Somatostatin receptor imaging by SPECT and PET in patients with chronic inflammatory disorders: a systematic review

The selection of appropriate publications was based on the PRISMA guidelines [15]. We searched in Cochrane Central Register of Controlled Trials (CENTRAL) published in the Cochrane Library, MedLine (1993–2017), EMBASE (1993–2017), PubMed (1993–2017), Google Scholar (1993–2017), OVID (1993–2017), EBSCO (1993–2017), and Scopus and Web of Science (1993–2017). All relevant studies published in English, Italian, and Spanish language were identified by using the following search strategy: (“receptors, somatostatin”[MeSH Terms] OR “receptors”[All Fields] AND “somatostatin”[All Fields]) OR (“somatostatin receptors”[All Fields] OR (“somatostatin”[All Fields] AND “receptor”[All Fields]) OR “somatostatin receptor”[All Fields]) AND (“Bildgebung”[Journal] OR “imaging”[All Fields]) AND (“adult”[MeSH Terms] OR “adult”[All Fields] OR “adults”[All Fields]) NOT (“neoplasms”[MeSH Terms] OR “neoplasms”[All Fields] OR “neoplasm”[All Fields]) NOT (“tumors”[All Fields] OR “neoplasms”[MeSH Terms] OR “neoplasms”[All Fields] OR “tumors”[All Fields]) somatostatin receptor imaging AND adults.

Studies or reports in which compounds were labeled with ⁶⁸Ga, ^99mTc, ⁶⁸Ga, and ⁶⁴Cu for applications other than in oncology-related indications were included. Every effort was made to include both the earliest and the most recent publications relating to a particular application, as well as any study with a significant new contribution. The decision to include or exclude an article was made by consensus.

Publications in which the focus was on SRS in oncology-related applications or in which the emphasis was on aspects related to generators, radiochemistry, animal models, experimental reports, or physics were excluded.

Literature search was broadened to all reference lists of all retrieved articles, including observational studies, systematic reviews, conference proceedings, and poster presentations. Studies or reports with more than three cases in which SRS was used in patients with inflammatory diseases were also included. Case reports allusive to just one patient where not taken in consideration because the results of the observations could not be compared between patients in order to highlight differences or similarities.

The decision to include or exclude an article was made by consensus between LKF, AWJMG, AS, and AM. We used the search strategies described to obtain titles and abstracts of studies of potential relevance to the review.

Two authors, one with expertise in the topic (LKAF), the second with expertise in the statistical methodology (SM), undertook the search. LKF, SM, and AM screened independently titles, abstracts, conferences proceedings, posters, and full texts for eligibility. We ensured the avoidance of multiple studies reporting on the same patient population by controlling the duplicate records in a peer review way, including the entire databases results; we also used standardized, pilot-tested forms, together with detailed instructions. Reviewers resolved disagreement through discussion or, if required, by adjudication by a third reviewer (SM).

The included studies were assessed by using the Cochrane Collaboration’s tool [16] and Quadas-2 [17] for potential source of bias and variation: patient selection bias (study design, patient recruitment, prospective data collection, consecutive patient enrollment), patient variation (demographic features, disease prevalence, disease severity, prior testing), index test bias (test review bias, threshold selection), index test variation (observer variation, availability of clinical information, test technology, test execution), reference standard bias (inappropriate reference standard, diagnostic review bias, incorporation bias), reference standard variation (definition of target condition), flow and timing bias (disease progression bias, treatment paradox, partial verification bias, differential verification bias, withdrawals, uninterpretable test results, sample size).

Because of differences between studies, it was not possible to pool data, and so a narrative synthesis was prepared. Study population was very heterogeneous, for this reason we classified patients according to the diagnosed disease: endothelial inflammation and atherosclerosis, rheumatoid arthritis, idiopathic pulmonary fibrosis, Graves’ ophthalmopathy, sarcoidosis, and other chronic inflammatory diseases.

We found 93 potential studies evaluating radiolabeled SST analogues in chronic inflammatory diseases. After checking for duplicates, 13 papers were immediately excluded; after reading the full-text records, 40 were excluded because they did not fulfill the inclusion criteria. We, therefore, analyzed 38 publications, as shown in Fig. 1. Results are summarized in Table 1.

