Proof-of-concept PET imaging of pulmonary sarcoidosis using VAP-1-targeted radiotracer [⁶⁸Ga]Ga-DOTA-Siglec-9

Abstract

Background

Sarcoidosis is a multisystem granulomatous disease of unknown etiology, with pulmonary involvement being the most common and clinically significant manifestation. Vascular adhesion protein-1 (VAP-1) plays a key role in leukocyte trafficking to inflamed tissues. [⁶⁸Ga]Ga-DOTA-Siglec-9 is a novel PET radiotracer that binds to VAP-1. This proof-of-concept study aimed to evaluate the feasibility of [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT for imaging pulmonary sarcoidosis.

Methods

Six patients with stage 2 pulmonary sarcoidosis (age 50.5 ± 13.1 years; bodyweight 84.2 ± 14.7 kg), diagnosed by clinical, radiological, and histological findings, underwent [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT. Control subjects included six healthy male volunteers (age 37.5 ± 10.3 years; bodyweight 80.3 ± 3.9 kg) and one female patient with lung cancer (age 77 years; bodyweight 62 kg). Tracer uptake was quantified in the lungs, mediastinal lymph nodes, and organs involved in systemic inflammation.

Results

Patients with sarcoidosis showed significantly higher [⁶⁸Ga]Ga-DOTA-Siglec-9 uptake in the lungs (SUV_mean 1.82 ± 0.52 vs. 0.41 ± 0.08; P = 0.00006) and mediastinal lymph nodes (SUV_mean 2.06 ± 0.46 vs. 0.89 ± 0.26; P = 0.0003) compared to healthy controls. Increased uptake was also observed in the liver (SUV_mean 1.18 ± 0.14 vs. 0.80 ± 0.10, P = 0.0003), spleen (SUV_mean 1.13 ± 0.09 vs. 0.82 ± 0.06, P = 0.00003), bone marrow (SUV_mean 0.30 ± 0.12 vs. 0.06 ± 0.05, P = 0.001), and bone (SUV_mean 0.27 ± 0.11 vs. 0.08 ± 0.04, P = 0.004), indicating systemic inflammation.

Conclusions

This proof-of-concept study demonstrates the potential of VAP-1-targeted [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT for imaging pulmonary sarcoidosis and associated inflammatory activity. Further validation in larger cohorts is warranted.

Trial registration

ClinicalTrials.gov, NCT03755245 (registered 27 November 2018), NCT05212103 (registered 27 January 2022).

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-025-03455-8.

Introduction

Sarcoidosis is a chronic multisystem inflammatory disorder, typically characterized by the presence of noncaseating granulomas in affected organs [1]. Although its precise etiology remains unknown, sarcoidosis most commonly affects the lungs and mediastinal lymph nodes, though any organ system including heart, liver, eyes, skin, spleen, or nervous system, can be involved [2, 3]. Diagnosis is based on a combination of clinical and radiological findings, supported by histological confirmation of granulomas [4, 5]. However, in the absence of a disease-specific biomarker, noninvasive and accurate imaging modalities are urgently needed to improve diagnostic precision, especially in cases where tissue biopsy is not feasible.

2-Deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) PET/CT is a widely used imaging modality for evaluating metabolically active granulomatous inflammation, offering high signal-to-noise ratios and good resolution [6, 7]. Despite its clinical utility, [¹⁸F]FDG is not specific to sarcoidosis, as it accumulates in various malignant and non-malignant inflammatory processes. This lack of specificity can complicate differential diagnosis, particularly when increased uptake is seen in lymph nodes or organs commonly involved in malignancy, such as lymphoma or metastatic disease [8, 9]. Thus, there is a pressing need for more specific molecular imaging tools that selectively target the inflammatory mechanisms unique to sarcoidosis.

Vascular adhesion protein-1 (VAP-1), also known as amine oxidase copper-containing 3 (AOC3), is a dual-function molecule involved in leukocyte trafficking. It mediates rolling, adhesion, and transendothelial migration of leukocytes during inflammation through both adhesive and enzymatic activity [10]. In resting tissues, VAP-1 resides in intracellular granules of endothelial cells, but rapidly translocates to the cell surface in response to inflammatory stimuli [11]. A soluble form of VAP-1 (sVAP-1), likely derived from the cleavage of its membrane-bound counterpart, is also detectable in circulation, with elevated levels reported in certain inflammatory diseases [12, 13].

