Gastrointestinal Stromal Tumors (GISTs)

Gastrointestinal (GI) stromal tumors (GISTs) account for less than 1% of all GI tumors, yet are still the most common mesenchymal tumors of the GI tract.  The diagnosis of GIST is based on histologic examination. Further ancillary testing, including immunohistochemical staining and mutational analysis, provides confirmation of diagnosis, prognostic information, and delivers therapeutic implications.   Mutational assessment is important because of the availability of specific, molecular-targeted therapy with KIT/PDGFRA tyrosine kinase inhibitors (TKIs) (eg, imatinib, sunitinib).   Most GISTs (roughly 85%) contain oncogenic mutations in KIT or PDGFRA, whereas the remainder lack KIT and PDGFRA mutations and are commonly referred to as wild-type GISTs.  The National Comprehensive Cancer Center Network (NCCN) and European Society for Medical Oncology (ESMO) consider mutational analysis standard practice and strongly encourage including it in the diagnostic workup for all GISTS.  

Quick Answers for Clinicians

How are gastrointestinal stromal tumors diagnosed?

Suspected gastrointestinal (GI) stromal tumors (GISTs) usually present initially with GI bleeding (the most common manifestation), abdominal pain, and obstructive symptoms. On rare occasions, these tumors are found incidentally. Radiographic studies to further characterize the lesion should be performed, followed by endoscopic evaluation with biopsy. Histologic examination by a pathologist is the gold standard for the diagnosis of GIST. The pathologist will direct the diagnostic workup and perform ancillary testing, such as immunohistochemical staining for KIT (CD117) or DOG1, and KIT and PDGFRA mutational analysis.  

What is the role of mutational analysis in determining therapy for gastrointestinal stromal tumors?

KIT/PDGFRA mutational assessment is important for decision-making and treatment purposes, as gastrointestinal (GI) stromal tumors (GISTs) have a variety of subtypes with different predictive and prognostic relevance.     Current therapeutic strategies for GISTs revolve around targeted therapy against PDGFRA, KIT, and related receptor tyrosine kinases. Imatinib resistance, either primary or secondary, is associated with specific tumor genotypes, so genotyping of individual patient tumors helps guide decisions concerning whether to offer imatinib and at what dose, or whether an alternative tyrosine kinase inhibitor (TKI) (ie, sunitinib) is more appropriate.     Mutational profiling, for example, can identify mutations that are associated with imatinib resistance, such as PDGFRA D842V, wild-type GIST, and KIT exon 9 mutations that confer partial imatinib resistance; tumors with these mutations may respond to higher doses of imatinib.     Mutational status assessment can also aid in diagnosis when the immunohistochemical workup is inconclusive. 

Indications for Testing

Individuals with a GI tract mass suspected to be GIST should undergo biopsy and histologic examination. The results of immunohistochemical staining and mutational analysis may aid the clinician in determining the most appropriate course of action based on the most recent therapeutic guidelines put forth by the National Institutes of Health/National Cancer Institute and other major health organizations.   

Laboratory Testing

Diagnosis

Immunohistochemical Staining

The vast majority of GISTs (95%) overexpress the receptor tyrosine kinase KIT (CD117); this overexpression is the most prominent diagnostic marker for GIST and allows for differentiation of GIST from other subepithelial tumors of the GI tract, which are nearly universally negative for CD117.  However, KIT is not specific for GIST, and results should be carefully interpreted in light of the anatomic location and morphology of the tumor to distinguish GIST from other mesenchymal, neural, and neuroendocrine neoplasms.  

Staining for DOG1, also known as anoctamin-1/ANO1, a voltage-gated calcium-activated anion channel highly expressed in the interstitial cells of Cajal, may be useful for cases that cannot be categorized as GIST based on CD117 immunostaining.   DOG1 is expressed strongly in GIST and rarely expressed in other soft tissue tumors.   

SDHB immunohistochemical staining should be considered for any tumors that lack KIT or PDGFRA mutations.  SDHB staining is a helpful ancillary test that aids in identifying SDH-deficient GIST.  Recent reports have shown that SDH-deficient GISTs overexpress insulin-like growth factor 1 receptor (IGF1R), which may have future therapeutic implications in SDH-mutated GISTs.   

Therapy Planning

KIT and PDGFRA Mutational Assessment

KIT and PDGFRA mutational analysis is strongly recommended during the initial diagnostic evaluation and should always be performed before beginning treatment.   Approximately 80% of GISTs have a mutation in the gene that encodes the KIT receptor tyrosine kinase; another 5-10% of GISTs have a mutation in the gene that encodes the related PDGFRA receptor tyrosine kinase. 

