[⁶⁸Ga]Ga-PSMA-11 in prostate cancer: a comprehensive review

Abstract

Imaging of the prostate-specific membrane antigen (PSMA) has become an important tool for managing patients with recurrent prostate cancer, and one of the most frequently employed radiopharmaceuticals is [⁶⁸Ga]Ga-PSMA-11. Herein, we summarize the preclinical development and the clinical applications of [⁶⁸Ga]Ga-PSMA-11 and present side-by-side comparisons with other radiopharmaceuticals or imaging modalities, in order to assist imagers and clinicians in recommending, performing, and interpreting the results of [⁶⁸Ga]Ga-PSMA-11 PET scans in patients with prostate cancer.

Prostate cancer

Prostate cancer is one of the leading causes of morbidity and death in men in the Western world, and the second most common cancer in men worldwide. With an ever-aging population, the absolute number of men being diagnosed with prostate cancer is constantly increasing. In 2018, 1,276,106 new cases of prostate cancer were registered worldwide, representing 7.1% of all cancers in men [1]. In certain areas of the world, such as in the UK, more men die from prostate cancer each year than women die of breast cancer [2].

Screening tools for prostate cancer remain limited, primarily by means of prostate specific antigen (PSA) level assessment [3-5], treatment on the other hand has greatly improved in recent years [6]. The latest therapies approved include androgen receptor signaling with abiraterone acetate, enzalutamide and apalutamide, radiotherapy of bone metastases with radium-223 dichloride, immunotherapy with sipuleucel-T, and chemotherapy with taxane-based drugs. If current treatments are built on the synergistic effects obtained when using a combination of the aforementioned therapies, the next major leap forward is expected to stem from developments in the molecular characterization of stage-dependent markers.

Recurrence after primary therapy for prostate cancer occurs in 20 to 60% of cases [7,8], and the 5-year survival rate in patients with high-volume metastatic disease is below 30% [9]. The historical mainstays of clinical examinations remain physical such as digital rectal examination, blood based in the form of PSA or tissue based in the form of trans-rectal or trans-perineal biopsies. However, these modalities present inherent diagnostic limitations.

Digital rectal examination has a positive predictive value between 5 and 30% in patients presenting low PSA values [10], it sometimes fails to identify clinically important prostate cancers, and it displays a high rate of high false positives [11]. Blood markers, such as PSA tend to be non-specific since they may be elevated by non-malignant clinical conditions such as prostatitis and benign prostatic hypertrophy. On the other hand, low PSA does not necessarily rule out the presence of prostatic malignancy [12].

Conventional imaging techniques, such as Computed Tomography (CT) or multi-parametric magnetic resonance imaging (mpMRI), have been used to substantiate the diagnostic value. Given its poor sensitivity and specificity, anatomical diagnosis with CT of the prostate gland has been primarily used to stage the disease once diagnosis has been established. Computed tomography may reveal metastatic spread to pelvic lymph nodes, seminal vesicles, osseous metastases but is inherently based on changes in anatomy, particularly with regard to size. Thus, the failure to provide information pertaining to tumor metabolic activity limits its use to the early stage of disease. In the limited context for lymph node diagnosis, a recent analysis showed an acceptable specificity of 82% but unacceptable sensitivity of only 42% with CT [13].

The use of mpMRI has been increasing in frequency given its higher sensitivity, specificity and predictive value [14]. In addition to the detection of changes in architecture and anatomy of the prostatic gland, this imaging modality gives insights into the potential transformation of the tumor by assessing certain key parameters such as diffusion restriction and is also more accurate than CT in assessing the lymph nodes within the pelvis [15]. For all these reasons, mpMRI is gradually being implemented in the classical clinical workup of Prostate cancer [16].

Ultimately, diagnosis can only be affirmed by pathological assessment, usually with prostatic biopsy, but tissues may be obtained from biopsy material originating from prostatic metastatic foyers.

The prostate specific membrane antigen (PSMA)

In spite of the steady shift toward molecular imaging in clinical diagnostics, clinical imaging modalities for diagnosing cancer and monitoring treatment response have mostly remained at the anatomical rather than molecular level. For example, the historical Response Evaluation Criteria in Solid Tumors (RECIST) criteria [17], which are based on anatomical size, are still considered the reference standard in spite of its mere representation of what is happening at an anatomical size level. The overexpression of Prostate Specific Membrane Antigen (PSMA) in prostate cancer, which increases angiogenesis and increases metabolism of polyglutamated folates and uptake of monoglutamated folates, thus imparting a clear proliferative advantage [18], has been exploited as a molecular marker in the diagnostics of prostate cancer.

PSMA is a 750-amino acid trans-membrane protein found within the apical epithelium of secretory ducts of benign prostatic tissue. While its physiologic role in the prostate remains unclear, its enzymatic role in the cleavage of α-linked glutamate from N-acetylaspartyl glutamate and γ-linked glutamates from polyglutamated folates has been demonstrated [18]. The malignant transformation sees the translocation of PSMA to the luminal surface of the ducts [19] in addition to its overexpression, which is not found in other benign diseases such as prostatic hyperplasia [20].

Several other functions, such as involvement in cellular migration and nutrition, transport and signal transduction, have been attributed to PSMA [21]. Upon binding of a ligand, PSMA is internalized into the cell. In spite of its name, PSMA is not prostate specific as it can be found within lacrimal and salivary glands, the kidneys, liver, spleen and small intestine [22]. Its expression can be detected in tumor associated angiogenesis, glioblastoma, thyroid cancer, gastric, breast, renal and colorectal cancers [22].

PSMA boasts features, which can be exploited as a molecular target for imaging and therapy. Its high level of overexpression (100-1000 fold in 95% of prostate cancer cells) [22] and internalization upon binding [23], lead to enhanced specific uptake and retention, both vital factors for image quality and therapeutic efficacy. In addition, from a disease staging point of view, PSMA expression appears to correlate with advanced disease, castration resistant disease, Gleason score and PSA level [24,25].

PSMA targeting agents

In spite of the typical issues related to antibody-based imaging agents, such as long circulation half-life, low signal to noise ratio and poor target tissue uptake, two monoclonal antibodies (mAb) were developed targeting both the extracellular and intracellular epitopes of PSMA and demonstrated high affinity, specific and efficient targeting in vivo. The murine mAb 7E11 binds an intracellular domain of PSMA and the humanized mAb hJ591 binds to an extracellular domain of PSMA [26]. 7E11 was developed as a theranostic agent with parallel radiolabeling with ([¹¹¹In]In ([¹¹¹In]In-7E11, ProstaScint™) as a potential SPECT imaging agent [27,28], and with [⁹⁰Y]Y ([⁹⁰Y]Y-7E11) as its therapeutic counterpart [29]. The high myelotoxic effect observed with [⁹⁰Y]Y-7E11 ultimately stopped further development while the overall poor sensitivity with ProstaScint™ as a SPECT imaging agent gradually lead to its clinical demise. J591 mAb was clinically investigated for PET/CT imaging as [⁸⁹Zr]Zr-hJ591 [30] and for therapy as [¹⁷⁷Lu]Lu-hJ591 [31].

