| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Medical Sciences/Endocrine Oncology (H.O., K.O., B.S., B.E.), Radiology (A.S.), Nuclear Medicine (U.G.), and Surgery (C.J.) and Uppsala University PET Center IMANET (A.S., B.L., M.B.), Uppsala University Hospital, S-751 85 Uppsala, Sweden
Address all correspondence and requests for reprints to: Hakan Orlefors, M.D., Ph.D., Department of Medical Sciences/Endocrine Oncology, Uppsala University Hospital, S-751 85 Uppsala, Sweden. E-mail: hakan.orlefors{at}medsci.uu.se.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
). For MGCs, 5-HT, urinary 5-HIAA, and chromogranin A are the main tumor markers (4), whereas in foregut carcinoids [bronchial carcinoids and endocrine pancreatic tumors (EPTs)] and hindgut carcinoids, 5-HT production is rare; therefore, 5-HIAA is rarely increased. Still, production of 5-HTP can also occur in these groups; therefore, immunohistochemical staining for 5-HT may be positive (1).
|
The standard positron emission tomography (PET) tracer in oncology, 18F-fluorodeoxyglucose, has shown to be of limited value in the imaging of NETs (14, 15). In contrast, with PET using the 11C-labeled 5-HT precursor 5-HTP as tracer, we have previously demonstrated a high tracer uptake in a limited number of patients, where the change of transport rate constant indicated PET also to be useful in therapy monitoring of NETs (16).
The dopamine precursor L-dihydroxyphenylalanine (DOPA), labeled with 11C or 18F, has also been used for PET imaging of NETs (17, 18). In the former of these studies, three patients were coexamined with 11C-5-HTP, and in these patients, higher standardized uptake values were seen for 5-HTP than for DOPA (Bergstrom, M., unpublished data), indicating that 5-HTP might be preferable for the amine uptake system of NETs. On the other hand, since these tracers are actively internalized and rapidly decarboxylated intracellularly (19, 20), a functionally inactive or necrotic tumor can be overlooked with these imaging techniques. In the study by Hoegerle et al. (18), using 18F-DOPA-PET in 17 patients with gastrointestinal carcinoids, the highest sensitivity was seen for the combination of nonfunctional techniques, including both CT and MRI.
For the diagnostic imaging of metastatic pheochromocytomas, PET with 18F-dopamine has been shown to be highly sensitive, and recent publications indicate that this imaging modality is superior to both SRS and 131I-metaiodobenzylguanidine scintigraphy (21, 22).
Our hypothesis was that the system for uptake and internalization of an amine precursor might be expressed to a certain degree in all APUD-omas. Therefore, by using a radiolabeled amine precursor as a PET tracer, our objective was to investigate whether this method can be used as a universal imaging technique for visualization of all tumors histopathologically characterized as being of neuroendocrine differentiation. We chose 11C-labeled 5-HTP as the tracer based on previous observations (see above) and aimed at including a mixture of different NETs for diagnostic imaging, of both primary tumors and metastases, with PET in comparison with SRS and CT.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Forty-two consecutive patients with NETs, referred to the Department of Endocrine Oncology (Uppsala University Hospital, Uppsala, Sweden) were included in the study (six bronchial, 16 midgut, two foregut, and two thymic carcinoids; one ectopic Cushing’s syndrome; four gastrinomas; one insulinoma; one gastric carcinoid [enterochromaffine like (ECL)-oma]; six nonfunctioning EPTs; two endocrine-differentiated pancreatic carcinomas; and one paraganglioma), including three patients with MEN-1.
Inclusion criteria were: 1) histopathological diagnosis of NET and detected lesion on any conventional radiology (CT, abdominal, or endosonography) or on SRS; or 2) biochemical evidence of NET and detected lesion on any conventional radiology or on SRS.
In all patients, biopsy or surgery was performed for histopathological diagnosis. Patient characteristics are shown in Table 1. Nineteen patients had been submitted to previous tumor surgery, and 17 patients were operated on after the biochemical and imaging work-up, allowing surgical comparison of the imaging results in this subgroup of patients.
