in Urology, we present a series of articles to overview the role of PET imaging in genitourinary oncology.

Testicular Tumors

Testicular tumors can be assessed using 18F-FDG, able to discriminate between the presence of metabolically active tumor cells from fibrosis/necrosis after therapy. Regarding initial staging, PET has been shown to identify more extensive disease than conventional imaging, avoiding unnecessary treatments.[1] Reported sensitivity, specificity, and accuracy were 70, 100, and 93% respectively in a study enrolling 46 patients with stage I nonseminomatous germ cell tumors, whereas negative predictive value was 92% for PET and 78% for standard staging imaging procedures.[2] The clinical impact of 18F-FDG PET was evaluated retrospectively by Ambrosini and colleagues in a cohort of 51 patients with seminoma and 70 patients with nonseminoma.[3] PET showed good sensitivity and specificity for seminoma lesion detection, i.e. 92% and 84% respectively, but its sensitivity was 77% for nonseminomatous tumors and specificity 95%. Additionally, the PET result influenced the clinical management of 92% of patients with seminoma and of 84% of patients with nonseminoma. A consolidated indication for PET is the imaging of residual masses after chemotherapy. In a meta-analysis including five studies and 130 patients with seminoma, 18F-FDG PET proved to be superior to standard imaging in predicting viable residual disease.[4] Positive and negative predictive value were 70 and 93% for PET and 28 and 86% for standard imaging, respectively. In another meta-analysis of nine studies and 375 patients with seminoma, the pooled analysis provided the following results: sensitivity 78%, specificity 86%, positive predictive value 58%, negative predictive value 94%, and accuracy 84%.[5] A better diagnostic accuracy in evaluating residual/recurrent lesions more than 3 cm compared to those less than 3 cm was also demonstrated. Considering nonseminomas, 18F-FDG PET ability to predict viable residual disease was evaluated in a prospective trial with 45 patients.[6] PET was compared to CT scans and changes in serum tumor markers. Sensitivity for PET versus CT versus serum changes was 59 versus 55 versus 42% and specificity was 92 versus 86 versus 100%. Therefore, the identification of recurrent disease is the key role for PET imaging in testicular tumors for 30 years even if disease staging and assessment of treatment response are other everyday indications.

Urothelial Cancer

The physiological urinary excretion of 18F-FDG is a major limitation for studying urothelial cancers and PET is not employed routinely for evaluation of primary tumor. Nevertheless, 18F-FDG PET has a consolidated role in staging or restaging distant metastases and a meta-analysis showed that pooled sensitivity and pooled specificity were 82 and 89% respectively for staging or restaging metastatic lesions of bladder cancer.[7] In order to overcome 18F-FDG intrinsic limitations for urothelial cancer imaging, other PET tracers have been proposed. In a prospective study of 14 patients before radical cystectomy and excision of pelvic lymph nodes, 11C-acetate and 11C-choline PET/CT were performed within 1 week. The two tracers showed equivalent results, importantly a high negative predictive value for lymph node involvement with the potential in selecting patients who would benefit from a more tailored treatment.[12]

Prostate Cancer

Several radiotracers have been proposed in the imaging of prostate cancer, namely 18F-FDG, 11C-choline, 18F-choline, 11C-acetate, 18F-acetate, 18F-fluciclovine, dihydrotestosterone-based radiotracers such as 16b-18F-fluoro-5a-dihydrotestosterone, and radiolabeled bombesin receptor antagonists that targets gastrin-releasing peptide receptors and other receptor-binding molecules. Since 2012 however, year of the first PET image obtained with a prostate-specific membrane antigen (PSMA) ligand, a multitude of trials are assessing this new family of radiotracers in the majority of clinical settings of prostate cancer.


First promising data about PSMA-ligand imaging were presented in retrospective or single-center studies, without direct comparison with conventional imaging. In the first multicenter randomized trial (proPSMA) published in 2020 and enrolling 302 patients with high-risk features, Hofman et al.[16] showed that PSMA-ligand PET offers greater accuracy than conventional imaging (i.e. CT and bone scan): 92% for detecting pelvic nodal or distant metastases compared with 65% for conventional imaging. PSMA-ligand PET led to management changes in 28% of patients compared with 15% for conventional imaging. Nevertheless, PSMA imaging still seems to have a confined role for intraprostatic detection of prostate cancer.[22]

Biochemical Recurrence

PSMA-ligand PET detects prostate cancer with unprecedented accuracy also after radical treatments even at very low PSA levels.[17,18,23] Accordingly with the results of a prospective single-center single-arm trial where 50 patients with prostate cancer with early biochemical recurrence (PSA range between 0.2 and 2.0 ng/ml) underwent 18F-fluciclovine and PSMA-ligand PET within 15 days, PSMA-ligand PET demonstrated a higher detection rate compared to 18F-fluciclovine PET, that is 56 and 26% respectively at a patient level.[15] Another single-arm prospective multicenter trial published in 2019 demonstrated high detection rate, positive predictive value, inter-reader reproducibility and safety for PSMA-ligand PET.[24] In detail, Fendler et al. enrolled 635 patients with biochemically recurrent prostate cancer after radical treatment and found a positive predictive value of 84% by histopathologic validation and 92% by composite reference standard whereas detection rate for PSA values less than 0.5 ng/ml was 38%.