In parallel, small molecule PSMA-peptide inhibitors, devoid of inherent antibody specific limitations, have been successfully developed and are nowadays the mainstay of current PSMA imaging and therapy modalities [32]. A rational approach was used to develop these agents, with high PSMA affinity and rapid blood clearance as key parameters.

[⁶⁸Ga]Ga-PSMA-11

Currently, [⁶⁸Ga]Ga-PSMA-11 (Glu-NH-CO-Lys-(Ahx)-[[⁶⁸Ga]Ga-HBED-CC] (HBED CC: N,N’-Bis(2-hydroxy-5-(ethylene-betacarboxy)benzyl)ethylenediamine N,N’-diacetic acid, Figure 1) is among the most widely used agents for prostate cancer PET/CT imaging. [⁶⁸Ga]Ga-PSMA-11 belongs to the substance class of peptidomimetic PSMA inhibitors, a class of urea-based PSMA inhibitors first reported in 2001 [33]. Following its initial description and preclinical evaluation in 2012 [34], hasted development yielded the first clinical reports on [⁶⁸Ga]Ga-PSMA-11 PET/CT imaging in 2012 and 2013 [35,36].

Figure 1.

Structure of [⁶⁸Ga]Ga-PSMA-11, with a representation of the two different substructures, HBED-CC for the chelation of [⁶⁸Ga]Gallium and the urea-based pharmacophore for PSMA binding specificity.

[⁶⁸Ga]Ga-PSMA-11 binding affinity

Upon radiolabeling, the size-demanding radiometal complexes often influence the affinity for the targeting molecule by changing the initial lipophilicity and/or charge. In particular in the case of small molecules, the pharmacological property can dramatically be reduced with respect to its binding properties. A study comparing linkers situated between the PSMA binding group 2-[3-(1,3-dicarboxypropyl)-ureido]pentanedioic acid (DUPA) and a ^99mTc complex revealed that linker lipophilicity correlates positively with improved binding properties. A study further supported the idea that the PSMA active site is a pocket containing multiple potential interaction sites.

The pharmacophore was proposed to ideally present three carboxylic groups capable of interacting with the respective side chains of PSMA, one oxygen as part of zinc complexation in the active center and an aromatic structure able to interact with a hydrophobic part of the binding pocket [34,37,38]. These interactions were found to have positive additive effects on binding efficiency, which instigated the successful development of the amphiphilic [⁶⁸Ga]Ga-PSMA-11, a urea-based pharmacophore combined with a [⁶⁸Ga]Ga-HBED-CC metal complex.

[⁶⁸Ga]Ga-PSMA-11 was subsequently competitively analyzed for its binding capacity by performing an enzyme-based assay on rhPSMA (Naaladase-Assay) and a binding assay on LNCaP cells, an androgen-sensitive human prostate adenocarcinoma PSMA positive cell line [34]. The affinity related IC₅₀ and calculated Ki values of both assays were found to be 7.5 ± 2.2 and 12.0 ± 2.8 nM respectively. For comparison purposes, in the same assay, IC₅₀ and calculated Ki for the direct DOTA analog were found to be 19.4 ± 7.1 and 37.6 ± 14.3 nM respectively.

[⁶⁸Ga]Ga-PSMA-11 uptake in LNCaP cells revealed that the HBED complex displays a significant increase in PSMA-specific internalization when compared with its DOTA analog. In a cell uptake experiment, three different concentrations of HBED and DOTA radiolabeled compounds were given to either LNCaP or PC-3 cells, a PSMA negative cell line derived from bone metastasis of grade IV of prostate cancer. Specific uptake in LNCaP was substantially higher for [⁶⁸Ga]Ga-PSMA-11, while unspecific uptake in PC-3 cells was significantly lower when compared to the DOTA analog. Thus, despite nearly identical binding affinity, the presence of HBED and DOTA complexes induce clear differences in specific and unspecific cell uptake.

Influence of diastereoisomers of HBED-CC on [⁶⁸Ga]Ga-PSMA-11 binding

As previously mentioned, the structure of the active site of a PSMA inhibitor consists of two independent main binding sites, a zinc-containing rigid site and a rather lipophilic efferent tunnel [38]. The urea-based portion of PSMA inhibitors typically interacts with the carboxylic groups and the carbonylic oxygen. However, efficient internalization of a PSMA-directed radiotracer requires the interaction of the linker region of the molecule with hydrophobic tunnel region. Due to the specific nature of the interaction, slight chemical differences caused by the known formation of the three different diastereoisomers of HBED-CC after gallium-complexation might influence the binding properties of the whole molecule.

High thermodynamic stability constants are observed for the complexation of gallium with HBED (>10³⁹), which structure is acyclic and requires low energy for complex formation. As a consequence, labeling is fast even at ambient temperature and yields a complex with high kinetic stability at physiological pH [39], in vivo [40] and in human serum for at least 72 hours [41]. These features render HBED extremely attractive as a gallium chelator for high-stability labeling of radiopharmaceuticals. However, in contrast to other clinically radiometal cyclic chelators, HBED-CC can form three NMR-identifiable diastereoisomers (namely RR, RS and SS) during gallium complexation, with RR thermodynamically favored [40]. In spite of the fact that experimental in vitro studies have shown that the two main species observed (RR and RS) have identical binding properties toward PSMA (IC₅₀ values: 24.8 ± 1.2 nM and 27.4 ± 1.3 nM respectively) [38], the presence of two different radioisomers in a variable ratio form batch to batch is unacceptable from a quality control perspective in a clinical setting.

For this reason, in a standard GMP-compliant synthesis labeling protocol, [⁶⁸Ga]Ga-PSMA-HBED-CC is incubated at 85°C to favor the formation of the thermodynamically more stable diastereoisomer RR, but RS is still present in small amount in the labeling reaction mixture even at high labeling temperature. Its presence, however, does not have any significant negative influence on the PSMA-binding properties and therefore on image quality.

[⁶⁸Ga]Ga-PSMA-11 in vivo biodistribution

One hour following tail vein injection of 1-2 MBq [⁶⁸Ga]Ga-PSMA-11 in mice (0.1-0.2 nmol) the animals were sacrificed and their organs of interest were dissected, blotted dry, and weighed. The radioactivity was measured with a gamma counter and calculated as % ID/g [34]. [⁶⁸Ga]Ga-PSMA-11 was cleared rapidly from the blood and PSMA negative tissue. Liver activity represented a mere 0.87 ± 0.05% ID/g as early as one hour after injection. Accumulation in kidney, spleen, and lung uptake was high with 139.4 ± 21.4% ID/g, 17.90 ± 2.87% ID/g, and 2.49 ± 0.27% ID/g respectively, and could be completely blocked to 4.02 ± 1.14% ID/g, 1.54 ± 0.33% ID/g, and 0.64 ± 0.32% ID/g respectively following the co-injection of 2 mg/kg 2-(phosphonomethyl)pentanedioic acid (PMPA), a PSMA inhibitor. Tumor uptake amounted to 7.70 ± 1.45% ID/g on LNCaP and 1.30 ± 0.12% ID/g on PC-3.