In total, we analyzed 9 studies and excluded 2 for the analysis because they did not complete the inclusion criteria; 7 studies with the use of SRS for endothelial inflammation were included, with a total of 262 patients (Table 1). In details, Tarkin et al. [18], in a descriptive study, compared PET with ¹⁸F-fluorodeoxyglucose (FDG) and ⁶⁸Ga-DOTA-TATE to detect culprit coronary and carotid arteries with acute coronary syndrome, transient ischemic attacks and stroke, and demonstrated best results with ⁶⁸Ga-DOTA-TATE in the diagnostic accuracy to detect stable and unstable inflamed coronary lesions and vascular inflammation in neighboring coronary and aortic vasculature. When age, total cholesterol, and BMI were evaluated with other relevant clinical factors using multi- variate linear regression, they remained significant predictors of ⁶⁸Ga-DOTATATE mTBRmax. In patients with TIA or stroke, increased ⁶⁸Ga-DOTATATE inflammatory signals were reliably differentiated between culprit carotid plaques and contralateral nonculprit carotid arteries. Ex vivo ⁶⁸Ga-DOTATATE carotid autoradiography showed high levels of specific ligand binding to SSTR2 receptors. SSTR2 and CD68 mRNA levels were highly correlated within carotid plaque. Carotid SSTR2 and CD68 mRNA levels also showed strong correlation with in vivo ⁶⁸Ga-DOTATATE TBRmax values measured at the corresponding level in clinical PET images. These data provided both histological and molecular validation of ⁶⁸Ga-DOTATATE as specific marker of atherosclerotic inflammation. Under QUADAS analysis, we did not find any significant source of risk bias in this report. Malmberg et al. [19] in an observational study in 60 consecutive patients described the differences between ⁶⁸Ga-DOTA-TOC and ⁶⁴Cu-DOTA-TATE in detecting atherosclerotic plaques in different vessels, showing better performance of ⁶⁴Cu-DOTA-TATE explained by the longer half-life which allows for early as well as late PET scanning, even the day after injection. ⁶⁴Cu has also a substantially shorter positron range than ⁶⁸Ga, 1 versus 4 mm, rendering it a much better spatial resolution, but a lower positron abundance [20] Uptake of ⁶⁴Cu-DOTATATE was significantly higher than ⁶⁸Ga-DOTATOC in the vascular regions both when calculated as maximum and mean uptake. In a multivariate analysis, there was a significant association between Framingham risk score and the overall maximum uptake of ⁶⁴Cu-DOTATATE using SUV as well as target-to-background ratio, whereas no association was found with ⁶⁸Ga-DOTATOC. The association of risk factors and maximum SUV of ⁶⁴Cu-DOTATATE was found driven by body mass index, smoking, diabetes, and coronary calcium score suggesting the potential use of this radiotracer as noninvasive biomarker of cardiovascular risk. According to QUADAS 2 analysis, we considered the reference test as the main source of risk of bias because it did not use the anatomopathological evidence to validate the results on 68Ga-DOTATOC but the calcium score on CT. Simon Wan et al. [21] conducted a descriptive study with 20 patients with recent stenotic carotid events and described the findings of ⁶⁸Ga-DOTA-TATE in carotid plaques versus histology. Surprisingly, they did not find any significant uptake of ⁶⁸GaDOTATOC in the plaques neither activated macrophages expressing SSTR2; and they could not support the use of ⁶⁸GaDOTATOC for evaluating the inflamed plaque; the source of bias in this report, (misclassification explained by heterogeneity of time between carotid event, research PET, and endarterectomy) could maybe explain the results. In QUADAS 2 analysis, this source of bias corresponds to flow and timing. Pedersen et al. [22], in a population of 10 patients with clinical symptoms of stroke or transient ischemic attack, described the findings of ⁶⁴Cu-DOTA-TATE uptake in atherosclerotic plaques. ⁶⁴Cu-DOTA-TATE uptake was significantly higher in symptomatic plaques and correlated with CD163 expression by activated macrophages at histology. We did not find any important source of risk of bias in this report according to QUADAS analysis. Xian Li et al. [23] in a retrospective series of 16 patients, analyzed simultaneously ¹⁸F-FDG and ⁶⁸Ga-DOTA-TATE PET/CT and showed significantly increased uptake in the fibrotic and vulnerable atherosclerotic plaques compared with normal coronary arteries, suggesting a potential role of this tracer for molecular assessment of coronary artery disease. They found higher values of focal uptake with ⁶⁸GaDOTATATE in patients with high cardiovascular risk and significant correlation between the mean uptake of both tracers and the patient’s score of risk factor. We found that the reference test and the index test were source of risk of bias in QUADAS 2 analysis. Mojtahedi et al. [24] generated a preliminary data by using ⁶⁸Ga-DOTATATE PET/CT, with in vulnerable or fibrotic atherosclerotic plaques in the coronary arteries. In a population of 44 patients with neuroendocrine tumors (NET) who underwent ⁶⁸Ga-DOTATATE PET/CT, they found that the mean TBR value in the normal group was 1.345 ± 0.58 while the mean TBR value in the fibrotic plaque group was 1.752 ± 1.50 (p 0.0043) and in atherosclerotic plaques group was 2.043 ± 1.76, p < 0.0001). There was a significant correlation (p = 0.0026) between ⁶⁸Ga-DOTATATE uptake and the progression to formation of atherosclerotic plaques, based on coronary CT calcium score, HU. Those findings suggested the potential of ⁶⁸Ga-DOTATATE PET/C for molecular assessment of coronary artery disease. According to QUADAS 2 analysis, we considered as potential source of risk bias the reference standard they used for validate the ⁶⁸Ga-DOTATATE PET results. Rominger et al. [25] in a descriptive analysis of 70 consecutive patients correlated ⁶⁸Ga-DOTATATE uptake in the left anterior descending coronary artery with the presence of calcified plaques and cardiovascular risk factors. Higher uptake lead distinguishes between patients with and without coronary calcifications. As in the abovementioned reports, the main source of bias of this series in QUADAS 2 analysis was related to the reference standard they used to validate the ⁶⁸Ga-DOTATATE results.

