Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Gastrointestinal Stromal Tumors
Epidemiology
Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal, or nonepithelial, neoplasms of the gastrointestinal (GI) tract. GISTs may be found anywhere in the GI tract, from the esophagus to the internal anal sphincter. The most common GI location is the stomach (56%), as reported in a 2015 review of multiple population-based studies, followed by the small intestine (32%), colon and rectum (6%), and esophagus (<1%). The remaining 5% of lesions occur in other less common locations, including the mesentery, pelvis, pancreas, liver, omentum, and genitourinary tract. Occurrence of these extragastrointestinal lesions, or EGISTs, is described predominantly in case reports and small series.
GISTs were historically thought to be of smooth muscle origin and therefore categorized as leiomyomas or leiomyosarcomas. Prior to 1983, when the term stromal tumor was first used, they remained largely unrecognized as a separate class of neoplasms. The development of specific pathologic criteria has allowed for more accurate categorization of mesenchymal tumors and reclassification of many of these tumors as GISTs. The exact incidence of GISTs was initially unclear because many studies included data from the era prior to their recognition as a distinct disease entity. In a review of population-based studies including more recent data from 19 countries, the incidence ranged from 7 to 15 cases per million people in most studies. A study published in 2015 based on Surveillance, Epidemiology, and End Results (SEER) data estimated the incidence to be approximately 6.8 cases per million, or approximately 2100 total cases in the United States. This study also reported 5-year overall and disease-specific survivals of 65% and 79%, respectively. The true incidence of GISTs may be underreported because SEER requires reporting of these tumors only if they are classified as malignant or metastatic. As the recognition of these tumors has grown, several groups are in the process of developing databases to better estimate the true incidence of GISTs.
Pathology and Diagnosis
Histology
GISTs are diagnosed based on morphologic features supported by immunohistochemical (IHC) studies. Histologically, GISTs can occur in three different types: spindle cell, which accounts for approximately 70% of tumors, epithelioid, and mixed, which contains a variable combination of spindle cells and epithelioid cells ( Fig. 81.1 ). Rarely, GISTs may have myxoid stroma, neuroendocrine features, signet ring variant, or marked lymphocytic infiltrate. The pathologic distinction may still be difficult to conclusively define in individual cases, with fibromatosis and leiomyosarcomas still being mistaken for GISTs.
Identification of Cells of Origin of Gastrointestinal Stromal Tumors
In 1983 the term stromal tumor was first used to describe a series of gastric tumors that appeared to arise not from smooth muscle but rather from nerve sheath cells that did not resemble typical Schwann cells. This initial distinction was based on IHC and electron microscopic analysis that demonstrated lack of desmin, positivity for the s100 protein, and lack of ultrastructural features typical of smooth muscle or Schwann cells.
Numerous studies were subsequently performed in an attempt to identify the cells of origin of these tumors. In the early 1990s it was found that the majority of (but not all) GISTs were positive for the myeloid progenitor cell antigen CD34, which is not typically seen in leiomyomas or Schwannomas. The identification of IHC staining for the CD117 protein, which is the tyrosine kinase receptor known as KIT , was a critical step in better classifying these tumors. This led to the hypothesis that GISTs arise from the interstitial cells of Cajal (ICC), which also express both CD34 and CD117.
The ICC were identified in 1893 by Santiago Ramon y Cajal, who aptly named them “interstitial” cells given their interposition between nerve endings and smooth muscle cells in the gut wall. These cells have very few contractile elements (unlike smooth muscle cells) and have a greater proportion of mitochondria. They arise from mesenchymal cells and are thought to play an integral role in the propagation of intrinsic slow-wave gut peristalsis. ICC, although primarily known for their role in gut motility, are also found outside of the GI tract, including the genitourinary system, portal vein, and pancreas. The higher frequency of tumors confined to the GI tract likely represents the increased proportion of ICC in the GI tract because no clear histologic distinction has been elucidated among the ICC of various locations.