Sialic acid-binding immunoglobulin-like lectin 9 (Siglec-9), primarily expressed on neutrophils, monocytes, macrophages, and dendritic cells [14], serves as a natural ligand for VAP-1 [15]. In response to inflammatory signals, surface expression of Siglec-9 is rapidly upregulated, making it a promising molecular probe for imaging VAP-1 activity [15]. A gallium-68-labeled peptide incorporating residues 283–297 of Siglec-9, conjugated to 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid ([⁶⁸Ga]Ga-DOTA-Siglec-9) has shown efficacy for PET imaging of inflammation in preclinical models of inflammatory bowel disease [16], myocarditis [17], arthritis [18], atherosclerosis [19], and acute lung injury [20]. To date, the only reported clinical use of this tracer has been in a patient with rheumatoid arthritis to image synovial inflammation [21].

In this proof-of-concept study, we investigated the feasibility of using [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT for VAP-1-targeted molecular imaging of pulmonary sarcoidosis and associated mediastinal lymphadenopathy.

Materials and methods

Supplementary materials and methods are available online.

Study population

The study enrolled six patients (3 females, 3 males; age, 50.5 ± 13.1 years; body weight, 84.2 ± 14.7 kg) diagnosed with active, stage 2 pulmonary sarcoidosis, as confirmed by histology and high-resolution computed tomography (HRCT) imaging. The inclusion criteria were as follows: age 18 years or older; a diagnosis of active pulmonary sarcoidosis confirmed by histology, HRCT, and/or a serum angiotensin convertase (ACE) concentration > 60 U/L (excluding patients on ACE inhibitors); and the ability and willingness to provide written informed consent and comply with the study protocol. Exclusion criteria included diagnosis of uncontrolled type 2 diabetes, and evidence of significant uncontrolled concomitant diseases such as neurological, renal, hepatic, endocrine, or gastrointestinal disorders. Six healthy male volunteers (age 37 ± 9 years; bodyweight 80 ± 4 kg) were included as controls [21], having been recruited through local advertisements and screened at the study center. The absence of significant medical, neurological, or psychiatric history, as well as alcohol or drug abuse, was confirmed through questionnaires. Additionally, each subject underwent a medical history review, routine blood tests, electrocardiography, and physical examination. A 77-year-old female patient (weight 62 kg) with adenocarcinoma in the upper lobe of the left lung was included as a diseased control. The study design is shown in Fig. 1.

Fig. 1.

Study design

The study was approved by the joint Ethics Committee of the University of Turku and Turku University Hospital (ETMK 35/1800/2018), and by the Finnish Medicines Agency (EudraCT 2018-001002-29). Each subject provided informed consent before entering the study. The study was registered as a clinical trial (NCT03755245; registered 27 November 2018). A lung cancer patient participated in another clinical trial NCT05212103 (registered 27 January 2022; ethical permission ETMK 101/2021), but dropped out after the first scan.

Radiopharmaceuticals

All PET radiopharmaceuticals used in the study were prepared according to Good Manufacturing Practices guidelines at the Radiopharmaceutical Laboratory of Turku PET Centre under Turku University Hospital Pharmacy control. The [⁶⁸Ga]Ga-DOTA-Siglec-9 tracer was synthesized using an automated module [22]. The [¹⁸F]FDG tracer was synthesized using a FASTlab synthesizer [23].

PET/CT scanners and radiation exposure

Imaging was conducted using the Discovery 690 and Discovery MI PET/CT systems (General Electric Medical Systems). The effective radiation dose for a 140 MBq injection of [⁶⁸Ga]Ga-DOTA-Siglec-9 was estimated to be 3.08 mSv (0.022 mSv/MBq) [21], and that for a 200 MBq dose of [¹⁸F]FDG was estimated to be 3.8 mSv (0.019 mSv/MBq) [24]. In the [¹⁸F]FDG PET/CT study, the total radiation dose ranged from 8.0 to 12.0 mSv. In the [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT study, the total radiation dose ranged from 7.6 to 11.3 mSv. Imaging of a lung cancer patient was performed using Vision Quadra (Siemens Healthineers).

PET/CT protocol

Before PET/CT, the patients underwent screening procedures and a physical examination. Females of childbearing potential had a urine pregnancy test at the time of screening and on the day of, or within 24 h prior to, the scheduled PET/CT scans. The PET/CT studies were conducted after a minimum fasting period of 6 h. A catheter was inserted into an antecubital vein for tracer injection, and another was inserted into the opposite radial vein for blood sampling to assess tracer stability and the plasma concentration of radioactivity. The subjects were positioned supine, with their arms at their sides. [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT and [¹⁸F]FDG PET/CT scans were performed on consecutive days. A low-dose whole-body CT scan (100 kV, 10−20 mA, noise index 170) was performed for attenuation correction and anatomical reference.

