Esophageal Gastrointestinal Stromal Tumor With MKRN1-BRAF Fusion: A Rare Mediastinal Case

Abstract

Gastrointestinal stromal tumors (GISTs) are rare mesenchymal neoplasms typically involving the upper gastrointestinal tract. Most GISTs are driven by activating mutations in the KIT or PDGFRA genes, which are primary molecular markers for diagnosis and treatment. In addition, CD117 and DOG1 (discovered on GIST 1) are important immunohistochemical diagnostic markers of GISTs. Esophageal GISTs are exceptionally rare, and those driven by BRAF gene fusions are even more uncommon. This report describes a 69-year-old woman who underwent surgical resection of an esophageal GIST harboring an MKRN1-BRAF gene fusion, offering insight into future treatment and surveillance strategies.

Gastrointestinal stromal tumors (GISTs), although rare, are the most common type of gastrointestinal mesenchymal tumors. GISTS are typically located in the stomach and small intestine, with only about 1% to 10% occurring in the esophagus, colon, or rectum.^1,2 Although most GISTs are driven by KIT or PDGFRA gene mutations, our case of an MKRN1-BRAF gene fusion represents an extremely rare clinical finding that has been scarcely documented in the literature.

A 69-year-old woman presented to the thoracic surgery office with unintentional weight loss and anemia. Computed tomography imaging revealed a 2.6 × 2.7-cm paraesophageal mass with areas of decreased attenuation and necrosis, and subsequent positron emission tomography demonstrated an avid mass without associated mediastinal lymphadenopathy (Figures 1, 2). Bronchoscopy demonstrated normal bronchial anatomy, and subsequent esophagogastroduodenoscopy with endoscopic ultrasound and fine-needle aspiration demonstrated extrinsic compression of the distal esophagus in the area adjacent to the lesion. Preliminary cytologic examination of the fine-needle aspiration specimen was consistent with a spindle cell neoplasm. Upfront esophagectomy was thought to be unnecessary, given the uncertainty of diagnosis. Instead, the patient was consented for excisional biopsy to determine the exact pathologic process. If final pathologic examination was concerning, there were plans for multidisciplinary collaboration to discuss esophagectomy with potential neoadjuvant therapies. The patient subsequently underwent robot-assisted esophageal mass resection approached from the left side of the chest. This was completed through a thoracoscopic 4-port technique in a vertical hockey-stick fashion similar to the chest portion of an Ivor-Lewis esophagectomy. In addition, an assistant port was placed for specimen extraction. The mass was adherent to both the esophagus and the left lower lobe and was carefully mobilized and resected from the surrounding structures without violation of the capsule. The operation was tolerated well, and the patient was discharged home on postoperative day 2.

Figure 1.

Computed tomography of the chest with intravenous administration of contrast material demonstrating a smoothly marginated 2.6 × 2.7-cm paraesophageal mass to the left of the midesophagus with extrinsic compression of the esophagus (mass shown within circle).

Figure 2.

Metabolic activity associated with the paraesophageal mass without other metabolically active disease.

Official surgical pathologic examination demonstrated a well-circumscribed low-grade spindle cell neoplasm with myxoid stroma (Figure 3), with all sampled lymph nodes negative for malignancy. The mitotic count was 0 to 1 per 50 high-power fields. Tumor necrosis was not present. On immunohistochemical analysis, the tumor cells were positive for DOG1, CD34, SMA, and desmin (focal) and negative for CD117, STAT6, S100, SOX10, EMA, and ALK. Given the unusual location, unusual morphologic appearance with myxoid stroma, and immunoprofile with CD117 negativity, this case was initially signed out as low-grade spindle cell neoplasm with myxoid stroma, not further classifiable. Next-generation sequencing assay detected a somatic MKRN1-BRAF chromosomal rearrangement, confirming the diagnosis of GIST with BRAF fusion. Overall, given the small tumor size (<5 cm) and low mitotic count (<0-1 per 50 high-power fields), the tumor was considered at low risk of recurrence per the Miettinen classification,³ which is supported by the College of American Pathologists risk stratification criteria for GISTs.

