Eadie Heyderman

Production of monoclonal antibodies against human epithelial membrane antigen for use in diagnostic immunocytochemistry

Two monoclonal murine antibodies have been raised against a delipidated extract of human cream. These antibodies were detected by immunohistological screening of hybridoma culture supernatants on sections of human breast tissue. One of those antibodies (E29) was subsequently screened against a range of normal and neoplastic human tissues and shown to react with a wide variety of human epithelia and with mesothelial cells. Antibody E29 was unreactive with other cell types, with the exception of occasional plasma cells. Antibody E29 is suitable for use on paraffin embedded tissue and represents a valuable reagent for the identification of tumours of epithelial origin.

Epithelial markers in primary skin cancer: an immunoperoxidase study of the distribution of epithelial membrane antigen (EMA) and carcinoembryonic antigen (CEA) in 65 primary skin carcinomas

Sixty‐five primary malignant skin tumours have been stained for carcinoembryonic antigen (CEA) and epithelial membrane antigen (EMA) using rabbit polyclonal affinity‐purified antibodies and an indirect immunoperoxidase technique. The tumours consisted of 15 invasive squamous carcinomas, 23 basal cell carcinomas, 16 malignant eccrine poromas (porocarrinomas), and 11 sebaceous carcinomas. The basal cell carcinomas were negative for CEA and EMA except where there was keratotic or sebaceous differentiation. All the sebaceous and squamous carcinomas and 15/16 porocarcinomas contained EMA. 12/15 squamous carcinomas were positive for CEA. The malignant poromas were negative for CEA except on the ulcerated surface of two. In tumours classified as sebaceous carcinomas there was positive staining for CEA in some cells, cyst contents and/or keratotic foci. These findings have implications for the use of immunoperoxidase localization of epithelial markers in the differential diagnosis of primary and metastatic skin cancer.

Epithelial markers in thyroid carcinoma: an immunoperoxidase study

Ten cases each of papillary, follicular, anaplastic and medullary carcinoma of the thyroid were stained for thyroglobulin, calcitonin, epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA) and cytokeratin (CAM 5.2). Monoclonal or affinity purified polyclonal antibodies, and an indirect immunoperoxidase technique were used. All the papillary and follicular tumours, 5/10 anaplastic and 3/10 medullary carcinomas contained thyroglobulin. Only the 10 medullary carcinomas stained positively for calcitonin. Three out of 10 papillary, 1/10 follicular, 0/10 anaplastic and 10/10 medullary carcinomas were positive for CEA. Nine out of ten papillary, 7/10 follicular, 2/10 anaplastic and 3/10 medullary carcinomas were positive for EMA. Ten out of 10 papillary, 10/10 follicular, 5/10 anaplastic and 10/10 medullary carcinomas were positive for cytokeratin. The presence of calcitonin and CEA is of value in the diagnosis of medullary carcinoma, and enable its distinction from anaplastic thyroid carcinoma. Thyroglobulin is a useful marker in thyroid carcinomas.

Torre‐muir syndromea. An association with isolated sebaceous carcinoma

Three cases of Torre‐Muir syndrome are described in which isolated sebaceous carcinomas were associated with internal malignancy. The first was a 66‐year‐old woman who had a sebaceous carcinoma in association with carcinoma of the uterus, two adenocarcinomas of colon, carcinoma of renal pelvis and bladder, and a squamous cell carcinoma of the skin. The second was a 78‐year‐old woman who had a sebaceous carcinoma associated with carcinomas of the colon, ovary, and pancreas. The third patient had a sebaceous carcinoma associated with breast cancer. The subject is discussed and the previously reported cases reviewed.