To substantiate the claim that reduced kidney accumulation of [⁶⁸Ga]Ga-PSMA-11 after PMPA co-administration is PSMA specific, a side-by side comparison of both D- and L-forms of [⁶⁸Ga]Ga-PSMA-11 was initiated. PET dynamic time−activity curves revealed that the D-form is cleared rapidly from the kidney while [⁶⁸Ga]Ga-PSMA-11 is accumulating with little bladder excretion. This result is most likely linked to the 103-fold difference in PSMA affinity of D-[⁶⁸Ga]Ga-PSMA-11 [34].

[⁶⁸Ga]Ga-PSMA-11 in vivo imaging

MicroPET studies were conducted by injection of 10-25 MBq of [⁶⁸Ga]Ga-PSMA-11 via a lateral tail vein into mice bearing LNCaP tumor xenografts [34]. The anesthetized animals were placed into a small animal PET scanner and 50 min dynamic microPET scans, starting at 1 min post injection followed by a 20 min static scan, were recorded. The organ and tumor uptake value of the [⁶⁸Ga]Ga-PSMA-11 was reflective of in vitro data since [⁶⁸Ga]Ga-PSMA-11 cleared rapidly from the blood and PSMA negative tissues.

Liver activity was limited to only 0.87 ± 0.05% ID/g as early as one hour following injection. Uptake was found to be high in kidney (139.4 ± 21.4% ID/g), spleen (17.90 ± 2.87% ID/g) and lung (2.49 ± 0.27% ID/g). These uptakes were nearly completely blocked down to 4.02 ± 1.14% ID/g, 1.54 ± 0.33% ID/g, and 0.64 ± 0.32% ID/g, respectively, after the co-injection of 2 mg/kg of PMPA. Tumor uptake amounted to 7.70 ± 1.45% ID/g on the PSMA positive LNCaP and 1.30 ± 0.12% ID/g on PSMA negative PC-3 cell lines.

Using a model of monoclonal cell lines, where PSMA expression was differential, but tumor sizes comparable at around 5 mm of diameter, the relationship between absolute surface PSMA target expression of biopsy samples of prostate cancer, and imaging signal with [⁶⁸Ga]Ga-PSMA-11 was assessed in a murine model [42]. The use of PROMISE criteria guided the visual interpretation based on reference organ uptake [43] of [⁶⁸Ga]Ga-PSMA-11.

PET/CT scans were then performed on days 7 and 8, and PSMA expression was quantified on days 7 and 8 by flow cytometry of fine needle aspiration tumor biopsies. In this model, where cell surface PSMA expression was correlated with PET signal, and about 20,000 PSMA molecules per tumor cell surface were identified as threshold for positive PET reading. This threshold is about 10-times lower than the known surface expression in typical human prostate cancer cell lines LNCaP and C4-2 (~190,000 and 240,000 receptors per cell, respectively).

These findings suggest that the threshold for preclinical PET positivity is quite low. On the other hand, while PSMA PET imaging seems to be able to detect small changes in PSMA molecules/cell at a low expression level, this sensitivity disappears at higher PSMA levels, with a mere 1.2-fold PET signal increase for an increase of 22,000 to 45,000 PSMA/cell. Limitations to the accuracy of quantitative PET imaging and the direct value of this side-by-side comparison depends on scanner-specific factors, such as spatial resolution, sensitivity; the characteristics of the radiopharmaceutical, e.g. specific activity (the ratio of radiolabeled and “cold” masses) as well as biological variables, e.g. receptor saturation.

[⁶⁸Ga]Ga-PSMA-11 toxicity

In the absence of regulatory guidelines, the mass amount of [⁶⁸Ga]Ga-PSMA-11 allowed to be injected in humans was, for the longest time, a subject of personal appreciation. However, a circulated draft of a European monography for [⁶⁸Ga]Ga-PSMA-11, indicates a maximum amount of 30 microg per injection.

[⁶⁸Ga]Ga-PSMA-11 dosimetry in humans

The effective dose and organ doses from injection of [⁶⁸Ga]Ga-PSMA-11 in a cohort of low-risk prostate cancer patients [44] was recently reported from an injection with 133-178 MBq of [⁶⁸Ga]Ga-PSMA-11 in a cohort of six patients, followed by PET/CT acquisitions, urine and venous blood collection up to 4h post injection. In this study, [⁶⁸Ga]Ga-PSMA-111 was rapidly cleared from the blood and accumulated preferentially in the kidneys and the liver, and the associated effective dose was 0.022 mSv/MBq. Kidneys and lacrimal glands receiving the highest organ dose, with 40 mGy and 0.12 mGy per MBq administered respectively.

Current joint Society of Nuclear Medicine and Molecular Imaging (SNMMI) and European Association of Nuclear Medicine (EANM) guidelines recommend a dose of 1.8-2.2 MBq (0.049-0.060 mCi) per kilogram (kg) body weight (BW) [45]. A recent attempt to assess image quality with decreased dose revealed a substantial negative impact on image quality and lesion detectability [46].

[⁶⁸Ga]Ga-PSMA-11 biodistribution in humans

The distribution of [⁶⁸Ga]Ga-PSMA-11 is linked to the epithelial expression of the target protein PSMA present in the various tissues and to the physiological excretion of the radiopharmaceutical [47]. Therefore, physiological [⁶⁸Ga]Ga-PSMA-11 uptake is mainly observed in the urinary bladder, the kidneys and the ureters, due to urinary excretion. It is also observed in parotid and submandibular glands due to salivary excretion, in lachrymal glands, and in the colon due to digestive excretion. Finally, it is found in the reticulo-endothelial system, e.g. the spleen and the liver, in the prostate gland, the pancreas, the adrenal glands, and autonomic ganglia. Indicative values of intensity of the activity (SUV_max) of the different tissues and background are summarized in Table 1 (adapted from [47]).

Table 1.

Physiological uptake of [⁶⁸Ga]Ga-PSMA-11 (adapted from [47])

		Median SUVmax	SUVmax Range
Lachrymal gland		7.5	3.0-25.9
Nasal mucosal lining		4.0	1.7-8.8
Parotid gland		16.1	5.5-30.9
Sub-mandibular gland		17.3	7.5-30.4
Liver		6.8	2.8-13.0
Spleen		9.1	3.8-36.7
Kidney		49.6	2.7-97.0
Duodenum		13.8	5.8-26.9
Pancreas	Head	2.9	1.1-7.6
	Body	2.7	1.2-8.6
	Tail	3.3	1.6-8.1
Colon		1.6	0.5-2.7
Blood pool (aorta)		1.8	0.8-3.2
Adrenal glands		1.8	0.6-3.4
Bone marrow (over the iliac bone)		0.7	0.2-1.8
Lymph nodes with fatty hilum		1.8	1.5-2.2
Prostate gland		2.2	1.7-2.9

False positive findings with [⁶⁸Ga]Ga-PSMA-11

The comprehensive pathophysiological mechanism of [⁶⁸Ga]Ga-PSMA-11 uptake is not defined for all tissues. Thus, in addition to the physiological distribution and specific uptake in prostate tumor tissues, also specific uptake in other tissues is known. This uptake can interfere with the image analysis, both in malignant and benign lesions. Therefore, as part of the image interpretation, radiopharmaceutical uptake intensity should be taken in consideration, since signal to background ratio is positively correlated with diagnostic accuracy, for the localization of radiopharmaceutical uptake, with regards to typical tumor drainage territory and for presence of underlying morphological abnormalities.