It is well known that monocytes and macrophages, being the first inflammatory cells associated with vascular inflammation, play an important role in the pathogenesis of atherosclerosis, contributing to necrotic core formation, fibrous cap thinning, and plaque vulnerability [26]. These cells express SSTR-2 on the cell surface in cultures [27] and therefore ⁶⁸Ga-DOTA-TATE uptake was thought to have the potential to be a surrogate marker of inflammation to study plaque biology [28]. Taking together all published data, it is likely that there are complex relationships between the stage of plaque evolution (chronic vs acute), macrophage density and activity, SSTR subtype expression, and ¹⁸F-FDG and ⁶⁸Ga-DOTA-TATE uptake. Evidences, so far, show a tendency to confirm the utility of SRS with ⁶⁸Ga-DOTA-TOC to detect high-risk, vulnerable, atherosclerotic plaques and an interesting correlation with risk factors. Added value of TOC (specific for inflammatory cells only provides an in vivo information on the vulnerability of plaques) vs FDG (global inflammatory burden but non-specific and does not provide a relevant histopathological information) and ¹⁸F-Na and calcium score [29].

In total, 11 publications were analyzed, but just 2 clinical studies were included within 36 patients (Table 1), since most were reviews, case reports or in vitro or animal studies. Anzola et al. [30] conducted a pilot study to evaluate the uptake pattern of ^99mTc-EDDA/HYNIC-TOC in the joints of 18 patients with rheumatoid arthritis refractory to treatment. They showed high uptake of the molecule in all evaluated joints and also in the salivary glands of some patients as possible evidence of Sjogren syndrome. There was a statistically significant reduction in uptake after treatment with anti-TNFα antibodies in all patients. The authors concluded that this scan can be useful to select and monitor RA patients for therapy with anti-TNFα antibodies. In QUADAS 2 analysis, we found as potential source of risk of bias the index test because the lack of standardized method to evaluate the results. Vanhagen et al. [31] reported the results of 14 patients with RA and 4 with osteoarthritis using ¹¹¹In-octreotide. They reported sensitivity (uptake of the tracer in swollen and painful joints) of 76% in RA patients and of 20% in osteoarthritis. As in the abovementioned series, we found lack of a standardized method to evaluate the results and weakness in the reference test used to compare the results (QUADAS 2, reference and index test source of bias).