|
|
PET
11C was produced using a Scanditronix 17 MeV cyclotron (General Electrics Medical Systems, Milwaukee, WI), and 11C-5-HTP was produced in a multienzymatic reaction according to previously described procedures (23, 24). The patients were examined using a Siemens ECAT HR+ PET scanner (Siemens, Munich, Germany) with four to five bed positions, each with a 13.6-cm axial field of view providing 2.5-mm slices with a resolution of approximately 5.5 mm. The emission times were typically 5, 6.7, 10, 10, and 15 min at bed positions 1–5, respectively. A 5-min segmented whole-body (WB) transmission scan was acquired at all bed positions using externally rotating 68Ge-pins. The images were corrected for scatter and attenuation, then reconstructed in a 128 x 128 matrix to represent radioactivity concentration using an iterative reconstruction algorithm with six iterations, 16 subsets, and an 8-mm ramp filter. WB-PET scanning was started 20 min after iv injection of 11C-5-HTP at a dose of 140–521 MBq (mean, 381 MBq). To reduce tracer decarboxylation by blocking the enzyme aromatic amino acid decarboxylase, all patients received 200 mg carbidopa as pretreatment 1 h before the PET examination as described in a recent communication (Orlefors, H., A. Sundin, L. Lu, K. Oberg, B. Langstrom, B. Eriksson, and M. Bergstrom, submitted manuscript).
SRS
SRS was performed using 111In-DTPA-D-Phe1-octreotide, labeled as previously described (25) and delivered by Mallinckrodt Medical (Petten, Netherlands). Patients were injected with one kit, 6 mCi (222 MBq), as recommended by the producer, and planar anteroposterior as well as lateral scintigrams were collected after 24 h with a large field-of-view 
-camera and a medium-energy collimator. Static WB images were obtained, and single photon emission CT was additionally performed after 24 h using a single-headed -scintillation camera (Nuclear Diagnostics, Hagersten, Sweden, and London, UK), and the data collection was performed using a 64-step rotation of 360° in a 64 x 64 word matrix and 40-sec acquisition per projection. Images were iteratively reconstructed in four subsets, four iterations, and no postfiltering (Hermes Ordered Subset Expectation Maximization, Hermes, Stockholm, Sweden). In patient 14, a regular-filtered back projection was used.
CT
WB-CT was performed, using two different scanners (Somatom Plus 4 and Somatom Plus S, Siemens), over the thorax and abdomen in all patients. CT was performed before and during iv contrast enhancement using 8-mm slice thickness and increment. For CT of the pancreas, 3-mm slice thickness and 4.5-mm increment were additionally used in the arterial contrast enhancement phase.
The image findings were assigned to the following categories: liver metastases (l.m.), abdominal lymph node metastases (a.lgll.m.), lung metastases (lung.m.), mediastinal lymph node metastases (m.lgll.m.), and bone metastases (bone.m.). When more than five lesions were detected in one of these categories, this was described as >5 lesions. The images were interpreted by radiologists (for CT) and nuclear medicine physicians (for SRS) who were blinded for the results of the other imaging modalities. Comparison between imaging techniques was made by two of the authors (H.O. and A.S). All examinations in each patient were performed during a maximum period of 8 wk (mean ± SD, 3.2 ± 2.8 wk).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
. Four patients (three MGCs and one EPT) had to be excluded from the study. For two of these patients, there was a change in the medical therapy between the examinations that altered the baseline conditions and made comparison of modalities unreliable. One patient did not want to continue the study after performing two of the three imaging modalities, and the last subject was excluded after one examination due to severe pain.
Overall comparison
The greatest number of lesions was visualized with 11C-5-HTP-PET. In 95% (36 of 38) of the patients, PET could visualize positive lesions. Patients 20 and 37 (see below) were negative on PET. In 58% (22 of 38) of the patients, more lesions were detected with PET than with SRS and CT. In 34% (13 of 38) of the patients, imaging with PET visualized equal numbers of lesions as SRS and/or CT. In three cases (patients 20, 35, and 37), SRS or CT showed more lesions than PET. Patient 20 had a recurrence of a nonfunctioning EPT, visible solely on SRS as a lymph node metastasis (Table 1). Patient 35, with a diagnosis of MEN-1, displayed liver metastases and mediastinal lymph node metastases of a thymic carcinoid as well as a primary tumor in the thymus on CT. Both SRS and PET could image the primary tumor but not the metastases. In patient 37, disseminated tumor disease of a pancreatic carcinoma, with some endocrine differentiation on immunohistochemistry, was detected on CT. Two lesions were slightly positive on SRS, whereas PET only showed areas as totally devoid of activity in the pancreas and liver. These three patients will be discussed further in the Discussion.
SRS could detect more lesions than CT in 37% of the patients (14 of 38), whereas in 21% (eight of 38) of the patients, CT showed more lesions than SRS.
