Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Oct 1.
Published in final edited form as: Eur J Nucl Med Mol Imaging. 2022 Jun 28;49(12):4000–4013. doi: 10.1007/s00259-022-05884-9

PET imaging of hepatocellular carcinoma by targeting tumor-associated endothelium using [68Ga]Ga-PSMA-617

Qiaomiao Lu 1,2, Yu Long 1,2, Kevin Fan 3, Zhiwen Shen 4, Yongkang Gai 1,2,5, Qingyao Liu 1,2,5, Dawei Jiang 1,2,5, Weibo Cai 3, Chidan Wan 4, Xiaoli Lan 1,2,5
PMCID: PMC9529836  NIHMSID: NIHMS1825602  PMID: 35763056

Abstract

Objective

Hepatocellular carcinoma (HCC) is a malignant tumor associated with high morbidity and mortality rates. In many non-prostate solid tumors such as HCC, prostate-specific membrane antigens (PSMA) are overexpressed in tumor-associated endothelial cells. Therefore, the aim of this study was to evaluate the performance of [68Ga]Ga-PSMA-617 PET imaging on HCC with different animal models, including cell line-derived xenografts (CDX) and patient-derived xenografts (PDX), and to explore its mechanisms of function.

Methods

[68Ga]Ga-PSMA-617 was prepared. The expression level of PSMA in two human hepatocellular cancer cells (HepG2 and HuH-7) was evaluated, and the cellular uptakes of [68Ga]Ga-PSMA-617 were assayed. HepG2 and HuH-7 subcutaneous xenograft models, HepG2 orthotopic xenograft models, and four different groups of PDX models were prepared. Preclinical pharmacokinetics and performance of [68Ga]Ga-PSMA-617 were evaluated in different types of HCC xenografts models using small animal PET and biodistribution studies.

Results

Low PSMA expression level of HepG2 and HuH-7 cells was observed, and the cellular uptake and blocking study confirmed the non-specificity of the PSMA-targeted probe binding to HepG2 and HuH-7 cells. In the subcutaneous xenograft models, the tumor uptakes at 0.5 h were 0.76 ± 0.12%ID/g (HepG2 tumors) and 0.78 ± 0.08%ID/g (HuH-7 tumors), respectively, which were significantly higher than those of the blocking groups (0.23 ± 0.04%ID/g and 0.20 ± 0.04%ID/g, respectively). In the orthotopic xenograft models, PET images clearly displayed the tumor locations based on the preferential accumulation of [68Ga]Ga-PSMA-617 in tumor tissue versus normal liver tissue, suggesting the possibility of using [68Ga] Ga-PSMA-617 PET imaging to detect primary HCC lesions in deep tissue. In the four different groups of HCC PDX models, PET imaging with [68Ga]Ga-PSMA-617 provided clear tumor uptakes with prominent tumor-to-background contrast, further demonstrating its potential for the clinical imaging of PSMA-positive HCC lesions. The staining of tumor tissue sections with CD31- and PSMA-specific antibodies visualized the tumor-associated blood vessels and PSMA expression on endothelial cells in subcutaneous, orthotopic tissues, and PDX tissues, confirming the imaging with [68Ga]Ga-PSMA-617 might be mediated by targeting tumor associated endothelium.

Conclusion

In this study, in vivo PET on different types of HCC xenograft models illustrated high uptake within tumors, which confirmed that [68Ga]Ga-PSMA-617 PET may be a promising imaging modality for HCC by targeting tumor associated endothelium.

Keywords: Hepatocellular carcinoma (HCC), Positron emission tomography (PET), Prostate-specific membrane antigen (PSMA), Tumor-associated endothelial cells, Diagnosis

Introduction

Hepatocellular carcinoma (HCC) is a malignant tumor that seriously endangers human health, accounting for ~90% of primary liver cancer cases [1]. According to a status report from 2018 on the cancer burden worldwide, liver cancer is predicted to be the sixth most commonly diagnosed cancer and the fourth leading cause of cancer-related mortality, with about 841,000 new cases and 782,000 deaths annually [2]. As one of the most high-risk HCC areas, China holds more than 50% of the world’s burden, and 47.1% of the deaths occurred in China [3].

HCC is characterized by a rich blood supply and strong angiogenic activity, both of which play a critical role in tumor growth and metastasis, and is therefore associated with an unfavorable clinical course. According to the European Association for the Study of the Liver (EASL) Clinical Practice Guidelines, many patients with HCC are diagnosed in advanced stages, meaning appropriate selections of treatment strategies are limited [4]. Effective treatment of HCC depends on early diagnosis and accurate staging, as both factors are essential for proper clinical management strategies of HCC patients. One dilemma in the management of HCC is the differential diagnosis, so that specific tracers for HCC are highly needed in clinical practice. However, imaging of HCC at diagnosis is challenged by many factors, including low metabolism, physiological liver activity, and false positive findings, which restrict the use of positron emission tomography integrated computed tomography (PET/CT) with 18F-fludeoxyglucose ([18F]-FDG) [5]. Although other PET molecular tracers such as [11C]-acetate or [11C]-choline can improve sensitivity to an extent when diagnosing well-differentiated hepatocellular carcinomas, the detection rate for small lesions of hepatocellular carcinoma remains low [6, 7]. Furthermore, due to the short half-life of 11C, its application is limited [8]. Phosphatidylinositol protein polysaccharide-3 (Glypican-3, GPC3) is a heparan sulfate proteoglycan with high expression in HCC tumor cells, suggesting an important target for precise diagnosis of HCC [9]. Preclinical studies have confirmed the excellent specificity of GPC3 in HCC, and showing rapid clearance from the normal organs, and good biosafety profiles of GPC3-targeted probes [10]. GPC3-targeted immunoPET imaging strategies has the ability to specifically localize hepatocellular carcinoma with high tumor-to-background ratio and high sensitivity in preclinical HCC models, laying further investigations for clinical translation [11–13].

