Trop2 immuno-PET/CT imaging of prostate cancer: a proof-of-concept translational study

Abstract

Background

Prostate cancer (PCa) is the second leading cause of death in men and is highly prone to metastasis. This study aims to develop the novel immuno-PET/CT tracer targeting trophoblast cell surface antigen 2 (Trop2), a transmembrane protein overexpressed in aggressive prostate malignancies, and to investigate its diagnostic value in preclinical studies as well as its potential utility in detecting metastases in PCa patients.

Methods

Integrated analysis of TCGA, GEO, and HPA datasets revealed Trop2 overexpression in prostate adenocarcinoma. By labeling two Trop2-targeted nanobodies (His-tagged T4 and His-tag-free RT4) with ⁶⁸Ga and ¹⁸F, we synthesized three radiotracers, [⁶⁸Ga]Ga-NOTA-T4, [¹⁸F]AlF-RESCA-T4, and [¹⁸F]AlF-RESCA-RT4. Preclinical validation included cellular assays and xenograft studies for ⁶⁸Ga/¹⁸F-T4 variants, followed by a first-in-human trial evaluating [¹⁸F]AlF-RESCA-RT4 in ten treatment-naïve PCa patients.

Results

Bioinformatics analysis demonstrated Trop2 as a promising target for PCa diagnosis and treatment. In preclinical studies, both [⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4 illustrated high affinity to Trop2, specific tumor uptake (4.33 ± 0.38 %ID/g and 5.50 ± 0.69 %ID/g at 60 min, respectively) in Trop2-positive xenograft mouse models with minimal background distribution. In the translational study, [¹⁸F]AlF-RESCA-RT4 outperformed standard [¹⁸F]-FDG imaging in metastatic prostate cancer patients, detecting 62.4% more lymph node metastases (SUVmax 21.58 ± 13.12 vs. 4.80 ± 1.77) and 69.4% more bone lesions (SUVmax 7.83 ± 4.32 vs. 4.72 ± 1.22). No adverse events were observed.

Conclusions

This work successfully establishes that Trop2 is a new biomarker for PCa, and Trop2 immuno-PET/CT imaging can detect PCa lymph node and osseous metastases. The superior tumor-to-background contrast and clinical safety of [¹⁸F]AlF-RESCA-RT4 support its translational potential to guide Trop2-targeted therapies and enhance personalized treatment strategies.

Trial registration

NCT06851663. Retrospectively registered 02/24/2025.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-025-04589-8.

Background

Prostate cancer (PCa) is one of the most common malignancies among males worldwide, ranking as the second leading cause of cancer-related deaths in this population [1–3]. Due to its subtle onset, approximately 10–20% patients are diagnosed with metastatic disease in western countries, with bone and lymph nodes being the most common sites of spread, and that is even higher in many developing countries [4, 5]. While androgen deprivation therapy (ADT) demonstrates initial effectiveness in most men with advanced PCa, almost all patients eventually develop castration-resistant prostate cancer (CRPC) [6]. Targeted therapeutic strategies against cell surface antigens, particularly those that deliver cytotoxic agents or elicit anti-tumor immune responses, have emerged as promising therapeutic approaches for advanced cancers [7, 8]. Therefore, selecting a target that is stably and consistently highly expressed on the surface of prostate cancer cells is critical for the success of targeted therapies and molecular imaging strategies.

Trop2, encoded by the TACSTD2 gene, is a trophoblast cell surface antigen that has emerged as a promising therapeutic target due to its overexpression in multiple epithelial cancers [9]. The anti-Trop2 antibody–drug conjugate (ADC) Sacituzumab Govitecan (SG) has demonstrated clinical efficacy in various cancers, including triple-negative breast cancer, advanced non-small cell lung cancer, and metastatic platinum-resistant urothelial carcinoma [10–12]. In PCa, Trop2 overexpression drives tumor progression and strongly correlates with clinical recurrence and metastasis [13]. Emerging evidence suggests that Trop2-targeted therapies may yield significant clinical benefits in patients with PCa [14]. Notably, Sperger et al. reported persistent Trop2 expression in CRPC patients both before and after androgen receptor (AR)-targeted therapies, suggesting potential therapeutic benefits in AR-resistant disease [15]. However, significant interpatient variability in Trop2 expression levels may influence ADC treatment outcomes, highlighting the need for non-invasive methods to monitor Trop2 expression dynamics during therapy.

The current gold standard for Trop2 detection is based on immunohistochemical staining of biopsy specimens, which faces inherent limitations due to tumor heterogeneity in both temporal expression dynamics and spatial distribution [16, 17]. Immuno-positron emission tomography (immuno-PET), a molecular imaging modality that combines antibody-derived targeting specificity with the quantitative capabilities of PET, addresses these limitations by enabling non-invasive monitoring of Trop2 expression on the whole-body level [18]. Over the past years, we have been focusing on developing Trop2-targeted molecular probes in the form of intact antibodies and nanobodies, demonstrating their utility in visualizing heterogeneous Trop2 expression for patient stratification and differential diagnosis [19–21]. In this study, we aim to evaluate the feasibility of Trop2-targeting nanobody probes for the non-invasive assessment of Trop2 expression in prostate cancer. To begin with, we employed bioinformatics analysis to demonstrate Trop2 as a promising target for PCa diagnosis and treatment. Following this, our preclinical data confirmed two probes’ ([⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4) superiority in tumor imaging with optimal biodistribution and clearance profiles. In a subsequent pilot translational study, we further assessed the diagnostic performance of [¹⁸F]AlF-RESCA-RT4 (a His-tag-free RT4 nanobody probe) for PCa, focusing on lymph node metastases and bone metastases. Clinical imaging confirmed that Trop2 immuno-PET/CT enabled precise detection of diagnosing metastatic PCa, offering a superior tool for patient stratification and disease monitoring. This novel imaging approach may guide patient stratification and promote precision prostate cancer treatment with Trop2-targeted ADCs or combinational therapies.