Sarcoidosis patients were injected intravenously with 157.7 ± 22.1 MBq of [⁶⁸Ga]Ga-DOTA-Siglec-9 as a bolus, followed immediately by dynamic PET acquisition of the lung area over 30 min. Thereafter, a whole-body PET scan was performed from head to toe, which required approximately 12 bed positions, with a 2-min acquisition time per position. Blood samples were collected post-injection to analyze [⁶⁸Ga]Ga-DOTA-Siglec-9 stability and the radioactivity concentration. For [¹⁸F]FDG PET/CT, patients were injected with 218.7 ± 59.1 MBq of the tracer, and a whole-body scan was performed 60 min later, as described above. [¹⁸F]FDG PET/CT was not performed on healthy control subjects.

Healthy control subjects received a bolus of [⁶⁸Ga]Ga-DOTA-Siglec-9 (162.2 ± 3.8 MBq) intravenously, followed by semi-dynamic whole-body PET/CT scans as previously described [21].

The injected dose of [⁶⁸Ga]Ga-DOTA-Siglec-9 for the diseased control subject was 150.1 MBq. After 30-min dynamic PET of the lung area, a whole-body scan was performed.

Dynamic [⁶⁸Ga]Ga-DOTA-Siglec-9 PET images were reconstructed using VUEPoint HD (VPHD), SharpIR (General Electric), and Q.Clear, with a β value of 600. Whole-body [⁶⁸Ga]Ga-DOTA-Siglec-9 images were reconstructed using SharpIR, as well as VUEPoint FX/Q.Clear, with a β value of 600. For the [¹⁸F]FDG PET scans, image reconstruction was performed using SharpIR, as well as VUEPoint FX/Q.Clear, with β value of 350.

Quantitative analysis of PET images

PET image analysis was performed by manually defining regions of interest (ROIs) along the anatomical boundaries of the lung parenchyma, from apex to base, on each transverse slice of the fused PET/CT images. The diaphragm, trachea, and mainstem bronchi were excluded from the ROIs. To assess lymphadenopathy, ROIs were placed on mediastinal lymph nodes at multiple stations, including the upper and lower paratracheal, subcarinal, paraesophageal, pulmonary ligament, and hilar regions (Fig. 2A). Carimas 2.10 software (Turku PET Centre) was used to define ROIs in various organs and tissues, including the cortical bone (femur), blood pool (left ventricle cavity), myocardium, kidneys, liver, lungs, skeletal muscle (triceps brachii), pancreas, femoral bone marrow, salivary glands, spleen, and urinary bladder content. The results were expressed primarily as the mean standardized uptake value (SUV_mean), which was calculated as a ratio of tissue radioactivity concentration (Bq/mL) and given radioactivity dose (Bq) divided by subject’s body weight (kg). In addition, SUV_max values and target-to-blood and target-to-muscle ratios were calculated to further characterize tracer uptake and lesion-to-background contrast. Carimas software was used to visually assess myocardial uptake of [⁶⁸Ga]Ga-DOTA-Siglec-9 on images fused with [¹⁸F]FDG.

Fig. 2.

(A) Representative transaxial high-resolution computed tomography (HRCT), [¹⁸F]FDG PET/CT, and [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT images from a patient with stage 2 pulmonary sarcoidosis. The lung parenchymal region of interest is delineated by the dashed white line. Mediastinal lymph nodes with high uptake are delineated by red dashed lines. Quantification of [⁶⁸Ga]Ga-DOTA-Siglec-9 PET imaging in (B) lung parenchyma and (C) mediastinal lymph nodes((

Histology and immunohistochemistry

Formalin-fixed paraffin-embedded lymph node, liver, and bronchial tissue samples were obtained from Auria Biobank (agreement #AB21-2675; Turku, Finland). For general histology, hematoxylin-eosin (HE) staining was performed on 4 μm sections. For immunohistochemistry, sections were stained with antibodies specific for VAP-1 (polyclonal VAP-1 antibody produced in rabbits against recombinant human VAP-1, dilution 1:5000, incubation overnight at + 4 °C; custom production by Harlan Sera-Lab Ltd.) or CD68 (monoclonal mouse anti-human CD68, prediluted, incubation 1.5 h at room temperature; ab845, Abcam). VAP-1 staining was performed without antigen retrieval, endogenous peroxidase blocked with Bloxall-solution (SP-6000, Vector Laboratories, Burlingame) followed by blocking with normal serum. Detection was performed with Vectastain rabbit ABC-HRP kit (PK-6101; Vector Laboratories) and 3,3’-diaminobenzidine (DAB) as a chromogen (K3468; Dako), followed by counterstaining with hematoxylin and mounting. CD68 staining was performed using fully-automated Ventana Benchmark staining system (Ventana Medical Systems), pre-treatment with Tris-based buffer (CC1 pH 8, #05424569001, Roche), and detection with Ventana ultraView Universal DAB detection kit (#05269806001, Roche). Stained sections were scanned with a digital slide scanner (Pannoramic P1000, 3DHISTECH Ltd.), and examined using Pannoramic Viewer 1.15 software (3DHISTECH Ltd.).

