- Email: [email protected]
EUS Detection of Pancreatic Endocrine Tumors in Asymptomatic Patients with Type 1 Multiple Endocrine Neoplasia
EUS detection of pancreatic endocrine tumors in asymptomatic patients with type 1 multiple endocrine neoplasia Erik-Jan Wamsteker, MD, Paul G. Gauger, MD, Norman W. Thompson, MD, James M. Scheiman, MD Ann Arbor, Michigan
Background: EUS is highly sensitive, specific, and cost-effective for localization of pancreatic neuroendocrine tumors. EUS screening of asymptomatic patients with multiple endocrine neoplasia type 1 has not been described. Methods: EUS was used to evaluate all patients with known or suspected pancreatic neuroendocrine tumors. Asymptomatic patients with either a confirmed genetic or clinical diagnosis of multiple endocrine neoplasia type 1 were evaluated with EUS. The results were correlated with surgical and histopathologic findings. Results: A total of 65 patients with multiple endocrine neoplasia type 1 underwent 132 EUS procedures over an 8-year period, including 13 asymptomatic patients. Five of the 13 asymptomatic patients had normal serum gastrin levels, and 8 had levels less than 4 times the upper limit of normal. EUS demonstrated pancreatic neuroendocrine tumors in 11 of these patients, ranging in size from 0.4 to 4 cm (mean 10 mm). Ten of the 11 patients with tumors at EUS underwent surgical exploration, and 23 of 28 (82%) tumors removed surgically had been identified by EUS. At a mean follow-up of 44 months, no liver or lymph node metastases were demonstrated in any patient. Conclusions: In this first cohort study of asymptomatic patients with multiple endocrine neoplasia type 1 undergoing screening EUS, pancreatic neuroendocrine tumors were identified before the development of significant biochemical test abnormalities. Aggressive early surgical treatment may improve the prognosis for these patients. (Gastrointest Endosc 2003;58:531-5.)
Multiple endocrine neoplasia type 1 (MEN 1), a condition with autosomal dominant inheritance, consists of benign and malignant tumors of the pituitary, the parathyroids, and the enteropancreatic neuroendocrine system, as well as other tissue types.1 Patients with MEN 1 are at high risk for the development of pancreatic endocrine tumors (PET). Of patients with MEN 1, 40% to 90% of them develop pancreaticoduodenal endocrine tumors. Approximately 50% of patients with MEN 1 will develop gastrinomas by 50 years of age. The gastrinomas are most often localized to the duodenal wall; gastrinoma is the most frequent functioning tumor. From 10% to 35% of patients with MEN 1 may Received November 19, 2002. For revision March 28, 2003. Accepted June 26, 2003. Current affiliations: Department of Internal Medicine, Division of Gastroenterology, and Department of Surgery, Division of Endocrine Surgery, The University of Michigan School of Medicine, Ann Arbor, Michigan. Abstract was presented at Digestive Diseases Week, May 19-23, 2002, San Francisco, California (Gastroenterology 2002;122:AB W1102, A-566). Reprint requests: James M. Scheiman, MD, 3912 Taubman Center, Box 0362, University of Michigan, Ann Arbor, MI 48109-0362. Copyright © 2003 by the American Society for Gastrointestinal Endoscopy 0016-5107/2003/$30.00 + 0 PII:S0016-5107(03)01965-5 VOLUME 58, NO. 4, 2003
develop insulinomas. Non-functioning PETs also are common, but the true incidence is unknown. Prognosis and impact on quality of life for patients with MEN 1 relates to direct tumor expansion, hormone excess syndromes, and the metastatic potential of these tumors.2 Thus, biochemical and radiologic localization screening tests are used for patients with a genetic or family history of MEN 1. Before the discovery of the gene mutation, all relatives of patients with MEN 1 underwent annual biochemical tests and imaging procedures for pituitary and gastroenteropancreatic neuroendocrine tumors throughout their lifetime.3 Genetic testing of families with MEN 1 now can be performed, which identifies at-risk members while excluding non-carriers.4 This has the potential to reduce health care costs by decreasing the numbers of patients who require expensive hormonal biochemical and imaging tests. Once a carrier of the MEN1 gene is identified, evaluation can be intensified. Biochemical/hormonal screening typically includes a serum calcium and levels of parathyroid hormone, gastrin, fasting glucose and insulin, chromogranin A, glucagon and proinsulin, prolactin, insulin-like growth factor, and pancreatic polypeptide, all of which are obtained annually.5 There is no consensus as to the most appropriate imaging modalities or the timing of GASTROINTESTINAL ENDOSCOPY
531
E-J Wamsteker, P Gauger, N Thompson, et al.
