Gastrointestinal stromal tumors: an ultrastructural study

Gastrointestinal stromal tumors (GISTs) represent an enigmatic group of lesions of uncertain phenotype and biologic potential. Although earlier studies suggested smooth muscle cells, schwann cells, or neuronal differentiation, more recent evidence indicates that these tumors show phenotypic features that are similar to the interstitial cells of Cajal. Recently, investigators have begun to evaluate these lesions in a site-specific manner and have found that, in addition to morphologic differences between them, their biologic behavior also appears to be linked to their anatomic location. Many of these studies have emphasized the histologic and immunophenotypic features of GISTs in relation to their sites of origin, however, their site-specific ultrastructural characteristics have received little attention in the literature. In this study, we evaluated 34 GISTs (15 gastric, 12 small intestinal, 4 colonic, and 3 omental) for a variety of ultrastructural features in an effort to identify site-specific similarities and differences. Tumors predominantly composed of epithelioid cells were more commonly seen in gastric (60%) and omental (67%) tumors than in those of the small intestine (33%) and colon (0%). Cytoplasmic filaments and intercellular junctions were commonly seen in tumors from all locations, the filaments frequently forming paranuclear aggregates in the epithelioid cells. Tumors from all sites were composed of cells with surface filopodia and interdigitating cell processes, but in tumors of the stomach and omentum the filopodia were usually short and minimally intertwined, whereas those of small and large intestinal GISTs were characteristically long and complex. Basal lamina, though poorly formed, was present only in tumors of gastric and omental origin (13% and 67%, respectively). Pinocytotic vesicles were also seen in tumors from these sites (33% of gastric tumors and 67% of omental lesions) as well as those of the small intestine (17%) and the colon (25%). None of the gastric or omental tumors had microtubules; they were, however, seen in small intestinal (33%) and colonic (25%) stromal tumors. Skenoid fibers were seen in 33% of small intestinal GISTs and 1 metastatic gastric GIST. Overall, gastric and omental tumors have better developed features of myogenic differentiation and have blunt filopodia and minimally intertwined cell processes. Indeed, these 2 groups are indistinguishable ultrastructurally, raising the possibility that the genesis of omental GISTs is similar to that of gastric stromal tumors. Small intestinal stromal tumors have characteristic interdigitating cell processes and numerous elongate filopodia-like structures harboring intercellular junctions as well as microtubules and extracellular skenoid fibers. The constituent cells in colonic stromal tumors, while more reminiscent of small intestinal stromal, were frequently more primitive in appearance. In conclusion, GISTs from different anatomic locations share many overlapping ultrastructural characteristics; however, a few features are distinctive. It is hoped that these findings will aid in their recognition and contribute to the classification of this heterogeneous group of neoplasms.     

Cocaine profiling method retrospectively developed with nontargeted discovery of markers using liquid chromatography with time-of-flight mass spectrometry data

Illicit drug profiling performed by forensic laboratories assists law enforcement agencies through providing information about chemical and/or physical characteristics of seized specimens. In this article, a model was developed for the comparison of seized cocaine based on retrospective analysis of data generated from ultrahigh performance liquid chromatography with time-of-flight mass spectrometry (UHPLC-TOF-MS) comprehensive drug screening. A nontargeted approach to discover target compounds was employed, which generated 53 potential markers using data from cocaine positive samples. Twelve marker compounds were selected for the development of the final profiling model. The selection included a mixture of commonly used cocaine profiling targets and other cocaine-related compounds. Combinations of pretreatments and comparison metrics were assessed using receiver operating characteristic curves to determine the combination with the best discrimination between linked and unlinked populations. Using data from 382 linked and 34,519 unlinked distances, a classification model was developed using a combination of the standardization and normalization transformations with Canberra distance, resulting in a linked cut-off with a 0.5% false positive rate. The present study demonstrates the applicability of retrospectively developing a cocaine profiling model using data generated from UHPLC-TOF-MS nontargeted drug screening without pre-existing information about cocaine impurities. The developed workflow was not specific to cocaine and thus could potentially be applied to any seized drug in which there are both sufficient data and impurities present.     