sVAP-1 and other laboratory tests

Prior to tracer injection for [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT, serum samples were collected and the concentration of circulating sVAP-1 was measured using an in-house sandwich enzyme-linked immunosorbent assay (ELISA), as previously described [25]. In addition, blood samples were collected for hematology, serology, and clinical chemistry.

Statistical analysis

Data are presented as the mean ± standard deviation (SD). The normality of continuous variables was assessed using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Variables that deviated significantly from normality were log-transformed before statistical analyses. Independent-samples t-tests were used to compare normally distributed [⁶⁸Ga]Ga-DOTA-Siglec-9 uptake values between patients with pulmonary sarcoidosis and healthy control subjects. Statistical analyses were performed using IBM SPSS version 26.0. GraphPad Prism version 8.0.2 was used to generate graphical representations of the results. A P-value < 0.05 was considered statistically significant.

Results

Characteristics of study population

The subject characteristics are summarized in Table 1. There were no significant differences between the patient and control groups regarding age (50.5 ± 13.2 vs. 37.5 ± 10.3 years; P = 0.088), bodyweight (84.2 ± 14.7 vs. 80.3 ± 3.9 kg; P = 0.561), or body mass index (BMI 26.8 ± 1.5 vs. 24.6 ± 2.6 kg/m²; P = 0.110). Similarly, there was no significant difference in the injected [⁶⁸Ga]Ga-DOTA-Siglec-9 radioactivity dose (157.7 ± 22.1 vs. 162.2 ± 3.8 MBq; P = 0.580).

Table 1.

Clinical and anthropometric characteristics of study subjects

	Pulmonary sarcoidosis (n = 6)	Healthy controls (n = 6)	P value	Lung cancer (n = 1)
Age (years)	50.5 ± 13.2	37.5 ± 10.3	0.088	77
Sex (female/male)	3/3	0/6	0.076	1/0
Weight (kg)	84.2 ± 14.7	80.3 ± 3.9	0.561	62
Body mass index (kg/m2)	26.8 ± 1.5	24.6 ± 2.6	0.110	21.0
Smoking status (smoker/non-smoker)	0/6	0/6	ND	1/0
[68Ga]Ga-DOTA-Siglec-9 dose (MBq)	157.7 ± 22.1	162.2 ± 3.8	0.641	150.1
Lysozyme (U/L)	22.8 ± 10.5	ND	ND	ND
Angiotensin-converting enzyme (U/L)	79.3 ± 31.9	ND	ND	ND
Calcium ions (mmol/L)	1.33 ± 0.20	ND	ND	ND
Sarcoidosis onset (months)	33.1 ± 50.8	ND	ND	ND
sVAP-1 (ng/mL)	1150.0 ± 150.2	874.0 ± 139.9	0.037	ND

The results are expressed as the mean ± SD. sVAP-1, soluble vascular adhesion protein-1

ND Not determined

P values are from the independent-samples t-test

The diagnosis of sarcoidosis was confirmed by the presence of granulomatous inflammation on biopsy, along with elevated levels of lysozyme (range 12−40; normal value < 12 U/L), ACE (range 49−140; normal value < 60 U/L), and calcium ions (range 1.19−1.67; normal range, 1.16−1.31 mmol/L). All patients had radiologically confirmed stage 2 pulmonary sarcoidosis. Additionally, HRCT findings in the lungs included pleural and peribronchovascular nodularity, parenchymal consolidation, ground-glass opacity, crazy paving, interlobular septal thickening, and calcification (Supplementary Table 1).

The concentration of sVAP-1 was significantly higher in subjects with pulmonary sarcoidosis than in healthy controls (1150.0 ± 150.2 vs. 874.0 ± 139.9 ng/mL; P = 0.037). The hematology, serology, and clinical chemistry results for the study subjects are shown in Supplementary Table 2.