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
Table 1. Patient demographic data Patient
Age (y)
Gender
Gastrin* (pg/mL)
Associated diseases
1 2 3 4 5 6 7 8 9 10 11 12 13
35 38 52 15 30 24 30 48 34 20 18 55 54
F M M F F M M F M F M F F
42 222 160 169 208 318 127 130 95 68 54 100 95
HPTH HPTH HPTH HPTH HPTH HPTH HPTH HPTH, PA HPTH HPTH, PA HPTH, PA HPTH HPTH, PA
HPTH, Hyperparathyroidism; PA, pituitary adenoma. *Gastrin: nmL < 100 pg/mL.
EUS imaging Figure 1. Radial scanning EUS imaging showing typical appearance of neuroendocrine tumors (NET) in head of pancreas.
these procedures as a screening test. This may be because of the limited sensitivity of CT and MR as well as OctreoScan imaging for small tumors, and/or a lack of expertise for certain procedures such as EUS. EUS has been relegated, in some centers, to a role as a secondary imaging test performed only to clarify the findings of alternative imaging modalities.6 However, our group demonstrated EUS to be the most sensitive, specific, accurate, as well as costeffective imaging procedure for preoperative localization of PETs.7 Because PETs that cause symptoms may already be locally or regionally metastatic, early identification may have the potential to decrease morbidity and mortality.8 This report comprises the findings of the first cohort study of presymptomatic patients with MEN 1 evaluated with EUS. PATIENTS AND METHODS A total of 65 patients with known or suspected MEN 1 underwent 132 EUS procedures over an 8-year period (1994-2002). This study was approved by the institutional review board of our medical school. EUS is the initial imaging modality in our diagnostic and therapeutic algorithm.8 Patients with a genetic diagnosis or family history of MEN 1 and/or involvement of more than one endocrine organ meeting clinical criteria for MEN 1 (at least two of the following: hyperparathyroidism, pituitary adenoma, or PETs) were evaluated. Data regarding clinical symptoms, results of imaging procedures, previous abdominal surgery, as well as biochemical tests at referral were collected in a standardized manner. To be included in the study, patients had to be asymptomatic without evidence of PET on other imaging procedures. 532
GASTROINTESTINAL ENDOSCOPY
Patients were placed in the left lateral position and sedated by intravenous administration of benzodiazepine and opioid drugs. Standard EGD was performed before EUS to evaluate the duodenum for submucosal tumors. A radial scanning echoendoscope (GF-UM130; Olympus America Corp., Melville, N.Y.) was advanced to the descending duodenum and then slowly withdrawn to the stomach. Regional anatomic structures were identified to ensure that all regions of the pancreas were visualized. Characteristic EUS features of PETs have been described.9 If visualized, tumor size, multiplicity, echotexture, location within the pancreas, and presence of regional lymphadenopathy were documented, and, when applicable, these findings were compared with surgical pathology. In all cases, the pancreas was thought to have been examined completely by EUS.
RESULTS Thirteen patients (6 men, 7 women; mean age 33 years) met inclusion criteria (Table 1). Mean follow-up for the patient cohort was 44 months (21-87 months). The gastrin level was normal in 5 patients and mildly elevated (<4 times the upper normal limit) in 8 patients. All patients had associated hyperparathyroidism; pituitary adenomas had been demonstrated in 4. Magnetic resonance imaging or CT had not been performed in any patient before the initial EUS. Three patients underwent an OctreoScan before EUS, one of which was positive and corresponded to a duodenal gastrinoma. EGD revealed no evidence for a subepithelial tumor in this patient cohort. EUS demonstrated PETs that ranged in size from 0.4 to 4 cm (mean 1.1 cm) in 11 of the 13 patients (Fig. 1). Multiple tumors were seen in 9 patients. The mean number of PETs identified per patient was 3.3 (1-5) (Table 2). The EUS findings were confirmed in the 10 patients who underwent surgical exploration. Two (patients 4,10) with tumors have VOLUME 58, NO. 4, 2003
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
E-J Wamsteker, P Gauger, N Thompson, et al.
Table 2. EUS findings Patient
No. of tumors
Location of tumors
Tumor size range (cm)
Sonographic features of NET
1 2 3 4 5 6 7 8 9 10 11 12 13
5 Abnormal panc 2 5 3 5 3 2 1 3 3 1 Normal panc
Neck, body, tail NA Periampullary, body, tail Head, uncinate, body, tail Uncinate, neck, tail Head, neck, body, tail Head, uncinate, body, tail Body, tail Head Uncinate, body Neck, body, tail Neck NA
0.9 each NA 0.5-2.0 0.5-0.7 0.4-1.0 0.4-0.9 1.1-1.6 0.8-2.3 4 0.5-0.8 0.5-1.4 0.6 NA
Hypoechoic Heterogeneous Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Iso/hypoechoic Hypoechoic Normal panc
NET, Neuroendocrine tumor; panc, pancreas; NA, not applicable.
