Somatostatin-receptor scintigraphy in the management of gastroenteropancreatic tumors

Vol. 93, No. 1, 1998 ISSN 0002-9270/98/$19.00 PII SGilO2-9270(97)0@324-5

THE AMERICAN JOURNALOF GA~~ROENTER~LOGV

Copyright 0 1998 by Am. COIL of Gastroenterology Published by Elsevier Science Inc.

Somtitostatin-Receptor Scintigraphy in the Management of Gastroenteropancreatic Tumors Yodphat

Krausz, M.D., Jacob Bar-Ziv, M.D., Rolf B. J. de Jong, M.D., Sophia Ish-Shalom, Roland Chisin, M.D., Nazeeh Shibley, C.N.M.T., and Benjamin Glaser, M.D.

M.D.,

Departments of Nuclear Medicine, Endocrinology and Metabolism, and Radiology, Hadassah University Hospital, Jerusalem, and Departmerlt qf Endocrinology, Rumbum Medical Center, Haifa, Israel; and Mallinckrodt Medical, Petten, Holland

INTRODUCTION

Objective: This study evaluates the diagnostic and therapeutic implications of somatostatin-receptor scintigraphy in the management of patients with proven or high clinical suspicion of gastroenteropancreatic endocrine tumors. Methods: Forty-one patients were studied by planar and tomographic imaging at 4 h and 24 h after ’ “In-pentetreotide injection. Scintigraphic findings were compared with computed tomography, and in several patients also with ultrasound, angiography, biopsy, and/or surgery, when performed. Results: Among 23 patients with carcinoid tumor, three of nine primary tumors were initially identified by scintigraphy. Unsuspected mesenteric metastases found on scintigraphy in three patients led to octreotide treatment. Scintigraphic detection of multiple metastases in a patient with thyroid metastasis of bronchial carcinoid spared her an unnecessary total thyroidectomy. Among 18 patients with 19 islet-cell tumors, scintigraphy detected three of live insulinomas, whereas computed tomography identified only one. Receptor positivity in an islet-cell tumor (vipoma?) with no metastases on the scan led to surgical removal of the primary tumor. Receptor-positive metastases of gastrinoma (two of three patients), glucagonoma (two of three patients), and parathyroid hormone-related peptide-producing tumor (one patient) led to octreotide treatment. Nonvisualization of metastases of a glucovipoma led to chemotherapy. Conclusions: Somatostatin-receptor scintigraphg contributes to the management of patients with gastroenteropancreatic tumors in the following ways: 1) localization of a primary occult tumor, allowing surgical removal; 2) staging of the disease for optimal therapy-surgical excision or systemic treatment; and 3) identification of receptor status of the metastases for octreotide treatment or chemotherapy. (Am J Gastroenterol 1998;93:66-70. 0 1998 by Am. Coll. of Gastroenterology)

Receilwd Apr. 22.

1997: accepterl Aug.

The gastroenteropancreatic (GEP) endocrine tumors (carcinoid and islet-cell tumors) originate from a variety of endocrine cells located in different organs and display various neuroendocrine characteristics (1). Carcinoid and isletcell tumors are rare (incidence of 2.1 per 100,000 and 0.4 per 100,000, respectively) (2), have a prolonged natural history, and may remain undetected for years despite the associated peptide hypersecretion. Localization of the primary tumor and metastatic foci, although frequently difficult, is of utmost importance because surgery remains the mainstay of treatment for localized and for some cases of advanced disease. Several radiological techniques have been suggested, but detection rates range from 13% to 85%, depending on type, site, size of the tumor, and the technique used (3). The discovery of dense high-affinity somatostatin receptors in membrane homogenates and tissue sections of GEP tumors (4, 5) led to the use of radiolabeled somatostatin analogues for in vitro autoradiography (6) and for in vivo scintigraphic detection (7) of these tumors. The in vivo technique provides whole-body screening of pathological findings and is associated with radiation exposure comparable to that of other imaging modalities (8). The present study evaluates the diagnostic and therapeutic implications of somatostatin-receptor scintigraphy (SRS) in the management of patients with GEP tumors. MATERIALS AND METHODS Patients