In vitro and in vivo metabolism and detection of 3‐HO‐PCP,a synthetic phencyclidine,in human samples and pooled human hepatocytes using high resolution mass spectrometry

The new psychoactive substance (NPS) 3‐HO‐PCP, a phencyclidine (PCP) analog, was detected in a law enforcement seizure and in forensic samples in Denmark. Compared with PCP, 3‐HO‐PCP is known to be a more potent dissociative NPS, but no toxicokinetic investigations of 3‐HO‐PCP are yet available. Therefore, 3‐HO‐PCP was quantified in in vivo samples, and the following were investigated: plasma protein binding, in vitro and in vivo metabolites, and metabolic targets. All samples were separated by liquid chromatography and analyzed by mass spectrometry. The unbound fraction in plasma was determined as 0.72 ± 0.09. After in vitro incubation with pooled human hepatocytes, four metabolites were identified: a piperidine‐hydroxyl‐and piperidine ring opened N‐dealkyl‐COOH metabolite, and O‐glucuronidated‐ and O‐sulfate‐conjugated metabolites. In vivo, depending on the sample and sample preparation, fewer metabolites were detected, as the O‐sulfate‐conjugated metabolite was not detected. The N‐dealkylated‐COOH metabolite was the main metabolite in the deconjugated urine sample. in vivo analytical targets in blood and brain samples were 3‐HO‐PCP and the O‐glucuronidated metabolite, with 3‐HO‐PCP having the highest relative signal intensity. The drug levels of 3‐HO‐PCP quantified in blood were 0.013 and 0.095 mg/kg in a living and a deceased subject, respectively. The 3‐HO‐PCP concentrations in deconjugated urine in a sample from a living subject and in post‐mortem brain were 7.8 and 0.16 mg/kg, respectively. The post mortem results showed a 1.5‐fold higher concentration of 3‐HO‐PCP in the brain tissue than in the post mortem blood sample.     

Endogenous aggregates of amyloidogenic cystatin C variant are removed by THP-1 cells in vitro and induce differentiation and a proinflammatory response

A mutation in the human cystatin C gene leads to familial cerebral amyloid angiopathy. This disease is known as “hereditary cerebral hemorrhage with amyloidosis-Icelandic type” or “hereditary cystatin C amyloid angiopathy.” The mutant cystatin C protein forms aggregates and amyloid, within the central nervous system almost exclusively in connection with the vascular system. It was not known whether immune cells could remove mutant cystatin C protein aggregates. Ex vivo mutant cystatin C protein aggregates, both in solution and dried onto a glass surface, induced adhesion to the substrate, differentiated the THP-1 monocyte cell line and led to a proinflammatory response. Aggregates were also taken up by both THP-1 cells and THP-1 derived macrophages. These are the same responses induced by other amyloidogenic protein species, such as amyloid β protein and amylin, supporting the model of all amyloidogenic proteins being toxic due to common structural motifs. Proinflammatory response induced by the ex vivo mutant cystatin C protein aggregates suggests that vascular inflammation plays an important role in hereditary cerebral hemorrhage with amyloidosis-Icelandic type. Ex vivo protein aggregates of cystatin C might better model cellular behavior than in vitro-generated aggregates or supplement in vitro material.     

Epithelioid angiosarcoma of the bone: a series of 10 cases

The clinical and pathologic features of 10 epithelioid angiosarcomas of bone were analyzed. There were eight males and two females who ranged in age from 26 to 83 years (mean 62 years). Four tumors were solitary and six were multifocal. In two consultation cases, the submitted diagnosis was metastatic carcinoma. Microscopically, the tumor cells were arranged in solid and infiltrative sheets, and in most cases vascular channels or cystically dilated spaces were present. The neoplastic cells had abundant eosinophilic cytoplasm and large nuclei with open chromatin and prominent eosinophilic nucleoli. Intratumoral hemorrhage, neutrophilic infiltrates, and intracytoplasmic lumina were frequently present. All 10 tumors stained positive for one or more endothelial markers, with CD31 being the most sensitive marker. Seven cases stained positive for cytokeratin. Ultrastructural examination in three tumors confirmed their endothelial differentiation. In the absence of obvious vascular differentiation, abundant intratumoral hemorrhage and intratumoral neutrophils are useful ancillary morphologic features that may suggest a vascular origin. Six patients are dead of disease, one is alive with metastasis, and two patients are currently disease free. Epithelioid angiosarcoma of bone should be included in the differential diagnosis of epithelioid neoplasms of bone, and endothelial markers should be a part of their immunohistochemical analysis to avoid the misdiagnosis of a metastatic carcinoma because of the significant differences in the treatment and clinical outcomes of these entities.