Table 3. Surgical and pathologic findings Patient
No. of PETs by surgery
1 2 3 5 6 7 8 9 11
5 3 2 3 5 3 2 1 4
12
2
Histology with stain NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–); (+) for serotonin* NET/gastrin (–)
Duodenal primary with stain NA (+) for (+) for (+) for (+) for NA NA (+) for NA NA
gastrin† gastrin† gastrin† gastrin†
gastrin†
Post-surgical follow-up (mo) 59 37 62 54 34 44 47 21 37 87
PET, Pancreatic neuroendocrine tumor; NET, neuroendocrine tumor; NA, not applicable. *Tumor type not known. †Consistent with gastrinoma.
not had surgical exploration because of their young age (15 and 20 years) and the small size of the tumors (5-8 mm), and are currently undergoing surveillance EUS. A third patient (patient 13) with a normal EUS was lost to follow-up. Twenty-three of 28 (82%) PETs removed at surgery were identified by preoperative EUS. The mean size of these tumors was 10 mm (Table 3). Multiple tumors were found in 9 of the 10 patients who underwent surgery. Five of the 13 patients also were noted to have duodenal gastrinomas, none of which were detected by EUS or EGD. One patient in whom PETs were not identified by EUS had a markedly heterogeneous pancreas, and, at surgery, 3 PETs were discovered, the largest being 1.1 cm. The two remaining tumors not seen by EUS were 4 to 5 mm in size and were located in the uncinate process of the pancreas. EUS identified a 5-mm mass, which was not palpable at surgery, in the head of the pancreas in one patient. This abnormality was seen on several subsequent EUS procedures and is felt to be VOLUME 58, NO. 4, 2003
a lymph node. Mean postoperative follow-up was 48 months. Only one patient had lymph node metastases identified at initial surgical exploration. This patient (patient 3) was discovered to have MEN 1 at age 52 years after his son was found to have the clinical syndrome consistent with MEN 1. All tumors were routinely stained for gastrin alone and were otherwise described histopathologically as neuroendocrine tumors (NET); one also was stained for serotonin (Table 3). The histopathologic features for all pancreatic and duodenal tumors were consistent with NETs. The only gastrinpositive tumors identified were duodenal in location. Four of the 8 patients with elevated preoperative serum gastrin levels had duodenal gastrinomas. DISCUSSION Although patients with MEN 1 syndrome and PETs may die as a consequence of the secretion of hormones and associated clinical syndromes, these tumors can metastasize and, thereby, have the potenGASTROINTESTINAL ENDOSCOPY
533
E-J Wamsteker, P Gauger, N Thompson, et al.
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
Figure 2. Algorithm for management of asymptomatic patients with MEN 1.
tial to shorten lifespan. Advances in genetic testing have allowed identification of the MEN1 gene before the development of clinical and other manifestations of the syndrome. Biochemical abnormalities suggestive of neoplasia also can be detected up to 2 decades before the disease becomes clinically overt.10 As mortality in these patients most frequently is related to complications from progression of the disease, early identification of the tumors may improve prognosis. Approximately 20% of PETs in patients with MEN 1 are “non-functional,” and some may, therefore, escape detection when biochemical screening alone is used, although measurement of pancreatic polypeptide level has decreased this number.11 Our group, as well as other investigators, has demonstrated the superiority of EUS for detection of PETs compared with cross-sectional imaging modalities and more invasive procedures such as angiography with and without hormone sampling. EUS, as a safe and relatively noninvasive technique that can detect even tiny tumors, is ideal for screening asymptomatic patients. For patients with symptoms from MEN 1 disease, EUS is critical for localization of PETs in the head of the pancreas, given the aggressive surgical approach that routinely includes distal pancreatectomy and enucleation of pancreatic head neuroendocrine tumors.12 EUS-guided FNA is not used because it is believed that a characteristic appearance in a patient with a high clinical likelihood for tumor development is diagnostic. The primary role of EUS is preoperative localization of these tumors, which guides surgical planning.13 At present, the role of preoperative EUS is localization of PETs in the head of the pancreas, which can change the operative approach and decrease operative time. The major limitation of EUS is the difficulty encountered in localizing lesions in the presence of 534
GASTROINTESTINAL ENDOSCOPY