Prostate-specific membrane antigen (PSMA) is a type II transmembrane protein encoded by the FOLH1 gene [14]. In almost all stages of prostate cancer (PCa), PSMA expression was found to be upregulated, especially in poorly differentiated, metastatic, and hormone-refractory cases [15–18]. Accordingly, PSMA has been considered as an ideal target for PCa imaging and therapy. In recent years, more literatures have reported that PSMA was also expressed in the vasculature of various non-prostate solid cancer types, including thyroid cancer, lung cancer, breast cancer, renal cancer, gastrointestinal cancer, and HCC, but not in the normal vascular endothelial cells of benign tissues [19–24]. Though PSMA expression levels in the tested non-prostatic cancer cell lines are nearly 30-fold less than what was observed in PSMA-positive LNCaP cells [25], histopathological studies have confirmed that the overexpression of PSMA on newly formed micro vessels of various non-prostate solid tumors, and nearly 95% of HCC tumor-associated vasculature generally show high levels of PSMA expression [26]. Therefore, PSMA may be a promising diagnosis, prognostic and even therapeutic target for HCC.

In recent years, the discovery of increased [68Ga]Ga-PSMA uptake in HCC has been described, but a majority have been from case reports [27–29]. Taneja S et al. [28] reported a case involving a patient who underwent a [68Ga] Ga-PSMA PET/MRI scan for prostate cancer evaluation, but a PSMA avid HCC lesion was incidentally detected. Several clinical trials have also reported [68Ga]Ga-PSMA accumulation in both hepatic lesions and extrahepatic metastases, suggesting a potential role of [68Ga]Ga-PSMA PET/CT in diagnosing and staging for patients with HCC [30, 31]. Despite these case reports and small-sized clinical trials, in-depth basic research is still needed. The aim of this study was to evaluate the potential of [68Ga]Ga-PSMA-617 PET imaging in HCC using different types of HCC models, including cell line-derived xenografts (CDX), subcutaneous xenograft models and orthotopic xenograft models, and patient-derived xenografts (PDX) models, and to explore the underlying molecular mechanism.

Materials and methods

Cell culture

Human prostate cancer cells (LNCaP and PC-3) and human hepatocellular cancer cells (HepG2 and HuH-7) were provided by Stem Cell Bank and the Chinese Academy of Sciences. LNCaP and PC-3 cells were maintained in culture medium RPMI 1640 with 10% fetal bovine serum and 1% penicillin-streptomycin. HepG2 and HuH-7 cells were maintained in culture medium DMEM (high glucose with L-glutamine and sodium pyruvate) with 10% fetal bovine serum and 1% penicillin-streptomycin.

Identification of the PSMA expression

The PSMA expression of LNCaP, PC-3, HepG2 and HuH-7 cells were confirmed by Western blot and flow cytometry assessment. Detail steps were provided in the Supplement data.

Preparation of [68Ga]Ga-PSMA-617

The PSMA-617 precursor was purchased from Med Cheme Express (USA). 68Ga was eluted using a solution of 0.05 mol/L hydrochloride from the 68Ge/68Ga generator. The 68Ga solution (185 MBq), 0.25 mol/L sodium acetate (pH 6.8, 0.125 mL), and 1 mmol/L PSMA-617 (0.01 mL, dissolved in DMSO) were added in a micro tube and reacted at 95°C for 15 min with 300 rpm agitation. After cooling, small ali-quots (10–30 μL) were withdrawn from the end products for quality control. Labeling efficiency and radiochemical purity were determined using Radio-HPLC (Shimadzu Corpporation, Japan). The radiochemical purity was ≥95%.

In vitro cellular uptake analysis

Cellular uptake analysis was performed to confirm that the expression of PSMA was not observed in the hepatocellular cancer cell itself. LNCaP, PC-3, HepG2 and HuH-7 cells were seeded onto 24-well plates (1 × 105 cells/well) and cultured overnight. [68Ga]Ga-PSMA-617 solution (1 nM, 2 μCi) was then added to each well and incubated for 0.5, 1, and 2 h. Wells for blocking studies were co-administered with the known PSMA inhibitor ZJ-43 (0.5 mg/well, Tocris Biocience, UK). Cells were washed twice with phosphate-buffered saline (PBS) and collected in solutions after they were lysed with 0.1 N NaOH (200 μL). Radioactivity of the collected supernatant and cell lysate were measured separately via automatic gamma counter (Perkin Elmer, USA).

Preparation of xenograft models

Animal housing

Nude mice and NOD-scid IL2Rgammanull (NOD-scid gamma, NSG) mice at 4–6 weeks of age were purchased from Beijing Vital River Laboratory Animal Technology (China). All mice in research were maintained in a barrier facility under HEPA-filtered air with food and water available. Temperature (20–21°C), humidity (50–60%) and 12 h light-dark cycles were controlled. Animals were manipulated under sterile conditions during surgery. Animal experiments complied with the standards of current ethical guidelines for humane animal care, and were approved by The Institutional Animal Care and Use Committee at Tongji Medical College, Huazhong University of Science and Technology ([2021] IACUC Number: 2688).

Subcutaneous cell-derived xenograft models

Subcutaneous cell-derived xenograft models were established by the subcutaneous injection of HepG2 or HuH-7 cells (1 × 107 cells) suspended in 0.1mL of PBS and Matrigel (1:1; Corning, Biocoat, USA).

Orthotopic xenograft models

When subcutaneous HepG2 cell xenograft tumor size reached approximately 1.0–1.2 cm in diameter, the tumor tissue was removed and cut into fragments roughly 1mm3 in size. Nude mice were injected intraperitoneal with pentobarbital sodium solution (1%, 0.01 mL) for anesthesia prior to all procedures. The abdominal skin was sterilized with povidone iodine solution and a 10 mm vertical incision was performed in the midline of the upper abdomen through the skin and peritoneum, and exposure of the left lobe of the liver was carefully performed. A single 1 mm3 tumor fragment was then implanted into the left lobe using an absorbable suture surgical method. After compression hemostasis, the liver was gently placed back. The abdominal wall and skin were closed with continuous surgical sutures. Mice were injected subcutaneously with an analgesic (carprofen, 5 mg/kg) once a day for 3–5 days after implantation of orthotopic tumors to the liver. All mice should be observed carefully to evaluate whether there had weight loss, abnormal appearance (rough fur, slow response, and arched back), and abnormal manifestations (dyspnea, tremor, spasm, crouching), to determine the administration time of analgesics. Post-operative anti-infection was treated with penicillin solution (400,000 units/mL, 0.2–0.25 mL). Experimental scheme of orthotopic xenograft models was shown in Figure S1.