Clinical use of PET with [⁶⁸Ga]Ga-PSMA-11 and other PSMA radiopharmaceuticals, revealed consistent and significant uptake in minor and major salivary glands [48]. This uptake is still not well understood, but could lie in the biology of the glands themselves, given the prevalence of secretory granules that potentiate formation of radiation-induced free radicals present in this type of tissue. With the steady increase in PSMA radioligand therapies, it is of vital importance to understand the underlying reasons since therapeutic radiations severely damage these glands. External cooling of the salivary glands was initially performed in the clinic with the expectation to reduce uptake due to vasoconstriction. However, the technique ultimately failed to prove relevant in a systematic analysis and probably finds its explanation in the form of local hyper-perfusion to restore the crucial blood supply to the organs near the head [49]. So far, the only autoradiography and immunohistochemistry study [50], focused on the accumulation of PSMA-targeting radioligands in samples of submandibular gland human tissues, recently provided evidence that this accumulation in submandibular gland is not primarily a result of PSMA-mediated uptake.

False positives are related to benign lesions that can mimic distant metastases or lymphatic dissemination. In vitro immune-histochemical expression of PSMA by the autonomous system was confirmed when [⁶⁸Ga]Ga-PSMA-11 increased uptake was shown in ganglia of the autonomic nervous system [51]. Frequently reported locations include celiac ganglia [51] and the sympathetic chains at the cervical, thoracic and sacral level [51-53]. Uptake at the celiac level can be mistaken for retroperitoneal metastases of prostate cancer, and is therefore more challenging to properly diagnose than isolated uptake which, without other pathological foci of uptake in the retroperitoneal or pelvic region, is more likely to be of benign origin [51,52].

Granulomatous inflammatory diseases such as Wegener’s disease [47] and sarcoidosis, with typical mediastino-hilar ganglionic involvement [54-57], can show increased [⁶⁸Ga]Ga-PSMA-11 uptake. Yet, in the latter, bilateral and symmetrical distribution of the radiotracer within mediastino-hilar lymph nodes is expected. In addition, selective endothelial expression of PSMA receptor may result in tracer uptake in pleura and heart valves [58,59].

Secondary bone dissemination of prostatic adenocarcinoma is relatively common. As such, benign bone lesions showing increased [⁶⁸Ga]Ga-PSMA-11 uptake might represent a diagnostic challenge, as reported for osseous hemangioma [60-62], fibrous dysplasia [63], Paget’s disease [64-70] and fractures [71,72]. The uptake observed in osseous and extra-osseous hemangioma is thought to be related to increased lesion vascularization and endothelial cells number. An example of false positive focal bone uptake is shown in Figure 2.

Figure 2.

Prostate cancer staging in a 54 years old patient (Gleason score 4 + 4, PSA value: 8.6 ng/ml) showing two blastic lesions with focal [⁶⁸Ga]Ga-PSMA-11 uptake (maximum Standardized Uptake Value measured at 13). The patient underwent surgery and PSA was undetectable after surgery, proving the non-specific nature of these lesions.

Systemic diseases might also mimic visceral metastases, such as tuberculosis [73] and sarcoidosis, e.g. in the lungs or the liver [56,57]. Nervous system lesions such as meningioma [74,75], schwannoma [76,77], peripheral nerve sheath tumor [78], and cerebral infarction [79,80] have also been reported to exhibit increased uptake. Finally, soft tissue lesions can wrongfully be diagnosed as metastatic sites for prostatic adenocarcinoma since focal uptake has been reported for fascitis nodularis [81], pseudo-angiomatous stromal hyperplasia [82], intramuscular myxoma [83], desmoid tumor [84], acrochordon [85] and dermatofibroma [86].

A summary of the published reports on detection of non-prostatic benign PSMA-avid lesions in the staging/restaging work-up of prostate cancer, i.e. false-positive findings, is provided in Table 2.

Table 2.

Non-prostatic benign PSMA-avid lesions (adapted from [104])

Diagnostic	Reference(s)
Sarcoidosis	[54-57]
Reactivated tuberculosis	[73]
Benign lung opacities and bronchiectases	[174]
Anthracosis	[175]
Paget’s disease	[64-70]
Vertebral body fracture	[72]
Healing sacral fracture	[71]
Benign fibrous dysplasia	[63]
Schwannoma	[76,77]
Meningioma	[74,75]
Peripheral nerve sheath tumor	[78]
Hemangioma	[60-62]
Intramuscular myxoma	[83]
Acrochordon	[85]
Dermatofibroma	[86]
Pseudo-angiomatous stromal hyperplasia of breast	[82]
Desmoid tumor	[84]
Fasciitis nodularis	[81]
Pancreatic serous cystadenoma	[176]
Follicular thyroid adenoma	[177,178]
Lipid-rich adrenal adenoma	[179]
Herniated spleen	[180]
Senile seminal vesicle amyloidosis	[181]
Cerebral infarction	[79,80]

False negative findings with [⁶⁸Ga]Ga-PSMA-11

Prostate cancer lesions lacking increased PSMA expression, leading to false negative findings, have been reported, and can be associated with primary histology and metastatic localization. Immunohistochemistry studies have shown that PSMA-negative primary prostate cancer have a rare occurrence of less than 3% [25,87] and can be correlated with the uptake of [⁶⁸Ga]Ga-PSMA-11 in primary prostate cancers [88]. In addition, neuroendocrine differentiation has been associated with negative PSMA-based imaging [89-91]. Immunohistochemistry also showed that PSMA expression is highest in primary cancer lesions in 88 to 100% of nodal metastases [92], while bone metastases can be negative in up to 15% of cases [25].

[⁶⁸Ga]Ga-PSMA-11 uptake in other malignant lesions

The main reason for specific [⁶⁸Ga]Ga-PSMA-11 uptake in non-prostatic malignancies is the epithelial expression of PSMA linked to neo-vascularization [93,94]. Several types of cancer have already been reported to display [⁶⁸Ga]Ga-PSMA-11 uptake. The histology most commonly reported for its elevated PSMA expression, confirmed by immunohistochemical studies, is renal cell carcinoma [95-102], particularly clear cell renal cell carcinoma, followed by chromophobe renal cell carcinoma [98,103]. A summary of the reported incidental detection of non-prostatic malignant PSMA-avid lesions in the staging/restaging work-up of prostate cancer is provided in Table 3 (adapted from [104]).

