To resect or observe? Conservative management of a large oesophageal granular cell tumour

Summary

Granular cell tumours (GCTs) (Abrikossoff tumours) are rare neoplasms of Schwann cell origin and are infrequently encountered in the oesophagus. Although most are benign, lesion size greater than 2 cm is often cited as an indication for surgical excision due to concern for malignant potential. We describe a middle-aged African American male in whom a large oesophageal GCT was discovered incidentally during endoscopic evaluation following traumatic injury. Despite exceeding commonly referenced size thresholds for resection, the tumour demonstrated no high-risk histological features and the patient remained asymptomatic. Given substantial operative risk from prior extensive abdominal surgery, a multidisciplinary team favoured close endoscopic surveillance. This case highlights the limitations of size-based management strategies and emphasises the importance of individualised risk assessment incorporating histology, endoscopic characteristics, symptoms and surgical risk.

Background

Granular cell tumours (GCTs), first described by Abrikossoff in 1926, were initially believed to arise from skeletal muscle but are now recognised as tumours of Schwann cell origin.^{1 2} These tumours are typically benign, slow-growing and asymptomatic, with malignant transformation reported in approximately 1%–2% of cases.³ The oesophagus is an uncommon site, accounting for roughly 2% of all GCTs.⁴

Due to the rarity of oesophageal involvement, standardised management guidelines are lacking. Lesion size greater than 2 cm is frequently cited as an indication for surgical excision, though this recommendation is derived largely from small case series rather than prospective data. As a result, clinicians are often faced with uncertainty when managing larger lesions that lack histological or clinical features of malignancy. This case illustrates the importance of individualised decision-making beyond lesion size alone.

Case presentation

A middle-aged African American male with no significant prior medical history presented to the emergency department following multiple thoracoabdominal gunshot wounds. He underwent emergent exploratory laparotomy with splenectomy, distal pancreatectomy, gastric injury repair and temporary abdominal closure with negative pressure therapy.

During his inpatient course, esophagogastroduodenoscopy (EGD) and endoscopic ultrasound (EUS) were performed to evaluate upper gastrointestinal anatomy. These studies incidentally identified a well-circumscribed, homogeneous, hypoechoic submucosal mass at the gastro-oesophageal junction measuring approximately 2.0 cm×1.7 cm. The lesion originated from the submucosal layer and abutted the adjacent liver capsule. A shave biopsy was obtained.

Following recovery from his traumatic injuries, the patient underwent repeat outpatient EGD/EUS approximately 7 months later. This demonstrated persistent irregular submucosal thickening with a hypoechoic appearance, measuring up to 26 mm×8 mm, consistent with prior findings. Repeat biopsies confirmed a GCT. Histopathological examination showed polygonal cells with abundant granular cytoplasm and strong S100 immunostain positivity, without evidence of mitotic activity or necrosis (figure 1).

Figure 1. Histopathological findings from shave biopsy (left) and corresponding S100 immunohistochemical staining (right).

The oncology and gastroenterology teams were consulted. Although the lesion exceeded commonly cited size thresholds for resection, the patient remained asymptomatic and biopsy findings did not suggest malignancy. Surgical oncology evaluation emphasised the patient’s extensive history of upper abdominal surgery and associated operative risk. After multidisciplinary discussion and shared decision-making, a strategy of close endoscopic surveillance was selected.

Approximately 1 year later, repeat EGD/EUS was performed to reassess resectability. The lesion remained stable in size, measuring approximately 26 mm×16 mm, with well-defined margins and submucosal origin. Endoscopic removal was deemed technically challenging. Given ongoing absence of symptoms, lack of high-risk histological features and stable imaging findings, surgical intervention was deferred in favour of continued surveillance.

Investigations

Initial EGD with EUS performed during the patient’s inpatient hospitalisation revealed a well-circumscribed, homogeneous, hypoechoic mass located at the gastro-oesophageal junction. The lesion originated from the submucosal layer of the oesophageal wall and measured approximately 2.0 cm×1.7 cm at initial evaluation. The mass demonstrated well-defined margins and abutted the adjacent liver capsule without evidence of invasion. A shave biopsy was obtained at that time.

Repeat outpatient EGD/EUS approximately 7 months later demonstrated persistent irregular thickening of the oesophageal submucosa with a hypoechoic appearance, measuring up to 26 mm×8 mm, consistent with prior findings. Subsequent histopathological evaluation confirmed a GCT. Microscopic examination showed polygonal cells with abundant granular cytoplasm, and immunohistochemical staining demonstrated strong S100 positivity, consistent with neural origin. No mitotic activity, necrosis, pleomorphism or spindling was identified on biopsy specimens. These findings suggested a benign lesion without histological features associated with malignant potential.

Follow-up EGD/EUS performed approximately 1 year later demonstrated stability of the lesion, with measurements of approximately 26 mm×16 mm and no change in echogenicity or margins. There was no evidence of mucosal ulceration, invasion of adjacent structures or regional lymphadenopathy.

