A predictive index for the efficacy of immunotherapy for lung cancer and its application

Abstract

The present invention discloses a predictive index for the efficacy of immunotherapy for lung cancer and its application, and belongs to the technical field of predictive markers. The present invention provides a predictive index for the efficacy of immunotherapy for lung cancer, and the predictive index includes a specific oral flora. The present invention also discloses the application of a quantitative detection reagent for the predictive index in the preparation of a lung cancer prognosis prediction reagent. The present invention analyzes salivary microbiome and metabolite data based on non-targeted metabolomics liquid chromatography mass spectrometry (LC / MS) of Metagenomic Next‑Generation Sequencing (mNGS). Through multi-omics analysis, it is found that the distribution of specific oral flora has an important regulatory effect on immunotherapy resistance, and it is confirmed that its related fat metabolism is highly enriched in lung cancer patients with immunotherapy resistance.

Description

Technical Field

[0001] The present invention relates to the technical field of predictive markers, and particularly to a predictive index for the efficacy of lung cancer immunotherapy and its application. Background Art

[0002] Lung cancer remains an important public health problem. It is estimated that by 2024, nearly 125,070 people will die of lung cancer in the United States. In the past decade, significant progress has been made in the treatment of lung cancer. In-depth understanding of lung cancer biology has led to the development of targeted therapy and immunotherapy. Immune checkpoint inhibitors (ICIs) have shown significant benefits in the treatment of non-small cell lung cancer (NSCLC) and are now considered first-line treatment products for advanced lung cancer. However, despite the significant progress made in NSCLC treatment with ICIs, it is clear that a substantial number of patients do not respond well to these treatments. Nearly 70% of advanced NSCLC patients do not obtain durable benefits from ICI-based treatments.

[0003] Therefore, it is crucial to investigate the mechanisms of immunotherapy resistance and search for potential predictive markers. The surface of the human body is home to complex bacterial communities. The functions encoded within the human microbiome play important roles in all aspects of human physiology. The oral microbiome is the second most diverse habitat in the human body after the gut microbiome. The oral microbiota interacts with its host's immune system and plays a role not only in oral health but also in overall health, including lung diseases. It is reasonable that the oral microbiome is related to the efficacy of immunotherapy in lung cancer patients. However, the specific distribution characteristics and mechanisms of the oral microbiota in immunotherapy-sensitive and immunotherapy-resistant lung cancer patients are still unclear. Summary of the Invention

[0004] The object of the present invention is to provide a predictive index for the efficacy of lung cancer immunotherapy and its application to solve the problems existing in the above-mentioned prior art. The present invention discloses the important regulatory role of specific oral microbiota distribution in immunotherapy resistance and confirms that the related lipid metabolism is highly enriched in immunotherapy-resistant lung cancer patients.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] Technical solution 1: A prediction index for the efficacy of lung cancer immunotherapy, the prediction index including specific oral flora; the specific oral flora includes Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, Actinomyces georgiae, Granulicatella adiacens and Streptococcus oralis.

[0007] Further, the sample of the prediction index is saliva.

[0008] Technical solution 2: Application of the quantitative detection reagent of the prediction index in the preparation of a reagent for predicting the prognosis of lung cancer.

[0009] Further, the efficacy of lung cancer immunotherapy is predicted by detecting the distribution concentration of the specific oral flora.

[0010] Further, the Youden index of the ROC curve is used as the critical value. Lung cancer patients with a detection score ≥ Youden index are defined as the high-concentration group, and lung cancer patients with a detection score < Youden index are defined as the low-concentration group; when the specific oral flora Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis and Actinomyces georgia are distributed at a high concentration, it is a response to lung cancer immunotherapy; when the specific oral flora Granulicatella adiacens and Streptococcus oralis are distributed at a high concentration, it is a non-response to lung cancer immunotherapy.

