The proliferation of tumor cells, a critical phase in cancer progression, is reliant on soluble growth factors and other regulatory molecules delivered by exosomes. Consequently, exosomes play an active role in modulating the tumor microenvironment and influencing tumor cell proliferation and progression. Long non-coding RNA (LncRNA) UFC1 is upregulated in serum exosomes, tumor tissues, and serum of patients with NSCLC, and shows a positive correlation with tumor infiltration. Silencing UFC1 results in the inhibition of NSCLC cell proliferation, migration, and invasion, while inducing cell cycle arrest and apoptosis (Zang et al., 2020). This suggests that exosomes derived from NSCLC cells facilitate NSCLC cell proliferation, migration, and invasion through the transfer of regulatory molecules. Additionally, circSATB2 is highly expressed in NSCLC cells and tissues and positively regulates the expression of fascin homolog 1, actin-bundling protein 1 (FSCN1) via miR-326. Exosomes transfer circSATB2 to enhance the proliferation, migration, and invasion of NSCLC cells, and also induce abnormal proliferation in normal human bronchial epithelial cells (Zhang et al., 2020). In a separate investigation, researchers demonstrated that circ-CPA4 modulates cell proliferation, stemness, motility, and drug resistance in NSCLC cells. Additionally, it inactivates CD8⁺ T cells within the tumor microenvironment through the let-7 miRNA/programmed death-ligand 1 (PD-L1) axis (Hong et al., 2020). Furthermore, the overexpression of Wnt5b, a protein belonging to the same family as Wnt5a, has been linked to increased cancer aggressiveness. Wnt5b-associated exosomes are biologically active, promoting the proliferation and migration of A549 lung AdCa cells and enhancing Wnt5b signaling in Chinese hamster ovary (CHO) cells (Harada et al., 2017).

Exosomes facilitate the transfer of immunosuppressive molecules either through paracrine signaling or direct interaction with immune cells, leading to the suppression of anti-tumor functions and the promotion of a tumor-supportive microenvironment. Consequently, tumor-derived exosomes modulate antitumor immune responses by delivering materials such as mRNA, miRNA, and DNA, which inhibit T-cell activation and proliferation, increase the presence of regulatory T-cells (Tregs) and myeloid-derived suppressor cells (MDSCs), and suppress the function of natural killer (NK) and CD8⁺ T-cells, thereby effectively evading the host immune system (Zandberg et al., 2024; Zabeti et al., 2024). Additionally, tumor cells evade immune surveillance by upregulating the surface expression of PD-L1, which interacts with programmed death-1 (PD-1) receptors on T cells to trigger the immune checkpoint response. Notably, tumor-derived exosomes express PD-L1 on their surface, which suppresses CD8 T cell function, thereby promoting tumor growth and influencing the outcome of anti-PD-1 therapy (Chatterjee et al., 2021). Furthermore, this mechanism is associated with the induction of systemic anti-tumor immunity and memory (Ayala-Mar et al., 2021). Exosomes are implicated in the induction of immune escape mechanisms in tumor cells, such as the evasion of immunogenic cell death through damage-associated molecular patterns (DAMPs) and the expression of surface recognition molecules including CD39/CD73, CD38, CD47, cytotoxic T-lymphocyte antigen 4 (CTLA-4, also known as CD152), lymphocyte-activation gene 3 (LAG-3), V-domain immunoglobulin suppressor of T cell activation (VISTA), and T cell immunoglobulin and mucin domain 3 (TIM-3) (Zhang et al., 2015; Wo et al., 2019). Furthermore, exosomes facilitate immune escape by delivering immunosuppressive molecules such as transforming growth factor beta (TGF-β), vascular endothelial growth factor (VEGF), interleukin 10 (IL-10), and prostaglandin E2 (PGE2) (Bojović et al., 2020; Mortezaee et al., 2023).

EMT represents a complex phenotypic transformation that enhances cellular motility and invasiveness by disrupting extensive epithelial cell-cell interactions, apical-basal polarity, and distinct cytoskeletal structures. As a pivotal event in tumor metastasis, EMT is associated with alterations in exosomal protein and RNA expression, which contribute to the promotion of EMT, as well as the migration and invasion of cancer cells (Lv and Xiong, 2024). Cancer-associated fibroblasts (CAFs) play a pivotal role in tumor progression by transporting exosomes to adjacent cells. These exosomes, expressed by CAFs, contain cargoes such as miR-210, which can facilitate EMT by targeting UPF1 and activating the PTEN/PI3K/AKT signaling pathway, thereby enhancing the migration and invasion of NSCLC (Yang et al., 2020). Similarly, miRNA-92a promotes EMT through the activation of the PTEN/PI3K/AKT signaling pathway in NSCLC metastasis (Lu et al., 2017). Other studies have demonstrated that exosome cargo reflects the TGF-β1-mediated EMT status in human lung AdCa cells (Kim et al., 2016) and contributes to resistance against osimertinib in epidermal growth factor receptor (EGFR)-mutant NSCLC (Hisakane et al., 2021).

Lung cancer remains the leading cause of cancer-related mortality, accounting for 24% of all cancer deaths. Metastasis, a critical phase in lung cancer progression, is associated with over 70% of these fatalities. Cancer-derived exosomes are regarded as significant drivers of cancer-induced pre-metastatic niche formation at distant sites, as well as metastasis (Li et al., 2024). Hoshino and colleagues have demonstrated that tumor-derived exosomes, including those originating from the lungs of both mice and humans, are taken up by organ-specific cells such as lung fibroblasts, epithelial cells, liver Kupffer cells, and brain endothelial cells, thereby facilitating the preparation of the pre-metastatic niche. Furthermore, treatment with exosomes from lung-tropic models was found to redirect the metastasis of bone-tropic tumor cells, with exosomal integrins α6β4 and α6β1 being associated with lung metastasis (Hoshino et al., 2015). Exosomal RNAs also play a significant role in the pre-metastatic and metastatic stages of lung cancer. For instance, circSATB2 is markedly upregulated in serum exosomes from lung cancer patients, demonstrating high sensitivity and specificity for clinical detection and being linked to lung cancer metastasis (Zhang et al., 2020).

Angiogenesis refers to the formation of new blood vessels from existing ones. Exosomes contribute to the angiogenic process in cancer progression by transporting various pro-angiogenic biomolecules, such as VEGF, matrix metalloproteinases (MMPs), and microRNAs, while also downregulating factors that inhibit hypoxia-inducible factor 1 (HIF-1) (Olejarz et al., 2020). Within the tumor microenvironment, the uptake of exosomes by normal endothelial cells activates angiogenic signaling pathways, leading to the formation of new blood vessels. Cancer-derived exosomal miR-25-3p facilitates the establishment of a pre-metastatic niche by enhancing vascular permeability and promoting angiogenesis (Zeng et al., 2018). The overexpression of miR-210 further augments angiogenesis, as it is transferred via exosomes derived from lung cancer cells, functioning as a proangiogenic factor in cancer-associated fibroblasts through modulation of the JAK2/STAT3 pathway (Fan et al., 2020). Additionally, miR-23a is markedly elevated in exosomes originating from lung cancer, directly targeting and suppressing prolyl hydroxylase 1 and 2 (PHD1 and 2) and inhibiting the tight junction protein ZO-1. This suppression leads to the accumulation of HIF-1α in endothelial cells, thereby enhancing angiogenesis under both normoxic and hypoxic conditions, and increasing vascular permeability and cancer transendothelial migration (Huang et al., 2024).

The concept of blood-based biomarkers is well-established. For several years, tumor protein biomarkers such as CEA, alpha-fetoprotein (AFP), and carbohydrate antigen 19–9 (CA19-9) have been extensively utilized in the screening and diagnosis of lung cancer, evaluation of treatment efficacy, postoperative monitoring, and prognostic assessment (Wojtalewicz et al., 2023; Bitter et al., 2023). Despite their utility in clinical practice, certain biomarkers lack specificity and may be upregulated in non-tumor conditions. Fortunately, alternative blood-based biomarkers, such as exosomes and their cargo, have demonstrated promising diagnostic accuracy, characterized by high sensitivity, specificity, reliability, and reproducibility for lung cancer. Consequently, these biomarkers continue to be actively investigated. Exosomes released into circulation possess the potential to provide insights into tumor status, as evidenced by the correlation between plasma exosomal microRNAs and the efficacy of immunotherapy in EGFR/ALK wild-type advanced NSCLC (Liu F. et al., 2024). Additionally, plasma exosomal proteins, including SRGN, TPM3, THBS1, and HUWE1, have been shown to differentiate lung AdCa from normal controls (Vykoukal et al., 2017).

However, tumor-derived exosomes represent only a small fraction of the total exosome population in blood, and the presence of abundant lipoproteins and platelets complicates the isolation of pure exosomes. To address these challenges, MISEV2023 guidelines recommend that studies report all relevant experimental conditions clearly, including anticoagulant type, centrifugation protocols, and methods used to isolate exosomes, ensuring that co-isolates like platelets and lipoproteins are effectively removed (Welsh et al., 2024). Despite many promising findings (e.g., protein or miRNA markers), most studies remain relatively small or single-center, and large-scale validation across diverse populations is still lacking.

Exosomes, along with other microvesicles, have been successfully isolated from human saliva and subjected to comprehensive characterization. These salivary extracellular vesicles contain informative proteins, RNAs, and DNAs that serve as potential biomarkers for the screening and early detection of lung cancer. Exosomes function as protective carriers for tumor cell-specific mRNAs and proteins in body fluids, potentially facilitating their transmission from blood to saliva. This suggests a link between distal tumor progression and biomarker discovery in saliva (Cui et al., 2024). Ultra-short circulating tumor DNA (usctDNA) has been predominantly localized within the exosomal fraction of saliva and plasma in patients with NSCLC, highlighting its potential as a novel target for liquid biopsy in lung cancer detection (Wang F. et al., 2024). One study successfully identified and quantified 785 proteins from salivary exosomes and 910 proteins from microvesicles, with 150 and 243 proteins, respectively, being identified as dysregulated candidates in the context of lung cancer (Sun et al., 2018). In a similar proteomic profiling study using liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) on saliva and serum exosomes from lung cancer patients, researchers identified 319 and 994 exosomal proteins, respectively, and discovered 11 potential lung cancer-specific candidates present in both body fluids (Sun et al., 2017).

Although saliva offers a non-invasive, easy-to-collect biofluid, its complexity poses challenges. Salivary exosomes contain proteins, RNAs, and DNAs that can serve as potential biomarkers. However, the complexity of saliva, which contains enzymes, antibodies, and microbial products, can obscure the detection of tumor-derived exosomes. Saliva flow rate, oral hygiene, food intake, and microbial contamination can all influence exosome isolation and cargo analysis. To minimize variability, studies should standardize collection protocols, assess the impact of food intake on saliva composition, and follow MISEV guidelines to ensure accurate reporting of collection methods and processing conditions (Welsh et al., 2024). To date, reports of salivary exosomal biomarkers for lung cancer remain scarce, and any findings should be viewed as preliminary until validated in larger, multi-center studies.

Urine, as one of the most accessible biofluids, has been increasingly utilized in clinical diagnostics and biomedical research. Over the past decade, urinary extracellular vesicles have been shown to reflect molecular processes and physiological and pathological conditions across various tissues. Consequently, several methods have been developed to isolate and characterize these vesicles to explore their clinical applications (Erdbrügger et al., 2021). Recent advancements in analytical methodologies, particularly those utilizing mass spectrometry (MS)-based proteomic analysis, have facilitated high-throughput and comprehensive proteome profiling of urinary extracellular vesicles. This has significantly contributed to the discovery, quantification, and characterization of cancer-specific exosome biomarkers. Urinary exosomes are emerging as a valuable source of biomarkers not only for lung cancer but also for bladder, prostate, kidney, esophageal, and colorectal cancers (Wang X. et al., 2024). A novel integrated microfluidic device has been developed specifically for the isolation and in situ detection of lung cancer-specific exosomes from patient urine samples. This device boasts a detection limit of fewer than 1,000 exosome particles per milliliter and has been validated using 500 μL urine samples from both lung cancer patients and control subjects. It demonstrates a promising capability to differentiate early-stage lung cancer patients from healthy individuals (Di Bella and Taverna, 2024).

Urine offers a completely non-invasive and easily repeatable source of exosomes, but in the context of lung cancer, its use faces several constraints. First, exosome yield and concentration in urine may be low, since vesicles must pass through renal filtration or derive from urinary tract cells rather than lung tissue, which reduces the “tumor fraction” — making lung-tumor–specific signals likely very diluted and potentially masked by exosomes from kidney or bladder cells (Martínez-Espinosa et al., 2024). Additionally, urine’s variable matrix (pH, osmolarity, presence of salts, urea, metabolites) may degrade or alter exosome integrity or cargo stability, complicating reproducible isolation (Martínez-Espinosa et al., 2024). Given the variability in urinary exosome levels, normalization strategies (e.g., urinary creatinine or total protein concentration) should be employed, as recommended by MISEV2023, to account for these differences (Welsh et al., 2024). Hence, while urinary exosomes are attractive for noninvasive diagnostics, evidence for lung cancer-specific markers in urine remains limited and requires rigorous validation in larger, well-controlled cohorts.

Over the past six decades, sputum cytology has been a fundamental component of lung cancer screening, complementing other diagnostic techniques such as chest X-ray, computed tomography (CT), fluorescence endoscopy, and LDCT (Adams et al., 2023). Recent research on pulmonary diseases and cancer has identified the presence of extracellular vesicles in airway secretions, including sputum, bronchoalveolar lavage fluid (BALF), nasal lavage (NL), and pharyngeal lavage (He et al., 2024). Given the pivotal role of extracellular vesicles from these secretions in the pathophysiology of pulmonary conditions, they are considered potential biomarkers. Substantial epidemiological evidence supports the association between idiopathic pulmonary fibrosis and lung cancer (Tzouvelekis et al., 2019). Exosomes isolated from the sputum of patients with idiopathic pulmonary fibrosis exhibited a significantly elevated level of exosomal miR-142-3p (8.06-fold increase, p < 0.0001), which was positively correlated with the percentage of macrophages in the sputum (Guiot et al., 2020). A related study similarly identified significant dysregulation in sputum exosomal miRNA levels between patients with lung fibrosis and healthy individuals, revealing a distinct miRNA signature (miR-142-3p, miR-33a-5p, let-7d-5p) that differentiates patients from healthy controls with high sensitivity and specificity. Notably, miR-142-3p and let-7d-5p are recognized for their roles in EMT, a critical process in lung tumorigenesis, while miR-33a-5p has been associated with liver fibrosis (Njock et al., 2019). These findings suggest the potential utility of sputum-derived exosomes as biomarkers for diagnosing and assessing the severity of lung conditions.

Sputum offers the advantage of being in direct contact with lung tissue, potentially providing a higher “tumor fraction” of exosomes. However, sputum samples are highly variable in volume and composition, often contaminated with saliva, oral microbes, and inflammatory cells, which can complicate exosome isolation (Welsh et al., 2024). To improve reproducibility and minimize contamination, standardized protocols for sputum collection, processing, and isolation are necessary. As with other biofluids, following MISEV guidelines and reporting all relevant metadata on sample collection and processing is essential to ensure consistency and transparency across studies (Welsh et al., 2024). Consequently, although this source is promising, existing data are preliminary, and large, well-controlled validation studies with rigorous methods are needed to confirm utility for early lung cancer diagnosis.

The exosomal lipid bilayer’s ability to protect miRNAs from degradation by cellular RNases in bodily fluids renders exosomal miRNAs an ideal source of non-invasive biomarkers for the early detection and prognosis of cancer (Linares-Rodríguez et al., 2025). The resemblance between circulating exosomal miRNAs and tumor-derived miRNAs enables the use of circulating exosomal miRNAs as a screening tool for lung cancer. A multicenter study including 3,102 participants derived from case–control and cohort datasets, using genome-wide miRNA profiling of blood samples, reported that a 15-miRNA signature distinguished individuals with lung cancer from non-cancer controls in a validation cohort, achieving an accuracy of 91.4% (95% CI, 91.0–91.9), a specificity of 93.5% (95% CI, 93.2–93.8), and a sensitivity of 82.8% (95% CI, 81.5–84.1) (Fehlmann et al., 2020). In the same study, a distinct 14-miRNA panel discriminated lung cancer patients from individuals with non-malignant lung diseases, with an accuracy of 92.5% (95% CI, 92.1–92.9), a specificity of 88.6% (95% CI, 88.1–89.2), and a sensitivity of 96.4% (95% CI, 95.9–96.9) (Fehlmann et al., 2020).

A separate serum-based investigation constructed a diagnostic model based on miR-1268b and miR-6075 using 2,588 miRNA expression profiles, and evaluated this model in a validation cohort of 1,358 lung cancer patients and 1,970 non-cancer participants, reporting both sensitivity and specificity of 99% (Asakura et al., 2020). These findings support the potential diagnostic value of exosomal miRNA signatures. Nevertheless, variations in study design, cohort composition, and validation strategies across studies may influence the reported performance and should be considered when interpreting these results. In particular, highly selected populations and modeling strategies may contribute to optimistic estimates of diagnostic accuracy. Further evaluation in large, prospective screening cohorts will be critical to determine their real-world clinical performance.

Beyond miRNAs, extracellular vesicle-associated miR-10b has been reported to show strong diagnostic performance for lung AdCa, achieving an AUC of 0.998 when compared with established serum tumor markers, including AFP, CEA, CYFRA21-1, NSE, and others (Yuan et al., 2021). In addition, the exosomal long non-coding RNA growth arrest-specific transcript 5 (GAS5) was found to be downregulated in non-small cell lung cancer compared with healthy controls, and the combination of exosomal GAS5 with CEA yielded an AUC of 0.929, with more modest performance for stage I disease (AUC 0.822) (Li et al., 2019a).

Exosomal lncRNAs AGAP2-AS1 and TBILA have also been reported to be overexpressed in non-small cell lung cancer compared with healthy individuals. Receiver operating characteristic analyses indicated moderate discriminatory performance, with reported sensitivity and specificity values ranging from approximately 65%–67% and 73%–81% (Tao et al., 2020), respectively. Although these findings are encouraging, differences in study populations, sample sizes, and analytical approaches across studies highlight the need for further validation. These observations underscore the importance of well-designed prospective screening studies to minimize potential bias and to establish the clinical utility of exosome-based biomarkers for early lung cancer detection.

Exosomes are known to contain specific surface and intracellular protein molecules that are implicated in the initiation and progression of lung cancer. Consequently, they are being investigated as potential biomarkers for the early diagnosis and prognosis of this disease (Kujtan and Subramanian, 2019). Several exosomal membrane protein markers, including EGFR, CD91, and CD317, have been identified as potential diagnostic biomarkers for lung cancer (Luo et al., 2020). In evaluating the utility of exosomal protein profiling for diagnosing lung cancers across all stages, the researcher isolated plasma exosomes from a cohort of 581 participants, comprising 431 individuals with lung cancer and 150 control subjects. The findings indicated that the markers CD151, CD171, and tetraspanin 8 were the most effective in distinguishing patients with cancer, regardless of histological subtype, from those without cancer. Additionally, a multi-marker model incorporating ten markers achieved an AUC of 0.76 specifically for AdCa and an AUC of 0.74 for all lung cancer histological subtypes, underscoring the potential of exosomal protein profiling in the early diagnosis of lung cancer (Sandfeld-Paulsen et al., 2016). A study examining the correlation between exosomes derived from NSCLC and potential protein markers utilized unique Raman scattering profiles and principal component analysis (PCA) for cancer diagnosis. This research identified distinct peaks that showed a strong correlation with the ratio of cancerous exosomes (R² > 90%), representing a unique Raman band specific to NSCLC exosomes. Further analysis of the Raman bands in conjunction with exosomal protein markers (CD9, EpCAM, CD81, and EGFR) demonstrated that EGFR exhibited a 1.97-fold greater similarity in Raman profiles compared to the other markers (Shin et al., 2018). These findings suggest the potential role of exosomal surface protein markers in cancer diagnosis.

Additionally, a Triple SILAC quantitative proteomic approach was employed to investigate differential protein abundance between exosomes from normal bronchial epithelial cells and those from NSCLC, as a potential screening tool for early lung cancer detection. The data revealed a protein profile associated with NSCLC exosomes, including proteins such as EGFR, GRB2, and SRC, among the 721 exosomal proteins quantified (Clark et al., 2016). The enrichment of these exosomal protein markers in NSCLC positions them as promising candidates for lung cancer detection through a multi-marker protein panel. In a separate study, 25 proteins from salivary exosomes and 40 from microvesicles in lung cancer patients were identified as originating from distal organ cells, with 5 out of the 25 and 9 out of the 40 proteins being associated with lung tissue. Proteins such as BPIFA, MUC5B, CRNN, and IQGAP warrant further investigation for their potential role in lung cancer screening (Sun et al., 2018). Additionally, exosomal leucine-rich α-2-glycoprotein (LRG1) has been reported as a potential biomarker for lung cancer detection, capable of effectively distinguishing early-stage lung cancer from normal controls using urine samples (Di Bella and Taverna, 2024; Li et al., 2011).

The lipid bilayer of exosomes, estimated to be 5 nm in thickness, comprises approximately 53,400 lipid molecules and constitutes about 37% of the total volume of a 70-nm-sized exosome. This bilayer, containing lipid classes such as cholesterol, phosphatidylcholine, sphingomyelin, phosphatidylserine, and phosphatidylethanolamine, plays a crucial role in protecting the internal components from rapid degradation (Zhou et al., 2023). Notwithstanding, there is a paucity of studies addressing the role of exosomal lipids in early tumor detection. However, recent investigations have highlighted the emerging potential of these lipids in cancer screening and diagnosis. For instance, exosomes isolated from patient blood samples have been subjected to direct analysis via MS fingerprinting without any preparatory treatment, facilitating their application in cancer classification through mathematical analysis and enabling the qualitative and semi-quantitative detection of cancer biomarkers (Zhou et al., 2023; Skotland et al., 2023). The lipid profiles of blood plasma exosomes have been investigated for the early detection of NSCLC utilizing ultra-high-resolution Fourier transform mass spectrometry (UHR-FTMS). The researchers employed multivariate statistical methods to identify the top 16 lipids, which included 9 phosphatidylcholines, 2 sphingomyelins, 2 triacylglycerides, 1 cholesterol ester, 1 lysophosphatidylcholine, 1 lysophosphatidylcholine-plasmalogen, and 7 additional lipid features. The findings indicated that the area under the receiver operating characteristic curve (AUROC) for distinguishing early and late-stage cancer from normal subjects, utilizing the selected lipid features, was 0.85 and 0.88 for the Random Forest (RF) model, and 0.79 and 0.77 for the Least Absolute Shrinkage and Selection Operator (LASSO) model, respectively (Fan et al., 2018). Table 1 provides a summary of studies investigating exosomal components in the context of early lung cancer detection.

MS is an analytical technique that ionizes chemical species and separates them according to their mass-to-charge ratio. Due to its high specificity and sensitivity, MS is capable of identifying and characterizing the molecular composition of exosomes, including proteins, lipids, and metabolites (Zhuang et al., 2024). LC-MS/MS has been utilized to conduct comparative proteomic analyses of exosomes and microvesicles in human saliva for lung cancer (Sun et al., 2018), to profile the proteome of human urinary exosomes (Pan et al., 2024), and to analyze plasma exosomes in cancer (Bhavsar et al., 2024). Ultra-high-resolution Fourier transform mass spectrometry (UHR-FTMS) was employed to examine exosomal lipids for the classification of early and late-stage NSCLC, utilizing multivariate statistical methods such as LASSO. This methodological approach successfully differentiated early-stage lung cancer patients from healthy individuals (Fan et al., 2018). Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) facilitates the detection of intact exosomes, producing exosomal fingerprints within minutes. This rapid detection method is proposed as a promising tool for cancer biomarker research. MALDI-TOF MS enables the analysis of intact exosomes without necessitating their lysis or labeling, thereby serving as a robust instrument for examining biological samples for diverse clinical applications (Duncan et al., 2016). Furthermore, this technique can be employed for exosome proteome profiling by detecting protein extracts, which holds potential for cancer diagnostic and prognostic biomarker identification (Alberca-Del Arco et al., 2024).

ELISA, a widely utilized immunological technique for the quantification of antibodies, antigens, proteins, and glycoproteins in biological samples, has been extensively applied in the study of cancer exosomes. The use of ELISA-based methods for the quantification of proteins specific to cancer-derived exosomes, such as CD44, CD44v6, CD9, EpCAM, and CD81, has been well-documented (Giampieri et al., 2019). Additionally, ELISA is employed to measure cytokines and chemokines expressed by exosomes (Wang et al., 2020). The quantification of low concentrations of specific exosomes in minimal volumes of clinical samples holds potential for noninvasive cancer diagnosis and prognosis. Liu et al. developed an immunosorbent assay for the digital quantification of target exosomes utilizing droplet microfluidics. In this method, exosomes are immobilized on magnetic microbeads via sandwich ELISA complexes, which are tagged with an enzymatic reporter that generates a fluorescent signal. This droplet-based single-exosome-counting ELISA technique enables the absolute quantification of cancer-specific exosomes, achieving unprecedented accuracy with a detection limit of 10 enzyme-labeled exosome complexes per microliter (approximately 10^-17 M) (Pengtao et al., 2024). Ongoing research and advancements in this technique have led to the development of more refined models for specific applications, such as tumor screening and diagnosis. For example, a cell surface ELISA in suspension method was evaluated for exosome detection, demonstrating its suitability for identifying exosome particles within biological samples. This method can be adapted to assess exosome interactions with target cells in both diagnostic and therapeutic contexts (Malaguarnera and Cabrera-Pastor, 2024).

Due to the absence of exosome-specific markers, proteins that are ubiquitously enriched in exosomes from various cellular origins are frequently utilized for exosome detection. Consequently, the markers used for exosome detection may differ depending on the originating cells. Typically, Western blot analysis of exosomes employs markers from each of the categories recommended by the International Society for Extracellular Vesicles (ISEV) for the characterization of extracellular vesicles. These include tetraspanins such as CD9, CD63, and CD81; intracellular proteins associated with multivesicular body (MVB) formation, such as tumor susceptibility gene 101 (TSG101); and chaperone proteins such as heat-shock proteins (HSP) 90 and HSP70. Additionally, some detection kits may include antibodies against the marker Alix, also known as PDCD6IP. As a conventional method for protein detection, Western blotting has been extensively employed in numerous studies to evaluate exosomes expressed in lung tumors through their protein constituents (Liang et al., 2020; Li et al., 2019b; Chen et al., 2024). In various applications, Western blotting has been employed to verify the distinct expression of tumor exosomal surface/membrane proteins, such as glypican 1 (GPC-1), glucose transporter 1 (GLUT-1), and disintegrin and metalloproteinase domain-containing protein 10 (ADAM10), which are considered potential biomarkers for cancer diagnosis and prognosis (Xu F. et al., 2024).

Flow cytometry is utilized to measure the light scattering and fluorescence signals of individual exosomes. As reported by Kuiper and colleagues, flow cytometry is the sole technique capable of detecting, sizing, and phenotyping millions of extracellular vesicles within a matter of minutes (Bettin et al., 2023). While single-particle measurement using flow cytometry is advantageous, the diminutive size of exosomes and the low abundance of surface antigens pose significant challenges to conventional flow cytometry methods. This has led to the development of vesicle-specific assays and experimental designs tailored for exosomal studies. For example, the implementation of fluorescence-triggered Vesicle Flow Cytometry (VFC), a comprehensive approach for the quantitative assessment of extracellular vesicle number, size, and surface marker expression, addresses these challenges effectively (Kolenc and Maličev, 2024). Quantification is crucial for comprehending the fundamental biological interactions between exosomes and their parent cells, as well as for interpreting exosome signaling. Furthermore, the quantification of the molecular cargo within exosomes via on-bead flow cytometry is imperative for elucidating their role in information transfer and human disease. Recent research has demonstrated that on-bead flow cytometry, standardized for use with conventional cytometers, is an effective method for the detection and quantification of proteins in exosomes isolated from cell line supernatants or cancer patient plasma (Hong et al., 2024). In this method, exosomes isolated through size-exclusion chromatography are captured using biotinylated antibodies specific to exosomal cargo antigens and immobilized on streptavidin-labeled beads. Detection is subsequently achieved with pre-titrated fluorochrome-labeled antibodies of the desired specificity. Additionally, subpopulations of extracellular vesicles and particles derived from cancer cells have been characterized using nano-flow cytometry (Liu et al., 2022).

Among the newly developed technologies for the detection of cancer-derived exosomes, Surface-Enhanced Raman Spectroscopy (SERS) is regarded as one of the most sensitive, reliable, and selective methods for non-destructive molecular analysis. This is achieved through the enhancement of electromagnetic fields and/or the formation of charge-transfer states between the chemisorbed analyte molecule and the SERS-active substrate (Su et al., 2023). In the context of lung cancer, researchers observed a significant variation in the SERS signals of proteins, nucleic acids, and lipids, distinguishing not only between exosomes from normal lung tissue and those from cancerous lung tissue but also between exosomes from fibroblasts and epithelial cells. Further investigation demonstrated that exosomes derived from lung cancer exhibited a pronounced protein SERS signal during time-dependent SERS analysis, whereas exosomes from normal lung tissue predominantly showed SERS signals associated with proteins, lipids, and nucleic acids (Lyu et al., 2024). A label-free and highly sensitive approach for classifying exosomes, which integrates SERS with statistical pattern analysis, effectively differentiated exosomes derived from lung cancer cells from those derived from normal cells, achieving a sensitivity of 95.3% and a specificity of 97.3%. Furthermore, Principal Component Analysis indicated that the observed differences were attributable to 11 distinct points in the SERS signals of exosomes from lung cancer cells (Park et al., 2017). Similarly, a recent investigation demonstrated the precise diagnosis of early-stage lung cancer through the application of deep learning-based SERS on exosomes derived from both normal and lung cancer cell lines. The model exhibited an accuracy of 95% and successfully predicted that 90.7% of plasma exosomes (including those from stage I and II cancer patients) exhibited greater similarity to lung cancer cell exosomes compared to the average healthy controls, with an AUC of 0.912 for the entire cohort and 0.910 for stage I patients (Shin et al., 2020). These findings suggest the potential utility of combining exosome analysis with deep learning as a method for early-stage liquid biopsy in lung cancer diagnostics.

EFIRM is a technique designed for the selective extraction of exosomes from biofluids, facilitating the release of their cargo for analysis of internal RNA and protein content. This method has been investigated for its potential in the early diagnosis of lung cancer. EFIRM has demonstrated the capability to show that exosome-like microvesicles transport tumor cell-specific mRNA and protein from blood to saliva in a xenografted mouse model of human lung cancer (Cui et al., 2024). Mutations in the EGFR gene are known to predict sensitivity to EGFR-targeted therapies for NSCLC. Recent advancements indicate that EFIRM can detect EGFR mutations with 100% sensitivity in both plasma and saliva samples from NSCLC patients. Deep sequencing analysis of circulating tumor DNA (ctDNA) enriched for the EGFR L858R mutation revealed a significant presence of EGFR L858R ctDNA as ultrashort ctDNA (usctDNA) with sizes ranging from 40 to 60 base pairs in patient plasma and saliva, with the majority of usctDNAs predominantly localized within the exosomal fraction (Wang F. et al., 2024). Saliva-based EFIRM liquid biopsy has been employed alongside other methodologies to serially assess ctDNA in lung AdCa, serving as a predictor of treatment response and resistance (Kim C. et al., 2021), as well as enabling the specific detection of both exosomal RNA and protein biomarker targets (Tu et al., 2015). The EFIRM method was utilized for the analysis of exosomal profiles in mice injected with human lung cancer H640 cells, a cell line engineered to express the exosome marker human CD63-GFP. The findings indicated elevated expression levels of exosomal biomarkers in both serum and saliva samples of the mice injected with human lung cancer cells, suggesting the potential of salivary exosome biomarkers for the detection of distal diseases (Tu et al., 2015).

Lateral flow immunoassay (LFIA), recognized as one of the most prevalent point-of-care testing techniques, is extensively used as portable strips for biological detection due to its rapidity, cost-effectiveness, and specificity for target biomarkers (Moghadam et al., 2015). This appealing immunochromatographic strip test facilitates easy use and rapid detection of exosomes, even by untrained personnel. Recent studies have incorporated this technique into rapid diagnostic test kits for lung cancer, highlighting its advantages of straightforward operation, high sensitivity, and affordability (Zhou et al., 2020). The techniques, which are predicated on the detection of exosomal miRNA in clinical salivary and urine samples, demonstrated high selectivity and sensitivity, aligning with the standard quantitative real-time polymerase chain reaction (qRT-PCR) method, and exhibited a detection limit of 7.76 × 10⁹ (Zhou et al., 2020). This represents a promising strategy for the screening, monitoring, and differentiation of lung cancer exosomes from normal cell exosomes, thereby serving as a potential diagnostic tool. Purified exosomes from human plasma and urine can be detected within 15 min, with a detection limit of 8.54 × 10^5 exosomes/µL when a combination of anti-CD81 and anti-CD9 is employed as capture antibodies, and anti-CD63 labeled with gold nanoparticles is utilized as the detection antibody (Oliveira-Rodríguez et al., 2016). The challenge of low exosome detection by colorimetric-based lateral-flow assays can be addressed by employing highly bright multi-quantum dots embedded in silica-encapsulated nanoparticles (M-QD-SNs), which enhances the sensitivity of the techniques, achieving an exosome detection limit of 117.94 exosomes/μL. Furthermore, the resultant method for exosome detection is not only highly sensitive and quantitative but also accurate, rapid, and selective (Kim H. M. et al., 2021). In the context of tumor-specific exosome detection, particularly in melanoma, techniques such as LFIA and ELISA have demonstrated efficacy in identifying scarce tumor antigens within exosomes. These methods also successfully detected the human NKG2D-Ligand, MICA, within tetraspanin-containing nanovesicles, as reported in study (López-Cobo et al., 2018), This suggests the potential utility of these techniques for analyzing biological samples to identify tumor-derived exosomes.

Microfluidics, a multidisciplinary field encompassing engineering, chemistry, physics, biochemistry, biotechnology, and nanotechnology, involves the behavior, manipulation, and precise control of fluids confined to small scales where surface forces predominate over volumetric forces. This field has significant practical applications, particularly in the development of systems designed for processing low volumes of fluids, thereby enabling automation, multiplexing, and high-throughput screening, with notable implications for cancer screening (Garcia-Cordero and Maerkl, 2020). An integrative microfluidic platform has been developed for the isolation and ultrasensitive detection of lung cancer-specific exosomes from patient urine. This device comprises a section modified with a capture antibody and a second antibody-conjugated gold (Au) nanorod probe, which is utilized to identify and quantify lung cancer-specific exosomes via dark field microscopy. The resulting AuNC-Exosome-AuR complex induces a notable scattering wavelength shift and enhanced scattering intensity, enabling the ultrasensitive detection of exosomes with a limit of detection (LOD) below 1,000 particles/mL (Di Bella and Taverna, 2024). Validation using 500 μL urine samples from lung cancer patients demonstrated the device’s efficacy in distinguishing early-stage lung cancer patients from healthy controls. Similarly, a magnetic-based microfluidic platform, known as the ExoPCD-chip, has been reported for the on-chip isolation and detection of tumor-derived exosomes. This device efficiently captures exosomes, achieving highly sensitive detection of CD63-positive exosomes at concentrations as low as 4.39 × 10^3 particles/mL (Xu et al., 2018). Utilizing human serum from patients with liver cancer, the ExoPCD-chip demonstrated a robust ability to distinguish between cancer patients and healthy controls. Comparable studies have documented the isolation and profiling of circulating tumor-associated exosomes through the use of an extracellular vesicular lipid-protein binding affinity-based microfluidic device (Kang et al., 2019) and a detachable microfluidic device equipped with an electrochemical aptasensor (DeMEA) for the sequential analysis of cancerous exosomes (Feng et al., 2023).

Numerous alternative methods for exosome detection have been documented and are continually evolving. A recent investigation introduced an electrochemical micro-aptasensor, designed as a highly sensitive approach for exosome detection by integrating a micropatterned electrochemical aptasensor with a hybridization chain reaction (HCR) for signal amplification. In this methodology, exosomes are concentrated on electrodes functionalized with CD63 aptamers and subsequently identified by HCR products linked to avidin-horseradish peroxidase (HRP) via EpCAM aptamers as intermediaries. Signal generation is achieved through the enzymatic reaction between the HRP enzyme and 3,3′,5,5′-tetramethylbenzidine (TMB)/H2O2, which is directly proportional to the quantity of HRP bound to the HCR products, thereby reflecting the concentration of target exosomes (Zhang et al., 2021). This technique exhibits a detection limit of 5 × 10² exosomes/mL and effectively identifies lung cancer exosomes in serum samples from patients with both early-stage and late-stage lung cancer. Alternative ultrasensitive electrochemical aptasensors for tumor exosome detection have been developed utilizing click chemistry, offering a detection range from 1.12 × 10² to 1.12 × 10⁸ particles/μL and a detection limit of 96 particles/μL (An et al., 2019). Additionally, a sensitive and selective colorimetric aptasensor has been engineered for the detection of cancer-derived exosomes, employing HRP-accelerated dopamine polymerization and in situ polydopamine deposition, achieving a detection limit of 7.7 × 10³ particles/mL, which represents an improvement of 3-5 orders of magnitude over conventional Dot-blot methods (Küçük et al., 2024). Furthermore, the integration of micronuclear magnetic resonance and microfluidics constitutes another sensitive platform for the detection of extracellular vesicles, including exosomes. In this approach, exosomes are preconcentrated, purified, and labeled with magnetic nanoparticles (MNPs), which are then directed into the microfluidics system, where the labeled exosomes are subsequently trapped on a membrane filter (Mohammadi et al., 2021). Table 2 presents some of the alternative techniques and their diagnostic features in lung cancer.