[⁶⁸Ga]Ga-DOTA-Siglec-9 uptake in lungs and mediastinal lymph nodes

In subjects with pulmonary sarcoidosis, the [¹⁸F]FDG scan results revealed consistently prominent and abnormal activity in the hilar and mediastinal lymph nodes, with SUV_mean ranging from 4.2−12.0. Multiple hypermetabolic lymph nodes were identified in the neck, predominantly in the supraclavicular region (SUV_mean 3.1−6.6). Active para-aortic lymph nodes were also observed, with SUV_mean ranging from 9.6−19.1 (Supplementary Figs. 1–6). For descriptive purposes, the corresponding SUV_max values for [¹⁸F]FDG and [⁶⁸Ga]Ga-DOTA-Siglec-9 in affected regions are summarized in Supplementary Table 3. It is important to note that these [¹⁸F]FDG images are presented for illustrative purposes only and are not intended for direct comparison. The focus of this study remains on demonstrating the imaging characteristics of [⁶⁸Ga]Ga-DOTA-Siglec-9 PET in sarcoidosis relative to healthy controls.

We found that the uptake of [⁶⁸Ga]Ga-DOTA-Siglec-9 in the lungs of subjects with pulmonary sarcoidosis was significantly higher than that in healthy controls (SUV_mean 1.82 ± 0.52 vs. 0.41 ± 0.08; P = 0.00006, Fig. 2B, Table 2). Additionally, uptake in mediastinal lymph nodes was significantly higher in patients than in controls (SUV_mean 2.06 ± 0.46 vs. 0.89 ± 0.26; P = 0.0003) (Fig. 2C, Table 2). In the diseased control, [⁶⁸Ga]Ga-DOTA-Siglec-9 accumulation was observed in the lung adenocarcinoma (SUV_mean 1.91) and in mediastinal lymph node (SUV_mean 2.46). For comparison, the SUV_max values for [⁶⁸Ga]Ga-DOTA-Siglec-9 uptake are summarized in Table 3. Fig. 3 shows [⁶⁸Ga]Ga-DOTA-Siglec-9 PET images comparing a patient with sarcoidosis and a healthy control.

Table 2.

Tissue-specific [⁶⁸Ga]Ga-DOTA-Siglec-9 uptake as SUV_mean

	Pulmonary sarcoidosis (n = 6)	Healthy controls (n = 6)	P value	Lung cancer (n = 1)
Blood, heart left ventricular cavity	1.48 ± 0.23	1.36 ± 0.13	0.271	2.90
Bone, cortical	0.27 ± 0.11	0.08 ± 0.04	0.004	0.59
Bone marrow	0.30 ± 0.12	0.06 ± 0.05	0.001	0.46
Kidneys	3.45 ± 0.85	6.22 ± 3.49	0.086	6.64
Liver	1.18 ± 0.14	0.80 ± 0.1	0.0003	1.81
Lung cancer lesion	ND	ND	ND	1.91
Lung parenchyma	1.82 ± 0.52	0.41 ± 0.08	0.00006	0.53
Mediastinal lymph nodes	2.06 ± 0.46	0.89 ± 0.26	0.0003	2.46
Muscle, triceps brachii	0.31 ± 0.07	0.30 ± 0.06	0.761	0.56
Myocardium	0.77 ± 0.24	0.83 ± 0.24	0.715	1.98
Pancreas	0.89 ± 0.32	0.66 ± 0.20	0.181	1.84
Parotid gland	0.61 ± 0.08	0.59 ± 0.18	0.822	1.59
Spleen	1.13 ± 0.09	0.82 ± 0.06	0.00003	1.95
Salivary gland, submandibular	0.82 ± 0.21	0.62 ± 0.17	0.093	1.95
Thymus	0.51 ± 0.16	0.91 ± 0.27	0.013	1.09
Urinary bladder content	96.52 ± 89.28	71.9 ± 45.31	0.865	33.17

The results are expressed as the mean standardized uptake values (SUV_mean, mean ± SD)

ND Not determined

P values are from the independent-samples t-test

Table 3.