Synovia of patients with active rheumatoid arthritis expresses a high density SSTR [32]. The experience published in these two clinical papers, shows the potential of SRS to localize and identify sites of active inflammation in joints, both symptomatic and asymptomatic, and extra-articular involvement as in salivary glands. The information provided by SRS is different from other imaging modalities and provides in vivo histopathological information on the activity of cell-mediated inflammation. This information has an important impact for therapy decision-making and therapy follow-up as also appears in other studies [33].

In total, 7 studies were analyzed, but only 3 studies could be included within 51 patients (Table 1). Ambrossini et al. [34] in an observational study evaluated the potential role of ⁶⁸Ga-DOTANOC PET/CT in patients with idiopathic pulmonary fibrosis (IPF) and non-specific interstitial pulmonary pneumonia (NSIP). ⁶⁸Ga-DOTANOC PET/CT findings were compared with HRCT results. In IPF patients, ⁶⁸Ga-DOTA-NOC uptake appeared as a typical subpleural and peripheral distribution involving both lung fields predominantly at the bases. Areas of ⁶⁸Ga-DOTA-NOC uptake directly corresponded to pathologic areas on HRCT. In contrast, ⁶⁸Ga-DOTANOC uptake was faint in NSIP patients and undetectable in healthy control subjects. No significant association was documented in NSIP cases. According to QUADAS 2 analysis, we found as source of potential risk of bias the index test because they did not use a previous standardized method to evaluate the results and they did not confirm the findings under histopathogical analysis.

Lebtahi et al. [35] investigated the expression in vivo of SSTR in the lungs of 11 patients with IPF, and 6 patients with pulmonary fibrosis associated with systemic sclerosis by using ¹¹¹In-octreotide scintigraphy and compared the results with 19 control patients. They reported higher uptake in all 11 IPF patients and in 4 of 6 systemic sclerosis patients related to normal control population. Increased uptake correlated significantly with different lung function and cytological tests. As in the abovementioned series, according to QUADAS 2 analysis, we found as potential source of risk of bias the index test and the reference test explained by the same conditions argued before.

Win Thida et al. [36] analyzed in an observational study 26 patients with diffuse parenchymal lung disease (10 patients with idiopathic pulmonary fibrosis, 12 with nonspecific interstitial pneumonia and 4 with other forms of interstitial lung disease), by using ⁶⁸Ga-DOTA-TATE and ¹⁸F-FDG by combined PET and HRCT. All patients demonstrated increased pulmonary signal with both ⁶⁸Ga-DOTA-TATE and ¹⁸⁻F-FDG.

The experience reported in this review shows the potential utility of the SSTR scintigraphy to localize and to detect the active inflammation mediated by an immunological response in idiopathic pulmonary fibrosis. In QUADAS 2 analysis, we found as source of risk of bias the reference test and index test explained by the same reasons argued above.

In total, 33 papers were reviewed, 14 studies with in total 451 patients were included (Table 1). Aguirre-Balsalobre et al. [37] in a series of 18 patients affected by thyroid orbitopathy with Graves’ diseases by using a semiquantitative analysis of the images with ¹¹¹In-pentreotide, demonstrated how patients with higher scores had better response to corticosteroid therapy; according to QUADAS 2, analyzing the source of bias of this experience was related to the index test because lack of standardized method. Colao et al. [38], Gerdin et al. [39], Kahaly et al. [40], Kahaly et al. [41], Krassas et al. [42], Krassas et al. [43], Moncayo et al. [44], Nocaudie et al. [45], Postema et al. [46], in similar study designs by using ¹¹¹In octreotide, described how patients with active orbitopathy in Graves’ disease had higher uptake of the tracer in the orbits, and how this finding was helpful to discriminate active disease Vs non-active disease and how it helped to choose better the candidates to immunotherapy, radiotherapy or somatostatin analogues therapy. In those series, it was possible to observe how patients with higher uptake in orbits responded well to the installed treatment. The common source of bias for the abovementioned authors according to QUADAS 2 analysis was related to the index test, and although none of the authors used a standardized method to analyze the images, the results were not only similar but also promissory; moreover, in Postema’s report, it was seen with a high frequency of source of risk bias related to patient selection. In Postema’s series, it also found the patient selection as source of risk of bias (Fig. 2).