Methods

Trop2 expression pattern analysis in prostate cancer

To investigate the complementary diagnostic and therapeutic potential of Trop2 (encoded by TACSTD2 gene) in PCa, the mRNA levels of Trop2 in pan-cancer were identified from the Tumor IMmune Estimation Resource (TIMER) database (https://cistrome.shinyapps.io/timer) and Gene Expression Profiling Interactive Analysis (GEPIA) database (http://gepia.cancer-pku.cn/). TIMER 2.0 database contains 10,897 tumor samples derived from 32 cancer types [22]. The diffexp module of TIMER 2.0 was used to show the expression of Trop2 in different cancers by box plots. The GEPIA database includes the unified standardized pan-cancer from the Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression databases (GTEx) dataset [23]. Complete RNA-Seq data of 492 tumors and 152 normal samples based on TCGA and GTEx were used to further elucidate the expression of Trop2 in PCa. Moreover, the level of Trop2 protein in PCa was observed by immunohistochemistry (IHC) images from The Human Protein Atlas (THPA) (https://www.proteinatlas.org/). In addition, the Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo/) collects a mass of high-throughput gene expression data by microarray technology, where we collected single-cell RNA sequencing data of 11 PCa patients (accession number: GSE176031) [24]. Single-cell sequencing data analysis was performed using the scanpy pipeline (v1.8.2) [25]. Raw counts were log-normalized (scanpy.pp.normalize_per_cell(counts_per_cell_after = 105), scanpy.pp.log1p()) and subsequently used to select 4000 highly variable genes (scanpy.pp.highly_variable_genes()). Then, we reduced the dimensionality of the data by performing principal component analysis (PCA). To remove the batch effect from the different datasets, batch removal was performed with the parameter “n_pcs = 36, sc.external.pp.harmony_integrate.” In addition, the dimensionality of the merged dataset was reduced using Uniform Manifold Approximation and Projection (UMAP) implemented by the “scanpy.tl.umap” function. Finally, for clustering the cells of the neighborhood graph, we used the Leiden clustering.

Production and characterization of radiotracers

[⁶⁸Ga]Ga-NOTA-T4: After adjusting pH to 9.0–10.0 with 0.2 M Na₂CO₃/NaHCO₃, 1 mg of T4 in PBS (PB180327, Procell, China) was mixed with the bifunctional chelator p-SCN-Bn-NOTA (B-605, Macrocyclics, USA) at a molar ratio of 1:20, which is freshly dissolved in DMSO (75927J, Adamas-beta, China). The reaction was incubated at 37 °C for the synthesis of NOTA-conjugated T4. After 2 h, the Nap™−5 Column (17085301, Cytiva, USA) and Amicon Ultra Centrifugal Filters (3 kDa; UFC5003BK, Merck, Germany) were used to purify and concentrate the mixture. The final production NOTA-T4 solution was used for subsequent radiolabeling experiments. ⁶⁸GaCl₃ was eluted from the ⁶⁸Ge/⁶⁸Ga radionuclide generator (ITM Medical Isotopes GmbH, Germany) with 0.05 M hydrochloride. The pH of the ⁶⁸GaCl₃ solution was adjusted to pH 4.0–4.5 with 0.25 M sodium acetate (pH 6.8), then about 2 nmol NOTA-T4 was added to the 37 MBq of ⁶⁸GaCl₃ solution and reacted at 37 °C for 15 min with 400 rpm agitation. The end production was obtained by purifying with PD-10 column (17085101, Cytiva, USA) using PBS as the mobile phase.

[¹⁸F]AlF-RESCA-T4/RT4: 1 mg of T4/RT4 in PBS was added to 0.1 M Na₂CO₃ to reach a final pH of 8.5–8.7. The chelator (±)-H3RESCA-TFP (BDDH-1–25 mg, CONFLUORE, China) was dissolved in DMSO and added to the nanobody solution thoroughly, meeting (±)-H3RESCA-TFP:T4/RT4 = 1:12. The reaction mixture was incubated at room temperature for 2 h. The reaction mixture was purified with a pre-equilibrated Nap™−5 Column desalting column using 0.1 M CH₃COONH₄ solution (pH 4.4–4.6) as the mobile phase and concentrated with Amicon Ultra Centrifugal Filters. The product RESCA-T4/RT4 was used for subsequent radiolabeling experiments. As for ¹⁸F-labeling, 3 mL of ¹⁸F solution (1480 MBq) was added to a QMA column (WAT023525, Waters GmbH, Germany) and 200 μL of 0.9% NaCl solution was used to rinse the QMA column and collect the ¹⁸F solution. Then, 10 μL of 2 mM aluminum chloride solution (pH 4.4–4.6) was added to the ¹⁸F solution (1100 MBq) and placed for 5 min at room temperature. At the end, 600 μg of RESCA-T4/RT4 in 50 μL of 0.1 M CH₃COONH₄ solution (pH 4.4–4.6) was added to the mixture. The reaction system was placed in a thermostatic shaker at room temperature for 12 min. After the labeling reaction, the final product was purified with a pre-equilibrated PD-10 column using 0.9% NaCl solution as the mobile phase.

We tested the final products’ radiochemical purity (RCP) by high-performance liquid chromatography (HPLC, SHIMADZU, Japan) with a gel column (B213-050015-07830S, NanoChrom, China). For serum stability testing, [⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4 were incubated with 10% fetal bovine serum (FBS) for 2 h, respectively. The quality control was performed by instant thin-layer chromatography (iTLC, B-MS-1000F, Ziegler, USA).

Cell culture

Three human PCa cell lines (PC3, DU145, 22RV1) were purchased from Procell (CL-0185/CL-0075/CL-0004, China). All the cell lines were cultured in RPMI 1640 medium (11875093, Gibco, USA) supplemented with 10% fetal bovine serum (FBS; FBS500-H, HycezmBio, China) and 1% antibiotic–antimycotic (PB180120, Procell, China). Cells were cultured in a humidified incubator at 37 °C with 5% CO₂ humidity.