The identification of GISTs with genotyping is important because of the availability of specific, molecular-targeted therapy with KIT or PDGFRA TKIs, such as imatinib or, in the case of imatinib-resistant GIST, sunitinib.   Similarly, mutations that confer secondary resistance to imatinib during treatment should be assessed in refractory or progressive disease. Multiple concurrent TKI-resistant mutations may be detected in a tumor.  Therapeutic drug monitoring for imatinib is available and may be performed to evaluate and optimize the treatment regimen. Roughly 90% of patients with a KIT exon 11 mutation respond to imatinib, whereas only ~50% of patients with a KIT exon 9 mutation respond (although in the latter case, response may improve with a higher dose of imatinib).  Most PDGFRA mutations described to date do not confer resistance to imatinib in various functional and clinical studies, with the exception of D842V, which is a well-described point mutation that confers resistance to imatinib.  

About 10-15% of GISTs lack mutations in the KIT or PDGFRA genes and are referred to as wild-type GISTs.  Wild-type GISTs are a heterogeneous group of GISTs that may have SDH, NF1, or BRAF (V600E) mutations.    Tumors that lack KIT or PDGFRA mutations should be considered for further mutational analysis, including BRAF and SDH gene mutation testing. 

SDH-deficient GISTs are often resistant to imatinib and may respond more favorably to second- and third-line therapies. Recent reports of IGF1R overexpression in SDH-deficient GISTs suggest the potential role for IGF1R and vascular endothelial growth factor receptor (VEGFR) inhibitor therapy. 

GIST Subgroups

GIST Subgroups Based on IHC and Mutational Analysis
Subgroups Description
KIT (CD117)-negative GIST

Approximately 5% of GISTs are negative for CD117 (KIT) expression

About 30% of these KIT-negative tumors harbor PDGFRA gene mutations, whereas more than half have KIT mutations

KIT-mutant GIST Approximately 80% of all GISTs contain a mutation in the KIT gene that results in constitutive activation of the protein
PDGFRA-mutant GIST

Roughly 5-8% of GISTs harbor a PDGFRA mutation

These GISTS differ from KIT-mutant GISTs in numerous ways, including predilection for stomach and variable expression of CD117

Wild-type GIST

12-15% of all GISTs are considered wild type and harbor no detectable KIT or PDGFRA mutations

Wild-type GISTs have no particular association with tumor location or clinical outcome

IHC, immunohistochemistry

Sources: NCCN, 2019 ; PDQ, 2019 

Familial Risk (Hereditary) Genetic Testing

Although the majority of GISTs appear to be sporadic, some patients, especially younger patients, have one of several associated syndromes, including neurofibromatosis type 1 (NF1), Carney-Stratakis syndrome (CSS), Carney triad (CT), and primary familial GIST syndrome. Genotyping can confirm these syndromes.  

Characteristics of Syndromes Associated with GIST
Clinical Syndromes Description
NF1

Patients with NF1 have an increased risk of GIST (most commonly in small bowel)

NF1-associated GISTs may present with multiple primary GISTs due to interstitial cells of Cajal hyperplasia; these tumors usually lack KIT and PDGRFA mutations

CSS

CSS is associated with familial paragangliomas and GISTs, and is autosomal dominant with incomplete penetrance

CSS GISTs occur due to a germline inactivating mutation in the SDH complex

CT

CT is associated with gastric GIST, paraganglioma, and pulmonary chondroma

CT GISTs lack mutations in KIT and PDGFRA and show evidence of epigenetic silencing of SDHC

CT is noninherited

Primary familial GIST syndrome

Primary familial GIST syndrome is autosomal dominant and is often caused by a germline KIT mutation and less often, a PDGFRA mutation

Primary familial GIST syndrome is marked by multiple GISTs throughout the GI tract (most commonly in the small bowel) and hyperpigmentation of skin (mast cell disorders)

ACS, American Cancer Society

Sources: Ricci, 2016 ; ACS, 2019 

Prognosis

GISTs may range from completely benign to aggressively malignant.  Prognosis and malignant potential of GISTs are partially assessed by size and mitotic rate of the tumor. Most gastric GISTs are indolent, and those that are <2 cm are considered to be of low risk, whereas micro-GISTs (<1 cm) are invariably benign.  However, it is difficult to predict the behavior of GISTs based on pathologic features alone; therefore, tumor site is also important to consider. GISTs in the small intestine or colorectum tend to be more aggressive than gastric GISTs. 

The mutational status of KIT and PDGFRA is not used to determine malignant potential. These mutations may be found in benign, indolent tumors and aggressive, high-grade tumors. 

For more information on prognosis and risk, refer to the most recent NCCN recommendations for proposed predictors of GIST biologic behavior. 

ARUP Laboratory Tests

Immunohistochemistry
Mutational Analysis (KIT and PDGFRA) to Predict Response to TKI Therapy
Therapeutic Drug Testing
Familial Risk (Hereditary) Genetic Testing

Medical Experts

Author
Author

References

Resources from the ARUP Institute for Clinical and Experimental Pathology®