Gastrointestinal Stromal Tumors - GIST

Primary Author Wallander, Michelle L., PhD.

Key Points

The advent of tyrosine kinase inhibitor (TKI) therapy for the treatment of gastrointestinal stromal tumors (GISTs) makes it imperative to distinguish GISTs from histologic mimics (eg, leiomyoma, leiomyosarcoma, schwannoma, high-grade sarcoma, and desmoid fibromatosis).

Immunohistochemistry (IHC) staining for KIT (CD117) identifies most GISTs; the remainder are frequently identified by DOG1 staining. Mutational analysis is most helpful if TKIs are considered for unresectable or metastatic disease, in tumors that test negative for CD117 or DOG1 by IHC, or to identify patients who will likely demonstrate TKI resistance (National Comprehensive Cancer Network [NCCN], 2014). The vast majority (90-95%) of GISTs have a gene mutation (primarily in the KIT gene). PDGFRA and KIT gene mutations are mutually exclusive and cause ligand-independent activation of signal transduction pathways. TKI therapy competitively inhibits the ATP-binding pocket at the catalytic binding site of tyrosine kinase.

Immunohistochemistry

CD117

ARUP Test 

CD117 (c-Kit) by Immunohistochemistry 2003806

Occurrence

  • 90-95% of GISTs
  • Excellent screen for KIT mutation

Characteristics

  • Strong staining with diffuse pattern in cytoplasm is typical – may also be dot-like, perinuclear, or membranous
    • Staining intensity does not correlate with treatment sensitivity or mutational status
  • Other tumors (eg, melanoma, synovial sarcoma) may also stain positive, so histology should be correlated with IHC

Limitations

  • Does not identify type of mutation – crucial for predicting responsiveness to TKI therapy

DOG1 (ANO1)

ARUP Test

DOG1 by Immunohistochemistry 2010168

Occurrence

  • >75% of GISTs

Characteristics

  • Most sensitive in spindle cell subtypes
  • Most useful in KIT-negative tumors – likely to harbor a PDGFRA mutation
    • Staining does not correlate with mutational status

Limitations

  • Does not identify type of PDGFRA gene mutation – crucial for predicting responsiveness to TKI therapy 
Other markers (may be helpful in questionable histology)

CD34

ARUP Test

CD34, QBEnd/10 by Immunohistochemistry 2003556

Occurrence

  • 60-70% of GISTs

Smooth Muscle Actin

ARUP Test

Smooth Muscle Actin (SMA) by Immunohistochemistry 2004130

Occurrence

  • 30-40% of GISTs

Characteristics

  • Weak pattern with focal staining
Molecular Mutations in KIT and PDGFRA

KIT gene

ARUP Test

Gastrointestinal Stromal Tumor Mutation 2002674

Characteristics

  • Mutations cluster most commonly on exon 11 – less commonly on exons 9, 13, 14, 17, and 18
    • D816V mutation on exon 17 – rarely detected
  • Acquired mutations occur during TKI treatment on exons 13,14, and 17
    • Mutations decrease binding capacity of TKIs

Therapeutic implications

  • Wild-type KIT GISTs
    • More responsive to sunitinib
  • Exon 9 mutation
    • Requires escalated dose of TKI for response
    • Better response to sunitinib than imatinib
  • Exon 11 mutation
    • Associated with TKI sensitivity
  • Exon 13 mutation
    • Primary (non-therapy associated) – associated with TKI sensitivity
    • Secondary (acquired during therapy) – associated with TKI resistance
  • Exon 14 mutation
    • Secondary (acquired during therapy) – associated with TKI resistance
  • Exon 17 mutation
    • Primary – associated with TKI sensitivity
    • D816V – associated with TKI resistance
    • Secondary – associated with TKI resistance
  • Exon 18 mutation
    • Rare
    • Secondary (acquired during therapy) – associated with TKI resistance

PDGFRA gene

ARUP Test

Gastrointestinal Stromal Tumor Mutation 2002674

Characteristics

  • Mutations cluster on exons 12, 14, and 18
    • Exon 18 – most common
      • D842V and D846V mutations

Therapeutic implications

  • Wild-type PDGFRA
    • Associated with TKI resistance
    • More responsive to sunitinib
  • Exons  12, 14
    • Associated with TKI sensitivity
  • Exon 18
    • Associated with TKI resistance

Diagnosis

Indications for Testing

  • Patient with gastrointestinal symptoms (satiety, abdominal discomfort due to pain or swelling, intraperitoneal hemorrhage) and suspicious mass on endoscopy or scanning

Laboratory Testing

  • CBC – may demonstrate anemia
  • Liver function tests

Histology

  • GISTs are soft and fragile tumors; biopsy may cause tumor hemorrhage and possible increased risk for tumor dissemination
    • Consideration of biopsy should be based on extent of disease and suspicion of a given histologic subtype
  • Immunohistochemistry and molecular testing
    • See Key Points
    • Tumors lacking KIT or PDGFRA gene mutations should be considered for the following
      • SDHB IHC, BRAF mutation, or SDH gene mutation
        • BRAF V600 – usually exon 15

Imaging Studies

  • Tumor is often discovered incidentally on imaging
  • Ultrasound – endoscopic ultrasound demonstrates hypoechoic mass that is contiguous with the muscularis propria
    • High-risk features include irregular border, cystic spaces, ulceration, echogenic foci, and heterogeneity
  • MRI and/or abdominal/pelvic CT with contrast – demonstrates mass and helps define extension of tumor
  • PET scan – may help differentiate active tumor from necrotic or inactive scar tissue, malignant from benign tissue, and recurrent tumor from nondescript benign changes
    • PET is not a substitute for CT but may clarify ambiguous CT or MRI findings
  • Percutaneous image-guided biopsy may be appropriate for confirmations of metastatic disease

Prognosis

Differential Diagnosis

  • Desmoid fibromatosis
  • Leiomyoma
  • Schwannoma
  • Leiomyosarcoma
  • Neurofibroma
  • Inflammatory fibroid polyps
  • Inflammatory myofibroblastic tumors
  • Ischemic bowel
  • Other gastrointestinal cancer – colorectal, gastric, pancreatic
  • Solitary fibrous tumor

Clinical Background

Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal tumors of the gastrointestinal (GI) tract, representing ~20% of all sarcomas. These tumors were historically identified as leiomyomas, leiomyosarcomas, leiomyoblastomas, and peripheral nerve sheath tumors.

Epidemiology

  • Incidence – ~7/million annually in the U.S.
  • Age – median is 60-65 years
    • Rarely <21 years – frequently associated with a syndrome
  • Sex – M:F, equal

Inheritance

  • Inherited tumor syndromes are uncommon
  • Carney triad – usually associated with loss of expression of SDHB by immunohistochemistry (IHC)
    • GIST, paraganglioma, and pulmonary chondroma
    • More common in women
    • Multifocal disease within stomach most common
    • Rare
    • Shares features similar to pediatric GIST – lacks defined mechanism of KIT gene activation
  • Carney dyad (Carney-Stratakis syndrome) – usually associated with loss of expression of SDHB by IHC
    • GIST and paraganglioma
  • Familial GIST (common)
    • Median age of onset is 47 years
    • 90% chance of GIST diagnosis by age 70
  • Neurofibromatosis type 1 – strongly KIT-positive by IHC but wild-type KIT genetically
    • Small percentage will develop GIST

Pathophysiology

  • Tumor originates from the interstitial cells of Cajal – pacemaker cells that regulate peristalsis in the GI tract
    • Classified as spindle cell (70%), epithelioid cell (20%), and occasionally mixed tumors of the GI tract that express the KIT gene proteins CD117 or CD34 on IHC
    • Mutations involve KIT or PDGFRA genes – both on chromosome 4q12
      • ~80% of GISTs have mutation in the gene encoding the KIT receptor tyrosine kinase
      • 5-10% of GISTs have mutation in the gene encoding the related PDGFRA receptor tyrosine kinase
      • 10-15% of GISTs have no detectable KIT or PDGFRA mutation; however, absence of mutation does not rule out diagnosis of GIST
      • Small number of wild-type GISTs (lacking either of above genes) have SDH or BRAF V600E mutations
  • Variable malignant potential from low to highly aggressive
  • Most common sites are stomach (>50%) and small intestine (30%)
  • Tumors usually involve the outer muscular layer; growth tends to be exophytic
  • Rarely found outside the GI tract
    • Known as extragastrointestinal stromal tumors (EGIST)
    • Sites include uterus, vagina, mesentery, omentum, and retroperitoneum
  • Metastasis
    • Rarely, lymph node,  liver, lung, or bone

Clinical Presentation

  • 30% are asymptomatic – usually small tumors (<2 cm)
  • Most common symptom is GI bleeding due to mucosal ulceration
  • Gastric GIST – nausea, emesis, weight loss, abdominal discomfort (60% of cases)
  • Small bowel GIST – melena, abdominal pain (30% of cases)
  • Colorectal GIST – change in bowel habits, hematochezia, abdominal pain, and distention (~10% of cases)
  • Esophageal GIST – odynophagia, dysphagia, retrosternal chest pain, and hematemesis (<1% of cases)
  • Carney triad – paraganglioma, pulmonary chondroma, and GIST
    • Indolent course with high rate of recurrence

Pediatrics

Clinical Background

Epidemiology

  • Prevalence – 1-2% of GISTs
  • Age – 10-20 years
  • Sex – M<F (marked)

Pathophysiology

  • Fundamentally different clinicopathologic entity from adult GISTs – majority have SDH gene mutations
  • Most tumors are in the stomach or small intestine
  • Predominant epithelioid morphology
  • Tumors often spread to liver and peritoneum – neither feature necessarily worsens prognosis
  • 90% of pediatric GISTs lack KIT or PDGFRA gene mutations; tyrosine kinase inhibitors (TKIs) are generally less effective

Clinical Presentation

  • Pediatric GISTs are generally more indolent than adult type
  • Abdominal symptoms – nausea, emesis, abdominal pain, and gastrointestinal bleeding
  • Fatigue, pallor, and weakness – due to anemia
  • May be associated with pulmonary chondromas or paragangliomas (Carney triad)

Treatment

  • Treatment algorithms for adults do not apply

Diagnosis

Indications for Testing

  • Patient with gastrointestinal symptoms and suspicious mass on endoscopy or scanning

Laboratory Testing

  • CBC – may demonstrate anemia
  • Liver function tests

Histology

  • Pathologic criteria for predicting malignancy (eg, size, mitotic activity) do not apply in pediatric GISTs
  • IHC – stain for KIT immunoreactivity and SDH mutation, which can be screened for by SDH IHC (usually has loss of expression of SDHA or SDHB)
  • Tissue –epithelioid histology most common; often has low-grade histologic features
  • Mutation analysis – required for all pediatric GISTs, especially those in young adults
    • Presence of KIT or PDGFRA gene mutations supports the diagnosis of GIST and aids in the prediction of response to imatinib
      • KIT and PDGFRA mutations are uncommon in pediatric patients
  • Lymph node metastasis – not common in contrast to adults with GISTs

Imaging Studies

  • Refer to Diagnosis tab

Indications for Laboratory Testing

  • Tests generally appear in the order most useful for common clinical situations
  • Click on number for test-specific information in the ARUP Laboratory Test Directory
Test Name and Number Recommended Use Limitations Follow Up
CD117 (c-Kit) by Immunohistochemistry 2003806
Method: Immunohistochemistry
Initial screening test in tumor that is morphologically and clinically suspicious for GIST

Not specific for GIST; may also be found in melanoma, angiosarcoma, and Ewing sarcoma

Molecular testing required to confirm KIT mutations

DOG1 by Immunohistochemistry 2010168
Method: Immunohistochemistry

Screening test in tumor that is morphologically and clinically suspicious for GIST when CD117 is negative

DOG1-positive and CD117-negative IHC stains – predictive for PDGFRA mutation

Does not identify type of mutation – crucial for predicting responsiveness to TKI therapy

Molecular testing for PDGFRA to confirm mutation

Gastrointestinal Stromal Tumor Mutation 2002674
Method: Polymerase Chain Reaction/Sequencing

Detects activating mutations in KIT and PDGFRA genes to predict response to TKI therapy

KIT gene exons tested – 9, 11, 13, 14, 17, 18

PDGFRA gene exons tested – 12, 14, 18

Clinical sensitivity – mutations detected in >85% of GISTs (70% KIT and 15% PDGFRA)

Analytical sensitivity – 25% mutant alleles (50% tumor)