Western blot

For western blot analysis, cells were dispersed and collected for protein extraction. The total protein concentration was measured using a BCA protein assay kit (BF0026, Boerfu, China) and protein samples were loaded onto 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (BF0006, Boerfu, China). After separation, all transferred proteins were transferred using PVDF membranes (IPVH00010, Millipore, USA). After blocking for 30 min, membranes were separately incubated with the primary antibody (anti-Trop2 antibody: ab214488, Abcam, UK; anti-PSMA antibody: A9547, Abclonal, China; 4 °C, overnight) and secondary antibody (goat anti-rabbit IgG: 111–035-003, Jackson, China; room temperature, 30 min) orderly. Proteins were visualized by the Enhanced Chemiluminescence reagent and methodology (BF0023, Boerfu, China). GAPDH was applied as internal reference.

Saturation and competition binding assays

Saturation binding assays were performed by incubating PC3 (Trop2+) cells with different concentrations (0–100 nM) of [⁶⁸Ga]Ga-NOTA-T4. To determine non-specific binding, a parallel group of cells was preincubated previously with unlabeled T4. After incubation at 37 °C for 60 min, cells were washed three times with cold PBS and lysed with 1 M NaOH for radioactivity counting. For competition binding assays, PC3 cells were treated with various concentrations of unlabeled T4 (1 × 10⁻³–1 × 10³ nM) using [⁶⁸Ga]Ga-NOTA-T4 as the radioligand. Following a 60-min incubation, cells were washed with cold PBS, harvested, and the radioactivity was measured to obtain the half-maximal inhibitory concentration (IC50) value.

Cell uptake assays

Seed the PC3 cells (Trop2+) and 22RV1 cells (Trop2−) in 24-well plates with a cell concentration of 1.0 × 10⁵ cells/well and cultured overnight. After complete attachment, replace 800 µL of serum-free medium containing [⁶⁸Ga]Ga-NOTA-T4 (74 kBq) or [¹⁸F]AlF-RESCA-T4 (74 kBq) to each PC3 well, while [⁶⁸Ga]Ga-NOTA-T4 (74 kBq) was also added to 22RV1 cells, and incubation was performed at 37 °C for multiple time points (30 min, 60 min, and 120 min). For blocking studies, PC3 cells were incubated previously with unlabeled T4 (15 µg to each well) for 60 min. At each time point, the radioactive medium was discarded, and rinsed the radiotracer-treated cells with 800 μL pre-cooling PBS once. Collect the rinsed PBS as supernatants. Then, 800 μL 1.0 M NaOH was added for cell lysis and collected as lysates. Radioactivity of the supernatants and lysates were measured separately via γ-counter (PerkinElmer, USA). After attenuation correction, the cell uptake rate was determined as A_lysate / (A_supernatant + A_lysate) × 1000‰.

Animals and tumor modeling

The animal study protocol was approved by the Institutional Animal Care and Use Committee at Tongji Medical College, Huazhong University of Science and Technology (Approval Number: 4494). Male Balb/c nude mice (4–5 weeks old) and C57BL/6 mice (4–5 weeks old) were purchased from Beijing Vital River Laboratory Animal Technology (Beijing, China). Balb/c nude mice were used to construct subcutaneous PCa models, and C57BL/6 mice were used for biotoxicity experiments. 0.1 mL cell suspension (containing PC3 or 22RV1 cells, 1 × 10⁷, Matrigel and sterile PBS mixed in a ratio of 1:1) was implanted subcutaneously into the right axilla of each Balb/c nude mouse. Mice with the tumor volume reached 300–500 mm³ were used for imaging or biodistribution studies.

Preclinical PET imaging and data analysis

All PET/CT images were conducted by Trans PET Discoverist 180 (RAYCAN Technology Co, Ltd. (Suzhou), China). PET/CT imaging was performed on PC3/22RV1 tumor xenograft mice, respectively. A dose of 3.7–7.4 MBq of either [⁶⁸Ga]Ga-NOTA-T4 or [¹⁸F]AlF-RESCA-T4 was administered via the tail vein. Subsequently, the mice were anesthetized with 2% isoflurane and scanned 60 min after intravenous injection. For the blocking study, PC3 xenograft tumor-bearing mice received excess unlabeled T4 (400 µg) at 30 min prior to the radiotracer injection. PET/CT images were obtained after attenuation corrected reconstruction, and regions of interests (ROIs) were analyzed using the Inveon Research Workplace (Siemens, Germany).

Ex vivobiodistribution study

After the above PET imaging studies, mice were euthanized, and samples (tumor, blood, brain, heart, lung, liver, spleen, kidney, stomach, intestines, muscle, and bone) were freshly collected and weighed for radioactivity counting using a γ-counter (PerkinElmer, USA). The radioactivity of each sample was computed and described as a percentage of injected dose per gram (%ID/g).

Histopathological analysis

To rigorously assess Trop2 expression in the tumors, PC3 and 22RV1 tumors were separated from sacrificed tumor-bearing mice and fixed in 4% formalin, then embedded in paraffin. Sections of 5 μm were cut and stained for H&E, Trop2 (ab214488, Abcam, UK) using standard protocols. For the IHC process, a horseradish peroxidase (HRP)-labeled rabbit anti-human IgG H&L was used as the secondary antibody.

Toxicity evaluation

Five healthy C57BL/6 mice (14–16 g, male) received an intravenous injection of approximately 13 MBq [¹⁸F]AlF-RESCA-T4 (867 MBq/kg), while a control group of five mice were injected with an equal volume of saline. The administered dose was 400 times the equivalent dose for a 70-kg human. Over an 18-day observation period, the mice were monitored for changes in various health indicators (including diet, respiration, activity, defecation, skin condition, pain response, and other health indicators), with daily weight measurements. At the end of the study, the mice were euthanized. Blood samples were collected, centrifuged, and analyzed for aspartate aminotransferase (AST), alanine aminotransferase (ALT), and blood urea nitrogen (BUN) levels. Additionally, vital organs including the heart, lung, liver, spleen, and kidney were harvested for histological examination.

