Final Diagnosis -- Sarcoma, Suspect Recurrent Malignant Gastrointestinal Stromal Tumor




Gastrointestinal stromal tumors (GISTs) are mesenchymal neoplasms of the gastrointestinal tract that formerly were often designated "smooth muscle tumors". The etiology and risk factors remain unclear. GISTs occur predominantly in middle-aged and older persons, approximately 70% in stomach, 20-30% in the small intestine, less than 10% in the esophagus, colon and rectum. The most common symptoms include vague upper GI pain, bleeding or signs of obstruction (1).

Grossly, GISTs are usually circumscribed, solitary, round or ovoid masses with relatively pinkish-white appearance, often with areas of hemorrhage, necrosis, myxoid change, or cavitary degeneration.

Microscopically, GISTs show a wide spectrum of histological features, including spindle (70%) or epithelioid (30%) cells. The majority of GISTs are composed of spindle cells arranged in short fascicles in several architectural patterns including storiform, herringbone palisades and broad sheets. The spindle cells usually have blunt-ended, sometimes contracted, nuclei with some or more nucleoli occasional perinuclear vacuoles and abundant eosinophilic cytoplasm. The epithelioid patterns are less common and arranged in organoid clusters or sheets and have variable amounts of eosinophilic cytoplasm or clear cytoplasm as well as round nuclei with inconspicuous nucleoli. Mitotic activity is variable. Cytoplasmic clearing appears to be the result of a fixation (2, 3).

The precise cellular origin of GISTs recently has been proposed to be the interstitial cell of Cajal, GI pacemaker cells (4). These cells are intercalated between the autonomic nerves and smooth-muscle cells and control the peristaltic action of GI tract. By immunohistochemical analysis, normal Cajal cells are positive for the c-kit protooncogene, which encodes for a transmembrane tyrosine-kinase receptor (CD117). This postulate is supported by the finding that GISTs have cell markers similar to those of the normal Cajal cell. Furthermore, the Cajal cell has characteristics of both smooth muscle and neural differentiation on the ultrastructural examination. Thus, neoplastic Cajal cells could preferentially express one, both, or neither of these features, accounting for the variants of GIST (5).

CD34, the hematopoietic progenitor cell antigen, is a transmembrane protein of unknown function. Most GISTs (70-80%) including benign and malignant GISTs are positive for CD34. The malignant GISTs may show a slightly lower frequency of CD34 expression than the benign ones (5, 6).

Smooth muscle actin, is expressed by the normal GI-smooth muscle and vascular smooth muscle cells with 20-40% frequency of positivity in GISTs and the reactivity is often focal. In addition, GISTs are generally negative for S100-protein (5).

In summary, the consistent expression of CD117 (c-kit protein) is the best defining feature of GISTs and is seen in nearly all cases. The CD117-positivity is seen in all histological variants and in benign and malignant GISTs at different sites.

The prediction of malignant potential of GISTs is often difficult based on the primary site histopathologic features alone. Although the tumor size, cellularity, necrosis, mitotic count are the most important prognostic variables, current criteria for diagnosing malignancy sometimes fail to predict the outcome, because even benign appearing tumor will metastasize sometimes. The concept developed newly is that integration of gene analysis into existing methods for pathologic evaluation (7). Preliminary studies by our group have shown that greater malignancy and poorer survival could be predicted by determining the extent of accumulated mutational damage of primary GISTs. Histologic sections of tumor are microdissected and tested for a panel of multiple genomic sites of loss of heterozygosity (LOH). The microdissection mutational profile is correlated with metastatic spread and outcome. Aggressive diseases with early metastasis and short survival are correlated with high acquired fractional mutation rate. It has been proposed that accumulated chromosomal changes may contribute to the progression and/or malignant transformation. So the molecular analysis is a useful prognostic tool to supplement histopathology.


  1. Sternberg SS, Antonioli DA, Carter D, Mills SE and Oberman HA. Diagnostic surgical pathology. 3rd edition. 1999; Vol 2:pp1337-1339 &1454-1457
  2. Rosai J. Ackerman's surgical pathology. 8th edition. 1996; Vol 1: pp1645-647
  3. Lewin KL, Appleman HD. Tumors of the esophagus and stomach. Atlas of tumor pathology, 3rd Series, Fascicle 8. Washington D.C.: AFIP1996. pp 405-435
  4. Kindblom LG, Remotti HE, Aldenberg F, Meis-Kindblom JM. Gastrointestinal pacemaker cell tumors (GIPACT): gastrointestinal stromal tumors show phenotypic characteristics of the interstital cells of Cajal. Am J Path 1998; 152:1259-1269.
  5. Miettinen M, Sarlomo-Rikala M, Lasota J. Gastrointestinal stromal tumors: Recent advances in understanding of their biology. Hum Path 1999; 30(10):1213-1220
  6. Van de Rijn M, Hendrickson MR, Rouse RV: The CD34 expression in by gastrointestinal stromal tumors. Hum Path 1994; 25:766-771.
  7. Finkelstein SD, Przygodzki R, Swalsky PA. Microdissection-based p53 genotyping: Concepts for molecular testing. Molecular Diagnosis 1998; 3(3): 179-191

Contributed by Xiaoyan Wang, MD, PhD and Sydney Finkelstein, MD


IndexCME Case StudiesFeedbackHome