Next Article in Journal
Updates on the Treatment of Richter’s Syndrome, Including Novel Combination Approaches
Next Article in Special Issue
Exploratory Algorithms to Aid in Risk of Malignancy Prediction for Indeterminate Pulmonary Nodules
Previous Article in Journal
Targeting Mediator Kinase Cyclin-Dependent Kinases 8/19 Potentiates Chemotherapeutic Responses, Reverses Tumor Growth, and Prolongs Survival from Ovarian Clear Cell Carcinoma
Previous Article in Special Issue
Neutrophil-to-Lymphocyte and Platelet-to-Lymphocyte Ratio: Side by Side with Molecular Mutations in Patients with Non-Small Cell Lung Cancer—The INOLUNG Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Cancers
Volume 17
Issue 6
10.3390/cancers17060942
cancers-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Development of a miRNA-Based Model for Lung Cancer Detection

by
Kai Chin Poh
1,*,
Toh Ming Ren
1,
Goh Liuh Ling
2,
John S Y Goh
3,
Sarrah Rose
4,
Alexa Wong
4,
Sanhita S. Mehta
4,
Amelia Goh
3,
Pei-Yu Chong
3,
Sim Wey Cheng
2,
Samuel Sherng Young Wang
5,
Seyed Ehsan Saffari
5,
Darren Wan-Teck Lim
6 and
Na-Yu Chia
4,*
1
Division of Respiratory Medicine, Sengkang General Hospital, Singapore 544886, Singapore
2
Molecular Diagnostic Laboratory, Tan Tock Seng Hospital, Singapore 308433, Singapore
3
Professional Officers Division, Singapore Institute of Technology, Singapore 828608, Singapore
4
Averywell Limited, Greater Manchester OL8 4QQ, UK
5
Duke-NUS Medical School, Singapore 169857, Singapore
6
National Cancer Center Singapore, Singapore 168583, Singapore
*
Authors to whom correspondence should be addressed.
Cancers 2025, 17(6), 942; https://doi.org/10.3390/cancers17060942
Submission received: 6 February 2025 / Revised: 2 March 2025 / Accepted: 6 March 2025 / Published: 10 March 2025
(This article belongs to the Special Issue Predictive Biomarkers for Lung Cancer)
Browse Figures
Review Reports Versions Notes

Simple Summary

Lung cancer remains the leading cause of cancer-related mortality worldwide, primarily due to late-stage diagnosis and limitations of current screening methods with low-dose computed tomography scans. Analysis of serum microRNA biomarkers may improve early detection of lung cancer. We aimed to discover a panel of miRNA biomarkers that is highly accurate in detecting lung cancer through a case-control study. A final panel of six miRNA biomarkers were selected with machine learning algorithms, achieving a high AUC of 0.86. The use of miRNA biomarkers shows promise in augmenting current lung cancer screening protocols and represent a significant step towards reducing lung cancer mortality.

Abstract

Background: Lung cancer is the leading cause of cancer-related mortality globally, with late-stage diagnoses contributing to poor survival rates. While lung cancer screening with low-dose computed tomography (LDCT) has proven effective in reducing mortality among heavy smokers, its limitations, including high false-positive rates and resource intensiveness, restrict widespread use. Liquid biopsy, particularly using microRNA (miRNA) biomarkers, offers a promising adjunct to current screening strategies. This study aimed to evaluate the predictive power of a panel of serum miRNA biomarkers for lung cancer detection. Patients and Methods: A case-control study was conducted at two tertiary hospitals, enrolling 82 lung cancer cases and 123 controls. We performed an extensive literature review to shortlist 25 candidate miRNAs, of which 16 showed a significant two-fold increase in expression compared to the controls. Machine learning techniques, including Random Forest, K-Nearest Neighbors, Neural Networks, and Support Vector Machines, were employed to identify the top six miRNAs. We then evaluated predictive models, incorporating these biomarkers with lung nodule characteristics on LDCT. Results: A prediction model utilising six miRNA biomarkers (mir-196a, mir-1268, mir-130b, mir-1290, mir-106b and mir-1246) alone achieved area under the curve (AUC) values ranging from 0.78 to 0.86, with sensitivities of 70–78% and specificities of 73–85%. Incorporating lung nodule size significantly improved model performance, yielding AUC values between 0.96 and 0.99, with sensitivities of 92–98% and specificities of 93–98%. Conclusions: A prediction model combining serum miRNA biomarkers and nodule size showed high predictive power for lung cancer. Integration of the prediction model into current lung cancer screening protocols may improve patient outcomes.

1. Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide, with approximately 1.8 million deaths annually [1]. Despite advancements in cancer therapeutics, including novel immunotherapies, lung cancer survival rates remain low [2,3]. This is primarily attributed to the asymptomatic nature of early-stage lung cancer, often resulting in late diagnosis [4]. Data from the Surveillance, Epidemiology, and End Results Program (SEER) indicates over half of lung cancer patients (53%) are diagnosed at a metastatic stage, where the 5-year survival rate is only 8.9% [5].
Early detection is critical for improving survival in lung cancer patients [6,7]. Low-dose computed tomography (LDCT) has proven to be a valuable tool for lung cancer screening, particularly among heavy smokers. The landmark National Lung Screening Trial (NLST) demonstrated a 20% relative risk reduction in lung cancer mortality with annual LDCT screening for heavy smokers aged 55–74 years [8]. This finding was also corroborated by another adequately powered lung cancer screening study, the NELSON trial [9].
Despite positive outcomes from these trials, widespread implementation of lung cancer screening with LDCT remains limited. LDCT screening is highly resource-intensive, requiring numerous medical specialists and readily accessible advanced imaging equipment [10,11,12,13]. Due to a high number needed to treat (NNT), the cost effectiveness of LDCT screening remains unclear. Additionally, LDCT has a high false-positive rate (7.9–49.3%) [14]. There are concerns regarding overdiagnoses, unnecessary investigations, risk from invasive biopsies and psychological distress amongst screened populations [15,16]. The limitations of LDCT highlight the need for adjunct tools to enhance lung cancer screening efficiency and accessibility.
Liquid biopsy, a minimally invasive diagnostic technique that analyses biomarkers in non-solid biological samples, has emerged as a promising approach for lung cancer detection. Various biomarkers, including circulating tumour cells (CTCs), circulating tumour DNA (ctDNA), cell free DNA (cfDNA), messenger RNA (mRNA) and microRNA (miRNA) have been evaluated for their association with lung cancer [17]. Among these, ctDNA and cfDNA assays are the most extensively investigated and are increasingly utilized for lung cancer treatment planning and post-treatment monitoring [18,19,20,21,22]. However, their sensitivity in early-stage cancer detection is limited due to the low abundance of circulating tumour DNA [21,22,23,24,25].
miRNAs demonstrate significant potential as biomarkers for the early detection of lung cancer. These small, non-coding RNA molecules (20–25 nucleotides) regulate gene expression at the post-transcriptional level by inducing translational repression or degradation of mRNA, thereby controlling key metabolic pathways and cellular processes [26,27,28]. The metabolomic alterations in lung cancer include disorders in glycolysis, the tricarboxylic acid cycle, fatty acid oxidation, glutaminolysis and amino acid synthesis [29,30,31]. The complex interplay between genetic alterations and metabolites derangement promotes tumorigenic processes such as cell proliferation, angiogenesis, migration, and resistance to apoptosis [32,33].
Unlike circulating DNA, miRNA dysregulation is frequently detectable in early-stage cancer, making them particularly valuable for early diagnosis [34,35]. Their stability in serum and ease of quantification using qPCR techniques further enhance their practicality for lung cancer screening [36]. miRNA sequencing analysis is usually less expensive as compared to ct-DNA and potentially a more cost-effective option for screening.
This study aims to develop and evaluate the predictive power of a serum miRNA-based model for lung cancer. By serving as an adjunct to current screening methods, the miRNA-based prediction model could significantly enhance early detection and improve lung cancer screening outcomes.

2. Materials and Methods

2.1. Study Design

We performed a case-control study in two tertiary teaching hospitals between 2020 and 2023. All participants were aged between 21 and 87 years of age and mentally competent to give informed consent for participation in the study. Patients with a known history of malignancy or active malignancy undergoing treatment were excluded from the study. Cases included patients with newly diagnosed lung cancer based on histological studies while controls included patients with no lung nodules, adenopathy, mass on radiographic imaging or benign lung nodules. A diagnosis of benign lung nodule was made following lung biopsy or if a lung nodule was less than 6 mm and had a less than 1% risk of malignancy.
Baseline patient characteristics and clinical features, including age, gender, ethnicity, smoking status, tumour site, stage, histology and molecular subtypes, lung nodule characteristics and radiological evidence of emphysema, were obtained from electronic health records.

