Gene Expression Profiling of EGFR, FGFR2, PIK3CA, PTEN, SMAD4, STK11, and TP53 in Cell-Free RNA From Exhaled Breath Condensate: Diagnostic, Prognostic, and Therapeutic Implications in Advanced Lung Adenocarcinoma.

BACKGROUND: Advanced non-small cell lung cancer (NSCLC), particularly lung adenocarcinoma, remains a leading cause of cancer-related mortality. Although molecular profiling has improved disease stratification, most biomarker assays rely on invasive tissue sampling. Exhaled breath condensate (EBC) is an attractive non-invasive biofluid that may provide clinically relevant molecular information. This study evaluated whether cell-free RNA (cfRNA) obtained from EBC can capture diagnostically and prognostically informative gene-expression changes in advanced lung adenocarcinoma.

METHODS: In this prospective study, 40 patients with histologically confirmed stage IIIB-IV lung adenocarcinoma and 20 healthy controls were enrolled. EBC samples were collected from all participants, and plasma samples were obtained from a subgroup of 10 patients and 10 controls. Expression of EGFR, FGFR2, PIK3CA, PTEN, SMAD4, STK11, and TP53 was quantified by RT-qPCR and normalized to GAPDH using the 2-ΔΔCT method. Receiver operating characteristic (ROC) analysis, correlation testing, and Kaplan-Meier survival analysis were performed.

RESULTS: The most clinically relevant findings were observed for PIK3CA, FGFR2, and EGFR. In EBC, PIK3CA emerged as the only gene with significant diagnostic performance for distinguishing patients from healthy controls (AUC = 0.8522; sensitivity 81.58%; specificity 76.92%). In survival analyses, high FGFR2 and high PIK3CA expression were associated with significantly shorter overall survival, supporting their prognostic relevance. In treatment-stratified analysis, low EGFR expression was associated with longer survival in patients receiving chemotherapy alone, suggesting potential predictive value in this subgroup. In plasma, PIK3CA was upregulated approximately five-fold, whereas EGFR was downregulated approximately five-fold. Notably, EGFR expression showed a strong positive correlation between EBC and plasma (R2 = 0.7849, p = 0.0006), indicating cross-matrix consistency for this marker. By contrast, alterations in PTEN, TP53, and SMAD4 should be considered exploratory, as they showed differential expression but lacked comparable diagnostic, prognostic, or treatment-associated strength.

CONCLUSIONS: EBC-based cfRNA profiling provides a feasible, non-invasive approach for molecular characterization of advanced lung adenocarcinoma. Among the genes examined, PIK3CA showed the strongest diagnostic signal, FGFR2 demonstrated prognostic significance, and EGFR showed the clearest cross-sample concordance and treatment-associated relevance. Larger independent validation studies are required before clinical implementation.