Patient-derived xenografts models

Tumors selected for the establishment of PDX mouse models were from patients with suspected HCC who had undergone tumor resection. Human HCC samples and peritumoral liver tissue were obtained from April to May 2021, after patients provided written informed consent. Resected HCC tissues from 5 patients were used for the establishment of PDX mouse models. The tumor tissue for xenografts were obtained at the time of tumor resection after confirmed histologic diagnosis. Excess fresh tumor tissue for implantation was placed in a culture medium immediately after resection and transported from the operating room to our laboratory on ice, with the time from collection to implantation being less than 2 h. Additional HCC tissue and peritumoral liver tissue was snap-frozen in liquid nitrogen for future studies and fixed in paraformaldehyde for histopathological analysis. Detailed information for further tumor model preparation was listed in the Supplement data, and Figure S2 and S3.

[68Ga]Ga-PSMA-617 PET imaging and analysis

Detailed information for imaging experimental setup was listed in the Supplement data (Table S1).

Subcutaneous cell-derived xenograft models

Tumor xenograft mice (n = 4/group) were investigated using a small animal PET/CT scanner (Novel Medical, China) when tumor size reached approximately 1.0–1.2 cm in diameter. After the intravenous injection of [68Ga]Ga-PSMA-617 (7.4–11.1 MBq) in saline, static PET scans were obtained at 0.5, 1, and 2 h post injection (p.i.) for all mice under isoflurane anesthesia. Acquisition time for each scan was roughly 15 min. Blocking studies were performed in tumor xenograft mice (n = 4/group) by co-injection of [68Ga]Ga-PSMA-617 (7.4–11.1 MBq) with ZJ-43 (25mg/Kg, Tocris). The detail information of image acquisition and processing data was listed in the Supplement data (Table S2). All PET data was reconstructed by three-dimensional-ordered-subset expectation maximization with attenuation and scatter correction. Regional uptake of radioactivity was decay-corrected to the injection time and expressed as %ID/g. Regions of interest (ROI) in the tumor, liver, muscle, and kidney were drawn on PET images to quantify the radioactive signals.

Orthotopic xenograft models

At day 21 after modeling, the tumor size reached approximately 1.0 cm in diameter. Tumor xenograft mice (n = 4/group) were investigated using a small animal PET/CT scanner. After the intravenous injection of [68Ga]Ga-PSMA-617 (7.4–11.1 MBq) in saline, static PET scans were obtained at 0.5 h post injection (p.i.) for all mice under isoflurane anesthesia. PET imaging acquisition and reconstruction are the same as above.

Patient-derived xenografts models

Tumor xenograft mice (n = 4/group) were investigated using a small animal PET/CT scanner when tumor size reached approximately 1.0–1.2 cm in diameter. After the intravenous injection of [68Ga]Ga-PSMA-617 (7.4–11.1 MBq) in saline, static PET scans were obtained at 0.5, 1, and 2 h post injection (p.i.) for all mice under isoflurane anesthesia. PET imaging acquisition and reconstruction are the same as above.

Biodistribution of [68Ga]Ga-PSMA-617

Biodistribution was carried out at 0.5, 1, and 2 h after injection of [68Ga]Ga-PSMA-617. Mice were sacrificed, and the tumor, blood, brain, heart, lung, liver, spleen, kidneys, stomach, intestine, muscle, bone, and pancreas were harvested and wet weighed. The amount of radioactivity was determined by a gamma counter, and results were expressed as percentages of the injected dose per gram of tissue (%ID/g). Detailed information for biodistribution experimental setup was listed in the Supplement data (Table S1).

Ex vivo blocking experiment and phosphorimaging

Ex vivo blocking experiment was performed. After the mice were sacrificed, tumor tissues were immediately removed and frozen in dry ice and sectioned using a cryomicrotome (PSZ–5330, Pasajie Medical, China) with a thickness of 30 μm. Tumor tissues sections were placed in petri dishes with high-glucose DMEM media. [68Ga]Ga-PSMA-617 solution (100 μCi) was then added and incubated for 0.5 h. In blocking studies, tumor tissue sections were co-administered with PSMA inhibitor ZJ-43 (5 mg/mL,100μL). After incubation, tumor tissue sections were washed with PBS and then were exposed on phosphorimaging plates that were subsequently scanned, and analyzed using a phosphorimager.

Tissue staining

The staining of tissue sections was performed to identify PSMA expression in the CDX and PDX tumor, healthy organs of mouse model, and peritumoral liver tissue of human HCC samples. Detailed information was shown in the Supplement data.

Statistical analysis

Results were shown with mean ± S.D. The differences between groups were analyzed using Prism software (Graph-Pad Software, Inc.). Student’s t-tests were used, and P values of less than 0.05 were considered statistically significant.

Results

Identification of the PSMA expression and in vitro cellular uptake analysis

The Western blot results of PSMA expression levels in LNCaP (PSMA+), PC-3 (PSMA−), HepG2, and HuH-7 cells were shown in Fig. 1A, B, and demonstrated that, like PC-3 (PSMA–) cells, the PSMA expression level of HepG2 and HuH-7 cells were significantly lower compared to LNCaP (PSMA+) cells. As shown in Figure S4, both HepG2 and HuH-7 were PSMA-negative assayed by flow cytometry.

Fig. 1.

Fig. 1

Open in a new tab

Identification of PSMA expression level in cells and cell uptake study. A and B Prostate-specific membrane antigen (PSMA) expression assay in prostate cancer cells LNCaP (PSMA+) and PC-3 (PSMA−), and hepatocellular carcinoma cells HepG2 and HuH-7 using Western blot. C Cell uptake of LNCaP, PC-3, HepG2, and HuH-7 cells after 0.5, 1, 2 h of incubation with [68Ga]Ga-PSMA-617 (n=4). D Blocking study with inhibitor ZJ-43 (n = 4). E HepG2 and F HuH-7 cell uptakes with or without co-administration of ZJ-43 (n = 4)

Figure 1C revealed much higher cell uptakes of [68Ga] Ga-PSMA-617 in LNCaP (PSMA+) (5.11% ± 0.52%, 2 h) compared to PC-3 (PSMA−), HepG2, and HuH-7 cells (1.17% ± 0.33%, 1.18% ± 0.25%, 1.21% ± 0.46%, 2 h, respectively). In the blocking assay (Fig. 1D), the uptakes of LNCaP (PSMA+) cells showed a significant decrease at all time points (2.76% ± 0.30%, 2 h) while the cell uptakes of PC-3 (PSMA–), HepG2 and HuH-7 cells had little changes (1.35% ± 0.14%, 1.58% ± 0.31%, 1.26% ± 0.35%, 2 h, respectively). The uptake and blocking study confirmed the non-specific uptake of [68Ga]Ga-PSMA-617 within HepG2 and HuH-7 cells (Fig. 1E, F).

PET images of the [68Ga]Ga-PSMA-617 in subcutaneous xenograft models and biodistribution

PET images of the subcutaneous tumor-bearing mice with two cell lines (HepG2 and HuH-7) were acquired at 0.5, 1, and 2 h after administration of [68Ga]Ga-PSMA-617 in order to study the tumor uptake and metabolism of the probe in vivo (Fig. 2). The tumors were clearly visible as early as 0.5 h post injection for the two different tumor-bearing mice, and tumor uptake was significantly reduced in the blocking mice, suggesting specific uptake. With the extension of injection time, tumor uptake decreased, and the radioactivity in other organs or tissues decreased more dramatically, resulting in a better signal-to-noise ratio. Significantly, high kidney uptake was observed at 0.5, 1, and 2 h. Tracer accumulation detected in the kidneys and bladders may be due to renal extraction. In mice-bearing subcutaneous HepG2 and HuH-7 xenografts, ROI analysis of the PET data was shown in Table S3.

Fig. 2.

Fig. 2

Open in a new tab

Representative static PET images of [68Ga]Ga-PSMA-617 in HCC subcutaneous xenograft models with two different tumor cell lines (HepG2 and HuH-7). White arrows point to the tumors

Figure 3 indicates biodistribution results. The tumor uptake was 0.76 ± 0.12%ID/g (HepG2 tumors), and 0.78 ± 0.08%ID/g (HuH-7 tumors) at 0.5 h post injection, respectively, which were significantly higher than those of the blocking groups (0.23 ± 0.04%ID/g for HepG2 tumors and 0.20 ± 0.04%ID/g for HuH-7 tumors, respectively). Although tumor uptake at 1 and 2 h were lower than that of 0.5 h post injection, higher tumor-to-blood ratios (T/B) (2.34 ± 0.38 for HepG2 tumors and 1.97 ± 0.34 for HuH-7 tumors, respectively, at 2 h) and similar tumor-to-muscle ratios (T/M) (3.75 ± 0.69 for HepG2 tumors and 2.72 ± 0.75 for HuH-7 tumors, respectively, at 2 h), were recorded. The tumor-to-liver ratios (T/L) of HuH-7 tumors were higher than that of HepG2 tumors (2.09 ± 0.91 vs. 0.96 ± 0.23, at 2 h).

Fig. 3.

Fig. 3

Open in a new tab

Biodistribution of [68Ga]Ga-PSMA-617 in HepG2 and HuH-7 subcutaneous xenograft models. A Biodistribution of [68Ga] Ga-PSMA-617 in HepG2 subcutaneous xenograft models at 0.5, 1, 2 h post injection with and without co-administration of ZJ-43 as a blocking agent. B Tumor-to-muscle (T/M), tumor-to-blood (T/B), and tumor-to-liver (T/L) ratios at the indicated time points in HepG2 subcutaneous xenograft models. C Biodistribution of [68Ga]Ga-PSMA-617 in HuH-7 subcutaneous xenograft models at 0.5, 1, 2 h post injection with and without co-administration of ZJ-43 as a blocking agent. D T/M, T/B, and T/L ratios at the indicated time points in HuH-7 subcutaneous xenograft models. The data are expressed as the mean ± SD (n = 4/group). (***P < 0.001; ****P < 0.0001)

PET images of the [68Ga]Ga-PSMA-617 in orthotopic xenograft models and biodistribution

Figure 4A showed the PET images acquired at 0.5 h after administration of [68Ga]Ga-PSMA-617 in the HepG2 orthotopic xenograft models. The tumors were clearly visible with good contrast to the surrounding normal liver tissue, and its location correlated with the tumor site (Fig. 4D). Biodistribution (Fig. 4B, C) results showed the tumor uptake at 0.5 h was 1.08 ± 0.13%ID/g, with high T/B (2.08 ± 0.37), T/M (3.57 ± 0.36), and T/L (2.43 ± 0.51) ratios.

Fig. 4.

Fig. 4

Open in a new tab

PET images of the [68Ga-PSMA-617 in orthotopic xenograft models and biodistribution results. A Representative static PET imaging and B biodistribution of [68Ga]Ga-PSMA-617 in HepG2 orthotopic xenograft models at 0.5 h post injection. C Tumor-to-blood (T/B), tumor-to-muscle (T/M), and tumor-to-liver (T/L) ratios at 0.5 h post injection. D Surgical exploration after PET/CT imaging and biodistribution. The data are expressed as the mean ± SD (n = 4/group). White arrows in A and blue arrow heads in D point to the tumor

PET images of the [68Ga]Ga-PSMA-617 in patient-derived xenografts models and biodistribution

Resected HCC tissue samples from 5 patients (P01, P02, P03, P04, and P05) were used to establish PDX models. The clinical data was listed in Table 1. Four tissue samples (P01, P02, P03, and P05) successfully grew in the subcutaneous areas of the F0 mice and transplanted into F1 mice. The duration of F0 tumor formations ranged from 52 to 127 days. Four different groups of HCC PDX models were successfully established and passaged up twice for current studies, denoted as PDX-1, PDX-2, PDX-3, and PDX-4. All groups showed histopathological characteristics consistent with their corresponding primary HCC (Figure S5), and were chosen as the experimental models to evaluate the in vivo behavior of [68Ga]Ga-PSMA-617.

