Keywords

34.1 Introduction

Although they have historically been considered as rare tumors, recent data suggest that neuroendocrine tumors (NETs) are more common than might be expected [1, 2]. Reported annual age-adjusted incidence is 5.25/100,000 [3]. NETs may arise anywhere in the human body, but the most common location of the primary lesion is the gastroenteropancreatic (GEP) tract, followed by the lungs. GEP NETs originate from the diffuse endocrine system of the gastrointestinal tract and pancreas [4]. A wide spectrum of biologically active peptides can be produced by NET cells (e.g., serotonin, gastrin, glucagon, and insulin) which are stored in vesicles, whose proteins (chromogranin A and synaptophysin) are common markers of GEP NETs [5].

The majority of GEP NETs (primaries and metastases) express somatostatin receptors, five distinct subtypes (sstr 1 to 5) of which have been identified, all of them binding native somatostatin. Somatostatin (sst) is a small, cyclic neuropeptide formed of 14 or 28 amino acids, both originating from the same preprotein, and is present in neurons and endocrine cells. It inhibits the secretion of a wide range of hormones. Its antiproliferative action controls cell growth with the potential for therapeutic application. Somatostatin actions are mediated by transmembrane domain G-protein-coupled receptors, and multiple subtypes of these receptors frequently coexist in the same cell [6, 7].

Naturally occurring sst has a very low metabolic stability in vivo with a half-life of less than 2 min. Therefore, more stable synthetic sst analogues have been developed [7, 8] which for in vivo diagnostic purposes have been labeled with gamma emitters: first with I-123 [9] and subsequently with In-111 [10], Tc-99 m [11,12,13], Ga-67 [14], and with positron emitters: C-11 [15], F-18 [16, 17], Ga-68 [14], Cu-64 [18], Sc-44 [19], and Tb-152 [20].

68Ga-DOTATOC and 68Ga-DOTATATE (Fig. 34.1) are the most established somatostatin receptor PET tracers and both have been recently approved in USA and Europe. Both radiopharmaceuticals enabled higher lesion detection rate than conventional 111In-DTPA-octreotide SPECT scintigraphy [21, 22], and changed the clinical management in most patients with negative or inconclusive findings on 111In-DTPA-octreotide scintigraphy [22]. Furthermore, dosimetric data showed the effective dose of 111In-DTPA-octreotide to be approximately three to five times higher than for the 68Ga-labeled somatostatin analogs [23].

Fig. 34.1
2 chemical structural formulas of 68 G a-DOTATOC on the left and 68 G a-DOTATATE on the right to highlight T O C and T A T E, respectively.

Structural formulas of 68Ga-DOTATOC and 68Ga-DOTATATE, where TOC and TATE stand, respectively, for D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Thr(OH) and D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Thr. Difference in structure is highlighted with blue square [29]

In GEP NET sstr 2 is overexpressed most abundantly, followed by sstr 1 and 5, rarely sstr 3 and 4 [24]. Higher presence of sstr 3 is revealed in pancreatic NET (incidence up to 71%) [25], compared with non-pancreatic NET. In vitro, it has been shown that 68Ga-DOTATATE binds to sstr 2, with an approximately tenfold higher affinity than 68Ga-DOTATOC. On the other hand, 68Ga-DOTATOC binds also to sstr 5 with significantly lesser affinity than to sstr 2, but with fivefold higher affinity compared to 68Ga-DOTATATE [26]. However, for various radionuclides, such an affinity difference between DOTATOC and DOTATATE could not be confirmed in humans. To clarify this discrepancy, Poeppel et al. [27] have conducted a study on 40 NET patients undergoing both 68Ga-DOTATATE and 68Ga-DOTATOC PET/CT. They also performed a separate subgroup analysis of 27 GEP-NET patients [28]. Both these studies showed a small but significantly higher number of lesions detected by 68Ga-DOTATOC and unexpectedly higher SUVmax in both primary tumors and metastases, but not in kidneys. Hence, a significantly higher kidney-to-tumor ratio for 68Ga-DOTATOC was reported. The authors conclude that the approximately tenfold higher in vitro affinity of 68Ga-DOTATATE for sstr 2 is not clinically relevant. Yet, in another study on 10 NET patients undergoing both 68Ga-DOTATATE and 68Ga-DOTATOC PET/CT, Velikyan et al. [29] state a preference for 68Ga-DOTATATE because of healthy organ distribution and excretion, although their conclusion is not fully supported by their data and their tumor-to-healthy organ ratios for liver, kidney, and spleen were all in favor of 68Ga-DOTATOC. Therefore, the aim of this study was to resolve this discrepancy by comparing in another group of GEP NET patients the in vivo distribution of the two radiopharmaceuticals in the same patients by determining their SUVmax values in primary tumors, metastases, and in normal liver.

