Search

A panel of six monoclonal antibodies (MAbs) was employed to evaluate antigen expression in pulmonary adenocarcinomas and mesotheliomas. Monoclonal anti-human milk fat globulin (HMFG-2), anti-carcinoembryonic antigen (NP-2), anti-epithelial membrane antigen (EMA), anti-cytokeratin (PKK-1), anti-tumor-associated antigen 72 (B72.3), and anti-human myelomonocytic antigen (Leu M-1) antibodies were used to localize their respective antigens in formalin-fixed, paraffin-embedded tumors by using the avidin-biotin-complex immunoperoxidase technique. In all, 28 mesotheliomas obtained from Ohio State University Anatomic Pathology files and from a Southwest Oncology Group (SWOG) protocol were compared to 22 pulmonary adenocarcinomas by using this MAb panel. None of the mesotheliomas demonstrated positive staining with MAbs NP-2 (anti-CEA) or Leu M-1. However, 95% (21/22) of adenocarcinomas stained with one of these two antibodies. Although neither of these two MAbs stained all adenocarcinomas, each antibody demonstrated positive immunostaining in more than 90% of the adenocarcinomas studied. Therefore, MABs NP-2 and Leu M-1 are, individually, quite useful for distinguishing mesothelioma from adenocarcinoma. However, in our study, no single MAb could be used to distinguish these two tumor types in every case. MAb B72.3 stained 91% (20/21) adenocarcinomas but also stained 7% (2/28) of mesotheliomas. MAb HMFG-2 reacted positively with 95% of adenocarcinomas, but also stained 39% of the mesotheliomas, usually in a membranous pattern. MAbs EMA and PKK-1 were not found useful in distinguishing mesothelioma from adenocarcinoma. We conclude that MAbs Leu M-1 and NP-2 were both useful in distinguishing mesothelioma from pulmonary adenocarcinoma in that positive staining was demonstrated in adenocarcinomas and not mesotheliomas.

Decalcified bone marrow biopsies containing metastatic tumor from 36 patients were stained for prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP) using the avidin biotin complex (ABC) immunoperoxidase technique. Of these patients, 22 had known prostate primaries, ten had known nonprostatic, and four female patients had unknown primaries. Prostate-specific antigen was identified in 86% (19/22) of the metastatic prostatic carcinomas. Prostatic acid phosphatase was present in only 36% (8/22). None of the patients with nonprostatic primaries or unknown primaries showed positive staining for either antigen (0/14). This study indicates that immunoperoxidase staining for PSA is very sensitive and specific in the diagnosis of metastatic prostate carcinoma, while PAP was less sensitive using decalcified bone marrow specimens. We believe that immunostaining with PSA should be of great value in diagnosis of prostatic carcinoma metastatic to the bone.

This report describes the findings of an extra-adrenal paraganglioma and chondroma of the lung in an elderly man. Tissue analysis of the paraganglioma showed a high level of catecholamines, as well as the presence of somatostatin and small amounts of other peptide hormones. Immunoperoxidase staining demonstrated the distribution of peptide hormones in the tumor. This case illustrates two important points. First, it confirms the ability of paraganglionic tissue to secrete peptide hormones. Secondary, it adds yet another case to the unusual group of patients described by Carney (Carney JA: Cancer 43:374-382, 1979).