[0011] The present invention discloses the following technical effects:

[0012] The present invention is based on the non-targeted metabolomics liquid chromatography-mass spectrometry (LC / MS) analysis of saliva microbiota and metabolite data by Metagenomic Next-Generation Sequencing (mNGS). Through multi-omics analysis, it discovers the important regulatory role of the distribution of specific oral microbiota in immune therapy resistance, and confirms that the related lipid metabolism is highly enriched in lung cancer patients with immune therapy resistance. When the distributions of specific oral microbiota Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, and Actinomyces georgia are at high concentrations, it is a response to lung cancer immunotherapy; when the distributions of specific oral microbiota Granulicatella adiacens and Streptococcus oralis are at high concentrations, it is a non-response to lung cancer immunotherapy. In addition, through in vitro experiments, the present invention further confirms that lung cancer immunotherapy resistance is closely related to abnormal lipid metabolism and verifies its effect on the expression of checkpoint PD-L1. Therefore, the present invention confirms and emphasizes the value of specific oral microbiota as a lung cancer immune regulator and determines its related lipid metabolism in regulating the immune signaling pathway. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0014] Figure 1 It shows the composition of the saliva microbiota of lung cancer patients. Among them, A shows the main phylum composition in lung cancer patients; B shows the main genus distribution in lung cancer patients; C shows the distribution of the top 20 oral bacterial species in lung cancer patients;

[0015] Figure 2 It shows the saliva microbiota structure between the NR and R groups. Among them, (A-D) α-diversity analysis shows that ACE (A), Chao1 (B), There was no significant difference in the Shannon index (D) between the two groups; (E - F) β-diversity analysis showed that there was also no significant difference in PCOA based on Bray-Curtis distance (E) and Jaccard distance (F) between the two groups;

[0016] Figure 3 Differential bacterial species between NR and R patients. Among them, LEfSe analysis in A identified 13 differential species; among them, 11 species were more enriched in the R group and 2 species were more enriched in the NR group; N: responder; NR: non-responder; Distribution and abundance of 12 major differential bacterial species identified by LEfSe in B; P values were calculated by Mann-Whitney U test; *p < 0.05, **p < 0.01; N: responder; NR: non-responder; Kaplan-Meier analysis of lung cancer patients according to the median abundance of each dominant differential species; High levels of Neisseria subflava C and Actinomyces meyeri D groups had a significant survival advantage compared to the low level groups; Low levels of Granulicatella adiacens (E) and Streptococcus oralis (F) groups had a significant survival advantage compared to the high level groups; P values were calculated by Log-rank test; *p < 0.05;

[0017] Figure 4 Changes in the 12 major differential bacterial species identified by Lefse in NR and NR-PD patients. Saliva samples of NR patients at disease progression were collected and classified as NR-PD, and the results showed that these 12 differential species remained stable during lung cancer treatment;

[0018] Figure 5 Relationship between the salivary microbiome and PD-L1 expression. Among them, A was LEfSe analysis to identify several differential bacterial species in patients with PD-L1 ≥ 10%; B - E were Spearman correlation analyses between the major differential bacterial species and PD-L1 expression; Neisseria subflava (B) and Actinomyces meyeri (C) were positively correlated with PD-L1 expression; Granulicatella adiacens (D) and Streptococcus oralis (E) had no significant correlation with PD-L1 expression;

[0019] Figure 6Metabolomics analysis revealed unique metabolic profiles between the NR and R groups; among them, PLS-DA analysis of the A-B positive ion model (A) and negative ion model (B) showed that the metabolic patterns between the two groups were significantly different; the differential metabolites of the C-D positive ion model (C) and negative ion model (D) were shown in the circular volcano plot, and most of them were lipids and lipid-like molecules;

[0020] Figure 7 It is a hierarchical clustering heatmap showing the abundances of differential metabolites for each patient; the data was normalized based on Z-score; among them, A is the positive ion model; the results showed that most differential metabolites were enriched in NR patients; all lipids and lipid-like molecules were enriched in the NR group and clustered together; B is the negative ion model, and the results showed that most differential lipids and lipid-like molecules were enriched in the NR group and clustered together;

[0021] Figure 8 It is the significant association between the sum of all lipids and lipid-like molecules in the negative ion model A and positive ion model B and the PD-L1 expression; the P-value and r-value were calculated by Spearman analysis;

[0022] Figure 9 It is the correlation heatmap of differential metabolites and differential species in the two groups. Spearman correlation analysis was performed, and the color intensity represents the r-value; among them, A is the positive ion model, and several differential metabolites including lipids and lipid-like metabolites were significantly correlated with certain bacterial species; B is the negative ion model;

[0023] Figure 10 It is the ROC curve result, analyzing the prediction sensitivity and specificity of the immune efficacy of the lung cancer patients by the flora of Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, Actinomyces georgiae, Lactobacillus fermentum, Granulicatella adiacens and Streptococcus oralis. Detailed implementation mode

[0024] The various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but should be understood as a more detailed description of certain aspects, characteristics and implementation schemes of the present invention.