Tissue-specific [⁶⁸Ga]Ga-DOTA-Siglec-9 uptake as SUV_max

	Pulmonary sarcoidosis (n = 6)	Healthy controls (n = 6)	P value	Lung cancer (n = 1)
Blood, heart left ventricular cavity	2.91 ± 0.86	2.17 ± 0.57	0.054	4.04
Bone, cortical	0.49 ± 0.08	0.23 ± 0.13	0.001	0.69
Bone marrow	0.58 ± 0.13	0.31 ± 0.22	0.012	0.62
Kidneys	15.56 ± 11.76	9.05 ± 2.89	0.109	6.95
Liver	2.61 ± 0.94	1.47 ± 0.46	0.012	2.30
Lung cancer lesion	ND	ND	ND	2.44
Lung parenchyma	2.01 ± 0.44	0.62 ± 0.20	0.0002	1.06
Mediastinal lymph nodes	2.49 ± 0.53	1.19 ± 0.29	0.0001	3.55
Muscle, triceps brachii	0.64 ± 0.21	0.53 ± 0.24	0.218	0.87
Myocardium	1.54 ± 0.38	1.11 ± 0.31	0.030	2.58
Pancreas	1.83 ± 0.56	1.13 ± 0.43	0.018	3.45
Parotid gland	1.09 ± 0.42	0.86 ± 0.29	0.152	2.18
Spleen	2.18 ± 0.62	1.19 ± 0.27	0.003	2.90
Salivary gland, submandibular	1.16 ± 0.48	0.94 ± 0.23	0.163	2.73
Thymus	1.10 ± 0.40	1.10 ± 0.50	0.491	1.54
Urinary bladder content	127.26 ± 140.92	80.03 ± 46.04	0.227	35.78

The results are expressed as the maximum standardized uptake values (SUV_max, mean ± SD)

ND Not determined

P values are from the independent-samples t-test

Fig. 3.

[⁶⁸Ga]Ga-DOTA-Siglec-9 PET maximum intensity projection (MIP) images of a patient with histologically confirmed sarcoidosis and a healthy control. The patient shows increased tracer uptake in the liver and mediastinum compared with the control

The time-activity curves extracted from 30-min dynamic [⁶⁸Ga]Ga-DOTA-Siglec-9 PET of the thoracic region confirmed higher uptake in the mediastinal lymph nodes than in the lung parenchyma of patients with sarcoidosis (Fig. 4).

Fig. 4.

Time-activity curves for 30-min dynamic [⁶⁸Ga]Ga-DOTA-Siglec-9 PET in the thoracic region of sarcoidosis patients show uptake in various tissues. SUV_mean: mean standardized uptake value

[⁶⁸Ga]Ga-DOTA-Siglec-9 PET findings in other tissues

Uptake of [⁶⁸Ga]Ga-DOTA-Siglec-9 in the ^corticalbone, bone marrow, liver, and spleen, were significantly higher in patients with pulmonary sarcoidosis than in healthy controls (Table 2). Whole-body [⁶⁸Ga]Ga-DOTA-Siglec-9 PET images (Fig. 3) highlight increased tracer uptake in the liver and mediastinum of the sarcoidosis patient compared with the healthy control. Conversely, the uptake in the thymus was significantly lower in patients with pulmonary sarcoidosis than in control subjects. The [⁶⁸Ga]Ga-DOTA-Siglec-9 signal within the myocardium was homogenous and uniformly low (lower than in the blood) throughout the left ventricle in all patients (Supplementary Fig. 7). Additionally, none of the patients had a clinical history of cardiac sarcoidosis.

There was no statistically significant difference in uptake of [⁶⁸Ga]Ga-DOTA-Siglec-9 by other organs such as the muscle, myocardium, pancreas, parotid gland, and salivary glands (Table 2), or in the plasma radioactivity concentration (Supplementary Fig. 8), between subjects with sarcoidosis and healthy controls. The time-activity curves of [⁶⁸Ga]Ga-DOTA-Siglec-9 in sarcoidosis patients (Fig. 4) show that radioactivity concentration was highest in the liver, followed by the myocardium, thymus, and muscle, and declined to a steady level approximately 10–15 min after injection.

While the above-mentioned results are based on SUV_mean values, the corresponding SUV_max results are presented in Table 3 for comparison. Supplementary Table 3 shows that [⁶⁸Ga]Ga-DOTA-Siglec-9 generally exhibited higher SUV_max than [¹⁸F]FDG across most tissues in patients with sarcoidosis, with significantly higher uptake observed in the blood and lung parenchyma. Tissue-to-blood and tissue-to-muscle ratios for patients with sarcoidosis and healthy controls are provided in Supplementary Tables 4 and 5, respectively.

Histology and immunostaining

Representative images showing HE, CD68 (a marker of macrophages and histiocytes), and VAP-1 staining of lymph node, liver, and bronchus biopsies collected from study subjects diagnosed with sarcoidosis are shown in Fig. 5. In all the samples, epithelioid histiocytes (constituting sarcoid granulomas), surrounding macrophages/histiocytes, and Kupffer cells in the liver sinusoids stained positive for CD68. Endothelial cells within vascular structures of the lymph nodes were positive for VAP-1. In the liver, the sinusoidal endothelium, and especially previously injured and fibrotic areas adjacent to the granulomas, were VAP-1- positive, as were vascular endothelial cells. In addition, strong staining of the stroma beneath the basement membrane was observed.