A new intraoperative approach to tumor localization using radiolabelled monoclonal antibody (MAb) B72.3 involves the use of a hand-held gamma-detecting probe (GDP) by the surgeon and, subsequently, the pathologist. We report here the use of 125I-labelled MAb B72.3 IgG and a GDP to localize primary and metastatic colorectal cancer in 31 patients. The patients were administered radiolabelled MAb i.v., and all underwent surgical exploration 5 to 35 days post-injection. In vivo localization of the MAb was evaluated using a GDP, with tumor and normal tissue counts being obtained. In each case, the subsequent tumor and normal tissue that were resected were analyzed in vitro for MAb localization; this was evaluated by calculating the radiolocalization index, i.e., the ratio of the injected dose per gram localized to tumor versus that of normal tissue. When the GDP was used intraoperatively, MAb B72.3 localized tumors in 68% (21/31) of the patients; the arbitrary criterion of tumor-to-normal tissue ratios higher than or equal to 2.0:1 in vivo being taken as positive. Resected tumor radiolocalization indices ranged from 0.5 to 543, and 71% (22/31) of the patients studied had tumors with radiolocalization indices higher than or equal to 3. Of 50 carcinoma biopsies, 34 that were probe-positive were antigen-positive when B72.3 was used in immunoperoxidase assays, while 4 carcinoma biopsies that were probe-negative were also antigen-negative. Twelve of 50 biopsies were probe-negative and antigen-positive, but many of these lesions only contained a few antigen-positive cells; none of the 50 was probe-positive and antigen-negative. Tumors of all histologic grades localized injected MAb and, in general, higher in vivo probe ratios and radiolocalization indices were obtained from patients who underwent surgery 20 to 35 days following injection of the MAb. MAb B72.3 localized tumor in all sites to which colon carcinoma commonly metastasizes, including mesenteric and peri-aortic lymph nodes, liver, lung, and peri-rectal soft tissue. There was a strong statistical correlation (p = 0.001) between detecting MAb B72.3 localization to tumors using the GDP intraoperatively and subsequent in vitro analysis of cpm/g for tumor versus normal tissues. These studies thus further validate the use of 125I-labelled MAb B72.3 IgG and of a hand-held gamma probe for the intraoperative detection of carcinomatous lesions.

This study was undertaken to evaluate the relative utility of histochemical and immunohistochemical stains in diagnosing malignant mesothelioma of the thorax. We performed a battery of histochemical stains, including periodic acid Schiff (PAS) with and without diastase, mucicarmine, colloidal iron (Coll Fe) with and without hyaluronidase, and immunohistochemical stains for keratin and carcinoembryonic antigen (CEA) on 12 pleural mesothelioma specimens obtained from 11 patients, five primary pulmonary adenocarcinomas, and one metastatic adenocarcinoma each to pleura and pericardium. All diagnoses were established by autopsy or thorough clinical and surgical evaluation. The diagnosis of mesothelioma was established following rigid anatomic criteria. All tissue was formalin fixed and paraffin embedded. Commercially available reagents and antisera were used in all cases. Results showed a high rate of positivity for keratin and hyaluronidase-sensitive Coll Fe in the mesotheliomas while adenocarcinomas were uniformly positive for CEA and keratin and generally positive for PAS-D (diastase) and mucicarmine. Mesotheliomas were negative for CEA, mucin, and PAS-D. Positive keratin staining was also seen in the spindle cell components of mesotheliomas. Immunohistochemical stains often added significantly to our ability to establish the diagnosis of mesothelioma with confidence, since they were more frequently and more clearly positive than histochemical stains.

Twenty-two inflammatory, benign, or malignant parotid lesions were studied by means of the specific red cell adherence test (SRCA), a modification of the Coombs' mixed cell agglutination reaction. In 22 normal parotid tissues the acinar structures were devoid of red cell agglutination, but it was present in ductal epithelium. Findings were similar in 2 cases of parotitis, 11 benign mixed tumors, and 1 malignant mixed tumor. All lacked red cell agglutination in areas of neoplastic change. Benign Warthin's tumors (4 cases) demonstrated antigenicity in the columnar epithelial component of the tumor, but lacked red cell agglutination in areas of the lymphoid component. One malignant Warthin's tumor showed agglutination in areas of normal columnar epithelium but not in areas of malignant dedifferentiation. Undifferentiated carcinoma (1 case) and adenoid cystic carcinoma (2 cases) did not possess detectable ABO (H) antigens in neoplastic areas of the gland. The absence of ABO antigens in normal acinar glands supports their suggested myoepithelial or mesenchymal derivation, as the absence of antigen in benign and malignant mixed tumors supports their proposed mesenchymal derivation. Ductular epithelium and the epithelial components of benign Warthin's tumors have ABO (H) antigens, while the loss of antigen in the epithelial portion of the malignant Warthin's tumor is characteristic of epithelial neoplastic dedifferentiation. Loss of antigen in adenoid cystic and undifferentiated carcinomas of the parotid supports the concept that antigen is absent in epithelially derived malignant neoplasms.