CN113186287B - Biomarker for non-small cell lung cancer typing and application thereof - Google Patents

Abstract

The invention relates to a biomarker for non-small cell lung cancer typing and application thereof, belonging to the technical field of medical detection. The biomarker comprises at least 5 genes such as TP53, STK11, PTEN, NFE2L and KRAS. By using the biomarkers, squamous carcinoma and adenocarcinoma in small cell lung cancer are classified, when the minimum number of the markers is 5, the AUC of a classification diagnosis ROC curve is 0.700, when the markers are further increased to 10, the AUC of the classification diagnosis ROC curve is 0.734, when all the biomarkers are used, the AUC can reach 0.786, and the diagnosis capability is excellent.

Description

Biomarkers for non-small cell lung cancer classification and their applications

Technical Field

The present invention relates to the field of medical detection technology, and in particular to a biomarker for non-small cell lung cancer typing and application thereof.

Background Art

Lung cancer is a heterogeneous disease. The main basis for the selection of existing lung cancer treatment methods is pathological classification and staging diagnosis. Pathological classification is generally determined by histology: important classifications include small cell vs non-small cell, adenocarcinoma vs squamous cell carcinoma, etc. The distinction between various morphological subtypes of lung cancer is necessary in guiding patient management. Different pathological subtypes have different corresponding treatment strategies. For example, small cell undifferentiated lung cancer is highly malignant, easy to metastasize in the early stage, sensitive to radiotherapy and chemotherapy, and non-surgical treatment with systemic chemotherapy and local radiotherapy is its main treatment method. Non-small cell lung cancer mainly includes squamous cell carcinoma and adenocarcinoma. Surgery is the main choice for stage I and II non-small cell lung cancer, and some can be cured by postoperative adjuvant radiotherapy and chemotherapy. For non-small cell lung cancer, the tumor cells of adenocarcinoma grow faster, and most of them metastasize in the early stage, mainly in the blood, so they are more sensitive to chemotherapy drugs, and the reflex treatment effect is poor, so surgery, chemotherapy, immunotherapy, targeted therapy and other methods are often chosen. Squamous cell carcinoma develops relatively slowly. In the early stages, most invasions are local, with lymph node metastasis being the main route. Distant metastasis occurs relatively late, so squamous cell carcinoma is more sensitive to radiotherapy, and surgery, radiation, immunotherapy, and other methods are generally used.

Conventional lung cancer pathology judgment mostly relies on histological pathological diagnosis results. However, due to the personal experience of doctors and various reasons, the pathological subtype classification is incorrect, the error rate is high, and there are no other relevant quality control measures.

In addition to conventional radiotherapy and chemotherapy, lung cancer treatment has gradually entered the era of precision medicine: traditional imaging diagnosis and pathological diagnosis can no longer meet the needs of precision medicine. At present, in addition to traditional diagnostic methods, the precision diagnosis of lung cancer can detect specific therapeutic target markers at the molecular level, that is, detecting tumor-specific gene mutations and characteristic proteins, RNA, metabolites and other molecular markers, and its importance is becoming increasingly prominent.

Gene mutation detection has become one of the essential processes in routine treatment. The most famous example is the detection of the epidermal growth factor receptor (EGFR) in cancer cells. Lung cancer patients with malfunction of this protein (usually caused by gene mutation) show a significant response to a drug that specifically targets the EGFR protein. The detection of related gene mutations of this target has been included in the relevant lung cancer treatment guidelines.

In addition, the detection of drug target-related variant genes is an almost mandatory test item for lung cancer patients, especially those in the advanced stage. However, limited tissue samples and the need to evaluate an increasing number of therapeutic target markers have greatly increased the current diagnostic needs. Studies on the reproducibility of histological diagnosis have shown differences in judgments within and between pathologists: incorrect results of pathological judgments, poorly differentiated tumors, and contradictory immunohistochemical results, etc., have challenged the accuracy of current precision medicine for lung cancer. Therefore, a reliable means of determining the pathological subtypes of lung cancer is needed.

Summary of the invention

Based on this, it is necessary to provide a biomarker for the classification of non-small cell lung cancer to address the above problems. By using the different expression profiles of variant genes in adenocarcinoma vs squamous cell carcinoma of non-small cell lung cancer, a biomarker that can be used to classify adenocarcinoma and squamous cell carcinoma in non-small cell lung cancer can be obtained, providing a molecular-based discrimination method for the two pathological subtypes.