Forty-one patients with histological or biochemical evidence of GEP tumors were studied: 17 men and 24 women ranging in age from 22 to 86 yr. They were included in the protocol, approved by the ethics committee of Hadassah University Hospital, after giving written informed consent. Twenty-three patients had carcinoid tumor and 18 patients had 19 islet-cell tumors: five insulinomas, three glucagonomas, three gastrinomas, one somatostatinoma, one parathyroid hormone-related peptide (PTHrP)-producing tumor, one suspected vipoma, three multiple endocrine neoplasia type 1 (MEN-l), and two nonsecretory tumors. The patient

26, 1997. 66

AJG - Januap

1998

SOMATOSTATIN-RECEPTOR

with suspected vipoma had a typical clinical presentation, but no hormonal confirmation was available. One patient had both an insulinoma and a nonsecreting islet-cell tumor. Diagnosis was based on biochemical findings (nine patients‘), biopsy (five patients), and/or surgical specimen (27 patients). Computed tomography (CT) of the abdomen and/& chest was performed initially in all patients according to the specific clinical indication. This was followed by ’ ’ ‘In-pentetreotide scintigraphy, which was interpreted independently. When indicated, CT was reevaluated or repeated. with emphasis placed on regional abnormalities identified on scintigraphy. The scintigraphic findings were compared with a synthesis of biochemical findings with those of CT, ultrasound (US), angiography, biopsy, and surgery, when available.

Scintigraphy was performed after intravenous administration of ’ ’ ‘In-pentetreotide (222 MBq) (Octreoscan 111; Mallinckrodt Medical, Petten, Holland). Labeling,efficiency was 196% in all patients (SEP-PAK C,, chromatography; Waters, Division of Millipore Corporation, Milford, MA). Images were obtained using a large field-of-view rotating gamma camera (APEX-4 15ECT or HELIX; Elscint, Haifa, Israel) equipped with a medium-energy collimator. Planar images were recorded for 15 min or up to 500,000 counts over the skull, chest, abdomen, and pelvis to mid-thigh at 4 h, 24 h, and occasionally at 48 h after injection. Single photon emission computed tomography (SPECT) was acquired at 4 h and/or at 24 h after injection in 29 patients (26 of the abdomen and 13 of the chest). Acquisition parameters included a 64 X 64 matrix, 60 projections over 360”, and an acquisition time of 40 s per step. The raw data were backprojected using a Hanning filter to form transaxial, coronal, and sagittal tomograms of 0.68 cm thickness per slice. No attenuation correction was used because of the energy of “‘In. RESULTS The scintigraphic and CT findings of the 41 patients, analyzed with respect to the primary tumor, local recurrence, and distant metastases, are summarized in Tables 1 and 2. Tumor sites were identified by scintigraphy in 35 and by CT in 23 patients. Six patients had no detectable abnormal findings by either technique. Among 23 patients with carcinoid (Table l), SRS detected the primary tumor in three of nine nonoperated patients: an ileal tumor, an ovarian carcinoid, and a 7-mm adrenocorticotropic hormone-producing lung carcinoid. Visualization spared the latter patient bilateral adrenalectomy (9). Local recurrence was visible close to the previously resected tumor in the liver and lung of one and two patients, respectively, > 10 months after initial surgery, whereas CT was negative.