[0025] It should be understood that the terms used in the present invention are only for describing specific embodiments and are not intended to limit the present invention. Additionally, for the numerical ranges in the present invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Any intermediate value within any stated value or stated range, as well as each smaller range between any other stated value or intermediate value within the stated range, is also included in the present invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0026] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention pertains. Although the present invention only describes preferred methods and materials, any methods and materials similar or equivalent to those described herein may also be used in the implementation or testing of the present invention. All documents mentioned in this specification are incorporated by reference to disclose and describe the methods and / or materials related to the said documents. In case of conflict with any incorporated document, the content of this specification shall prevail.

[0027] Without departing from the scope or spirit of the present invention, various improvements and changes can be made to the specific embodiments of the present invention specification, which are obvious to those skilled in the art. Other embodiments obtained from the present invention specification are obvious to those skilled in the art. The present invention specification and examples are merely exemplary.

[0028] Regarding the use of "comprising", "including", "having", "containing", etc. herein, they are all open-ended terms, meaning including but not limited to.

[0029] Example 1

[0030] I. Experimental Methods

[0031] 1. Patient Recruitment: From July 2021 to September 2022, 20 newly diagnosed patients with stage IIIB-IV NSCLC were recruited from the Department of Respiratory and Critical Care Medicine, Nanfang Hospital, Southern Medical University, China.

[0032] 2. Inclusion Criteria are as follows: (1) The diagnosis of advanced non-small cell lung cancer was confirmed by histopathological biopsy and imaging; (2) No anti-tumor treatment had been carried out before sampling, including surgical treatment, chemotherapy, radiotherapy, targeted therapy, immunotherapy, traditional Chinese medicine treatment, etc.; (3) The age was between 18 and 79 years old; (4) The PS score was 0-1; (4) No antibiotic treatment had been received within 1 month before sampling.

[0033] 3. Exclusion criteria are as follows: (1) Patients with a history of primary malignancies in other systems; (2) Patients with acute oral infections, acute pulmonary infections, acute exacerbations of chronic obstructive pulmonary disease, bronchial asthma, infectious bronchiectasis, active pulmonary tuberculosis, and other acute infectious diseases within the past 4 weeks; or patients with severe cardiac, respiratory, renal, or liver insufficiency; (4) Patients who have used oral hormones, immunosuppressants, or microbial agents within the past 4 weeks; (5) Patients with irregular anti-tumor treatment cycles; (6) Patients with lung adenocarcinoma with mutations in EGFR, ALK, ROS1, BRAF V600, NTRK, or MET exon 14. All enrolled patients received immunotherapy combined with chemotherapy as first-line treatment.

[0034] 4. Tumor efficacy was evaluated using the immune-modified Response Evaluation Criteria in Solid Tumors (ImRECIST). Treatment response was evaluated every 2 treatment cycles. When iUPD occurred, iUPD needed to be confirmed by a follow-up evaluation at least 4 weeks after the first recording date. Patients in this invention were defined as responders (R) (complete remission, partial remission, or stable disease of tumor lesions lasting more than 9 months after the first cycle of treatment) or non-responders (NR) (disease progression within 9 months during the first cycle of treatment).

[0035] 5. Sample collection and storage: All study subjects were instructed to brush their teeth thoroughly and then refrain from eating or taking any oral hygiene measures until the next morning when saliva samples were collected. Saliva samples were collected from each participant before morning brushing and placed in a 20-ml disposable sterile sample cup before treatment. Approximately 5 ml of saliva was collected from each person. The saliva did not contain impurities such as blood, food residues, or sputum. Otherwise, re-sampling was required. Saliva samples were collected before the first cycle of anti-tumor treatment. For patients in the NR group, saliva samples were collected again when tumor treatment efficacy evaluation showed disease progression (PD). Samples were stored in a -80°C refrigerator until further processing.