A biomarker for non-small cell lung cancer classification, including: NFE2L2, TP53, CDKN2A, PTEN, MUC16, PIK3CA, RYR2, ATP10A, SLCO1B1, RASA1, ZFHX4, KMT2D, KRAS, EGFR, ANK1, BRAF, PCLO, PTPRD, ASTN1, ADGRG4, NOTCH4, FAT3, PCDH15, ROBO2, KEAP1, TENM 2, TSHZ3, SETBP1, CACNA1E, XIRP2, ASXL3, ZNF804A, NALCN, FBN2, SPTA1, MUC17, RBM10, SETD2, MXRA5, ST6GAL2, RP1L1, ASPM, FLG, HECW1, COL12A1, COL14A1, AFF2, SMARCA4, SDK1, EPHB6, UBA6, SF1, MGAM,PCDH11X,CO L5A1, ALK, ATM, VCAN, ZNF536, EPHA6, PDZRN3, PTPRC, ITGA4, DRD5, MYH11, DACH1, CTNNB1, TNR, NPAP1, TLR4, F8, ABCB5, RYR1, OR4C15, MYOM2, DMD, FOLH1, FRMPD4, ADAMTS17, SHANK1, FAM171A1, CCKBR , TRPS1, HMCN1, At least five of the following genes: NTRK3, ATRX, AHNAK, SALL1, PRUNE2, CSMD1, PDGFRA, ADAMTS12, GRM1, SYNE2, OR8H2, TEP1, CCDC178, STK11, NID1, DCSTAMP, STAG2, MET, BCL11B, ZNF226, NTRK2, NEDD4, BTK, TMTC3, RBM15, KLK2, ITK, CMKLR1.

The present inventors have conducted a thorough study of the gene map of non-small cell lung cancer and obtained the related genes (NFE2L2, TP53, CDKN2A, PTEN, MUC16, PIK3CA, RYR2, ATP10A, SLCO1B1, RASA1, ZFHX4, KMT2D) that have a higher mutation frequency in squamous cell carcinoma than in adenocarcinoma, and the related genes (KRAS, EGFR, ANK1, BRAF, PCLO, PTPRD, ASTN1, ADGRG4, NOTCH4, FAT3, P CDH15, ROBO2, KEAP1, TENM2, TSHZ3, SETBP1, CACNA1E, XIRP2, ASXL3, ZNF804A, NALCN, FBN2, SPTA1, MUC17, RBM10, SETD2, MXRA5, ST6GAL2, RP1L1, ASPM, FLG, HECW1, COL12A1, COL14A1, AFF2, SMARCA4, SD K1, EPHB6, UBA6, SF1, MGAM, PCDH11X, COL5A1, A LK, ATM, VCAN, ZNF536, EPHA6, PDZRN3, PTPRC, ITGA4, DRD5, MYH11, DACH1, CTNNB1, TNR, NPAP1, TLR4, F8, ABCB5, RYR1, OR4C15, MYOM2, DMD, FOLH1, FRMPD4, ADAMTS17, SHANK1, FAM171A1, CCKBR, TRPS1, HMCN1, NTRK3, ATRX, AHNAK, SALL1, PRUNE2, CSMD1 , PDGFRA, ADAMTS12, GRM1, SYNE2, OR8H2, TEP1, CCDC178), and mutant genes only found in adenocarcinoma (STK11, NID1, DCSTAMP, STAG2, MET, BCL11B, ZNF226, NTRK2, NEDD4, BTK, TMTC3, RBM15, KLK2, ITK, CMKLR1), and a differentiation and judgment model for the above genes was established. By using the above biomarkers, the two pathological subtypes of squamous cell carcinoma and adenocarcinoma in non-small cell lung cancer can be effectively distinguished.

In one embodiment, the biomarkers include: TP53, STK11, PTEN, NFE2L and KRAS genes.

For molecular diagnosis, minimizing the number of diagnostic markers while ensuring diagnostic sensitivity and specificity can effectively reduce the complexity of the operation, ensure the repeatability and reliability of the test results, and greatly reduce the economic burden on patients. The above biomarkers provide a combined application of 5 biomarkers, which reduces the application cost, and the AUC value of the RCO curve used for the typing judgment model can reach 0.700.

In one embodiment, the biomarkers include: TP53, STK11, PTEN, NFE2L, KRAS, EGFR, CDKN2A, BRAF, TSHZ3 and PIK3CA genes. The AUC value of the RCO curve of the judgment model obtained by typing with the above 10 biomarkers can reach 0.734.