SCINTIGRAPHY

67

TABLE 1

Imaging Data of SRS and CT in 23 Patients With Carcinoid SRS Primary Ileum (2) ovary (1) Lung (1) Local recurrence Lung (2) Liver (1) Metastases Liver (11) Mesentery (5) Mediastinuni (2) Lung (2) Bone (1) Thyroid (1) Brain (1)

Tumor

CT

1 1 1

2 (1;). 1* 1*

2 1

0 0

11 5 2 1 1 1 1

9 2 1 1 * 1 1

Numbers in parentheses (first column) represent total number of patients with lesions known to be present at each location, using all available techniques. * Identified on CT only when the CT was reevaluated in light of the scintigraphic findings. TABLE 2

Imaging Data of SRS and CT in I8 Patients With 19 Islet-Cell SRS Insulinoma Pancreas (5) Glucagonoma Pancreas (2) Liver (3) Somatostatinoma Pancreas (1) Liver (1) Lung (1) Gastrinoma Pancreas ( 1) Antrum (1) Liver (2) Mesentery (1) PTHrP Liver (1) Vipoma Pancreas (1) MEN- 1 Pancreas (2) Liver (1) Mesentery ( 1)

Lung (1) Pancreas

(2)

Tumors

CT

1* 1 3 1 0 (surgery) 1 0

0 (surgery) 2 0 1

1

1

1

2 1 0

1* 1 1* 1

1

0 (2-surgery)

I

Numbers in parentheses represent total number of patients with lesions known to be present at each location, using all available techniques. * Identified on CT only when the CT was reevaluated in light of the scintigraphic findings.

Carcinoid metastases were observed in the liver of 11 patients and in the mesentery of nine patients by SRS, and in five and two patients, respectively, by CT. Diffuse metastatic disease was identified on whole-body screening of a

68

KRAUSZ et al.

FIG. 1. Abdominal SRS-SPECT (at 24 h) of a patient with insulinoma demonstrates a focus of pathological uptake in the head of the pancreas (+).

for total thyroidectomy because of intrathyroidal metastasis, sparing her unecessary surgery (10). Among 19 patients with islet-cell tumors (Table 2), SRSSPECT detected three of five insulinomas (Fig.’ l), CT localized only one, and in two patients all imaging techniques failed to localize the primary tumor. A primary glucagonoma was visible on SRS, and residual tumor was detected in a patient with glucovipoma several months after initial surgery. Liver metastases developed in the latter patient within months of initial diagnosis, visualized as photon-deficient areas on SRS (Fig. 2A), and the patient was treated with chemotherapy rather than with octreotide. In contrast, two patients with glucagonoma with slow-growing SRS-positive liver metastases responded clinically to octreotide. Likewise, visualization of liver metastases of PTHrP-secreting tumor on SRS (Fig. 2B) led to long-term octreotide treatment, with control of hypercalcemia. A primary somatostatinoma and its lung metastases were identified on both CT and SRS, but liver foci of
scheduled