Pilot clinical [¹⁸F]AlF-RESCA-RT4 PET/CT imaging in prostate cancer patients

The human study was approved by the Ethics Committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University (Approval Number: LY2024-307-A) and registered in ClinicalTrials.gov (NCT06851663). The inclusion criteria were as follows: age between 18 and 75 years; histologically confirmed diagnosis of solid tumors or suspected solid tumors by diagnostic imaging; capable of giving signed informed consent; willingness and ability to cooperate with all programs of this study. The exclusion criteria were as follows: severe hepatic and renal insufficiency; allergic to single-domain antibody radiopharmaceuticals. Written informed consent forms were obtained from all the included patients.

In this study, ten patients (males, age 69 ± 9.2) with pathologically confirmed PCa were included. All patients underwent whole-body PET/CT 1 h after [¹⁸F]AlF-RESCA-RT4 or [¹⁸F]-FDG administration, with these two scans performed on two separate days within 1 week of each other. The total-body PET/CT scanner featured a longitudinal FOV of 194 cm and executed continuous PET scans for 10 min using the 3D-list mode (uEXPLORER, United Imaging). The CT scan equipped with a fixed tube voltage of 120 kV and a dose modulation technique to automatically adjust mAs. The maximum standardized uptake value (SUVmax) was used to quantify the uptake of radiopharmaceuticals in the primary and metastatic foci of the tumors. A 10-mm-diameter ROI was drawn in the mediastinum (blood pool) as background, and the SUVmax was recorded. The tumor-to-background uptake ratio was calculated as the SUVmax of the tumor lesion/SUVmax of the mediastinal background.

Statistical analysis

The data are presented as mean ± standard deviation (SD) percentages. Statistical analysis and charting were conducted using GraphPad Prism software (version 10.0, USA, RRID: SCR_002798‌‌), with a significance threshold set at P < 0.05.

Results

Expression pattern of Trop2 in human prostate cancer

To evaluate the diagnostic and therapeutic potential of Trop2 in PCa, we analyzed the expression pattern of Trop2 (encoded by TACSTD2) in PCa. First, the expression of Trop2 at mRNA level in various cancers was calculated by using TIMER database which revealed that Trop2 was significantly higher in several tumor samples including PARD (prostate cancer) (Fig. 1A). Furthermore, the results of GEPIA database also showed that Trop2 expression was significantly increased in PCa tissues compared to normal prostate tissues, which was consistent with the TIMER database results (Fig. 1B). Representative immunohistochemistry images from clinical PCa samples further confirmed elevated Trop2 protein levels in PCa (Fig. 1C). Subsequently, we analyzed the GSE176031 mRNA-seq dataset, which comprised 21,743 cells derived from PCa tissues of 11 patients [24]. Using marker genes for different cell types, we annotated the cells into ten significant clusters, namely the B cell, endothelial cell, epithelial cell, fibroblast, macrophage, mast cell, neutrophil, plasma cell, smooth muscle cell, and T cell (Fig. 1D). Of the ten cell types, Trop2 expression was found higher in epithelial cells (Fig. 1E, F). Based on these findings, we hypothesize that Trop2-targeted immuno-PET/CT imaging has the potential to significantly enhance the diagnosis and therapeutic management of PCa.

Fig. 1.

Expression pattern of Trop2 in human prostate cancer (PCa). A Trop2 (encoded by the TACSTD2 gene) expression levels in various cancer types were analyzed using the TIMER database, which is based on the TCGA database. B The differential mRNA level of Trop2 was analyzed between 492 PCa tissues (red) and 152 normal prostate tissues (gray) from the TCGA and GTEx databases using the GEPIA database. *P < 0.05. C The difference in Trop2 protein expression at the protein level between PCa tissue and normal prostate tissue was demonstrated by immunohistochemical results obtained from the HPA database (antibody CAB072852). D–F Trop2 expression levels in the single-cell transcriptome. The UMAP dimensional plot showed the cell types (D) and the Trop2 expression level (E) in each cell. F The violin plot exhibited the expression of Trop2 in different cell types

Preparation of the Trop2-targeted radiotracers and cell-based experiments

In preclinical evaluation, we developed two Trop2-targeted radiotracers [⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4, both derived from the high-affinity nanobody T4 [19, 21]. Both radiotracers exhibited excellent radiochemical purity (> 99% after purification) with non-decay-corrected radiolabeling yields of 57.93% and 15.93%, respectively (Fig. 2A and B). Radiolabeled T4 remained stable in serum for up to 2 h (Additional file 1: Fig. S1). To establish model systems for subsequent studies, we first characterized Trop2 expression across three prostate cancer cell lines (PC3, DU145, and 22RV1). Western blot analysis (Fig. 2C) revealed significantly elevated Trop2 levels in PC3 cells compared to DU145 and 22RV1 (P < 0.01). Notably, PC3 is a PSMA-low-expressing cell line, making it suitable for further evaluation of the complementary roles of Trop2 and PSMA in PCa theranostics. The binding affinity of [⁶⁸Ga]Ga-NOTA-T4 to the Trop2 receptor was assessed on PC3 cells, revealing a dissociation constant of 4.31 ± 1.67 nM and a 50% inhibitory concentration of 0.70 ± 0.48 nM (Fig. 2D and E). Cellular uptake studies demonstrated time-dependent accumulation of [⁶⁸Ga]Ga-NOTA-T4 in PC3 cells from 30 to 120 min post-incubation. At 60 min, PC3 cell uptake (11.04 ± 0.70‰) substantially exceeded that of 22RV1 cells (3.28 ± 0.68‰) (Fig. 2F). Targeting specificity was further validated through competitive binding assays: pre-treatment with unlabeled T4 reduced [¹⁸F]AlF-RESCA-T4 uptake in PC3 cells by 63% (6.13 ± 0.56‰ vs. 2.26 ± 0.10‰ at 60 min post-incubation) (Fig. 2G). These findings collectively confirmed the selective Trop2-targeting capability of [⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4.