Differential diagnosis

The differential diagnosis for a submucosal oesophageal mass includes leiomyoma, gastrointestinal stromal tumour, lipoma and neuroendocrine tumour. Leiomyomas typically arise from the muscularis propria and may demonstrate different echogenic characteristics on EUS. Gastrointestinal stromal tumours were considered; however, the lesion’s well-defined margins, submucosal origin and characteristic histopathological findings were not supportive of this diagnosis. Lipomas generally appear hyperechoic on EUS and lack the histological features observed in this case. Neuroendocrine tumours were also considered but were less likely given the absence of characteristic immunohistochemical staining and the lesion’s submucosal rather than mucosal origin. The combination of endoscopic appearance, histopathology and immunohistochemical profile supported the diagnosis of oesophageal GCT.

Treatment

Given the absence of symptoms, lack of high-risk histological features and stable appearance on serial endoscopic evaluations, no immediate endoscopic or surgical intervention was pursued. Although lesion size exceeded commonly cited thresholds for resection, multidisciplinary evaluation involving gastroenterology, oncology and surgical oncology teams concluded that the risks of operative intervention outweighed the potential benefits in this patient. The patient’s extensive history of upper abdominal surgery following traumatic injury, including splenectomy, distal pancreatectomy and gastric repair, was felt to significantly increase perioperative morbidity.

Endoscopic removal was considered but deemed technically challenging due to lesion size and submucosal location. After shared decision-making with the patient, a strategy of close endoscopic surveillance was selected, with plans for periodic EGD and EUS to monitor for interval growth, development of symptoms or emergence of features concerning for malignant transformation.

Outcome and follow-up

The patient remains asymptomatic with stable lesion size on serial EGD/EUS examinations. Continued periodic surveillance has been planned.

Discussion

GCTs of the oesophagus are rare and most often identified incidentally.^{4 5} Although malignant behaviour is uncommon, concern for malignancy has led to recommendations for resection of larger lesions, particularly those exceeding 2 cm.⁶ However, reliance on size alone may be overly simplistic and may expose patients to unnecessary procedural risk.

Histologically, GCTs are characterised by polygonal or spindle-shaped cells with abundant granular cytoplasm. Immunohistochemical staining typically demonstrates positivity for S100, vimentin, neuron-specific enolase and CD57, supporting neural origin.^{7 8} Several histological criteria have been proposed to identify malignant potential. Fanburg-Smith et al described six features associated with malignancy, including necrosis, spindling, vesicular nuclei with prominent nucleoli, increased nuclear-to-cytoplasmic ratio, increased mitotic activity and pleomorphism.⁹ Nasser et al later emphasised that metastasis remains the most definitive criterion for malignancy, with Ki-67 and p53 serving as adjunctive markers.¹⁰

In the present case, despite lesion size exceeding 2 cm, none of the histological features associated with malignancy were present. Furthermore, the patient’s extensive prior abdominal surgery significantly increased operative risk. These factors supported a conservative approach, highlighting the importance of individualised management that incorporates histology, endoscopic appearance, symptom burden and patient-specific surgical risk. In scenarios where resection is not feasible or carries prohibitive risk, alternative local ablative therapies deserve consideration, although data specific to oesophageal GCTs remain limited to isolated case reports. Argon plasma coagulation has been reported to achieve complete ablation of a 2 cm benign oesophageal GCT with immediate symptom resolution and durable follow-up in a single case.¹¹ Percutaneous cryoablation has also been described as a minimally invasive alternative for a benign GCT at a non-oesophageal site, with reduction in lesion size and no internal enhancement on follow-up imaging.¹² Endoscopic Nd-YAG laser therapy and injection of dehydrated alcohol have additionally been reported for GCTs at various sites. These approaches may represent reasonable alternatives in carefully selected patients in whom endoscopic or surgical resection is not achievable, although their routine application to oesophageal GCTs cannot be recommended given the absence of comparative or prospective data. In the present case, the lesion remained asymptomatic and stable without intervention, and the substantial perioperative risk associated with prior abdominal surgery did not favour an ablative procedure.

To our knowledge, there are no registered clinical trials evaluating the management of oesophageal GCTs, a reflection of the rarity of this entity and the absence of prospective evidence to inform management guidelines. Current recommendations are therefore derived from small case series and isolated reports, underscoring the need for multi-institutional registries to better define surveillance and resection thresholds in high-risk patients. This case further underscores the pivotal role of multidisciplinary team-based decision-making, with gastroenterology, oncology and surgical oncology jointly weighing histological risk, endoscopic findings and operative risk before arriving at a shared management plan.

Learning points.

• Oesophageal granular cell tumours are rare and usually benign, with malignant transformation reported in approximately 1%–2% of cases.

• Lesion size alone may be insufficient to dictate management decisions.

• Histological risk stratification using Fanburg-Smith or Nasser criteria is critical in assessing malignant potential.

• Conservative surveillance may be appropriate for select large, asymptomatic lesions lacking high-risk features, particularly in patients with elevated surgical risk.