Table 1.

Characteristics of patients and carcinoma tissues

Number Gender Age Subtype TNM stage* Histological classification Immunohisto chemistry PDX**
P01 M 51 HCC T2N0M0 II Poorly differentiate (macrotrabecular/solid pattern with pseudoglandular structure) PSMA (+) + (PDX-1)
P02 M 62 HCC T1bN0M0 IB Moderately differentiate (trabecular pattern, with pseudoglandular structure) PSMA (+) + (PDX-2)
P03 M 49 HCC T2N0M0 II Poorly differentiate (macrotrabecular/solid pattern) PSMA (+) + (PDX-3)
P04 M 65 HCC T1bN0M0 IB Moderately-to-poorly differentiated (trabecular/macrotrabecular/solid pattern) PSMA (−) –
P05 F 47 HCC T3N0M0 IIIA Moderately differentiated (trabecular/macrotrabecular pattern) PSMA (+) + (PDX-4)
Open in a new tab
*

Classification is based on the 8th edition of the AJCC cancer staging classification [37]

**

+ indicates PDX models were successfully established and denoted as PDX-1, PDX-2, PDX-3, PDX-4; − indicates F0 tumor was not obtained successfully for passage in another NSG mice

F, female; M, male; HCC, hepatocellular carcinoma; PDX, patient-derived xenograft

Figure 5 showed the PET images of the PDX models. Tumor uptakes were clearly noticed at 0.5 h after injection of [68Ga]Ga-PSMA-617, and gradually decreased with time for all PDX models. The main concentration organs were kidney and bladder, and other organs demonstrated low uptakes and quickly decreased, resulting in low background signaling and favorable target-to-background ratios. Target specificity was evaluated by simultaneous administration of the known PSMA inhibitor ZJ-43 as a competitor with [68Ga]Ga-PSMA-617. The tumor uptake at 0.5 h after injection was suppressed greatly by blocking in PDX-1, and the radiotracer clearance in most organs was faster than when blocking was not used.

Fig. 5.

Fig. 5

Open in a new tab

Representative static PET imaging of [68Ga]Ga-PSMA-617 in HCC PDX-1, PDX-2, PDX-3, and PDX-4 models. White arrows point to the tumors

The biodistribution of [68Ga]Ga-PSMA-617 in four different groups of HCC PDX models were also examined (Fig. 6). The results were similar when compared to imaging, which both showed that the tumor was the highest uptake tissue apart from the kidney. [68Ga]Ga-PSMA-617 in the blood, tumor, and main organs decreased as time progressed. Figure 7 showed excellent T/M, T/B, and T/L ratios of PDX models. Regarding the blocking of PDX-1, a decrease of radioactivity was detected in most organs, and the tumor uptake decreased significantly (1.05 ± 0.17%ID/g vs. 0.28 ± 0.08% ID/g, at 0.5 h, P<0.001).

Fig. 6.

Fig. 6

Open in a new tab

Biodistribution of [68Ga]Ga-PSMA-617 in HCC PDX-1, PDX-2, PDX-3, and PDX-4 models at 0.5, 1, and 2 h post injection. The data are expressed as the mean ± SD (n = 4/group). (***P < 0.001)

Fig. 7.

Fig. 7

Open in a new tab

T/M, T/B, and T/L ratios at the indicated time points in HCC PDX-1, PDX-2, PDX-3, and PDX-4 models. The data are expressed as the mean ± SD (n = 4/group)

Ex vivo blocking experiment and phosphorimaging

Ex vivo blocking experiment and phosphorimaging results showed that the tumor uptake was significantly blocked by co-administered with PSMA inhibitor ZJ-43, which indicated that the tumor uptake was not affected by tumor blood flow and perfusion (Figure S6).

Staining of tissue sections

Staining of tumor tissue sections with CD31- and PSMA-specific antibodies visualized the tumor-associated blood vessels and the PSMA expression on endothelial cells in subcutaneous, orthotopic tissues, as well as PDX tissues (Fig. 8). In healthy organs of mouse model, the results showed much PSMA expression in kidney, few in brain, but no or low PSMA expression in other organs (Figure S7). Peritumoral liver tissue of human HCC samples (P01, P02, P03, and P05) also showed no obvious PSMA expression (Figure S8).

Fig. 8.

Fig. 8

Open in a new tab

Histological staining analysis. A Staining of subcutaneous and orthotopic xenografts tissue sections with CD31- and PSMA-specific antibodies. Original magnification, ×400. B Staining of PDX tumor tissue sections with PSMA-specific antibodies indicated clear PSMA expression on the tumor vasculature, but not on the tumor cells. Original magnification, ×400

Discussion

Imaging diagnosis of HCC continues to remain a challenge. In this study, we established different types of animal models, including two kinds of subcutaneous xenograft models, orthotopic xenograft models, and four different groups of PDX models, with the goal of evaluating the performance of [68Ga]Ga-PSMA-617 PET imaging for HCC and its mechanism. PET imaging and biodistribution studies confirmed obvious tumor uptake of [68Ga] Ga-PSMA-617 in different animal models, and the uptake could be blocked by a larger amount of PSMA inhibitors, suggesting the specificity of PSMA-targeted PET imaging. Further histological study confirmed PSMA expression on tumor associated blood vessels. These results suggested that [68Ga]Ga-PSMA-617 PET imaging could be used to detect HCC by targeting tumor-associated endothelium, and provided basis for further clinical application in HCC.

For in vivo analysis, we first established HCC subcutaneous xenograft models from two cell lines (HepG2 and HuH-7). Human HCC samples displayed a rich network of blood vessels, whereas xenografts in mice might display relatively lower blood vessel density. Considering this, we used relatively larger-sized xenograft models. PET imaging and biodistribution studies showed obvious radioactive accumulation in the tumors with good T/B, T/M ratios, and the uptakes could be largely blocked by co-injection using the PSMA inhibitor ZJ-43, suggesting the selectivity and specificity of imaging. For other organs, rapid excretion through the kidney was observed. The majority of [68Ga]Ga-PSMA-617 could be swept quickly from the blood, excreted foremost via the urinary system, and collected in the bladder. The tracer accumulation detected in the kidneys are mainly due to its renal excretion, and PSMA expression in the kidney was shown to be increased in mice [32].

The orthotopic xenograft model reproduces critical tumor-stroma interactions in HCC, as well as the specific pathophysiological features of HCC lesions, such as tumor initiation and progression, and the orthotopic liver transplantation tumors are generally present in deeper tumor sites. These promoted further in vivo PET imaging with orthotopic xenografts in this study. Unsurprisingly, the orthotopic xenografts tumors also had good uptake of [68Ga] Ga-PSMA-617 (1.08 ± 0.13%ID/g, at 0.5 h after injection of the tracer), which was significantly higher than the normal liver tissue (0.46 ± 0.09%ID/g, P<0.0001). In fact, from the PET images and biodistribution results, slight accumulation was present in the liver, indicating that a small proportion of the tracer may be cleared through the hepatobiliary system. Another possible reason for the increased liver uptake in mice may be free radionuclides that dissociated in vivo, because macrocyclic chelates have limited in vivo stability against superoxide dismutase in the liver, and such structures are more stable in humans than in mice, as observed in other studies on different DOTA-conjugated tracers [33–35]. Nevertheless, PET images clearly displayed the tumor locations based on the preferential accumulation of [68Ga]Ga-PSMA-617 in tumor tissue versus normal liver tissue, suggesting the possibility of using [68Ga]Ga-PSMA-617 PET imaging to detect primary HCC lesions with normal physiological liver uptake and in deep-tissue.

Though the CDX experiment preliminarily confirmed the potential of [68Ga]Ga-PSMA-617 PET for detecting HCC, the lack of original biological characteristics may yield inconsistent results between preclinical and clinical studies. Thus, establishing more appropriate preclinical models is quite critical. PDX models enable the examination of tumor tissue in a native environment without significantly affecting the heterogeneity and stromal architecture of the neoplasms, providing an ideal preclinical model for cancer research [36]. In the present study, we established a large panel of HCC PDX models that maintain the principal histopathological characteristics of the original tumors even after serial passages (F0–F2), which exhibit great promise in preclinical settings for investigating the potential of [68Ga]Ga-PSMA-617 PET imaging. In four HCC PDX groups, PET imaging provided clear tumor images early after administration of [68Ga]Ga-PSMA-617. Biodistribution studies also confirmed the excellent T/M, T/B, and T/L ratios. The specificity of [68Ga]Ga-PSMA-617 was strongly confirmed by effective uptake inhibition in the presence of the known PSMA inhibitor ZJ-43 in PET scan and biodistribution experiments. In vivo studies on PDX models indicated obvious accumulation of [68Ga]Ga-PSMA-617 in HCC xenografts and further demonstrated its potential for clinical imaging with [68Ga]Ga-PSMA-617 PET for HCC lesions.

It is worth noting that the targeting principle of [68Ga]Ga-PSMA-617 PET imaging for HCC detection was possibly related to the expression of PSMA in endothelium, but not the hepatocellular carcinoma itself, which was verified by in vitro cell study, immunofluorescence (CD31 and PSMA) and immunohistochemistry staining (PSMA). To identify the PSMA expression in HCC, we selected HepG2 and HuH-7, the major cell lines of HCC, comparing them to known PSMA-positive and negative prostate cancer cell lines (LNCaP and PC-3, respectively). The results of Western blot and flow cytometry assessment confirmed that PSMA expression level in HCC cells was relatively low. The cellular uptakes of [68Ga]Ga-PSMA-617 in HepG2 and HuH-7 were also very low, and could not be blocked by a PSMA inhibitor, suggesting that non-specific cell uptakes existed. Furthermore, according to the histopathological studies, PSMA expression could be observed mainly in tumor-associated vasculature in the xenografts. This suggests that the mechanism of [68Ga]Ga-PSMA-617 PET for HCC imaging might be the specific binding with PSMA expressed in tumor associated endothelium.

There were limitations present in this study, one of which being the relatively low tumor uptake and rapid washout of [68Ga]Ga-PSMA-617 from the tumor. This may be due to the rapid excretion and relatively low binding affinity of low-molecular-weight ligands compared to antibodies. Moreover, PSMA is expressed in HCC tissue at a relatively lower density in tumor associated endothelium compared to the cancer cell surface. Dynamic acquisition may be very helpful for better understanding of tumor uptake and washout for the tracer. Though relatively low tumor uptake in mice was observed in our study, published papers have shown high [68Ga]Ga-PSMA accumulation in HCC lesions on both visual and quantitative evaluation in clinical trials [30, 31]. Another limitation is the lack of comparison with the PSMA-negative HCC models. We tried to establish a group of PSMA-negative PDX models, but the F0 tumor was not obtained successfully for serial transplantation. Though PSMA-negative HCC models were not successfully established to obtain the negative control data, the blocking studies still demonstrated the specificity of [68Ga]Ga-PSMA-617 for PSMA-expressing xenografts in vivo. Thirdly, it should be noted that there was no enough evidence to undoubtedly conclude that the mechanism of [68Ga]Ga-PSMA-617 PET for HCC imaging was mediated by targeting tumor associated endothelium. Although the ex vivo blocking and phosphorimaging studies (Figure S6) confirmed that the tumor uptake could be significantly blocked by co-administered with PSMA inhibitor ZJ-43, suggesting it may not be relevant to the effect of tumor blood flow and perfusion, it was still not able to determine whether reduced signal intensity of the tumor tissue resulted from the blocked vessels. Correlation analysis is needed to provide stronger evidence.

Conclusions

In this study, we prepared different cell line-derived and patient-derived HCC xenograft models, and verified positive uptakes of [68Ga]Ga-PSMA-617 in HCC tumors, which suggested that [68Ga]Ga-PSMA-617 PET may be a promising imaging method for HCC by targeting tumor associated endothelium. These results need further clinical prospective confirmation.

Supplementary Material

SI

Acknowledgements

We would also like to acknowledge the service provided by Beijing Novel Medical Equipment Ltd. for image acquisition.

Funding

This work was funded by the National Natural Science Foundation of China (No. 82030052).

Footnotes

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s00259-022-05884-9.

Ethics approval and consent to participate All experimental schemes were performed under the guidance and approved by the Institutional Animal Care and Use Committee of Tongji Medical College of Huazhong University of Science and Technology. Extensive efforts were made to ensure minimal suffering of the animals used during the study.

Competing interests Weibo Cai is a scientific advisor, stockholder, and grantee of Focus-X Therapeutics, Inc. All other authors declare no conflict of interest.

References

  • 1.Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet. 2018;391(10127):1301–14. 10.1016/S0140-6736(18)30010-2. [DOI] [PubMed] [Google Scholar]
  • 2.Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394–424. 10.3322/caac.21492. [DOI] [PubMed] [Google Scholar]
  • 3.Omata M, Cheng AL, Kokudo N, Kudo M, Lee JM, Jia J, et al. Asia-Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int. 2017;11(4):317–70. 10.1007/s12072-017-9799-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.European Association for the Study of the Liver. Electronic address: easloffice@easloffice.eu; European Association for the Study of the Liver. EASL Clinical Practice Guidelines: management of hepatocellular carcinoma. J Hepatol. 2018;69(1):182–236. 10.1016/j.jhep.2018.03.019. [DOI] [PubMed] [Google Scholar]
  • 5.Lin CY, Liao CW, Chu LY, Yen KY, Jeng LB, Hsu CN, et al. Predictive value of 18F-FDG PET/CT for vascular invasion in patients with hepatocellular carcinoma before liver transplantation. Clin Nucl Med. 2017;42:e18–e187. 10.1097/RLU.0000000000001545. [DOI] [PubMed] [Google Scholar]
  • 6.Castilla-Lièvre MA, Franco D, Gervais P, Kuhnast B, Agostini H, Marthey L, Désarnaud S, Helal BO. Diagnostic value of combining 11C-choline and 18F-FDG PET/CT in hepatocellular carcinoma. Eur J Nucl Med Mol Imaging. 2016;43(5):852–9. 10.1007/s00259-015-3241-0. [DOI] [PubMed] [Google Scholar]
  • 7.Chotipanich C, Kunawudhi A, Promteangtrong C, Tungsuppawattanakit P, Sricharunrat T, Wongsa P. Diagnosis of hepatocellular carcinoma using C11 CHOLINE PET/CT: comparison with F18 FDG, contrast enhanced MRI and MDCT. Asian Pac J Cancer Prev. 2016;17:3569–73. [PubMed] [Google Scholar]
  • 8.Filippi L, Schillaci O, Bagni O. Recent advances in PET probes for hepatocellular carcinoma characterization. Exp Rev Med Devices. 2019;16(5):341–50. 10.1080/17434440.2019.1608817. [DOI] [PubMed] [Google Scholar]
  • 9.Zhou F, Shang W, Yu X, Tian J. Glypican-3: a promising biomarker for hepatocellular carcinoma diagnosis and treatment. Med Res Rev. 2018;38(2):741–67. 10.1002/med.21455. [DOI] [PubMed] [Google Scholar]
  • 10.Xu H, Tang Y, Zhao Y, Wang F, Gao X, Deng D, et al. SPECT imaging of hepatocellular carcinoma detection by the GPC3 receptor. Mol Pharm. 2021;18(5):2082–90. 10.1021/acs.molpharmaceut.1c00060. [DOI] [PubMed] [Google Scholar]
  • 11.Sham JG, Kievit FM, Grierson JR, Chiarelli PA, Miyaoka RS, Zhang M, et al. Glypican-3-targeting F(ab’)2 for 89Zr PET of hepatocellular carcinoma. J Nucl Med. 2014;55(12):2032–7. 10.2967/jnumed.114.145102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Labadie KP, Ludwig AD, Lehnert AL, Hamlin DK, Kenoyer AL, Sullivan KM, et al. Glypican-3 targeted delivery of 89Zr and 90Y as a theranostic radionuclide platform for hepatocellular carcinoma. Sci Rep. 2021;11(1):3731. 10.1038/s41598-021-82172-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.An S, Zhang D, Zhang Y, Wang C, Shi L, Wei W, Huang G, Liu J. GPC3-targeted immunoPET imaging of hepatocellular carcinomas. Eur J Nucl Med Mol Imaging. 2022. 10.1007/s00259-022-05723-x. [DOI] [PubMed] [Google Scholar]
  • 14.Bacich DJ, Pinto JT, Tong WP, Heston WD. Cloning, expression, genomic localization, and enzymatic activities of the mouse homolog of prostate-specific membrane antigen/NAALADase/folate hydrolase. Mamm Genome. 2001;12(2):117–23. 10.1007/s003350010240. [DOI] [PubMed] [Google Scholar]
  • 15.Ross JS, Sheehan CE, Fisher HA, Kaufman RP Jr, Kaur P, Gray K, et al. Correlation of primary tumor prostate-specific membrane antigen expression with disease recurrence in prostate cancer. Clin Cancer Res. 2003;9:6357–62. [PubMed] [Google Scholar]
  • 16.Elsasser-Beile U, Buhler P, Wolf P. Targeted therapies for prostate cancer against the prostate specific membrane antigen. Curr Drug Targets. 2009;10:118–25. 10.2174/138945009787354601. [DOI] [PubMed] [Google Scholar]
  • 17.Bühler P, Wolf P, Elsässer-Beile U. Targeting the prostate-specific membrane antigen for prostate cancer therapy. Immunotherapy. 2009;1:471–81. 10.2217/imt.09.17. [DOI] [PubMed] [Google Scholar]
  • 18.Slovin SF. Targeting novel antigens for prostate cancer treatment: Focus on prostate-specific membrane antigen. Expert Opin Ther Targets. 2005;9:561–70. 10.1517/14728222.9.3.561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chang SS, Reuter VE, Heston WD, Bander NH, Grauer LS, Gaudin PB. Five different anti-prostate-specific membrane antigen (PSMA) antibodies confirm PSMA expression in tumor-associated neovasculature. Cancer Res. 1999;59:3192–8. [PubMed] [Google Scholar]