Fig. 2.

Synthesis of Trop2-targeted radiotracers and cellular studies. A Radiochemistry purity of [⁶⁸Ga]Ga-NOTA-T4 (left) and B [¹⁸F]AlF-RESCA-T4 (right) after purification. C Trop2 and PSMA expression assay in prostate cancer cells PC3, DU145, and 22RV1 using western blot. Uncropped gel images shown in Additional file 2. D Saturation binding curve illustrating interaction of [⁶⁸Ga]Ga-NOTA-T4 with PC3 tumor cells (n = 4). E Competitive binding curve demonstrating displacement of [⁶⁸Ga]Ga-NOTA-T4 by unlabeled T4 on PC3 tumor cells (n = 4). F Uptake with [⁶⁸Ga]Ga-NOTA-T4 in PC3 cells and 22RV1 cells (n = 3). ****P < 0.0001. G Uptake with [¹⁸F]AlF-RESCA-T4 and blocking experiment with unlabeled T4 in PC3 cells (n = 3). ****P < 0.0001

[⁶⁸Ga]Ga-NOTA-T4 immuno-PET/CT imaging in prostate cancer model

The diagnostic performance of [⁶⁸Ga]Ga-NOTA-T4 immuno-PET/CT for PCa was first interrogated in Trop2-positive PC3 tumor and Trop2-negative 22RV1 tumor-bearing mice at 60 min post-injection. The PET/CT imaging results showed that [⁶⁸Ga]Ga-NOTA-T4 rapidly accumulated in the PC3 tumor with an excellent tumor-to-background ratio (TBR) at 60 min post-injection (Fig. 3A). Apart from high uptake in the kidney, only minimal radioactivity was detected in other healthy tissues and organs. Region of interest (ROI) quantitative analysis showed an average tumor uptake of 4.33 ± 0.38 %ID/g, followed by the kidney of 176.30 ± 13.25%ID/g (Fig. 3B). The subsequent biodistribution analysis data confirmed the highest uptake in the kidney accumulation value of 169.90 ± 39.91 %ID/g and tumor uptake of 2.80 ± 0.11 %ID/g (Fig. 3C). In comparison, the tumors of the control group had lower [⁶⁸Ga]Ga-NOTA-T4 uptake, with ROI and biodistribution results of 0.87 ± 0.06 %ID/g and 0.99 ± 0.12 %ID/g, respectively (Fig. 3D–F). Moreover, the immunohistochemical staining results confirmed the overexpression of Trop2 in PC3 xenograft tumors, which was significantly more pronounced than in 22RV1 tumors (Fig. 3G, H). Meanwhile, as shown in Fig. 3I, negative Trop2 expression in PC3 xenograft tumor was further reported. All these findings further verified the PET/CT imaging results.

Fig. 3.

[⁶⁸Ga]Ga-NOTA-T4 Immuno-PET/CT Imaging in prostate cancer models. A, D Representative [⁶⁸Ga]Ga-NOTA-T4 immuno-PET/CT images of Trop2-positive PC3 tumor (n = 3) and Trop2-negative 22RV1 tumor (n = 3) bearing mice at 60-min points after injection of the tracer. The tumors were displayed on the images with yellow arrows. B, E The ROI analysis of the Trop2-positive group and the Trop2-negative group. C, F Ex vivo biodistribution data of 2 groups. G, H The Trop2 immunohistochemistry (IHC) staining results of PC3 and 22RV1 tumor tissues. I The PSMA IHC staining results of PC3 tumor tissue

[¹⁸F]AlF-RESCA-T4 immuno-PET/CT imaging in Trop2-positive model

While [⁶⁸Ga]Ga-NOTA-T4 demonstrated robust tumor targeting, its diagnostic utility was partially limited by persistent renal retention, a phenomenon attributed to nanobody reabsorption via renal tubules. Building on reports suggesting reduced kidney uptake with ¹⁸F-labeled nanobodies, we evaluated the tropism of [¹⁸F]AlF-RESCA-T4 in PC3 xenograft models through integrated PET imaging and competitive blockade studies [26]. As shown in Fig. 4A, [¹⁸F]AlF-RESCA-T4 visualized subcutaneous PC3 tumors at 60 min post-injection. ROI analysis revealed as lightly higher tumor uptake compared to the Ga-68 counterpart (5.50 ± 0.69 %ID/g vs. 4.33 ± 0.38 %ID/g), though unexpectedly high renal accumulation persisted (199.10 ± 29.85 %ID/g) (Fig. 4B). Ex vivo biodistribution analysis indicated a lower renal uptake (120.10 ± 17.26 %ID/g) and a stable tumor retention (2.31 ± 0.20 %ID/g) of [¹⁸F]AlF-RESCA-T4 (Fig. 4C). Specificity was conclusively demonstrated through pre-administration of unlabeled T4 (400 µg): blocking experiments eliminated tumor visualization on PET (Fig. 4D), with ROI quantification showing 75.15% reduced tumor uptake (Fig. 4E) and biodistribution revealing 81.76% inhibition (Fig. 4F). Notably, comprehensive toxicity assessments revealed no acute physiological changes or organ damage (Additional file 1: Fig. S2), supporting further clinical translation of [¹⁸F]AlF-RESCA-T4 as a theranostic tool.

Fig. 4.

[¹⁸F]AlF-RESCA-T4 immuno-PET/CT imaging of Trop2-positive PC3 model. A, D At 60-min points, images of the non-blocking (n = 3) and 400 µg-blocking (n = 3) groups after [¹⁸F]AlF-RESCA-T4 injection. B, E ROI quantitative analysis of 2 groups. C, F Ex vivo biodistribution data of 2 groups. Yellow arrows: tumor

Head-to-head evaluation of [¹⁸F]-FDG and [¹⁸F]AlF-RESCA-RT4 PET/CT imaging in prostate cancer patients

Building on promising preclinical results, we translated Trop2-targeted immuno-PET/CT to clinical evaluation in PCa patients (n = 10, patient characteristics summarized in Additional file 1: Table 1). To address the high renal retention observed with [⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4, we further developed [¹⁸F]AlF-RESCA-RT4, a His-tag-free RT4 nanobody engineered for reduced kidney accumulation [19]. The average administered [¹⁸F]-FDG and [¹⁸F]AlF-RESCA-RT4 activities were 119 ± 26 and 284 ± 28 MBq, respectively. No adverse reactions were observed in all subjects during the injection or 4-h follow-up period of [¹⁸F]AlF-RESCA-RT4. On account of the high expression of Trop2 in both benign and malignant epithelial cells of prostate, physiological uptake of Trop2-targeted RT4 by benign prostate tissue significantly affects identification of primary lesion (Fig. 5). Despite this, [¹⁸F]AlF-RESCA-RT4 demonstrated excellent diagnostic performance for metastatic lesions. A total of 358 lesions were found in 10 patients, including 109 lymph node metastases and 249 bone metastases (Additional file 1: Table 2). Sixty-eight (62.4%) lymph node metastases and 172 (69.4%) osseous lesions were detected only by [¹⁸F]AlF-RESCA-RT4 (Fig. 6A). Among all lesions, the mean SUVmax was higher on [¹⁸F]AlF-RESCA-RT4 than on [¹⁸F]-FDG (Fig. 6B). The mean SUVmax of lymph node metastases on [¹⁸F]-FDG was 4.80 ± 1.77, and that on [¹⁸F]AlF-RESCA-RT4 was 21.58 ± 13.12 (P < 0.001). The mean SUVmax values of 4.72 ± 1.22 and 7.83 ± 4.32 were seen in osseous lesions on [¹⁸F]-FDG and [¹⁸F]AlF-RESCA-RT4, respectively (P < 0.001). Moreover, no significant correlations were found between the SUVmax values of [¹⁸F]AlF-RESCA-RT4 and [¹⁸F]-FDG (Fig. 6C, P > 0.05). A 53-year-old man with PCa underwent PET/CT for initial staging (Fig. 7). [¹⁸F]AlF-RESCA-RT4 and [¹⁸F]-FDG PET/CT visualized different imaging features. More metastases were found on [¹⁸F]AlF-RESCA-RT4 PET/CT imaging, with 25 lesions detected on [¹⁸F]AlF-RESCA-RT4 and 15 lesions detected on [¹⁸F]-FDG. [¹⁸F]AlF-RESCA-RT4 PET/CT imaging had the highest contrast in both bones and lymph nodes with SUVmax of 7.87 ± 3.16, which were higher than [¹⁸F]-FDG with SUVmax of 4.32 ± 1.51. And other representative [¹⁸F]AlF-RESCA-RT4 and [¹⁸F]-FDG images of PCa patients were presented in Additional file 1: Fig. S3. These findings confirm the clinical feasibility of [¹⁸F]AlF-RESCA-RT4 for mapping systemic metastases of PCa.

Fig. 5.

Physiological high uptake of [¹⁸F]AlF-RESCA-RT4 by prostate affects the diagnosis of primary tumor. A 58-year-oldman with prostate cancer. A–C Multi-parametric magnetic resonance imaging suggested a nodule in the left peripheral band of the prostate, with high signal intensity on T2-weighted image and markedly restricted diffusion (red arrows). D [¹⁸F]AlF-RESCA-RT4 showed intense physiological uptake in prostate (red dotted line), which significantly affected identification of primary lesion

Fig. 6.

The comparison of [¹⁸F]AlF-RESCA-RT4 and [¹⁸F]-FDG uptake in prostate cancer metastases. A The number and B SUVmax distribution of lymph node and bone metastases of [¹⁸F]AlF-RESCA-RT4 and [¹⁸F]-FDG. C The relationship between [¹⁸F]AlF-RESCA-RT4 SUVmax and [¹⁸F]-FDG SUVmax was analyzed with curve fittings on lesion-based analysis. All metastases were detected from 10 PCa patients. ****P < 0.0001

Fig. 7.

Representative PET/CT images in a patient with prostate cancer. A 53-year-old man with prostate cancer underwent PET/CT for initial staging. A [¹⁸F]AlF-RESCA-RT4 showed intense tracer uptake in widespread bone metastases (b and c, red arrowheads) and multiple regional lymph nodes (including left clavicular region, mediastinum, posterior diaphragm, para-abdominal aorta, and bilateral para-iliac vessel lymph nodes; e and g, blue arrowheads). B Corresponding lesions showed low or mild FDG uptake

Discussion

For patients with advanced metastatic PCa, targeted therapeutic strategies are essential for prolonging survival. Prostate-specific membrane antigen (PSMA, also known as FOLH1) is the most extensively validated theranostic target in PCa [27]. Despite demonstrated clinical utility, up to 40% of CRPC patients show heterogeneous or diminished PSMA expression during disease progression, with near-universal loss observed in neuroendocrine prostate cancer (NEPC) [28–30]. Furthermore, prolonged ADT has been shown to reduce PSMA expression in PCa lesions [31]. These limitations underscore the critical need for alternative targeted strategies to ensure effective therapeutic management.

