Objective Developing a new diagnostic strategy for Malignant Pleural Mesothelioma measuring soluble proteins in plasma and pleural fluid using ELISA and qRT-PCR.

Methods Case-control study, in which the cases were incident patients with malignant pleural mesothelioma in two hospitals from Mexican Social Security (Neumology and Oncology) and there were 4 controls for each case. A questionary was applied for asbestos exposure. ELISA with double antibody was used to determine the presence of thrombomodulin and mesothelin, using known concentration curves for each one. An statistical descriptive analysis was made as well as logistic regression models and ROC curves.

Results In this study were enrolled 35 cases and 136 controls. The mean concentrations, (SD) and [min-max] in plasma for thrombomodulin were: 10.2 (7) ng/mL [1.6–32.2] for the cases, and 5.4 (5.1) ng/mL [0.78–32.5] for the controls. For mesothelin values were: 104.3 (78.2) ng/mL, [12.2–351.2] for the cases, and 25.2 (3.2) ng/mL, [6.5–330.3] for the controls. Both biomarkers were statistically significant. In the ROC curve for both biomarkers sensibility of 82.3% and specificity of 95.5% were achieved. In the nonconditional logistic regression model, mesothelin and thrombomodulin had an OR = 1.03 (CI: 1.02–1.05) and OR = 1.08 (CI: 1.01–1.16), respectively, adjusted for age and sex.

Conclusions A unique parameter to evaluate a diagnostic test in every situation doesn’t exist, but following Sackett and Álvarez-Cáceres guidelines, our two biomarkers have a high specificity (95.5%) and sensibility (82.35%) with a cut-off point, in this sample, ≥83 ng/mL for mesothelin and ≥10.5 ng/mL. Nevertheless we consider that our sample is small and that we also need to control the factors, such as diabetes mellitus, hypercholesterolemia and renal failure, that have influence in the level of both biomarkers.