Prostate-specific membrane antigen (PSMA) is a 100kDa type II-transmembrane glycoprotein which is overexpressed in nearly all prostate cancer cells (1). PSMA has been validated as a diagnostic and therapeutic target in prostate cancer (PCa), and radionuclide probes-targeted PSMA like ¹⁸F-PSMA, ⁶⁸Ga-PSMA and ¹⁷⁷Lu-PSMA was used in diagnosis and treatment of PCa (2). While PSMA is predominantly recognized in PCa, it is also expressed in various other solid tumors, such as thyroid, breast, liver, lung cancer, and glioblastoma (3–6). Primary liver cancer has shown the most rapid rise in mortality in decades. Hepatocellular carcinoma (HCC) is the most common primary liver cancer, followed by intrahepatic cholangiocarcinoma (ICC). Chen et al. (7) analyzed PSMA expression in 446 formalin-fixed paraffin-embedded (FFPE) liver tumors (213 HCC, 203 ICC, and 30 liver cirrhosis) and found that PSMA was expressed in 86.8% of HCC, 79.3% of ICC, and only 6.6% of liver cirrhosis. Few PSMA-targeted PET imaging studies of HCC, ICC, and combined hepatocellular-cholangiocarcinoma (CHC) have been reported. To date, only two cases have reported PSMA-PET imaging of the ICC. Herein, we report a case of PCa with incidental ICC detection using ¹⁸F-PSMA and ¹⁸F-fluorodeoxyglucose (FDG) PET/MRI.

A 71-year-old man presented with more than 4 years history of urinary urgency and frequency, nocturia, and dysuria. The serum prostate-specific antigen (PSA) level was 5.54 ng/ml. Other tumor markers, including carbohydrate antigen (CA) 72–4 (75 U/ml), CA125 (37.3 U/ml), were elevated. Alpha-fetoprotein (AFP, 3.35 ng/ml), carcinoembryonic antigen (CEA, 2.98 ng/ml and CA19–9 (< 2.0 U/ml) levels were within normal ranges. Prostate biopsy was performed. Histopathological examination revealed prostatic adenocarcinoma with a Gleason score of 6 (3 + 3) ( Figure 1A ). Immunostaining revealed that the tumor cells were positive for P504S ( Figure 1B ). For preoperative staging, ¹⁸F-PSMA-1007 PET/MR imaging was performed.

Whole-body PSMA-PET revealed diffuse increased uptake in the prostate (maximum standard uptake value [SUVmax]=5.53) ( Figures 2A, C, D ). T2-weighted MR revealed an area of low signal intensity in the bilateral peripheral zone and heterogeneous signal intensity in the transitional and central zones ( Figure 2B ). In addition to the prostate lesions, PSMA-PET imaging showed abnormally increased radiotracer uptake in the liver mass (SUVmax=18.76) ( Figures 3B, C ). MR imaging showed that the mass was located at the left-right lobe junction with a low signal on T1-weighted images ( Figure 3G ) and a slightly high signal on T2-weighted images ( Figure 3A ). To further determine the nature of liver lesions, ¹⁸F-FDG PET/MRI was performed. The prostate showed no FDG uptake ( Figures 2E–H ). The liver lesion had a low FDG uptake with an SUVmax of 3.23 ( Figures 3D–F ). Contrast-enhanced MRI of the abdomen was conducted revealing significant enhancement of the margin of the mass in the arterial phase, with further enhancement in the portal phase. The center of the mass was a necrotic area with no enhancement or radiotracer uptake ( Figures 3B, G–I ).

Subsequently, the liver lesion was surgically removed. Histopathological examination of the lesion revealed a moderately to poorly differentiated cholangiocarcinoma ( Figure 1C ); further, the tumor cells were positive for CK7 ( Figure 1D ) and negative for PSA ( Figure 1E ) and P504S ( Figure 1F ).