Advanced Prostate Cancer

Despite 18F-FDG is the most widely employed radiotracer, its role in prostate cancer is limited because of the habitual low FDG avidity of prostate cancer cells in contrast with the greater number of other neoplasms. However, 18F-FDG potentially provide useful information in poorly differentiated prostate cancer, suggesting an adjuvant role in castration-resistant prostate cancer and a prognostic role for patients. The cell surface protein PSMA, increasingly expressed in prostate cancer compared to other tissues, is a promising target not only for imaging but also for a therapeutic approach in a theragnostic concept. It refers to the possibility of combining a diagnostic biomarker with a therapeutic agent and finds its first clinical application in metastatic thyroid cancer employing 131-iodine. Currently, the β-emitting radioisotope 177-Lutetium has gained popularity as the therapeutic radionuclide of choice for the treatment of neuroendocrine tumors and prostate cancer. The first prospective phase II trial in prostate cancer was published in 2018 by Hofman et al. (LuPSMA trial).[9] Their cohort involved 30 patients with castration-resistant prostate cancer heavily pretreated (87% received one or more lines of prior chemotherapy) and more than 50% PSA reduction was achieved in 57% of patients. Interestingly, median PSA progression-free survival and overall survival were 7.6 and 13.5 months, respectively. The evidence generated by this trial and multiple retrospective studies led to two currently underway randomized trials in metastatic castrate-resistant prostate cancer, that is TheraP (NCT03392428) and VISION (NCT03511664).

Renal Cell Carcinoma

PET does not play a major role in local diagnosis of renal cell carcinoma (RCC), as most of the radiotracers, for example 18F-FDG, have a urinary excretion. Furthermore, renal parenchyma has baseline 18F-FDG uptake with an overall inadequacy in differentiating malignant lesions. RCC is the most common malignant kidney tumors and 18F-FDG PET is widely employed for metastases imaging despite the possibility of false negative results because of the clear cell subtype of the disease. In a meta-analysis pooled sensitivity and specificity was 91 and 88% for the detection of extra renal lesions, respectively.[11] However, most of the included studies were PET scans without CT. Interestingly, in a bicentric retrospective study with 104 patients with RCC after surgery, positive versus negative 18F-FDG PET result was associated with worse cumulative survival rates over a 5-year period.[12] Moreover, 18F-FDG PET influenced therapeutic strategy in 43% of patients. Several PET radiotracers have been proposed besides 18F-FDG, both metabolic tracers and ligands for specific proteins. 11C-acetate was proposed in 1995 to detect RCC in 18 patients.[14] The authors found that renal uptake of 11C-acetate is prompt and without any urinary excretion. More interesting, they reported that RCC had 11C-acetate uptake similar to the normal tissue, but that the rate of clearance was significantly lower in cancer cells allowing for ready differentiation of RCC from nonneoplastic renal tissue on images acquired beyond 10 min of radiotracer administration. Additionally, 11C-acetate PET was proposed for evaluating sunitinib response, a multitargeted tyrosine kinase inhibitor with antiangiogenic properties for RCC lesions.[15] Antibody-based molecular imaging (immune-PET) has shown potential for RCC imaging improving and the most promising results have been reported employing girentuximab-based radiotracers. The chimeric antibody cG250 (girentuximab) binds with carbonic anhydrase IX, a cell-surface antigen highly and homogeneously expressed in more than 95% of clear-cell RCC.[21] In the multicenter REDECT trial published in 2013, 195 patients with clear-cell RCC underwent 124I-girentuximab PET/CT showing an average sensitivity if 86% for PET/CT and 76% for contrast-enhanced CT and an average specific of 86% for PET/CT versus 47% for contrast-enhanced CT.[25] More recently, in a prospective multicenter study published in 2019, 42 newly diagnosed metastatic clear-cell RCC with good or intermediate prognosis underwent CT, 18F-FDG PET/CT, and 89Zr-DFO-girentuximab PET/CT at baseline.[26] A total of 449 lesions were identified by at least one modality and with the addition of 89Zr-DFO-girentuximab PET/CT lesion detection by CT alone increased from 56 to 91% and was better than CT combined with 18F-FDG PET/CT. PSMA-ligand PET is an emerging tool for RCC imaging and particularly for clear-cell RCC. This subtype is characterized by a considerable neovascularization resulting in a high PSMA expression on the cell surface of the tumor microvasculature.[20] PSMA-ligand PET is not useful for primary staging because of physiological expression in the proximal convoluted tubules and renal excretion of the radiotracer resulting in the high uptake seen in the kidney parenchyma.[20,27] In a prospective pilot study enrolling 10 patients, 86 PSMA-avid lesions were identified with 36 samples being available for analysis. A total of 35/36 samples were positive for RCC deposits, whilst 1 sample was inconclusive for diagnosis on biopsy. For the histologically confirmed lesions, there were no false-negative PSMA-ligand PET lesions; however, CT was false negative in 11.[28] On the other hand, eight patients with metastatic nonclear cell RCC underwent PSMA-ligand PET and employing CT or MRI as reference demonstrated that only 14% of lesions had high radiotracer avidity.[29] In a series of 257 patients with RCC, immunohistochemistry revealed low PSMA expression for nonclear cell RCC but demonstrated increased PSMA expression with increased grade and stage, metastatic disease and aggressive phenotype.[30] Finally, in a theragnostic scenario, PSMA-targeting therapeutic radiopharmaceuticals might be another therapeutic option for RCC which deserve to be explored.