Fig. 5.

Hematoxylin-eosin (HE) staining, and immunohistochemical staining of cluster of differentiation 68 (CD68; macrophages) and vascular adhesion protein-1 (VAP-1) in lymph node (cases 1 and 2), liver (case 4), and bronchus (case 5) samples from different patients with sarcoidosis. Black arrows indicate CD68-positive sarcoid granulomas. Black arrowheads indicate VAP-1-positive endothelium and stroma

Discussion

This proof-of-concept study is the first to demonstrate the use of [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT for evaluating pulmonary sarcoidosis and associated mediastinal lymphadenopathy. Our findings reveal significantly higher tracer uptake in the lung parenchyma and mediastinal lymph nodes of patients with pulmonary sarcoidosis compared with healthy controls, indicating the potential utility of this VAP-1-targeted imaging approach for detecting active granulomatous inflammation.

Histologically, sarcoidosis is characterized by well-defined, non-necrotizing granulomas composed of epithelioid histiocytes, typically following a lymphangitic distribution [26, 27]. Although [¹⁸F]FDG PET/CT is widely used to evaluate sarcoidosis and other leukocyte-mediated disorders [28–32], its clinical utility is limited by non-specific uptake in any metabolically active tissue [8, 9]. This hampers differentiation between inflammatory and malignant lesions, such as lymphoma or metastatic disease. In contrast, [⁶⁸Ga]Ga-DOTA-Siglec-9 targets VAP-1, a molecule specifically upregulated in inflamed endothelium, potentially offering greater specificity for imaging of inflammatory lesions in sarcoidosis. Our findings suggest this approach may enhance diagnostic accuracy and provide a more targeted method for monitoring disease activity.

Increased uptake of [⁶⁸Ga]Ga-DOTA-Siglec-9 in the lungs and mediastinal lymph nodes of patients with sarcoidosis aligns with prior observations from a porcine model of acute lung injury, which also showed tracer accumulation in inflamed pulmonary tissue [33]. Similarly, VAP-1-targeted imaging has demonstrated utility in preclinical models of autoimmune myocarditis [17], where immunohistochemical studies confirmed VAP-1 expression in inflamed cardiac tissue, highlighting potential future applications in cardiac sarcoidosis.

In our study, myocardial uptake of [⁶⁸Ga]Ga-DOTA-Siglec-9 was low in all participants, consistent with the absence of known cardiac involvement. In contrast, increased tracer uptake in the liver, spleen, bone marrow, and bone among patients with sarcoidosis suggests systemic inflammation, extending the relevance of VAP-1-targeted imaging beyond localized pulmonary findings.

[⁶⁸Ga]Ga-DOTA-Siglec-9 was developed as a specific ligand for VAP-1, a copper-containing semicarbazide-sensitive amine oxidase (AOC3) that mediates leukocyte adhesion transmigration at sites of inflammation [34, 37]. While VAP-1 is predominantly expressed on vascular endothelial cells [34], it is also present in smooth muscle cells [38], adipocytes [39], and constitutively in the endothelial lining of organs such as the liver, kidney, and peripheral lymph nodes [34]. In our study, patients with sarcoidosis exhibited elevated serum levels sVAP-1 compared to healthy controls. Increased sVAP-1 levels have been observed in several chronic inflammatory conditions, including psoriasis, atopic eczema, chronic kidney disease, inflammatory liver disease, and multiple sclerosis [13, 40–42], though not uniformly across all inflammatory disease. Notably, sVAP-1 levels appear to vary with age and sex, with slightly elevated levels observed in younger women but no significant sex-based differences in older adults [43]. These findings support the relevance of VAP-1 both as an imaging target and potential biomarker for inflammatory activity in sarcoidosis.

Interestingly, in the diseased control with lung adenocarcinoma, [⁶⁸Ga]Ga-DOTA-Siglec-9 uptake in the tumor and mediastinal lymph nodes was higher than the average uptake observed in patients with sarcoidosis. While VAP-1-targeted imaging shows promise for detecting inflammatory activity in sarcoidosis, it is important to note that VAP-1 is not exclusively expressed in this condition. Previous studies, including Viitanen et al. [44] have demonstrated upregulation of VAP-1 in certain malignancies such as melanoma, and our current study also observed elevated VAP-1 expression in lung adenocarcinoma. Therefore, the specificity of VAP-1-targeted imaging for sarcoidosis may be limited in contexts where malignant or other inflammatory conditions are present. Furthermore, compared to [¹⁸F]FDG PET, VAP-1-targeted signals were generally lower and less sensitive, highlighting that while this imaging approach provides complementary mechanistic information about inflammation, it may not detect all disease activity. Future studies are warranted to better define the sensitivity and specificity of VAP-1-targeted molecular imaging and to establish its potential role in differentiating sarcoidosis from malignant or other inflammatory conditions.

Study limitations

This study has several limitations. Most notably, the sample size was small, and the findings must be considered preliminary. Longitudinal studies with larger and more diverse populations are needed to validate the diagnostic and prognostic value of [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT, particularly in monitoring disease progression and treatment response. Second, comparison was limited to healthy controls and a single diseased control. Future studies should include a broader spectrum of control groups, including patients with other inflammatory and neoplastic lung diseases, to better assess the specificity and potential clinical utility of this imaging approach. Finally, the study population lacked ethnic diversity. Sarcoidosis has well-documented variations in presentation and severity across ethnic groups. For example, African American patients are more likely to experience severe pulmonary involvement, multiorgan disease, and worse outcomes—including higher hospitalization and mortality rates—than white patients [45]. These demographic factors should be considered in future research to improve generalizability and clinical relevance.

We chose SUV_mean to reflect the average tracer uptake across the ROIs, minimizing the influence of single-voxel noise or extreme values. We acknowledge that SUV_mean can be affected by ROI definition and delineation methods, which may introduce variability. Including SUV_max alongside SUV_mean provides a more comprehensive assessment and helps account for potential methodological variability.

Conclusion

This proof-of-concept study demonstrates that VAP-1-targeted [⁶⁸Ga]Ga-DOTA-Siglec-9 PET/CT imaging can detect inflammatory changes in pulmonary sarcoidosis and mediastinal lymphadenopathy with high contrast relative to healthy controls. The findings suggest a potential role for this tracer in noninvasive diagnosis and disease monitoring. In addition, our results contribute to the growing understanding of VAP-1’s role in sarcoidosis pathophysiology and support further development of VAP-1-targeted molecular imaging in inflammatory lung diseases.

Abbreviations

[¹⁸F]FDG

2-deoxy-2-[18F]fluoro-D-glucose

[⁶⁸Ga]Ga-DOTA-Siglec-9

Gallium-68-labeled peptide incorporating residues 283–297 of Siglec-9, conjugated to 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid

ACE

Angiotensin convertase

AOC3

Amine oxidase copper-containing 3

BMI

Body mass index

CT

Computed tomography

DAB

3,3’-diaminobenzidine

DOTA

1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid

ELISA

Enzyme-linked immunosorbent assay

HE

Hematoxylin-eosin

HRCT

High-resolution computed tomography

PET

Positron emission tomography

ROI

Region of interest

SD

Standard deviation

Siglec-9

Sialic acid-binding immunoglobulin-like lectin 9

SUV

Standardized uptake value

VAP-1

Vascular adhesion protein 1

sVAP-1

Soluble VAP-1

Authors’ contributions

Conception and design: H.Y-O., T.Tu., M.S., T.To., P.N., S.J., A.S., T.S., A.R.; acquisition, analysis and interpretation of data: P.D., H.Y-O., K.K., T.Tu., S.M., M.K., R.V., N.R., A.S., P.T., A.R. All authors were involved in drafting or critically reviewing the manuscript for important intellectual content, and all approved the final manuscript for submission.

Funding

The study was supported financially by grants from the Jane and Aatos Erkko Foundation, the Research Council of Finland (#350117), and the State Research Funding of Turku University Hospital (#13856).

Data availability

The original data of the work can be obtained from Prof. Anne Roivainen upon rational request.

Declarations

Ethics approval and consent to participate

The study was approved by the joint Ethics Committee of the University of Turku and Turku University Hospital (ETMK 35/1800/2018), and by the Finnish Medicines Agency (EudraCT 2018-001002-29). Each subject provided informed consent before entering the study. The study was registered as a clinical trial (NCT03755245). A lung cancer patient participated in another clinical trial NCT05212103 (ETMK 101/2021), but dropped out after the first scan.

Consent for publication

Not applicable.

Competing interests

Maria Silvoniemi received speaker fees and/or honoraria for advisory boards from MSD, BMS, Roche, Pfizer, AstraZeneca, Amgen, Novartis, and Boehringer-Ingelheim (not related to the current study). Antti Saraste has consultancy fees from AstraZeneca, Novo Nordisk, and Pfizer, and speaker fees from Abbott, AstraZeneca, BMS, Janssen, and Pfizer (not related to the current study). Sirpa Jalkanen owns patent regarding VAP-1 as a target for imaging. The remaining authors have no conflicts of interest to disclose.