Initial cancer evaluation includes assessment of histologic appearance, tumor grading, assessment of lymph node status, and presence of metastasis. However, traditional diagnostic methods such as histopathology and radiology are not sensitive enough to detect small numbers of cancer cells and are limited in their ability to predict response to treatment. Recently, there has been considerable progress in molecular diagnostics in these areas. Using molecular-based technologies, it is now possible to detect cancer early in asymptomatic individuals, identify minimal residual disease at histopathologic normal surgical margins, more precisely assess tumor burden in cancer patients, and more accurately assess the prognosis of the patients. Examples of these applications in the evaluation of head and neck cancer are reviewed here. However, despite the great promise of these new molecular approaches for cancer detection, much of the current technology limits their implementation into routine clinical use. A 2005 Wiley Periodicals, Inc. Head Neck 27: 995–1003, 2005

Young men dying suddenly and autopsied by the coroner sometimes have coronary thrombosis at a relatively early stage of arteriosclerosis. The plaques under such thrombi often have a complex of features, a) rupture, b) hemorrhage, c) medial destruction, d) nodular collections of foam cells, e) calcification, f) cellular infiltrates of the fibrous cap, fibrous base and adventitia, and g) a newly described kind of phagocytic activity at the boundary between the necrotic core and the fibrous base of the plaque. Commonplace innocuous plaques in most middle and old aged subjects without heart disease also often have some of these features. What structural characteristics might distinguish rare thrombogenic from commonplace innocuous plaques? Twenty-one thrombotic plaques from 18 cases of sudden coronary heart disease (CHD) death were histologically compared with 129 nonthrombotic plaques from these same 18 cases, 85 plaques from 23 cases of CHD death due to arteriosclerotic occlusion, and 94 plaques from 22 cases having no CHD. Plaques with thrombosis all had necrotic cores; plaques for comparison with these were therefore chosen all to have necrotic cores. Rupture and hemorrhage were found in 90% of thrombotic plaques, with mixing of plaque gruel and blood in the thrombus. Medial destruction, foam cells and calcification (features c, d, and e) were commonplace in all types of plaques. Small-cell infiltrates and atherophagocytosis (features f or g) were found in 72–94% of the 21 thrombotic plaques, but only in 18–24% of the 94 not CHD plaques. The necrotic core, characterized by crystalline cholesterol, appears to incite cellular responses in some plaques but not others; those responses distinguish thrombogenesis. The findings imply that thrombogenicity and its accompanying plaque cellularity are incited not by cholesterol, but by some trace or minor component of the plaque gruel of the necrotic core. The possibility of testing these hypotheses by practical methods has been shown to be feasible.

Sinonasal lymphomas represent a distinct subset of extranodal head and neck lymphomas. While sinonasal lymphomas are relatively rare in Western countries, in Asian populations they are the second most frequent group of extranodal lymphomas, after gastrointestinal lymphomas. With advances in immunohistochemistry, these lymphomas have been separated into B-cell, T-cell, and most recently into natural killer (NK) cell phenotypes. The B-cell phenotype is typically located in the paranasal sinuses and has a slight predominance in Western countries. The T/NK-cell phenotype is the most common in Asian and South American countries. These tumors are typically located in the nasal cavity and have an aggressive, angioinvasive growth pattern that often results in necrosis and bony erosion. Thus, sinonasal lymphomas have been included in the past with other destructive malignant and benign lesions under the descriptive and nonspecific name lethal midline granuloma. Patients are classically in the sixth to eighth decades, with a 2:1 male-to-female ratio. The prognosis is generally better than that of nodal-based lymphomas of similar histologic grade. Treatment is a combination of local irradiation and chemotherapy with an anthracycline-based regimen.

We describe the clinicopathologic features and biologic behavior of 16 cases of histologically benign hemangiopericytoma containing a variable amount of mature fat as an intrinsic part of the neoplasm. These so-called lipomatous hemangiopericytomas occurred primarily in men (12 men and 4 women) with a mean age of 54 years (range, 33-74 years). All occurred in deep soft tissue and had an average size of 10 cm when first detected. All were characterized by a relatively sharp border and typical histologic features of hemangiopericytomas, including oval to round cells surrounding a sinusoidal and staghorn vasculature often with perivascular hyalinization. Mature fat varied in amount but usually occupied approximately one quarter to three quarters of the area of tumor. Mitotic activity was low, with more than half the cases having no mitotic activity. Five cases showed moderate nuclear atypia. In four cases, the pericytic regions had sclerotic zones. In contrast to liposarcoma, neither lipoblasts nor isolated atypical hyperchromatic cells within mature fat, as are seen in well-differentiated liposarcoma, were present. Immunohistochemistry performed in four cases showed factor XIIIa in tumor cells and an intricate pattern of immunoreactivity around cells for type IV collagen. CD34 and smooth-muscle actins were identified in two of four cases. Follow-up in seven cases showed no recurrences or metastases within the follow-up period of 1 to 7 years. Because these lesions are located in deep soft tissue and contain large amounts of mature fat, they could be mistaken for well-differentiated liposarcomas in limited biopsy material, although the distinction is easily made in examining the entire specimen. The lipomatous hemangiopericytoma represents yet another example of a bimodal mesenchymal tumor containing mature fat and raises the question of whether a common cytogenetic abnormality can explain the emergence of two clonal populations in this hybrid tumor.

