Subepithelial rectal gastrointestinal stromal tumor – the use of endoscopic ultrasound-guided fine needle aspiration to establish a definitive cytological diagnosis: a case report

GISTs are mesenchymal tumors that display differentiation toward the lineage of interstitial cells of Cajal [7] and typically express CD117/c-kit. Recent studies showed that these cells play a fundamental role in gastrointestinal motility because they conduct inputs from enteric motor neurons, generating intrinsic electrical rhythmicity in phasic smooth muscles [8]. On histological examination, the cells of Cajal are at the fourth layer (muscularis propria) [8, 9]. For that reason, GISTs present as SETs.

A definitive diagnosis of a GIST tumor depends on the association of histological and immunohistochemical findings. On histological examination, there are three types of GIST: spindle cell type (70%), epithelioid (25%), and mixed subtype (5%) [10]. The most sensitive and specific immunohistochemistry marker is CD117/c-kit while other standard markers are platelet-derived growth factor receptor-alpha (PDGFR-α) and CD34 [7]. Nonetheless, recent studies showed that a new marker, Discovered On GIST-1 (DOG-1), plays a significant role in diagnostic yield of GIST tumors, especially when CD117 expression has not been demonstrated. Therefore, DOG1 has recently been recommended to be part of the routine diagnostic panel [11].

GISTs are equally distributed between genders, with a mean age of presentation of 60 to 65-years old (ranging from 10 to 100 years) [12]. Although GISTs are a rare type of cancer, they are the most common sarcoma of the gastrointestinal tract [1] with an estimated incidence of 1/100,000 per year [13]. Treatment is responsible for an enormous financial impact; the estimated cost per month per patient is approximately US$4000 for non-recurrent cases and more than US$8000 for recurrent cases [14].

Approximately 80% of patients are symptomatic, despite the non-specificity of symptoms (mild abdominal pain, gastrointestinal bleeding, and fatigue due to iron deficiency anemia). The most common primary affected site is the stomach while esophagus, colon, and rectum are the rarest [12]. GISTs account for only 0.1% of cases of rectal tumors [15].

Characteristics identified clinically and on esophagogastroduodenoscopy (EGD) are well known to help differentiate between GISTs and non-GIST SETs. Patients of an older age, tumor site location outside the gastric cardia, large tumor size, exophytic growth pattern, and ulceration or dimpling are independent preoperative predictive factors for GISTs versus non-GISTs [16].

Metastatic progression of GISTs it is not uncommon. It is expected that almost 30% of all completely resected GISTs will recur within 2 years [17]. Most of the recurrences are disseminated, especially involving the liver and peritoneum [18]. Patients presenting a metastatic GIST are treated mainly with imatinib (tyrosine-kinase inhibitor) therapy and the median overall survival rate is 5 years [19]. In this context, positron emission tomography (PET) with ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) tracer may provide valuable information. A recent systematic review showed that ¹⁸F-FDG PET was a strong predictor of clinical outcome and in assessing treatment response [20]. Other studies found that it provides a timelier and more accurate response assessment compared to a computed tomography (CT) scan and magnetic resonance imaging (MRI) [21]. Moreover, ¹⁸F-FDG PET may provide information about the malignancy: a tumor presenting maximum standardized uptake value (SUV) greater than 3.0 has a high malignant potential even in small tumors (<2 cm) [22]. Therefore, this imaging modality should be considered in all cases of GISTs, especially in metastatic disease, and while monitoring treatment response.