PET-positive tumor lesions in 33 of 36 PET-positive patients (92%), a total of 51 lesions, were histopathologically confirmed as tumors with biopsy or surgery (Fig. 2), and no false-positive PET lesions were found among the examined samples. In patient 12, with suspected residual disease after primary surgery, the lesions were not large enough to be transabdominally biopsied, and surgery was not indicated. In patients 27 and 34, with residual or recurrent disease after primary tumor surgery, there was a discussion about whether this could represent postoperative changes, but based on the functional information of the positive PET images, active treatment intervention was suggested in both cases.
|
). The primary tumors not visualized on PET were two MGCs (patients 5 and 9) and one endocrine pancreatic cancer (patient 37). For three of the MGC patients, the area of the primary tumor was included in a larger PET-positive tumor mass.
|
), and the contrast in the images was higher on PET than on CT.
|
). Sixteen of these patients displayed PET-positive lesions (patient 20 was negative on PET), and in 15 of those cases, surgery could confirm tumor findings. In patient 23, with a nonfunctioning EPT, surgery could not verify PET findings except for a liver lesion that was radiofrequency ablated during operation. In this patient, who had been submitted to previous abdominal surgery, a new PET scan was performed postoperatively, where three PET-positive tracer uptakes were still present. Twelve of the operated patients had a remaining primary tumor. Nine of these primary tumors were confirmed surgically, and all of them were positive on PET. The sizes of the surgically removed PET-positive primary tumors were in the range of 6–30 mm. In two patients with MGC, the operation was not directed toward the primary gut tumor; therefore, this lesion was not verified. In patient 31, only the lymph node metastasis could be removed at reoperation.
Additionally, biopsy could confirm four of the nonoperated PET-positive primary tumors (patients 15, 24, 30, and 35), whereas the final patient with liver metastases from a foregut carcinoid and a previously unknown primary tumor (patient 32), now displaying a positive PET image over the pancreas, has not yet been confirmed and therefore could not be accounted for.
PET and anatomical imaging (CT)
No tumor could be imaged with CT in nine patients (Table 1). Two patients expressed no tracer uptake and therefore could not be imaged on the PET scans (patients 20 and 37).
In patients with MGC (patients 1–13), PET imaged abdominal lymph node metastases in all 13 patients, liver metastases were imaged in 10 of 13 patients, mediastinal lymph node metastases in seven of 13 patients, and bone metastases in three of 13 patients. The corresponding numbers for imaging with CT were six of 13, six of 13, four of 13, and one of 13 patients.
In the group of EPTs (patients 14–23), four patients had a remaining primary tumor (patients 14, 15, 17, and 19). All of these tumors were visualized by PET, whereas CT could visualize only one (patient 15). Abdominal lymph node metastases were imaged with PET in six of 10 patients, liver metastases in five of 10, and lung metastases in one of 10, compared with three of 10, four of 10, and one of 10 patients, respectively, for CT.
In bronchial carcinoids (patients 25–31), PET detected the primary tumor in four of four patients, liver metastases in two of seven, bone metastases in one of seven, and mediastinal lymph node metastases in three of seven. Corresponding numbers for CT were three of four, one of seven, one of seven, and one of seven patients.
Both modalities imaged the ECL-oma (patient 24), whereas the paraganglioma (patient 36) only was positive on PET. For tumors initially classified as foregut carcinoids (patients 32 and 33), later specified as EPT and duodenal carcinoid, PET could image the primary as well as the liver metastases in both cases, whereas only the liver metastases in patient 32 could be imaged by CT. The primary thymic carcinoid of patient 35 was imaged by both modalities as well as the mediastinal lymph node metastases of the second patient with thymic carcinoid (patient 34), whereas the liver and lymph node metastases of this patient could only be imaged by CT. These liver metastases were rapidly progressive and expressed a high proliferation index.
In the two patients with pancreatic cancers with some endocrine differentiation on immunohistochemistry and a high proliferation index, PET was totally negative in patient 37, whereas CT imaged extensive disease, and in patient 38, PET could image liver metastases that were not visible on CT. Both of these cases are tumors with a lower degree of differentiation than what is common in NETs.
PET and SRS
In MGCs, SRS imaged abdominal lymph node metastases in 10 of 13 patients, liver metastases in four of 13, mediastinal lymph node metastases in seven 13, and bone metastases in two of 13. Corresponding numbers for PET were, as seen above, 13 of 13, 10 of 13, seven of 13, and three of 13 patients.
Primary EPTs could be visualized in two of four cases with SRS, whereas PET imaged all four. Abdominal lymph node metastases were imaged with SRS in three of 10 EPTs, liver metastases in three of 10 patients, and lung metastases in zero of 10 patients. This is to be compared with six of 10, five of 10, and one of 10 patients, respectively, for PET. However, bone metastases were seen in one case (patient 18) on SRS but not PET.