Figure 3.

Esophageal gastrointestinal stromal tumor with BRAF fusion. (A) Low-magnification view of well-circumscribed nodule, showing (B) hypocellular cytology bland spindle cell proliferation with myxoid stroma and (C) positivity for DOG1. The tumor cells were negative for CD117 (not shown).

Comment

Esophageal GISTs are extremely rare and account for <1% of GISTs. Whereas most GISTs express mutations in the KIT or PDGFRA genes, a small subset is characterized by other oncogenic drivers, such as BRAF mutations and fusions.

The pathologic finding of the MKRN1-BRAF translocation represents an exceedingly rare genomic finding. BRAF is a serine-threonine kinase in the RAF family of kinases and is a central component of the RAS/MAPK pathway involved in cell differentiation, proliferation, migration, and apoptosis. The fusion of BRAF with different 5′ partners, in this case MKRN1, replaces the BRAF autoregulatory domain with a dimerization domain from the partner gene, which results in the production of constitutively active BRAF homodimers that are capable of driving tumor progression.

BRAF-mutated GISTs are morphologically similar to and immunohistochemically different from KIT-mutated GISTs. On the basis of limited data, CD117 can be negative or weakly positive in BRAF-mutated GISTS. For KIT-mutated cases, all would be positive for CD117. Negativity for CD117 can pose diagnostic challenges. Given this, CD117 negativity and abundant myxoid stroma should suggest unusual molecular subtypes of GISTs. In working up spindle cell neoplasms, adding DOG1 to CD117 and other immunohistochemical markers, such as CD34, SMA, desmin, S100, SOX10, SS18, or STAT6, can improve diagnostic accuracy⁴

From a clinical perspective, the patient’s MKRN1-BRAF fusion–positive GIST would fall under the wild-type category, known to be resistant to standard tyrosine kinase inhibitors such as imatinib. These GISTs do not respond to standard tyrosine kinase inhibitor therapy because the BRAF oncogenic signal originates downstream of KIT. Therefore, even if imatinib blocks KIT completely, the BRAF-MEK-ERK pathway remains fully active because of the constitutively active BRAF mutation. There are several Food and Drug Administration–approved BRAF and MEK1/2 inhibitors on the market for treatment of BRAF or MEK mutation-positive cancers, and they are often used in combination to increase treatment efficacy.⁵

In respect to tumor immunohistochemistry in the context of prognosis and surveillance, because BRAF-mutant GISTs are rare, current literature is insufficient to assess impact on prognosis. Given the rarity of BRAF fusion GISTs, primary drivers of recurrence and stratification should be based on established risk stratification, such as size, mitotic count, and location. Unusual locations for GISTs, such as esophagus, mesentery, or peritoneum, should be classified according to the criteria for jejunal and ileal GISTs. As a result, this patient will undergo surveillance per institutional standard of care with restaging computed tomography of the chest, abdomen, and pelvis every 6 months for 24 months, followed by annual imaging through year 5 because of the low-risk tumor profile.

This report illustrates unusual morphology and the immunohistochemical variability seen in BRAF-mutated GISTs, emphasizing that all surgical pathology specimens, especially from unusual locations, should undergo comprehensive molecular profiling for accurate diagnosis and potential use of targeted therapy. Identifying actionable targets such as MKRN1-BRAF fusions may open opportunities for targeted therapy, improve postoperative surveillance strategies, and contribute to a growing understanding of GIST heterogeneity. In conclusion, we present the case of a patient with a paraesophageal GIST that was revealed to harbor an incredibly rare MKRN1-BRAF translocation mutation. To date, only a handful of GISTs with BRAF fusions and even fewer with esophageal involvement have been identified and documented in the literature. This report serves to add to the growing body of literature regarding fusion proteins and cellular signaling pathways and to help guide future clinicians with treatment and surveillance strategies when faced with unique tumor pathology.