another pancreatic abnormality, mainly chronic pancreatitis.14 One patient in the present study had a markedly heterogeneous pancreas on EUS, likely because of chronic pancreatitis as a result of excessive ingestion of alcohol. This patient (patient 2) underwent surgery because of the subsequent development of severe dyspepsia and a mild increase in gastrin levels. A second limitation of EUS, as well as EGD, is poor accuracy for localization of duodenal tumors. This is not problematic because duodenotomy is routine at surgical exploration. EUS is part of a multidisciplinary approach to the diagnosis and management of patients with PETs, and, although our experience is limited, the addition of the OctreoScan may lead to the identification of some duodenal gastrinomas; but further study regarding this possibility is required. Early screening and surveillance strategies, by using EUS as well as early surgical intervention, may decrease morbidity and mortality because of PETs, a hypothesis supported by the absence of regional and distant metastases in the present study. Only one patient was found to have locoregional lymph node metastases at surgical exploration. The age of this patient was approximately 20 years greater than the mean age for the cohort, and the patient was discovered to have MEN 1 only after his son developed the clinical syndrome. However, further follow-up of the cohort is necessary to determine the impact of early screening and surveillance on overall survival. The present study demonstrates that EUS in experienced hands can identify PETs in asymptomatic patients with MEN 1, despite normal or equivocal results of biochemical testing. The algorithm currently used to evaluate asymptomatic patients with MEN 1 is shown in Figure 2. In addition to EUS screening, patients should undergo annual biochemical screening and CT of the abdomen and the pelvis every 3 years, examining for extra-intestinal complications of MEN 1. If the initial EUS is normal, our recommendation is that screening EUS be repeated in 3 years. If a small PET is identified (<1 cm), then consideration can be given to further surveillance, including annual biochemical tests and EUS. However, this approach raises questions that are not addressed by the current study: what is the appropriate interval for screening EUS after an initial normal EUS, as well as what should be the timing for surveillance EUS and surgical treatment for patients with small PETs. These are questions that must be addressed in future studies. REFERENCES 1. Marx S, Spiegel AM, Skarulis MC, Doppman JL, Collins FS, Liotta LA. NIH conference-multiple endocrine neoplasia: clinical and genetic topics. Ann Intern Med 1998;129:484-94. 2. Doherty GM, Olson JA, Frisella MM, Lairmore TC, Wells SA VOLUME 58, NO. 4, 2003
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
3.
4.
5.
6.
7.
8.
Jr, Norton JA. Lethality of multiple endocrine neoplasia type I. World J Surg 1998;22:581-7. Karges W, Schaaf L, Cralle H, Boehm BO. Concepts for screening and diagnostic follow-up in multiple endocrine neoplasia type 1 (MEN 1). Exp Clin Endocrinol Diabetes 2000;108:334-40. Schussheim DH, Skarulis MC, Agarwal SK, Simonds WF, Burns AL, Spiegel AM, Marx SJ. Multiple endocrine neoplasia type 1: new clinical and basic findings. Trends Endocrinol Metab 2001;12:173-8. Granberg D, Stridsberg M, Seensalu R, Eriksson B, Lundqvist G, Oberg K, Skogseid B. Plasma chromogranin A in patients with multiple endocrine neoplasia type 1. J Clin Endocrinol Metab 1999;84:2712-7. Kopp I, Bartsch D, Wild A, Schilling T, Nies C, Bergenfelz A, et al. Genetic screening in MEN-1 families. World J Surg 2001;25:610-6. Bansal R, Tierney W, Carpenter S, Thompson N, Scheiman J. Cost effectiveness of EUS for preoperative localization of pancreatic endocrine tumors. Gastrointest Endosc 1999;49:19-25. Skogseid B, Oberg K, Eriksson B, Juhlin C, Granberg D, Akerstrom G, Rastad J. Surgery for asymptomatic pancreatic lesion in multiple endocrine neoplasia type 1. World J Surg 1996;20:872-7.
VOLUME 58, NO. 4, 2003
E-J Wamsteker, P Gauger, N Thompson, et al.
9. Anderson MA, Carpenter S, Thompson NW, Nostrant T, Elta GH, Scheiman JM. Endoscopic ultrasound is highly accurate and directs management in patients with neuroendocrine tumors of the pancreas. Am J Gastroenterol 2000;95:2271-7. 10. Oberg K, Skogseid B. The ultimate biochemical diagnosis of endocrine pancreatic tumours in MEN-1. J Intern Med 1998; 243:471-6. 11. Brandi ML, Gagel RF, Angeli A, Bilezikian JP, Beck-Peccoz P, Bordi C, et al. Consensus statement: guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001;86:5658-71. 12. Thompson N. Management of pancreatic endocrine tumors in patients with multiple endocrine neoplasia type I. Surg Oncol Clin N Am 1998;4:881-91. 13. Gauger PG, Scheiman JM, Wamsteker EJ, Richards ML, Doherty GM, Thompson NW. Role of Endoscopic ultrasonography in screening and treatment of pancreatic endocrine tumours in asymptomatic patients with multiple endocrine neoplasia type I. Br J Surg 2003;90:748-54. 14. Delvalle J, Scheiman J. Zollinger-Ellison Syndrome. In: Yamada T, editor. Textbook of gastroenterology. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 1999. p. 144562.
GASTROINTESTINAL ENDOSCOPY
535
Background: EUS is highly sensitive, specific, and cost-effective for localization of pancreatic neuroendocrine tumors. EUS screening of asymptomatic patients with multiple endocrine neoplasia type 1 has not been described. Methods: EUS was used to evaluate all patients with known or suspected pancreatic neuroendocrine tumors. Asymptomatic patients with either a confirmed genetic or clinical diagnosis of multiple endocrine neoplasia type 1 were evaluated with EUS. The results were correlated with surgical and histopathologic findings. Results: A total of 65 patients with multiple endocrine neoplasia type 1 underwent 132 EUS procedures over an 8-year period, including 13 asymptomatic patients. Five of the 13 asymptomatic patients had normal serum gastrin levels, and 8 had levels less than 4 times the upper limit of normal. EUS demonstrated pancreatic neuroendocrine tumors in 11 of these patients, ranging in size from 0.4 to 4 cm (mean 10 mm). Ten of the 11 patients with tumors at EUS underwent surgical exploration, and 23 of 28 (82%) tumors removed surgically had been identified by EUS. At a mean follow-up of 44 months, no liver or lymph node metastases were demonstrated in any patient. Conclusions: In this first cohort study of asymptomatic patients with multiple endocrine neoplasia type 1 undergoing screening EUS, pancreatic neuroendocrine tumors were identified before the development of significant biochemical test abnormalities. Aggressive early surgical treatment may improve the prognosis for these patients. (Gastrointest Endosc 2003;58:531-5.)
Multiple endocrine neoplasia type 1 (MEN 1), a condition with autosomal dominant inheritance, consists of benign and malignant tumors of the pituitary, the parathyroids, and the enteropancreatic neuroendocrine system, as well as other tissue types.1 Patients with MEN 1 are at high risk for the development of pancreatic endocrine tumors (PET). Of patients with MEN 1, 40% to 90% of them develop pancreaticoduodenal endocrine tumors. Approximately 50% of patients with MEN 1 will develop gastrinomas by 50 years of age. The gastrinomas are most often localized to the duodenal wall; gastrinoma is the most frequent functioning tumor. From 10% to 35% of patients with MEN 1 may Received November 19, 2002. For revision March 28, 2003. Accepted June 26, 2003. Current affiliations: Department of Internal Medicine, Division of Gastroenterology, and Department of Surgery, Division of Endocrine Surgery, The University of Michigan School of Medicine, Ann Arbor, Michigan. Abstract was presented at Digestive Diseases Week, May 19-23, 2002, San Francisco, California (Gastroenterology 2002;122:AB W1102, A-566). Reprint requests: James M. Scheiman, MD, 3912 Taubman Center, Box 0362, University of Michigan, Ann Arbor, MI 48109-0362. Copyright © 2003 by the American Society for Gastrointestinal Endoscopy 0016-5107/2003/$30.00 + 0 PII:S0016-5107(03)01965-5 VOLUME 58, NO. 4, 2003
develop insulinomas. Non-functioning PETs also are common, but the true incidence is unknown. Prognosis and impact on quality of life for patients with MEN 1 relates to direct tumor expansion, hormone excess syndromes, and the metastatic potential of these tumors.2 Thus, biochemical and radiologic localization screening tests are used for patients with a genetic or family history of MEN 1. Before the discovery of the gene mutation, all relatives of patients with MEN 1 underwent annual biochemical tests and imaging procedures for pituitary and gastroenteropancreatic neuroendocrine tumors throughout their lifetime.3 Genetic testing of families with MEN 1 now can be performed, which identifies at-risk members while excluding non-carriers.4 This has the potential to reduce health care costs by decreasing the numbers of patients who require expensive hormonal biochemical and imaging tests. Once a carrier of the MEN1 gene is identified, evaluation can be intensified. Biochemical/hormonal screening typically includes a serum calcium and levels of parathyroid hormone, gastrin, fasting glucose and insulin, chromogranin A, glucagon and proinsulin, prolactin, insulin-like growth factor, and pancreatic polypeptide, all of which are obtained annually.5 There is no consensus as to the most appropriate imaging modalities or the timing of GASTROINTESTINAL ENDOSCOPY
531
E-J Wamsteker, P Gauger, N Thompson, et al.
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
Table 1. Patient demographic data Patient
Age (y)
Gender
Gastrin* (pg/mL)
Associated diseases
1 2 3 4 5 6 7 8 9 10 11 12 13
35 38 52 15 30 24 30 48 34 20 18 55 54
F M M F F M M F M F M F F
42 222 160 169 208 318 127 130 95 68 54 100 95
HPTH HPTH HPTH HPTH HPTH HPTH HPTH HPTH, PA HPTH HPTH, PA HPTH, PA HPTH HPTH, PA
HPTH, Hyperparathyroidism; PA, pituitary adenoma. *Gastrin: nmL < 100 pg/mL.
