Gastrointestinal schwannomas: a rare but important differential diagnosis of mesenchymal tumors of gastrointestinal tract

Schwannomas are slow-growing, homogeneous, mostly benign tumors arising from the schwann cells of the nerve sheath [1, 2]. They are most commonly found in the cranial vault including the myelin-forming cells of the 8th cranial nerve [1, 3]. They can rarely occur in the gastrointestinal (GI) tract, representing about 2–6% of all mesenchymal tumors [4]. The most common site (60–70% of all GI cases) is the stomach, followed by the colon and rectum (3%) [5‐10]. Small-intestinal and esophageal schwannomas have been very rarely reported [7].

Schwannomas of the GI tract, firstly presented by Daimaru et al. in 1988 [11], are usually benign and notably different neoplasms from conventional schwannomas that arise in the soft tissue or the central nervous system [7]. They are classified under a heterogeneous group of mesenchymal or neuroectodermal neoplasms which arise from the wall of the gastrointestinal tract and include schwannomas, gastrointestinal stromal tumors (GIST), leiomyomas, leiomyosarcomas, neurofibromas, lipomas, ganglioneuromas, paragangliomas, granular cell tumors, and glomus tumors [7]. The GI schwannomas are most frequently asymptomatic presenting as submucosal lesions, usually detected incidentally during a laparoscopy / laparotomy, esophagogastroduodenoscopy (EGD), endoscopic ultrasonography (EUS) or at imaging [6, 7]. A preoperative diagnosis of schwannomas cannot always be carried out with high accuracy and the diagnosis is established by the pathological examination of the surgical specimen [5].

The purpose of our study was the description of clinical, histopathological and immunohistochemical features of GI schwannomas, providing long-term follow-up and confirming the benign nature of these tumors.

Seven patients with histologically identified GI schwannoma during a 10 year period, were involved in this retrospective study. Clinical, histopathological, immunohistochemical and surgical data were reviewed. A complete surgical excision of the tumors was performed in all patients. Specimens were obtained from all patients. The diagnosis of GI schwannomas was based on histopathological analysis accompanied by immunohistochemical staining performed for S-100 protein, glial fibrillary acidic protein (GFAP), c-kit (CD117), CD34, discovered on GIST-1 (DOG-1), smooth muscle actin (SMA) and desmin. The patients were regularly followed up on an outpatient basis. Follow-up included physical examination, blood counts, serum chemistries, endoscopy and computed tomography (CT)-scan. Median follow-up was 60 months (range, 24–185).

Gastrointestinal schwannomas are rare mesenchymal tumors [5‐7], firstly reported by Daimaru, who identified a schwannoma as a primary GI tumor entity based on the positive S-100 immunostaining [11]. They represent ca. 2 to 6% of all mesenchymal tumours of the GI tract [4, 12, 13]. They are of clinical importance since GI schwannomas are notably different neoplasms from conventional soft-tissue and central nervous system schwannomas, some of which may be associated with neurofibromatosis 2.

In terms of pathologic evaluation, GI schwannomas are classified as non-epithelial tumours, and represent a separate, homogenous entity, distinct from leiomyomas, leiomyosarcomas, gastrointestinal autonomic nerve tumors (GANTs) and GISTs [12, 14, 15]. They can be identified in every part of the GI tract but they occur predominantly in stomach. The second most common site seems is colon, whereas schwannomas located to small intestine or esophagus are rare [5‐9, 16, 17].

Our series shows a remarkable regularity regarding to the characteristics of GI-schwannomas. They occur more frequently in the sixth decade of life and there is a predominance of females [11, 12, 15, 17‐21]. Our results showed that all of our patients except one woman (39 years-old), were over 70 years of age at the time of diagnosis. These tumors do not usually cause any symptoms and can be found out incidentally, as presented in our series. However, when symptomatic, they represent with unspecific clinical symptoms, and particularly Bruneton et al. reported that most common symptoms of these patients are bleeding and abdominal pain [22].

A preoperative diagnosis of GI schwannomas is generally difficult and challenging, if not impossible, since the tumor have no specific clinical symptoms and no diagnostic modality can show any pathognomonic features unique to this tumor [22]. On CT examination, these tumors mostly represent as exophytic masses displaying homogeneous enhancement in most cases whereas cystic change, cavity formation, necrosis or calcification are uncommon [7, 16, 24]. Our CT findings were comparable to previous reports in terms of enhancement patterns and growth patterns. On endoscopy, gastric schwannomas appear grossly as elevated submucosal lesions, and a central ulcer can be seen in 25–50% due to ischemic changes in the covering mucosa [16, 25]. It should be noted that there is an important false-negative rate of endoscopic biopsy diagnosis because of normal mucosa overlying a submucosal lesion [22]. Therefore, endoscopic biopsy is supposed to be not as effective as expected in the diagnosis of gastric schwannomas, since the results mostly demonstrate nonspecific spindle cells. EUS-guided fine needle aspiration (FNA) biopsies of submucosal lesions in the upper GI tract can be helpful in preoperative diagnosis of spindle cell tumors but are sometimes not representative of deeper submucosal tissue [25‐29]. Two patients with gastric schwannomas underwent an EUS-guided FNA without representative findings. However, in one case (case Nr. 3) a CT-guided biopsy of a 2.2 cm mass of duodenum was performed and detected a GI schwannoma. Generally, the CT and endoscopic findings in our patients were similar to those from previous studies as far as the location and mucosal change.

GI-schwannomas are supposed to arise from the myenteric plexus of the GI wall because of their immunophenotypic similarities; both schwannomas and myenteric plexus cells express S-100 protein and GFAP [7, 11]. Pathologically, GI schwannomas are considered to be notably different tumors from conventional schwannomas, which arise from central nervous system and soft tissues [7, 21]. Macroscopically, these tumors are round or fusiform and their cut surface reveals smooth, glistening, grey-white appearance. Microscopically, unlike conventional schwannomas, GI schwannomas are not always encapsulated, but mostly well circumscribed. Cystic areas of hemorrhage and calcification may be present. These tumors are composed of uniformly spindled schwann cells arranged in a interlacing fascicles, often with germical centers [4, 21].

The pathologic findings of the GI schwannomas in the present analysis were consistent with the previously described findings. On immunohistochemistry, vimentin, GFAP and S100 protein are diffusely and strongly expressed by the cells of GI schwannomas [21]; the S100 immunostaining pattern is both in a nuclear and cytoplasmic distribution [4]. Expression of CD34 is rare, whereas there is lack of CD117, SMA and desmin positivity [30]. This immunohistochemical pattern is very important because it can differentiate GI schwannomas from other GI tract mesenchymal neoplasms.

The main differential diagnosis for an exophytic lesion arising in the wall of the GI tract is a GIST, as it is the most common mesenchymal tumor located in GI tract [7, 31, 32]. Voltaggio et al. estimated that the ratio of gastric GIST to gastric schwannoma is approximately 45 to 1 [33]. Some other smaller studies have shown lower gastric GIST frequencies (8-14 to 1) in relation to schwannoma [11, 19]. In these cases, immunohistochemistry is extremely helpful. Another tumor entity included in the differential diagnostic of GI schwannomas is the primary and secondary lymphomas because of their similar CT appearance to schwannomas. Both of them arise from the GI wall and tend to appear as a homogeneous mass in CT. However, the presence of adenopathy in most of lymphomas, in contrast to GI schwannomas, is an important distinguishing feature [7]. Other entities included in the differential diagnosis are GI variant of clear cell sarcoma, metastatic malignant melanoma and GI adenocarcinoma [7, 34].

Schwannomas of the GI tract are generally benign since previous follow-up studies have not identified any malignant variants [4, 9, 11, 19‐21]. Therefore, pathologic parameters such as tumor size and mitotic rate seem not to have any prognostic significance. The complete surgical resection seems to be the treatment of choice because of the preoperative diagnostic uncertainty and the excellent long-term outcome, as these tumors are uniformly benign [23, 35]. Tumor recurrence is generally associated with an incomplete surgical margin. The outcome after surgical resection in our study was excellent with no malignant variant and no recurrences or metastases. However, a very few cases of gastric malignant schwannomas have been reported in the literature [25, 36, 37]. Malignant GI schwannomas are extremely rare and they cannot be distinguished from benign schwannomas only on the basis of the histopathological examination of the resected specimens. Furthermore, cases of “malignant schwannomas” also called malignant peripheral nerve sheath tumors have been reported [38]. These malignant tumors with neural differentiation are considered to be distinct tumors from GI schwannomas and are called gastrointestinal autonomic nerve tumors (GANTs) [38]. The usefulness of molecular therapy for gastric malignant schwannomas is not clearly established because only a few cases have been reported. Further molecular therapy studies will be useful in order to determine its role in the treatment of GI malignant schwannoma.

Schwannomas are rare, slow-growing and mostly asymptomatic gastrointestinal mesenchymal tumors, notably different from conventional soft-tissue and central nervous system schwannomas. They present a challenge in GI tract surgical oncology due to difficulty in the preoperative diagnosis, as the clinical, endoscopic and imaging findings are nonspecific. The definitive diagnosis can only be established through the histopathological and immunohistochemical examination, which is confirmatory. Complete surgical excision is the gold standard in the treatment of schwannomas and the prognosis of patients is excellent since these lesions are mostly benign.