Prostate Embryology and Anatomy

During the embryonic period in a 50‐mm male fetus, under the influence of testosterone produced by the Leydig cells of the fetal testis, the prostate gland starts to develop as epithelial buds from the walls of the urogenital sinus at the site of the Mullerian tubercle [1]. Resembling the shape of an inverted pyramid, the prostate gland base is located below the neck of the urinary bladder and the apex is located inferiorly at the level of the urogenital diaphragm. The normal adult prostate gland measures approximately 5 × 3 × 3 cm with a volume of 25 mL or less. The gland consists of four zones: the peripheral zone (PZ) is the outermost zone, comprising 70–80% of the glandular tissue, the transition zone (TZ) wraps around the urethra proximal to the verumontanum comprises 5% of the glandular tissue, the central zone (CZ) surrounds the ejaculatory ducts and comprises 20% of the glandular tissue, and the anterior fibromuscular stroma (AFMS) does not contain any glandular tissue. The TZ is the main site of benign prostatic hyperplasia accounting for a higher proportion of the glandular tissue. In contrast, most (70%) prostate cancers arise from PZ, followed by TZ, and rarely CZ. Prostate cancer does not arise from AFMS, and the presence of prostate cancer in CZ or AFMS is usually the result of extension from PZ or TZ.

Zonal anatomy might be appreciated on computed tomography (CT) scan with a more hypodense appearance of PZ. Prostate gland zonal anatomy is best delineated on magnetic resonance imaging (MRI), with PZ showing a higher T1‐weighted signal than TZ due to its greater water content. The normal CZ is visualized on T2‐weighted and diffusion‐weighted images as symmetric low‐signal areas surrounding the ejaculatory ducts bilaterally. The ejaculatory ducts join the prostatic urethra at the mid‐gland level of the prostate at a urothelial elevation called the verumontanum. AFMS is seen as a bilateral crescentic‐shaped low‐signal area on T2‐weighted imaging [2].

The term “prostate capsule” has been colloquially used, since it is a pseudo‐capsule formed by the fibromuscular tissue. The prostate pseud‐capsule is often seen as a rim of T2 hypointensity surrounding the gland. It fuses anteriorly with AFMS at the apex with the external urethral sphincter and at the base with the bladder detrusor. The posterior surface of the gland is separated from the rectum by a membrane called the rectovesical (Denovillier’s) fascia. The paired seminal vesicles are located superior and posterior to the gland and are seen as multicystic structures on MRI. The paired neurovascular bundles run at the posterolateral aspects of the gland at 5 and 7 o’clock, and comprise branches of inferior vesicular artery, accompanying veins and nerves. The venous plexus of Santorini surrounds the prostate anteriorly and anterolaterally. The prostatic venous complex drains to the paravertebral venous plexus and the inferior vesicular vein. The communication between the prostatic and prevertebral venous complexes is responsible for the high frequency of vertebral metastasis from prostate cancer.

Prostate Diseases

Benign prostatic hyperplasia (BPH) is the most common symptomatic disease of the prostate gland, with a prevalence of 50–60% by the age of 60. BPH occurs as the result of glandular and stromal proliferation, and often leads to urinary tract symptoms [3]. The diagnosis is clinical and a variety of medical and interventional treatment options are available.

Prostatitis is a general term which encompasses four distinct clinical entities: acute bacterial prostatitis, chronic bacterial prostatitis, inflammatory or noninflammatory pelvic pain syndrome, and asymptomatic prostatitis. Overall prevalence is 8.2%, the diagnosis is clinical, and imaging is rarely requested for diagnosis [4].

Prostate cancer is the leading cause of malignancy and the second leading cause of cancer‐related mortality in men in the United States. Twelve‐core trans‐rectal ultrasound biopsy (TRUS) is typically performed if there is clinical suspicion for malignancy based on elevated or rising serum prostate specific antigen (PSA) level or abnormal digital rectal examination (DRE). Developments in the multiparametric MRI of the prostate have revolutionized the work‐up of prostate cancer, with MRI‐guided biopsy methods now widely used in clinical practice.