[0036] 6. Nucleic acid extraction and sequencing: DNA was extracted from all samples using the TIANamp Bacteria DNA Kit (DP302-02, TIANGEN, Beijing) and operated according to the manufacturer's instructions. The DNA concentration was calibrated using Qubit reagent (Q33230, ThermoFisher Scientific, Invitrogen, Carlsbad, CA, USA). The Fragmentation Mix (FRM) reaction solution (SQK-RPB004 Rapid PCR Barcoding Kit (Nanopore), Oxford Nanopore Technologies, Oxford, UK) was prepared in a 0.2 mL PCR tube, gently stirred and mixed evenly, incubated at 30 °C for 5 min, incubated at 80 °C for 1 min, and then quickly cooled on ice. After the PCR product was transferred to a 1.5 mL centrifuge tube, an equal volume of 50 μL purified magnetic beads was added, mixed evenly, left standing at room temperature for 5 min, and then centrifuged immediately. When the solution became clear, the supernatant was discarded. It was washed twice with 180 μL of 80% ethanol, the sample was immediately centrifuged at 3000 rpm for 1 min, placed on a magnetic stand to absorb the residual ethanol, dried with the lid open at room temperature for 30 s, and 10 μL of a mixed solution of 10 mM Tris-HCl (pH = 8.0, PHG0002, SAFC Biosciences, Lenexa, KS, USA) and 50 mM NaCl (S8210, Solarbio Life sciences, Beijing, China) was added, and the suspended magnetic beads were gently rotated. After incubating at room temperature for 2 min, the sample was placed on a magnetic stand. When the solution became clear, the eluate for subsequent use was aspirated. The concentration of each sample eluate was calibrated using Qubit reagent. According to the relevant concentrations, templates with the same absolute quantity were mixed into 10 μL, with a total concentration range of 100 to 200 ng / μL, 1 μL of RapidAdapter (RAP) solution (SQK-RPB004 Rapid PCR Barcoding Kit (Nanopore), Oxford Nanopore Technologies, Oxford, UK) was added, gently mixed, and incubated at room temperature for 15 min. When the loading mixture was ready within 15 min, sequencing was performed. The sequencing data volume was 500 Mb.

[0037] 7. Nanopore sequencing and microbial data analysis: The raw data files were generated in fast5 format using a MinION sequencer (Oxford Nanopore Technologies, Oxford, UK), and real-time identification and the generation of fastq files were completed using MinKnow software (version 1.11.5, Oxford Nanopore Technologies, Oxford, UK). Low-quality sequences were filtered using MinKnow software. Host DNA in the filtered data was removed using Minimap2 software (version 2.17.r941, Broad Institute, Cambridge, MA, USA) and the human genome reference sequence Hg38. The filtered sequences were subjected to multiple sequence alignment and the identification of pathogenic microorganisms using Centrifuge v1.0.3 (http: / / www.ccb.jhu.edu / software / centrifuge / , Center for Computational Biology, Johns Hopkins University, Baltimore, MD, USA) and the non-redundant nucleic acid database of NCBI. The species count and relative abundance table were input into R-base V.4.1.0 for statistical analysis. For alpha diversity, the present invention selected the ACE index, Chao index, and Simpson and Shannon indices for evaluation. For beta diversity analysis, the present invention used Bray-curits distance and Jaccard distance and visualized them through principal coordinate analysis (PCoA). Differentially classified taxa were identified in http: / / huttenhower.sph.harvard.edu / galaxy through linear discriminant analysis effect size (LEfSe) analysis.

[0038] 8. LC / MS Analysis: UHPLC-MS / MS analysis was performed using a Vanquish UHPLC system (ThermoFisher, Germany) and an Orbitrap Q Exactive TM HF mass spectrometer (Thermo Fisher, Germany). The sample was injected onto a Hypesil Gold column (100×2.1 mm, 1.9 μm) with a 17-min linear gradient at a flow rate of 0.2 mL / min. Solvent A in the positive polarity mode was 0.1% FA in water, and solvent B was methanol. Solvent A in the negative polarity mode was 5 mM ammonium acetate at pH 9.0, and solvent B was methanol. The solvent gradient was set as follows: 2% B for 1.5 min; 2 - 85% B for 3 min; 85 - 100% B for 10 min; 100 - 2% B for 10.1 min; 2% B for 12 min. The Q ExactiveTM HF mass spectrometer was operated in positive / negative polarity modes with a spray voltage of 3.5 kV, a capillary temperature of 320 °C, a sheath gas flow rate of 35 psi, an auxiliary gas flow rate of 10 L / min, an S-lens RF level of 60, and an auxiliary gas heater temperature of 350 °C.

[0039] 9. Data Processing and Metabolite Identification: The raw data files generated by UHPLC-MS / MS were processed using Compound Discoverer 3.1 (CD3.1, ThermoFisher) for peak alignment, peak selection, and quantification of each metabolite. The main parameter settings were as follows: retention time tolerance, 0.2 min; actual mass tolerance, 5 ppm; signal intensity tolerance, 30%; signal-to-noise ratio, 3; and minimum intensity, etc. After that, the peak intensities were normalized to the total spectral intensity. The normalized data were used to predict the molecular formula based on the added ions, molecular ion peaks, and fragment ions. Then, the peaks were matched with the mzCloud (https: / / www.mzcloud.org), mzVault, and MassList databases to obtain accurate characteristics and relative quantification results. Statistical analysis was performed using statistical software R (R version R-3.4.3), Python (Python version 2.7.6), and CentOS (CentOS release 6.6).

[0040] 10. Metabolite data analysis: These metabolites were annotated using the HMDB database and the LIPIDMaps database. Orthogonal partial least squares discriminant analysis (OPLS-DA) was performed using the OmicStudio tool (https: / / www.omicstudio.cn / tool). The Mann-Whitney U test was applied in the present invention to calculate the statistical significance (p-value). Partial least squares discriminant analysis (PLS-DA) was performed to calculate the VIP score. Metabolites with VIP > 1 and p-value < 0.05 and fold change ≥ 1.5 or ≤ 0.67 were considered differential metabolites. For the clustering heatmap, the data was standardized using the intensity region z-score of the differential metabolites and performed on https: / / www.omicstudio.cn / tool.

[0041] 11. Statistical analysis: Independent sample continuous variables were compared by the Mann-Whitney U test or the independent t-test. The Wilcoxon signed-rank test was used for paired sample tests. Categorical variables were compared by Fisher's exact test. A p-value < 0.05 was considered statistically significant. Spearman analysis was used to explore the correlations between different categories, as well as the correlations between categories and metabolites.

[0042] II. Results

[0043] 1. Patient characteristics: A total of 20 newly diagnosed stage III B-IV NSCLC patients participated in the study. The basic information and clinical characteristics of these patients are listed in Table 1. All patients were followed up until disease progression or death, and the last follow-up was conducted on January 3, 2023.

[0044] Table 1 Basic information and clinical characteristics of patients

[0045]

[0046]

[0047] According to clinical evaluation, the patients were divided into two groups. 10 patients were evaluated as R (immunotherapy responders), and the remaining 10 (n = 10) had disease progression within 9 months after the first cancer treatment cycle and were classified as NR (immunotherapy non-responders). The average age of the 20 NSCLC patients was approximately 60 years old, and most patients were male. No significant differences were observed between the patient groups in terms of age, BMI, gender, disease history, disease stage, smoking history, antibiotic use, PD-L1 expression, pathological type, and clinical stage.

