Somatostatin Receptor PET/MR Imaging of Inflammation in Patients With Large Vessel Vasculitis and Atherosclerosis

Background Assessing inflammatory disease activity in large vessel vasculitis (LVV) can be challenging by conventional measures. Objectives We aimed to investigate somatostatin receptor 2 (SST2) as a novel inflammation-specific molecular imaging target in LVV. Methods In a prospective, observational cohort study, in vivo arterial SST2 expression was assessed by positron emission tomography/magnetic resonance imaging (PET/MRI) using 68Ga-DOTATATE and 18F-FET-βAG-TOCA. Ex vivo mapping of the imaging target was performed using immunofluorescence microscopy; imaging mass cytometry; and bulk, single-cell, and single-nucleus RNA sequencing. Results Sixty-one participants (LVV: n = 27; recent atherosclerotic myocardial infarction of ≤2 weeks: n = 25; control subjects with an oncologic indication for imaging: n = 9) were included. Index vessel SST2 maximum tissue-to-blood ratio was 61.8% (P < 0.0001) higher in active/grumbling LVV than inactive LVV and 34.6% (P = 0.0002) higher than myocardial infarction, with good diagnostic accuracy (area under the curve: ≥0.86; P < 0.001 for both). Arterial SST2 signal was not elevated in any of the control subjects. SST2 PET/MRI was generally consistent with 18F-fluorodeoxyglucose PET/computed tomography imaging in LVV patients with contemporaneous clinical scans but with very low background signal in the brain and heart, allowing for unimpeded assessment of nearby coronary, myocardial, and intracranial artery involvement. Clinically effective treatment for LVV was associated with a 0.49 ± 0.24 (standard error of the mean [SEM]) (P = 0.04; 22.3%) reduction in the SST2 maximum tissue-to-blood ratio after 9.3 ± 3.2 months. SST2 expression was localized to macrophages, pericytes, and perivascular adipocytes in vasculitis specimens, with specific receptor binding confirmed by autoradiography. SSTR2-expressing macrophages coexpressed proinflammatory markers. Conclusions SST2 PET/MRI holds major promise for diagnosis and therapeutic monitoring in LVV. (PET Imaging of Giant Cell and Takayasu Arteritis [PITA], NCT04071691; Residual Inflammation and Plaque Progression Long-Term Evaluation [RIPPLE], NCT04073810)

L arge vessel vasculitis (LVV) is a chronic relapsing and remitting systemic inflammatory disease that comprises giant cell arteritis (GCA) and Takayasu arteritis (TAK), which causes panarterial granulomatous infiltration of the aorta and its major branches. The initial clinical presentation of LVV is often confounded by nonspecific constitutional symptoms that can lead to diagnostic uncertainty. However, dangerous clinical sequelae such as acute visual loss and myocardial infarction (MI) can occur in GCA and TAK, respectively.
Although 18  shown high SST 2 binding affinity and favorable tracer kinetics. 3 In this proof-of-concept study, we tested the hypothesis that SST 2 could be a useful imaging target for the diagnosis and therapeutic monitoring of LVV using PET/magnetic resonance imaging (MRI) with 68 Ga-DOTATATE and 18 F-FET-bAG-TOCA.   Values are median (IQR) or n (%). a 1 patient with active Takayasu arteritis also had a non-ST-segment elevation myocardial infarction secondary to atherosclerotic plaque rupture 5 days before baseline positron emission tomography/magnetic resonance imaging. b The patient with unspecified LVV was a 55-year-old man with history of Sjögren disease and stroke in whom vasculitis was diagnosed based on constitutional symptoms, raised C-reactive protein level, and classical imaging findings. c Blood results shown are from the day of baseline positron emission tomography/magnetic resonance imaging.

Research
BMI ¼ body mass index; GCA ¼ giant cell arteritis; HDL ¼ high-density lipoprotein; LDL ¼ low-density lipoprotein; LVV ¼ large vessel vasculitis; n/a ¼ not applicable; NR ¼ normal range. maximum tissue-to-blood ratios (TBR max ) values. 4 The mds was defined as the highest arterial TBR max slice, averaged with contiguous slices above and below. The index vessel was the artery with the highest mean TBR max . Intraobserver and interobserver repeatability of these methods has previously been demonstrated using 68 Ga-DOTATATE. 2 RNA SEQUENCING. Bulk RNA-sequencing (RNAseq) data are from the UKGCA Consortium Study (NCT04102930), which is linked to the National Institute for Health Research Rare Diseases Bio-Resource. Single-cell and single-nucleus RNAseq were performed using Chromium (10x Genomics).