The prostate cancer grading system has long been based on the greatest Gleason score on pathology. A new prostate cancer grading system was introduced [5] by the International Society of Urological Pathology (ISUP) in 2014 which assigns Gleason groups 1–5 derived from the more traditional Gleason scores. This grading system has been validated in large cohorts and is now widely accepted [6, 7]. Cohort studies have found that the cancer‐related mortality is low in patients with low‐grade prostate cancer, and they do not benefit from invasive treatments. These patients are typically followed with active surveillance, while those with unfavorable intermediate‐grade or high‐grade prostate cancer require more extensive diagnostic workup and treatment planning. Radical prostatectomy, external beam radiation therapy or brachytherapy is used for the treatment of localized nonmetastatic prostate cancer. After curative‐intent treatment, patients undergo regular follow‐up with serum PSA to detect possible biochemical recurrence (BCR), which occurs in about 40% of men over a decade post initial definitive therapy (Table 19.1). When higher serum PSA level (>10 ng/mL) is associated with BCR of treated prostate cancer, CT scan and bone scan are employed. In lower serum PSA levels, the sensitivity of these conventional imaging studies decreases markedly, necessitating the use of more sophisticated modalities. Early and accurate detection of the sites of recurrent disease is of paramount importance for patient management and treatment, since local salvage therapy is used for pelvic recurrence and systematic treatments are used for metastatic disease [10, 11]. Locoregional salvage therapies are most effective with serum PSA <1 ng/mL [12]. Metastasis‐directed therapy (MDT) may also be considered in patients with oligometastatic disease (typically up to five lesions identified on imaging). Comprehensive guidelines for the initial diagnosis, management, and treatment of prostate cancer can be found at the National Comprehensive Cancer Network (NCCN) website [13].

Ultrasound and Ultrasound‐guided Biopsy

With a distended urinary bladder, the prostate gland can be grossly assessed on pelvic ultrasound with a curved array transducer. On ultrasound, BPH is identified with visualization of an enlarged prostate (volume >30 mL), heterogeneous TZ, bladder wall hypertrophy and trabeculation, and post‐micturition urine residue in the urinary bladder.

Gray‐scale transrectal ultrasound (TRUS) was introduced in 1968, but despite the development of higher frequency transducers and improved signal reception, it is still not an optimal modality for detection of prostate cancer at early stages, with many prostate cancer foci appearing isoechoic on TRUS. Color Doppler ultrasound improves the diagnostic performance of gray‐scale TRUS. In 1989, the first TRUS‐guided systematic sextant biopsy protocol was described, and gradually the 12‐sysmetic biopsy turned into the standard of care for the diagnosis of prostate cancer. Biopsy is performed after initiation of prophylactic antibiotics (fluroquinolones by choice, a widely used regimen is 500 mg of ciprofloxacin 1 day prior to biopsy and for 2 days after) and under local anesthesia. Contrast‐enhanced TRUS and elasto‐sonography can augment visualization of suspicious foci for targeted TRUS biopsy. Biopsy is carried out using a guided biopsy system attached to the transrectal probe.

CT Scan

Overall, CT scan is an insufficient modality for assessment of prostatic disease. Prostate volume measurement might be overestimated on CT scan. In daily practice, a prostate gland with a transverse diameter of greater than 4.5 cm in the axial plane is considered enlarged. Prostate calcifications are frequently seen on CT scan, without clear clinical significance. In the setting of BPH, enlarged prostate gland, heterogenous appearance of TZ with visible high‐ or low‐density nodules, and, in severe cases, protrusion of the gland into the urinary bladder might be seen. Bladder wall thickening and trabeculation are indirect clues for the presence of BPH. With acute prostatitis, the prostate gland becomes enlarged and edematous, and if a prostate abscess develops, a rim‐enhancing unilocular or multilocular hypodense area in the PZ is expected on contrast‐enhanced CT scan.

CT scan is limited for the primary detection of prostate cancer, but focal increased enhancement in the PZ of prostate gland is described in some patients with prostate cancer. In high‐grade disease or when there is suspicion for locoregional or distant metastasis, chest and abdominopelvic CT scan are commonly used for detection of nodal and soft tissue metastasis, and planning for radiotherapy.

Multiparametric MRI and MRI‐directed Biopsy

MRI is not routinely used for the diagnosis of benign prostate conditions. MRI, however, is superior to ultrasound for classification of BPH subtypes and estimation of glandular/stromal ratio, findings which can help to tailor treatment. On MRI, chronic prostatitis and granulomatous prostatitis can cause signal abnormality and mimic prostate cancer [14].