2.2. miRNA Selection

Candidate miRNAs potentially involved in lung cancer pathogenesis were identified through a comprehensive literature review of major scientific databases, including PubMed, Embase, MEDLINE, Web of Science, Cochrane Library and ScienceDirect. The search employed key terms such as “miRNA”, “lung”, “pulmonary”, “cancer”, “malignancy”, “neoplasm” and “tumor”. To ensure a robust selection, we shortlisted miRNAs that had been reported in at least two independent studies and demonstrated functional relevance to tumorigenic processes like angiogenesis, apoptosis, invasion, and metastasis. The miRNAs expressed in a discovery cohort were subsequently tested in the final study cohort.

2.3. Sample Processing and miRNA Analysis

All subjects underwent venepuncture and had 5 mL of peripheral blood samples drawn in plain serum tubes (BD, Franklin Lakes, NJ, USA). Blood samples were left to clot for 45–60 min and centrifuged at 2000× g for 15 min at room temperature. Serum was aliquoted and stored in cryotubes at −80 °C for long term storage. If a blood sample was haemolyzed, the specimen was not analysed due to potential contamination by miRNAs released from red blood cells [37].
The expression levels of each miRNAs were quantified and reported as Ct values. Total RNA, including miRNA, was isolated from 200 µL of patient blood serum (Averywell, UK). Reverse transcription (RT) and qPCR were performed using Averywell’s mirLung DxTM Kit (Averywell, UK). TaqMan primers and probes were then used to quantitate each miRNA target via real time qPCR. The reactions were incubated in a 96-well plate at 95 °C for 30 s, followed by 40 cycles of 95 °C for 3 s and 60 °C for 30 s. The qPCR reaction was carried out in Applied Biosystems QuantStudio 3 and 5 (ThermoFisher Scientific, Waltham, MA, USA).

2.4. Statistical Analysis

Clinical characteristics between cases and controls were compared using two-sample t-tests for continuous variables and Chi-square tests for categorical variables, with statistical significance set at p < 0.05.
Clinical variables and miRNA expression levels were evaluated as predictive features and ranked separately using the Random Forest (RF) algorithm. The RF algorithm was chosen for its ability to handle complex interactions and nonlinear relationships, making it suitable for assessing the variable importance of miRNAs. RF’s ensemble structure enables a robust ranking of biomarkers based on their discriminative power while accounting for heterogeneous effects across variables. The clinical variables assessed included patient demographics (age, gender, race, smoking status and the presence of emphysema/COPD) and characteristics of lung nodules (number, size of the largest nodule, location and spiculation). The mean importance score from RF was used as a threshold, with miRNA biomarkers above this threshold selected for inclusion in the final risk model.
To evaluate the predictive performance of the top miRNA biomarkers and clinical variables, four machine learning models were employed: Neural Networks (NNET), Support Vector Machine (SVM), K-Nearest Neighbours (KNN), and Naive Bayes (NB). We selected KNN, NNET, SVM and NB because they represent a diverse set of machine learning approaches, balancing interpretability, flexibility and predictive performance. Each method has its own strengths and weaknesses: KNN is simple but sensitive to high-dimensional data, NNET captures complex patterns but requires careful tuning, SVM is robust to small datasets but computationally intensive, and NB is efficient but relies on strong independence assumptions.
The full dataset was used for training, and model performance was validated using five-fold cross-validation to mitigate overfitting. Hyperparameter tuning was also performed for each model through five-fold cross-validation to optimize performance.
Two feature sets were tested: (1) Model 1, consisting solely of the top biomarkers and (2) Model 2, which combined the top miRNAs with the size of the largest nodule. Model performance was primarily evaluated using the Area Under the Receiver Operating Characteristic Curve (AUC) as the accuracy metric. Additional performance metrics, such as sensitivity and specificity, were used to assess the model’s practical applicability.
Data analysis was performed in R software, R version 4.3.2 (R Core Team (2023). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ accessed on 31 October 2023.

3. Results

3.1. Patient Demographics and Clinical Characteristics

The study included 205 participants, with 82 lung cancer cases and 123 controls. The mean age of the participants was 60 years, with the majority being male (61%) and of Chinese ethnicity (75%). Lung cancer cases were older than the controls (66.5 vs. 57.0 years, p < 0.005) and were more likely to be heavy smokers (50 vs. 24.6%, p < 0.005). Emphysema was more commonly observed in lung cancer patients (15.9 vs. 7.3%, p = 0.016). This indicates age and smoking history as significant risk factors for lung cancer. Notably, malignant nodules were larger than benign nodules (40 vs. 15 mm, p < 0.005). Malignant nodules were predominantly located in the upper lobes of the lungs (48.8 vs. 19.5%, p < 0.005) and more commonly spiculated (31.7 vs. 2.4%, p < 0.005) (Table 1).
Adenocarcinoma was the predominant histological subtype (71%) of lung cancer. A significant proportion (39%) of lung cancer cases were never-smokers. In terms of cancer staging, most lung cancer cases were diagnosed at advanced stages (Stage 3: 9.8%, Stage 4: 48.8%). Small cell lung cancer (SCLC) was present in 8.5% of cases (n = 7) and were mostly diagnosed (71.4%) in the extensive stage (Table 1).

3.2. Identifying Top Clinical Features and miRNA Biomarkers for Lung Cancer Detection

In total, 25 candidate miRNAs were identified as potential lung cancer biomarkers after a comprehensive literature review. From this list, 16 miRNAs demonstrated a minimum two-fold increase in expression levels in a discovery cohort (n = 8) and were included in the full study cohort (Supplementary Materials Table S1).
The RF analysis produced a ranked list of clinical features according to their predictive importance for lung cancer risk. Notably, the size of the largest lung nodule emerged as the most important clinical predictor, highlighting its strong association with malignancy. Additional significant clinical features included age, nodule spiculation, smoking status, race, presence of emphysema/COPD and gender, which were sequentially ranked in decreasing order of importance (Figure 1).
In parallel, RF analysis was also applied to the 16 miRNAs to evaluate their discriminative value. In total, 6 miRNAs—miR-196a-5p, miR-1268, miR-130b-5p, miR-1290, miR-106b-5p, and miR-1246 (Figure 2)—had RF importance scores above the mean index and were selected for further analysis. These six miRNAs displayed significantly higher expression in lung cancer cases compared to the controls, underscoring their potential role as key biomarkers in disease detection. The combined results of clinical features and miRNA rankings emphasize the importance of both nodule characteristics and specific miRNAs in predicting lung cancer risk.

3.3. Clinical Performance Index

Model 1 (miRNA-only): By using only the top six miRNA biomarkers, NNET achieved the highest AUC of 0.863, demonstrating strong discriminative ability for lung cancer based solely on miRNA expression (Figure 3). Machine learning using SVM achieved an AUC of 0.802, with KNN and NB showing slightly lower AUCs of 0.781 and 0.790, respectively (Supplementary Materials Figure S2). Sensitivity in Model 1 ranged from 70% to 78%, with NNET showing the highest sensitivity (0.775) and KNN the lowest (0.700). Specificity varied from 73% to 85%, with NNET again achieving the highest specificity (0.850) (Table 2).
Model 2 (miRNA + nodule size): Incorporating the largest nodule size as an additional feature significantly improved diagnostic performance. KNN achieved the highest AUC (0.989) in this configuration (Figure 4), followed closely by NB (AUC = 0.978). SVM and NNET also performed well, with AUCs of 0.983 and 0.961, respectively (Supplementary Materials Figure S3). Sensitivity in Model 2 ranged from 92% to 98%, with SVM and KNN attaining the highest sensitivity scores (0.975 and 0.921, respectively). Specificity increased to 93–98%, with KNN achieving the highest specificity (0.975), followed closely by NB (0.976). The substantial improvement in predictive performance in Model 2 underscores the added value of including clinical information, particularly nodule size, in a lung cancer risk prediction model (Table 2).

4. Discussion

4.1. Synergistic Role of a miRNA Panel in Lung Cancer Detection

The six miRNAs in the final panel enhanced diagnostic accuracy as they target interrelated oncogenic pathways implicated in the pathogenesis of lung cancer. miR-196a-5p and miR-106b-5p directly promote uncontrolled cell proliferation through the PI3K/AKT pathway [38,39,40]. Similarly, miR-130b-5p, miR-1246, and miR-1290 regulate epithelial-mesenchymal transition (EMT), a key process in metastasis, by altering expression of proteins such as DPP-4 [41] and metallothionein [42]. Additionally, miR-106b-5p and miR-1290 suppress pro-apoptotic genes such as BTG3 [43], allowing lung cancer cells to evade programmed cell death.
The six miRNAs also work synergistically through differential expression patterns across disease stages. miR-130b-5p, miR-1290 and miR-1246 are consistently overexpressed in advanced NSCLC and correlate with disease progression and cancer stemness [44,45]. miR-106b-5p and miR-130b-5p are elevated in early-stage lung cancer, making them suitable for early detection [44,46]. The integration of these miRNAs into a single diagnostic panel provides better specificity and sensitivity than any individual miRNA biomarker.