  • 20.Wernicke AG, Varma S, Greenwood EA, Christos PJ, Chao KS, Liu H, et al. Prostate-specific membrane antigen expression in tumor-associated vasculature of breast cancers. APMIS. 2014;122(6):482–9. 10.1111/apm.12195. [DOI] [PubMed] [Google Scholar]
  • 21.Baccala A, Sercia L, Li J, Heston W, Zhou M. Expression of prostate-specific membrane antigen in tumor-associated neovasculature of renal neoplasms. Urology. 2007;70(2):385–90. 10.1016/j.urology.2007.03.025. [DOI] [PubMed] [Google Scholar]
  • 22.Haffner MC, Kronberger IE, Ross JS, Sheehan CE, Zitt M, Mühlmann G, et al. Prostate-specific membrane antigen expression in the neovasculature of gastric and colorectal cancers. Hum Pathol. 2009;40(12):1754–61. 10.1016/j.humpath.2009.06.003. [DOI] [PubMed] [Google Scholar]
  • 23.Wang HL, Wang SS, Song WH, Pan Y, Yu HP, Si TG, et al. Expression of prostate-specific membrane antigen in lung cancer cells and tumor neovasculature endothelial cells and its clinical significance. PLoS ONE. 2015;10(5):e0125924. 10.1371/journal.pone.0125924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kasoha M, Unger C, Solomayer EF, Bohle RM, Zaharia C, Khreich F, et al. Prostate-specific membrane antigen (PSMA) expression in breast cancer and its metastases. Clin Exp Metastasis. 2017;34(8):479–90. 10.1007/s10585-018-9878-x. [DOI] [PubMed] [Google Scholar]
  • 25.Nimmagadda S, Pullambhatla M, Chen Y, Parsana P, Lisok A, Chatterjee S, et al. Low-level endogenous PSMA expression in nonprostatic tumor xenografts is sufficient for in vivo tumor targeting and imaging. J Nucl Med. 2018;59(3):486–93. 10.2967/jnumed.117.191221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Denmeade SR, Mhaka AM, Rosen DM, Brennen WN, Dalrymple S, Dach I, et al. Engineering a prostate-specific membrane antigen-activated tumor endothelial cell prodrug for cancer therapy. Sci Transl Med. 2012;4(140):140ra86. 10.1126/scitranslmed.3003886. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Patel D, Loh H, Le K, Stevanovic A, Mansberg R. Incidental detection of hepatocellular carcinoma on 68Ga-labeled prostate-specific membrane antigen PET/CT. Clin Nucl Med. 2017;42(11):881–4. 10.1097/RLU.0000000000001832. [DOI] [PubMed] [Google Scholar]
  • 28.Taneja S, Taneja R, Kashyap V, Jha A, Jena A. 68Ga-PSMA uptake in hepatocellular carcinoma. Clin Nucl Med. 2017;42(1):e69–70. 10.1097/RLU.0000000000001355. [DOI] [PubMed] [Google Scholar]
  • 29.Sasikumar A, Joy A, Nanabala R, Pillai MR, Thomas B, Vikraman KR. (68)Ga-PSMA PET/CT imaging in primary hepatocellular carcinoma. Eur J Nucl Med Mol Imaging. 2016;43(4):795–6. 10.1007/s00259-015-3297-x. [DOI] [PubMed] [Google Scholar]
  • 30.Kuyumcu S, Has-Simsek D, Iliaz R, Sanli Y, Buyukkaya F, Akyuz F, et al. Evidence of prostate-specific membrane antigen expression in hepatocellular carcinoma using 68Ga-PSMA PET/CT. Clin Nucl Med. 2019;44(9):702–6. 10.1097/RLU.0000000000002701. [DOI] [PubMed] [Google Scholar]
  • 31.Kesler M, Levine C, Hershkovitz D, Mishani E, Menachem Y, Lerman H, et al. 68Ga-PSMA is a novel PET-CT tracer for imaging of hepatocellular carcinoma: a prospective pilot study. J Nucl Med. 2019;60(2):185–91. 10.2967/jnumed.118.214833. [DOI] [PubMed] [Google Scholar]
  • 32.Schmittgen TD, Zakrajsek BA, Hill RE, Liu Q, Reeves JJ, Axford PD, et al. Expression pattern of mouse homolog of prostate-specific membrane antigen (FOLH1) in the transgenic adenocarcinoma of the mouse prostate model. Prostate. 2003;55(4):308–16. 10.1002/pros.10241. [DOI] [PubMed] [Google Scholar]
  • 33.Bass LA, Wang M, Welch MJ, Anderson CJ. In vivo transchelation of copper-64 from TETA-octreotide to superoxide dismutase in rat liver. Bioconjug Chem. 2000;11(4):527–32. 10.1021/bc990167l. [DOI] [PubMed] [Google Scholar]
  • 34.Rylova SN, Stoykow C, Del Pozzo L, Abiraj K, Tamma ML, Kiefer Y, et al. The somatostatin receptor 2 antagonist 64Cu-NODAGA-JR11 outperforms 64Cu-DOTA-TATE in a mouse xenograft model. PLoS ONE. 2018;13(4):e0195802. 10.1371/journal.pone.0195802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hu LY, Bauer N, Knight LM, Li Z, Liu S, Anderson CJ, et al. Characterization and evaluation of (64) Cu-labeled A20FMDV2 conjugates for imaging the integrin αvβ 6. Mol Imaging Biol. 2014;16(4):567–77. 10.1007/s11307-013-0717-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hu B, Li H, Guo W, Sun YF, Zhang X, Tang WG, et al. Establishment of a hepatocellular carcinoma patient-derived xenograft platform and its application in biomarker identification. Int J Cancer. 2020;146(6):1606–17. 10.1002/ijc.32564. [DOI] [PubMed] [Google Scholar]
  • 37.Chun YS, Pawlik TM, Vauthey JN. 8th Edition of the AJCC cancer staging manual: pancreas and hepatobiliary cancers. Ann Surg Oncol. 2018;25(4):845–7. 10.1245/s10434-017-6025-x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SI
NIHMS1825602-supplement-SI.docx (6.7MB, docx)

RESOURCES