Search

The diagnosis of cervical intraepithelial neoplasia (CIN) has low interobserver reproducibility. The pathogenesis of human papillomavirus (HPV) from infection to high-grade CIN is well understood. In benign lesions, HPV-DNA is often packaged into virions, whereas malignant transformation disrupts virion assembly. It is conceivable that if cervical lesions were exposed to endonuclease digestion, HPV virions would alter nuclear susceptibility to DNA degradation. We propose that susceptibility to endonuclease digestion can serve as a simple marker to identify CIN grade. From paraffin-embedded tissue blocks, condyloma accuminata, CIN I-III, and cervical carcinoma cases were identified. Sections were placed in a bath containing DNAse I for DNA digestion. Residual DNA was stained by a Feulgen process. Endonuclease-resistant DNA (erDNA) staining was correlated to disease grade. In addition, 10 HPV (+) patients whose infection regressed and 8 whose infection progressed to CIN II or above had their initial HPV lesions stained for erDNA. erDNA was observed in 81% condylomas and 80% CIN I cases. All CIN II, III, and cancer cases were endonuclease sensitive with 100% of lesions showing no staining. Eighty percent of HPV lesions that regressed had erDNA staining, whereas 75% lesions that progressed had no erDNA staining. The spectrum of cervical disease caused by HPV has different susceptibilities to endonuclease digestion, which may aid in the diagnosis of CIN. Furthermore, in our small pilot study, erDNA status was associated with the clinical outcomes. Prospective studies are needed to confirm this observation. erDNA status is a promising novel biomarker.

Medullary carcinoma of the breast has attracted attention because of its relatively good prognosis, in spite of its high cytologic grade. It has, by definition, a consistent, florid tumor infiltrating lymphocyte (TIL) population, probably the result of cytotoxic T-lymphocytes recognizing tumor cells in an HLA-DR-restricted manner. HLA-DR tends to be more highly expressed on primary medullary carcinoma cells than on ductal carcinoma cells; however, the MHC-class II antigenicity of the tumor cells themselves has not been analyzed extensively, and as yet there has been no comparative study of HLA-DR expression in medullary and ductal carcinomas metastatic to lymph nodes. Eleven cases of medullary carcinoma and 15 cases of ductal carcinoma, primaries, and respective lymph node metastases were analyzed by immunoperoxidase staining for HLA-DR and lymphocytes antigens. Polymerase chain reaction (PCR) analysis to identify HLA-DR subtypes from the paraffin blocks was performed on selected cases of primaries and nodal metastases of both tumor types. Immunoperoxidase staining for HLA-DR antigen revealed a marked difference in antigen expression between medullary and ductal carcinomas. In the medullary carcinomas, the mean percentage of cells staining for HLA-DR was 74.5% in the primary tumors and 67.3% in the nodal metastases. For the ductal carcinomas, the mean percentage of cells staining was 17.7% in the primaries and 7% in the metastases. There was a tendency for the level of HLA-DR expression to remain high in medullary carcinoma metastatic to nodes, whereas whatever HLA-DR was present within ductal primaries tended to diminish when cells metastasized to regional nodes. PCR analysis of the HLA-DR within the two tumor types revealed no emerging subtype or variant that could be associated with either the medullary or the ductal carcinomas. Medullary carcinoma cells express much greater quantities of HLA-DR, on the whole, than ductal carcinomas. Expression of HLA-DR is retained on medullary carcinoma cells that have spread to lymph nodes, whereas the smaller quantities of HLA-DR present within ductal primaries tend to diminish even further when the tumor cells are found in lymph nodes. No discernible HLA-DR mutations or predominant subtypes emerged on PCR analysis, and the authors therefore conclude that it is the quantity and not the quality of HLA-DR expression in medullary carcinoma that maintains the characteristic TIL infiltrate, not seen in ductal carcinomas.

Objective To investigate the possibility that various salivary gland tumors that look histologically similar could express p53 oncoprotein and Ki-67 proliferation antigen differentially and possibly aid in distinguishing benign from malignant lesions. Study design Intraoral paraffin-embedded biopsy specimens of salivary gland tumors were used. Thirty-eight pleomorphic adenomas, 17 adenoid cystic carcinomas, 23 monomorphic adenomas, and 17 polymorphous low-grade adenocarcinomas were stained with p53 and Ki-67 antibodies by using an immunoperoxidase detection system. Each case was evaluated in terms of staining intensity and percentage of cells staining. Results Ki-67 and p53 antigens are expressed in generally low levels in the histologically well-differentiated salivary tumors that were studied here, both benign and malignant. Only 1 solid-type adenoid cystic carcinoma showed a high percentage of cells expressing p53. Conclusions The histologically well-differentiated salivary tumors studied do not show differential expression of p53 and Ki-7, in spite of their differing courses of biologic behavior. These antibodies should not be relied on to distinguish benign from malignant lesions of the salivary glands; however, they might be markers for those lesions that are dedifferentiating histologically and, therefore, might be displaying more aggressive behavior.