Burggasser et al. [47] in a series of 44 consecutive patients, by using ^99mTcP-829, demonstrated statistically the differences in the orbital uptake between control and disease patients and between active and inactive orbitopathy, in Graves’ disease patients with orbitopathy. Even though they used two different methods to analyze the images, we considered the index test as source of bias according to QUADAS 2 analysis. Huang Sung et al. [48], by using ^99mTc HYNIC-TOC, reported similar results of the above authors with the same previous described source of bias.

Rong Zhao et al. in a prospective study [49] by using SPECT/CT system as novel technique, investigated the predictive role of the orbital somatostatin receptor scintigraphy with ^99mTc-EDDA/HYNIC-TOC (^99mTc-TOC) to detect clinical stage of Graves’ ophthalmopathy (GO) and the response to corticosteroid therapy in a sample of 46 patients with GO and four volunteers without eye disease. The treatment effect was evaluated both by the orbital ^99mTc-TOC uptake and NOSPECS. Orbital images were quantified by region of interest (ROI) analysis. Semi-quantitative evaluation of retrobulbar uptake was performed with irregular ROIs which were placed over the orbits (O) and the reference area over the occipital area (OC). This prospective study demonstrated marked orbital accumulation of ^99m Tc-TOC in 22 GO patients, showing higher activity score and clinical improvement; they demonstrated good correlation of orbital ^99m Tc-TOC scintigraphy with CAS. The value of this study was to identify the anatomical structures depicted with ^99m Tc-TOC in patients with GO by means of SPECT and CT image fusion analysis. Thus, in the absence of histology, because of a favorable target-to-background ratio and fusion imaging of SPECT/CT in this study, orbital 99mTc-TOC fusion imaging was able to determine the pathological phase of Graves’ disease, giving a high positive scan in the active early phase and a low positive or negative scan in the stable end phase of the disease. We consider that the analysis of the images under hybrid images diminishes the risk of bias related in index test (QUADAS 2).

Lincke et al. [50] is the only report so far that used PET study in thyroid-related pathologies. He compared the characteristics of uptake between ¹¹¹In-pentreotide and ⁶⁸Ga-DOTA-TOC in normal and pathologically altered thyroid glands and found that both tracers showed higher uptake in active Hashimoto and Graves’ disease most likely caused by SSTR expressing lymphocytes in the thyroid tissue; however, the physiologic or pathophysiologic relevance of the increased In-111 pentetreotide and Ga-68 DOTATOC uptake in normal thyroid glands, hot and cold nodules, and goiters and nodular thyroids remain to be determined. According to QUADAS 2 analysis in this report, the index test and the reference test were considered as unclear sources of bias.

During the last 20 years, it demonstrated the utility of SSTRs scintigraphy identifying active orbitopathy in Graves’ exophthalmopathy. No matter as a common source of bias in those series it was found to be lacking of validated method to evaluate the SSTR scintigraphies, all the authors reported similar results showing scintigraphic patterns consistently with high uptake of the molecule in the periorbitary zone as evidence of active disease with prognostic implications.

