C 2005) Digestive Diseases and Sciences, Vol. 50, No. 7 (July 2005), pp. 1315–1318 (
DOI: 10.1007/s10620-005-2779-2

CASE REPORT

Gastrointestinal Stromal Tumor (GIST) of the Esophagus Detected by Positron Emission Tomography/Computed Tomography WEI-CHOU CHANG, MD,* CHING TZAO, MD, PhD,† DANIEL HUENG-YUAN SHEN, MD, PhD,‡ CHENG-YI CHENG, MD, PhD,‡ CHENG-PING YU, MD, PhD,§ and HSIAN-HE HSU, MD* KEY WORDS: gastrointestinal stromal tumor; esophagus; positron emission tomography/computed tomography.

Although rare elsewhere in the gastrointestinal tract, leiomyoma (LM) is the most common esophageal mesenchymal neoplasm. As a comparison, gastrointestinal stromal tumor (GIST) predominates in the stomach and intestines, but esophageal GIST has been reported less frequently. In contrast to other esophageal mesenchymal tumors, GIST is typically immunoreactive for KIT protein (CD117) in more than 95% of cases and is frequently coexpressed with CD34 (60 to 70%) (1, 2). Mutation of the c-kit proto-oncogene is associated with increasing risk for malignant transformation (3). Positron emission tomography (PET) has recently been employed to monitor progression of GIST from gastrointestinal tracts other than the esophagus (4–6) but its role in the diagnosis of GIST remains unclear. Herein we present a case of GIST of the esophagus confirmed by immunohistochemical stain. Conventional radiological studies including barium meal and computed tomography (CT) were unable to differentiate reliably LM or leiomyosarama (LMS) from GIST of the esophagus. Fused PET/CT was performed in this patient and provided additional information preoperatively in terms of its metabolic activity and the likelihood of a submucosal tumor other than LM. Manuscript received October 16, 2004; accepted November 3, 2004. From the *Department of Radiology, †Division of Thoracic Surgery, Department of Surgery, ‡Department of Nuclear Medicine, and §Department of Pathology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, Republic of China. Drs. Tzao and Chang contributed equally to this work. Address for reprint requests: Ching Tzao, MD, PhD, Division of Thoracic Surgery, Department of Surgery, Tri-Service General Hospital, 325, Sec. 2, Cheng Gong Road, Nei Hu, Taipei 114, Taiwan, Republic of China; [email protected].

CASE REPORT A 36-year-old man presented with grade I dysphagia over 3 months before this admission. No remarkable past medical condition was noted. He denied any history of cigarette smoking or alcohol use, and the family history was also unremarkable. Laboratory results of a complete blood cell count as well as biochemical investigations of electrolytes and liver enzymes were normal. Prior to admission, an endoscopic ultrasonography barium swallowing study of the esophagus showed a segmental filling defect with intact mucosa, 6.5 cm long, over the lower third of the esophagus, which extended to the gastric cardia just beyond the esophagogastric junction (Figure 1A). Consistent with barium swallowing, endoscopic ultrasonography (EUS) disclosed a hypoechoic submucosal tumor located predominantly in the lower esophagus, with extension to the cardia of the stomach (Figure 1B). CT with contrast enhancement revealed a poorly enhanced submucosal tumor over the lower third of the esophagus without an apparent enlarged mediastinal lymph node. These studies suggested a submucosal tumor such as a LM or a GIST. As malignant tumor was considered because of the relatively large size of this lesion (<5 cm), endoscopic biopsy was performed to determine its nature but did not yield a conclusive result. PET/CT was therefore performed to obtain more information on the metabolic activity of this tumor and showed moderately increased glucose uptake by 18 F-FDG-PET (Figure 1C), raising the suspicion of a submucosal tumor other than LM. The patient underwent an exploratory thoracotomy with complete resection by enucleation of the tumor mass followed by closure of the muscularis. Upon microscopic examination with hematoxylin and esoin (H&E) stain, the sections showed pictures of a well-circumscribed submucosal esophageal tumor, which was hypocellular and consisted of benign-looking spindle cells without atypism or mitosis (Figure 2A). The pathologists reported it as a GIST, as they observed positive immunoreactivity for actin, c-kit (Figure 2B), and CD34 (Figure 2C) but negative staining for desmin, S-100 protein, vimentin and cytokeratin. The patient had an uneventful recovery and was followed

Digestive Diseases and Sciences, Vol. 50, No. 7 (July 2005) C 2005 Springer Science+Business Media, Inc. 0163-2116/05/0700-1315/0


1315

CHANG ET AL.

Fig 1. (A) Barium esophagogram demonstrated a segmental filling defect with normal mucosa, 6.5 cm long, over the lower third of the esophagus, extending to just beyond the esophagogastric junction. (B) EUS of the esophagus disclosed a lobulated, hypoechoic submucosal tumor with eccentric annular localization, arising from the mucosa propia (MP) layer (arrow). (C) 18 F-FDG PET/CT scan showed a well-delineated, homogeneous, low-density mass over the lower third of the esophagus on cross-sectional images, with increased 18 F-FDG uptake (arrow).

regularly at 3-month intervals for a year without evidence of recurrence.

DISCUSSION GIST, previously misclassified as LM or LMS, has recently been identified as a separate clinicopathologic entity. Throughout the course of the human digestive tract, this newly defined stromal tumor occurs predominantly in the stomach (60–70%), followed by the small intestine (20–30%), colon (5%), and esophagus (<5%) (7). In general, 70% of GISTs are benign, and 30% are malignant (5). As far as we know, no report regarding the application of PET scan in differentiating a LM from a GIST of the esophagus has been published before. Similar to a LM on barium esophagrogram, our patient’s GIST showed a segmental smooth indentation of

1316

the mucosa along the longitudinal axis of the esophagus without obvious luminal obstruction. In contrast, LMS tends to show exophytic growth and usually appears as a large intraluminal mass enlarging the esophageal lumen because of its expansive growth (8). On CT scans, esophageal GIST shows an eccentric intraluminal mass with a smooth margin and homogeneous soft tissue attenuation. Further EUS results supported a submucosal tumor but were unable to differentiate LM from GIST. The value of preoperative endoscopic biopsy in differentiating the pathology for a larger submucosal esophageal tumor remains controversial. It is not generally recommended, as mucosal tear caused by biopsy may increase the risk of esophageal leak after a submucosal resection (9). It was not surprising that we were unable to obtain a representative tissue specimen for histologic confirmation of the tumor, as it laid entirely beneath the mucosa. PET/CT Digestive Diseases and Sciences, Vol. 50, No. 7 (July 2005)

GASTROINTESTINAL STROMAL TUMOR OF THE ESOPHAGUS

Fig 2. H&E stain (A) and immunostaining for c-kit (CD117; B) and CD34 (C) for the resected esophageal GIST. (A) Microscopically, the sections showed a well-circumscribed submucosal esophageal tumor with hypocellularity, consisting of benign-looking spindle cells (arrows) without atypism or mitosis. Immunohistochemical stain shows positive immunoreactivity within tumor cells (arrows) for CD117 (B) and CD34 (C).

in this patient showed moderate uptake of 18 F-FDG-PET by the tumor, making us consider it a submucosal tumor other than LM. Further, it showed almost-equivalent uptake intensity to the liver, usually with a normal standard uptake value (SUV) of about 2 to 3 mg/ml, indicating that it is a relatively less aggressive tumor compared with malignant GIST, which generally presents with a SUV of about 5 mg/ml (5). Enucleation of the tumor was planned, as a possible benign submcosal lesion was suggested based on findings of imaging studies and PET/CT scan. Although there are no specific criteria for benign or malignant GIST, in general, a few rules may be useful based on analysis of multiple morphologic features, including site, size, cellularity, mitotic rate, and invasive characteristics (1, 10). Our patient appeared to have a benign GIST except for the relative larger size of the tumor. Historically, most GISTs are diagnosed as LMs or LMSs, largely because of their site of origin within the muscluaris as well as their morphologic resemblance to true smooth muscle neoplasms. Most GISTs can be reliably diagnosed using a combination of morphologic features and immunostaining results including KIT and CD34 (3). Whether or not there is a correlation between 18 F-FDG uptake and KIT expression by GIST needs to be further investigated. Enucleation Digestive Diseases and Sciences, Vol. 50, No. 7 (July 2005)

of the tumor was performed, as we considered it to be a benign GIST based on preoperative workup, which showed a moderately increased 18 F-FDG uptake as mentioned, a relatively smaller size, and an absence of necrosis that often characterizes a malignant mesenchymal tumor. We recommend, however, that the surgeon inform the patient that a radical resection such as esophagectomy should be performed if the pathology turns out to be a malignancy or, alternatively, that imatinib mesylate (Glivec) should be used as an adjuvant therapy. In conjunction with conventional radiological evaluation, PET might help us better to detect GIST, its progression and recurrence (5, 6). In conclusion, PET/CT may help us to define the nature of an esophageal mesenchymal tumor preoperatively and in follow-up of tumor recurrence after compete resection. Definitive diagnosis of these tumors depends on a comprehensive immunopathologic examination of the surgical specimen. REFERENCES 1. Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, Miettinen M, O’Leary TJ, Remotti H, Rubin BP, Shmookler B, Sobin LH, Weiss SW: Diagnosis of gastrointestinal stromal tumors: A consensus approach. Hum Pathol 33:459–465, 2002

1317

CHANG ET AL. 2. Joensuu H, Fletcher C, Dimitrijevic S, Silberman S, Roberts P, Demetri G: Management of malignant gastrointestinal stromal tumours. Lancet Oncol 3:655–664, 2002 3. Miettinen M, Sarlomo-Rikala M, Sobin LH, Lasota J: Esophageal stromal tumors: a clinicopathologic, immunohistochemical, and molecular genetic study of 17 cases and comparison with esophageal leiomyomas and leiomyosarcomas. Am J Surg Pathol 24:211–222, 2000 4. Antoch G, Kanja J, Bauer S, Kuehl H, Renzing–Koehler K, Schuette J, Bockisch A, Debatin JF, Freudenberg LS: Comparison of PET, CT, and dual-modality PET/CT imaging for monitoring of imatinib (STI571) therapy in patients with gastrointestinal stromal tumors. J Nucl Med 45:357–365, 2004 5. Gayed I, Vu T, Iyer R, Johnson M, Macapinlac H, Swanston N, Podoloff D: The role of 18F–FDG PET in staging and early prediction of response to therapy of recurrent gastrointestinal stromal tumors. J Nucl Med 45:17–21, 2004

1318

6. Lehnert T: Gastrointestinal sarcoma (GIST)—-A review of surgical management. Ann Chir Gynaecol 87:297–305, 1998 7. Lau S, Lui CY, Yeung YP, Lam HS, Mak KL: Gastrointestinal stromal tumor of rectum: a report of 2 cases. J Comput Assist Tomogr 27:609–615, 2003 8. Lyons AS, Garlock JH: Leiomyosarcoma of the esophagus; report of first successful resection. Surgery 29:281–287, 1951 9. Burkill GJ, Badran M, Al–Muderis O, Meirion TJ, Judson IR, Fisher C, Moskovic EC: Malignant gastrointestinal stromal tumor: distribution, imaging features, and pattern of metastatic spread. Radiology 226:527–532, 2003 10. Yu CC, Fletcher CD, Newman PL, Goodlad JR, Burton JC, Levison DA: A comparison of proliferating cell nuclear antigen (PCNA) immunostaining, nucleolar organizer region (AgNOR) staining, and histological grading in gastrointestinal stromal tumours. J Pathol 166:147–152, 1992

Digestive Diseases and Sciences, Vol. 50, No. 7 (July 2005)