As a pan-cancer biomarker, Trop2 exhibits limited expression in normal tissues but is overexpressed across various solid tumors [9]. The effectiveness of SG in managing breast cancer validates Trop2 as a promising therapeutic target for cancer treatment [11]. For PCa, Trop2 is not only significantly elevated in PCa cell surface but also drives tumor growth and metastasis [13]. Weiten et al. reported substantial Trop-2 expression in cerebral metastases and provided preclinical in vitro evidence supporting the efficacy of SG in treating cerebral metastatic castration-resistant prostate cancer (mCRPC) [32]. Moreover, Trop2 has also been linked to the neuroendocrine differentiation of PCa cells, which are resistant to standard therapies and are associated with poor outcomes [33]. After approval for the treatment of metastatic triple-negative breast cancer, Trop2-based ADCs are being tested in a phase II trial (NCT03725761) on CRPC patients with progression on abiraterone or enzalutamide. However, the variable Trop2 expression level in PCa may potentially affect the efficacy of ADC treatment response, necessitating periodic reassessment before and after treatment [34]. Trop2 expression in tumors has primarily been assessed through IHC staining of pathological specimens, a method that is invasive and fails to capture the spatial and temporal heterogeneity of tumors [17]. This limitation highlights a growing need for non-invasive imaging techniques to identify patients most likely to benefit from Trop2-targeted therapies and to inform future clinical trials.

Immuno-PET is a paradigm-shifting molecular imaging modality that combines high specificity with the inherent sensitivity of PET technology [18]. In recent years, nanobodies have provided the advantage of rapid targeting and clearance, making them ideal for designing novel immuno-PET probes [35]. Our previous studies have demonstrated the feasibility of Trop2-targeted radiotracers, including [⁶⁸Ga]Ga-NOTA-T4, [¹⁸F]AlF-RESCA-T4, and [¹⁸F]AlF-RESCA-RT4, in pancreatic, nasopharyngeal, and lung cancers [19–21]. In this study, we further demonstrated the value of these probes in prostate cancer, broadening the clinical application of Trop2-targeted immuno-PET imaging. High and specific tumor uptake of [⁶⁸Ga]Ga-NOTA-T4 and [¹⁸F]AlF-RESCA-T4 was observed in the prostate cancer PC3 model, with ROI values of 4.33 ± 0.38 %ID/g and 5.50 ± 0.69 %ID/g, respectively. In addition, our preclinical data verified that the potential application of Trop2-targeted immuno-PET imaging in PSMA-negative PCa. However, significant tracer accumulation in the kidneys was also noted. Most nanobodies are filtered by the kidneys and reabsorbed by renal tubular epithelial cells, where they are degraded in lysosomes [36]. This process leads to prolonged accumulation of radioactivity in the renal cortex, which may reduce diagnostic accuracy for kidney-related conditions and raise concerns about potential toxicity if the probe were adapted for therapeutic applications. To address this issue, we have explored several strategies to reduce renal retention [37–39]. In our previous work, we removed the His-tag from T4 and developed the His-tag-free nanobody tracer [¹⁸F]AlF-RESCA-RT4, which substantially reduced kidney accumulation without significantly compromising tumor uptake [19]. Building on these findings, we used the His-tag-free [¹⁸F]AlF-RESCA-RT4 probe in this clinical study, which effectively identified metastatic lesions. On the other hand, the background physiological uptake in the prostate must be considered when interpreting our imaging results. The limited diagnostic value of [¹⁸F]AlF-RESCA-RT4 for detecting primary intra-prostatic lesions might be attributed to several factors. While bioinformatics analysis revealed Trop2 overexpression in PCa at the transcriptomic level, the heterogeneity of primary tumors and the presence of benign prostatic hyperplasia or prostatitis within the same gland may lead to variable and non-specific signals. Additionally, the spatial resolution of PET/CT may hinder the visualization of small, intra-prostatic tumors against this potential background. In contrast, [¹⁸F]AlF-RESCA-RT4 exhibited high uptake in metastatic sites with higher SUVmax values compared to [¹⁸F]-FDG, highlighting the potential of Trop2 immuno-PET/CT as a valuable tool for detecting metastatic PCa. The strong tumor uptake observed with [¹⁸F]AlF-RESCA-RT4 suggested that Trop2 imaging may offer clinical benefits in staging and treatment monitoring.

In addition to Trop2, efforts have been made to develop new targets for PCa, such as membrane cofactor protein (CD46), delta-like ligand 3 (DLL3), erythropoietin-producing hepatocellular A2 (EphA2), neurotensin receptor subtype 1 (NTR1), and carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAM5) [40–43]. Several tracers targeting these molecules have shown promising results in prostate cancer models, demonstrating high tumor uptake in preclinical evaluations. Tendler et al. suggested that an ⁸⁹Zr-labeled monoclonal antibody tracer, [⁸⁹Zr]Zr-DFO-SC16, could serve as a non-invasive PET imaging agent for detecting DLL3-positive NEPC lesions [44]. However, the large molecular weight of monoclonal antibodies leads to a prolonged circulation time and a low tumor-to-background ratio, necessitating the use of long half-live radionuclides for immuno-PET, which limits their clinical application. Gan et al. demonstrated the potential of EphA2 as a target for detecting PSMA-negative PCa and developed a radiolabeled bicyclic peptide, [¹⁸F]AlF-ETN, which specifically visualized EphA2-expressing PCa with high contrast [40]. Wu et al. reported that [⁶⁸Ga]Ga-DOTA-NT-20.3, which exhibits a high affinity for NTR1, has the potential for detecting poorly differentiated PCa or cases with neuroendocrine differentiation [45]. Despite these advances, translating novel targets into clinical practice remains challenging. Furthermore, integrating these new tracers into the current prostate cancer diagnostic workflow requires careful consideration of their complementary role alongside established imaging agents, such as PSMA-targeted radiotracers. For instance, in patients with suspected or biopsy-proven neuroendocrine differentiation, or in those with clinically aggressive disease and negative or low PSMA-PET expression, Trop2-PET can be deployed as a next-line imaging strategy to identify otherwise occult metastases. As research continues, the development of dual-targeted imaging strategies, combining PSMA with emerging biomarkers like Trop2, may offer a more comprehensive approach to characterizing tumor heterogeneity in prostate cancer. Such advancements could enhance precision oncology by enabling more personalized treatment strategies, ultimately improving outcomes for patients with treatment-resistant diseases.

In our study, Trop2-targeted immuno-PET has exhibited favorable pharmacokinetics and promising diagnostic potential in PCa. Trop2 is recognized a pan-cancer biomarker, and Trop2-ADCs have shown clinical efficacy in various tumors such as triple-negative breast cancer and non-small cell lung cancer. Therefore, Trop2-targeted radiotracers are likely to have broader utility beyond PCa. A key next step will be to evaluate their application in these additional malignancies, with the aim of identifying patients who may benefit from Trop2-targeted therapies and monitoring treatment response. Furthermore, with the development of the concept of cancer theranostics, Trop2-targeted nanobodies are expected to perform more therapeutic functions. However, the rapid blood clearance and renal retention of nanobodies are not conducive to transition these agents to targeted radionuclide therapies directly. The engineering of nanobody derivatives, including multivalent formats or albumin-binding mutants, represents a promising strategy to establish a versatile Trop2-targeted platform for subsequent radiotherapy. With an optimal profile of tumor uptake versus background toxicity established, nanobody radiotheranostics are poised to become a cornerstone of precision medicine.

Limitations still exist for the current study. First, this study includes few patients, and further validation in larger, multi-center cohorts is needed to confirm the tracer’s sensitivity, specificity, and clinical generalizability. Second, we cannot obtain complete IHC staining results for all the patients, which limits our ability to perform a fully correlative analysis between tracer uptake and target expression across the cohort. Third, we observed modest physiological uptake of the tracer in the prostate gland, indicating that further structural refinement is needed to optimize the tracer’s biodistribution. Finally, comprehensive dosimetry studies are also required to fully assess the tracer’s safety profile for potential clinical translation. Future prospective studies are warranted to expand our patient cohort to address these limitations. Furthermore, we plan to apply this imaging platform in patients undergoing Trop2-ADC therapy, which will allow direct assessment of its utility in guiding ADC treatment selection and monitoring therapeutic response.

Conclusions

We demonstrate that Trop2 is a promising biomarker for advanced PCa, and [¹⁸F]AlF-RESCA-RT4 shows potential for non-invasive, Trop2-specific imaging in detecting PCa lymph node and bone metastases. Further studies with larger, multicenter cohorts are needed to better confirm the clinical value of Trop2 immuno-PET/CT imaging in diagnosing PCa and guiding treatment decisions.

Abbreviations

PCa

Prostate cancer

Trop2

Trophoblast cell surface antigen 2

TACSTD2

Tumor-associated calcium signal transducer 2

ADT

Androgen deprivation therapy

CRPC

Castration-resistant prostate cancer

ADC

Antibody-drug conjugate

SG

Sacituzumab Govitecan

AR

Androgen receptor

IHC

Immunohistochemistry

Immuno-PET

Immuno-positron emission tomography

TIMER

Tumor IMmune Estimation Resource

GEPIA

Gene Expression Profiling Interactive Analysis

TCGA

The Cancer Genome Atlas

GTEx

Genotype-Tissue Expression databases

THPA

The Human Protein Atlas

GEO

Gene Expression Omnibus

PCA

Principal component analysis

UMAP

Uniform Manifold Approximation and Projection

p-SCN-Bn-NOTA/NOTA

P-Isothiocyanatobenzyl-1,4,7-triazacyclononane-1,4,7-triacetic acid

(±)-H3RESCA-TFP/RESCA

(±)-2,2′,2″-(1,4,7-Triazonane-1,4,7-triyl)triacetic acid-4-((2,3,5,6-tetrafluorophenoxy)methyl)benzyl)

RCP

Radiochemical purity

AST

Aspartate aminotransferase

ALT

Alanine aminotransferase

BUN

Blood urea nitrogen

PARD

Prostate cancer

TBR

Tumor-to-background ratio

ROI

Region of interest

mCRPC

Metastatic castration-resistant prostate cancer

CD46

Membrane cofactor protein

DLL3

Delta-like ligand 3

NTR1

Neurotensin receptor subtype 1

CEACAM5

Carcinoembryonic antigen-related cell adhesion molecule 5

EphA2

Erythropoietin-producing hepatocellular A2

Authors’ contributions

DJ, LD, and XL2 designed and conceived the project. XL1 and XN performed the preclinical experiments. SA and LD recruited the patients for clinical trials. SA, DX, and WW performed the PET/MR image acquisition and interpretation. XL1 and SA verified the data. XL1 and SA drafted the original manuscript. DJ, LD, and XL2 reviewed and edited the manuscript. All authors read and approved the final manuscript.

Funding

This work was funded by the Innovative Drug Research and Development-National Science and Technology Major Project (No. 2025ZD1803403), the National Natural Science Foundation of China (22277031 and 82030052), Hubei Science and Technology Innovation Talent Program (2023DJC162), and Wuhan Union Hospital.

Data availability

Data supporting the findings of bioinformatics analysis in this study are available in the Gene Expression Profiling Interactive Analysis (GEPIA) database (http://gepia.cancer-pku.cn/), the Tumor IMmune Estimation Resource (TIMER2.0) database (http://timer.cistrome.org/), and the Human Protein Atlas (THPA) (https://www.proteinatlas.org/). Single-cell RNA-sequencing data are derived from the NCBI Gene Expression Omnibus (GEO) repository under accession number GSE176031 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE176031). All other data that underlie the results reported in this article will be available from the corresponding author, Dawei Jiang, upon reasonable request.

Declarations

Ethics approval and consent to participate

The animal study protocol was approved by the Institutional Animal Care and Use Committee at Tongji Medical College, Huazhong University of Science and Technology (Approval Number: 4494). The human study was approved by the Ethics Committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University (Approval Number: LY2024-307-A) and registered in ClinicalTrials.gov (NCT06851663). Written informed consent forms were obtained from all the included patients.

Consent for publication

All patients gave written informed consent. All authors have read and approved of its submission to this journal.

Competing interests

The authors declare no competing interests.