Molecular Mechanism of Carcinogenesis
After the IHC and phenotypic distinction of GISTs were established, significant attention was directed to the molecular mechanism of carcinogenesis. The basic underlying pathophysiology in the majority of lesions appears to involve a gain of function mutation leading to increased activity of the KIT tyrosine kinase receptor on the cell membrane. Normal physiology consists of a monomeric tyrosine kinase receptor that binds to an extracellular ligand, stem cell factor (SCF), which then causes dimerization of the receptor. This dimerization allows for subsequent autophosphorylation, activating downstream intracellular signaling pathways, and leads to increased cell proliferation. Some lesions may also develop due to gain of function mutations in platelet-derived growth factor receptor A (PDGFRA), another tyrosine kinase receptor, or BRAF, a protooncogene serine/threonine kinase, each of which may contribute to the development of approximately 10% of GISTs.
Immunohistochemical Staining
CD117 IHC staining has now been widely accepted as the primary criterion for a pathologic diagnosis of GIST, with a reported sensitivity greater than 95% and a high specificity. PDGFRA staining has also been identified as a marker of GIST, with a frequency of up to 15%. PDGFRA may be used in combination with c-Kit or as a marker for GISTs that lack the c-Kit mutation. An additional IHC marker, DOG1, which stands for discovered on GIST, also has utility in the diagnosis of tumors that lack c-Kit or PDGFRA mutations. DOG1 has an overall sensitivity of 95%, similar to Kit. There is growing evidence of a significant correlation between immunophenotype and histology, as well as other features of GISTs.
Clinical Presentation of Gastrointestinal Stromal Tumors
GISTs are often asymptomatic and discovered only as incidental findings on imaging studies or endoscopy performed for other indications. Patients with a symptomatic GIST often have symptoms that are related to the size and location of the tumor. Early in disease progression, symptoms may be very nonspecific, such as mild pain, bloating, or dyspepsia. More significant symptoms generally do not occur until much later in the course of the disease, when they have grown quite large. This is likely due to their submucosal location, exophytic growth, and propensity to displace rather than invade adjacent organs. Patients presenting with large GISTs may have palpable tumors and present with pressure or pain or symptoms related to compression of adjacent organs. Patients may also present with acute GI blood loss or symptoms of chronic anemia and associated fatigue resulting from overlying mucosal ulceration, or with peritonitis related to tumor perforation.
Imaging and Diagnosis
The increased awareness of GISTs has improved preoperative diagnosis. Suspected GISTs should be evaluated with contrast-enhanced computed tomography (CT) and/or magnetic resonance imaging (MRI). Typical findings are variable, depending on the size and aggressiveness of the lesion. Large GISTs appear as hypervascular, enhancing masses and are often heterogeneous because of necrosis, hemorrhage, or cystic degeneration. They often displace but rarely invade adjacent organs. Smaller GISTs are usually more homogeneous and polypoid in appearance.
Many GISTs are also identified during endoscopy for evaluation of upper GI symptoms. A submucosal mass ( Fig. 81.2 ) may be seen that is smooth in appearance and often appears to form a bulge into the lumen of the stomach ( Fig. 81.3 ). Ulceration may be seen in lesions that present with bleeding or anemia. Cross-sectional imaging should be performed on all patients with suspected GIST identified endoscopically because endoscopic findings may underestimate the full extent of disease.
Biopsy of a suspected GIST is not required if it appears surgically resectable. However, if neoadjuvant or palliative therapy is being considered for large, unresectable, or metastatic tumors, a tissue diagnosis is needed. Endoscopic biopsies are preferable, if possible, to percutaneous techniques because of the risk of bleeding and tumor rupture.
Endoscopic ultrasound (EUS) may also aid in the differentiation of a submucosal gastric mass versus impingement from surrounding organs (i.e., pancreatic mass, pseudocyst). EUS is also useful in biopsy of these lesions due to their submucosal location. EUS is particularly important in the management of very small (<2 cm) gastric GISTs. In these patients, EUS is indicated to evaluate for high-risk features. These include irregular borders, cystic spaces, ulceration, echogenic foci, and internal heterogeneity. EUS–fine-needle aspiration may be performed on these lesions to confirm the diagnosis of GIST.
Chest imaging should be considered for staging of patients with GISTs. Positron emission tomography (PET) scans may also have utility in preoperative staging. PET/CT scans have already been shown to be useful in the early determination of the response of GISTs to systemic therapy. Ring-shaped uptake on preoperative PET scan has recently been shown to be an independent adverse prognostic factor for postoperative recurrence.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here