Mutations outside of targeted exons are not detected

Test alone cannot be used for diagnosis of malignancy

 
CD34, QBEnd/10 by Immunohistochemistry 2003556
Method: Immunohistochemistry
Aid in histologic diagnosis of GIST    
Caldesmon by Immunohistochemistry 2003484
Method: Immunohistochemistry

Aid in histologic diagnosis of GIST

Stained and returned to client pathologist; consultation available if needed

   
Smooth Muscle Actin (SMA) by Immunohistochemistry 2004130
Method: Immunohistochemistry
Aid in histologic diagnosis of GIST    
Desmin by Immunohistochemistry 2003863
Method: Immunohistochemistry

Aid in histologic diagnosis of GIST

Use to differentiate GISTs from smooth muscle tumors

   
S-100 Protein by Immunohistochemistry 2004127
Method: Immunohistochemistry

Aid in histologic diagnosis of GIST

Use to differentiate GIST from malignant melanoma and schwannoma

   
Glial Fibrillary Acidic Protein (GFAP) by Immunohistochemistry 2003899
Method: Immunohistochemistry

Aid in histologic diagnosis of GIST

Stained and returned to client pathologist; consultation available if needed

   
Beta-Catenin-1 by Immunohistochemistry 2003454
Method: Immunohistochemistry

Aid in histologic diagnosis of GIST

Differentiate GIST from desmoid-type fibromatosis

Stained and returned to client pathologist; consultation available if needed

   
Gastrointestinal Hereditary Cancer Panel, Sequencing and Deletion/Duplication, 15 Genes  2010198
Method: Massive Parallel Sequencing/Exonic Oligonucleotide-based CGH Microarray

Confirm a diagnosis of hereditary GI cancer in individuals with a personal or family history of GI cancer and/or polyposis

Preferred test for individuals with suspected hereditary GI cancer syndrome

Genes included – APC, BMPR1A, CDH1, EPCAM, MLH1, MSH2, MSH6, MUTYH, PTEN, SDHB, SDHC, SDHD, SMAD4, STK11, TP53

Analytical sensitivity/specificity – 99%

Diagnostic errors can occur due to rare sequence variations

Not determined or evaluated:

  • Mutations in genes not included on the panel
  • Deep intronic and regulatory region mutations
  • Breakpoints for large deletions/duplications
  • PMS2 gene (associated with Lynch syndrome) is not included on this panel
  • Sequence changes in EPCAM will not be evaluated
  • Deletions/duplications may not be detected in
    • Exon 1 in CDH1 and MSH2 genes
    • Exons 4, 6, and 7 in STK11 gene
    • Exon 8 in PTEN gene
    • Exon 9 in BMPR1A gene

Individuals with hematological malignancy and/or a previous allogenic bone marrow transplant should not undergo molecular genetic testing on peripheral blood specimen

  • Testing of cultured fibroblasts or buccal specimen is required for accurate interpretation of test results

Lack of a detectable gene mutation does not exclude a diagnosis of hereditary GI cancer syndrome

Not all predisposing genes are analyzed

 
Additional Tests Available
 
Click the plus sign to expand the table of additional tests.
Test Name and NumberComments
CBC with Platelet Count and Automated Differential 0040003
Method: Automated Cell Count/Differential

Use to determine presence of anemia

Hepatic Function Panel 0020416
Method: Quantitative Enzymatic/Quantitative Spectrophotometry

Use to monitor liver function

Solid Tumor Mutation Panel by Next Generation Sequencing 2007991
Method: Massively Parallel Sequencing

Prognosis/treatment of individuals with solid tumor cancers at initial diagnosis or with refractory disease

SDHB with Interpretation by Immunohistochemistry 2006948
Method: Immunohistochemistry

Consider for cases of PDGRA and KIT negative GIST

BRAF Codon 600 Mutation Detection by Pyrosequencing 2002498
Method: Polymerase Chain Reaction/Pyrosequencing

Consider for cases of PDGRA and KIT negative GIST

Hereditary Paraganglioma-Pheochromocytoma (SDHB, SDHC, and SDHD) Sequencing and Deletion/Duplication Panel 2007167
Method: Polymerase Chain Reaction/Sequencing/Multiplex Ligation-dependent Probe Amplification

Consider for cases of PDGRA and KIT negative GIST