Lung cancer represents a heterogeneous group of malignant neoplasms, among which two main forms can be distinguished — the non-small-cell and the small-cell lung cancer. These subtypes significantly differ by the histological, the molecular-genetic and the clinical characteristics, which defines the necessity of precise differential diagnostics for selecting the optimal treatment tactics. The review highlights the modern methods of diagnostics for the non-small-cell and the small-cell lung cancer, including the instrumental diagnostics, the histological and immunohistochemical examinations. Special attention was paid to the pros and cons of the promising non- and minimally invasive approaches, such as the analysis of circulating tumor cells, of the extracellular DNA, of the miRNA, of the marker proteins, of the volatile organic compounds and of the modern medical visualization (radiomics). Despite the significant progress in developing new diagnostic approaches, the problems remain that are related to the heterogeneity of tumors, the limited accessibility of the materials of small-cell lung cancer and the necessity of standardizing the new methods. The promising direction seems to the integration of multimodal approaches, combining the fluid biopsy, radiomics and the algorithms of machine learning, which can increase the precision of diagnostics and optimize the personalized treatment of the patients with various subtypes of lung cancer.

The stability of miRNAs in biological fluids and their association with pathological conditions position them as promising biomarkers for non-invasive cancer diagnostics, including those for non-small cell lung cancer (NSCLC). However, variability in miRNA expression and technical aspects of quantitative reverse transcription PCR (RT-qPCR) require effective normalization to accurately assess miRNA levels and identify biological differences. Despite the availability of multiple normalization methods, there is no consensus on the optimal approach for circulating miRNA data. This study provides a systematic methodological benchmarking of five normalization strategies—pairwise normalization, Tres normalization, Quadro normalization, normalization to the overall mean, and normalization to the exclusive mean—applied to RT-qPCR data of 17 miRNAs from plasma EVs of NSCLC patients and healthy donors. The aim was to evaluate their impact on the performance of machine learning models. For each normalization method, Random Forest models were built. Models were evaluated using accuracy, F1 score, AUC-ROC, AUPRC, precision, sensitivity, specificity, MCC, and overfitting gap. Pairwise normalization consistently outperformed all other methods across most metrics, achieving the highest AUC-ROC and AUPRC and the lowest overfitting gap, with statistically significant differences compared to other methods. Pairwise normalization offers a computationally simple, high-performing option for such benchmarking applications.

Introduction. It is known that the treatment of oncological diseases including prostate cancer (PCa) causes changes in the expression of oncogenic and oncosuppressive miRNAs. The analysis of miRNA expression dynamics can be used to predict the course of the disease and its response to therapy. However, the effect of PCa therapy on the expression of extracellular miRNAs is just beginning to be investigated. Aim. To study the expression dynamics of 14 miRNAs (miR-19b, -22-3p, -30e, -31, -92a, -125b, -144, -200b, -205, -222, -375, -378a, -425, -660) in urine extracellular vesicles of PCa patients after radical prostatectomy and to reveal prognostic miRNA ratios. Materials and methods. Urine samples of 18 donors and 18 PCa patients, obtained before radical prostatectomy, 1 week and 3 months after surgery, were examined. Extracellular vesicles were isolated by aggregation-precipitation protocol; extracellular vesicles miRNAs were isolated using fiberglass sorbents and octane acid. Data on threshold detection cycles of 14 miRNAs were obtained using reverse transcription – loop polymerase chain reaction (TaqMan). Results. It was found that prostatectomy causes a significant change in the relative expression of 44 miRNA ratios in the urine of PCa patients. Four groups of miRNA ratios can be distinguished: 1) miRNA ratios, which expression level significantly differed between donors and PCa patients before surgery and significantly changed in PCa patients 3 months after prostatectomy in the direction of the level of donors (6 pairs); 2) miRNA ratios, which expression did not significantly differ between donors and PCa patients before surgery, but significantly differed from the baseline in PCa patients and donors 3 months after prostatectomy (5 pairs); 3) miRNA ratios, based on expression ratios of which PCa patients can be divided into two or three significantly different subgroups 3 months after prostatectomy (19 pairs); 4) miRNA ratios that did not significantly change their expression after prostatectomy (30 pairs). Conclusion. Prostatectomy causes a significant change in the level of expression of miRNA in urine. 6 pairs of miRNAs, the relative expression of which after surgery significantly changed towards that of healthy donors and 19 pairs of miRNAs, according to the level of relative expression of which patients with prostate cancer were divided into two significantly different subgroups 3 months after prostatectomy, were identified based on the analysis of the dynamics of miRNA expression after prostatectomy.

BACKGROUND: Asymptomatic development and untimely diagnostics of lung cancer are the main reasons of high mortality caused by this disease. The development of tests based on “fluid biopsy”, which analyze circulating nucleic acids in blood, including the microRNA (miRNA) derived from the microvesicles, can be implemented as an effective screening test for lung cancer assuring the timely detection of the disease. AIM: To analyze the relative expression of 17 miRNAs in the microvesicles fraction from blood plasma samples from patients with non-small-cell lung cancer (NSCLC) and healthy donors. METHODS: The blood plasma microvesicles were isolated by aggregation-precipitation. miRNA was then isolated from the obtained blood microvesicles fraction using the glass fiber filters. Using the method of reverse transcription polymerаse chain reaction, an evaluation was carried out for the diagnostic efficiency of previously proposed miRNA pairs. This analysis was also used to select the most effective diagnostic pairs and panels. RESULTS: The obtained data have confirmed the diagnostic efficiency of four miRNA pairs (-31/-125b; -133b/-374a; -133b/-425; -133b/-222). The sensitivity and the specificity of distinguishing NSCLC patients from donors were 79% and 100%, respectively. Further analysis using paired normalization, compiling the regression models and multiple sample generations yielded three independent minimal panels of miRNA pairs, capable of diagnosing NSCLC with 100% accuracy: (1) -19b/-425, -125b/-378a, -205/-660; (2) -324/-374a, -30e/-92a, -125b/-378a; (3) -324/-374a, -125b/-378a, -205/-660. CONCLUSION: The diagnostic performance of the proposed miRNA panel across independent cohorts, along with the observed inter-cohort variability, supports its promising potential for NSCLC screening test development. These data demonstrate the necessity to identify additional multi-purpose miRNA markers and to validate the panel in further independent patient groups. Such research is essential to establish a reliable and timely miRNA-based screening tool for NSCLC.