4.2. Limitations of Existing Lung Cancer Screening Strategy: Identifying Risk Amongst Non-Smokers

While LDCT screening reduces lung cancer mortality among heavy smokers, its utility for non-smokers and individuals with non-smoking-related risk factors remains unclear. This issue is especially relevant in Asia, where a substantial proportion (45–70%) of lung cancer cases occur in non-smokers [47,48,49]. A simulation study suggest that lung cancer screening limited to heavy smokers in Asian populations reduces mortality by only 3.76–4.74%, emphasizing the need to expand screening criteria to better capture high-risk populations [16].
Clinical prediction models offer a promising solution to improve lung cancer screening criteria. Tools such as PLCOm2012, the Liverpool Lung Project (LLP), the Lung Cancer Death Risk Assessment Tool (LCDRAT) and the Henan Lung Cancer Risk Model enable personalized risk assessments based on demographic and clinical factors. These models enhance lung cancer risk stratification, reducing the number needed to treat (NNT), and increasing the proportion of screen-preventable deaths [50,51,52,53,54]. However, these existing clinical prediction models are still heavily reliant on smoking history for risk assessment and may not accurately predict lung cancer risk in non-smokers.
Expanding LDCT screening criteria to include non-smokers with risk factors such as family history, passive smoke exposure, or chronic respiratory conditions has been shown to increase rates of lung cancer detection. The Taiwan Lung Cancer Screening in Never-Smoker Trial (TALENT) study demonstrated the effectiveness of LDCT in detecting lung cancer amongst non-smokers with these risk factors. The lung cancer incidence in the study was twice of that reported in the NLST and NELSON trials, underscoring the importance of incorporating individuals with non-smoking-related risk factors into lung cancer screening programmes [6].

4.3. Challenges in Management of Screen-Detected Lung Nodules

In addition, lung cancer screening with LDCT alone is associated with a high rate of false-positive results. Most screen-detected lung nodules are benign, yet they often necessitate further evaluation, including invasive biopsies, which contribute significantly to healthcare costs. A cost analysis study estimated that 43.1% of the total cost of lung cancer workup is attributed to invasive biopsies of benign lung nodules [55].
Initial risk assessment of lung nodules often involves the use of prediction models. Subsequent investigations, including functional imaging with positron emission tomography (PET) scans, are guided by the risk of malignancy (Supplementary Materials Figure S1). The Brock and Mayo Clinic prediction models are widely recommended in major guidelines, including American College of Chest Physicians [56], the British Thoracic Society [57], and Fleischner Society [58]. These models utilize demographic, clinical, and radiologic features, including patient age, gender, family history of lung cancer, nodule size and location, and the presence of emphysema to estimate risk of lung malignancy. While effective in estimating malignancy risk, these models have primarily been validated in heavy-smoking, predominantly Caucasian cohorts.
The predictive performance of these models in Asian or non-smoking populations remains uncertain [56,59]. Their reliability in assessing indeterminate nodules is moderate, with AUC values ranging from 0.67 to 0.70 [60]. PET scans, which are often used to refine malignancy risk in indeterminate nodules, also has many limitations. They are expensive, often not readily accessible, and have reduced diagnostic specificity in regions with high tuberculosis prevalence, including many parts of Asia [61].

4.4. Improving Lung Cancer Screening with miRNA-Based Prediction Models

The panel of serum miRNA biomarkers, with its ease of application and high predictive ability for lung cancer, could be utilised to identify high-risk individuals who may not meet conventional screening criteria (heavy smoking history). The miRNA test could identify individuals with non-smoking-related risk factors and improve their enrolment into lung cancer screening programs (Supplementary Materials Figure S1). Compared to traditional clinical models such as the PLCOm2012, the panel of six serum miRNA biomarkers exhibits superior discriminative power, which could translate into improved screening outcomes.
Integrating the serum miRNA biomarker panel with nodule size from LDCT also significantly enhanced predictive power, achieving an AUC of 0.989 (Figure 4). This approach is consistent with findings from the BioMILD trial, which demonstrated that combining LDCT with an miRNA signature classifier (MSC) outperformed conventional risk models, including LCDRAT, PLCOm2012, and the Brock model [62]. The combined model holds promise for optimizing lung nodule management, reducing the need for unnecessary invasive procedures, and improving the cost-effectiveness of lung cancer screening programs (Supplementary Materials Figure S1).

4.5. Study Strength and Limitations

This study addresses limitations of many prior research, including failure to include all lung cancer subtypes and using techniques such as RNA-Seq or microarrays that are less suitable for adoption in large-scale lung cancer screening [63,64,65]. We utilized a highly accurate qPCR method to quantify serum miRNA levels, included all lung cancer subtypes, and used machine learning to identify the optimal combination of miRNA biomarkers. These methodological choices enhance the reliability and applicability of our findings. The lung cancer cases in our study cohort closely reflected the regional epidemiology, characterized by a high proportion of non-smokers (40.2%) and a predominance of adenocarcinomas.
A study limitation was that majority of subjects were of South-East Asian ethnicity. This limits the generalizability of our findings, especially as lung cancer subtypes and genetic mutations vary across populations. Furthermore, we did not control for smoking status, a potential confounder in miRNA-based lung cancer detection [66]. The study included both early and late-stage cases due to limited sample size. While this demonstrates the test’s applicability across all stages of lung cancer, the sensitivity for detection of early-stage cancer needs to be affirmed in future studies with larger cohorts stratified by cancer stage.

4.6. Challenges and Future Directions for miRNA-Based Screening

While miRNA biomarkers show promise as a novel lung cancer screening tool, larger studies are required to validate technical reliability and diagnostic accuracy. With an improved understanding of lung cancer metabolomics, the interplay between miRNA and lung cancer metabolism represents a novel field for further research.
There is a vital need to close the knowledge gap between miRNA as potential lung cancer biomarkers and practical implementation in cancer-screening programs. Clear guidelines are needed for optimal management of false-positive miRNA results (e.g., cases with benign lung nodules), which may arise from pre-cancerous conditions, extrapulmonary cancers, or infections.
Another critical consideration is the cost-effectiveness of miRNA-based lung cancer screening. Although miRNA assays may improve diagnostic accuracy and reduce rates of false positives, they come with high costs. Comprehensive health economic analyses will be necessary to determine the feasibility of incorporating miRNA assays into lung cancer screening at a national or global scale.

5. Conclusions

Our study highlights the potential of using miRNA biomarkers to improve lung cancer screening, particularly when combined with LDCT. This approach may help address the limitations of current screening strategies and LDCT, offering a more efficient and effective means of detecting lung cancer. Further validation in larger, ethnically diverse cohorts is necessary before miRNA biomarkers can be integrated into routine clinical practice. Continued research will be critical to refining this approach and maximizing its impact on lung cancer detection.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cancers17060942/s1, Table S1: miRNA biomarker selection. Figure S1: Use of miRNA-based prediction model to improve lung cancer screening. Figure S2: ROC curves for Model. Figure S3: ROC curves for Model 2. References [28,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103] are for references in Supplementary Table S1.

Author Contributions

Conceptualization, K.C.P.; data curation, K.C.P., T.M.R., S.R., A.W. and S.S.M.; formal analysis, K.C.P., T.M.R., S.R., S.E.S. and N.-Y.C.; funding acquisition, N.-Y.C.; investigation, A.G., S.S.Y.W., S.E.S. and N.-Y.C.; methodology, K.C.P., T.M.R. and N.-Y.C.; project administration, N.-Y.C.; resources, G.L.L., J.S.Y.G., A.G., P.-Y.C., S.W.C. and S.S.Y.W.; supervision, D.W.-T.L.; validation, T.M.R.; writing—original draft, K.C.P.; writing—review and editing, K.C.P., T.M.R., A.W., D.W.-T.L. and N.-Y.C. All authors have read and agreed to the published version of the manuscript.

Funding

The study was funded by Averywell Limited.

Institutional Review Board Statement

The study was approved by Singhealth IRB Board (Ref 2021/2526) on 17 September 2021 and National Health Group IRB Board (Ref 2018/01200) on 5 February 2022.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Dataset available on request from the authors. The raw data supporting the conclusions of this article will be made available by the authors on request.