[0048] 2. Bacterial composition and diversity between the NR and R groups: Overall, a total of 3,433 pathogenic bacteria were identified from the saliva samples of 20 patients through in-depth etiological metagenomic sequencing. The sequence reads of each saliva sample were normalized to 1 million. The taxonomic composition of all participants is shown in Figure 1 A of Figure 1 . This invention elucidates the distribution characteristics of the microbiota in lung cancer patients receiving immunotherapy. At the phylum level, the most abundant phyla are Actinobacteria, Bacillota, and Bacteroidota ( Figure 1 A of Figure 1 ). At the genus level, the top five genera are Rothia, Streptococcus, Prevotella, Porphyromonas, and Actinomyces ( Figure 2 B of Figure 2 ). At the species level, Rothia species, Streptococcus species, Actinomyces species, and Prevotella species are common ( Figure 2 C of Figure 2 ). Among them, the top 10 species are Rothia mucilaginosa, Streptococcus parasanguinis, Porphyromonas somerae, Prevotella melaninogenica, Streptococcus salivarius, Streptococcus mitis, Streptococcus oralis, Neisseria subflava, Actinomyces graevenitzii, and Prevotella pallens. To compare the differences in bacterial diversity related to immunotherapy sensitivity and drug resistance, this invention compared the alpha and beta diversities of the saliva microbiomes of the NR and R groups. At the species level, the ACE index ( Figure 2 A of Figure 2 F) was performed at the species level to visualize the dissimilarities between the two groups of saliva taxonomic communities. The results showed that the two groups shared common species and had similar species abundances. There was no significant difference in oral bacterial diversity between immunotherapy responders and non-responders in NSCLC patients.

[0049] 3. Differential species related to immunotherapy response: To further explore specific oral bacterial species related to the efficacy of immunotherapy, LEfSe analysis was performed on the NR and R groups. A total of 11 R-enriched species and 2 NR-enriched species were found. Among them, species of Neisseria and Actinomyces were common ( Figure 3 as shown in A). Specifically, the species significantly increased in the R group were Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, Actinomyces georgiae, and Lactobacillus fermentum, while the species significantly increased in the NR group were Granulicatella adiacens and Streptococcus oralis ( Figure 3 as shown in A). Except for Lactobacillus fermentum, the abundances of the other 12 differential species were ≥1000 and were used for further analysis. The distribution and abundances of these 12 differential species are shown in Figure 3 as shown in B. Through the Mann-Whitney U test, significant differences in species distribution between the two groups were further revealed. In the R group, Neisseria subflava was the most abundant species in the genus Neisseria, while Actinomyces meyeri was the most abundant species in the genus Actinomyces. The abundances of the two enriched species - Granulicatella adiacens and Streptococcus oralis were the highest in the NR group. These 12 differential species are listed in Table 2. The present invention shows that high concentrations of Neisseria and Actinomyces species are characteristics of lung cancer immunotherapy responders. In contrast, high concentrations of Granulicatella adiacens and Streptococcus oralis are characteristics of lung cancer immunotherapy non-responders.

[0050] Table 2 Twelve differential bacterial species related to immunotherapy response

[0051] Bacterial strain Average abundance (NR group) Average abundance (R group) LDA score P value Neisseria subflava 7562.64 30752.474 3.96 0.008a Neisseria perflava 977.60 3889.804 3.06 0.013a Neisseria flavescens 951.17 4023.372 3.10 0.016a Neisseria meningitidis 261.72 928.441 2.42 0.023a Actinomyces meyeri 1714.63 2620.344 2.79 0.028a Neisseria lactamica 147.50 567.081 2.20 0.028a Neisseria cinerea 306.61 1117.884 2.51 0.034a Neisseria polysaccharea 139.36 527.855 2.16 0.041a Actinomyces hongkongensis 818.49 2314.780 2.80 0.041a Actinomyces georgiae 506.14 1162.273 2.48 0.049a Streptococcus oralis 29954.22 14749.977 3.99 0.049a Granulicatella adiacens 9481.12 4577.990 3.55 0.041a

[0052] Given the potential variability of salivary microbiota abundance over time, the aim of the present invention was to determine whether these 12 differential bacteria were stable during lung cancer immunotherapy. To this end, the present invention collected saliva samples from all non-responding patients at disease progression (PD) and performed mNGS sequencing. The results showed that the abundances of these 12 differential species remained stable ( Figure 4 ), indicating that they could be used as predictive indicators of immunotherapy efficacy. Spearman correlation analysis was performed to measure the relationships between these 12 differential species ( Figure 4 ), and the results showed that there was a strong correlation among 7 Neisseria species. Similarly, there was also a strong correlation among 3 Actinomyces species. In addition, a moderate correlation was shown between Granulicatella adiacens and Streptococcus oralis.

[0053] The present invention applied the Youden index in the ROC curve as the optimal diagnostic cut-off value, i.e., the cut-off value ( Figure 10 ), which showed the high and low grouping of Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, Actinomyces georgiae, Lactobacillus fermentum, Granulicatella adiacens, and Streptococcus oralis flora. The area under the ROC curve (AUC) was used to evaluate the predictive sensitivity and specificity of the above 12 bacterial species indicators. As Figure 10 shown, it had good predictive ability for the immune efficacy of lung cancer patients. The present invention tested whether Neisseria subflava, Actinomyces meyeri, Granulicatella adiacens, and Streptococcus oralis were associated with the progression-free survival (PFS) of lung cancer immunotherapy. All patients were divided into two groups according to the median level of each species. Kaplan-Meier analysis showed that Neisseria subflava ( Figure 3 in C) and Actinomyces meyeri ( Figure 3Patients with a higher abundance of D) tended to have a longer PFS. In contrast, patients with a higher abundance of Granulicatella adiacens( Figure 3 E) and Streptococcus oralis( Figure 3 F) tended to have a shorter PFS. This indicates that the distribution of specific bacteria is closely related to the treatment outcome of patients, further suggesting that the detection of specific bacteria can be used as a predictive indicator for immunotherapy.

[0054] 4. Differential species related to PD-L1 expression: PD-L1 is an important therapeutic target and detection marker for immunotherapy. To further support the close link between oral bacterial distribution and immunotherapy, the present invention analyzed the expression and distribution of PD-L1 and different major bacteria in lung cancer of the N and NR groups. Patients were divided into two groups according to PD-L1 < 10% and PD-L1 ≥ 10%. Interestingly, LEfSe analysis showed that most R-enriched species, including: Neisseria subflava, Neisseria perflava, Neisseria flavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, and Actinomyces georgiae were also significantly increased in the high PD-L1 group( Figure 5 A). Spearman correlation analysis showed that Neisseria subflava was strongly correlated with PD-L1 expression (P < 0.001, R = 0.925)( Figure 5 B), while Actinomyces meyeri was moderately correlated with PD-L1 expression (P < 0.05, R = 0.571)( Figure 5 C), while Granulicatella adiacens( Figure 5 D) and Streptococcus oralis( Figure 5E) There was no significant correlation with PD-L1 expression. In the R group, Neisseria and Actinomyces species were significantly enriched and showed a positive correlation with PD-L1 expression, indicating that these bacteria are associated with immunotherapy sensitivity. In contrast, Granulicatella adiacens and Streptococcus orali were significantly enriched in the NR group and were associated with low PD-L1 expression, suggesting that they are bacteria related to lung cancer immunotherapy resistance. These results emphasize the importance and significance of the distribution differences of specific oral bacteria in lung cancer immunotherapy.

[0055] 5. Analysis of metabolic pattern differences between the NR and R groups: Nutrients act as chemical signals that regulate cell growth and metabolism. Excessive nutrition not only disrupts metabolic balance but also leads to excessive nutrient signaling, which in turn promotes abnormal cell metabolism and proliferation, contributing to the development of various diseases, including cancer. To study the metabolic patterns of populations with different immunotherapy effects and to search for metabolites related to the microbiome, the present invention performed LC / MS analysis. A total of 1008 metabolites in positive ion mode and 507 metabolites in negative ion mode were identified in saliva samples. Given the sensitivity and rapid changes of metabolomics to external factors, quality control (QC) samples are usually used to ensure data reliability. Based on the relative quantitative values of metabolites in the two ion modes, the Pearson correlation coefficient between QC samples was calculated, and the R 2 value was greater than 0.9, indicating high data quality ( Figure 6 A-B). In the PLS-DA model, the metabolic profile of NR could be distinguished from that of R ( Figure 6 A-B). Differentially expressed metabolites between the NR and R groups were determined based on specific criteria: VIP of the first two principal components of the PLS-DA model ≥ 1; fold change ≥ 1.5 or ≤ 0.67; P value < 0.05. The differentially expressed metabolites between the NR and R groups are listed in the circular volcano plot ( Figure 6 C-D). Eighty-four potential biomarker metabolites were identified in the negative ion mode, while 40 potential biomarker metabolites were identified in the positive ion mode. To determine whether these differentially expressed metabolites were stable during lung cancer treatment, the present invention collected saliva samples from all NR patients at PD and performed LC / MS analysis. The Wilcoxon rank sum test was used to compare the differentially expressed metabolites at different treatment time points. Finally, 70 differentially expressed metabolites in the negative ion mode and 28 differentially expressed metabolites in the positive ion mode remained stable. These stable differentially expressed metabolites are listed in Table 3 and were used for further analysis. In the negative ion mode (26 / 70, 37%) and positive ion mode (14 / 28, 50%), most of the differentially expressed metabolites were lipids and lipid-like molecules.