Regarding bronchial carcinoids, SRS imaged the primary tumor in two of four patients, compared with four of four patients with PET. Liver metastases, bone metastases, and mediastinal lymph node metastases were visualized by SRS in one of five, one of five, and three of five patients, respectively. Corresponding numbers for PET were two of five, one of five, and three of five patients, respectively.
Both techniques imaged the ECL-oma, the paraganglioma, and the primary thymic carcinoids, whereas in the patients initially classified as foregut carcinoids (patients 32 and 33), PET could image lesions at more sites than SRS in both cases. Two lesions in patient 37 were seen on SRS but not on PET, whereas in patient 38, PET could image both primary and liver metastases, whereas only liver lesions were seen on SRS.
Tumor subgroup results
All four gastrinomas were readily imaged with 11C-5-HTP-PET. Two of these patients displayed tumors that were found to be less than 2 cm in diameter at surgery (patients 14 and 17; Figs. 3 and 4). Their peptide hormone production was markedly elevated. In contrast, four of five nonfunctioning EPTs, i.e. tumors that cause no hormonal symptoms, revealed nevertheless elevated 5-HTP uptake and consequently were imaged by PET.
All seven bronchial carcinoids were depicted by PET, including the patient with Cushing’s syndrome due to ectopic ACTH production (patient 29). This tumor was surgically removed and measured approximately 2 cm in diameter. This is interesting since ectopic ACTH-producing tumors are known to be difficult to image (26, 27). Additional studies are needed to support this observation. However, the residual/recurrent tumor of a patient that previously had been operated for an ACTH-producing bronchial carcinoid causing Cushing’s syndrome (patient 31), now with normal ACTH and cortisol levels, could also be visualized by PET.
In patient 28, who previously was submitted to primary surgery for a bronchial carcinoid and now with a recurrence in thorax, PET could, apart from the positive mediastinal lesion (also seen on SRS, but only retrospect at CT), visualize multiple liver metastases (Fig. 5). After repeated ultrasonography examinations, tumor spread to the liver was confirmed with biopsy, and the patient was subjected to medical treatment instead of surgery.
|
) as well as several small intraabdominal lesions. In retrospect, by rereading the CT images with knowledge of the PET findings, many of these previously overlooked lesions could be detected.
In contrast, also, the less common NETs, such as a gastric carcinoid (patient 24), a cervical nonfunctioning paraganglioma (patient 36), and two thymic carcinoids (patients 34 and 35), were clearly depicted by 11C-5-HTP-PET. For patient 36, with a recurrence of a paraganglioma, the PET images were very supportive in the clinical management since both CT and MRI initially were negative. In this case, SRS also clearly imaged the lesion in the neck (Fig. 6).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The two patients in whom PET could not visualize any tumor (patients 20 and 37) had a recurrence of a nonfunctioning EPT and an endocrine pancreatic carcinoma with poor differentiation and high proliferation index (>40%). In one other case, PET displayed less tumor lesions than SRS and CT. This was in patient 35, with a thymic carcinoid and liver metastases with a high proliferation rate. In common for these three cases is a low peptide hormone production and, therefore, almost normal biochemical markers. This might indicate that the amine precursor uptake system is not as well expressed in these tumors, resulting in a low uptake of the radiolabeled amine precursor 5-HTP and thereby a poor tumor imaging. In these patients, 18F-fluorodeoxyglucose could have been a better choice of PET tracer for tumor visualization. Since 11C-5-HTP is incorporated in a biochemical pathway, PET imaging with this tracer reflects the metabolic activity of a tumor concerning processing of biogenic amines. Consequently, some lesions show lower tracer uptake, and occasional lesions such as necrotic tumors were shown to lack tracer accumulation. Also, tumors with very low peptide production, i.e. nonfunctioning tumors or poorly differentiated tumors, can be difficult to detect using this concept. This was illustrated by patient 19, with multiple EPTs as part of a MEN-1 syndrome. This patient only showed a slight elevation of pancreatic polypeptide (PP) levels and chromogranin A; consequently, the 5-HTP uptake was only slightly increased in two pancreatic lesions, although sufficient for visualization by PET. On the other hand, when surgery was performed in this patient, three EPT lesions were found, and the smallest tumor measured 4 mm in diameter.