In one embodiment, the biomarkers include: TP53, STK11, PTEN, NFE2L, KRAS, EGFR, CDKN2A, BRAF, TSHZ3, PIK3CA, SETD2, MUC16, RYR2, PTPR, PCLO, RP1L1, ASTN1, SPTA1, ASXL3 and XIRP2. The AUC value of the RCO curve of the judgment model obtained by typing with the above 20 biomarkers can reach 0.747.

In one embodiment, the biomarkers include: NFE2L2, TP53, CDKN2A, PTEN, MUC16, PIK3CA, RYR2, ATP10A, SLCO1B1, RASA1, ZFHX4, KMT2D, KRAS, EGFR, ANK1, BRAF, PCLO, PTPRD, ASTN1, ADGRG4, NOTCH4, FAT3, PCDH15, ROBO2, KEAP1, TENM 2, TSHZ3, SETBP1, CACNA1E, XIRP2, ASXL3, ZNF804A, NALCN, FBN2, SPTA1, MUC17, RBM10, SETD2, MXRA5, ST6GAL2, RP1L1, ASPM, FLG, HECW1, COL12A1, COL14A1, AFF2, SMARCA4, SDK1, EPHB6, UBA6, SF1, MGAM,PCDH11X , COL5A1, ALK, ATM, VCAN, ZNF536, EPHA6, PDZRN3, PTPRC, ITGA4, DRD5, MYH11, DACH1, CTNNB1, TNR, NPAP1, TLR4, F8, ABCB5, RYR1, OR4C15, MYOM2, DMD, FOLH1, FRMPD4, ADAMTS17, SHANK1, FAM171A1, CC KBR,TRPS1,H MCN1, NTRK3, ATRX, AHNAK, SALL1, PRUNE2, CSMD1, PDGFRA, ADAMTS12, GRM1, SYNE2, OR8H2, TEP1, CCDC178, STK11, NID1, DCSTAMP, STAG2, MET, BCL11B, ZNF226, NTRK2, NEDD4, BTK, TMTC3, RBM15, KLK2, ITK and CM KLR1.

Using all biomarkers for typing, the AUC value of the RCO curve of the judgment model can reach 0.786.

The present invention also discloses the application of the above biomarker in the diagnosis of adenocarcinoma and squamous cell carcinoma in patients with non-small cell lung cancer.

It is understandable that when detecting the above biomarkers (i.e., gene mutations), various methods that can be used to detect gene mutations in the art can be used, for example:

1) Sequencing technology, including second-generation sequencing technology and third-generation sequencing technology. The principle of second-generation sequencing technology is massive parallel sequencing (MPS). In some embodiments, the second-generation sequencing technology can be: 1. Based on DNA polymerase synthesis sequencing technology (Sequencing by synthesis technology, SBS), represented by Illumina (reversible terminator sequencing), Thermo Fisher/Life Technologies (Ion Torrent), GenapSys, Roche Diagnostics (454 pyrophosphate sequencing), etc.; 2. Based on DNA ligase ligation sequencing technology (Sequencing by ligation technology, SBL), represented by BGI/Complete Genomics (composite probe-anchor molecule ligation, cPAL), Thermo Fisher/Applied Biosystems (Sequencing by Oligonucleotide Ligation and Detection, SOLiD), etc. The third-generation sequencing technology is a single-molecule sequencing technology. In some embodiments, the third-generation sequencing technology can be: single-molecule real-time fluorescence sequencing technology (SMRT, Pacific Biosciences), nanopore sequencing technology [Oxford Nanopore Technologies (ONT), Genia Technologies and Stratos Genomics (Roche Diagnostics)], nanogate sequencing technology (Nanogate, Quantum Biosystems), DNA hydrolysis-based sequencing technology (Sequencing by de-synthesis, pyrophosphorolysis, Base4) and any other sequencing methods known in the art.

2) Microarray hybridization technology: such as SNP microarray;

3) PCR-based detection technologies for variant sites: such as KASP typing method, ligase detection reaction (LDR) typing method, Taqman probe method, etc.

In one embodiment, the biomarker is detected as a biomarker in blood and/or tissue.

It is understandable that the above detection is also applicable to other types of biological samples. However, the use of the above samples has the advantages of easy sample acquisition and wide applicability.

The present invention also discloses the use of a reagent for detecting the above-mentioned biomarkers in a biological sample in the preparation of a non-small cell lung cancer typing diagnostic reagent or diagnostic equipment.

It is understandable that the above reagents, such as a test kit, and the equipment, such as an integrated detection equipment, can be adjusted according to specific application requirements.

The present invention also discloses a detection kit for non-small cell lung cancer typing diagnosis, comprising a reagent for detecting the above-mentioned biomarker in the claim.

The present invention also discloses a system for diagnosing non-small cell lung cancer typing, comprising:

Analytical device: used to obtain the gene variation of the above-mentioned biomarkers in the biological sample of the subject to be evaluated, and input it into the evaluation model for typing evaluation;

Output device: used to output the above evaluation results.

In one embodiment, the evaluation model is established by the following method: the evaluation model is established by the following method: obtaining a number of adenocarcinoma and squamous cell carcinoma biological samples, sequencing to obtain the gene mutation of the biomarkers, and establishing a typing model using a random forest model, in which importance = TRUE and ntree = 100, mtry = 2, to obtain a non-small cell lung cancer typing diagnosis model.

Compared with the prior art, the present invention has the following beneficial effects:

The biomarkers for non-small cell lung cancer classification of the present invention can utilize the combination of the above-mentioned biomarkers to classify squamous cell carcinoma and adenocarcinoma in small cell lung cancer. When at least 5 markers are used, the AUC of the classification diagnosis ROC curve is 0.700. When the markers are further increased to 10, the AUC of the classification diagnosis ROC curve is 0.734. When all biomarkers are used, the AUC can reach 0.786, which has excellent diagnostic ability.

For patients with small cell lung cancer, the genetic test report results of the above-mentioned biomarkers can be used to conduct a mutual verification process with the pathological results to ensure that the pathological diagnosis results are correct, which plays an important role in the next step of precise treatment.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a ROC-AUC diagram of the model established using all markers to distinguish the effects in the validation set.

FIG2 is a ROC-AUC graph showing the effect of differentiating between the models established using a combination of 20 markers in the validation set.

FIG3 is a ROC-AUC graph showing the effect of distinguishing the models established using a combination of 10 markers in the validation set.

FIG4 is a ROC-AUC graph showing the effect of distinguishing the models established using a combination of five markers in the validation set.

DETAILED DESCRIPTION

In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and/or" used herein includes any and all combinations of one or more related listed items.

The reagents used in the following examples, unless otherwise specified, are all commercially available; the methods used in the following examples, unless otherwise specified, are all implemented by conventional methods.

illustrate:

TCGA: The full name is The Cancer Genome Atlas, which includes data on more than 30 types of tumors. It is a Cancer Genome Atlas (TCGA) project initiated by the National Cancer Institute (NCI) and the National Human Genome Research Institute (NHGRI). It is a comprehensive, multidimensional atlas for a variety of cancer genomes. The areas involved include not only genome sequencing, but also epigenomic sequencing such as transcriptome and methylation, as well as final integrated analysis, and linking them with clinical and imaging data.

In the present invention, an adenocarcinoma patient refers to a non-small cell adenocarcinoma patient whose pathological test results are jointly confirmed by two or more pathology experts; a squamous cell carcinoma patient refers to a non-small cell squamous cell carcinoma patient whose pathological test results are jointly confirmed by two or more pathology experts.

Example 1

Based on the public database, the initial screening of variant gene markers for pathological subtype classification of non-small cell lung cancer includes the following steps:

1. Screening of candidate mutation sites.

The whole genome sequencing data of tumor tissues of patients with non-small cell tumors were obtained from the TCGA database (https://portal.gdc.cancer.gov/): In this study, the whole genome sequencing data of 561 patients with non-small cell lung cancer (including 286 cases of adenocarcinoma and 275 cases of squamous cell carcinoma) were downloaded, and the mutation sites were calculated using four different software (mutect, varscan, muse and somaticsniper). The mutation sites that were simultaneously called by at least two of the four mutation software in the sample were taken as candidate mutation sites.

2. Screening of potential markers.

Differential analysis was performed based on the adenocarcinoma group data set and the squamous cell carcinoma group data set: fisher.test analysis was used, and variant genes with p≤0.05 were selected as potential markers, see the table below.

Table 1. Potential markers

Example 2

This example analyzes and verifies the potential markers obtained above on clinical samples, including the following steps:

1. Tissue sample acquisition:

A total of 374 FFPE section samples were collected from Jinan University, which were identified as non-small cell lung cancer (191 adenocarcinomas and 183 squamous cell carcinomas) by relevant experts.

2. Sample sequencing analysis:

FFPE tissue samples were analyzed by whole genome sequencing by a third party (NextCode Biotech).

3. Model establishment.

3.1 Preliminary establishment of the model.

All potential markers obtained in Example 1 above were used to detect and judge an independent validation set, i.e., the tissue samples of the 191 adenocarcinoma and 183 squamous cell carcinoma non-small cell lung cancer patients mentioned above. The random forest model was used for modeling analysis (R package randomForest), and 20 repetitions were performed according to the 6:4 split. Through repeated exploration and trial of the model establishment conditions, importance = TRUE, ntree = 100, and mtry = 2 were set, and the ROC curve AUC of the model was obtained to be as high as 0.786, as shown in Figure 1.

3.2 Screening of markers.

By using the random forest model for modeling and analysis, 191 adenocarcinoma and 183 squamous cell carcinoma tissue samples of non-small cell lung cancer patients were tested and judged. The random forest model was used for modeling and analysis, and 20 repetitions were performed according to the 6:4 split. Pearson correlation was used to perform pairwise correlation analysis on all markers, and the optimal combination of 20 MARKERs was obtained through the exhaustive random combination method. The ROC curve AUC of the model can be as high as 0.747, as shown in Figure 2.

3.2 Optimize markers.

By using the random forest model for modeling and analysis, 191 adenocarcinoma and 183 squamous cell carcinoma tissue samples of non-small cell lung cancer patients were tested and judged. The random forest model was used for modeling and analysis, and 20 repetitions were performed according to the 6:4 split. At the same time, all markers were analyzed pairwise by Pearson correlation, and the optimal combination of 10 MARKERs was obtained by exhaustive random combination method. The ROC curve AUC of the model can be as high as 0.734, as shown in Figure 3.

3.3 Further optimization of markers.

By using the random forest model for modeling and analysis, 191 adenocarcinoma and 183 squamous cell carcinoma tissue samples of non-small cell lung cancer patients were tested and judged. The random forest model was used for modeling and analysis, and 20 repetitions were performed according to the 6:4 split. Pearson correlation was used to analyze the correlation between all markers, and the optimal combination of 5 MARKERs was obtained through the exhaustive random combination method. The ROC curve AUC of the model can be as high as 0.700, as shown in Figure 4.

Example 3

Thirteen samples from Jinan University that were clinically diagnosed as non-small cell lung cancer, which were different from the sample set in Example 2, were selected for analysis using the 20 MARKER combination models established in Example 2. The analysis results were compared with the clinical expert judgment results. The results are shown in the following table.

Table 2. Clinical validation results

Case Model typing results Expert judgment results No. 1 Adenocarcinoma Adenocarcinoma No. 2 Squamous cell carcinoma Squamous cell carcinoma No.3 Adenocarcinoma Squamous cell carcinoma No. 4 Squamous cell carcinoma Adenocarcinoma No. 5 Adenocarcinoma Adenocarcinoma No. 6 Adenocarcinoma Adenocarcinoma No.7 Squamous cell carcinoma Squamous cell carcinoma No. 8 Adenocarcinoma Adenocarcinoma No. 9 Squamous cell carcinoma Adenocarcinoma No. 10 Adenocarcinoma Adenocarcinoma No. 11 Adenocarcinoma Adenocarcinoma No. 12 Adenocarcinoma Adenocarcinoma No. 13 Squamous cell carcinoma Squamous cell carcinoma

From the above results, it can be seen that the biomarkers of the present invention and the above model are more accurate in the classification of squamous cell carcinoma or adenocarcinoma in small cell lung carcinoma, with a consistency of more than 76.9% with the expert judgment.

The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims (2)

1. A biomarker for the classification of squamous cell carcinoma and adenocarcinoma in non-small cell lung cancer, characterized in that it consists of TP53, STK11, PTEN, NFE2L, KRAS, EGFR, CDKN2A, BRAF, TSHZ3, PIK3CA, SETD2, MUC16, RYR2, PTPR, PCLO, RP1L1, ASTN1, SPTA1, ASXL3 and XIRP2 genes. 2. Use of a reagent for detecting the expression amount of the biomarker as claimed in claim 1 in a biological sample in the preparation of a diagnostic reagent or diagnostic device for the classification of squamous cell carcinoma and adenocarcinoma in non-small cell lung cancer.

Images