EUS imaging Figure 1. Radial scanning EUS imaging showing typical appearance of neuroendocrine tumors (NET) in head of pancreas.
these procedures as a screening test. This may be because of the limited sensitivity of CT and MR as well as OctreoScan imaging for small tumors, and/or a lack of expertise for certain procedures such as EUS. EUS has been relegated, in some centers, to a role as a secondary imaging test performed only to clarify the findings of alternative imaging modalities.6 However, our group demonstrated EUS to be the most sensitive, specific, accurate, as well as costeffective imaging procedure for preoperative localization of PETs.7 Because PETs that cause symptoms may already be locally or regionally metastatic, early identification may have the potential to decrease morbidity and mortality.8 This report comprises the findings of the first cohort study of presymptomatic patients with MEN 1 evaluated with EUS. PATIENTS AND METHODS A total of 65 patients with known or suspected MEN 1 underwent 132 EUS procedures over an 8-year period (1994-2002). This study was approved by the institutional review board of our medical school. EUS is the initial imaging modality in our diagnostic and therapeutic algorithm.8 Patients with a genetic diagnosis or family history of MEN 1 and/or involvement of more than one endocrine organ meeting clinical criteria for MEN 1 (at least two of the following: hyperparathyroidism, pituitary adenoma, or PETs) were evaluated. Data regarding clinical symptoms, results of imaging procedures, previous abdominal surgery, as well as biochemical tests at referral were collected in a standardized manner. To be included in the study, patients had to be asymptomatic without evidence of PET on other imaging procedures. 532
GASTROINTESTINAL ENDOSCOPY
Patients were placed in the left lateral position and sedated by intravenous administration of benzodiazepine and opioid drugs. Standard EGD was performed before EUS to evaluate the duodenum for submucosal tumors. A radial scanning echoendoscope (GF-UM130; Olympus America Corp., Melville, N.Y.) was advanced to the descending duodenum and then slowly withdrawn to the stomach. Regional anatomic structures were identified to ensure that all regions of the pancreas were visualized. Characteristic EUS features of PETs have been described.9 If visualized, tumor size, multiplicity, echotexture, location within the pancreas, and presence of regional lymphadenopathy were documented, and, when applicable, these findings were compared with surgical pathology. In all cases, the pancreas was thought to have been examined completely by EUS.
RESULTS Thirteen patients (6 men, 7 women; mean age 33 years) met inclusion criteria (Table 1). Mean follow-up for the patient cohort was 44 months (21-87 months). The gastrin level was normal in 5 patients and mildly elevated (<4 times the upper normal limit) in 8 patients. All patients had associated hyperparathyroidism; pituitary adenomas had been demonstrated in 4. Magnetic resonance imaging or CT had not been performed in any patient before the initial EUS. Three patients underwent an OctreoScan before EUS, one of which was positive and corresponded to a duodenal gastrinoma. EGD revealed no evidence for a subepithelial tumor in this patient cohort. EUS demonstrated PETs that ranged in size from 0.4 to 4 cm (mean 1.1 cm) in 11 of the 13 patients (Fig. 1). Multiple tumors were seen in 9 patients. The mean number of PETs identified per patient was 3.3 (1-5) (Table 2). The EUS findings were confirmed in the 10 patients who underwent surgical exploration. Two (patients 4,10) with tumors have VOLUME 58, NO. 4, 2003
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
E-J Wamsteker, P Gauger, N Thompson, et al.
Table 2. EUS findings Patient
No. of tumors
Location of tumors
Tumor size range (cm)
Sonographic features of NET
1 2 3 4 5 6 7 8 9 10 11 12 13
5 Abnormal panc 2 5 3 5 3 2 1 3 3 1 Normal panc
Neck, body, tail NA Periampullary, body, tail Head, uncinate, body, tail Uncinate, neck, tail Head, neck, body, tail Head, uncinate, body, tail Body, tail Head Uncinate, body Neck, body, tail Neck NA
0.9 each NA 0.5-2.0 0.5-0.7 0.4-1.0 0.4-0.9 1.1-1.6 0.8-2.3 4 0.5-0.8 0.5-1.4 0.6 NA
Hypoechoic Heterogeneous Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Hypoechoic Iso/hypoechoic Hypoechoic Normal panc
NET, Neuroendocrine tumor; panc, pancreas; NA, not applicable.