HISTOLOGY AND
AUTORADIOGRAPHY. Arterial specimens were analyzed using methods for immunofluorescence microscopy as previously described, 5 with primary antibodies for SST 2 and CD68. 68 Ga-DOTATATE autoradiography was conducted using established methods. 2  Receiver-operating characteristic (ROC) analysis was used to assess diagnostic accuracy and identify optimal TBR thresholds based on the Youden index.
Linear mixed-effects models were used to account for hierarchical data structure and multiple observations within patients, with patient and vessel included as random effects and tracer as a fixed effect. A randomeffects regression model was used to assess DTBR at follow-up compared with baseline. Results of the regression models are reported as effect size AE SEM or absolute change (95% CI) for log-transformed data, as well as the percent difference between groups. Potential confounding factors associated with mean TBR max (P # 0.10) in univariable linear mixed-effects models were included in multivariable sensitivity analyses. A 2-sided P value of <0.05 was considered significant.
Baseline clinical data are summarized in Table 1.  Table 1). Data for the 2 tracers were pooled because there was comparable image quality and ability to distinguish active/ Receiver-operating characteristic analyses of SST 2 PET mean TBR max and mdsTBR max for differentiating active/grumbling from inactive LVV. Image scale bars indicate standardized uptake value; error bars in D indicate median (IQR). 18 F-FDG ¼ 18 F-fluorodeoxyglucose; AUC ¼ area under the curve; CT ¼ computed tomography; F ¼ female; m ¼ mean; mds ¼ most diseased segment; MR ¼ magnetic resonance; sens ¼ sensitivity; spec ¼ specificity; SST 2 ¼ somatostatin receptor 2; TAK ¼ Takayasu arteritis; TBR max ¼ maximum tissue-to-blood ratio; other abbreviations as in Figure 1. Inflammation Imaging in Arterial Disease  Table 2). For the thoracic aorta, the difference was 26.6% (95% CI: 12.6%-42.3%; P < 0.0001), and for all vessels, it was 11.6% (95% CI:    Table 3. Although it was intended to repeat imaging for all patients after 6 months, some scans were delayed during the COVID-19 pandemic, and 12 patients declined to attend for the second scan because they remained in isolation when public restrictions were lifted. Ten of the 15 LVV patients who did undergo repeat imaging had newly initiated or escalated treatment after their baseline scan, which was Inflammation Imaging in Arterial Disease  Table 4. Image scale bars indicate SUV. BL ¼ baseline; CRP ¼ C-reactive protein; ESR ¼ erythrocyte sedimentation rate; FU ¼ follow-up; other abbreviations as in Figures 1 to 3.  PET/MRI REPEATABILITY. Scan-scan repeatability was also evaluated. Importantly, there was no difference in index vessel mean TBR max in the 5 patients with LVV whose treatment remained unchanged (scan-scan interval: 9.2 AE 3.8 months) ( Figures 5B and 5D). The single measure intraclass correlation coefficient, using a 2-way mixed-effects model with absolute agreement for index vessel    Figure 8A). Aortic mean TBR max was also correlated with maximum wall thickness (r ¼ 0.68; P ¼ 0.002) (Supplemental Figure 8B) in these patients.
Single-cell RNAseq data (n To further corroborate these findings, singlenucleus RNAseq was performed in temporal artery biopsy samples from patients with histologically proven GCA (n ¼ 2) ( Figure 6C) and a carotid endarterectomy specimen (n ¼ 1) ( Figure 6D). Patient details for these and other specimens are in Supplemental Table 6. Uniform Manifold Approximation and Projections and numbers of nuclei for cell clusters are shown in Supplemental Figure 9 and Supplemental Table 7. Macrophages again emerged as the dominant SSTR2-expressing cell type in the carotid atherosclerotic plaque, and pericytes were also identified in temporal biopsy samples. SSTR2 expression was not detected by single-nucleus RNAseq in a healthy aortic specimen (n ¼ 1; not shown).
Cell clusters in the single-cell and single-nucleus experiments were not directly comparable. However, SSTR2-expressing macrophages identified in temporal arteries expressed proinflammatory markers (S100A8 and S100A9) (Supplemental Figure 10A). SSTR2-expressing macrophages in the carotid artery expressed markers of resident and/or alternatively activated macrophages (MERTK, SOD2, LGALS3) but also CXCL3, which is an inflammatory cytokine (Supplemental Figures 10C and 10D). Pericytes were not distinguished as a distinct population in the single-cell RNAseq data set.
analyzed for single-nucleus RNAseq (n ¼ 2), as well as additional temporal artery samples (GCA: n ¼ 2; control artery with no abnormality: n ¼ 1) and an aortic LVV specimen (n ¼ 1). Histologic findings were consistent with RNAseq data. There was specific costaining of SST 2 and the pan-macrophage marker CD68 within inflamed regions of temporal arteries ( Figures 7A and 7B) and aortic tissue (Supplemental in the endothelium or aSMA in the media of the main vessels. There was also no overlap with CD3 þ or CD4 þ T lymphocytes. However, cellular localization of SST 2 did occur with pericytes identified by neuron-glial antigen 2 þ and aSMA around neovessels in the adventitia ( Figures 8A and 8B), as well as cells with the morphologic appearance of perivascular adipocytes. SST 2 staining was not detected in the control artery and was low in perivascular tissue ( Figure 8C).

CENTRAL ILLUSTRATION Continued
Patients with large vessel vasculitis (LVV) and recent atherosclerotic myocardial infarction (MI) underwent somatostatin receptor 2 (SST 2 ) positron emission tomography/magnetic resonance imaging (PET/MRI) in a prospective observational cohort study. In parallel, ex vivo mapping of the imaging target was performed using RNA sequencing, histology, and autoradiography. The research methods and main study findings are summarized. Arterial SST 2 signal (arrow) measured by the maximum tissue-to-background ratio (TBR max ) using PET/MRI accurately differentiated patients with active/grumbling LVV from those with inactive LVV and recent MI, as well as control subjects.
There was also a strong correlation between SST 2 mean TBR max and para-aortic thickening in LVV patients. SST 2 expression was identified in macrophages (dashed arrow), pericytes, and perivascular adipocytes with arterial specimens from patients with LVV. DOTATATE ¼ DOTA-  15 We also found that in patients with LVV, SST 2 PET signals could additionally originate from pericytes and adipocytes within inflamed periadventitial tissue. Somatostatin receptor expression has been identified in pericytes from patients with interstitial lung disease 16 and retinal disease, 17 and avid 68 Ga-DOTATATE uptake has also been reported in a patient with a rare metastatic hemangiopericytoma. 18 Although endothelial cell SSTR2 expression has also been reported, 19