Mesothelioma cells grow diffusely throughout pleural and peritoneal cavities and do not typically form discreet tumor masses. Consequently, it is difficult to establish the extent of disease and to monitor the response to chemotherapy by conventional radiographic techniques. Mesothelin is a PI-linked glycoprotein that is produced by mesothelial and ovarian cells, but it is not widely produced by other tissues or cells. A significant fraction of mesothelin is known to be shed by tumor cells into culture media and has been previously observed to be elevated in serum from patients with mesothelioma, using a semiquantitative ELISA assay. The objective of this project is to produce a quantitative immunoassay for mesothelin and assess its possible clinical utility. A two-site sandwich type ELISA for mesothelin was developed in the previous year and has been further modified and improved in the past year. Based on a study of a normal control population (N=40), 9 ng/mL was decided as the optimum cutoff, which yielded a specificity of approximately 90% and a sensitivity of 50% for the diagnosis of mesothelioma. All false positive subjects detected so far were discovered to have renal failure, which may preclude the use of the test in this population. All 6 patients, who underwent IP chemotherapy treatment for peritoneal mesothelioma, showed a marked reduction in serum mesothelin, starting at least 1 week after treatment. In the upcoming year, we plan to do the following: (1) Develop a more sensitive and precise mesothelin assay, using an automated chemiluminescent immunoassay system. (2) Assess the clinical utility of the assay in a large collection of subjects with pleural and peritoneal mesothelioma. (3) Characterize the molecular forms of serum mesothelin by Western blot/MALDI-TOF Mass spectroscopy.