Table 3. Surgical and pathologic findings Patient
No. of PETs by surgery
1 2 3 5 6 7 8 9 11
5 3 2 3 5 3 2 1 4
12
2
Histology with stain NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–) NET/gastrin (–); (+) for serotonin* NET/gastrin (–)
Duodenal primary with stain NA (+) for (+) for (+) for (+) for NA NA (+) for NA NA
gastrin† gastrin† gastrin† gastrin†
gastrin†
Post-surgical follow-up (mo) 59 37 62 54 34 44 47 21 37 87
PET, Pancreatic neuroendocrine tumor; NET, neuroendocrine tumor; NA, not applicable. *Tumor type not known. †Consistent with gastrinoma.
not had surgical exploration because of their young age (15 and 20 years) and the small size of the tumors (5-8 mm), and are currently undergoing surveillance EUS. A third patient (patient 13) with a normal EUS was lost to follow-up. Twenty-three of 28 (82%) PETs removed at surgery were identified by preoperative EUS. The mean size of these tumors was 10 mm (Table 3). Multiple tumors were found in 9 of the 10 patients who underwent surgery. Five of the 13 patients also were noted to have duodenal gastrinomas, none of which were detected by EUS or EGD. One patient in whom PETs were not identified by EUS had a markedly heterogeneous pancreas, and, at surgery, 3 PETs were discovered, the largest being 1.1 cm. The two remaining tumors not seen by EUS were 4 to 5 mm in size and were located in the uncinate process of the pancreas. EUS identified a 5-mm mass, which was not palpable at surgery, in the head of the pancreas in one patient. This abnormality was seen on several subsequent EUS procedures and is felt to be VOLUME 58, NO. 4, 2003
a lymph node. Mean postoperative follow-up was 48 months. Only one patient had lymph node metastases identified at initial surgical exploration. This patient (patient 3) was discovered to have MEN 1 at age 52 years after his son was found to have the clinical syndrome consistent with MEN 1. All tumors were routinely stained for gastrin alone and were otherwise described histopathologically as neuroendocrine tumors (NET); one also was stained for serotonin (Table 3). The histopathologic features for all pancreatic and duodenal tumors were consistent with NETs. The only gastrinpositive tumors identified were duodenal in location. Four of the 8 patients with elevated preoperative serum gastrin levels had duodenal gastrinomas. DISCUSSION Although patients with MEN 1 syndrome and PETs may die as a consequence of the secretion of hormones and associated clinical syndromes, these tumors can metastasize and, thereby, have the potenGASTROINTESTINAL ENDOSCOPY
533
E-J Wamsteker, P Gauger, N Thompson, et al.
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
Figure 2. Algorithm for management of asymptomatic patients with MEN 1.
tial to shorten lifespan. Advances in genetic testing have allowed identification of the MEN1 gene before the development of clinical and other manifestations of the syndrome. Biochemical abnormalities suggestive of neoplasia also can be detected up to 2 decades before the disease becomes clinically overt.10 As mortality in these patients most frequently is related to complications from progression of the disease, early identification of the tumors may improve prognosis. Approximately 20% of PETs in patients with MEN 1 are “non-functional,” and some may, therefore, escape detection when biochemical screening alone is used, although measurement of pancreatic polypeptide level has decreased this number.11 Our group, as well as other investigators, has demonstrated the superiority of EUS for detection of PETs compared with cross-sectional imaging modalities and more invasive procedures such as angiography with and without hormone sampling. EUS, as a safe and relatively noninvasive technique that can detect even tiny tumors, is ideal for screening asymptomatic patients. For patients with symptoms from MEN 1 disease, EUS is critical for localization of PETs in the head of the pancreas, given the aggressive surgical approach that routinely includes distal pancreatectomy and enucleation of pancreatic head neuroendocrine tumors.12 EUS-guided FNA is not used because it is believed that a characteristic appearance in a patient with a high clinical likelihood for tumor development is diagnostic. The primary role of EUS is preoperative localization of these tumors, which guides surgical planning.13 At present, the role of preoperative EUS is localization of PETs in the head of the pancreas, which can change the operative approach and decrease operative time. The major limitation of EUS is the difficulty encountered in localizing lesions in the presence of 534
GASTROINTESTINAL ENDOSCOPY
