1987The American Journal of Surgical PathologyRequires access

The Immunohistochemistry of Gastrointestinal Stromal Tumors

Scott H. Saul, Mark L. Rast, John J. Brooks

Open publisher page 99 citations

Abstract

To study the histogenesis of gastrointestinal stromal tumors, 79 cases were evaluated for desmin (DES), vimentin (VIM), and S-100 protein immunoreactivity by the avidin-biotin immunoperoxidase procedure on paraffin-embedded, Bouin's-fixed tissue sections. All tumors showed weak vimentin positivity. Trapped non-neoplastic smooth muscle and nerve twigs were often noted, particularly at the tumor periphery. Significant tumor S-100 positivity was not identified in our series. Similarly, glial fibrillary acidic protein (GFAP) immunoreactivity (performed in 11 desmin-negative tumors) was not detected within either gastrointestinal stromal tumors or enteric glial cells. Fifty-three percent (53%) of all gastrointestinal stromal tumors (GIST) displayed positive tumor cell desmin immunoreactivity. All 10 esophageal and all four colorectal tumors were diffusely desmin positive and unequivocal smooth muscle lesions. In contrast, only 17 of 37 (46%) benign and six of nine (67%) malignant gastric tumors were desmin positive. Similarly, four of 10 (40%) benign and one of nine (11%) malignant small-bowel tumors expressed desmin. Several gastric neoplasms with prominent nuclear palisading resembling schwannian Antoni A regions were nonetheless desmin positive. Epithelioid gastric tumors were more frequently desmin positive than nonepithelioid tumors; however, this positivity did not attain statistical significance. Two gastrointestinal stromal tumors that were desmin negative in paraffin-embedded material had detectable antigen in frozen sections. Gastric and small-bowel tumors measuring less than 3 cm were significantly more often desmin positive than those 3 cm or greater. We conclude that the method of fixation, tissue preparation, and immunostaining may significantly affect the expression of desmin. Although the histogenesis of gastrointestinal stromal tumors remains controversial, most of these tumors show evidence of smooth muscle differentiation.

About this research paper

What this paper is about

To study the histogenesis of gastrointestinal stromal tumors, 79 cases were evaluated for desmin (DES), vimentin (VIM), and S-100 protein immunoreactivity by the avidin-biotin immunoperoxidase procedure on paraffin-embedded, Bouin's-fixed tissue sections. All tumors showed weak vimentin positivity. Trapped non-neoplastic smooth muscle and nerve twigs were often noted, particularly at the tumor periphery. Significant tumor S-100 positivity was not identified in our series. Similarly, glial fibrillary acidic protein (GFAP) immunoreactivity (performed in 11 desmin-negative tumors) was not detected within either gastrointestinal stromal tumors or enteric glial cells. Fifty-three percent (53%) of all gastrointestinal stromal tumors (GIST) displayed positive tumor cell desmin immunoreactivity. All 10 esophageal and all four colorectal tumors were diffusely desmin positive and unequivocal smooth muscle lesions. In contrast, only 17 of 37 (46%) benign and six of nine (67%) malignant gastric tumors were desmin positive. Similarly, four of 10 (40%) benign and one of nine (11%) malignant small-bowel tumors expressed desmin. Several gastric neoplasms with prominent nuclear palisading resembling schwannian Antoni A regions were nonetheless desmin positive. Epithelioid gastric tumors were more frequently desmin positive than nonepithelioid tumors; however, this positivity did not attain statistical significance. Two gastrointestinal stromal tumors that were desmin negative in paraffin-embedded material had detectable antigen in frozen sections. Gastric and small-bowel tumors measuring less than 3 cm were significantly more often desmin positive than those 3 cm or greater. We conclude that the method of fixation, tissue preparation, and immunostaining may significantly affect the expression of desmin. Although the histogenesis of gastrointestinal stromal tumors remains controversial, most of these tumors show evidence of smooth muscle differentiation.

Why it matters

OpenAlex reports 99 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

To study the histogenesis of gastrointestinal stromal tumors, 79 cases were evaluated for desmin (DES), vimentin (VIM), and S-100 protein immunoreactivity by the avidin-biotin immunoperoxidase procedure on paraffin-embedded, Bouin's-fixed tissue sections. All tumors showed weak vimentin positivity. Trapped non-neoplastic smooth muscle and nerve twigs were often noted, particularly at the tumor periphery. Significant tumor S-100 positivity was not identified in our series. Similarly, glial fibrillary acidic protein (GFAP) immunoreactivity (performed in 11 desmin-negative tumors) was not detected within either gastrointestinal stromal tumors or enteric glial cells. Fifty-three percent (53%) of all gastrointestinal stromal tumors (GIST) displayed positive tumor cell desmin immunoreactivity. All 10 esophageal and all four colorectal tumors were diffusely desmin positive and unequivocal smooth muscle lesions. In contrast, only 17 of 37 (46%) benign and six of nine (67%) malignant gastric tumors were desmin positive. Similarly, four of 10 (40%) benign and one of nine (11%) malignant small-bowel tumors expressed desmin. Several gastric neoplasms with prominent nuclear palisading resembling schwannian Antoni A regions were nonetheless desmin positive. Epithelioid gastric tumors were more frequently desmin positive than nonepithelioid tumors; however, this positivity did not attain statistical significance. Two gastrointestinal stromal tumors that were desmin negative in paraffin-embedded material had detectable antigen in frozen sections. Gastric and small-bowel tumors measuring less than 3 cm were significantly more often desmin positive than those 3 cm or greater. We conclude that the method of fixation, tissue preparation, and immunostaining may significantly affect the expression of desmin. Although the histogenesis of gastrointestinal stromal tumors remains controversial, most of these tumors show evidence of smooth muscle differentiation.

Key concepts: Desmin, Pathology, Vimentin, Stromal cell, Immunohistochemistry, Immunoperoxidase, Biology, Glial fibrillary acidic protein

Related papers

Back to paper searchBrowse research topicsOriginal source
The Immunohistochemistry of Gastrointestinal Stromal Tumors — Research Paper | ScholarLens