Table 3.

Non-prostatic malignant PSMA-avid lesions (adapted from [104])

Diagnosis	References
Follicular lymphoma	[182,183]
Follicular thyroid carcinoma	[184]
Papillary thyroid carcinoma	[97,185]
Hurthle cell adenoma	[185]
Multiple myeloma	[186]
Gastrointestinal stromal tumor	[187]
Pancreatic neuroendocrine tumor	[188]
Hepatocellular carcinoma	[189,190]
Rectal adenocarcinoma	[191,192]
Squamous cell carcinoma of the oropharynx	[193]
Primary lung cancer	[73,194]
Penile squamous cell carcinoma	[195]
Colon adenocarcinoma	[196]
Urothelial carcinoma of ureter	[197]
Renal cell carcinoma	[97,198,199]

[⁶⁸Ga]Ga-PSMA-11 for restaging of prostate cancer

Since measuring sensitivity and specificity for patients with recurrent prostate cancer is limited by the lack of a reference standard, studies often use detection rate as outcome in evaluation the usefulness of [⁶⁸Ga]Ga-PSMA-11 in prostate cancer restaging, considering by definition positive all patients in biochemical recurrence, namely with a PSA above 0.2 ng/mL [105].

The detection rate of [⁶⁸Ga]Ga-PSMA-11 in recurrent prostate cancer has been extensively investigated. Findings from the two largest meta-analyses and a large prospective study are that the detection rate ranges from 74 to 81%, and that the pooled estimated rate of positive scans are correlated with the PSA level [106-109]. Specifically, the rate of positive scans was 42-57% for PSMA levels of 0.2-0.99 ng/mL, 58-84% for PSMA levels of 1.0-1.99 ng/mL, 76% for PSMA levels of 2.0-2.99 ng/mL, and 95% for PSMA levels above 2 ng/mL.

Sensitivity and specificity in the context of recurrent prostate cancer has been measured only in limited patient cohorts using histopathology as gold standard. Here, salvage lymphadenectomy after [⁶⁸Ga]Ga-PSMA-11 imaging of 308 lesions in 28 patients was correlated with 87% per-lesion sensitivity and 93% specificity [110]. [⁶⁸Ga]Ga-PSMA-11 diagnostic performance estimates, using a lymphatic main region-based approach and a subregion-based approach, were derived from for 965 resected lymph nodes in 30 patients [111]. Sensitivity, specificity, negative predictive value, positive predictive value, and accuracy for the main region-based approach were 92%, 100, 100%, 89%, and 96, and for the subregion-based approach 81%, 100%, 99%, 93%, and 94%.

The clinical nomogram, proposed to predict [⁶⁸Ga]Ga-PSMA-11PET/CT positivity in different clinical settings of PSA failure proved good accuracy in predicting a positive scan with values ≥ 40% [112], providing the most informative cutoff in counseling patients to [⁶⁸Ga]Ga-PSMA-11 PET/CT and could be used as an important tool to guide to clinicians in the best use of PSMA-based PET imaging.

In an effort to assess the frequency of [⁶⁸Ga]Ga-PSMA-11 positive lesions outside the standard salvage radiotherapy planning volumes using the Radiation Therapy Oncology Group (RTOG) guidelines, 270 subjects with recurrent prostate cancer after radical prostatectomy and PSA levels < 1 ng/mL were investigated [113]. Fifty-two patients (19%) had at least one [⁶⁸Ga]Ga-PSMA-11-positive lesion not covered by the consensus target volumes, consisting mainly of bone lesions (in 23/52) and perirectal lymph nodes (16/52).

On the other hand, in order to evaluate the impact of [⁶⁸Ga]Ga-PSMA-11 imaging on TNM stage and radiotherapy planning as compared with conventional imaging using bone scan and/or CT or MRI, two cohorts consisting of 11 patients with persistent PSA after radical prostatectomy and 60 with PSA increase after primary treatment were studied [114]. The latter consisted of 23 subjects after RP, 5 after RT and 32 after radical prostatectomy followed by salvage radiotherapy. Respective mean PSA levels were 1.27 ng/mL and 1.1 ng/mL for the two groups. The identification of additional lesions with [⁶⁸Ga]Ga-PSMA-11 scans resulted in a change in TNM stage in 51% and change in radiotherapy plan in 56% of cases. An example of nodal metastasis in a case of biochemical recurrence only detected by [⁶⁸Ga]Ga-PSMA-11 PET imaging is shown in Figure 3.

Figure 3.

Restaging in a 74 years old patient in biochemical recurrence (PSA value: 0.2 ng/ml) showing only one lymph node measuring 4 mm and with a focal [⁶⁸Ga]Ga-PSMA-11 uptake (maximum Standardized Uptake Value measured at 15).

A retrospective review of patients scanned with [⁶⁸Ga]Ga-PSMA-11 for biochemical recurrence following radical prostatectomy with PSA ≤ 2.0 ng/mL was performed to assess if the recurrent disease was within standard radiation target volumes [115]. Through a comparison of patients and clinical variables between men with recurrences covered by standard salvage radiation fields and those with recurrences outside of standard fields, PSMA-avid disease was observed in 53% of patients. For these patients, 38% had PSMA-avid recurrence found outside of the pelvis, 50% lesions confined to the pelvic lymph nodes and prostatic bed, and 12% in the prostate bed only. In addition, salvage radiation including standard Intensity Modulated Radiation Therapy (IMRT) pelvic nodal volumes did not cover PSMA-avid nodal disease in 30% of patients. Therefore, routine use of PSMA-PET imaging in the early salvage setting may potentially lead to treatment optimization by improving target coverage, by using dose escalation to the local or nodal relapse [116-118] or by performing metastasis-directed therapies for oligometastatic patients [119].

A meta-analysis including over a thousand patients showed an overall change in management in 54% of cases (95% CI: 47-60%) following [⁶⁸Ga]Ga-PSMA-11 imaging [120]. In particular, in the population of patients with recurrent prostate cancer, there was an increase in the proportion of patients treated with curative approaches including radiotherapy, surgery, focal therapy and multimodal treatment, and reduction of patients treated solely with systemic medications or untreated.

Ongoing prospective phase III trials randomizing between standard salvage radiotherapy with or without a restaging PSMA PET/CT [121] will certainly help to determine in the near future if molecular imaging can improve outcome in patients with early biochemical relapse after radiotherapy.