34.2 Methods

Thirty eight histologically confirmed well-differentiated GEP NET patients with clinically, biochemically, and morphologically stable disease (selected from 800 NET patients), 19 female and 19 male (mean age 61.8 ± 12.1 years; age range 24–79 years) were submitted to first a 68Ga-DOTATATE and subsequently a 68Ga-DOTATOC PET/CT study on two consecutive visits as part of a usual workup, with 197 days (117–311 days range) in between. Well-differentiated GEP NET primary tumors included: 1 duodenal, 18 pancreatic, 2 cecal, 12 ileal, 3 jejunal, 1 mesenteric, and 1 GEP NET in appendix.

Patients had either not been on octreotide therapy or had octreotide therapy suspended for 6 weeks prior to 68Ga-DOTATOC/68Ga-DOTATATE PET/CT. Labeling and quality control of the radiopharmaceuticals were performed according to methods described previously [30]. 68Ga-DOTATOC and 68Ga-DOTATATE were prepared in the radiopharmacy of the Zentralklinik Bad Berka under GMP conditions and used in patients in agreement with specific German regulations for the use of in-house prepared radiopharmaceuticals and in accordance with regulations of the Federal Office for Radiation Protection (Bundesamt für Strahlenschutz). Patients were given 1.5 L of water-equivalent oral contrast dispersion Gastrografin 1 h before the start of acquisition. To increase renal washout and decrease radiation exposure to the urinary bladder, 20 mg of furosemide was given i.v. after injection of 68Ga-DOTATOC/68Ga-DOTATATE.

All patients were examined on a dual-modality PET/CT scanner (Biograph duo; Siemens Medical Solutions). On average, acquisition started 87 min after injection of 123 ± 9 MBq of 68Ga-DOTATATE, and 90 min after injection of 119 ± 8 MBq of 68Ga-DOTATOC, each with a peptide mass dose of 12 μg.

First, a topogram was acquired. The patients were given 100 mL of intravenous contrast (by an automated injection pump), followed by computed tomography (CT) scanning in the craniocaudal direction with 30 s delay after injection, and PET scanning in the caudocranial direction. After scatter and attenuation correction, PET emission data were reconstructed using an attenuation-weighted ordered-subsets maximization expectation approach with two iterations and eight subsets on 128 × 128 matrices and with a 5 mm Gaussian post-reconstruction filtering.

The PET/CT images were assessed using E.soft (syngo-based nuclear medicine software). In each of 76 patient PET/CT studies (38 PET/CT studies for each radiopharmaceutical), regions of interest (ROIs) were outlined in normal liver tissue, in the primary tumor as well as in all liver, lymph node, soft tissue, and bone metastases on PET/CT fusion images. For the liver ROI, the normal tissue of the liver was chosen with care as not to include possible metastases, present in the liver tissue of some GEP NET patients. The ROIs positioned were verified in all three planes (transversal, coronal, and sagittal).

SUVmax for all outlined ROIs in both studies (68Ga-DOTATATE and 68Ga-DOTATOC) were determined and mean SUVmax values for 68Ga-DOTATATE and 68Ga-DOTATOC in normal liver tissue, primary tumors, and metastases (in liver, lymph nodes, soft tissues, and bones) were calculated and compared.

Statistical analysis was performed using a General Linear Model for Repeated Measures procedure. Radiopharmaceutical had two variables, and the type of tissue had six variables (primary GEP NET, metastases in the liver, soft tissue, lymph nodes, bones, and normal liver tissue). Sidak’s test enabled comparison between accumulation of both radiopharmaceuticals (SUVmax values) in primary tumor, in metastases in liver, lymph nodes, bones, soft tissues, and in normal liver tissue.

34.3 Results

The intravenous injections of 68Ga-DOTATATE and 68Ga-DOTATOC were well tolerated. No local or systemic side effects were evident during the time of observation (up to 180 min post injection). Examples of data reconstruction and image analysis are displayed in Fig. 34.2a for 68Ga-DOTATATE and in Fig. 34.2b for 68Ga-DOTATOC.

Fig. 34.2
2 groups of readings of maximum intensity projection and fused P E T C T scans in 2 rows on the sagittal and coronal plane of the liver metastasis, with 68 G a-DOTATATE in group A and 68 G a-DOTATOC in group B.

(a) Maximum intensity projection, transversal, sagittal and frontal (coronal) fused PET/CT images obtained with 68Ga-DOTATATE in the same patient with pancreatic corpus and tail NET displayed. A ROI around a metastasis in S5/6 of the liver is outlined on all fused images. SUVmax = 21.5. (b) Maximum intensity projection, transversal, sagittal, and frontal (coronal) fused PET/CT images obtained with 68Ga-DOTATOC in a patient with pancreatic corpus and tail NET displayed. A ROI around a metastasis in S5/6 of the liver is outlined on all fused images. SUVmax = 26.0

On 76 PET/CT studies (38 68Ga-DOTATATE and 38 68Ga-DOTATOC in 38 GEP NET patients), 548 regions of interest (36 over primary tumors, 196 over liver metastases, 134 over lymph nodes metastases, 86 over bone metastases, 34 over soft tissue metastases, 62 over normal liver tissue) were outlined and SUVmax values were determined and compared (Table 34.1).

Table 34.1 Results of SUVmax measurements in 68Ga-DOTATATE and 68Ga-DOTATOC PET/CT studies on 38 GEP NET patients with stable disease

For both 68Ga-DOTATATE and 68Ga-DOTATOC, the highest SUVmax values were measured in primaries, followed by liver-, soft tissue- and lymph node-metastases. Of all metastases, bone metastases had the lowest SUVmax for both radiopharmaceuticals (Fig. 34.3). Almost identical values for 68Ga-DOTATATE and 68Ga-DOTATOC SUVmax were registered in normal liver tissue.

Fig. 34.3
A graph illustrates the S U V max values of 68 G a-DOTATATE and 68 G a-DOTATOC, where the labels read, primary tumor, metastasis in the liver, soft tissue metastasis, lymph node metastasis, bone metastasis, and normal liver tissue.

Comparison between mean SUVmax values of 68Ga-DOTATATE and 68Ga-DOTATOC in the normal liver, primary tumor and its metastases (in the liver, lymph nodes, soft tissues, and bones)

68Ga-DOTATOC had higher SUVmax values (mean SUVmax = 15.10) compared to 68Ga-DOTATATE (mean SUVmax = 12.67) in all measured tumor tissues (the difference was statistically highly significant: F = 27.174; p < 0.001) (Tables 34.1 and 34.2). The significance of the differences between 68Ga-DOTATOC and 68Ga-DOTATATE accumulation (SUVmax values) in primary GEP NET tumors (Sidak’s test results), in metastases in the liver, lymph nodes, bones, and soft tissues, and in normal liver tissue are displayed in Table 34.2.

Table 34.2 Significance of the difference between 68Ga-DOTATOC and 68Ga-DOTATATE accumulation (SUVmax values) in primary GEP NET tumors, in metastases in the liver, lymph nodes, bones and soft tissues, and in normal liver tissue

Comparison of radiopharmaceutical uptake (SUVmax) in different tumor tissues, as well as in tumor tissue and normal liver tissue, where a significant difference was found (for both 68Ga-DOTATATE and 68Ga-DOTATOC), is shown in Table 34.3. Highly significant difference was registered in primary tumor, in liver metastases, and in lymph node metastases. Significant difference was registered in bone metastases. Notwithstanding these highly significant differences in SUV max values, considerable variability in the preferred tracer uptake was still observed, as shown in Table 34.4. No significant difference existed in soft tissue metastases (higher values were obtained for 68Ga-DOTATOC compared to 68Ga-DOTATATE, but without statistically significant difference), and in normal liver tissue.

Table 34.3 Comparison of radiopharmaceutical uptake (SUVmax) in different tumor tissues, as well as in tumor tissue and normal liver tissue, where a significant difference was found
Table 34.4 Frequency of higher SUVmax for 68Ga-DOTATATE and 68Ga-DOTATOC in 243 GEP NET lesions

34.4 Discussion

In vivo molecular imaging by somatostatin analogue-based PET/CT has become the gold standard in clinical practice for diagnostics of GEP NET [31, 32]. These peptides exhibit fast pharmacokinetics, fast target localization, fast blood clearance, and fast renal excretion [33]. Scanning time is short and radiation dose is low [33]. Additional qualities are high sensitivity, high resolution, high detection rate, high image contrast, and the possibility of accurate quantification [33].

In this study, we have demonstrated a significantly higher accumulation of 68Ga-DOTATOC compared to 68Ga-DOTATATE, not only in primary tumors, but also in metastases (hepatic, lymph node and bone). In soft tissue metastases, 68Ga-DOTATOC SUVmax tended to be higher too, but the difference was not statistically significant. The results from this study are very similar to the results obtained by Poeppel et al. [27, 28], who also found significantly higher SUV max for 68Ga-DOTATOC compared to 68Ga-DOTATATE, although actual SUVmax values for different lesion types show differences between both studies, which may be explained by individual variations in receptor densities. Nevertheless, for 68Ga-DOTATOC, Poeppel et al. [28] reported the highest SUVmax for primary tumors, followed in decreasing order by metastases in liver, lymph nodes and bone, which is identical to the order found in this study. For 68Ga-DOTATATE, a similar sequence was observed in this study, albeit with lower SUVmax values than for 68Ga-DOTATOC, whereas Poeppel et al. reported a marginally higher SUVmax in liver metastases than in primary tumors for 68Ga-DOTATATE. However, it needs to be emphasized that the difference in SUVmax between primary tumors and hepatic metastases in the present study was not significant, whereas this was not tested by Poeppel et al.

Explanations for the different SUVmax values of 68Ga-DOTATOC and 68Ga-DOTATATE [27, 29] have been sought in the different in vitro affinity profiles for the various sst receptor subtypes of these ligands [6, 7]. 68Ga-DOTATATE has the highest in vitro affinity for sstr 2 (IC50 = 0.20 ± 0.04 nM/L), even higher than natural somatostatin-28 (IC50 = 2.7 ± 0.3 nM/L). Affinity of 68Ga-DOTATOC for sstr 2 (IC50 = 2.5 ± 0.5 nM/L) is similar to natural somatostatin-28, but tenfold lower than the affinity of 68Ga-DOTATATE. However, 68Ga-DOTATOC has 39% higher affinity for sstr 3 (IC50 = 613 ± 140 nM/L) and fivefold higher for sstr 5 (IC50 = 73 ± 21 nM/L) compared to DOTATATE (IC= > 1000 nM/L for sstr 3 and IC50 = 377 ± 18 nM/L for sstr 5) [26, 34]. However, at the peptide mass dose used in this study, the highest tracer plasma concentration that can be reached is in the low nanomolar, if not the subnanomolar, range. At this concentration, the only appreciable binding that may be expected for both tracers is with sstr 2 and any binding to other sst receptor subtypes will be very minor or negligible, even when the protein mass of the different receptor subtypes would be comparable. Predicted binding will be even less when taking into account that only a very small fraction of the peptide is actually radiolabeled and that the in vitro affinity of the unlabeled tracers is lower than for the 68Ga-labeled tracers [26].

These latter considerations would certainly correspond with the significant correlation of membranous sstr 2 expression as determined by immunohistochemistry and 68Ga-DOTATOC PET/CT SUVmax in patients with NETs [35]. Interestingly, Kaemmerer et al. demonstrated highly significant correlations between the SUVmax in 68Ga-DOTANOC PET/CT scans and immunoreactive scores of both sstr 2A and sstr 5 of NET patients [36]. Importantly, Wild et al. have reported for 68Ga-DOTANOC an sstr 2 IC50 = 1.9 ± 0.4 nM/L and an sstr 5 IC50 = 7.2 ± 1.6 nM/L, both values in the low nanomolar range [37]. However, even if for 68Ga-DOTATOC and 68Ga-DOTATATE only binding to sstr 2 plays a role, the higher SUVmax of 68Ga-DOTATOC is in remarkable contrast with the reported >10 times higher in vitro affinity of 68Ga-DOTATATE for sstr 2. As Poeppel et al. concluded, this >10 times higher in vitro affinity of 68Ga-DOTATATE did not prove to be clinically relevant. The affinity profiles for the various somatostatin receptor subtypes have been established in vitro in transfected cell cultures. The in vivo affinity for these somatostatin receptor subtypes expressed in their native environment may be further determined by, for instance, allosteric modulation and/or receptor clustering, which may impact the various ligand affinities differently. In this respect, it is highly interesting to note that in transfected cells expressing both sstr 2 and sstr 5, substantial heterodimerization was observed depending on the type of sst-agonist. The sstr 2/sstr 5 heterodimerization resulted in augmented receptor recycling and an approximate tenfold increase in efficiency for G-protein-coupling and MAPK activation [38, 39]. Unfortunately, the effects of DOTATOC and DOTATATE on formation of sstr homo- and heterodimers and their functional consequences are unknown.

In therapeutic applications with higher plasma concentrations of somatostatin analogues, binding to sstr subtypes with lower affinities may become more relevant. The most commonly expressed subtypes in GEP NETs are known to be sstr 2A and sstr 5. The expression of sstr 2 and sstr 5 has a prognostic role as reported by Corleto et al. [40]. They observed a significantly better survival rate in patients with well-differentiated endocrine carcinomas expressing sstr 2, sstr 5 and Ki-67 < 2%, treated with somatostatin analogues, compared to those with sstr 2 and sstr 5 negative tumors and Ki-67 ≥ 2% (p < 0.038). Five years survival rates were 91% vs. 43%, respectively. The positive prognostic role of expression of sstr 2 and sstr 5 is possibly related to the high affinity that the available somatostatin analogs display for these two specific sstr subtypes [40]. Interestingly, when going from well-differentiated endocrine tumors to poorly differentiated endocrine carcinomas, the densities of sstr 1, sstr 2, sstr 3, and sstr 4 all decrease, but the density of sstr 5 shows a substantial increase [25] (Fig. 34.4).

Fig. 34.4
A line graph differentiates the endocrine tumors from the endocrine carcinomas, where a decrease in the densities of s s t r 1 through 4, is noted, but the density of s s t r 5 increases substantially.

Percentage of GEP-NETs with sstr 1–5 expression according to WHO classification. WDET well-differentiated endocrine tumor, WDEC well-differentiated endocrine carcinoma, PDEC poorly differentiated endocrine carcinoma. (Adapted with permission from [25], copyright Elsevier)

We could not register differences in uptake between the two radiopharmaceuticals in normal liver tissue, which had significantly lower SUVmax values than primaries and liver- and lymph node metastases. In a previous publication [41], we have compared in vivo uptake of 68Ga-DOTATOC in liver and prostate (being threefold lower in the prostate) with in vitro sstr 2 expression (being sixfold greater in prostate) [42]. Higher liver uptake of 68Ga-DOTATOC than expected on the basis of sstr 2 expression has been related to normal peptide metabolism in the liver [42, 43]. Our result of almost identical uptake of 68Ga-DOTATOC and 68Ga-DOTATATE in normal liver tissue supports this explanation. 68Ga-DOTATOC SUVmax values for renal parenchyma at 1 h, 2 h, and 3 h and for liver at 1 h and 2 h were found to be lower than 68Ga-DOTATATE SUVmax values [29], while the tumor-to-kidney ratio has been reported to be higher than that of 68Ga-DOTATATE [29, 44]. Added to the higher 68Ga-DOTATOC SUVmax values for primary tumors and metastases, this would make 68Ga-DOTATOC, the preferred peptide for somatostatin receptor imaging. However, we have also shown considerable variability in preferred peptide uptake, suggesting that for therapy planning somatostatin receptor imaging with both peptides would be optimal, ideally with the same peptide amount as planned for therapy. Unfortunately, although 177Lu is currently the most widely used radionuclide for peptide receptor radionuclide therapy in NET patients, the affinity profiles of 177Lu-DOTATOC and 177Lu-DOTATATE for the various sstr have not been published to date.

34.4.1 Potential Limitations

For each patient, the first somatostatin receptor imaging reported in this study was performed with 68Ga-DOTATATE, the second with 68Ga-DOTATOC. However, as already mentioned, we have only selected patients, with clinically, biochemically, and morphologically stable disease on both occasions.

Partial volume effect in this study, described for lesions whose diameter is smaller than two or three times the scanner resolution [45], had no influence on the results, since both radiopharmaceuticals were compared under the same conditions in the same patients, in the same primaries and in the same metastases.

34.5 Conclusions

On average, 68Ga-DOTATOC shows significantly higher uptake in GEP-NET primary tumors and metastases than 68Ga-DOTATATE. However, we have also observed considerable variability in preferred peptide uptake. Optimal therapy planning would therefore require somatostatin receptor imaging with both these peptides.