A JG - Vol. 93, No. 1, I998

studied. One had residual tumor after excisional biopsy, not visualized on SRS, CT, magnetic resonance imaging, endoscopic retrograde cholangiopancreatography, or angiography, but removed at reoperation. The second patient, with biochemical evidence of insulinoma, did not respond to a test dose of octreotide with either suppression of insulin or an increase in serum glucose. SRS-SPECT revealed a pamological focus in the body of the pancreas, but a 1S-cm tumor removed at surgery did not stain for insulin, and fasting hyperinsulinemic hypoglycemia persisted. At reoperation, a l.l-cm tumor, identified in an adjacent area by intraoperative US, was removed and stained positively for insulin. No SRS was performed between the two surgical procedures. It is likely, however, that the nonfunctioning tumor was the one visible on SRS because the hypoglycemia was not relieved by octreotide. DISCUSSION ’ ’ ‘In-labeled pentetreotide detected pathological uptake consistent with tumor in 35 of 41 patients with GEP tumors, indicating an overall detection rate of 85%, which is in the range of 82-92% reported in the literature (8, 11-13). This scintigraphic technique aids in clinical decision making in the following situations. 1) In patients with a peptide-producing tumor, SRS may help identify the primary lesion and its neuroendocrine characteristics, as evident in the ovarian and lung carcinoid and as reported previously by others (14). The 7-mm adrenocorticotropic hormone-secreting lung carcinoid (9) was smaller than the one reported by Joseph et al. (1 l), and we have previously identified a 3-mm SRS-positive growth hormone-producing pituitary tumor (unpublished data). Thus, the sensitivity of SRS does not depend exclusively on tumor size, but on a high target-to-background ratio associated with dense, high-affinity receptors on the tumor surface compared with adjacent tissue. Markedly increased uptake was also observed in all of our gastrinoma patients. In fact, a gastrinoma was found in the pancreaticoduodenal area on SRS only, when other techniques, including endoscopic US, were negative. De Kerviler er al. (15) also found SRS to be superior to endoscopic US in detecting primary gastrinomas, in contrast to Hammond et al. (16). Scintigraphy was also superior to CT, US, and angiography in locating primary insulinomas, despite expression of octreotide-avid somatostatin receptors in only 46-618 of the tumors (17). Endoscopic and intraoperative US, the procedures of choice for such cases (18, 19) were not routinely available in our longitudinal study started several years ago. 2) Local recurrence, not identifiable by other modalities, was reflected in focal tracer uptake adjacent to the resection area of the primary carcinoid. This uptake may have been due to postoperative changes, although it seems unlikely because of the long delay since surgery (> 10 months). Similar findings were documented previously in recurrent medullary thyroid carcinoma (20).

A J G - Januar3' 1 9 9 8

SOMATOSTATIN-RECEPTOR SCINTIGRAPHY

69

o :,P

1

FIG. 2. (A) Anterior planar view of the abdomen (at 24 h) of a patient with glucovipoma demonstrates several focal defects in the liver, compatible with receptor-negative metastases. (B) Anterior planar view of the abdomen (at 3 h) of a patient with a PTHrP-producing tumor demonstrates multiple receptor-positive metastases in the liver.

3) Metastatic spread is more likely to be detected by scintigraphic whole-body screening, obviating unnecessary surgery (10, 21), whereas nonvisualization of metastases on SRS suggests that curative surgical resection of the primary tumor is feasible, as for the patient with vipoma-like syndrome. In our series, SRS was superior to CT in detecting liver and mesenteric metastases of carcinoid origin, in contrast to the results of Kwekkeboom et al. (12), who reported a 50% sensitivity in detecting hepatic metastases. The CT-negative liver metastases of somatostatinoma were not visualized by SRS, probably because of tumor dedifferentiation or production of somatostatin, which competes with the labeled ligand (22). Several investigators observed an improved sensitivity of SRS while patients are taking octreotide (23. 24). Our scintigraphic studies, however, detected all CTdemonstrable liver foci, despite withdrawal of octreotide 72 h before the scan. SPECT also was found to improve lesion detectability (21) and is essential for localization of tumors with low target-to-background ratio and negative planar views, mainly in the head and tail of the pancreas, in fact, two of three insulinomas and one nonsecretory isletcell tumor were delineated only by SPECT, in contrast to the findings of Zimmer et al. (18), who claimed that tumors not detected on planar views are also not identifiable on SPECT. A medium-energy collimator is also recommended for improved delineation of tumors, rather than the 360-KeV collimator used by Scherubl et al. (25). In patients with known metastatic disease, visualization of these foci on SRS may justify chronic octreotide therapy, even when no hormone-related symptoms are present. Octreotide therapy may have a beneficial effect on tumor growth p e r se, although the highly variable growth rate of these tumors hampers accurate assessment of the effect of octreotide on tumor size. No tumor regression was observed by Lamberts et al. ( 1 3 ) and Trautmann et aL (26). However, Kvols (27) and Arnold et al. (28) claimed that previously

progressive disease often became stable on octreotide therapy, leading to improved long-term survival. Among our patients, transient regression of receptor-positive liver metastases was evident in a patient with glucagonoma over a 2-yr period, and no progression of hepatic involvement was observed in a patient with carcinoid during a 7-yr follow-up, while both patients were on octreotide therapy. The latter patient developed carcinoid valvular disease while taking octreotide, which suggests that treatment may have stopped tumor growth but did not effectively block the secretion of vasoactive peptides. The ability of certain tumors to bind labeled octreotide, as demonstrated on scintigraphy, suggests that similar agents designed to deliver a high radiation dose may be beneficial for receptor-positive metastases, as documented recently by Krenning et al. (29) in a patient with glucagonoma. In contrast, nonvisualization of known metastatic spread by SRS suggests that chronic octreotide therapy or a labeled somatostatin analogue may not be of benefit. Rather, nonvisualization may reflect tumor dedifferentiation requiring aggressive chemotherapy, as suggested by Kvols (30). In our series of patients, SRS detected both the primary tumor and the metastases in a high percentage of cases. The technique is complementary to other imaging modalities, as it is effective for early detection of occult primary tumor, staging of disease, and assessment of somatostatin-receptor status. All of these aid in choosing the optimal mode of therapy for a given patient--surgery, octreotide, or chemotherapy. ACKNOWLEDGMENT This study was initially supported by Mallinckrodt Medical. Petten. Holland. Reprint requests and correspondence: Yodphat Krausz, M.D., Department of Nuclear Medicine, Hadassah University Hospital, P.O. Box 12000, Jerusalem 91120. Israel.

70

KRAUSZ

et al. REFERENCES

1 Kloppel Cl, Heitz PU. Classification of normal and neoplastic neuroendocrine cells. Ann NY Acad Sci 1994;733:18-23. 2. Oberg K. Expression of growth factors and their receptors in neuroendocrine gut and pancreatic tumors, and prognostic factors for survival. Ann NY Acad Sci 1994;733:46-55. 3. Lamberts SWJ, Chayvialle J-A, Krenning EP. The visualization of gastroenteropancreatic endocrine tumors. Metabolism 1992;41(suppl 2): 11 l-5. 4. Reubi JC, Kvols L, Krenning EP, et al. In vitro and in vivo detection of somatostatin receptors in human malignant tissues. Acta Oncol .1991;30:463-8. 5. Reubi JC, Krenning EP, Lambens SWJ, et al. In vitro detection of somatostatin receptors in human tumors. Metabolism 1992;4l(suppl 2):104-10. 6. Reubi JC, Laissue J. Waser B, et al. Expression of somatostatin receptors in normal, inflamed, and neoplastic human gastrointestinal tissues. Ann NY Acad Sci 1994;733: 122-37. 7. Krenning EP, Kwekkeboom DJ. Reubi JC, et al. In1 1 I-octreotide scintigraphy in oncology. Metabolism 1992;41(suppl 2):83-6. 8. Krenning EP, Kwekkeboom DJ. Bakker WH, et al. Somatostatin receptor scintigraphy with [In1 1 I-DTPA-D-Phel]and [1123-Tyr3]octreotide: The Rotterdam experience with more than 1000 patients. Eur J Nucl Med 1993;20:716-31. 9. Weiss M, Yellin A, Husza’r M, et al. Localization of adrenocorticotropic hormone-secreting bronchial carcinoid by somatostatin-receptor scintigraphy. Ann Intern Med 1994:121:198-9. scintig10. Krausz Y, Pfeffer MR. Glaser B. et al. Somatostatin-receptor raphy of subcutaneous and thyroid metastases from bronchial carcinoid. J Nucl Med 1996;37: 1537-9. with 11. Joseph K. Stapp J. Reinecke J. et al. Receptor scintigraphy In1 1 I-pentetreotide for gastroenteropancreatic tumors. Horm Metab Res Suppl 1993;27:28-35. 12. Kwekkeboom DJ. Krcnning EP. Bakker WH. et al. Somatostatin analogue scintigraphy in carcinoid tumors. Eur J Nucl Med 1993;20:283-92. 13. Lambens SWJ, Krenning EP. Reubi JC. The role of somatostatin and its analogs in the diagnosis and treatment of tumors. Endocr Rev 1991;12:450-82. 14. Pauwels S. Leners N. Fiasse R. et al. Localization of gastroenteropancreatic tumors with In1 1 I-pentetreotide scintigraphy. Semin Oncol 1994:21(suppl 13): 15-20. receptor 15. de Kerviler E. Cadiot G. Lebtahi R. et al. Somatostatin scintigraphy in forty-eight patients with the Zollinger-Ellison syndrome. Eur J Nucl Med 1994:31:1 191-7.

AJG - Vol. 93, No. 1, 1998 16. Hammond PJ, Arka A, Peters AM, et al. Localization of metastatic gastroenteropancreatic turnouts by somatostatin receptor scintigraphy with [In1 I I-DTPA-D-PhenI]-octreotide. Q J Med 1994;87:83-8. 17. Krenning EP, Kwekkeboom DJ, Oei HY, et al. Somatostatin-receptor scintigraphy in gastroenteropancreatic tumors: An overview of European results. Ann NY Acad Sci 1994;733:416-24. 18. Zimmer T, Ziegler K, Bader M, et al. Localization of neuroendocrine tumors of the upper gastrointestinal tract. Gut 1994;35:471-5. 19. Rothmund M, Angelini L, Brunt M, et al. Surgery for benign insulinoma: An international review. World J Surg 1990,14:393-9. 20. Krausz Y, De Jong R, Glaser B. Differential detection of metastatic foci of medullary thyroid carcinoma. Turk J Nucl Med 1994;3:4 (abstract). 21. Jamar F, Fiasse R, Leners N, et al. Somatostatin receptor imaging with In1 11-pentetreotide in gastroenteropancreatic neuroendocrine tumors: Safety, efficacy and impact on patient management. J Nucl Med 1995;36:542-9. 22. Krenning EP, Kwekkeboom DJ, de Jong M. et al. Essentials of peptide receptor scintigraphy with emphasis on the somatostatin analog octreotide. Semin Oncol 1994;2l(suppl 13):6-14. 23. Dorr U, Wurm K, Horing E, et al. Diagnostic reliability of somatostatin receptor scintigraphy during continuous treatment with different somatostatin analogs. Horm Metab Res Suppl 1993;27:36-43. 24. Soresi E. Bombardieri E, Chiti A, et al. In 111 -DTPA-octreotide scintigraphy modulation by treatment with unlabelled somatostatin analogue in small-cell lung cancer. Tumori 1995;81:125-7. 25. Scherubl H, Bader M, Fett U. et al. Somatostatin-receptor imaging of neuroendocrine gastroenteropancreatic tumors. Gastrocnterology 1993;105:1705-9. 26. Trautmann ME, Neuhaus CH. Lenze H, et al. The role of somatostatin analogs in the treatment of endocrine gastrointestinal tumors. Horm Metab Res Suppl 1993;27:24-7. 27. Kvols LK. Metastatic carcinoid tumors and the malignant carcinoid syndrome. Ann NY Acad Sci 1994;733:464-70. 28. Arnold R, Neuhaus C, Benning R. et al. Somatostatin analog sandostatin and inhibition of tumor growth in patients with metastatic endocrine gastroenteropancreatic tumors. World J Surg 1993;17:51 l-9. 29. Krenning EP, Kooij PPM, Bakker WH, et al. Radiotherapy with a radiolabeled somatostatin analogue, [In1 1 I-DTPA-D-Phenll-octreotide: A case history. Ann NY Acad Sci 1994;733:496-506. 30. Kvols L. Medical oncology considerations in patients with metastatic neuroendocrine carcinomas. Semin Oncol 1994;5(suppl 13): 56-60.