[⁶⁸Ga]Ga-PSMA-11 for initial staging of prostate cancer

The excellent diagnostic performance of [⁶⁸Ga]Ga-PSMA-11 in restaging motivated its investigation also in the initial staging of the disease, namely in patients at high risk for metastatic disease. Multiple studies suggested high diagnostic accuracy also in this indication [122-124]. An example of metastatic nodal and bone spread at staging detected by [⁶⁸Ga]Ga-PSMA-11 is shown in Figure 4. This was recently confirmed by a prospective randomized multicenter study assessing the impact of [⁶⁸Ga]Ga-PSMA-11 PET for initial staging of high-risk prostate cancer prior to curative treatment, compared with conventional imaging by CT and bone scanning. On the basis of these findings, PSMA PET/CT should be the imaging modality of choice in the primary staging of high-risk prostate carcinoma, given the superior accuracy as compared with conventional imaging, combined with lower overall radiation exposure and higher reporter agreement. PSMA imaging can also be used to guide radiotherapy treatment of oligometastatic de novo prostate cancer [119], the next investigational step in the management of low burden synchronous disease after the evidence of an overall survival benefit of a local radiotherapy [125]. Of note, ongoing clinical trial such as the EORTC 1414 PEGASUS trail (ClinicalTrials.gov Identifier: NCT02799706) already implement modern imaging techniques in the curative treatment of de novo oligometastatic prostate cancer patients.

Figure 4.

Prostate cancer staging in a 68 years old patient (Gleason score 4 + 3, PSA value: 13.5 ng/ml) showing local disease associated with multiple nodal (red arrows) and bone (green arrows) metastatic lesions. PET imaging induced a change in management towards docetaxel and androgen deprivation.

[⁶⁸Ga]Ga-PSMA-11 versus other radiopharmaceuticals and imaging modalities

[⁶⁸Ga]Ga-PSMA-11 vs. [¹⁸F]F-, [¹⁸F]F-Methyl-, [¹⁸F]F-Ethyl- or [¹¹C]-choline

[⁶⁸Ga]Ga-PSMA-11 was first used in humans in 2011 [35], and shortly thereafter became the new PET imaging reference standard for prostate cancer, as it clearly outshone [¹⁸F]F-choline, the primary clinical diagnostic PET radiopharmaceutical of the time, or some of its analogs such as [¹⁸F]F-Methylcholine, [¹⁸F]F-Ethylcholine or [¹¹C]choline. A clear superiority was demonstrated in various aspects of side-by-side comparisons.

Parallel injections of [⁶⁸Ga]Ga-PSMA-11 and [¹⁸F]F-choline, [¹⁸F]F-Methylcholine, [¹⁸F]F-Ethylcholine or [¹¹C]choline were performed to assess their respective lesion detection performance in several studies. Overall significant superior diagnostic performance of [⁶⁸Ga]Ga-PSMA-11 was consensual [110,126-129]. [⁶⁸Ga]Ga-PSMA-11 also allowed systematical identification of more lesions at lower PSA values than [¹⁸F]F-choline [126,127], as summarized in Table 4.

Table 4.

Detection rate of [⁶⁸Ga]Ga-PSMA-11 vs. [¹⁸F]F-choline in prostate cancer

PSA level (ng/mL)	[68Ga]Ga-PSMA-11 vs. [18F]F-choline (%)
	Ref [126]	Ref [126]	Ref [127]
<0.5		50/12
<1			61/46
<2		71/36
1-2			81/66
<2.82	69/44
>2.82	100/90
>2		88/63	97/89
>5

In a prospective study of prostate cancer patients with biochemical relapse, histology of all lesions indicated by imaging was performed [110]. Patients underwent [¹⁸F]F-Ethylcholine and [⁶⁸Ga]Ga-PSMA-11 PET scans. All patients with positive lymph nodes on imaging were submitted to pelvic and/or retroperitoneal lymphadenectomy. Per-lesion analysis showed an accuracy of 92% (95% CI, 88%-95%) for [⁶⁸Ga]Ga-PSMA-11 versus 82% (95% CI, 88%-95%) for [¹⁸F]F-Ethylcholine. The negative predictive value (NPV) was 97% (95% CI, 93%-98%) for [⁶⁸Ga]Ga-PSMA-11 versus 88% (95% CI, 84%-92%) for [¹⁸F]F-Ethylcholine. There was a clear trend towards higher sensitivity, specificity and negative predictive value. Per-patient, there was a positive predictive value of 82% for [⁶⁸Ga]Ga-PSMA-11 and 79% for [¹⁸F]F-Ethylcholine.

Side-by-side comparison of uptake of [⁶⁸Ga]Ga-PSMA-11 with [¹⁸F]F-choline showed a higher value for [⁶⁸Ga]Ga-PSMA-11 in 79% of lesions, lower in 15% and equal in 5% of all cases [126]. Tumor-to-background ratio was clearly superior in 95% of lesions with [⁶⁸Ga]Ga-PSMA-11 as increased uptake observed with [¹⁸F]F-choline, can be hampered by relatively high background activity. The most significant differences observed between the two radiopharmaceuticals regarding tumor uptake, and even more when it comes to tumor-to-background ratio, were observed in lymph node metastases followed by the bone lesions, local recurrences and soft tissue metastases.

Patient management after [⁶⁸Ga]Ga-PSMA-11 and [¹⁸F]F-Methylcholine imaging [128] was impacted in 63% of cases overall, 54% based on [⁶⁸Ga]Ga-PSMA-11 results alone and 29% on [⁶⁸Ga]Ga-PSMA-11 and [¹⁸F]F-Methylcholine as equals. Patients with early biochemical relapse after radical prostatectomy are usually treated with salvage radiotherapy of the prostatic bed even in the absence of imaging findings. However, in the patient cohort of this study, 75% of the [⁶⁸Ga]Ga-PSMA-11 positive patients with low PSA had disease outside the prostatic bed.

Among a pool of bone lesions detected by imaging from a cohort of 103 patients, a per-lesion-analysis showed that 62% were identified by both [⁶⁸Ga]Ga-PSMA-11 and [¹¹C]choline, 36% were visible with [⁶⁸Ga]Ga-PSMA-11 solely and 2% with [¹¹C]choline alone [129]. Overall, the 98% detection rate observed with [⁶⁸Ga]Ga-PSMA-11 was significantly higher as the 64% detection rate of [¹¹C]choline. The per-patient-analysis revealed that 31% of lesions were detected by both radiopharmaceuticals, 3% by [⁶⁸Ga]Ga-PSMA-11 alone, and 1% by [¹¹C]choline alone.

For distant metastases, [⁶⁸Ga]Ga-PSMA-11 and [¹¹C]choline seem to have complementary roles. [⁶⁸Ga]Ga-PSMA-11 detected significantly more patients with N1 stage in a cohort of patients with local lymph node metastases detected by imaging [129]. Thereby, 70% were classified N1 by both radiopharmaceuticals, 25% additional patients were upstaged to N1 after [⁶⁸Ga]Ga-PSMA-11 imaging, and 1.5% were positive with [¹¹C]choline alone. For distant metastases, staging was in agreement with both radiopharmaceuticals in 77% and discordant in 11% of all cases. Patients were upstaged after [⁶⁸Ga]Ga-PSMA-11 PET scan in 6% of all cases as compared to 5% on [¹¹C]choline results alone.

