Diagnostic Potential of Exosomal and Non-Exosomal Biomarkers in Lung Cancer: A Comparative Analysis Using a Rat Model of Lung Carcinogenesis

Abstract

Background: Identifying liquid biopsy biomarkers with high efficacy is crucial for cancer diagnosis. Exosomal cargo, including miRNAs and proteins, offers enhanced stability in biofluids compared with their free circulating forms, but direct comparisons of their diagnostic performance remain limited. This study evaluates and compares the diagnostic value of selected miRNAs and protein markers in exosomal versus non-exosomal fractions across stages of lung carcinogenesis in a rat model.

Methods: Lung cancer was induced in rats, and blood and lung tissue samples were collected at consecutive stages of tumor induction. We investigated the expression patterns of key miRNAs (miR-19b, miR-21, and miR-145) in exosomes, serum, and tissue and quantified levels of tumor biomarkers CEA and CYFRA 21-1 in exosomal and serum fractions.

Results: Our results revealed distinct expression patterns of the evaluated miRNAs across exosomes, serum, and tissue, throughout different stages of tumor induction. The expression of exosomal miRNAs dynamically changed in parallel with the tumor induction process, demonstrating high diagnostic efficacy. Specifically, exosomal miR-19b and miR-21 were significantly upregulated from an early induction stage, whereas their serum and tissue forms increased only during the late stages of induction. On the other hand, miR-145 was consistently downregulated across all fractions at every stage. Both exosomal and serum CEA levels increased significantly during tumor induction, while serum CYFRA 21-1 outperformed its exosomal counterpart. Strong positive correlations linked exosomal miR-19b and miR-145 with their non-exosomal counterparts, while moderate correlations were seen for miR-21 and the protein markers.

Conclusions: Our findings underscore the value of integrating exosomal biomarkers in liquid biopsies, highlighting their potential to improve early detection and monitoring of lung cancer development.

Keywords: diagnosis; exosomes; extracellular vesicles; liquid biopsy; lung cancer; miRNAs; tumor marker.

Figure 1

H&E-stained photomicrographs for lung tissue sections. (a) Normal control group showing normal alveoli and interalveolar septa. (b,c) Stage I group showing mild thickening of the interalveolar septa (arrow) and edema in the pulmonary blood vessels’ walls (dashed arrow). (d–f) Stage II rats showing thickening of the interalveolar walls (arrow) with infiltration of mononuclear inflammatory cells and eosinophils, and also in the perivascular area (dashed arrow), and desquamated bronchial epithelium (short arrow). (g–k) Stage III rats showing alveolar collapse (area) with marked infiltration of inflammatory cells and hyperplastic alveolar epithelium, which has scattered mitosis (arrow), polypoid hyperplasia of the bronchiolar epithelium (dashed arrow), lung consolidation, and vascular intimal proliferation with medial hyperplasia (M).

Figure 2

H&E-stained photomicrographs for lung tissue sections. (a–f) Stage IV rats showing multifocal areas of alveolar lining hyperplasia (square) with increased cellular atypia, polypoid hyperplasia of the bronchial epithelium (arrow), hyperplasia of the peri-bronchial lymphoid follicles (BALT). (g–l) Stage V rats showing multifocal alveolar atypical dysplasia with complete obliteration of the alveolar lumen (OB), solid adenocarcinoma (i,j), marked vascular intimal proliferation obliterating the vessels’ lumen (arrow) with medial hyperplasia (M), and focal metaplastic changes (dotted arrow) of the bronchial epithelium.

Figure 3

Exosomes isolated from serum samples were characterized by (A) transmission electron microscopy (TEM), where cup-shaped vesicle structures were identified as exosomes. (B) dynamic light scattering (DLS), where particle sizes were normally distributed with a mean value of approximately 77 nm. Size (d.nm) = Size in diameter nanometers, St Dev (d.nm) = Standard Deviation in diameter nanometers. (C) Western blot analysis for exosomal surface marker CD63 in a set of isolated exosome samples.

Figure 4

Box plot whiskers representing the relative expression levels of selected miRNAs (miR-19b, miR-21, and miR-145) in extracted exosome, serum, and tissue samples at each time point of collection. Statistical significance between groups was assessed using one-way ANOVA followed by Tukey’s post hoc multiple comparison test. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001.

Figure 5

Box plot whiskers representing the mean concentration values (ng/mL) for Carcinoembryonic Antigen (CEA) and Cytokeratin Fragment Antigen-21 (CYFRA 21-1) in exosomal and matched serum samples at each time point of collection. Significance was estimated between groups using one-way ANOVA, followed by Tukey’s post hoc multiple comparison test. * p > 0.05; ** p > 0.01; *** p > 0.001; **** p > 0.0001.

Figure 6

Spearman’s correlation analysis of selected miRNAs and protein tumor markers across exosomal, serum, and tissue fractions. The color intensity represents the density of data points.

Figure 7

Schematic diagram illustrating the experimental timeline and design for the chemical-induced lung carcinogenesis model. The diagram shows the timing of five intraperitoneal DEN injections (150 mg/kg body weight) administered at weeks 0, 4, 8, 12, and 16. Simultaneously, treated animals received 0.05% PB in their drinking water continuously for 20 weeks. Sample collection points from the DEN/PB-treated groups are indicated at 4, 8, 12, 16, and 20 weeks, representing different durations of carcinogen exposure. A parallel control group received equivalent volume saline IP injections and regular drinking water, and was sampled at the end of the 20-week study.

Figure 8

A flow chart of the miRNA selection process.