another pancreatic abnormality, mainly chronic pancreatitis.14 One patient in the present study had a markedly heterogeneous pancreas on EUS, likely because of chronic pancreatitis as a result of excessive ingestion of alcohol. This patient (patient 2) underwent surgery because of the subsequent development of severe dyspepsia and a mild increase in gastrin levels. A second limitation of EUS, as well as EGD, is poor accuracy for localization of duodenal tumors. This is not problematic because duodenotomy is routine at surgical exploration. EUS is part of a multidisciplinary approach to the diagnosis and management of patients with PETs, and, although our experience is limited, the addition of the OctreoScan may lead to the identification of some duodenal gastrinomas; but further study regarding this possibility is required. Early screening and surveillance strategies, by using EUS as well as early surgical intervention, may decrease morbidity and mortality because of PETs, a hypothesis supported by the absence of regional and distant metastases in the present study. Only one patient was found to have locoregional lymph node metastases at surgical exploration. The age of this patient was approximately 20 years greater than the mean age for the cohort, and the patient was discovered to have MEN 1 only after his son developed the clinical syndrome. However, further follow-up of the cohort is necessary to determine the impact of early screening and surveillance on overall survival. The present study demonstrates that EUS in experienced hands can identify PETs in asymptomatic patients with MEN 1, despite normal or equivocal results of biochemical testing. The algorithm currently used to evaluate asymptomatic patients with MEN 1 is shown in Figure 2. In addition to EUS screening, patients should undergo annual biochemical screening and CT of the abdomen and the pelvis every 3 years, examining for extra-intestinal complications of MEN 1. If the initial EUS is normal, our recommendation is that screening EUS be repeated in 3 years. If a small PET is identified (<1 cm), then consideration can be given to further surveillance, including annual biochemical tests and EUS. However, this approach raises questions that are not addressed by the current study: what is the appropriate interval for screening EUS after an initial normal EUS, as well as what should be the timing for surveillance EUS and surgical treatment for patients with small PETs. These are questions that must be addressed in future studies. REFERENCES 1. Marx S, Spiegel AM, Skarulis MC, Doppman JL, Collins FS, Liotta LA. NIH conference-multiple endocrine neoplasia: clinical and genetic topics. Ann Intern Med 1998;129:484-94. 2. Doherty GM, Olson JA, Frisella MM, Lairmore TC, Wells SA VOLUME 58, NO. 4, 2003
EUS detection of pancreatic endocrine tumors in MEN 1, asymptomatic patients
3.
4.
5.
6.
7.
8.
Jr, Norton JA. Lethality of multiple endocrine neoplasia type I. World J Surg 1998;22:581-7. Karges W, Schaaf L, Cralle H, Boehm BO. Concepts for screening and diagnostic follow-up in multiple endocrine neoplasia type 1 (MEN 1). Exp Clin Endocrinol Diabetes 2000;108:334-40. Schussheim DH, Skarulis MC, Agarwal SK, Simonds WF, Burns AL, Spiegel AM, Marx SJ. Multiple endocrine neoplasia type 1: new clinical and basic findings. Trends Endocrinol Metab 2001;12:173-8. Granberg D, Stridsberg M, Seensalu R, Eriksson B, Lundqvist G, Oberg K, Skogseid B. Plasma chromogranin A in patients with multiple endocrine neoplasia type 1. J Clin Endocrinol Metab 1999;84:2712-7. Kopp I, Bartsch D, Wild A, Schilling T, Nies C, Bergenfelz A, et al. Genetic screening in MEN-1 families. World J Surg 2001;25:610-6. Bansal R, Tierney W, Carpenter S, Thompson N, Scheiman J. Cost effectiveness of EUS for preoperative localization of pancreatic endocrine tumors. Gastrointest Endosc 1999;49:19-25. Skogseid B, Oberg K, Eriksson B, Juhlin C, Granberg D, Akerstrom G, Rastad J. Surgery for asymptomatic pancreatic lesion in multiple endocrine neoplasia type 1. World J Surg 1996;20:872-7.
VOLUME 58, NO. 4, 2003
E-J Wamsteker, P Gauger, N Thompson, et al.
9. Anderson MA, Carpenter S, Thompson NW, Nostrant T, Elta GH, Scheiman JM. Endoscopic ultrasound is highly accurate and directs management in patients with neuroendocrine tumors of the pancreas. Am J Gastroenterol 2000;95:2271-7. 10. Oberg K, Skogseid B. The ultimate biochemical diagnosis of endocrine pancreatic tumours in MEN-1. J Intern Med 1998; 243:471-6. 11. Brandi ML, Gagel RF, Angeli A, Bilezikian JP, Beck-Peccoz P, Bordi C, et al. Consensus statement: guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001;86:5658-71. 12. Thompson N. Management of pancreatic endocrine tumors in patients with multiple endocrine neoplasia type I. Surg Oncol Clin N Am 1998;4:881-91. 13. Gauger PG, Scheiman JM, Wamsteker EJ, Richards ML, Doherty GM, Thompson NW. Role of Endoscopic ultrasonography in screening and treatment of pancreatic endocrine tumours in asymptomatic patients with multiple endocrine neoplasia type I. Br J Surg 2003;90:748-54. 14. Delvalle J, Scheiman J. Zollinger-Ellison Syndrome. In: Yamada T, editor. Textbook of gastroenterology. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 1999. p. 144562.
GASTROINTESTINAL ENDOSCOPY
535