Microglandular carcinoma of the pancreas: immunohistochemical and ultrastructural study of an unusual variant of pancreatic carcinoma that may closely resemble a neuroendocrine neoplasm.

Two cases are described of an unusual form of primary adenocarcinoma of the pancreas characterized histologically by their striking resemblance with a neuroendocrine neoplasm. The tumors were composed of a population of relatively small, uniform cells arranged in sheets admixed with small microglandular structures resulting in a cribriform pattern of growth. The tumor cells displayed scant cytoplasm with indistinct cell borders and round to oval nuclei with irregular clumping of chromatin and small, inconspicuous nucleoli. Immunohistochemical studies in both cases showed positivity of the neoplastic cells with CAM 5.2 antibodies and negative staining with a battery of neuroendocrine-related markers including chromogranin, NSE and synaptophysin, as well as with a variety of peptide hormones including insulin, glucagon, vasoactive intestinal polypeptide, gastrin and serotonin. Ultrastructural examination revealed a cohesive population of cells forming abortive glandular lumens lined by imperfectly formed microvilli and showing well-developed junctional complexes. No dense core neurosecretory granules or zymogen granules could be identified in any of the cells, supporting a ductal type of differentiation for these tumors. The main importance of recognizing this rare variant of pancreatic adenocarcinoma lies in avoiding misdiagnosis with other primary and metastatic neuroendocrine neoplasms of this organ. Immunohistochemical and ultrastructural examination will be of value in such cases for differential diagnosis.

Morphometric estimation of superoxide generation in allergen-induced airway hyperresponsiveness.

BACKGROUND: We previously reported that the airway hyperresponsiveness (AHR) that follows antigen challenge is mediated, in part, by the generation of reactive oxygen species. The purpose of this study was to provide quantitative morphologic evidence of oxygen radical production in the airways during antigen-induced AHR. EXPERIMENTAL DESIGN: Allergic sheep with a history of early and late bronchial responses to inhaled Ascaris suum antigen were used. The sheep were either challenged with antigen (N = 5) or saline (N = 5) and then euthanized 24 hours later when AHR had been documented to occur. Complete transverse sections of the right cranial lobar bronchus were obtained from the animals and were washed three times in Tris buffer and then incubated for 20 minutes in an oxygenated solution of diaminobenzidine which, in the presence of superoxide and manganese, forms an insoluble amber reaction product. Superoxide dismutase (SOD, 2250 units/ml) and phorbol ester (phorbol myristate acetate, 0.5 microgram/ml) were added to some tissues and used as negative and positive controls, respectively. The bronchial samples were then fixed and embedded in paraffin for light microscopy. The diaminobenzidine reaction product was quantified by determining the volume fraction of reaction product by point counting with a differential interference contrast microscope without counterstain. RESULTS: Diaminobenzidine reaction product increased 5-fold (p < 0.05) in challenged animals, and this response was blocked by SOD. The reaction product was localized in and around the airway epithelium. Antigen challenge also resulted in a 2.4- and 2.0-fold increase in eosinophils and metachromatic cells in the airway wall. There were no differences in the number of neutrophils between groups. Pretreatment of animals (N = 2) with the combination of the 5-lipoxygenase inhibitor (zileuton, 10 mg/kg, po) and the platelet-activating factor antagonist (WEB-2086, 3 mg/kg, i.v.) agents, which have been shown to block AHR and antigen-induced inflammation in the sheep model, also blocked the antigen-induced superoxide formation. CONCLUSIONS: These data suggest that superoxide and increased numbers of mediator-containing inflammatory cells are present in sheep airways 24 hours after antigen challenge. The interaction of this reactive oxygen species with these cells could contribute to the AHR seen at this time.

Gastrointestinal stromal tumors with prominent myxoid matrix. Clinicopathologic, immunohistochemical, and ultrastructural study of nine cases of a distinctive morphologic variant of myogenic stromal tumor.