Regarding EUS evaluation, GISTs are seen as fourth layer tumors. The main differential diagnoses are leiomyoma, leiomyosarcoma, and schwannoma [23]. Precise diagnosis is essential because each of the pathologies mentioned above has a different treatment, follow-up, and prognosis. For small leiomyomas (<2 cm), surveillance annually with EGD may be performed [24, 25]. If the lesion is larger, symptomatic, or growing, then surgical resection is advised. Small leiomyomas (<2 cm) may be treated with endoscopic resection instead of surgical resection. The endoscopic resection of leiomyomas has lesser perforation rates than GISTs resection [26]. Leiomyosarcomas have a worse prognosis compared to GISTs, as their recurrence rate and subsequent metastasis rate may reach 70% and 80% respectively [27]. Schwannomas are benign tumors that tend to recur locally and to become malignant. Their response to chemotherapy and radiotherapy remains uncertain [28]; consequently, the treatment of choice for schwannomas is surgical resection with free margin. If they are uniformly benign, the long-term outcome is excellent [29]. Furthermore, the standard treatment for GISTs is open or laparoscopic surgical resection with free margins [30], although recent studies demonstrated the feasibility and safety of endoscopic resection through an endoscopic submucosal dissection (ESD) technique for small tumors. GISTs have been associated with a relatively low recurrence rate during long-term follow-up despite little R₀ resection rate [31]. Other prospective multicenter studies are needed to assess the real role of endoscopy in the treatment of GISTs, but initial data are encouraging. Such differences among SETs enhance the need of a precise histological diagnosis. GISTs in a rectal location are related to worse outcomes compared to other sites [13]; this adds emphasis to the importance of precise and early diagnosis as in our reported case.

In this context, EUS is an important method that presents excellent diagnostic rates and may provide material for cytological and histological evaluation [4]. As previously described, GISTs are fourth layer tumors, which are hypoechoic and homogeneous. Differential diagnoses are leiomyoma and schwannoma [23]. Some echographic characteristics to differentiate GISTs from other fourth layer tumors have already been described in the literature. GISTs and schwannomas usually exhibit a complete or incomplete marginal hypoechoic halo, while leiomyomas do not show any clear marginal halo. Also, GIST echogenicities, in general, are low but slightly higher than that of the normal surrounding fourth muscle layer (muscularis propria), whereas the level of echogenicity in leiomyomas was nearly equal to that of the surrounding typical muscularis propria layer, and the echogenicity in schwannomas was extremely low. The difference in echogenicities among the fourth layer tumors (mesenchymal) might reflect the structural components and pathological differences of cellularity of these tumors [5].

Despite these characteristics, diagnostic accuracy for third and fourth layer tumors by EUS alone is around 50% [32]. A recent study proposed a scoring system for diagnosis of GISTs and other SETs based on EUS characteristics and found sensitivity and specificity to be 75% and 85%, respectively, for GISTs. However, this study only enrolled gastric SETs and had not been validated yet [33]. Therefore, a complementary biopsy is mandatory. Despite its high rates of inadequate material acquisition [6], FNA biopsy is still currently considered the procedure of choice for preoperative diagnosis of GISTs. The average diagnostic accuracy rate of EUS-guided FNA ranges from 60 to 80% in SETs [34]. Recent studies showed that Trucut biopsy is related to greater accuracy rates, and that should be preferred over FNA [35, 36]. Others studies propose dual biopsy (FNA and Trucut) to increase diagnostic yield [37]. However, there is no consensus yet, and hence FNA remains as the most performed procedure. To the best of our knowledge, no study assessed the sensitivity and specificity of the association of EUS features and Trucut or EUS-guided FNA.

Besides diagnostic yield, EUS also provides prognostic information: specific ultrasonography characteristics related to malignancy in GIST. Size larger than 20 mm, presence of cystic spaces, surface ulceration, irregular borders, and echogenic foci are unique features predictive of the malignant potential of GISTs and, therefore, worse outcomes [6, 38]. Among them, tumor size is the most important factor.

Other imaging methods may be useful for identification and to make the correct diagnosis of rectal GISTs. Because the rectum is an unusual location for a GIST, few studies assessed the role of MRI and a CT scan, but features such as large well-circumscribed masses and exophytic masses with moderate and heterogeneous enhancement on CT and MRI are associated with rectal GISTs. Furthermore, invasion of adjacent organs is uncommon, as well as regional lymph node enlargement, and small bowel obstruction [39]. If present, however, these must be identified and treated accordingly.

Finally, a combination of clinical, radiological, endoscopic, ultrasonographic, and cytological characteristics could eventually compose a scoring system, which would objectively enhance the preoperative diagnostic rates of GISTs and other SETs. Which would, therefore, help physicians to decide more accurately whether to, how to, and when to treat their patients.