Search

Some 30 lectins in combination with glycosidase digestion and immunohistochemistry with 5 antibodies directed against antigens of the ABO and Lewis blood group systems were used to analyse the distribution and synthesis of glycoconjugates in the epithelium of the large airways in man. Both mucous gland cells and goblet cells were labelled by 12 of 30 lectins and by the antibodies, dependent on the ABO, Lewis, and secretor status. The corresponding binding patterns of the serous gland cells differed markedly from those of goblet and mucous gland cells and in general were not dependent on the ABO, Lewis, and secretor status. After digestion with neuraminidase and fucosidase, binding of soy bean agglutinin and peanut agglutinin to goblet and mucous gland cells was increased. Binding of peanut agglutinin to serous gland cells was stronger only after the digestion with neuraminidase. Digestion with O-glycosidase after the use of neuraminidase or fucosidase resulted in a decrease of peanut agglutinin binding to goblet and mucous gland cells. The present results show that the secretory products of goblet and mucous gland cells on the one hand and those of serous cells on the other differ considerably with respect to their terminal glycosylation. The glycosyltransferases coded by genes of the ABO and Lewis blood group and secretor systems are active only in goblet and mucous gland cells, resulting in the presence of the corresponding antigens. Precursor substances of blood group antigens types 1, 2, and 3 are found only in these cell types. In serous gland cells, blood group systems do not influence the glycosylation of glycoproteins. The results of the digestion with O-glycosidase indicates the presence of O-glycosylation in mucous gland and goblet cells, but not in serous gland cells.

There is an increasing demand for accurate preoperative and intraoperative staging of bronchial carcinoma with respect to neoadjuvant therapy protocols and parenchyma-sparing operations. This study prospectively evaluated accuracy of computed tomographic scan and surgical assessment for staging of bronchial carcinoma in 108 consecutive patients. The stage of the primary tumor (T stage) was correctly determined in 85% of the patients, and surgical evaluation correctly determined the T stage in 92%. Invasion of major mediastinal structures posed a major problem for computed tomographic scan. On a node-by-node basis, computed tomographic scan predicted involvement of lymph nodes in 81% (sensitivity 29%, specificity 93%, positive predictive value 49%, negative predictive value 85%). The surgeon correctly determined the lymph node status in 69% of lymph nodes (sensitivity 90%, specificity 63%, positive predictive value 39%, negative predictive value 96%). On a patient-by-patient basis, computed tomographic scan correctly predicted the nodal status in 58% of patients. Accuracy of computed tomographic scan and surgical assessment in determination of the lymph node status strongly depended on tumor type and lymph node region (hilar or mediastinal region) studied. This was partly due to the fact that adenocarcinomas exhibited a high proportion of tumor-positive normal-sized lymph nodes, whereas squamous cell carcinomas showed a high proportion of enlarged tumor-free lymph nodes. In conclusion, computed tomographic scan and surgical assessment are sufficiently accurate for determination of the tumor stage but are insufficient in determining the nodal status.