Regarding oligometastatic disease, a significant difference between the results with [⁶⁸Ga]Ga-PSMA-11 and [¹¹C]choline was observed since 16% considered oligometastatic with [¹¹C]choline alone were found to have more than 3 metastases with [⁶⁸Ga]Ga-PSMA-11. On the other hand, 4% of patients deemed oligo-metastatic with [⁶⁸Ga]Ga-PSMA-11 were found to be multi-metastatic with [¹¹C]choline. Overall, 45% of the oligometastatic patients were identified with both radiopharmaceuticals, while 35% were found to have more than 3 lesions by both compounds.

Regarding initial staging, significantly more suspicious lymph nodes and bone lesions were detected by [⁶⁸Ga]Ga-PSMA-11 compared to [¹¹C]choline in the lesion-based analysis (P<0.004), but without a significant difference in the patient-based analysis (P=0.625).

[⁶⁸Ga]Ga-PSMA-11 vs. [¹⁸F]F-DCFPYL

[⁶⁸Ga]Ga-labelled PSMA-radiopharmaceuticals have been systematically phased out by fluorine-18-labeled analogs given the advantages provided by [¹⁸F]Fluoride as compared to [⁶⁸Ga]Gallium: key features are the longer half-life (109 min vs. 68 min), the cyclotron produced large centralized batches (vs. generator-produced [⁶⁸Ga]Gallium), and the lower positron energy in favor of spatial resolution and reduced blurring effects.

Introduced in 2015, [¹⁸F]F-DCFPYL is a front runner [¹⁸F]F-labeled candidate for targeting PSMA with PET in the clinic. Systematic head-to-head comparison of the number of lesion positive results obtained, as compared to [⁶⁸Ga]Ga-PSMA-11, in 14 prostate cancer in biochemical relapse was performed [130]. Outcome measures, such as number of detected PSMA-positive lesions, tumor uptake value (SUV_max) and lesion to background ratio were assessed. All suspicious lesions identified by [⁶⁸Ga]Ga-PSMA-11 were also detected with [¹⁸F]F-DCFPYL while in three patients, the latter allowed identifying additional lesions. [¹⁸F]F-DCFPYL also significantly outperformed [⁶⁸Ga]Ga-PSMA-11 in the mean SUV_max measures (14.5 vs. 12.2, P=0.028), as well as mean tumor to background ratio. However, the differences in SUV_max were only found to be significant with the use of kidney, spleen, or parotid as reference organs (P=0.006, P=0.002, P=0.008), but not using the liver (P=0.167) or the mediastinum (P=0.363).

[⁶⁸Ga]Ga-PSMA-11 vs. [¹⁸F]F-fluciclovine

Since 2016, [¹⁸F]F-fluciclovine (Axumin^®, Blue Earth Diagnostics Ltd.) is the only PET imaging agent approved by the FDA in the US in the limited context of localization of recurrent prostate cancer. It is deemed “usually appropriate” by the American College of Radiology Appropriateness Criteria in the post-prostatectomic follow-up of prostate cancer patients, and after nonsurgical pelvic and local treatment in case of concern for recurrence.

Head-to-head comparison studies are still limited [131-133] and the relative values of each imaging modality were debated [134,135]. [⁶⁸Ga]Ga-PSMA-11 demonstrated overall higher rates detection but with high variability between cohorts and depending of sites of recurrent cancer. The key advantage of [¹⁸F]F-fluciclovine lies in its capacity to detect localized foyers in close anatomical relation to the urinary bladder, an area where the accumulation of [⁶⁸Ga]Ga-PSMA-11 hinders the detection. On the other hand, [⁶⁸Ga]Ga-PSMA-11 alone was able to detect recurrences in bone, other organs and extra-pelvic lymph node sites.

[⁶⁸Ga]Ga-PSMA-11 vs. [¹⁸F]F-PSMA-1007

[¹⁸F]F-PSMA-1007, another [¹⁸F]F-PSMA targeting agent has been recently introduced in the clinic. In addition to the aforementioned advantages provided by fluorine-18-fluoride, its key advantages lie in its rapid blood clearance combined with minimal urinary excretion. Both features yield clear advantages for local tumor assessment, as high radiopharmaceutical retention in the bladder and ureters is known to impair image interpretation.

102 patients with biochemical recurrent prostate cancer after RP were matched based on various clinical variables patients with corresponding [⁶⁸Ga]Ga-PSMA-11 scans [136]. In doing so, fluorine-18-PSMA-1007 PET revealed approximately 5 times more lesions attributed to benign origin compared to [⁶⁸Ga]Ga-PSMA-11 PET. Highest frequencies were observed in ganglia, unspecific lymph nodes and bone lesions with 43%, 31%, 24% with fluorine-18-PSMA-1007 and 29%, 42%, 27% with [⁶⁸Ga]Ga-PSMA-11.

In addition to the number of detected lesions, the SUV_max of lesions attributed to benign origin was also significantly higher (P<0.0001) with [¹⁸F]F-PSMA-1007 PET (5.3 with a range of 3.0-42.7 vs. 4.4 with a range of 2.8-7.5 respectively). Further, a similar number of lesions was attributed to recurrent prostate cancer, 124/369 lesions for [¹⁸F]F-PSMA-1007 PET and 126/178 lesions for [⁶⁸Ga]Ga-PSMA-11 PET. Therefore, in spite of key advantages, the considerable higher number of lesions with increased PSMA-ligand uptake attributed to benign lesions, as compared to [⁶⁸Ga]Ga-PSMA-11 PET, emphasizes the need for paramount reader training and caution with [¹⁸F]F-PSMA-1007 as a prostate cancer imaging agent in the clinical context.

[⁶⁸Ga]Ga-PSMA-11 vs. MRI

Multiparametric pelvic MRI is considered to be the standard imaging modality for staging and restaging local occurrence as well as for the detection of pelvic nodal metastases in prostate cancer patients. Several studies have assessed the respective performance of mpMRI and [⁶⁸Ga]Ga-PSMA-11, including multiple PET/MRI hybrid studies.

Initial staging of patients with prostate cancer is paramount in the therapeutic decision-making. A number of studies [137-143] have assessed the diagnostic performance of [⁶⁸Ga]Ga-PSMA-11 compared with conventional imaging in this context, especially for lymph node assessment and finally to evaluate management impact.

Comparison of [⁶⁸Ga]Ga-PSMA-11 PET/CT with mpMRI for loco-regional prostate cancer staging in patients who were candidates for RP, using histopathology as reference standard, showed that PSMA PET/CT provided superior detection of prostate cancer lesions than mpMRI. For primary staging, another study focused on patients with high-risk prostate cancer, compared mpMRI combined with diffusion weighted whole body MRI to [⁶⁸Ga]Ga-PSMA-11 imaging [144]. PET imaging allowed identifying nodal pelvic and extra-pelvic lesions as well as skeletal, liver and lung lesions that were not identified on MRI. However, the results obtained did not add value for T staging [144,145]. Importantly, these results were counterbalanced by another study which found no significant differences between mpMRI and [⁶⁸Ga]Ga-PSMA-11 for nodal staging in a series of 42 patients [146].

