Gastrointestinal autonomic nerve tumors: A clinicopathological, immunohistochemical and ultrastructural study of 10 cases

Gastrointestinal Autonomic Nerve Tumors (GANTs) are an underrecognized group of gastrointestinal stromal tumors (GISTs) putatively arising from the neural plexuses of the bower wall. Approximately 24 cases have been previously reported. Their histogenesis, malignant potential, morphology and phenotypic features are not well defined. We present details of 10 GANTs iterating features, predominantly ultrastructural, allowing distinction from other GISTs. Clinical details are: sex-7M, 3F; age range 31-79 yrs, mean 53; symptoms/signs - abdominal pain 3, GI bleeding 3, mass 2, anemia 2. Follow-up ranged from 1-102 mths, mean 29. Seven tumors involved the smalt intestine and 3 were gastric. Tumor size ranged from 30-160 mm, mean 79. They were solid and cystic, often transmural and usually involved mesentery and retroperitoneum. Spindled and epithelioid cells were ''compartmentalized'' by a branching microvasculature. Eosinophilic, PAS positive stromal globules were prominent. Paraffin immunostaining results were (number positive/total): vimentin (8/9), NSE (10/10), S100 protein (6/10), neurofilament protein (0/9), synaptophysin (3/9), desmin (2/9, focal), smooth-muscle actin (0/9). Ultrastructural diagnostic features were elaborate, branching cytoplasmic processes containing microtubules, intermediate filaments and varying numbers of neurosecretory granules. Characteristic features were elaborate smooth endoplasmic reticulum enmeshed with intermediate filaments, pleomorphic mitochondria with lamellar cristae, mitochondrial-RER complexes, confronting RER cisternae, and circumscribed collections of stromal ''skeinoid'' fibres. There were no features of smooth muscle, Schwannian or perineurial differentiation. We conclude that: 1) GANTs can be distinguished from other GISTs, electron microscopy being essential for diagnosis: 2) biological behaviour and predictors must await study of further cases with longer follow-up; our findings indicate less aggressive behaviour than that described previously; and 3) immunohistochemistry on fixed material is of limited value without ultrastructural assessment; focal desmin positivity is not indicative of smooth muscle differentiation without other supporting evidence.