Conflicts of Interest

Author Sarrah Rose, Alexa Wong, Sanhita S. Mehta and Na-Yu Chia were employed by the company Averywell Limited. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Siegel, R.L.; Miller, K.D.; Fuchs, H.E.; Jemal, A. Cancer Statistics, 2021. CA Cancer J. Clin. 2021, 71, 7–33. [Google Scholar] [CrossRef] [PubMed]
  2. Howlader, N.; Forjaz, G.; Mooradian, M.J.; Meza, R.; Kong, C.Y.; Cronin, K.A.; Mariotto, A.B.; Lowy, D.R.; Feuer, E.J. The Effect of Advances in Lung-Cancer Treatment on Population Mortality. N. Engl. J. Med. 2020, 383, 640–649. [Google Scholar] [CrossRef] [PubMed]
  3. Herbst, R.S.; Garon, E.B.; Kim, D.W.; Cho, B.C.; Perez-Gracia, J.L.; Han, J.Y.; Arvis, C.D.; Majem, M.; Forster, M.D.; Monnet, I.; et al. Long-Term Outcomes and Retreatment Among Patients With Previously Treated, Programmed Death-Ligand 1—Positive, Advanced Non—Small-Cell Lung Cancer in the KEYNOTE-010 Study. J. Clin. Oncol. 2020, 38, 1580–1590. [Google Scholar] [CrossRef]
  4. Bade, B.C.; Dela Cruz, C.S. Lung Cancer 2020: Epidemiology, Etiology, and Prevention. Clin. Chest. Med. 2020, 41, 1–24. [Google Scholar] [CrossRef]
  5. National Cancer Institute. Surveillance Epidemiology, and End Results Program (SEER). Cancer Stat Facts: Lung and Bronchus Cancer. 2022. Available online: https://seer.cancer.gov/statfacts/html/lungb.html (accessed on 11 April 2023).
  6. Chang, G.C.; Chiu, C.H.; Yu, C.J.; Chang, Y.C.; Chang, Y.H.; Hsu, K.H.; Wu, Y.C.; Chen, C.Y.; Hsu, H.H.; Wu, M.T.; et al. Low-dose CT screening among never-smokers with or without a family history of lung cancer in Taiwan: A prospective cohort study. Lancet Respir. Med. 2024, 12, 141–152. [Google Scholar] [CrossRef]
  7. Yang, C.Y.; Lin, Y.T.; Lin, L.J.; Chang, Y.H.; Chen, H.Y.; Wang, Y.P.; Shih, J.Y.; Yu, C.J.; Yang, P.C. Stage Shift Improves Lung Cancer Survival: Real-World Evidence. J. Thorac. Oncol. 2023, 18, 47–56. [Google Scholar] [CrossRef]
  8. Aberle, D.R.; Adams, A.M.; Berg, C.D.; Black, W.C.; Clapp, J.D.; Fagerstrom, R.M.; Gareen, I.F.; Gatsonis, C.; Marcus, P.M.; Sicks, J.D. Reduced lung-cancer mortality with low-dose computed tomographic screening. N. Engl. J. Med. 2011, 365, 395–409. [Google Scholar] [CrossRef]
  9. de Koning, H.J.; van der Aalst, C.M.; de Jong, P.A.; Scholten, E.T.; Nackaerts, K.; Heuvelmans, M.A.; Lammers, J.J.; Weenink, C.; Yousaf-Khan, U.; Horeweg, N.; et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. N. Engl. J. Med. 2020, 382, 503–513. [Google Scholar] [CrossRef] [PubMed]
  10. Lam, D.C.-L.; Liam, C.-K.; Andarini, S.; Park, S.; Tan, D.S.W.; Singh, N.; Jang, S.H.; Vardhanabhuti, V.; Ramos, A.B.; Nakayama, T.; et al. Lung Cancer Screening in Asia: An Expert Consensus Report. J. Thorac. Oncol. 2023, 18, 1303–1322. [Google Scholar] [CrossRef]
  11. Wait, S.; Alvarez-Rosete, A.; Osama, T.; Bancroft, D.; Cornelissen, R.; Marusic, A.; Garrido, P.; Adamek, M.; van Meerbeeck, J.; Snoeckx, A.; et al. Implementing Lung Cancer Screening in Europe: Taking a Systems Approach. JTO Clin. Res. Rep. 2022, 3, 100329. [Google Scholar] [CrossRef]
  12. van Meerbeeck, J.P.; Franck, C. Lung cancer screening in Europe: Where are we in 2021? Transl Lung Cancer Res. 2021, 10, 2407–2417. [Google Scholar] [CrossRef] [PubMed]
  13. Poon, C.; Wilsdon, T.; Sarwar, I.; Roediger, A.; Yuan, M. Why is the screening rate in lung cancer still low? A seven-country analysis of the factors affecting adoption. Front. Public Health 2023, 11, 1264342. [Google Scholar] [CrossRef]
  14. Force, U.S.P.S.T.; Krist, A.H.; Davidson, K.W.; Mangione, C.M.; Barry, M.J.; Cabana, M.; Caughey, A.B.; Davis, E.M.; Donahue, K.E.; Doubeni, C.A.; et al. Screening for Lung Cancer: US Preventive Services Task Force Recommendation Statement. JAMA 2021, 325, 962–970. [Google Scholar] [CrossRef]
  15. Behar Harpaz, S.; Weber, M.F.; Wade, S.; Ngo, P.J.; Vaneckova, P.; Sarich, P.E.A.; Cressman, S.; Tammemagi, M.C.; Fong, K.; Marshall, H.; et al. Updated cost-effectiveness analysis of lung cancer screening for Australia, capturing differences in the health economic impact of NELSON and NLST outcomes. Br. J. Cancer 2023, 128, 91–101. [Google Scholar] [CrossRef] [PubMed]
  16. Chen, Y.; Watson, T.R.; Criss, S.D.; Eckel, A.; Palazzo, L.; Sheehan, D.F.; Kong, C.Y. A simulation study of the effect of lung cancer screening in China, Japan, Singapore, and South Korea. PLoS ONE 2019, 14, e0220610. [Google Scholar] [CrossRef]
  17. Freitas, C.; Sousa, C.; Machado, F.; Serino, M.; Santos, V.; Cruz-Martins, N.; Teixeira, A.; Cunha, A.; Pereira, T.; Oliveira, H.P.; et al. The Role of Liquid Biopsy in Early Diagnosis of Lung Cancer. Front. Oncol. 2021, 11, 634316. [Google Scholar] [CrossRef] [PubMed]
  18. Bestvina, C.M.; Garassino, M.C.; Neal, J.W.; Wakelee, H.A.; Diehn, M.; Vokes, E.E. Early-Stage Lung Cancer: Using Circulating Tumor DNA to Get Personal. J. Clin. Oncol. 2023, 41, 4093–4096. [Google Scholar] [CrossRef]
  19. Schrag, D.; Beer, T.M.; McDonnell, C.H., 3rd; Nadauld, L.; Dilaveri, C.A.; Reid, R.; Marinac, C.R.; Chung, K.C.; Lopatin, M.; Fung, E.T.; et al. Blood-based tests for multicancer early detection (PATHFINDER): A prospective cohort study. Lancet 2023, 402, 1251–1260. [Google Scholar] [CrossRef]
  20. Poh, J.; Ngeow, K.C.; Pek, M.; Tan, K.H.; Lim, J.S.; Chen, H.; Ong, C.K.; Lim, J.Q.; Lim, S.T.; Lim, C.M.; et al. Analytical and clinical validation of an amplicon-based next generation sequencing assay for ultrasensitive detection of circulating tumor DNA. PLoS ONE 2022, 17, e0267389. [Google Scholar] [CrossRef]
  21. Marinello, A.; Tagliamento, M.; Pagliaro, A.; Conci, N.; Cella, E.; Vasseur, D.; Remon, J.; Levy, A.; Dall′Olio, F.G.; Besse, B. Circulating tumor DNA to guide diagnosis and treatment of localized and locally advanced non-small cell lung cancer. Cancer Treat. Rev. 2024, 129, 102791. [Google Scholar] [CrossRef]
  22. Bittla, P.; Kaur, S.; Sojitra, V.; Zahra, A.; Hutchinson, J.; Folawemi, O.; Khan, S. Exploring Circulating Tumor DNA (CtDNA) and Its Role in Early Detection of Cancer: A Systematic Review. Cureus 2023, 15, e45784. [Google Scholar] [CrossRef] [PubMed]
  23. Kan, C.F.K.; Unis, G.D.; Li, L.Z.; Gunn, S.; Li, L.; Soyer, H.P.; Stark, M.S. Circulating Biomarkers for Early Stage Non-Small Cell Lung Carcinoma Detection: Supplementation to Low-Dose Computed Tomography. Front. Oncol. 2021, 11, 555331. [Google Scholar] [CrossRef]