As far as sarcoidosis is concerned, were reviewed 18 publications, of which 6 were excluded (Table 1). Lapa et al. [51] in an observational study conducted a pilot prospective study in 6 patients with active peri−/myocarditis and 6 with sub-acute myocardial infarction, who underwent SSTR-PET/CT, ⁶⁸Ga DOTA-TOC and cardiac MRI within 3–10 days after onset of symptoms. They compared the results under semiquantitative analysis with 12 oncologic patients with no history of coronary artery disease. In patients with clinical evidence of myocarditis, PET/CT returned 26 positive segments and MRI 13 segments, respectively. In patients with myocardial infarction, there were 29 ⁶⁸Ga-DOTA-TOC positive segments and 31 abnormal segments in MRI. On a head-to-head comparison, SSTR-PET and MRI were concordantly positive in 36 segments (76.6%). Nineteen segments were SSTR-PET positive and MRI-negative (9.3%; 19/204), and 11 SSTR-PET negative and MRI-positive (5.4%; 11/204). Both modalities returned negative results in 138 segments, thus leading to an overall concordance of 85.3%. Agreement of the modalities was higher in patients with myocardial infarctions. SUV mean and SUV max were significantly higher in the infarcted/ inflamed myocardium as compared with remote myocardium or the left ventricular (LV) cavity. We did not find any relevant risk of bias according to QUADAS 2 analysis (Fig. 3). The overall concordance of the 2 modalities was 96.1%. Kwekkeboom et al. [52] in a cross sectional study in 46 patients with known mediastinal, hilar, and interstitial sarcoidosis, by using SRS with ¹¹¹In-pentetreotide showed positive results in 36 out of 37 patients with known pathology, and in 7 with normal X-rays. In 13 patients who had no evidence of extrapulmonary disease involvement with physical examination and conventional imaging, somatostatin receptor imaging revealed abnormal uptake of radioactivity outside the chest. Neither the degree of radioactive accumulation nor a specific pattern of pathological uptake was correlated with disease severity or clinical course. They repeated the scintigraphy on 13 patients, 5 of 6 patients who showed X-rays improvement, the SSTRs also improved. Two of 5 patients whose X-rays did not improve, lung function and SSTRs improved. They could not conclude that pentetreotide scintigraphy has a role in predicting prognosis or clinical course in patients with sarcoidosis. According to QUADAS 2 analysis, the source of risk of bias in this series was related to the lack of a standardized method to evaluate the results in the index test, and also we found a low concern related to the reference test used in some situations. Lebtahi et al. [53] in an observational study demonstrated the superiority of ¹¹¹In pentetreotide compared with ⁶⁷Ga scintigraphy in the evaluation of pulmonary and extrapulmonary involvement in 18 patients with proven sarcoidosis especially in the hilar or mediastinal area. Nine were or recently had been receiving steroid therapy at the time of the examination. Gallium scintigraphy found abnormalities in 89% of patients and detected 65% of the clinically involved sites. SRS found abnormalities in 100% of patients and detected 83% of clinically involved sites. Of the 9 treated patients, gallium scintigraphy found abnormalities in 78%, detecting 59% of the clinically involved sites whereas SRS found abnormalities in 100%, detecting 82% of the clinically involved sites. SRS images were consistently better than gallium images, with well-delineated lesions, especially in the hilar or mediastinal area, on either planar or SPECT images. They concluded that SSTRs seems to be a promising alternative to gallium scintigraphy for evaluating the extent of sarcoidosis, detects significantly more sites of sarcoidosis involvement with lesion contrast significantly higher than with gallium scintigraphy, especially for lung and mediastinal involvement. Despite the good results reported, we found as potential source of bias in QUADAS 2 analysis the lack of standardized method to evaluate the images on ¹¹¹In pentetreotide and also the lack of a well-defined reference test. Kamphuis et al. [54] in a retrospective study evaluated the additive value of SRS scintigraphy with ¹¹¹In pentetreotide in the clinical evaluation of sarcoidosis and compared the results with chest X-ray and CT. In both histologically proven and unproven sarcoidosis, in all but one SSTR uptake was demonstrated. In the thoracic region SRS increased the yield with 36% and 32% in comparison with X-ray and CT, respectively. In the histologically proven group, there were no negative SRS results, and the SRS increased the yield for thoracic localization in 30% and 14% of the patients in comparison with X-ray and CT, respectively. As in the abovementioned series, although the results reported were encouraging, we found in QUADAS 2 analysis that the lack of a standardized method for evaluating the SRS scintigraphy with ¹¹¹In pentetreotide was a source of potential bias as well as the reference test was source of potential risk of bias. Gormsen et al. [55]in a pilot study in 19 patients with suspected cardiac sarcoidosis compared the diagnostic accuracy and inter-observer variability of ⁶⁸Ga-DOTANOC vs ¹⁸F-FDG PET/CT. Cardiac sarcoidosis (CS) was diagnosed in 3/19 patients. By consensus, 11/19 ¹⁸F-FDG scans were rated as inconclusive, resulting in low agreement among reviewers (Fleiss combined kappa 0.27) and correspondingly poor diagnostic accuracy; for ⁶⁸Ga-DOTANOC, no scan was rated as inconclusive but it was reported as a poor interobserver agreement (Fleiss combined kappa 0.46). The sensitivity of ¹⁸F-FDG PET for diagnosing CS was 33%, specificity was 88%, PPV was 33%, NPV was 88%, with 79% of diagnostic accuracy; in contrast to ⁶⁸Ga-DOTANOC, and accuracy was 100%. ⁶⁸Ga-DOTANOC PET/CT looks very promising as an alternative CS PET tracer. When we analyzed under QUADAS 2 protocol, we found in the index test and reference test the potential source of risk of bias.