The potential application of MRI in prostate cancer was first described in 1982 [15]. With remarkable advances in image acquisition techniques, postprocessing software, and standardized interpretation guidelines during the past decade, multiparametric prostate MRI (mpMRI) has turned into the imaging modality of choice when assessment of intraprostatic malignancy or image‐guided prostate biopsy is indicated. Prostate mpMRI is generally recommended when there is either (i) clinically high suspicion for intermediate or high‐grade prostate cancer (based on serum PSA measures and findings of DRE) in patients who had a prior negative transrectal ultrasound guided biopsy or (ii) restaging or re‐biopsy planning in patients with known low‐grade prostate cancer under active surveillance [16]. The American Urologic Association currently does not recommend mpMRI for the purpose of prostate cancer screening or surveillance [17].

Adapted from a similar scoring system in breast cancer, the European Society of Urogenital Radiology (ESUR) introduced the first Prostate Imaging‐Reporting and Data System (PI‐RADS) in 2012, which included guidelines for prostate MRI acquisition, interpretation, and reporting [18]. PI‐RADS version 1 (v1) was succeeded by recommendations of a steering committee from the American College of Radiology (ACR), ESUR, and the AdMeTech Foundation, formulated as PI‐RADS v2. in 2016 [19] followed by PI‐RADS v2.1 in 2019 [16]. Since their introduction, PI‐RADS guidelines have been widely validated and right now are widely applied to daily practice in the United States and Europe. PI‐RADS guidelines assess the probability of clinically significant prostate cancer on a five‐point scale for each lesion on a standard prostate mpMRI (Table 19.2).

For the acquiring prostate mpMRI, most experts recommend using a 3 T MRI scanner when available. Compared to 1.5 T scanner, a 3 T scanner is more prone to the susceptibility artifact, which might compromise the image quality (particularly in diffusion‐weighted imaging [DWI]), and therefore in specific situations (such as the presence of metallic hip prosthesis) a 1.5 T scanner might be preferred. An external phased array coil is used with or without an endorectal coil. An endorectal coil increases signal‐to‐noise ratio (SNR), particularly in larger patients, but it increases the time and cost of the scan, and is associated with patient discomfort. Rectal evacuation before examination is recommended. The use of antispasmodic agents (such as glucagon and hyoscine bromide) to decrease bowel peristalsis and rectal gas is controversial.

Prostate multiparametric MRI initially included T1‐weighted imaging (T1WI), T2‐weighted imaging (T2WI), DWI with its corresponding apparent diffusion coefficient (ADC) map, dynamic contrast‐enhanced imaging (DCE), and magnetic resonance spectroscopy (MRS). With the widespread use of DWI, and based on the recommendations of PIRADS v.2, MRS has now fallen out of favor. For interpretation of prostate mpMRI, the PZ and TZ should be separately scrutinized for the presence of focal lesions, with the assessments primarily based on DWI/ADC at PZ and T2WI at TZ. DCE has a secondary role in PZ lesions with PI‐RADS score 3 (Figure 19.1). T1‐weighted imaging is not part of the PIRADS criteria but it is useful for identification of intraprostatic hemorrhage and for detection of pelvic lymph node or bone involvement. The details of PI‐RADS guidelines for the acquisition, interpretation, and reporting of prostate mpMRI are available at the American College of Radiology website (https://www.acr.org/‐/media/ACR/Files/RADS/Pi‐RADS/PIRADS‐V2‐1.pdf).

In a meta‐analysis of 21 studies (3857 patients), PIRADS v2 had pooled sensitivity and pooled specificity of 89% and 73%, respectively, for detection of prostate cancer, using prostatectomy or prostate biopsy as the reference standard [20]. Prostate mpMRI has significantly higher sensitivity and negative predictive value for the detection of clinically significant prostate cancer compared to TRUS biopsy, which indicates that acquiring prostate mpMRI could potentially avoid biopsy in up to 27% of men at risk [21]. Systematic TRUS‐guided biopsy commonly oversamples clinically insignificant disease and has a limited role for the diagnosis of prostate cancer in TZ. Based on the results of the PRECISION trial (which included 500 biopsy‐naïve patients), performing MRI with or without MRI‐targeted biopsy results in 12% higher detection of clinically significant prostate cancer and 13% lower detection of clinically insignificant cancers compared to TRUS‐guided biopsy [22].

MRI‐targeted biopsy is usually indicated with prostate lesions PIRADS categories 3, 4, and 5. Several approaches are used for performing MRI‐targeted biopsy of prostate without an absolute consensus on the preferred method. MRI‐targeted prostate biopsy might be performed via three approaches: MRI‐ultrasound fusion biopsy, cognitive MRI‐guided biopsy, and direct in‐bore MRI‐guided biopsy. In MRI‐ultrasound fusion biopsy, the previously acquired MRI data are fused on the real‐time TRUS on a fusion platform and sampling is performed under the tracking system. In cognitive MRI‐guided biopsy, the operator mentally maps (i.e. cognitively fuses) the previously acquired MRI target to the real‐time TRUS and guides the needle to this location. Finally, in‐bore MRI‐guided biopsy is performed under direct MRI guidance while the patient is lying in the MRI gantry.

Overall, with MRI‐targeted approaches there is a significantly higher detection rate of clinically significant prostate cancer (relative risk [RR] 1.16) and a significantly lower detection rate of clinically insignificant cancer (RR 0.47) over TRUS‐guided biopsy [23]. There is not any significant difference in detection of clinically significant prostate cancer between MRI‐ultrasound fusion biopsy, cognitive MRI‐ultrasound fusion biopsy, and direct in‐bore MRI‐guided biopsy. Direct MRI‐guided biopsy, however, allows direct visualization of suspicious lesions with higher core positivity rate (47.7%) compared to cognitive MRI‐guided biopsy (33.3%) and MRI‐ultrasound fusion biopsy (31.3%) [24]. Cost‐effectiveness studies have found both MRI‐based approaches to have higher quality‐adjusted life‐year benefits [25, 26]. The optimal biopsy approach should be chosen per patient and per center basis, considering the availability of biopsy platforms and the experience of the performing operators.

Fluorodeoxyglucose

Fluorodeoxyglucose (FDG) is a synthetic glucose analog that is taken up by glucose transporters (GLUTs), phosphorylated by hexokinase, and trapped in the cells. Most malignant tumors demonstrate enhanced glucose metabolism, increased hexokinase enzymatic activity, and upregulation of GLUTs (Warburg effect) compared to normal tissue. With relatively low glucose metabolism in the majority of prostate cancers, FDG PET/CT has a limited role in the primary detection, staging, and restaging of prostate cancer [27]. Castration‐resistant prostate cancer is typically FDG avid, and therefore FDG PET/CT might be useful for assessment of its treatment response and prognostication (time to hormonal treatment failure in castrate‐sensitive disease and overall survival in castrate‐resistant metastatic disease) (Figure 19.2) [28]. PET Response Criteria in Solid Tumors (PERCIST) 1.0, particularly in combination with PSA response criteria [29], sum of SUVmax, and the number of FDG‐avid lesions [30], can independently provide prognostic information in this group of patients.

Incidental focal intraprostatic FDG uptake is nonspecific but may carry a risk of malignancy. The prevalence of incidental high intraprostatic activity is 1.8%, with at least 17% risk of malignancy [31]. Some investigators have suggested that focal incidental intraprostatic FDG activity with SUV more than 6 should be further assessed by prostate mpMRI or serum PSA level [32].

Choline

Choline, as a component of the cytoplasmic membrane phospholipid layer, is phosphorylated by the enzyme choline kinase and is trapped in the cell membrane in the form of phosphatidylcholine [33]. The upregulation of choline kinase in prostate cancer is the most likely mechanism leading to enhanced expression of choline in prostate cancer. The most available radiolabeled choline ligands are ¹¹C‐choline, ¹⁸F‐fluoroethylcholine, and ¹⁸F‐fluoromethylcholine (latter two are commonly known as ¹⁸F‐fluorocholine or FCH). Normal biodistribution of ¹¹C‐choline demonstrates relatively high accumulation in the pancreas, liver, kidneys, and salivary glands, and variable uptake in the bowel, with minimal urinary excretion (latter only with ¹¹C‐choline) [34].

The short half‐life (~20 minutes) of ¹¹C limits its utility to institutions with an on‐site cyclotron. ¹⁸F‐labeled choline radiotracers have a longer half‐life (~110 minutes) and allow transportation from off‐site cyclotrons. With a shorter position range of ¹⁸F‐labeled choline ligands, the spatial resolution of images improves, but the higher urine activity with ¹⁸F‐choline in contrast to ¹¹C‐choline is disadvantageous. Physiologic ureteric or urinary bladder activity might limit assessment of prostate bed or pelvic nodal involvement. With ¹¹C‐choline, imaging starts immediately after injection, while with ¹⁸F‐choline imaging starts at, most commonly, 60 minutes post injection. When ¹⁸F‐choline is used, additional post‐micturition images may help distinguish pathologic activity from physiologic urine activity.

Choline PET/CT is not indicated for the detection of primary prostate cancer or routine initial staging. Tracer uptake in prostatic intraepithelial neoplasia and benign conditions such as prostatitis and BPH lowers its specificity for assessment of intraprostatic malignancy. ¹⁸F‐fluorocholine PET/CT has slightly higher sensitivity and specificity compared to ¹¹C‐choline. Choline PET/CT is more accurate than conventional imaging for detection of pelvic nodal disease at initial staging with pooled sensitivity and specificity of 62% and 92%, respectively [35].

When there is biochemical recurrence of treated prostate cancer, choline PET/CT has higher accuracy than conventional imaging modalities; however, the disease detection is highly influenced by serum PSA level and the sensitivity of the scan drops sharply, with 80% of scans being negative at PSA levels <1 ng/mL [36]. In a meta‐analysis of 2126 patients with biochemical recurrence of treated prostate cancer, the pooled detection rate of ¹¹C‐choline was 62%. The pooled sensitivity and specificity in 1270 patients were both 89% [37]. On September 12, 2012, ¹¹C‐choline was approved by the United States Food and Drug Administration (FDA) for imaging of patients with suspected prostate cancer recurrence and noninformative bone scan, CT, or MRI.

Acetate

Acetate, a key substrate in energy metabolism, provides the only carbon source for fatty acid synthesis. Fatty acid synthesis is enhanced in prostate cancer cells as a response to increased demand for energy and cell membrane lipid production [38]. The tracer primarily accumulates in the myocardium, kidneys, pancreas, spleen, and bone marrow. An advantage of prostate imaging, is that urinary bladder activity is absent as a result of active reabsorption of ¹¹C‐acetate in the renal proximal convoluted tubule [39].

¹¹C‐acetate PET is overall less investigated than ¹¹C‐choline PET. In a similar fashion to choline‐based imaging, ¹¹C‐acetate PET cannot reliably distinguish between BPH and prostate cancer, and therefore is not a practical modality for localization of intraprostatic, specifically low‐volume disease [38]. Some studies have shown that ¹¹C‐acetate PET may be less sensitive than ¹¹C‐choline PET for the detection of pelvic lymph nodes [40]. The diagnostic performance of ¹¹C‐acetate PET/CT in patients with biochemical recurrence of prostate cancer is similar to that of choline PET/CT with low yield when serum PSA level is under 1–2 ng/mL [38]. In a study of 358 patients with biochemical recurrence of prostate cancer, the detection rate varied from 19% in those with a PSA level between 0.2 and 1 ng/mL, to 46% in those with a PSA level between 1 and 3 ng/mL, and to 82% in those with a PSA level of more than 3 ng/mL [41].

¹⁸F‐fluciclovine

Amino acid metabolism and transport is upregulated in certain types of cancers, including prostate cancer. L‐¹¹C methionine (MET) was one of the first tracers used for this purpose but it was not widely investigated and it is not currently used in clinical practice. Anti‐1‐amino‐3‐¹⁸F‐fluorocyclobutane‐1‐carboxylic acid (¹⁸F‐FACBC, also known as ¹⁸F‐fluciclovine or commercially as Axumin; Blue Earth Diagnostics, Inc.) is a synthetic L‐leucine analog. This synthetic amino acid is transported into cells via specific amino acid transporters, such as alanine‐serine‐cysteine transporter 2 (ASCT2). ASCT2 is an essential transporter of glutamine, an amino acid which is upregulated in prostate cancer and plays an important role in the cancer signaling pathway. Fluciclovine does not undergo further intracellular metabolism and is not incorporated into protein structure [42]. Tracer activity is highest in the pancreas, with less activity in the liver, salivary glands, pituitary gland, marrow, lymphoid tissue, breast, gastrointestinal (GI) tract, adrenal glands, and renal parenchyma, with minimal urine activity (Figure 19.3a) [43].

Due to rapid influx and efflux of amino acids in prostate cancer, the peak activity happens very early after tracer injection, and therefore image acquisition should begin 3–5 minutes after injection to enhance the target‐to‐background ratio. The uptake is mainly assessed visually rather than by measuring SUV [44]. Table 19.3 provides a summary of ¹⁸F‐FACBC PET/CT scan interpretation. Additional comprehensive guidelines can be found in the literature.

Fluciclovine is most beneficial in individuals with biochemical recurrence of treated prostate cancer, but it has low sensitivity for nodal staging and is not recommended or approved for initial staging of prostate cancer. A metanalysis of 13 studies (563 patients) showed that fluciclovine PET/CT has pooled sensitivity and specificity of 87% and 84%, respectively, for the detection of primary prostate cancer, pooled sensitivity and specificity of 56% and 98%, respectively, for nodal staging, and pooled sensitivity and specificity of 79% and 69%, respectively, for localization of recurrent disease [45]. Fluciclovine is superior to choline PET/CT for the detection of intraprostatic, lymph node, and bone lesions in biochemical recurrence after radical treatment of prostate cancer [46]. In a limited evaluation, prostate ¹⁸F‐fluciclovin PET/MRI was found to have high specificity but low sensitivity for lymph node detection in high‐risk prostate cancer before undergoing radical prostatectomy, and lymph nodes smaller than 8 mm might not show tracer activity [47]. On May 27, 2016, ¹⁸F‐FACBC was approved by the United States FDA for imaging prostate cancer in patients with suspected recurrence based on rising serum PSA after prior definitive primary treatment. With remarkable advances in PSMA‐based tracers, ¹⁸F‐FACBC has started to fall out of favor (Figure 19.4).

Prostate‐Specific Membrane Antigen Ligands

Prostate‐specific membrane antigen (PSMA), also known as glutamate carboxypeptidase II or folate hydrolase I, is a type II transmembrane glycoprotein (100–120 kDa) which is mostly localized to the intracellular compartment in normal prostate gland cells, but in 90% of prostate cancers it highly expresses in the extracellular luminal compartment [48]. PSMA expression is 100–1000 times higher in density in prostate cancer compared to normal non‐neoplastic prostatic tissue [49], and the expression enhances as the cancer becomes castrate‐resistant. PSMA expression in prostate cancer is probably related to its role in folate transportation and metabolism, which further facilitates cell proliferation [50]. The fraction of prostate cancers that do not express PSMA are typically more aggressive. PSMA has shown unique characteristics which have made it a practical molecular target for the development of small‐molecule labeled inhibitors.

The first commercially available PSMA targeting antibody was ¹¹¹In‐ capromab‐pendetide (ProstaScint, AYTU Bioscience, Englewood, USA), which was approved by the United States FDA in 1996 for single‐photon emission computed tomography (SPECT) imaging of biopsy‐proven prostate cancer localized to the prostatic bed but at high risk for pelvic lymph node metastasis. Unlike the modern PSMA ligands, ProstaScint binds only to the intracellular epitope of PSMA. ProstaScint showed moderate sensitivity for localization of primary and metastatic prostate cancer even with added SPECT/CT, and with emerging of more sensitive radiotracers it is no longer used in clinical practice. The development of PSMA ligands dates to as early as 2001, with the first clinical experiences reported between 2012 and 2015. The most studied and currently FDA‐approved PSMA‐targeted ligands are ¹⁸F‐DCFPyl, and ⁶⁸Ga‐PSMA‐11 [51]. Another notable tracer is ¹⁸F‐PSMA‐1007, which has the advantage of predominant hepatobiliary excretion and minimal urine activity [52].

Extraprostatic PSMA expression is noted in the salivary glands, liver, spleen, kidneys, neural ganglia, duodenal mucosa, and other parts of the GI tract. Slight differences in uptake distribution exist between ⁶⁸GA‐PSMA‐11 and ¹⁸F‐DCFPyL (Figure 19.3b,c), with ¹⁸F‐DCFPyL showing higher liver uptake and lower renal, spleen, and salivary glands uptake [53]. One of the most common pitfalls in the interpretation of PSMA‐based PET is the ganglia uptake, which might mimic pathologic lymph nodes. Commonly visualized ganglia on PSMA‐targeted PET/CT are cervicothoracic/stellate, coeliac, lumbar, and sacral ganglia (Figure 19.5). Correlation to the typical anatomic location and the more linear/triangular shape of ganglia on CT scan is helpful. PSMA expression is also described in other nonprostate pathologies, including, but not limited to, pulmonary sarcoidosis, Wegner’s granulomatosis, benign neurogenic tumors (i.e. meningioma, neurofibroma, paraganglioma), benign bone diseases (e.g. fibro‐osseous disease, Paget’s disease), gynecomastia, and a few nonprostate malignancies (e.g. renal cell carcinoma, hepatocellular carcinoma) [54].

PSMA PET/CT is consistently found to outperform other molecular and nonmolecular imaging modalities, with overall more accurate initial staging, higher detection rate of lymph node metastasis, and higher lesion uptake with lower background activity (Figures 19.6 and 19.7) [55]. ⁶⁸Ga‐PSMA‐11 is more sensitive than ¹¹C/¹⁸F‐choline and ¹⁸F‐fluciclovine in detecting recurrence at PSA levels <2.0 ng/mL [42]. PSMA ligands have moderate‐to‐high sensitivity and very high specificity for lymph node detection, and have enabled localization of positive lymph nodes in stations which are otherwise overlooked, including deep posterior pelvic, mesorectal, and supraclavicular stations. The mesorectum/posterior pelvic area is not usually included in the lymph node dissection or the radiation field unless the involvement is known prior to the treatment (Figures 19.8 and 19.9) [56].

Imaging usually begins 60 minutes after injection of PSMA ligand, however with ¹⁸F‐DCFPyL a 2‐hour and with ⁶⁸Ga‐PSMA‐11 a 3‐hour post‐injection image might improve tumor localization [56]. ¹⁸F labeling carries several advantages over ⁶⁸Ga labeling, including longer half‐life (110 vs. 68 minutes) and more availability of cyclotron‐produced ¹⁸F than ⁶⁸Ga, which needs a germanium generator. Higher image resolution and lower urinary excretion with ¹⁸F‐labeled ligands are also advantageous [57].

In a meta‐analysis of 37 studies (4790 patients with BCR of prostate cancer), the detection rate of ⁶⁸Ga‐PSMA PET/CT scan at PSA categories 0–0.19, 0.2–0.49, 0.5–0.99, 1–1.99, and ≥2 ng/mL was 33%, 45%, 59%, 75%, and 95%, respectively [58]. According to the results of the CONDOR trial, ¹⁸F‐DCFPyL PET/CT results in a correct cancer localization rate of 84.8–87% with a change in the initially intended management plan in 63.9% of patients with BCR of prostate cancer [59]. ¹⁸F‐DCFPyL might allow detection of additional lesions, specifically at serum PSA levels <0.5 ng/mL, in comparison with ⁶⁸Ga‐PSMA‐11 [60]. European Association of Urology (EAU) guidelines on prostate cancer recommend the use of PSMA PET imaging for any case of biochemical recurrence after radical prostatectomy (PSA >0.2 ng/mL) [61] On December 1, 2020, ⁶⁸Ga‐PSMA‐11 was approved by the United States FDA, followed by the approval of ¹⁸F‐DCFPyL on May 27, 2021. According to the approval statements, both agents can be used for the initial staging of prostate cancer with suspected metastasis in patients who are considered for initial definitive therapy or in the setting of suspected recurrent prostate cancer based on the rising serum PSA after definitive radiotherapy or radical prostatectomy. A multidisciplinary workgroup has published appropriate use criteria which describe the clinical appropriateness of PSMA PET imaging in different scenarios of prostate cancer [62].

Several guidelines have attempted to standardize the interpretation and reporting of PSMA PET/CT. Based on Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE), the molecular imaging TNM (miTNM) version 1.0 was proposed in 2018. miTNM staging combines the anatomic distribution of the disease and the degree of PSMA expression (miPSMA scores) (Tables 19.4 and 19.5). For intraprostatic tumor evaluation, miTNM suggests the use of complementary mpMRI using a hybrid PET/MRI or separate PET/MRI fusion [63].

The PSMA Reporting and Data System (PSMA‐RADS) has been proposed, which uses a five‐point scale for categorization of positive findings, considering the anatomic location, the degree of PSMA expression, and anatomic image correlation. Up to five target lesions can be defined by the reader based on the expression intensity and diameter, and then the final overall PSMA‐RADS score is defined as the highest scored target lesion among these five representative target lesions. A high inter‐reader agreement is found for the use of PSAM‐RADS in the interpretation of PSMA‐targeted PET/CT [64].