  24. Bettegowda, C.; Sausen, M.; Leary, R.J.; Kinde, I.; Wang, Y.; Agrawal, N.; Bartlett, B.R.; Wang, H.; Luber, B.; Alani, R.M.; et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci. Transl. Med. 2014, 6, 224ra224. [Google Scholar] [CrossRef]
  25. Li, R.Y.; Liang, Z.Y. Circulating tumor DNA in lung cancer: Real-time monitoring of disease evolution and treatment response. Chin. Med. J. 2020, 133, 2476–2485. [Google Scholar] [CrossRef] [PubMed]
  26. Iorio, M.V.; Croce, C.M. MicroRNAs in cancer: Small molecules with a huge impact. J. Clin. Oncol. 2009, 27, 5848–5856. [Google Scholar] [CrossRef]
  27. Nana-Sinkam, S.P.; Croce, C.M. MicroRNAs as therapeutic targets in cancer. Transl. Res. 2011, 157, 216–225. [Google Scholar] [CrossRef]
  28. Wu, K.-L.; Tsai, Y.-M.; Lien, C.-T.; Kuo, P.-L.; Hung, J.-Y. The Roles of MicroRNA in Lung Cancer. Int. J. Mol. Sci. 2019, 20, 1611. [Google Scholar] [CrossRef] [PubMed]
  29. Carrà, G.; Petiti, J.; Tolino, F.; Vacca, R.; Orso, F. MicroRNAs in metabolism for precision treatment of lung cancer. Cell. Mol. Biol. Lett. 2024, 29, 121. [Google Scholar] [CrossRef] [PubMed]
  30. Tang, Z.; Liang, D.; Deubler, E.L.; Sarnat, J.A.; Chow, S.S.; Diver, W.R.; Wang, Y. Lung cancer metabolomics: A pooled analysis in the Cancer Prevention Studies. BMC Med. 2024, 22, 262. [Google Scholar] [CrossRef]
  31. Madama, D.; Martins, R.; Pires, A.S.; Botelho, M.F.; Alves, M.G.; Abrantes, A.M.; Cordeiro, C.R. Metabolomic Profiling in Lung Cancer: A Systematic Review. Metabolites 2021, 11, 630. [Google Scholar] [CrossRef]
  32. Boeri, M.; Verri, C.; Conte, D.; Roz, L.; Modena, P.; Facchinetti, F.; Calabrò, E.; Croce, C.M.; Pastorino, U.; Sozzi, G. MicroRNA signatures in tissues and plasma predict development and prognosis of computed tomography detected lung cancer. Proc. Natl. Acad. Sci. USA 2011, 108, 3713–3718. [Google Scholar] [CrossRef] [PubMed]
  33. Nadal, E.; Zhong, J.; Lin, J.; Reddy, R.M.; Ramnath, N.; Orringer, M.B.; Chang, A.C.; Beer, D.G.; Chen, G. A MicroRNA Cluster at 14q32 Drives Aggressive Lung Adenocarcinoma. Clin. Cancer Res. 2014, 20, 3107–3117. [Google Scholar] [CrossRef]
  34. Bianchi, F.; Nicassio, F.; Marzi, M.; Belloni, E.; Dall′olio, V.; Bernard, L.; Pelosi, G.; Maisonneuve, P.; Veronesi, G.; Di Fiore, P.P. A serum circulating miRNA diagnostic test to identify asymptomatic high-risk individuals with early stage lung cancer. EMBO Mol. Med. 2011, 3, 495–503. [Google Scholar] [CrossRef]
  35. Ying, L.; Du, L.; Zou, R.; Shi, L.; Zhang, N.; Jin, J.; Xu, C.; Zhang, F.; Zhu, C.; Wu, J.; et al. Development of a serum miRNA panel for detection of early stage non-small cell lung cancer. Proc. Natl. Acad. Sci. USA 2020, 117, 25036–25042. [Google Scholar] [CrossRef] [PubMed]
  36. Montani, F.; Marzi, M.J.; Dezi, F.; Dama, E.; Carletti, R.M.; Bonizzi, G.; Bertolotti, R.; Bellomi, M.; Rampinelli, C.; Maisonneuve, P.; et al. miR-Test: A Blood Test for Lung Cancer Early Detection. J. Natl. Cancer Inst. 2015, 107, djv063. [Google Scholar] [CrossRef] [PubMed]
  37. Kirschner, M.B.; Kao, S.C.; Edelman, J.J.; Armstrong, N.J.; Vallely, M.P.; van Zandwijk, N.; Reid, G. Haemolysis during sample preparation alters microRNA content of plasma. PLoS ONE 2011, 6, e24145. [Google Scholar] [CrossRef] [PubMed]
  38. Liu, X.H.; Lu, K.H.; Wang, K.M.; Sun, M.; Zhang, E.B.; Yang, J.S.; Yin, D.D.; Liu, Z.L.; Zhou, J.; Liu, Z.J.; et al. MicroRNA-196a promotes non-small cell lung cancer cell proliferation and invasion through targeting HOXA5. BMC Cancer 2012, 12, 348. [Google Scholar] [CrossRef]
  39. Liu, Q.; Bai, W.; Huang, F.; Tang, J.; Lin, X. Downregulation of microRNA-196a inhibits stem cell self-renewal ability and stemness in non-small-cell lung cancer through upregulating GPX3 expression. Int. J. Biochem. Cell Biol. 2019, 115, 105571. [Google Scholar] [CrossRef]
  40. Guerriero, I.; D′Angelo, D.; Pallante, P.; Santos, M.; Scrima, M.; Malanga, D.; De Marco, C.; Ravo, M.; Weisz, A.; Laudanna, C.; et al. Correction: Analysis of miRNA profiles identified miR-196a as a crucial mediator of aberrant PI3K/AKT signaling in lung cancer cells. Oncotarget 2022, 13, 755. [Google Scholar] [CrossRef]
  41. Wu, C.-Y.; Ghule, S.S.; Liaw, C.-C.; Achudhan, D.; Fang, S.-Y.; Liu, P.-I.; Huang, C.-L.; Hsieh, C.-L.; Tang, C.-H. Ugonin P inhibits lung cancer motility by suppressing DPP-4 expression via promoting the synthesis of miR-130b-5p. Biomed. Pharmacother. 2023, 167, 115483. [Google Scholar] [CrossRef]
  42. Zhang, W.C.; Chin, T.M.; Yang, H.; Nga, M.E.; Lunny, D.P.; Lim, E.K.H.; Sun, L.L.; Pang, Y.H.; Leow, Y.N.; Malusay, S.R.Y.; et al. Tumour-initiating cell-specific miR-1246 and miR-1290 expression converge to promote non-small cell lung cancer progression. Nat. Commun. 2016, 7, 11702. [Google Scholar] [CrossRef]
  43. Wei, K.; Pan, C.; Yao, G.; Liu, B.; Ma, T.; Xia, Y.; Jiang, W.; Chen, L.; Chen, Y. MiR-106b-5p Promotes Proliferation and Inhibits Apoptosis by Regulating BTG3 in Non-Small Cell Lung Cancer. Cell Physiol. Biochem. 2017, 44, 1545–1558. [Google Scholar] [CrossRef]
  44. Kim, Y.; Kim, H.; Bang, S.; Jee, S.; Jang, K. MicroRNA-130b functions as an oncogene and is a predictive marker of poor prognosis in lung adenocarcinoma. Lab. Investig. 2021, 101, 155–164. [Google Scholar] [CrossRef] [PubMed]
  45. Kim, G.; An, H.J.; Lee, M.J.; Song, J.Y.; Jeong, J.Y.; Lee, J.H.; Jeong, H.C. Hsa-miR-1246 and hsa-miR-1290 are associated with stemness and invasiveness of non-small cell lung cancer. Lung Cancer 2016, 91, 15–22. [Google Scholar] [CrossRef] [PubMed]
  46. El-Aal, A.E.A.; Elshafei, A.; Ismail, M.Y.; El-Shafey, M.M. Identification of miR-106b-5p, miR-601, and miR-760 Expression and Their Clinical Values in Non-Small Cell Lung Cancer (NSCLC) Patients’ Serum. Pathol. Res. Pract. 2023, 248, 154663. [Google Scholar] [CrossRef]
  47. Tang, Y.; Zhou, L.; Wang, F.; Huang, Y.; Wang, J.; Zhao, S.; Qi, L.; Liu, L.; Liang, M.; Hou, D.; et al. Assessing the efficiency of eligibility criteria for low-dose computed tomography lung screening in China according to current guidelines. BMC Med. 2024, 22, 267. [Google Scholar] [CrossRef] [PubMed]
  48. Loh, C.H.; Koh, P.W.; Ang, D.J.M.; Lee, W.C.; Chew, W.M.; Koh, J.M.K. Characteristics of Singapore lung cancer patients who miss out on lung cancer screening recommendations. Singap. Med. J. 2024, 65, 279–287. [Google Scholar] [CrossRef]