The trefoil gene family of mucus cell-secreted proteins is a critical mediator of gastrointestinal mucosal restitution. Transcription of trefoil genes is induced during mucosal repair, but the regulatory mechanisms involved are unknown. Mice deficient in the intestine-specific peptide intestinal trefoil factor (ITF), in which colonic restitution is lethally impaired, showed reduced expression of the gastric trefoil genes SP and pS2, suggesting that trefoil peptides may individually regulate transcription of the entire family. In gastric cell lines, the trefoils were shown to act in a manner suggestive of immediate-early genes capable of auto- and cross-induction through cis-acting regulatory regions. Trefoil-mediated transcriptional regulation required activation of the Ras/MEK/MAP kinase signal transduction pathway. EGF receptor (EGF-R) activation was also necessary for trefoil auto- and cross-induction, and both spasmolytic polypeptide (SP) and ITF stimulation of gastric cell lines led to phosphorylation of EGF-R. Nevertheless, ITF and ITF-thioredoxin cell surface binding at 4 degrees C colocalized not with EGF-R, but with CD71, which is found in clathrin-coated pits, suggesting that integration of trefoil peptide responses may occur after internalization. As EGF-R expression is itself strongly induced after mucosal damage, the trefoil/EGF-R relationship may be pivotal in the generation and maintenance of the mucosal repair phenotype.

Synovial chondromatosis (SC) of the temporomandibular joint (TMJ) is a rare disease that is characterized by the development of nodules of cartilage within the synovial connective tissues of articulating joints. Reports of extracapsular TMJ SC are rare. A case is presented of primary SC of the TMJ with extension to the pterygoid plates, with a suspected traumatic etiology. The differences between primary and secondary SC are discussed.

Sixty-nine round cell lesions of the sinonasal region (22 olfactory neuroblastomas [ONBs], 17 malignant lymphomas, nine Ewing's sarcomas [ES], nine rhabdomyosarcomas, three sinonasal undifferentiated carcinomas, five malignant melanomas, and four pituitary adenomas) were studied in an attempt to define the differential diagnostic capabilities of antibody to MIC2 and bcl-2 in paraffin-embedded tissue in the distinction of these lesions. In addition, antibody to p53 was applied in each case to define the incidence of p53 positivity among these various tumor types. Each of the ES cases was MIC2 positive; each of the other cases was MIC2 negative. Positivity for bcl-2 was confined to two cases, one of them a malignant lymphoma (85% of cells positive) and one an ONB (5% of cells positive). Small numbers of scattered p53-positive cells appeared in the majority of cases studied, without regard for the specific tumor type; only a single case, a malignant lymphoma, showed a majority (approximately 90%) of p53-positive cells. These results indicate that the MIC2 antibody is a useful method by which to distinguish ES from a variety of other round cell lesions that may be encountered in the sinonasal region. The practical applications of antibody to bcl-2 and p53 seem to be much more limited; by contrast, neither bcl-2 positive cells nor abundant p53 cells identified by immunohistochemical analysis seemed to be frequent findings in any of the tumor types studied. Although ONBs have been included with the peripheral primitive neuroectodermal tumors for classification purposes, these tumors diverge from the ES/primitive neuroectodermal tumor family in that they do not seem to share either the MIC2 positivity or the t(11;22) chromosomal translocation that typify the ES/primitive neuroectodermal tumor family of lesions. Although bcl-2 positivity has been associated with a light microscopic finding of an unfavorable histologic pattern in retroperitoneal neuroblastomas, it does not seem that bcl-2 positivity in ONB will select for a clinically distinctive subset of patients.

Perineurioma is a rare, probably benign peripheral nerve sheath neoplasm. Perineurial cell origin is most efficiently demonstrated immunohistochemically by positivity for epithelial membrane antigen and negativity for S-100 protein and Leu 7.