Nobashi et al. [56] in a descriptive study compared the utility of ⁶⁸Ga-DOTA-TOC with conventional ⁶⁷Ga-citrate in the visualization of active sarcoidosis and correlated quantitative parameters on ⁶⁸Ga-DOTA-TOC PET/CT with clinical data. Twenty patients with sarcoidosis underwent both studies. The diagnosis was confirmed in 12 patients histologically and in the other 8 on clinical and paraclinical data. Active lesions were identified by visual interpretation when focal increases in radioactivity were higher than the normal biodistribution of tracers. Quantitative analysis on ⁶⁸Ga-DOTATOC was made by using SUVmax values. DOTATOC-PET/CT showed abnormal findings in 19 patients, whereas ⁶⁷Ga showed abnormal findings in 17. DOTATOC was superior than ⁶⁷Ga identifying lymph nodes (p < 0.046). SUVmax in each organ did not show statistical difference between involved organs; however, the average SUVmax of lymph nodes was highest among the involved organs. It was not seen as statistically different in SUVmax between patients with and without symptoms. Active lesion volume calculated by DOTATOC-PET moderately correlated with serum ACE concentrations. ⁶⁸ GaDOTATOC-PET was superior to ⁶⁷Ga in visualizing lesions in the uvea and muscle as well in the lymph nodes. As a source of bias, there was partial verification bias because the lack of in vitro analysis of somatostatin receptor in the lesions and because not all patients were diagnosed by the same reference standard. There is patient variation bias in some patients who were receiving treatment for the clinical condition in the same time when there were acquired studies. According to QUADAS 2 analysis, the potential source of bias in this series relied on the index test and reference standard explained by the same arguments in the above references.

Piotrowoski et al. [57], in an observational study, evaluated 32 patients with sarcoidosis and used ^99mTc-HYNIC-TOC scintigraphy as a reference method to evaluate the clinical usefulness of traditional markers, such as angiotensin-converting enzyme (SACE), C-reactive protein (CRP), serum calcium (S-Ca²⁺ level, and 24-h Urinary-Ca²⁺ los, bronchoalveolar lavage fluid (BALF lymphocytes) as well as of a novel marker of lipid peroxidation-8-IP in exhaled breath condensate (EBC 8-IP). They divided the patients according to the ^99mTc-HYNIC-TOC results in two groups: grade 1 (20 patients) when the scintigraphy was positive for an abnormal uptake in thorax and grade 0 (12 patients) when the scan was negative. The only significant difference between scintigraphic negative vs positive results was found in the level of EBC 8-IP; unfortunately, this marker is not specific for sarcoidosis. Furthermore, it observed a trend towards higher levels of SACE in the group with positive radiotracer uptake. Although the tendency was noticed towards higher percentage in patients with positive scintigraphy results, the statistical significance was not reached. It was not possible to demonstrate the capacity of any laboratory test to estimate the intensity of the inflammatory process. The SSRTs could discriminate between positive and negative studies but further studies are needed to find the utility of those results. The reference standard was the most frequent source of bias and applicability concern in QUADAS analysis.

