Evaluation And Screening Of Molecular Markers For Lung Cancer Diagnosis Based On Surface-enhanced Raman Spectroscopy

Lung cancer（LC）has replaced liver cancer as the leading malignancy death in China and has become a major public health problem.The early-stage of LC is mostly in the form of lung nodules and lacks specific clinical symptoms;moreover,there is still a lack of accurate and safe early diagnosis methods as well as highly accurate and specific biomarkers for early diagnosis of LC.For tumor markers of LC,establishing a sensitive,rapid,and specific detection method to achieve accurate diagnosis of early-stage LC is a critical challenge that needs to be solved.This work focuses on exosomes and DNA methylation,novel tumor markers for LC.It develops new methods for the detection of early LC,adenocarcinoma in situ（AIS）,and benign lung diseases（BLD）based on surface-enhanced Raman spectroscopy（SERS）technology,with the following main content:Firstly,a pretreatment plasma-based label-free SERS assay,combined with machine learning algorithms,was proposed to develop a study of differentiation between early-stage LC groups and healthy controls（HC）.Plasma was diluted and adjusted to a weakly acidic pH,and subsequently mixed with silver nanoparticles（Ag NPs）and detection by SERS method.Additionally,liquid paraffin is combined to effectively avoid potential aerosol contamination of plasma samples during the detection process.The diagnostic results were further validated by principal component analysis-linear discriminant analysis（PCA-LDA）multivariate statistical analysis.The results showed that a 10-fold dilution of plasma concentration and a weak acidic detection condition significantly improved the diagnostic effect,with the area under the operating characteristic curve（AUC）values of 0.975.This study indicates that the study of plasma SERS detection in LC based on pH adjustment can reduce the effect of uric acid molecules and proteins in plasma on Raman signal;the experimental method has the advantages of easy operation and high efficiency,which is expected to provide a potential new strategy for non-invasive diagnosis of SERS in early-stage LC.A new method for detecting plasma exosomes by the multi-receptor SERS technique was constructed and allowed the differentiation of AIS and early-stage invasive adenocarcinoma（IAC）.This method has high reliability and sensitivity and provides a more effective tool for the early diagnosis of LC.Firstly,stable and uniform nanoparticles of gold nanoparticles loading on bacterial cellulose with two-dimensional SERS-enhanced substrate（BC/Au NPs film）were synthesized and modified low specificity and physicochemical selectivity molecules on BC/Au NPs film to obtain fingerprinting of exosomes.The signals of different exosome components are amplified selectively according to the characteristics of each functionalized molecule,improving the discrimination between two groups of samples with similar chemical composition.As a proof of concept,the functionalized BC/Au NPs film was validated based on cell-derived exosomes to obtain more spectral information to distinguish between LC cell lines and human lung epithelial cell lines by interacting with exosomes.Furthermore,we used multiple molecular receptors to generate multiple abundant SERS spectral features combined with the partial least squares regression discriminant analysis（PLS-DA）model to improve the discrimination between AIS and early-stage IAC（sensitivity of 90%and specificity of 95%）.Comparison with a single receptor sensing platform,the functionalized SERS sensor demonstrated superiority in the discrimination of early-stage IAC from AIS,with high reliability and sensitivity.Our developed SERS sensor method in this paper promises to provide a more effective means for early diagnosis of LC and pave a new direction for promoting early screening and diagnosis of LC.A sensitive and straightforward SERS method to detect the level of global DNA methylation was proposed and used to identify patients with early-stage LC and BLD.First,a reliable spectral marker for cytosine methylation was identified by comparing the SERS spectra of base and methylated DNA sequences.To validate the feasibility of the method for clinical application,we evaluated the methylation patterns of genomic DNA（gDNA）extracted from cell line models and formalin-fixed paraffin-embedded（FFPE）tissue samples using the SERS technique.The results showed different methylation patterns in gDNA from patients with early-stage LC（n=65）and BLD（n=41）.Combined with PLS-DA discriminant analysis,the AUC value of patients with early-stage LC and BLD was 0.85.We believe that SERS analysis based on altered DNA methylation combined with machine learning offers a new way for early detection of LC.A novel method based on the combination of magnetic substrates and SERS optical probes was constructed to detect SHOX2 and RASSF1A gene methylation in LC tissues and plasma.The method utilizes antibody-modified gold-coated magnetic nanospheres（Fe₃O₄@Au NPs）as a capture substrate for methylated DNA,and core-shell nanospheres functionalized with aptamers and internal standard molecules as SERS optical probes to specifically detect methylation of different genes.The method can be used to determine synthetic methylated SHOX2（m-SHOX2）and methylated RASSF1A（m-RASSF1A）with low detection limits of 1.15 PM and 0.43 PM,respectively,and can successfully detect methylation levels down to 1%.It has been successfully used to detect m-SHOX2 and m-RASSF1A of circulating free DNA（cfDNA）in FFPE samples and plasma,demonstrating high sensitivity and excellent specificity.Additionally,the method can use portable Raman as the Raman signal reading device,which has the advantage of easy operation and short analysis time,and also does not require bisulfite treatment and does not have tedious and time-consuming problems such as PCR amplification,which is expected to provide a new solution for clinical DNA methylation for Point-of-Care Testing（POCT）analysis applications.