Nine cases are presented of a distinctive morphologic variant of myogenic gastrointestinal stromal tumor characterized by an unusually prominent myxoid stromal background reminiscent of a neural neoplasm but lacking the immunohistochemical or ultrastructural features of peripheral nerve sheath or ganglionic differentiation. The patients included six women and three men aged 42 to 86 years (mean, 70). The lesions occurred in the stomach (seven cases) and small intestine (two cases) and ranged in size from 2.5 to 9.5 cm. They were described grossly as well circumscribed, unencapsulated, with a prominently myxoid and often cystic cut surface. Histologically, the lesions were composed of a proliferation of round, spindle, or stellate cells embedded in an abundant myxoid stroma. Histochemical stains showed strong positive reaction of the myxoid stromal background with alcian blue at pH 2.5; this staining reaction was abolished by treatment with hyaluronidase, indicating an abundance of connective tissue mucosubstances rich in hyaluronic acid. Immunohistochemical stains showed strong positivity of the tumor cells with vimentin antibodies in all cases and focal weak to moderate positive staining with muscle actin (HHF35) in eight cases and with desmin in two. Stains for keratin, S-100; epithelial membrane antigen, and collagen type IV were uniformly negative. Ultrastructural examination carried out in all cases showed features consistent with those previously described for myogenic gastrointestinal stromal tumors, namely, scattered mitochondria and prominent Golgi apparati, strands of rough endoplasmic reticulum, focal accumulation of intracytoplasmic microfilaments with occasional focal condensations, subplasmalemmal attachment plaques and immature cell junctions, focal extracellular basal lamina material, and surface-oriented micropinocytotic activity. The myxoid changes observed in these tumors may represent a secondary, nonspecific reaction pattern of the tumor cells to some noxious stimulus, or they may be a form of degenerative phenomenon such as that commonly observed in smooth-muscle tumors of the uterus and other sites. Myogenic gastrointestinal stromal tumors with prominent myxoid stroma should be distinguished from benign schwannoma of the stomach and gastrointestinal autonomic nerve tumors. Because of the differences in prognosis for these entities, immunohistochemical and ultrastructural examinations are recommended for the evaluation of gastrointestinal stromal neoplasms with prominent myxoid features.

Heparin prevents antigen-induced airway hyperresponsiveness: interference with IP3-mediated mast cell degranulation?

We hypothesized that heparin, because of its antiallergic and/or anti-inflammatory properties, modifies airway hyperresponsiveness (AHR). We studied the effects of inhaled heparin on AHR induced by specific antigen or by platelet-activating factor (PAF), a proinflammatory mediator. Specific lung resistance (sRL) was measured in 17 allergic sheep before, immediately after, and serially for up to 2 h after airway challenge with either specific antigen or PAF. Airway responsiveness was expressed as the cumulative provocative dose of carbachol that increased sRL to 4 cmH2O/s [PD4, in breath units (BU; 1 BU = 1 breath of 1% carbachol solution)]. PD4 was determined on a baseline day and on various experimental days 2 h after airway challenge with antigen or PAF, without or after pretreatment with inhaled heparin (1,000 U/kg). Pretreatment with inhaled heparin prevented antigen-induced bronchoconstriction and postantigen AHR. PD4 was 26 +/- 2.6 (SE), 12 +/- 1.7, and 22 +/- 2.8 BU on baseline, antigen control, and postheparin days, respectively. Heparin given immediately after the antigen challenge failed to modify the magnitude and/or duration of antigen-induced bronchoconstrictor response or postantigen AHR. Heparin also failed to prevent PAF-induced changes in sRL and AHR. In vitro heparin inhibited anti-immunoglobin E- and 1,4,5-inositol triphosphate-mediated degranulation of rat peritoneal mast cells without attenuating the effects of the Ca2+ ionophore A-23187. These data suggest that in "acute responders" heparin prevents antigen-induced bronchoconstriction and AHR, possibly by inhibiting 1,4,5-inositol triphosphate-dependent mast cell mediator release and not by its anti-inflammatory action.