Studies of antigen expression in dysplastic nevi have been limited to some extent by difficulties in obtaining frozen nevi with which to react monoclonal antibodies. Accordingly, we obtained a panel of antibodies that binds to antigens preserved in paraffin-embedded tissue. This panel of monoclonal antibodies, raised against nevomelanoma antigens, was used on 26 dysplastic compound nevi and additionally on 14 invasive superficial spreading melanomas. Among the dysplastic nevi, two basic dermal staining patterns emerged. One pattern designated "Type 1" shows a histologically well-developed dermal component that tends to be antigenically well stratified, with the cells in the upper dermis expressing the most antigen, and gradual loss of antigenicity in the lower dermis as maturation of nevic cells occurs. A second ("Type 2") pattern was seen in which nevic cells tended to remain in the upper dermis, with less downgrowth, and to express antigen in a diffuse or patchy, but nonstratified distribution. Some differential distribution of the antigens was noted, with antibodies 404-101, HMB-45, and ME 109 binding to the activated junctional zone, but showing lower binding affinity within the dermis. ME 491, NKI-C3, and 506 bind to antigens abundant in the junctional zone as well as the dermis. The antibody ME 67-6 binds to both junction and dermis, but is more useful for delineation of antigenic stratification and presence of abnormal "clones."

Twenty-five patients with primary presentation of malignant astrocytoma, astrocytoma with anaplastic foci, and glioblastoma multiforme were treated with surgical resection and definitive radiation therapy followed by intravenous or infra-arterial administration of Iodine-125 labeled monoclonal antibody-425, which binds specifically to human epidermal growth factor receptor. The patients presented with primary untreated disease, positive contrast enhanced computed tomography scans of the brain, and compatible clinical symptoms. In this Phase II clinical trial, the patients had surgical debulking or biopsy followed by definitively administered external beam radiation therapy and one or multiple doses (35 to 90 mCi per infusion) of radiolabeled antibody. The total cumulative doses ranged from 40 to 224 mCi. The administrations of the radiolabeled antibody were performed in most cases 4–6 weeks following completion of the primary surgery and radiation therapy. Ten patients had astrocytoma with anaplastic foci and 15 had glioblastoma multiforme. No significant life-threatening toxicities were observed during this trial. At 1 year 60% of the patients with astrocytoma with anaplastic foci or glioblastoma multiforme are alive. The median survival for both groups was 15.6 months.

Although the salient features of renal oncocytomas and renal cell carcinomas have been discussed in the recent literature, renal masses with features of both entities will present diagnostic difficulty, especially when the cells are diffusely eosinophilic on microscopic examination. A case of a firm, tan, rounded mass replacing the lower pole of the kidney is discussed. The final diagnosis of renal cell carcinoma, granular cell type, was made after multiple sections of the tumor were examined, and after electron microscopy was performed. A thorough search by light microscopy should be made for clear cell foci, necrosis, mitotic activity, and vascular or capsular invasion, features generally accepted as pathognomonic for renal cell carcinoma. Cellular and especially nuclear pleomorphism is typically focal or mild in renal oncocytomas. True oncocytic tumors will be packed with mitochondria on electron microscopy; however, granular renal cell carcinomas will contain mitochondria as well as other cellular organelles, lipid, and glycogen. Electron microscopy should be performed on tumors suspected of being oncocytomas because eosinophilia on hematoxylin and eosin stain, as demonstrated by this case, is not a predictable measure of mitochondria content. Immunoperoxidase staining for vimentin in oncocytomas has recently been shown to be negative, and may offer a method of ruling out oncocytoma in vimentin-positive tumors, pending further studies.

We present a case of a 24-year-old man with an original diagnosis of embryonal rhabdomyosarcoma, presumably of the right masseter or pterygoid muscle, in which the tumor progressed to involve a portion of the mandible and the adjacent intraoral tissues, including the retromolar gingiva and the roots of the third molar. Ultimately, there was widespread metastatic disease

Immunoperoxidase stains were performed on normal and neoplastic tissue from prostate, colon, thyroid, lung, nerve, uterus, and placenta embedded in both plastic (glycolmethacrylate [GMA]) and paraffin. Positive results in plastic section were obtained for carcinoembryonic antigen (CEA), keratin, epithelial membrane antigen (EMA), thyroglobulins, S-100, prostate-specific antigen, human chorionic gonadotrophin (HCG), and beta-HCG. More delicate staining with more precise localization of antigens is noted. Superior (paraformaldehyde) fixation and cold processing followed by GMA polymerization (4 degrees C) allow for optimum antigen survival. After fixation, tissue processing involves a series of 0.1 mol/L phosphate buffer rinses with sucrose and ammonium chloride in a conventional dip-and-dunk processor placed in a 4 degrees C cold room. Acetone dehydrations are used before GMA infiltration, cold polymerization, and sectioning. Before immunoperoxidase staining, the plastic section is digested in .25% bovine trypsin for ten minutes. The immunoperoxidase methods described can be useful when small biopsies are routinely embedded in plastic to obtain improved histologic (hematoxylin-eosin) sections. There may also be research applications in quantifying antigen expression in benign, dysplastic, and neoplastic tissues by examining the stains under high power.