PCa ranks first for estimated new cancer cases and second for estimated deaths for all cancer in men (8). Cancer metastasis is a major cause of mortality. Direct spread, hematogenous metastasis, and lymph node metastasis are metastatic pathways. PCa most frequently metastasizes to bones (84%), distant lymph nodes (10.6%), liver (10.2%), and thorax (9.1%). In patients with bone metastases, only 19.4% have multiple sites involved, and the most common sites of secondary metastases are the liver (39.1%), thorax (35.2%), distant lymph nodes (24.6%), and brain (12.4%) (9). In our patient, PSMA-PET showed multiple bones with PSMA uptake (data not shown) and the liver lesion with high PSMA and low FDG uptake. Therefore, we falsely concluded that the patient had bone and liver metastases. The misdiagnosis of this case led us to further search and summarize positive PSMA-PET images of liver lesions in PubMed. From 2015 to 2022, a total of 17 cases underwent PSMA-PET imaging (15 of ⁶⁸Ga-PSMA and 2 of ¹⁸F-PSMA-1007), comprising eight cases of HCC (10–17), two of ICC (18, 19), one of combined hepatocellular-cholangiocarcinoma (CHC) (20), one liver metastasis of cholangiocarcinoma (CCA) (21), and five benign liver lesions (22–26), as detailed in Table 1 . The mean age was 74.0 ± 6.4. The main purposes (76%, 13/17) of PSMA imaging were related to PCa, including elevated PSA levels and PCa staging or follow-up. The liver lesions in these cases were accidentally detected. Other reasons include research, staging, and therapy for malignant liver tumors. Most cases (n = 12) were concurrent with PCa. Two patients developed a third primary malignancy. Liver lesions were benign in five patients, including benign liver hemangioma (23), focal nodular hyperplasia (24), focal inflammation and steatosis (22), vascular malformation (25), and fatty sparing (26). All these 17 liver lesions had high PSMA uptake, and the mean SUVmax was 17.05 ± 6.31 (9.9–29.4). The median SUVmax of malignant tumor and benign lesion was 17.62 ± 7.62 and 16.1 ± 4.60, separately (P=0.76). Research had reported there was difference in SUVmax in ⁶⁸Ga-PSMA and ¹⁸F-PSMA-1007 (27), so we just analyzed the SUVmax in ⁶⁸Ga-PSMA. Five patients with malignant liver tumors underwent FDG examinations. The uptake in FDG-PET was lower than that in PSMA-PET in four patients. Three of them had HCC and one had ICC. Same as our case, Kang et al. reported incidental ICC in a 69-year-old man with PCa and the SUVmax of the liver lesion was 12.8 and 6.7 in ⁶⁸Ga-PSMA and ¹⁸F-PSMA-1007 respectively (19). Another case also reported high PSMA uptake in ICC (18). Therefore, PSMA-PET imaging may be better than FDG-PET imaging for diagnosing and alternative staging of ICC. Until now, there was just three cases reported ICC in PSMA-PET, so that large sample should be further studied in the future.

Three large-sample studies analyzed two different radiotracer PET imaging methods for detecting HCC. Kusymptcu et al. (28) found that FDG was positive in 15 patients and PSMA was positive in 16 patients. The mean SUVmax and tumor-to-background ratio of liver lesions on PSMA-PET were higher than those on FDG-PET. In nine patients, PSMA uptake was higher on visual and quantitative evaluations, whereas FDG uptake was observed in only four patients. A prospective pilot study analyzed 37 suspected malignant lesions in 7 patients with HCC and found that 36 of them showed increased PSMA uptake and only 10 were FDG-avid. A study with 14 HCC patients demonstrated that 36% of them had low FDG uptake, and 21% and 43% had moderate and high FDG uptake, respectively. On PSMA-PET, only 7% and 14% showed low and moderate uptake respectively, and 79% showed high uptake. The mean SUVmax and tumor-to-abdominal aorta or tumor-to-gluteal muscle ratios on PSMA-PET were higher than those observed on FDG-PET. These three studies showed that PSMA-PET imaging was superior to FDG-PET for detecting HCC. Its sensitivity, specificity, positive predictive value, negative predictive value, and accuracy are 97%, 100%, 100%, 80%, and 97%, respectively (10). Therefore, PSMA-PET may be used for the initial staging of HCC and as a potential 177Lu-PSMA therapy. Other than ICC and HCC, the CHC and liver metastasis of CCA also had high PSMA uptake (20, 21). Chen et al. (7) found that PSMA is primarily expressed in the neovascular endothelium associated with tumors. In benign liver lesions, this expression is probably due to increased local blood flow, local vascular permeability, and PSMA-expressing folate receptors in macrophages (29–31). Our study, along with others, revealed PSMA uptake in non-prostatic tissues and lesions, such as physiological uptake, benign pathological uptake, and non-prostatic uptake. Accordingly, when PET shows abnormal PSMA uptake in non-prostatic lesions in patients with prostate cancer, benign or malignant lesions other than metastases should be considered. A biopsy or surgery can be performed. If the non-prostatic lesion is located in the liver, it may be HCC, ICC, CHC, or other benign lesions. Although PSMA-PET has no advantage in the differential diagnosis of metastases of PCa and other lesions, it may be used as alternative staging and to identify patients with liver primary malignant for PSMA-targeted therapy.

Liver lesions with PSMA-avid in PCa cancer may not be metastasis of PCa but benign or malignant liver tumors, which should be further identified through pathology of biopsy or surgical specimens. PSMA-PET is superior to FDG-PET in detecting ICC and HCC and may be used as an alternative staging method.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving humans were approved by Medical Ethics Committee of Shanghai East Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

YS: Funding acquisition, Writing – original draft. HW: Writing – original draft. YY: Writing – original draft. ZY: Writing – original draft. JZ: Writing – review & editing.