The clinical significance of circulating microRNAs as biomarkers in lung cancer diagnosis and prognosis

Abstract

Lung cancer continues to be a major cause of deaths associated with cancer, with a considerable number of patients being diagnosed at more advanced stages of the disease. Circulating microRNAs (miRNAs), which are stable in body fluids, offer promising avenues for non-invasive biomarker identification. Recent advancements have highlighted the utility of specific miRNA signatures in treatment decisions and prognostic evaluations. This review has explored the significant function of circulating miRNAs as promising diagnostic and prognostic biomarkers for lung cancer. Integration of miRNA profiles with clinical parameters has shown potential in predicting treatment responses and patient outcomes. Nevertheless, obstacles persist in standardizing the protocols for isolating and quantifying miRNAs, along with the variability of miRNA expression across various cancer types and stages. Future research focusing on larger, multi-center studies is essential to validate the diagnostic and prognostic capacities of circulating miRNAs and enhance their applicability in clinical practice.

Introduction

Lung cancer is one of the most prevalent cancers, accounting for 11% of all newly diagnosed cases. It remains the primary cause of cancer-related mortality, responsible for over 15% of all cancer deaths [1]. Based on histopathological classification, lung cancer is classified into non–small-cell lung carcinoma (NSCLC) and small-cell lung carcinoma (SCLC) [2]. NSCLC constitutes approximately 85% of all cases, which mainly includes lung adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC). Clinically, many lung cancer patients may be asymptomatic in the early stages of the disease [3]. The current main treatment modalities include surgery, immunotherapy, radiotherapy, chemotherapy and targeted therapy. However, the five-year survival rate for individuals diagnosed with lung cancer continues to be notably low [4]. Patients diagnosed with lung cancer frequently present at an advanced stage, primarily because there are no early specific symptoms [3]. This leads to a poor prognosis, with a five-year survival rate of approximately 7% among those in the advanced stage [3, 5, 6]. To markedly enhance lung cancer outcomes, it is essential to implement robust early screening strategies, ensure precise diagnostic assessments, and tailor therapeutic strategies to each patient.

Early implementation of screening protocols is critical for reducing lung cancer–related mortality and enhancing patient survival [7, 8]. Moreover, alternative combination regimens such as pairing immune checkpoint inhibitors with chemotherapy and the refinement of treatment schedules, for instance through perioperative interventions and biomarkers driven adjuvant protocols, are poised to deliver further improvements in lung cancer prognosis [9, 10]. Nonetheless, these developments highlight the pressing need for biomarkers capable of assessing treatment response, detecting emerging resistance, and signaling postoperative relapse.

MicroRNAs (miRNAs) are single-stranded non-coding RNAs with approximately 21–25 nucleotides in length [11]. MiRNAs play an essential role in cancer progression by modulating gene expression post-transcriptionally by hybridizing to complementary regions located in the 3′untranslated region (UTR) of target messenger RNAs (mRNAs) [12]. Emerging evidence has revealed that miRNAs serve as master regulators of oncogenesis through regulating the onset, advancement, and spread of tumors [13]. And their dysregulated expression across multiple malignancies, including lung cancer, has highlighted their promise as diagnostic and prognostic indicators [14–16]. In the present review, we provide an overview of the function of circulating miRNAs in lung cancer, highlighting their clinical significances in the diagnosis, prognosis, and prediction of lung cancer, and evaluate their potential to tailor lung cancer therapies.

Circulating micrornas: biosynthesis, functional roles and detectable methods

MiRNAs were first described by Lee et al. in 1993, are 21–25 nucleotide long hairpin-loop RNAs transcribed by RNA polymerase II as primary miRNA (pri-miRNA) transcripts [17, 18]. Then, nuclear RNase III DROSHA cleaves pri-miRNAs at the base of their hairpin stems, producing 60–70 nucleotide long precursor miRNAs (pre-miRNA), which are subsequently transported into the cytoplasm by exportin-5 receptor [19]. In the cytoplasm, pre-miRNAs undergo further cleavage by the RNase III enzyme Dicer, which operates in conjunction with RNA-binding cofactors such as TAR RNA binding protein, which resulting mature miRNA duplex [19]. Mature miRNAs are generated from either the 5′ or 3′end of the pre-miRNA hairpin, yielding the − 5p or − 3p isoforms of miRNAs [20]. Mature miRNAs associate with an Argonaute (AGO) protein to form the mature miRNA-induced silencing complex (miRISC), which represses translation and promotes decay of target mRNAs by outcompeting eukaryotic initiation factor 4 F (eIF4F) for the 7-methylguanosine cap, it blocks cap-dependent initiation and recruits deadenylase complexes that remove the 3′poly(A) tail, leading to target mRNA destabilization [21, 22] (Fig. 1).

Fig. 1.

Mechanism of miRNA biogenesis and function in tumor biology. The transcription of miRNA genes by RNA Polymerase II (Pol II) produces primary miRNAs (pri-miRNAs) in the nucleus. These pri-miRNAs are processed into precursor miRNAs (pre-miRNAs) and subsequently matured into functional miRNAs. Mature miRNAs are incorporated into the miRNA-induced silencing complex (miRISC) in the cytoplasm. The miRISC complex targets specific messenger RNAs (mRNAs) leading to either their degradation or translational repression. Through these mechanisms, miRNAs play critical roles in various cellular processes relevant to cancer, including inducing angiogenesis, facilitating immune escape, resisting apoptosis, and activating metastasis, thereby contributing to tumor progression and development

MiRNAs regulate the expression of approximately 60% of human genes by binding complementary sites with target mRNAs [23]. Formation of these miRNA–mRNA duplexes promotes transcript degradation or directly represses translation [24, 25]. Each miRNA has the ability to regulate various mRNA targets, with target specificity primarily determined by the base-pairing interactions between the miRNA seed region and complementary sequences in the 3′ UTR of the transcript [26]. For example, miR-137 could regulate the expression of several genes, including RNF4 [27], triggering transposable element-derived 1 (TIGD1) [28], and lysosome Associated Protein Transmembrane 4B [29]. In contrast, individual mRNAs can simultaneously target mRNAs, creating a complex regulatory network that governs gene expression [30]. In cancer, miRNAs act as either tumor suppressors or oncogenes depending on their targets and cellular context. Aberrant miRNA expression contributes to unlimited proliferation, tumor angiogenesis, invasion and metastasis [31, 32]. Many miRNA loci reside in genomic regions prone to deletion or amplification in tumors, and microarray-based profiling consistently reveals marked differences in miRNA signatures between malignant and normal tissues across multiple types of human cancers [33–35].

MiRNAs could be released into extracellular fluids referred to as circulating microRNAs, which have been identified in human plasma, urine, and serum [36–39]. In the context of cancer, the mature miRNAs detected in plasma and serum have been largely ascribed to passive release from apoptotic or necrotic cells [40]. Beyond this mechanism, active cellular secretion pathways for miRNAs have also been documented [41]. Many circulating miRNAs are reported to be secreted into the extracellular environment in complex with AGO2 [42] and high-density lipoproteins [43]. Additionally, miRNAs could be packaged into cell membrane–derived extracellular vesicles (EVs) and released into the extracellular milieu [44]. These interactions enable the stability of circulating miRNAs in the extracellular environment.

Recent investigations have applied quantitative reverse transcription polymerase chain reaction (qRT-PCR) and Northern blotting to quantify circulating miRNA levels in the blood of cancer patients [45, 46]. Emerging platforms such as microarrays and next-generation sequencing permit comprehensive profiling of these miRNAs [47, 48]. The lineage and tissue-specific expression profiles of circulating miRNAs render them effective disease biomarkers, and their exceptional stability enhances their reliability as markers in liquid biopsy.

Circulating miRNAs for lung cancer subtype classification

Lung cancer is a heterogenetic malignant disease with various pathological subtypes, while miRNA expression profiles can facilitate the classification of its subtypes. Research has shown that miRNA signatures are highly specific to histological subtypes, enabling accurate differentiation among various forms of lung cancer. These advancements are essential for guiding treatment decisions—especially between LUSC and LUAD, which arise from different lung cell types and require distinct therapeutic strategies. For example, miR-375, miR-203, and miR-205 were differentially expressed miRNAs that could differentiate LUSC from other subtypes of NSCLC cases [49]. Moreover, a recent study using machine learning approaches identified miR-944 and miR-205 as significant markers for categorizing tumors into LUAD and LUSC subtypes [50]. Notably, certain miRNAs not only distinguished SCLC from healthy individuals but also differentiated SCLC from other types of lung cancer. A model that incorporated miR-375-3p, miR-320b, and miR-144-3p, along with race and age, might be utilized to diagnose patients with metastatic SCLC [51]. Additionally, miR-17, miR-190b, and miR-375 demonstrated the ability to accurately differentiate SCLC from NSCLC, yielding an area under the curve (AUC) value of 0.869 [52]. Additionally, SCLC and SCC might exhibit similarities in imaging, as tumors could exhibit similar locations in lungs. On the other hand, SCC could occasionally share characteristics with SCLC cases, including small cells with hyperchromatic nuclei and scant cytoplasm, which may result in misdiagnosis. In this context, Saviana et al. identified that miR-375 was potent classifier for distinguishing between SCLC and SCC [51]. Moreover, miRNA expression analysis could successfully distinguish primary lung tumors from metastatic lesions arising in other organs. MiR-182 exhibited significant upregulation in primary lung tumors, while miR-126 was more commonly found in lung metastatic tissues that were originating from other organs [53] (Table 1). However, there are several key issues that are needed to be addressed. Researches should validate cross-cohort reproducibility and cross-platform/matrix harmonization of subtype signatures. Moreover, whether a minimal, generalizable “core” panel exists that truly complements or outperforms cytology, IHC, and imaging that could contribute to lung cancer subtype classification [54].

Table 1.

Circulating MiRNA biomarkers in lung cancer

MiRNA	Sample Type	Clinical significance	AUC	References
miR-375, miR-203, miR-205	Serum	Differentiated LUSC from other types of NSCLC cases	0.833	[49]
miR-944	Plasm	Categorizing tumors into LUAD and LUSC subtypes	0.916	[49]
miR-375-3p, miR-320b, miR-144-3p	plasma	Diagnosed patients with metastatic SCLC	0.882	[51]
miR-17, miR-190b, miR-375	plasma	Differentiated SCLC from NSCLC	0.878	[52]
miR-182	Plasm	Differentiated primary lung tumors	/	[53]
miR-126	Plasm	Differentiated metastatic lung tumors	/	[53]
miR-146a, miR-222, miR-223	Serum	Distinguishing LUAD patients from healthy patients	0.951	[61]
miR-492, miR-590-3p, miR-631	Serum	Diagnosed the early stage of NSCLC	0.828	[62]
miR-193b, miR-200b	Serum	Diagnosed NSCLC	0.985	[63]
miR-21, miR-486, miR-375, miR-200b	Sputum	Diagnosed LUAD	0.917	[64]
miRNA-196b-5p	Plasm	Diagnosed NSCLC	/	[65]
miR-3565, miR-3126, miR-200b	Serum	Diagnosed SCLC	0.93	[67]
miR-483-3p	Plasma	Diagnosed early-stage SCL	0.758	[68]
miR-1228	Serum	Diagnostic biomarker for SCLC	/	[69]
miR-93-5p, miR-29c-3p, miR-449-5p,	Serum	Diagnosed NSCLC	0.76	[73]
miR-15a-5p, miR-93-5p, miR-29c-3p, miR-449-5p	Serum	Diagnosed LUAD	0.84	[73]
miR-93-5p, miR-29c-3p, miR-15a-5p	Serum	Diagnosed LUSC	0.7363	[73]
miRNA-148a	Plasm	Associated with lymph node metastasis and poor clinical outcomes	/	[77]
miRNAs-19a-3p, 126-5p, 556-3p, 671-5p, 937-3p, 4664-3p, 4746-5p	Plasm	Associated with overall survival in LUAD patients	0.617	[78]
miR-21, miR-141, miR-490	Serum	Associated with poor clinical outcomes	/	[79]
miR-155	Serum	Associated with poor clinical outcomes	0.87	[80]
miR-942, miR-601	serum	Associated with poor clinical outcomes	0.813	[81]
miR-637	Plasm	Downregulation of miR-637 in NSCLC was associated with poor prognosis of patients	/	[82]
miR-1249-3p	Serum	Increased expression of miR-1249-3p was associated with better responses to chemotherapy	/	[85]
miR-25, miR-145, miR-210	Serum	Functioned as predictors for the efficacy of maintenance treatment with pemetrexed in LUAD patients	/	[86]
miR-30c	Serum	Predicted chemoradiotherapy resistance	0.872	[89]
Hsa-miR-320d, hsa-miR-320c, and hsa-miR-320b	Plasm	Predicted the efficacy of immunotherapy	/	[90]
miR-105-5p, miR-767-5p	Plasm	Predicted the efficacy of immunotherapy	0.85	[91]
miR-323-3p, miR-1468-3p, miR-5189-5p, miR-6513-5p	Exosomal microRNAs as potential biomarkers for osimertinib resistance of non-small cell lung cancer patients	Assocaited with Osimertinib resistance	0.846	[92]

MiRNA microRNA, NSCLC non–small-cell lung carcinoma, SCLC small-cell lung carcinoma, LUAD lung adenocarcinoma, LUSC squamous cell carcinoma

Effectiveness of circulating MiRNAs for lung cancer diagnosis

Tissue biopsy remains the definitive method for confirming lung malignancy, and obtaining sufficient lung parenchymal material at first sampling is essential for accurate histopathological subtyping [55]. Securing an adequate initial specimen minimizes the need for repeat procedures—which carry additional risk and delay therapy—and ensures timely initiation of treatment. Commonly employed techniques for obtaining lung tissue or fluid samples include fiber-optic bronchoscopy, which can be performed with or without transbronchial needle aspiration. Other techniques are endobronchial ultrasound-guided biopsy, image-guided percutaneous transthoracic needle aspiration, mediastinoscopy, and thoracentesis followed by pleural fluid analysis. Additionally, thoracoscopic biopsy and both open and video-assisted surgical methods are also employed [56, 57]. Although these modalities provide high diagnostic yield, they are resource-intensive, associated with procedural complications, and may sometimes yield insufficient tissue, necessitating further sampling [58]. It is important to note that despite the standardization of immunohistochemical (IHC) analysis of tissue specimens, diagnosing lung cancer remains challenging. A review of pathology for 338 lung cancer patients revealed that different pathologists provided uncertain diagnoses in 16 cases, which could lead to variations in treatment plans [59]. Inaccurate diagnoses often result from the exclusion of squamous carcinoma markers in the IHC panel, the presence of neuroendocrine tumors or poor differentiation [59]. Consequently, there has been an increasing emphasis on evaluating and implementing adjunctive diagnostic tools in clinical practice, such as gene expression profiling to support the diagnosis of complex cases [60].

Recently, several studies have identified miRNAs in tissue samples of NSCLC, and some promising studies have detected changes in circulating miRNAs in blood samples. These findings may pave the way for the development of non-invasive detection methods. For example, Lv and colleagues have investigated the profiles of serum miRNAs that were obtained from early-stage LUAD patients [61]. Serum samples were collected from 180 LUAD patients and 180 healthy controls. Profiling data indicated that a collection of miR-146a, miR-222, and miR-223 achieved an AUC of 0.951, demonstrating high accuracy (87.27%) in distinguishing LUAD patients from healthy subjects [61]. Notably, significant expression differences were observed in these miRNAs among I/II stage LUAD patients compared to healthy controls, suggesting their promise as diagnostic indicators for early-stage LUAD [61]. In line with this, Duan and colleagues demonstrated that the collections of miR-492, miR-590-3p, and miR-631 could effectively differentiate between control groups and patients at the early stage of NSCLC [62]. Additionally, Nadal et al. investigated 60 selected miRNAs in a study involving 70 patients with NSCLC and 22 healthy individuals. In their ROC analysis, miRNAs such as miR-193b, and miR-200b emerged as key biomarkers [63]. In a different study utilizing an independent dataset, researchers demonstrated that the expression levels of four microRNAs in sputum (miR-21, miR-486, miR-375, and miR-200b) yielded a sensitivity of 81% and a specificity of 92% in differentiating 64 patients with LUAD from 58 healthy individuals [64]. Recently, Liu and colleagues reported that circulating miRNA-196b-5p was potential biomarkers for NSCLC patients [65].

SCLC has a particularly poor prognosis, mainly due to the fact that most cases are first identified as extensive stage with multiple-organ metastasis. Thus, early detection of SCLC is vital, as it may enhance patient outcomes [66]. A recent study has investigated exosomal miRNAs present in the serum of SCLC patients. Researchers initially analyzed the results from miRNA array assays, from which 51 miRNAs were selected based on their p-values. This was followed by an assessment of the top ten differentially expressed genes. Of these, miR-3565, miR-3126, and miR-200b exhibited upregulation, whereas miR-92b showed downregulations. The AUC values for each miRNA group ranged from 0.64 to 0.76, but the combined use of three miRNAs significantly improved diagnostic performance, achieving the AUC value of 0.93, thus this 3-miRNA panel might function as potent diagnostic biomarkers in SCLC [67]. In another study, Jiang et al. utilized deep sequencing to examine the EV cargo and identified 22 EV-miRNAs that were differentially expressed when comparing lung cancer patients to control subjects. Subsequent validation confirmed that potential significance of miR-483-3p could function as a candidate marker for early-stage SCLC (AUC = 0.758) [68]. Moreover, researchers reported that exosomal miR-1228 was elevated among aggressive SCLC patients, which might function as a potent diagnostic biomarker for SCLC [69]. Although circulating miRNAs hold promise as supplementary tools for pathological assessment, their clinical application remains quite limited. Currently, the miRNA diagnostic tool in clinical use is for thyroid nodules that cannot be definitively diagnosed through cytology. This involves the combined use of the ThyGeNEXT gene mutation testing and the ThyraMIR v2 miRNA panel, achieving a sensitivity of 96% along with a specificity of 99% [70, 71].

While plasma or serum biomarkers offered essential insights, they only represented a fraction of the intricate blood matrix, which could restrict the sensitivity of potential biomarkers [40, 72]. Recent research has suggested that different blood cells can rapidly adjust to stress by altering the concentrations of signaling molecules, playing a critical role in tumorigenesis. Geng et al. developed diagnostic panels that integrate miRNAs from red blood cells (RBCs). Their study revealed that the use of NSCLC miRNA panel (miR-93-5p、miR-29c-3p and miR-449-5p) could achieve an AUC of 0.76, while miR-15a-5p、miR-93-5p、miR-29c-3p and miR-449-5p were consisted of a LUAD panel, achieving an AUC value of 0.84. Specifically, RBC-derived four miRNAs including miR-93-5p、miR-29c-3p、miR-15a-5p and miR-449b-5p functioned as a group of biomarker that achieved an AUC of 0.763 for LUSC patients [73].

In general, miRNAs have yet to demonstrate consistent and dependable diagnostic effectiveness across broader clinical settings, and their use remains largely exploratory. To fully harness the diagnostic potential of miRNA signatures—particularly for lung cancer subtyping and early detection—large-scale, multi-center cohort studies that reflect diverse populations are essential. These studies should incorporate rigorous clinical annotation, adhere to standardized pre-analytical and analytical protocols, and integrate multi-omics layers (genomics, cfDNA methylation/fragmentomics, transcriptomics, proteomics, metabolomics), as well as radiologic and digital pathology features, to enhance robustness and interpretability. Such a strategy aims to improve reproducibility, increase clinical relevance, and clarify the incremental value of miRNA markers beyond current standards of care [54].

Circulating miRNAs as potent prognostic biomarkers in lung cancer

Recent developments, particularly in next-generation sequencing (NGS), have exposed substantial heterogeneity within lung cancer [74]. Consequently, a range of pathological subtypes has been identified, emphasizing the complexity of lung cancer and reinforcing the importance of personalized treatment strategies for each patient [75]. The outcomes for lung cancer patients can vary widely, with considerable risks of recurrence. Consequently, there is a growing interest in the development of consistent prognostic biomarkers for all stages of lung cancer to support informed treatment decision-making. Evidence has revealed that miRNAs have been linked to specific clinical outcomes in lung cancer patients [76]. Notably, significantly reduced levels of miRNA-148a were linked to lymph node metastasis and more advanced stages of the disease [77]. Moreover, the expression levels of miRNAs-19a-3p, 126-5p, 556-3p, 671-5p, 937-3p, 4664-3p, and 4746-5p were associated with OS in patients with LUAD, revealing higher mortality rates among high-risk groups [78]. Wu et al. reported that miR-21, miR-141, and miR-490 could sever as prognostic markers for lung cancer [79]. Shao et al. conducted a meta-analysis indicating that, for miR-155 in lung cancer, the pooled sensitivity and specificity were 0.82 and 0.78, respectively. Furthermore, the area under the curve (AUC) was 0.87, underscoring the important role of miR-155 in the detection of lung cancer [80].

Additionally, one study has indicated that the serum levels of circulating miR-942 and miR-601 were significantly elevated in NSCLC. Furthermore, elevated levels of serum miR-942 and serum miR-601 were associated with negative clinical variables and poorer survival outcomes, whereas those patients with low levels of both miRNAs exhibited the most favorable outcomes [81]. Furthermore, NSCLC patients with lower miR-637 expression exhibited a decreased survival rate compared to those with higher expression levels of this miRNA. Additionally, miR-637 acted as a tumor suppressor by inhibiting tumor growth in NSCLC [82]. Liu et al. established a 7-miRNA signature, which could predict survival of LUAD patients and showed improved performance when integrated with the N stage, accurately forecasting survival outcomes in LUAD patients [83]. Emerging evidence underscores the pivotal role of microRNAs in the diagnosis, subtyping, and prognosis of lung cancer, highlighting their potential as noninvasive biomarkers for early detection and tailored therapy. However, consistent, reproducible performance across real-world clinical settings has not yet been established. To translate promise into practice, the field must shift from proof-of-concept studies to rigorously designed, multi-center investigations that clarify clinical value and implementation pathways [54].

Circulating miRNAs serve as potential predictive biomarkers for assessing therapeutic responses in lung cancer

It is crucial to recognize that the levels of miRNAs might have a significant relationship with the response to treatment [84]. Specifically, variations in miRNA expression could serve as potential indicators of how well a patient responds to specific therapies. This correlation suggests that monitoring circulating miRNA levels could provide valuable insights into treatment efficacy and help tailor more personalized approaches for patients based on their individual molecular profiles [84]. Kumar et al. reported that patients with NSCLC who responded to chemotherapy showed increased expression of miR-1249-3p when compared to those who did not respond [85]. One recent study revealed that circulating miRNAs might act as biomarkers for forecasting the efficacy of maintenance therapy with pemetrexed in patients with stage IIIb or IV LUAD who had previously undergone first-line treatment with pemetrexed combined with platinum [86]. Serum levels of miR-25, miR-145, and miR-210 were linked to treatment responses in patients with advanced non-small cell lung cancer (NSCLC) receiving pemetrexed [86].

The cornerstone of treatment for inoperable stage III NSCLC is concurrent cCRT [87]. However, resistance mechanisms, such as the upregulation of autophagy, could render this therapy ineffective [88]. In non-responders to cCRT, the expression levels of miR-375, miR-200c, and miR-30c in EVs were found to be reduced. These miRNAs played a role in influencing the phosphatidylinositol-3-kinase (PI3K) signaling pathway, which was involved in the regulation of autophagy and is also linked to immune regulation. These results indicate that miRNAs have the potential to serve as predictive markers for treatment response to cCRT [89]. Further study could reveal the predictive role of these miRNAs in the treatment response to PI3K inhibitors. Peng et al. investigated the correlation between exosomal miRNAs and the efficacy of immunotherapy in advanced NSCLC patients with wild-type EGFR/ALK. Their study identified that circulating miR-320d, miR-320c, miR-320b were significantly upregulated in the progressive disease (PD) group at baseline before treatment compared to the partial response (PR) group. Additionally, they observed that the T-cell suppressor hsa-miR-125b-5p showed an increasing trend in expression within the PD group at baseline and was significantly downregulated in the post-treatment plasma exosomes compared to the pre-treatment samples from PR patients [90]. One recent study evaluated the predictive value of individual differentially expressed miRNAs in NSCLC patients. Their study revealed that miR-105-5p and miR-767-5p exhibited the greatest significance in predicting progression-free survival (PFS) time following anti-PD1 immune-checkpoint inhibitor treatment [91]. Researchers utilized next-generation sequencing to profile exosome-derived miRNAs from the plasma of eight NSCLC patients at the point they became sensitive to osimertinib and after they developed resistance to therapy. They found that elevated levels of circulating miR-323-3p, miR-1468-3p, miR-5189-5p, and miR-6513-5p were linked to resistance to Osimertinib in patients with NSCLC [92]. While these findings are indeed promising, current research emphasizes the necessity for larger patient populations and potentially multi-center trials to thoroughly evaluate and confirm the effectiveness and reliability of these biomarkers in lung cancer. Expanding the cohort size would enhance the statistical power of the studies, allowing for more robust conclusions regarding the applicability of these biomarkers across diverse patient demographics and treatment settings. Additionally, multi-center trials could help mitigate any biases associated with a single institution and provide a more comprehensive understanding of the biomarkers’ roles in different clinical environments. Such efforts would ultimately contribute to refining patient stratification and personalizing treatment approaches, thereby improving overall outcomes in lung cancer management [93, 94].

The “core set” of robustly validated miRNAs in lung cancer

Small-scale studies often lack the statistical power to generate robust conclusions. Larger studies with well-defined patient cohorts are crucial because they minimize the risks of false positives that smaller studies might yield [95]. Thus, miRNAs identified in these larger cohorts that exhibit consistent expression levels are more likely to be biologically significant. The most reliable miRNA biomarkers are those that have been replicated in independent studies. If a specific miRNA is identified as a biomarker in various large-scale studies with different populations and methodological approaches, it provides substantial evidence for its biological relevance and utility in clinical settings [96].

A miRNA’s clinical significance can also be evaluated by correlating its expression levels with clinical outcomes such as response to treatment or overall survival. For instance, certain miRNAs, like miR-21 and miR-155, have been linked to poor clinical outcomes, reinforcing their candidacy as significant biomarkers [97–100]. miR-21 frequently associated with poor prognosis across various studies and implicated in critical oncogenic pathways [97, 98]. Moreover, miR-155 was documented as a significant marker in cancer diagnostics and prognostics [99, 100]. Additionally, miR-126 and miR-486 also were demonstrated potential in differentiating between various lung cancer subtypes and have clinical relevance [16, 97]. These miRNAs have often correlated with notable clinical outcomes, suggesting they could form a core set of biomarkers used in lung cancer diagnosis and prognosis.

Challenges in the application of circulating miRNAs in clinical management of lung cancer

MiRNAs are essential regulators of gene expression and significantly impact various cellular processes, such as cell growth, differentiation, and programmed cell death [101]. Recent advances in miRNA research have highlighted their promising role in precision medicine, especially concerning lung cancer treatment [102, 103]. These small RNAs have the potential to affect the metabolism and effectiveness of cancer therapies within the body, thereby influencing therapeutic outcomes and the risk of adverse effects [104].

Circulating miRNAs are emerging as promising candidates for biomarkers in lung cancer diagnosis. Nonetheless, many studies concentrate primarily on analyzing the altered levels of miRNAs in the plasma or serum of cancer patients [54]. The lack of standardized methods for preparing plasma and serum samples, isolating RNA, and selecting appropriate internal controls significantly hinders the ability to compare findings across various laboratories [105]. This inconsistency in methodologies can lead to substantial variations in miRNA quantification, which may be attributed to differences in sample handling, extraction techniques, and the choice of normalization strategies [106]. For instance, the way plasma or serum samples are processed can affect the integrity and yield of the extracted RNA. Variations in factors such as the speed and duration of centrifugation, the temperature at which samples are stored, and the timing of sample processing can introduce biases that ultimately impact the analysis of miRNA levels [107]. Furthermore, the method employed for RNA isolation whether it’s a column based extraction, magnetic bead-based approach, or a phenol-chloroform extraction can also lead to discrepancies in RNA quality and quantity [108]. To address this issue, it is crucial to develop a comprehensive set of reference protocols for quantifying circulating miRNAs. Additionally, another factor affecting the technical reliability of “omics” is the variability in laboratory methods and the range of platforms employed [109, 110]. The inherent heterogeneity within and among samples results in measurement discrepancies, stemming from the nature of cancer itself, which is marked by both intra-tumor and inter-tumor variability that is challenging to manage [110]. Therefore, it is essential to develop real-time profiling, follow-up biomarker panels, and personalized therapeutics tailored to individual tumors, taking into account their unique characteristics (Fig. 2).

Fig. 2.

Challenges in the application of miRNAs in lung cancer

Additionally, from a translational viewpoint, there is an urgent necessity to explore the relationships between circulating miRNAs and existing biomarkers for cancer diagnosis and prognosis [111]. Conducting discovery and validation studies in larger, well-characterized patient cohorts that specifically examine miRNA panels as diagnostic and prognostic tools will greatly enhance the field [112]. In the context of lung cancer, miRNAs serve as valuable diagnostic and prognostic instruments, and ongoing clinical trials are vital for evaluating the specificity of these miRNA-based biomarkers [112, 113]. Challenges persist in the application of circulating miRNAs for targeted therapies. Artificially packaged and modified miRNAs within exosomes may enhance their stability in vivo; however, significant issues related to tissue specificity and permeability continue to be problematic [114]. Currently, there is ongoing development of ligands, antibodies, and nanoparticles tailored to deliver miRNAs, with improvements aimed at achieving better specificity and reduced immunotoxicity [115]. Nevertheless, extensive preclinical studies in animal models remain essential for evaluating their effectiveness. Gaining a deeper understanding of these mechanisms will not only validate circulating miRNAs as dependable biomarkers for cancer diagnosis but also contribute to the development of innovative cancer therapies.

Moreover, there is a lack of systematic and synthesized analysis of miRNA biomarkers in the current literature. And the variability in study designs and analysis techniques has resulted in a lack of comparability among results. This complexity poses challenges in conducting a thorough comprehensive analysis of the role of miRNAs in lung cancer diagnosis and prognosis. Given these limitations, we recommend that future research should consider conducting a meta-analysis of miRNA biomarkers. Such an analysis would be invaluable in synthesizing findings from diverse studies, revealing the true clinical significance of miRNAs in lung cancer diagnostics and prognostics. These efforts would not only provide more reliable diagnostic tools but may also inform individualized treatment strategies, ultimately leading to improved patient outcomes.

Conclusions and perspectives

In conclusion, circulating miRNAs represent a promising avenue for enhancing the diagnosis, prognosis, and treatment of lung cancer. Their stability in body fluids and the ability to reflect the pathological state of tumors make them suitable candidates for non-invasive biomarkers. This review highlights the critical role of specific miRNA signatures in distinguishing between various subtypes of lung cancer and predicting treatment responses, which are crucial for tailoring personalized therapeutic strategies. However, despite the potential of circulating miRNAs, several challenges remain, including the need for standardized methods of sample collection and analysis, as well as the inherent heterogeneity of miRNA expression profiles in different cancer types and stages. Moreover, when detecting circulating miRNAs in clinical settings, the choice of platform quantitative reverse transcription polymerase chain reaction qRT-PCR, NGS, or microarray technology significantly impacts results [116]. qRT-PCR is highly sensitive, cost-effective, and provides established protocols, making it ideal for targeted analyses, though it has limited multiplexing capability and depends on reference genes [117]. NGS enables comprehensive profiling of thousands of miRNAs and is suitable for large-scale studies, but it is more costly and generates complex data requiring advanced analysis [118]. Microarrays allow high-throughput evaluation and have established applications in biomarker discovery, but they suffer from lower sensitivity and issues with non-specific binding [119]. Ultimately, the platform selection should align with the specific research question, available resources, and clinical context to ensure reliable and relevant outcomes in miRNA research.

Additionally, circulating long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) also function as biomarkers for cancer [120–122]. Advances in sequencing have characterized lncRNAs, such as H19 and LINC-PINT, which were differentially expressed in lung cancer and might serve as diagnostic and therapeutic targets [123, 124]. Specific circRNAs like circFASRA, and circSATB2 were linked to immunosuppression and tumor progression, suggesting their potential as diagnostic markers and therapeutic targets in lung cancer [125, 126]. Overall, these techniques offer diverse options for noncoding RNA detection and present substantial possibilities for clinical lung cancer screening.

Moreover, ongoing blood testing may enhance the management of suspicious findings from low-dose computed tomography (LDCT), potentially facilitating timely interventions or preventive neo-adjuvant or adjuvant therapies [127]. Ultimately, incorporating these strategies into screening programs could lead to earlier cancer diagnoses, increase adherence to screening protocols, and reduce both costs and associated risks [127]. To fully realize the clinical utility of miRNAs, further research is needed. This should focus on large-scale, multi-center studies to validate miRNA biomarkers in diverse populations and contexts. By continuing to explore the complex regulatory networks involving miRNAs, as well as their interactions with existing biomarkers, we can develop more effective diagnostic tools and treatment approaches. Ultimately, the integration of miRNA profiling in clinical practice has the potential to significantly improve patient outcomes and advance our understanding of lung cancer biology.

Author contributions

Jun Fan, DongMing Shen and Lei He contributed to Conceptualization, Data curation, Investigation, Methodology, Project administration, Writing—original draft, and Writing—review & editing. ChunXia Yan, Hu Li, Hu Li contributed to Data curation, Resources, and Supervision. XiaoSong Bai contributed to Conceptualization, Supervision, and Writing—review & editing. All authors read and approved the final manuscript.

Funding

Not applicable.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval

The present article did not require institutional review board approval because we do not report human participant data.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.