Human chorionic gonadotropin and human placental lactogen in extragonadal tumors an immunoperoxidase study of ten non-germ cell neoplasms

The immunoperoxidase localization of the alpha and beta subunits of human chorionic gonadotropin (hCG) and of human placental lactogen (hPL) was studied in ten extragonadal nontrophoblastic tumors associated with raised serum levels of one or more of these placental proteins. Three of the tumors were bronchial carcinomas, one was a gastric carcinoma, two were malignant carcinoids (one bronchial and one gastric), two were pancreatic islet cell carcinomas, and two were metastatic carcinomas with an unknown primary site. The maximum alpha subunit serum level was 33,000 ng/ml (gastric carcinoid), the maximum hCG/hCG‐beta level was 705,000 ng/ml, and the maximum hPL level was 50 ng/ml (both in the gastric carcinoma). An indirect immunoperoxidase technique and rabbit polyclonal affinity‐purified antibodies and peroxidase conjugates were used on formalin‐fixed, paraffin‐embedded sections. Five blocks (eight cases) or six blocks (two cases) from various sites were obtained from each patient at surgery and/or autopsy. Positive stains for hCG/hCG‐beta were seen in six of seven tumors (25/37 blocks) with raised levels, for the alpha subunit in nine of nine tumors (30/47 blocks), and for hPL in two of five tumors (4/26 blocks). Only a relatively minor number of the cells were positive, and within the same case, there was considerable site‐to‐site variation in the number of positive cells. Large bizarre cells contained hCG/hCG‐beta as well as the alpha subunit, if it was demonstrated in the same tumor as the beta subunit. Otherwise, the alpha subunit was found in small unremarkable cells. Giant cells that were smaller than those positive for hCG/hCG‐beta contained in hPL. In some serial sections, hCG‐alpha, hCG/hCG‐beta, and hPL were segregated in different cell populations, supporting the concepts of their separate genetic control.

Benign dermal Schwannoma with glandular elements—true heterology or a local ‘organizer’ effect?

A unique case of intradermal Schwannoma containing five morphologically distinct types of gland, which had arisen in the forearm of a 16‐year‐ald girl, is presented. The glandular elements have been investigated both histochemically and immunohistochemically. While some of these structures probably represent entrapped dermal appendages, it is argued that this tumour shows true heterologous glandular differentiation and, as such, is the second convincing reported example of benign glandular Schwannoma which has been reported. The literature with regard to glandular Schwarmomas is also reviewed.

A novel hybridoma antibody (PASE/4LJ) to human prostatic acid phosphatase suitable for immunohistochemistry.

A murine monoclonal antibody PASE/4LJ to prostatic acid phosphatase (PAP) was used to immunostain a wide variety of sections of benign and malignant tissues (654 blocks). Non-neoplastic adult and fetal prostatic glands, primary and metastatic prostatic carcinomas, and scattered cells in prostatic and penile urethra were positive. Rat, dog and rabbit prostates were negative. Nine of 400 tumours of non-prostatic origin showed some positivity: 6/36 carcinoids, 1/9 islet cell tumours, 1/55 ovarian adenocarcinomas (serous) and one carcinosarcoma of the lung (epithelial portion). Positive staining was seen in islet cells in 4/5 specimens of normal pancreas, and in 4/9 blocks of normal pancreas surrounding a pancreatic tumour. Loops of Henle, maculae densae, and distal tubules in 10/10 fetal and 2/9 adult kidneys were also positive, with proximal tubules and collecting ducts negative. All other 159 blocks of non-neoplastic adult and fetal tissues were negative. The antibody was also affinity purified from ascitic fluid, and shown not to inhibit the enzyme activity of prostatic acid phosphatase.

