|
| |||||||||||||||||||||||||||||||
Background Intestinal adenocarcinomas and various endocrine tumors express large numbers of high-affinity receptors for vasoactive intestinal peptide (VIP). We have evaluated the usefulness of scanning with VIP labeled with iodine-123 for tumor localization in patients with gastrointestinal tumors.
Methods Radioiodinated VIP was purified by high-pressure liquid chromatography and administered as a single intravenous bolus injection (300 pmol [1 microg]). Scanning with radiolabeled VIP was compared with computed tomography and scanning with somatostatin analogues in 79 patients with colorectal cancer, pancreatic carcinoma, gastric cancer, carcinoid tumor, or insulinoma.
Results Visualization of gastrointestinal tumors and metastases was obtained with radiolabeled VIP. Binding of the labeled peptide by primary tumors and metastases was visible shortly after the injection and was still demonstrable at 24 hours. In patients with colorectal adenocarcinomas, primary or recurrent tumors were visualized in 10 of 10, liver metastases in 15 of 1, lung metastases in 2 of 3, and lymph-node metastases in 4 of 4. Primary pancreatic adenocarcinomas were visualized by imaging in 10 of 12 patients, and liver metastases were seen in 7 of 7. Primary or recurrent gastric adenocarcinomas were visualized in five of five patients, and liver metastases were seen in two of two patients. VIP scans were positive in 9 of 10 patients with carcinoid tumors and in 4 of 4 patients with insulinomas. Some tumors with positive VIP scans were also visualized with somatostatin analogues (4 of 17 colorectal adenocarcinomas, 8 of 9 carcinoids, and 2 of 2 insulinomas). In vitro binding studies confirmed the presence of VIP receptors on gastrointestinal tumors.
Conclusions Scanning with radiolabeled VIP can visualize intestinal tumors and metastases that express receptors for VIP.
During the past few years, radiolabeled peptides have been introduced for the in vivo visualization of solid tumors13. Radiolabeled octreotide, for example, has been used to localize endocrine tumors14,15,16. In this study, in vivo scanning with radioiodinated VIP was performed successfully in patients with gastrointestinal adenocarcinomas, carcinoids, and insulinomas.
Methods
Patients
The 79 patients enrolled in this study gave written informed consent to participate. The study protocol was approved by the ethics committee of the Medical Faculty of the University of Vienna. Thirty-five patients had colorectal adenocarcinomas, 20 had pancreatic adenocarcinomas, 8 had gastric cancers, 12 had carcinoids, and 4 had insulinomas. The diagnoses and stages of the disease were established according to the criteria of the World Health Organization (WHO). The location and size of the primary tumors and metastases were determined by conventional computed tomography (CT), radiography, ultrasonography, colonoscopy, gastroscopy, or surgery.
Preparation of Radioiodinated VIP
Synthetic VIP was labeled with iodine-123 according to a modified Iodogen method17. To obtain a high specific activity, 123I-labeled VIP was purified by preparative high-pressure liquid chromatography (column: reverse-phase C18, 5 microm, 4 x 250 mm; eluent: 74 percent [vol/vol] aqueous triethylammonium formate [0.25 mol per liter, pH 3.0] and 26 percent [vol/vol] acetonitrile at 1 ml per minute). The column eluate passed through a scintillation radioactivity detector and ultraviolet (280-nm) detector in series. After calibration with unlabeled VIP, the system allowed the collection of pure radioiodinated VIP, free of unlabeled VIP, reagents, and inorganic species of iodine. The eluate was evaporated at reduced pressure and dissolved in phosphate-buffered saline containing 0.1 percent (weight/vol) Tween 80 (Koch-Light, Colnbrooke, United Kingdom). When analyzed by analytic high-pressure liquid chromatography, the labeled product contained less than 3 percent of unbound iodide in all preparations and remained stable for at least 24 hours. The biologic activity of labeled VIP was identical to that of unlabeled VIP, as determined by its ability to enhance the formation of cyclic AMP (cAMP) (Table 1), the incorporation of 32P-labeled ATP, and the uptake of 3H-labeled thymidine by tumor cells6.
|
Gamma-Camera Scanning
Planar acquisitions and single-photon-emission CT acquisitions were performed with a large-field-of-view gamma camera (Toshiba, Japan) equipped with a low-energy general-purpose collimator. At the time of the injection of radiolabeled VIP, the field of view covered the abdomen and some of the thorax, unless otherwise stated. Standard techniques of recording and visualization were used. Sequential images were recorded every minute for 30 minutes (matrix, 128 x 128 pixels). Planar images in the anterior, posterior, and lateral views of three regions, covering the brain and neck, thorax, and abdomen, were acquired at 30 minutes, 2 to 4 hours, and 18 to 24 hours after the intravenous injection (matrix, 128 x 128 pixels; approximately 150 to 300 kilocounts were acquired; scanning time, 10 to 20 minutes). For whole-body kinetics, scanning was performed in the anterior and posterior views with a double-headed Toshiba camera (matrix, 256 x 1024 pixels; 20 cm per step). To evaluate the tumors or metastases as visualized during scintigraphy with radiolabeled VIP, the results of scanning were judged independently by three observers using a yes-or-no system.
In 38 of the 79 patients, somatostatin-receptor scanning was performed in addition to VIP-receptor scanning. Somatostatin-receptor scanning used either 123I-labeled Tyr3-octreotide14,15 or 111In-labeled DTPA-d-Phe1-octreotide16,18. The interval between VIP and octreotide scanning ranged from two to eight weeks. Samples of tumors from 10 patients were studied in vitro for their ability to bind to 123I-labeled VIP and 123I-labeled Tyr3-octreotide, as described elsewhere6.
Results
Biodistribution and Safety of 123I-Labeled VIP
Serial images recorded after the intravenous injection of radiolabeled VIP showed that after an initial rapid clearance of the compound from the circulation, the lung was the primary site of VIP binding. Within 30 minutes, approximately 45 percent (range, 32 to 56 percent) of the total radioactivity was found in the lungs. The amount of radioactivity in the lungs decreased thereafter to 25 percent (range, 17 to 38 percent) at 4 hours and to 10 percent (range, 6 to 18 percent) after 24 hours. Virtually no binding to the normal liver, spleen, or intestine was seen during the observation period of 24 hours. Radioactivity was subsequently excreted into the urine and amounted to approximately 35 percent (range, 27 to 58 percent) of the injected dose after 4 hours and to 90 percent (range, 72 to 97 percent) after 24 hours.
Visualization of primary tumors or metastases was obtained within an hour after the injection of radiolabeled VIP and was still demonstrable at 24 hours (Figure 1). The ratio of tumor to intestinal background ranged from 2.4 to 7.1 at 30 minutes, and from 2.1 to 7.9 at 4 hours after the injection.
|
Patients with Colorectal Adenocarcinomas
VIP scanning was performed in 35 patients with colorectal carcinomas (Table 2). A primary tumor or a recurrent local tumor mass (without metastases) was known to be present at the time of the scan in 10 patients; 3 patients had local recurrences with metastases to the liver, the lymph nodes, or both. In 18 patients the primary tumor had been resected and liver, lung, or lymph-node metastases were known. In seven patients, the primary tumor had been resected and at the time of 123I-labeled VIP imaging no metastases were known, as assessed by conventional CT scanning, radiography, ultrasonography, colonoscopy, or surgery. The primary or recurrent tumors ranged from 3 to 16 cm (median, 7.6) in diameter, and the liver metastases ranged from 1 to 13 cm in diameter.
|
|
|
Twenty patients with pancreatic adenocarcinomas underwent VIP scanning (Table 2). The tumor masses were visualized in 10 of 12 patients with primary or recurrent pancreatic adenocarcinomas after the injection of radiolabeled VIP. The tumors ranged from 2 to 10 cm (median, 5.5) in diameter as verified by CT or surgery. In all seven patients with known liver metastases, the hepatic deposits were visualized by VIP-receptor scanning. In five other patients the primary pancreatic tumors had been resected and no metastases were known. The VIP scans were negative in these patients.
Patients with Gastric Cancer
Eight patients with gastric cancer were studied by radiolabeled-VIP scintigraphy (Table 2). Three patients had primary adenocarcinomas, one had a nonlymphoid gastric sarcoma, and two had locally recurrent adenocarcinoma. In three patients liver or lymph-node metastases were known. The primary and recurrent gastric tumors ranged from 1 to 6 cm (median, 4.0) in diameter as verified by CT or gastroscopy.
The tumors were visualized by radiolabeled-VIP scanning in the five patients with primary or recurrent gastric adenocarcinomas (Table 2). Liver and lymph-node metastases were also seen. However, the gastric sarcoma was not visualized with radiolabeled VIP.
Patients with Carcinoid Tumors
Scanning with radiolabeled VIP was performed before the removal of a primary carcinoid tumor in 10 patients (Table 2). In two other patients the primary carcinoid had been resected, and liver metastases were known. In 3 of the first 10 patients, a primary carcinoid tumor had been resected, but the clinical symptoms were highly suggestive of a new carcinoid lesion. Three patients were studied after the removal of the primary carcinoid. In two patients CT scanning before surgery indicated the site of the primary carcinoid lesion. However, in all the other patients the primary site could not be detected by CT. The primary tumors ranged from 1 to 5 cm (median, 1.7) in diameter (Table 2) as assessed after surgery.
Scintigraphy with radiolabeled VIP demonstrated the primary carcinoid tumor in 9 of 10 patients. Liver metastases were detected in three of five patients, and a single lung metastasis was detected in one patient. After resection of the primary VIP-positive carcinoids, postoperative VIP scans were negative in three of three patients. In all the patients the primary site of the carcinoid was verified by surgery.
Patients with Insulinomas
Four patients with insulinomas were studied before surgery (Table 2). The VIP scans showed the site of the primary tumor in all four (Figure 4), whereas CT scans were positive in only two of the patients. The insulinomas ranged from 1 to 4 cm (median, 2.1) in diameter.
|
Somatostatin-receptor scanning with 123I-labeled Tyr3-octreotide or 111In-labeled DTPA-d-Phe1-octreotide has also been used to visualize endocrine tumors18. Therefore, in 38 of 79 patients, VIP-receptor and somatostatin-receptor scanning were compared. There was an interval of two to eight weeks between the scans. In patients with colorectal adenocarcinomas, the somatostatin-receptor scans were positive in two of five patients with primary tumors and in two of seven patients with liver metastases (Table 2). Radiolabeled-VIP scanning was also positive in these patients. Pancreatic adenocarcinomas in five patients (Table 2) were not visualized by radiolabeled octreotide; by contrast, radiolabeled VIP detected the tumors readily. The somatostatin-receptor scans were positive in eight of nine patients with primary carcinoids and in two of two patients with insulinomas (Table 2).
In Vitro Binding of Radiolabeled VIP
In vitro binding of radiolabeled VIP and of 123I-labeled Tyr3-octreotide to tumor tissue was investigated in surgical specimens of colorectal, pancreatic, and carcinoid tumors and in two insulinomas (Table 3). In vitro binding of radiolabeled VIP and of 123I-labeled Tyr3-octreotide was demonstrated in all cases. Each sample had high-affinity and low-affinity binding sites. The tumor cells bound at least 1000 times more radiolabeled VIP than did platelets or peripheral-blood leukocytes (Table 3).
|
We have developed a tumor-scanning technique that uses radioiodinated VIP. We found that VIP scintigraphy may be useful for localizing intestinal carcinomas and metastases that express receptors for VIP. In vitro studies have shown that normal intestinal epithelial cells and various tumor cells, including intestinal carcinomas and adenocarcinoma cell lines, express VIP receptors6,7,8,9,10,11,12. These observations prompted us to develop receptor scintigraphy using 123I-labeled VIP as a radioligand. After the intravenous injection of this peptide, primary and metastatic intestinal carcinomas were readily visualized. The high quality of the scans was probably due to the high specificity of binding of VIP to the tumor cell membranes6. Specificity for tumor tissue is most likely due to the very high numbers of VIP-binding sites expressed on these tumors as compared with normal tissues and blood cells (Table 3)6.
The cases in which the VIP scan was negative may be explained by the low numbers or absence of VIP receptors or by the blockade of VIP receptors by locally produced ligands. The possibility that these negative results were due to small tumors or to the lack of sensitivity of the VIP scan seems unlikely, since the VIP scan detected endocrine tumors less than 2 cm in diameter.
In our study, VIP-receptor scanning had an advantage over CT scanning in patients with small carcinoid tumors. CT scanning indicated the site of a primary carcinoid tumor in only 2 of 10 patients (these tumors were 2 and 5 cm in diameter), whereas VIP-receptor scanning was positive in 9 of these patients.
The long-acting somatostatin analogue octreotide has been successfully used to localize endocrine and other tumors14,15,16,18. In our study we were able to visualize carcinoid tumors and insulinomas with either radioactive octreotide or VIP, suggesting that some intestinal tumors may express both somatostatin and VIP receptors6. However, the octreotide scan was positive in only 4 of 17 patients with VIP scan-positive colonic adenocarcinomas and negative in 5 patients with VIP scan-positive pancreatic adenocarcinomas. VIP scans may thus have an advantage over octreotide scans in the visualization of intestinal adenocarcinomas. Another advantage of the VIP scan may be its use in patients whose tumors have been treated with octreotide. Furthermore, in some patients liver metastases were visualized better by scanning with radiolabeled VIP than by somatostatin-receptor scanning. In this respect, it is noteworthy that VIP is cleared through the urinary tract, which may add to its usefulness in localizing intestinal tumors.
VIP is a neuroendocrine mediator that causes the Verner-Morrison syndrome of watery diarrhea in patients with VIP-secreting tumors2,3,19. Therefore, we took special precautions in performing the scintigraphy and used radiolabeled VIP that was purified by high-pressure liquid chromatography to very high specific activity. No clinical side effects during or after the injection of VIP were noted, except for a mild and transient drop in blood pressure. The absence of reactions is probably due to the small amount of VIP injected into our patients.
Supported by the Commission for Oncology and the Anniversary Foundation of the University of Vienna, the Foundation of the Mayor of the City of Vienna, the Austrian National Bank, and the Fonds zur Forderung der wissenschaftlichen Forschung in Osterreich.
We are indebted to the members of the Vienna University Hospital nursing staff in the Departments of Gastroenterology (Trude Rothmayr and team) and Oncology (Brigitte Maca and team) for their excellent cooperation during the performance of the study.
Source Information
From the Departments of Nuclear Medicine (I.V., A.K., S.B., Q.Y., S.L., M.B.), Clinical Pharmacology (I.V.), and Surgery (B.N.); the Divisions of Oncology (M.R., W.S.) and Hematology (P.V.); Internal Medicine I; and the Division of Gastroenterology, Internal Medicine IV (J.P.), the University of Vienna, Vienna; and the Department of Radiochemistry, Research Center, Seibersdorf (P.A.) -- both in Austria.
Address reprint requests to Dr. Virgolini at the Department of Nuclear Medicine, University of Vienna, Wahringer Gurtel 18-20, Ebene 3L, A-1090 Vienna, Austria.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |