ИСТИНА |
Войти в систему Регистрация |
|
ИСТИНА ИНХС РАН |
||
Lung cancer currently remains the leading cause of cancer deaths worldwide in men and women. In particular, the reason of such state of affairs is associated with late diagnosis, when cancer symptoms may be absent or non-specific, screened by the symptoms of associated diseases. The development of new approaches in the differential diagnosis of lung cancer on the background of other lung and respiratory tract diseases is an important socially significant task.Exhaled breath condensate (EBC) is collected non-invasively and contains a large number of components - from gases to pro-tein molecules. The use of modern methods, such as high resolution mass spectrometry, has great potential in the field of profiling biomarkers of EBC.We have investigated EBC of 32 healthy volunteers, 30 patients with chronic obstructive pulmonary disease (COPD) and com-munity-acquired pneumonia, 46 patients diagnosed with lung cancer.Based on the study of EBC protein composition in healthy donors, cytoskeletal keratins were shown to be the major proteins of exhaled breath condensate. The set of proteins identified in samples from healthy donors served as a protein background for EBC for further research.Based on the study of EBC protein composition for patients with COPD, pneumonia and lung cancer, it was shown that the results of proteome analysis were consistent with the clinical picture of the diseases under consideration. 19 non-keratin proteins were identified only in EBC obtained from donors with diagnosed lung cancer stage 1-2 and proposed as a diagnostic panel for lung cancer.Based on the study, a linear analytical model was developed for predicting the presence of lung cancer in the donor. The model was tested with a group of donors not included in machine learning, and showed good prognostic ability (AUC = 0.99), identifying cancer samples.