Immunocytochemistry today—problems and practice

Most histopathology departments use immunocytochemical techniques for the localization of a wide variety of antigens. Reynolds, in this issue of Histoputhology, reports on a national external quality assurance scheme designed to increase standardization and the quality of results. The improvements in diagnosis and classification achieved with the use of these techniques can make other more expensive and potentially hazardous investigations of the patient unnecessary, and enable the most appropriate therapy to be used. However, a number of problems in methodology and interpretation remain. Immunocytochemical localization does not distinguish between antigens synthesized by the cell and those taken up from elsewhere. Some antigens are bound to receptors, while others are passively adsorbed. Tumour cells, as well as macrophages, may be phagocytic, and endocytosis of a variety of substances occurs. Conversely, cells may appear negative if they export their synthesized products without significant storage. In .vim hybridization can be used to demonstrate mRNA and DNA in fixed or unfixed tissue sections, using complementary probes labelled with radioactive isotopes or with non-isotopic labels such as biotin and digoxigenin (Herrington et d. 1989). Localization of the specific mRNA virtually excludes phagocytosis, endocytosis and receptor-binding but, since the demonstration of mRNA does not necessarily indicate translation, immunocytochemistry may be required to confirm protein synthesis. The two techniques are complementary and, within the near future, both should be equally available in histopathology departments. The effect of fixation must be considered when designing immunocytochemistry protocols and assessing results. Antigens in unfixed frozen material are probably closest to their in zlizio state but most tissues are already fixed before arrival at histopathology departments. Fixation may be almost instantaneous, or it may take hours or days, depending on the size of the specimen and on the fixative. Autolysis can continue in unfixed portions, while fixation may crosslink and change the structure of antigenic determinants. For example, the D5 antibody (Amersham International, UK) can be used to localize the oestrogen receptor-related protein, P24, in methacarn-fixed materal, but positive areas appear almost negative in adjacent formalin-fixed blocks (King et a/. 1985). The majority of laboratories use buffered formalin for fixation (Angel, Heyderman & Lauder 1989a) and predigestion with proteolytic enzymes such as trypsin, pepsin or pronase may be required to unmask some antigens, particularly immunoglobulins. However, proteolytic degradation could ‘unmask’ unrelated antigens, and observer bias is introduced if enzyme incubation protocols are varied to produce the expected result. Moreover, enzyme

Which antibodies for diagnostic pathology

The large number of monoclonal antibodies now available is currently a source of considerable confusion to the diagnostician. Pathology journals surely carry a responsibility not to add to this by the publication of seemingly endless reports on new antibodies without good reason. How then, for example, does Histopathology justify the article by Pallesen et al. in the current issue? Pallesen’s paper was reviewed and accepted as a clear account of the production and evaluation of a new monoclonal antibody which may be of interest to histopathologists. Adequate information is given about the preparation of the antibody in question, named BG3C8, and about the way in which the molecule which it detects was characterized by biochemical techniques (oneand twodimensional gel immunoblotting) . Its tissue reactivity has been comprehensively evaluated and well illustrated and the relationship of BG3C8 to other antibodies of comparable specificities is discussed. In summary, any histopathologist interested in stratified epithelium has all the information necessary to judge how this antibody compares with any reagents he has used in the past and to decide whether it would be useful in his own work. Nevertheless, even if individual papers can all be justified from some particular view it remains true that the present plethora of commercially available antibodies is a minefield for the uninitiated. The view was therefore expressed by the editorial board of Histopathology that it would be of value to write a commentary on a ‘top 10’ of such antibodies which have immediate and useful applications for the general and diagnostic histopathologist and which can be used on formalin fixed paraffin embedded tissue. We were invited to submit such a list independently of each other. The editor has prepared a composite table in which these antibodies have been partially grouped and some relevant comments included. In the event the list is a top dozen and it is worth noting that in the final eclectic selection only three have been omitted from the original submissions made, one from each contributor’s list and in respect of which no strong plea was made for their retention. It is clear that the immunohistochemical techniques have still to be used as adjuncts to, rather than substitutes for, traditional morphological skills. Some major problems persist and include the following. 1 How do we know that an antibody which can distinguish between obvious cases (e.g. carcinomas from lymphomas) can still make this distinction in difficult cases? The difficulty in diagnosis may reflect oddity not only in morphology but also in phenotype.

Immunoperoxidase techniques: the deleterious effect of sodium azide on the activity of peroxidase conjugates.

The effect of including sodium azide as a bacteriostatic agent in solutions used to dilute antibodies conjugated with the enzyme horseradish peroxidase was examined. An enzyme-linked immunosorbent assay (ELISA) and an immunohistochemical method were used and both techniques demonstrated an inhibitory effect of sodium azide on the activity of the peroxidase conjugates. It is concluded that the use of sodium azide in solutions used to dilute peroxidase conjugates is to be avoided.