SRS has been proposed also for studying other chronic inflammatory conditions, in particular in histiocytosis, tuberculosis, cardiac allograft rejection, and small vessel vasculitis. Weinmann et al. [58] evaluated 13 patients with histiocytosis. Lung uptake of ¹¹¹In-pentreotride visually and semiquantitative analyzed was statistically significantly higher in patients compared with controls. This report showed concern as potential source of bias in the reference and index test. Oztürk et al. [59] described three similar cases: 2 patients with pulmonary sarcoidosis and one with tuberculosis by ¹¹¹In-pentreotide scintigraphy; this report generated in QUADAS 2 analysis concern about the source of risk of bias in the three items contained in the evaluation: patient selection, index test, and reference standard. Vanhagen et al. [60] reported in a consecutive series of 20 patients the utility of ¹¹¹In-octreotide localizing sites of granuloma infiltration in patients with sarcoidosis and tuberculosis. In vitro autoradiography of fresh tissue biopsies showed bonding of the molecule at sites that were microscopically identified as granulomatous disease. The major contribution of this study was that this scintigraphy procedure demonstrated the expression of somatostatin receptor in such kind of granulomas. Under QUADAS 2 analysis, we found index test as a potential risk of bias because of the lack of a standardized method to evaluate the results.

Aparici et al. [61] conducted a prospective study in 13 patients with suspected cardiac allograft rejection to assess the feasibility of SRS with ¹¹¹In-pentreotide to target activated lymphocytes in transplanted heart as a possible early marker of rejection. A high cardiac uptake was observed in 8 patients: 3 had an acute rejection and 5 had mild or no rejection. However, in 4 of the 5 patients with no rejection at time of study, a biopsy, performed 1 week later, demonstrated a significant rejection requiring treatment. These preliminary results indicate the feasibility of targeting activated lymphocytes for the detection of cardiac allograft rejection and suggest a possible predictive role of SRS. These preliminary results indicate the feasibility of targeting activated lymphocytes with somatostatin receptor imaging in the detection of cardiac allograft rejection.

Index test generated in QUADAS analysis the high concern as source of bias and eligibility concern. The method used to evaluate the pattern uptake was not validated previously.

Neumann et al. [62], by using 111 In-pentreotide, analyzed 32 consecutive patients with ANCA-associated small vessel disease (AASV). Disease activity was evaluated with the Birmingham Vasculitis Activity Score (BVAS). For pulmonary AASV, SRS showed a sensitivity of 86% and a specificity of 96% with a positive predictive value of 97% for active disease. False negative scans were seen in patients under immunosuppressive therapy. In patients with ear/nose/throat involvement, SRS showed a sensitivity of 68% and a specificity of 100% with a positive predictive value of 100%. In patients who responded to therapy and went into full remission, repeat SSTR scintigraphy demonstrated the absence of previously present tracer accumulation; patients with aggressive disease who responded poorly to immunosuppressive therapy remained positive at repeat scintigraphy. Immunohistochemical analysis for SSTR2 and SSTR3 was performed on three open lung biopsies in active disease obtained from two patients with Wegener’s granulomatosis (WG) and one patient with microscopic polyangiitis as well as on nasal and lung tissue of one autopsy case with WG. All specimens demonstrated STR2 y and STR3 expression on monocytes-macrophages and giant cells surrounding granulomas and occasionally in the center of the granulomatous reactions. The reference standard generated in QUADAS analysis high concern as source of bias and applicability concern, because the immunohistochemical analysis was obtained just in three patients.

Reported evidence so far shows the utility of ⁶⁸Ga-DOTA-TOC for detecting sarcoidosis lesions, especially for lymph nodes, uvea, and muscles, as well as for vasculitis and other chronic inflammatory diseases. In particular, the results obtained for monitoring cardiac allografts open a new important clinical application of SRS to early detect lymphocytic infiltration in transplanted tissues and to monitor the effect of therapies in preventing rejection.