[0056] Table 3 Differentially expressed metabolites in the cation mode and anion mode

[0057]

[0058]

[0059]

[0060] 6. Among the differential lipids and lipid-like metabolites, most (15 in the negative ion mode and 14 in the positive ion mode) were upregulated in the NR group. The abundance heatmap of the stable differential metabolites in the two ion modes is shown ([ Figure 7 A-B). Most of the differential lipids and lipid-like metabolites clustered together. Correlation analysis showed that the total abundance of all lipids and lipid-like metabolites was negatively correlated with PD-L1 expression ([ Figure 8 )(negative ion mode P = 0.029, R = -0.534 and positive ion mode P = 0.018, R = -0.572). These results revealed that abnormal lipid metabolism was highly enriched in the NR group and was closely related to low PD-L1 expression. In summary, the present invention elucidated the metabolic characteristics of lung cancer patients receiving immunotherapy.

[0061] 7. Specific correlations between different salivary microbial community species and different lipid metabolites: To further reveal the specific correlations between differential bacteria and differential metabolites, the present invention performed Spearman correlation analysis ([ Figure 9 A-B). Several differential lipids and lipid-like metabolites were related to differential bacteria. In the positive ion mode, most of the differential phosphocholines and acylcarnitines were negatively correlated with differential Actinomyces species. Most of the phosphocholine metabolites were positively correlated with Granulicatella adiacens or Streptococcus oralis. PC(14:0e / 26:2) and glycoursodeoxycholic acid were negatively correlated with differential Neisseria species. In the negative ion mode, most of the lipids and lipid-like metabolites enriched in the NR group were positively correlated with Granulicatella adiacens. All differential monoacylglycerol phosphatidylinositols were negatively correlated with differential Actinomyces species. LPS20:2, myristic acid, SM(d14:0 / 22:0) and tetraene prostaglandin E2 were negatively correlated with differential Neisseria species, while most of the lipids and lipid-like metabolites enriched in the R group were positively correlated with differential Neisseria species ([ Figure 9 A-B). In summary, certain lipids and lipid-like metabolites enriched in the NR group were closely related to specific salivary species, which were related to the efficacy of immunotherapy. This indicates that the specific oral microbial community distribution related to lung cancer immunotherapy is closely related to the abnormal regulation of lipid metabolism signals.

[0062] The embodiments described above are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A predictive indicator for the efficacy of immunotherapy for lung cancer, characterized in that: The predictive indicators include specific oral flora; the specific oral flora include Neisseria subflava, Neisseria perflava, Neisseriaflavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis, Actinomyces georgiae, Granulicatella adiacens and Streptococcus oralis.

2. The prediction indicator according to claim 1, characterized in that The sample of the predictive indicator is saliva.

3. Use of the quantitative detection reagent for prediction indexes according to claim 1 in the preparation of a reagent for predicting the prognosis of lung cancer.

4. The use according to claim 3, characterized in that: The efficacy of lung cancer immunotherapy is predicted by detecting the distribution concentration of the specific oral flora.

5. The use according to claim 4, characterized in that: The Youden index of the ROC curve was used as the critical value, and the lung cancer patients with a test score ≥ Youden index were defined as a high concentration group, and the lung cancer patients with a test score < Youden index were defined as a low concentration group; when the specific oral flora Neisseria subflava, Neisseria perflava, Neisseriaflavescens, Neisseria meningitidis, Neisseria lactamica, Neisseria cinerea, Neisseria polysaccharea, Actinomyces meyeri, Actinomyces hongkongensis and Actinomyces georgia were distributed in high concentrations, it was a response to lung cancer immunotherapy; when the specific oral flora Granulicatella adiacens and Streptococcus oralis were distributed in high concentrations, it was a non-response to lung cancer immunotherapy.

Images