A drawback in this study is the fact that surgical confirmation of the imaging results could only be achieved in 17 of 38 patients. For the remaining 19 patients, there was no clinical indication to perform surgery. In this subgroup of 17 patients, the sensitivity of 11C-5-HTP-PET was shown to well surpass that of both SRS and CT.
In all but three cases (patients 12, 27, and 34), material from lesions that were positive on the PET scans were found to represent tumor tissue, indicating that the uptake seen on PET truly represent NET. Indeed, surgery is the gold standard for verification of tumor lesions. However, this study was not designed to compare PET with surgery; therefore, a full surgical lesion mapping was not performed.
When 11C-5-HTP-PET was compared with SRS and CT, more tumor lesions were detected with PET in a majority of cases. SRS, however, still defends its place as the functional imaging method of choice for NETs due to its availability and capacity to reflect the expression of somatostatin receptors, which forms the basis for therapy with nonradioactive and ß-emitting-labeled somatostatin analogs (peptide receptor radionuclide therapy). In combination with conventional radiology, SRS probably is sufficient as work-up in a majority of patients with NETs (5), but as this study indicates, 11C-5-HTP-PET contributes in a majority of cases, especially with regard to small tumor lesions. This new and fairly expensive technology is probably most beneficial in selected patients such as those with biochemical evidence of an endocrine tumor or tumor recurrence but with negative imaging work-up, as well as to find possible metastases in patients where the aim is curative surgery. Another situation where 11C-5-HTP-PET can contribute is when considering liver transplantation in a patient with solely liver metastases of a NET. In that case, it is crucial to exclude extra hepatic tumor sites before introducing potent immunosuppressive treatment.
So far, 11C-5-HTP-PET can only be performed at centers with access to a cyclotron for synthesis of the radionuclide 11C due to its short half-life (20 min). On the other hand, with the use of 11C, the tracer molecule remains structurally and biologically intact, which is important when targeting endocrine pathways using small molecules such as 5-HTP. Nevertheless, a labeling, e.g. with 18F, is needed to facilitate the spread of this imaging technique.
To conclude, this study indicates that 11C-5-HTP-PET is a sensitive method for detection of NETs, and it exceeds both SRS and CT in tumor visualization. The contribution in information regarding patient tumor status with this technique is considerable. With the exception of poorly differentiated NETs and possibly nonfunctioning tumors, we believe that 11C-5-HTP can be used as a universal technique for imaging of NETs, with the greatest benefit in imaging of small tumor lesions. This study also reflects that NETs process biogenic amines to some degree regardless of functionality and endocrine syndrome.
|
Footnotes |
|---|
First Published Online March 8, 2005
Abbreviations: APUD, Amine precursor uptake and decarboxylation; CT, computed tomography; DOPA, dihydroxyphenylalanine; ECL, enterochromaffine like; EPT, endocrine pancreatic tumor; 5-HIAA, 5-hydroxyindole acetic acid; 5-HT, serotonin; 5-HTP, 5-hydroxytryptophan; MEN-1, multiple endocrine neoplasia type 1; MGC, midgut carcinoid; MRI, magnetic resonance imaging; NET, neuroendocrine tumor; PET, positron emission tomography; SRS, somatostatin receptor scintigraphy; WB, whole body.
Received September 30, 2004.
Accepted March 2, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. P. Koopmans, O. C. Neels, I. P. Kema, P. H. Elsinga, W. J. Sluiter, K. Vanghillewe, A. H. Brouwers, P. L. Jager, and E. G.E. de Vries Improved Staging of Patients With Carcinoid and Islet Cell Tumors With 18F-Dihydroxy-Phenyl-Alanine and 11C-5-Hydroxy-Tryptophan Positron Emission Tomography J. Clin. Oncol., March 20, 2008; 26(9): 1489 - 1495. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gabriel, C. Decristoforo, D. Kendler, G. Dobrozemsky, D. Heute, C. Uprimny, P. Kovacs, E. Von Guggenberg, R. Bale, and I. J. Virgolini 68Ga-DOTA-Tyr3-Octreotide PET in Neuroendocrine Tumors: Comparison with Somatostatin Receptor Scintigraphy and CT J. Nucl. Med., April 1, 2007; 48(4): 508 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F. Scarsbrook, A. Ganeshan, J. Statham, R. V. Thakker, A. Weaver, D. Talbot, P. Boardman, K. M. Bradley, F. V. Gleeson, and R. R. Phillips Anatomic and Functional Imaging of Metastatic Carcinoid Tumors RadioGraphics, March 1, 2007; 27(2): 455 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Findling and H. Raff Cushing's Syndrome: Important Issues in Diagnosis and Management J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3746 - 3753. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