A side-by-side comparison of the diagnostic accuracy and inter-rater agreement of mpMRI and [⁶⁸Ga]Ga-PSMA-11 PET/MRI for the detection of extracapsular extension and seminal vesicles infiltration was recently reported [145]. Both modalities performed equally for local staging of prostate cancer in patients with intermediate- to high-risk prostate cancer, since the slightly reduced specificity of [⁶⁸Ga]Ga-PSMA-11 PET/MRI for the detection of extracapsular extension offset its increase in sensitivity.

When lesion volume estimate on imaging and histopathology were compared, both mpMRI and [⁶⁸Ga]Ga-PSMA-11 PET showed good diagnostic performance, with a significant improvement when combining the areas identified as pathological on the two modalities [147,148]. In the detection of local recurrence, mpMRI holds a significant advantage for local lesions over [⁶⁸Ga]Ga-PSMA-11 PET, as excretion in the bladder reduces the aforementioned ability to detect focal uptake in the prostatic bed [149].

In the limited context of high-intensity focused ultrasound treatment of localized prostate cancer [150-153], patient follow-up typically includes mpMRI along with biopsy, which, in the post-interventional setting, often yields false-negative results. A study, aimed at investigating if [⁶⁸Ga]Ga-PSMA-11 was used to localize recurrent disease in a cohort of 10 PET/MR patients with positive template biopsy and negative mpMRI after high-intensity focused ultrasound [154]. Predictive values of PET/MRI for sensitivity, specificity, and positive and negative were 55%, 100%, 100% and 85%, respectively. In addition, patient-based PET/MRI was negative in 40% of cases with Gleason scores 3 + 4 and a tumor length between 0.1 and 3 mm and all lesions with Gleason scores 4 + 3 or higher were detected on PET/MRI. Taken together, these results indicate that [⁶⁸Ga]Ga-PSMA-11-PET/MR has the potential to localize prostate cancer recurrence after high-intensity focused ultrasound occult on mpMRI.

Whole body MRI is an emerging image modality for the detection of bone metastasis in patients with prostate cancer, mainly in case of bone marrow lesions, while sclerotic lesions might less visible [155]. However, for the detection of bone metastases [156,157], the accuracy of [⁶⁸Ga]Ga-PSMA-11 was shown to be significantly higher, with 100% vs. 80% [156] and 90% vs. 63% [157].

[⁶⁸Ga]Ga-PSMA-11 vs. bone scan

Bone scanning, with [^99mTc]Tc-labeled disphosphonates or [¹⁸F]F-NaF, is a reference imaging modality for the evaluation of bone metastases, namely in prostate cancer. The use of [⁶⁸Ga]Ga-PSMA-11 imaging both in staging and restaging has shown an incidence of bone metastases higher than expected with conventional imaging on the basis of PSA levels and disease stage [158], motivating direct comparison studies with bone scan.

Multiple groups have consistently reported the superior diagnostic accuracy of [⁶⁸Ga]Ga-PSMA-11 over technetium-99m-based bone scan [159,160]. The comparison with fluorine-18-sodium fluoride, on the other hand, did not show a clear superiority of one modality over the other [161,162], suggesting that the superior spatial resolution and sensitivity provided by the PET technology are a key factor in bone lesion detection.

As benign bone lesions might exhibit moderate [⁶⁸Ga]Ga-PSMA-11 binding (see paragraph on false positive findings), reporting recommendations based on the absolute uptake value or relative uptake as compared with the physiologic uptake in other organs have been proposed [43,163,164].

Summary

From an imaging point of view, [⁶⁸Ga]Ga-PSMA-11 PET/CT is unquestionably one of the most useful tools for the therapeutic management of patients with prostate cancer in the clinical setting in 2020 and foreseeable future. When compared with other imaging modalities, [⁶⁸Ga]Ga-PSMA-11 targeted imaging appears to offer higher sensitivity along with higher levels of specificity as examplified in Figure 5. The sensitivity of radiopharmaceuticals targeting PSMA generally correlates positively with serum PSA levels, [⁶⁸Ga]Ga-PSMA-11 PET/CT follows this pattern and performs relatively well at low PSA levels. Head-to-head comparisons with other molecular agents, such as [¹¹C]-choline or [¹⁸F]F-PSMA-1007 in patients with biochemically recurrent prostate cancer, proved that [⁶⁸Ga]Ga-PSMA-11 shows on-par or superior overall performance (Figure 2 includes representative in-house images of head-to-head direct comparisons of different imaging modalities or radiopharmaceuticals).

Figure 5.

Upper panel left: [⁶⁸Ga]Ga-PSMA-11 PET (left) vs. CT (right): Specific PSMA Uptake in a nodal recurrence in a millimetric size lymph node; Upper panel right: [⁶⁸Ga]Ga-PSMA-11 PET vs. MRI: Specific PSMA Uptake not detected by MRI imaging. Lower panel left: [⁶⁸Ga]Ga-PSMA-11 PET vs. technetium-99m-MDP SPECT performed 2 months apart, [⁶⁸Ga]Ga-PSMA-11 PET shows a higher number of metastatic bone sites. Lower panel right: [⁶⁸Ga]Ga-PSMA-11 PET vs. [¹¹C]choline of a local recurrence.

Other clinical applications of [⁶⁸Ga]Ga-PSMA-11 imaging are already being considered, namely as imaging tool to guide targeted treatment. Intraprocedural detection of local [⁶⁸Ga]Ga-PSMA-11 uptake might facilitate biopsies or surgery. Several studies [137,140,165-167] suggest that [⁶⁸Ga]Ga-PSMA-11 PET/CT or PET/MRI guided prostate biopsy could have an added value, namely in patients with contraindications to or negative multi-parametric MRI and could contribute to the optimization of the diagnostic/therapeutic algorithm with benefits for patients. In addition, the in-situ detection of small sub-centimeter nodal metastases was reported during PSMA-radio-guided surgery, during which additional lesions, not detected with preoperative [⁶⁸Ga]Ga-PSMA-11 PET/CT and close to known tumor deposits, were identified [168].

Dose escalated radiotherapy protocols have been demonstrated to improve the long-term biochemical control of prostate cancer patients. Focal boosts to the dominant intraprostatic lesion have been investigated as treatment strategy to improve disease control and optimize treatment-related side effects [169]. Noteworthy, complementary information in the definition of the target volume has been observed by IMRT dose escalation on the gross target volume based on the combination of mpMRI and [⁶⁸Ga]Ga-PSMA-11 PET/CT imaging [170,171]. Dose painting by boosting the gross target volume-union resulted in an estimated higher tumor control probability with no or minimal increase of normal tissue complication probability.

Last but not least, [⁶⁸Ga]Ga-PSMA-11 PET has been shown to increase consensus with histopathology compared to mpMRI for intraprostatic gross target volume delineation [172]. Therefore, use of [⁶⁸Ga]Ga-PSMA-11 PET finds an interest in the treatment planning of salvage therapies for a local relapse after a primary radiotherapy [173], including PSMA-dose painting stereotactic radiotherapy to the intraprostatic focal recurrence.

Disclosure of conflict of interest

None.