  49. Kakinuma, R.; Muramatsu, Y.; Asamura, H.; Watanabe, S.-i.; Kusumoto, M.; Tsuchida, T.; Kaneko, M.; Tsuta, K.; Maeshima, A.M.; Ishii, G.; et al. Low-dose CT lung cancer screening in never-smokers and smokers: Results of an eight-year observational study. Transl. Lung Cancer Res. 2020, 9, 10–22. [Google Scholar] [CrossRef]
  50. Liu, Y.; Xu, H.; Lv, L.; Wang, X.; Kang, R.; Guo, X.; Wang, H.; Zheng, L.; Liu, H.; Guo, L.; et al. Risk-based lung cancer screening in heavy smokers: A benefit–harm and cost-effectiveness modeling study. BMC Med. 2024, 22, 73. [Google Scholar] [CrossRef]
  51. Ten Haaf, K.; Bastani, M.; Cao, P.; Jeon, J.; Toumazis, I.; Han, S.S.; Plevritis, S.K.; Blom, E.F.; Kong, C.Y.; Tammemagi, M.C.; et al. A Comparative Modeling Analysis of Risk-Based Lung Cancer Screening Strategies. J. Natl. Cancer Inst. 2020, 112, 466–479. [Google Scholar] [CrossRef]
  52. Field, J.K.; Vulkan, D.; Davies, M.P.A.; Baldwin, D.R.; Brain, K.E.; Devaraj, A.; Eisen, T.; Gosney, J.; Green, B.A.; Holemans, J.A.; et al. Lung cancer mortality reduction by LDCT screening: UKLS randomised trial results and international meta-analysis. Lancet Reg. Health Eur. 2021, 10, 100179. [Google Scholar] [CrossRef]
  53. Kovalchik, S.A.; Tammemagi, M.; Berg, C.D.; Caporaso, N.E.; Riley, T.L.; Korch, M.; Silvestri, G.A.; Chaturvedi, A.K.; Katki, H.A. Targeting of low-dose CT screening according to the risk of lung-cancer death. N. Engl. J. Med. 2013, 369, 245–254. [Google Scholar] [CrossRef] [PubMed]
  54. Robbins, H.A.; Cheung, L.C.; Chaturvedi, A.K.; Baldwin, D.R.; Berg, C.D.; Katki, H.A. Management of Lung Cancer Screening Results Based on Individual Prediction of Current and Future Lung Cancer Risks. J. Thorac. Oncol. 2022, 17, 252–263. [Google Scholar] [CrossRef] [PubMed]
  55. Lokhandwala, T.; Bittoni, M.A.; Dann, R.A.; D′Souza, A.O.; Johnson, M.; Nagy, R.J.; Lanman, R.B.; Merritt, R.E.; Carbone, D.P. Costs of Diagnostic Assessment for Lung Cancer: A Medicare Claims Analysis. Clin. Lung Cancer 2017, 18, e27–e34. [Google Scholar] [CrossRef] [PubMed]
  56. Gould, M.K.; Donington, J.; Lynch, W.R.; Mazzone, P.J.; Midthun, D.E.; Naidich, D.P.; Wiener, R.S. Evaluation of Individuals with Pulmonary Nodules: When Is It Lung Cancer?: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2013, 143 (Suppl. S5), e93S–e120S. [Google Scholar] [CrossRef]
  57. Callister, M.E.J.; Baldwin, D.R.; Akram, A.R.; Barnard, S.; Cane, P.; Draffan, J.; Franks, K.; Gleeson, F.; Graham, R.; Malhotra, P.; et al. British Thoracic Society guidelines for the investigation and management of pulmonary nodules: Accredited by NICE. Thorax 2015, 70 (Suppl. S2), ii1. [Google Scholar] [CrossRef]
  58. MacMahon, H.; Naidich, D.P.; Goo, J.M.; Lee, K.S.; Leung, A.N.C.; Mayo, J.R.; Mehta, A.C.; Ohno, Y.; Powell, C.A.; Prokop, M.; et al. Guidelines for Management of Incidental Pulmonary Nodules Detected on CT Images: From the Fleischner Society 2017. Radiology 2017, 284, 228–243. [Google Scholar] [CrossRef]
  59. Baldwin, D.R.; Brain, K.; Quaife, S. Participation in lung cancer screening. Transl. Lung Cancer Res. 2021, 10, 1091–1098. [Google Scholar] [CrossRef]
  60. Heideman, B.E.; Kammer, M.N.; Paez, R.; Swanson, T.; Godfrey, C.M.; Low, S.-W.; Xiao, D.; Li, T.Z.; Richardson, J.R.; Knight, M.A.; et al. The Lung Cancer Prediction Model “Stress Test”: Assessment of Models’ Performance in a High-Risk Prospective Pulmonary Nodule Cohort. CHEST Pulm. 2024, 2, 100033. [Google Scholar] [CrossRef]
  61. Bai, C.; Choi, C.-M.; Chu, C.M.; Anantham, D.; Chung-man Ho, J.; Khan, A.Z.; Lee, J.-M.; Li, S.Y.; Saenghirunvattana, S.; Yim, A. Evaluation of Pulmonary Nodules: Clinical Practice Consensus Guidelines for Asia. Chest 2016, 150, 877–893. [Google Scholar] [CrossRef]
  62. Pastorino, U.; Boeri, M.; Sestini, S.; Sabia, F.; Milanese, G.; Silva, M.; Suatoni, P.; Verri, C.; Cantarutti, A.; Sverzellati, N.; et al. Baseline computed tomography screening and blood microRNA predict lung cancer risk and define adequate intervals in the BioMILD trial. Ann. Oncol. 2022, 33, 395–405. [Google Scholar] [CrossRef] [PubMed]
  63. Zhang, Y.K.; Zhu, W.Y.; He, J.Y.; Chen, D.D.; Huang, Y.Y.; Le, H.B.; Liu, X.G. miRNAs expression profiling to distinguish lung squamous-cell carcinoma from adenocarcinoma subtypes. J. Cancer Res. Clin. Oncol. 2012, 138, 1641–1650. [Google Scholar] [CrossRef]
  64. Geng, Q.; Fan, T.; Zhang, B.; Wang, W.; Xu, Y.; Hu, H. Five microRNAs in plasma as novel biomarkers for screening of early-stage non-small cell lung cancer. Respir. Res. 2014, 15, 149. [Google Scholar] [CrossRef]
  65. Siddika, T.; Heinemann, I.U. Bringing MicroRNAs to Light: Methods for MicroRNA Quantification and Visualization in Live Cells. Front. Bioeng. Biotechnol. 2021, 8, 619583. [Google Scholar] [CrossRef]
  66. Karabegović, I.; Maas, S.C.E.; Shuai, Y.; Ikram, M.A.; Stricker, B.; Aerts, J.; Brusselle, G.; Lahousse, L.; Voortman, T.; Ghanbari, M. Smoking-related dysregulation of plasma circulating microRNAs: The Rotterdam study. Hum. Genom. 2023, 17, 61. [Google Scholar] [CrossRef] [PubMed]
  67. Huang, D.; Qu, D. Early diagnostic and prognostic value of serum exosomal miR-1246 in non-small cell lung cancer. Int. J. Clin. Exp. Pathol. 2020, 13, 1601–1607. [Google Scholar] [PubMed]
  68. Cordoba-Lanus, E.; Dominguez de-Barros, A.; Oliva, A.; Mayato, D.; Gonzalvo, F.; Remirez-Sanz, A.; Zulueta, J.J.; Celli, B.; Casanova, C. Circulating miR-206 and miR-1246 as Markers in the Early Diagnosis of Lung Cancer in Patients with Chronic Obstructive Pulmonary Disease. Int. J. Mol. Sci. 2023, 24, 12437. [Google Scholar] [CrossRef]
  69. Mo, D.; Gu, B.; Gong, X.; Wu, L.; Wang, H.; Jiang, Y.; Zhang, B.; Zhang, M.; Zhang, Y.; Xu, J.; et al. miR-1290 is a potential prognostic biomarker in non-small cell lung cancer. J. Thorac. Dis. 2015, 7, 1570–1579. [Google Scholar] [CrossRef]
  70. Sozzi, G.; Boeri, M.; Rossi, M.; Verri, C.; Suatoni, P.; Bravi, F.; Roz, L.; Conte, D.; Grassi, M.; Sverzellati, N.; et al. Clinical utility of a plasma-based miRNA signature classifier within computed tomography lung cancer screening: A correlative MILD trial study. J. Clin. Oncol. 2014, 32, 768–773. [Google Scholar] [CrossRef]
  71. Yin, G.; Zhang, B.; Li, J. miR-221-3p promotes the cell growth of non-small cell lung cancer by targeting p27. Mol. Med. Rep. 2019, 20, 604–612. [Google Scholar] [CrossRef]
  72. Dou, L.; Han, K.; Xiao, M.; Lv, F. miR-223-5p Suppresses Tumor Growth and Metastasis in Non-Small Cell Lung Cancer by Targeting E2F8. Oncol. Res. 2019, 27, 261–268. [Google Scholar] [CrossRef] [PubMed]
  73. Zhang, H.; Mao, F.; Shen, T.; Luo, Q.; Ding, Z.; Qian, L.; Huang, J. Plasma miR-145, miR-20a, miR-21 and miR-223 as novel biomarkers for screening early-stage non-small cell lung cancer. Oncol. Lett. 2017, 13, 669–676. [Google Scholar] [CrossRef] [PubMed]
  74. D′Antona, P.; Cattoni, M.; Dominioni, L.; Poli, A.; Moretti, F.; Cinquetti, R.; Gini, E.; Daffrè, E.; Noonan, D.M.; Imperatori, A.; et al. Serum miR-223: A Validated Biomarker for Detection of Early-Stage Non–Small Cell Lung Cancer. Cancer Epidemiol. Biomark. Prev. 2019, 28, 1926–1933. [Google Scholar] [CrossRef]
  75. Asakura, K.; Kadota, T.; Matsuzaki, J.; Yoshida, Y.; Yamamoto, Y.; Nakagawa, K.; Takizawa, S.; Aoki, Y.; Nakamura, E.; Miura, J.; et al. A miRNA-based diagnostic model predicts resectable lung cancer in humans with high accuracy. Commun. Biol. 2020, 3, 134. [Google Scholar] [CrossRef]
  76. Foss, K.M.; Sima, C.; Ugolini, D.; Neri, M.; Allen, K.E.; Weiss, G.J. miR-1254 and miR-574-5p: Serum-Based microRNA Biomarkers for Early-Stage Non-small Cell Lung Cancer. J. Thorac. Oncol. 2011, 6, 482–488. [Google Scholar] [CrossRef] [PubMed]
  77. Ling, B.; Liao, X.; Tang, Q.; Ye, G.; Bin, X.; Wang, J.; Pang, Y.; Qi, G. MicroRNA-106b-5p inhibits growth and progression of lung adenocarcinoma cells by downregulating IGSF10. Aging 2021, 13, 18740–18756. [Google Scholar] [CrossRef]
  78. Wu, T.; Chen, W.; Liu, S.; Lu, H.; Wang, H.; Kong, D.; Huang, X.; Kong, Q.; Ning, Y.; Lu, Z. Huaier suppresses proliferation and induces apoptosis in human pulmonary cancer cells via upregulation of miR-26b-5p. FEBS Lett. 2014, 588, 2107–2114. [Google Scholar] [CrossRef]
  79. Liu, Y.; Zhang, G.; Chen, H.; Wang, H. Silencing lncRNA DUXAP8 inhibits lung adenocarcinoma progression by targeting miR-26b-5p. Biosci. Rep. 2021, 41, BSR20200884. [Google Scholar] [CrossRef]
  80. Li, D.; Wei, Y.; Wang, D.; Gao, H.; Liu, K. MicroRNA-26b suppresses the metastasis of non-small cell lung cancer by targeting MIEN1 via NF-κB/MMP-9/VEGF pathways. Biochem. Biophys. Res. Commun. 2016, 472, 465–470. [Google Scholar] [CrossRef]
  81. Xue, X.; Liu, Y.; Wang, Y.; Meng, M.; Wang, K.; Zang, X.; Zhao, S.; Sun, X.; Cui, L.; Pan, L. MiR-21 and MiR-155 promote non-small cell lung cancer progression by downregulating SOCS1, SOCS6, and PTEN. Oncotarget 2016, 7, 84508. [Google Scholar] [CrossRef]
  82. Lv, D.; Bi, Q.; Li, Y.; Deng, J.; Wu, N.; Hao, S.; Zhao, M. Long non-coding RNA MEG3 inhibits cell migration and invasion of non-small cell lung cancer cells by regulating the miR-21-5p/PTEN axis. Mol. Med. Rep. 2021, 23, 191. [Google Scholar] [CrossRef] [PubMed]
  83. Hirono, T.; Jingushi, K.; Nagata, T.; Sato, M.; Minami, K.; Aoki, M.; Takeda, A.H.; Umehara, T.; Egawa, H.; Nakatsuji, Y.; et al. MicroRNA-130b functions as an oncomiRNA in non-small cell lung cancer by targeting tissue inhibitor of metalloproteinase-2. Sci. Rep. 2019, 9, 6956. [Google Scholar] [CrossRef]
  84. Shao, C.; Yang, F.; Qin, Z.; Jing, X.; Shu, Y.; Shen, H. The value of miR-155 as a biomarker for the diagnosis and prognosis of lung cancer: A systematic review with meta-analysis. BMC Cancer 2019, 19, 1103. [Google Scholar] [CrossRef]
  85. Hetta, H.F.; Zahran, A.M.; Shafik, E.A.; El-Mahdy, R.I.; Mohamed, N.A.; Nabil, E.E.; Esmaeel, H.M.; Alkady, O.A.; Elkady, A.; Mohareb, D.A.; et al. Circulating miRNA-21 and miRNA-23a Expression Signature as Potential Biomarkers for Early Detection of Non-Small-Cell Lung Cancer. Microrna 2019, 8, 206–215. [Google Scholar] [CrossRef]
  86. Hsu, Y.L.; Hung, J.Y.; Chang, W.A.; Lin, Y.S.; Pan, Y.C.; Tsai, P.H.; Wu, C.Y.; Kuo, P.L. Hypoxic lung cancer-secreted exosomal miR-23a increased angiogenesis and vascular permeability by targeting prolyl hydroxylase and tight junction protein ZO-1. Oncogene 2017, 36, 4929–4942. [Google Scholar] [CrossRef] [PubMed]
  87. Reis, P.P.; Drigo, S.A.; Carvalho, R.F.; Lopez Lapa, R.M.; Felix, T.F.; Patel, D.; Cheng, D.; Pintilie, M.; Liu, G.; Tsao, M.S. Circulating miR-16-5p, miR-92a-3p, and miR-451a in Plasma from Lung Cancer Patients: Potential Application in Early Detection and a Regulatory Role in Tumorigenesis Pathways. Cancers 2020, 12, 2071. [Google Scholar] [CrossRef] [PubMed]
  88. Chen, X.; Hu, Z.; Wang, W.; Ba, Y.; Ma, L.; Zhang, C.; Wang, C.; Ren, Z.; Zhao, Y.; Wu, S.; et al. Identification of ten serum microRNAs from a genome-wide serum microRNA expression profile as novel noninvasive biomarkers for nonsmall cell lung cancer diagnosis. Int. J. Cancer 2012, 130, 1620–1628. [Google Scholar] [CrossRef]
  89. Chen, W.; Li, X. MiR-222-3p Promotes Cell Proliferation and Inhibits Apoptosis by Targeting PUMA (BBC3) in Non-Small Cell Lung Cancer. Technol. Cancer Res. Treat. 2020, 19, 1533033820922558. [Google Scholar] [CrossRef]
  90. Kanaoka, R.; Iinuma, H.; Dejima, H.; Sakai, T.; Uehara, H.; Matsutani, N.; Kawamura, M. Usefulness of Plasma Exosomal MicroRNA-451a as a Noninvasive Biomarker for Early Prediction of Recurrence and Prognosis of Non-Small Cell Lung Cancer. Oncology 2018, 94, 311–323. [Google Scholar] [CrossRef]
  91. Tian, F.; Wang, J.; Ouyang, T.; Lu, N.; Lu, J.; Shen, Y.; Bai, Y.; Xie, X.; Ge, Q. MiR-486-5p Serves as a Good Biomarker in Nonsmall Cell Lung Cancer and Suppresses Cell Growth With the Involvement of a Target PIK3R1. Front. Genet. 2019, 10, 688. [Google Scholar] [CrossRef]
  92. Wu, T.; Hu, H.; Zhang, T.; Jiang, L.; Li, X.; Liu, S.; Zheng, C.; Yan, G.; Chen, W.; Ning, Y.; et al. miR-25 Promotes Cell Proliferation, Migration, and Invasion of Non-Small-Cell Lung Cancer by Targeting the LATS2/YAP Signaling Pathway. Oxid. Med. Cell Longev. 2019, 2019, 9719723. [Google Scholar] [CrossRef] [PubMed]
  93. Wang, W.; Li, X.; Liu, C.; Zhang, X.; Wu, Y.; Diao, M.; Tan, S.; Huang, S.; Cheng, Y.; You, T. MicroRNA-21 as a diagnostic and prognostic biomarker of lung cancer: A systematic review and meta-analysis. Biosci. Rep. 2022, 42, BSR20211653. [Google Scholar] [CrossRef]
  94. Ma, F.; Xie, Y.; Lei, Y.; Kuang, Z.; Liu, X. The microRNA-130a-5p/RUNX2/STK32A network modulates tumor invasive and metastatic potential in non-small cell lung cancer. BMC Cancer 2020, 20, 580. [Google Scholar] [CrossRef]
  95. Chen, S.; Li, P.; Yang, R.; Cheng, R.; Zhang, F.; Wang, Y.; Chen, X.; Sun, Q.; Zang, W.; Du, Y.; et al. microRNA-30b inhibits cell invasion and migration through targeting collagen triple helix repeat containing 1 in non-small cell lung cancer. Cancer Cell Int. 2015, 15, 85. [Google Scholar] [CrossRef] [PubMed]
  96. Wu, R.; Zhang, B.; He, M.; Kang, Y.; Zhang, G. MicroRNA biomarkers and their use in evaluating the prognosis of lung cancer. J. Cancer Res. Clin. Oncol. 2023, 149, 16753–16761. [Google Scholar] [CrossRef]
  97. Zhang, W.; Lin, X.; Li, X.; Zhang, H.; Wang, M.; Sun, W.; Han, X.; Sun, D. Transcriptional identification of potential biomarkers of lung adenocarcinoma. J. Shanghai Jiaotong Univ. (Med. Sci.) 2020, 40, 1598–1606. [Google Scholar]
  98. Yong-Ming, H.; Ai-Jun, J.; Xiao-Yue, X.; Jian-Wei, L.; Chen, Y.; Ye, C. miR-449a: A potential therapeutic agent for cancer. Anti-Cancer Drugs 2017, 28, 1067–1078. [Google Scholar] [CrossRef]
  99. Wang, F.; Lou, J.F.; Cao, Y.; Shi, X.H.; Wang, P.; Xu, J.; Xie, E.F.; Xu, T.; Sun, R.H.; Rao, J.Y.; et al. miR-638 is a new biomarker for outcome prediction of non-small cell lung cancer patients receiving chemotherapy. Exp. Mol. Med. 2015, 47, e162. [Google Scholar] [CrossRef]
  100. Xia, Y.; Wu, Y.; Liu, B.; Wang, P.; Chen, Y. Downregulation of miR-638 promotes invasion and proliferation by regulating SOX2 and induces EMT in NSCLC. FEBS Lett. 2014, 588, 2238–2245. [Google Scholar] [CrossRef]
  101. Izzotti, A.; Coronel Vargas, G.; Pulliero, A.; Coco, S.; Vanni, I.; Colarossi, C.; Blanco, G.; Agodi, A.; Barchitta, M.; Maugeri, A.; et al. Relationship between the miRNA Profiles and Oncogene Mutations in Non-Smoker Lung Cancer. Relevance for Lung Cancer Personalized Screenings and Treatments. J. Pers. Med. 2021, 11, 182. [Google Scholar] [CrossRef]
  102. Dutkowska, A.; Szmyd, B.; Kaszkowiak, M.; Domańska-Senderowska, D.; Pastuszak-Lewandoska, D.; Brzeziańska-Lasota, E.; Kordiak, J.; Antczak, A. Expression of inflammatory interleukins and selected miRNAs in non-small cell lung cancer. Sci. Rep. 2021, 11, 5092. [Google Scholar] [CrossRef] [PubMed]
  103. Zhang, T.X.; Duan, X.C.; Cui, Y.; Zhang, Y.; Gu, M.; Wang, Z.Y.; Li, W.Y. Clinical significance of miR-9-5p in NSCLC and its relationship with smoking. Front. Oncol. 2024, 14, 1376502. [Google Scholar] [CrossRef] [PubMed]
Cancers 17 00942 g001
Figure 1. Clinical features selection with RF (training).
Figure 1. Clinical features selection with RF (training).
Cancers 17 00942 g001
Cancers 17 00942 g002
Figure 2. miRNA biomarkers ranking with RF (training).
Figure 2. miRNA biomarkers ranking with RF (training).
Cancers 17 00942 g002
Cancers 17 00942 g003
Figure 3. Machine learning using NNET showed the best-performing ROC curve for Model 1.
Figure 3. Machine learning using NNET showed the best-performing ROC curve for Model 1.
Cancers 17 00942 g003
Cancers 17 00942 g004
Figure 4. Machine learning using KNN showed the best performing ROC curve for Model 2.
Figure 4. Machine learning using KNN showed the best performing ROC curve for Model 2.
Cancers 17 00942 g004
Table 1. Patient demographics and clinical characteristics.
Table 1. Patient demographics and clinical characteristics.
Clinical CharacteristicsControls (N = 123)Cases (N = 82)p Value
Patient demographics
Age (mean ± SD) 57.0 ± 14.266.5 ± 10.7<0.005
Gender (%)
Male72 (58.5) 54 (65.9)0.17
Female50 (40.7)28 (34.1)
Race (%)
Chinese95 (77.2)59 (72.0)0.23
Malay14 (11.4)10 (12.2)
Indian8 (6.5)4 (4.9)
Others5 (4.1)9 (11.0)
Missing values
Smoking history (%)
Never smoker89 (72.4)32 (39.0)
Smoker/ex-smoker29 (23.6)42 (51.2)<0.005
Missing values4 (3.3)8 (9.8)
Emphysema (%)
Yes9 (7.4)13 (15.9)0.018
No110 (89.4)61 (74.4)
Missing values4 (3.3)8 (9.8)
Cancer stage at diagnosis (%)
1N.A.22 (26.8)N.A.
2N.A.5 (6.1)
3N.A.8 (9.8)
4N.A.40 (48.8)
Limited (for SCLC)N.A.2 (2.4)
Extensive (for SCLC)N.A.5 (6.1)
Nodule characteristics
Number of nodules <0.005
None75 (61.0)0 (0.0)
Single29 (23.6)48 (58.5)
Multiple (>1)19 (15.4)34 (41.5)
Size (of the most suspicious/malignant nodule) in mm14.7 ± 24.939.7 ± 27.6 <0.005
Nodule type (%)
Ground glass opacity5 (4.1)4 (4.9)<0.005
Solid40 (32.5)64 (78.0)
Part solid2 (1.6)6 (7.3)
No nodule76 (61.8)0 (0.0)
Spiculation (%)
Spiculated/lobulated3 (2.4)26 (31.7)<0.005
Not spiculated44 (35.8)48 (58.5)
No nodule76 (61.8)0 (0.0)
Missing values0 (0.0)8 (9.8)
Location (%)
Upper lobe24 (19.5)40 (48.8)<0.005
Non-upper lobe23 (18.7)34 (41.5)
No nodule76 (61.8)0 (0.0)
Missing values0 (0.0)8 (9.8)
Histology (%)
AdenocarcinomaN.A.57 (69.5)N.A.
Squamous cell carcinomaN.A.11 (13.4)
Small cell lung cancerN.A.7 (8.5)
Other malignanciesN.A.6 (7.3)
Nodule biopsied, benign9 (7.3)N.A.
Nodule not biopsied38 (30.9)N.A.
No nodule75 (61.0)0 (0.0)
Table 2. Clinical performance index of the risk models using ML algorithms.
Table 2. Clinical performance index of the risk models using ML algorithms.
ML Method Model 1Model 2
KNNAUC0.7810.989
Sensitivity0.7000.921
Specificity0.7320.975
NNETAUC0.8630.961
Sensitivity0.7750.937
Specificity0.8500.929
SVMAUC0.8020.983
Sensitivity0.7620.975
Specificity0.7320.937
NBAUC0.7900.978
Sensitivity0.7120.912
Specificity0.7870.976
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Poh, K.C.; Ren, T.M.; Ling, G.L.; Goh, J.S.Y.; Rose, S.; Wong, A.; Mehta, S.S.; Goh, A.; Chong, P.-Y.; Cheng, S.W.; et al. Development of a miRNA-Based Model for Lung Cancer Detection. Cancers 2025, 17, 942. https://doi.org/10.3390/cancers17060942

AMA Style

Poh KC, Ren TM, Ling GL, Goh JSY, Rose S, Wong A, Mehta SS, Goh A, Chong P-Y, Cheng SW, et al. Development of a miRNA-Based Model for Lung Cancer Detection. Cancers. 2025; 17(6):942. https://doi.org/10.3390/cancers17060942

Chicago/Turabian Style

Poh, Kai Chin, Toh Ming Ren, Goh Liuh Ling, John S Y Goh, Sarrah Rose, Alexa Wong, Sanhita S. Mehta, Amelia Goh, Pei-Yu Chong, Sim Wey Cheng, and et al. 2025. "Development of a miRNA-Based Model for Lung Cancer Detection" Cancers 17, no. 6: 942. https://doi.org/10.3390/cancers17060942

APA Style

Poh, K. C., Ren, T. M., Ling, G. L., Goh, J. S. Y., Rose, S., Wong, A., Mehta, S. S., Goh, A., Chong, P.-Y., Cheng, S. W., Wang, S. S. Y., Saffari, S. E., Lim, D. W.-T., & Chia, N.-Y. (2025). Development of a miRNA-Based Model for Lung Cancer Detection. Cancers, 17(6), 942. https://doi.org/10.3390/cancers17060942

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop